Perfluorchemicals (PFCs)

Figure 3-1: Fluorine and carbon; the two building blocks of the remarkable molecules knows as the perflurochemicals (PFC)s.

Physical Chemistry and Synthesis

Perfluorchemicals (PFCs) are derived from hydrocarbons by replacing hydrogen atoms with fluorine atoms, typically using common organic hydrocarbons as substrates. This is accomplished by one of three methods; the oldest of which is via a highly exothermic vapor-phase reaction employing fluorine gas. An alternative method is the more stable cobalt trifluoride technique which was developed during the Manhattan Project in World war II (WWII) [290]. Electrochemical fluorination, developed by Simmons in1950 [291] has increasingly replaced the earlier techniques.

Ectrochemical fluorination yields more homogeneous products with less carbon-carbon bond cleavage and is better suited to smaller scale production of molecules for use in research and development applications. However, whether done by electrolysis or by using the pre-Simmons method of reaction with high-valent metal fluorides, both laboratory and industrial scale PFC manufacturing and synthesis have in the past resulted in impure and poorly characterized compounds (unless perfluorinated building blocks are employed as starters). This occurs because the large difference (~15 kcal per M^-1) in the carbon-hydrogen bond versus the carbon-fluorine bond energies (and the release of this energy during the synthetic process) results in undesired side reactions, isomerisations, and  polymerizations creating a mélange of compounds which are not fully characterized, let alone purified.[292]

PFC carbon chain length is variable, and these, in addition to the attached moieties, determine the individual properties of a given molecule. Liquid PFCs that are useful as gas and/or heat exchange media in the lungs all exploit the utterly unique properties of the carbon-fluorine (C-F) bond and the larger size of fluorine atoms compared to hydrogen atoms.  The C-F bond is the strongest covalent bond found in organic chemicals (average 485 kJ mol^-1, compared to ~413 kJ mol^-1 for a standard C-H bond)  [293] and this bond-strength is amplified as the number of fluorine atoms on each carbon atom increases.[293]

The electroattracting nature of the fluorine atoms further increases the strength of the C-C bonds and the larger size of fluorine atoms (estimated van der Waals radius of 147 versus 120 pm) [294] and their high electron density result in a compact electron shield that ensures effective protection of the molecule’s backbone. The dense electron shell of the fluorine atoms may also provide protection against nucleophilic attack. The PFCs thus have very high intra-molecular (covalent) bonding and very low intermolecular forces (van der Waals interactions).[295]

Physical Properties

These liquids are clear, colorless, odorless, non-conducting, nonflammable, and are both hydrophobic and lipophobic. Replacement of hydrogen atoms by fluorine atoms results in high thermal stability and chemical inertness. PFCs do not undergo decomposition (except at temperatures above ~350ºC) and they are not metabolized or acted upon by any enzymatic system. They are ~1.8 times as dense as water; and are capable of dissolving large amounts of physiologically necessary gases (45 to 55 ml O₂/dl and16 to 210 ml CO₂/dl).[296]  O₂ dissolution in PFCs is via O₂ occupying intermolecular sites in the liquid, unlike the pH and concentration dependent porphyrin binding in hemoglobin which is responsible for the sigmoidal oxyhemoglobin dissociation curve.[297]   O₂ solubility in PFCs is a linear function of the pO₂ (per Henry’s law) and the structure of the particular PFC.[298]  Solubility of O₂ and other gases in PFCs is a function of the NMR T1 relaxation constant which determines the intermolecular cavity size and the presence and character of large channels within the liquid. Linear aliphatic structures more easily allow the formation of such channels while planar structures result in more closely interlocked layers which accommodate less gas. While the solubility of O₂ is greater in aliphatic than in aromatic PFCs this is not the case with CO₂. O₂ solubility increases with the degree of fluorination and the O₂-dissolving capacity of aliphatic PFCs having 9-11 carbons is in the range of 40 to 60 mole fractions of O₂, or roughly twice that of aromatic molecules.

The O₂ content of perfluocarbons at 1 atmosphere (atm) of 100% O₂ is approximately 20 times that of water, twice that of blood, and 1.5 times that of an equal volume of gaseous O₂. If the PFCs are compared to blood in terms of O₂ carrying capacity under normal atmospheric conditions it is immediately obvious that PFCs carry only a fraction of the O₂ that hemoglobin does. However, it is not simply the O₂ dissolving capacity of PFCs that is important, but also their O₂ delivery capability. The O₂ diffusion rate from the PFC-filled alveoli to the alveolar capillaries is quite high when saturated with O₂ at 760 torr [299] and due to their 50% greater O₂ carrying capacity than O₂ delivered as a gas at an FiO₂ of 100%, the delivered O₂ to the blood is 25% to 50% greater than is possible using 100% gaseous O₂ for ventilation. The use of such a high FiO₂ is sustainable with PFCs because, for reasons not yet understood, O₂ toxicity does not occur in their presence as a liquid in the lung under normobaric conditions.

A remarkable and essential feature of the PFCs in liquid assisted ventilation is that they are essentially insoluble in both water (7 to 11 ppm at STP) and alcohols, and are only sparingly soluble in some lipids.[299] The PFCs have a kinematic viscosity (ratio of viscosity to density) similar to that of water [300] and an extremely low surface tension (15 to 19 dyn/cm²) and dielectric constant; again secondary to weak intermolecular forces resulting from the exterior fluorine atoms.[301] The volatility of PFCs varies widely depending upon the molecular weight of the molecule, with vapor pressures ranging from ~1 to over 80 torr at 25ºC. Vapor pressure is the critical determinant of the half-life of elimination of a given PFC from the lungs following both intrapulmonary and intravenous administration. Vapor pressure also dictates the viscosity of a given PFC and thus PFC-gas and PFC-surface shear interactions.[302],[303]  Vapor pressure is also a critical determinant of toxicity as will be discussed under the heading Toxicology, below.

Commercially Available PFCs

Historically, the requirements of a PFC for use as a gas exchange medium in liquid ventilation are excellent solubility for respiratory and putative therapeutic gases such CO, NO, and H₂S, kinematic viscosity in the range of ~ 0.50 to 2.2, low to moderate vapor pressure (1 to 15 torr) with a cutoff of 20 torr, a high spreading coefficient and a very low surface tension.[303]   Additionally, a PFC that is to be used for intrapulmonary heat exchange must have a viscosity and freezing point appropriate to the application. These criteria, particularly with respect to vapor pressure and viscosity, may change in the future, depending upon the application, medical condition being treated, or large airway diameter of the patient.

No existing PFC combines all of these desirable properties, and certainly no single PFC can be tailored to have the widely varying physical properties required for a particular pathology or patient. In addition, from a physical chemistry standpoint, no single PFC is likely to combine all these properties, even in the sphere of providing assisted ventilation in ARDS; and recent research has begun to focus not only on the creation of novel of PFCs for liquid assisted ventilation applications, but also on investigating mixtures of different PFCs to provide an optimum ventilating medium which can be formulated to meet the needs of a given application.[304]

Unfortunately, de novo flurochemical synthesis involves the use of extremely toxic and hazardous materials such as fluorine gas, hydrogen fluoride, silver difluoride, or the halogen fluorides, and yields a poor ratio of desired end product to undesired (and costly to dispose of) side-products and waste.[293] Even flurochemical synthesis using per- and poly-fluroinated reagents, the ‘building block’ approach is costly, has low selectivity for many compounds and requires formidable expertise [305], although this is changing.[306]

Once the synthesis is completed, numerous and costly purification steps such as lengthily refluxing, spinning band distillation and reparative vapor phase chromatography must be undertaken, adding greatly to the cost, and making well characterized synthesis and purification of quantities sufficient for use in liquid assisted ventilation or blood substitutes a full-time effort and the province of expert chemists and dedicated facilities.[293]  Historically, this has severely limited the development and commercial production of high purity, completely chemically characterized novel PFCs.

As a result, most of the initial work in liquid ventilation (including that done for liquid assisted pulmonary cooling (LAPC)) was carried out using commercially available PFCs such as perflurodecalin, Rimar-101™ (Miteni Corporation, Milan, Italy), or the Fluorinert™ liquids (3M Company, St. Paul, MN). Table 3, below, shows some of the physical properties of a number of commercially available PFCs used for leak testing, heat exchange and for cleaning applications in the electronics industry marked by 3M Corporation as the Fluorinert™  ‘liquids,’ as well as those of Rimar-101 and Perflubron™ (Alliance Pharmaceuticals, San Diego, CA).

A serious disadvantage to Fluorinert™ PFCs and all other industrial grade PFCs (as well as most reagent grade materials available from laboratory chemical suppliers) is that they are not chemically defined in terms of chain length or even precise chemical composition. As examples, FC-75 has been shown to have as many as 6 peaks when evaluated by gas chromatography and F-tripropylamine (FTPA), one of the components in the first FDA approved blood substitute, Flusol-DA, contained only 27% of perfluorinated FTPA in addition to a number of other uncharacterized compounds.[307]  FC-43, a PFC used extensively as an experimental oxygen carrying blood substitute and for liquid assisted ventilation is chemically ~ 85% perfluoro-tri-n-butylamine (C₁₂F₂₇) with the unit structure shown in Figure 3-2. However, the perfluoro-tri-n-butylamine molecules may be present as polymers of varying lengths, and other related fluorinated molecular species are also present.

The degree of polymerization as well as the physical properties of the species which comprise the ~15% balance of FC-84, are such that the average vapor pressure, boiling point, melting point, thermal conductivity and other physical properties of FC-84 are fairly uniform from lot-to-lot.[308],[184] However, two of the most critically important determinants of the utility of PFCs in liquid ventilation applications are their vapor pressure (and thus their viscosity) and their direct chemical toxicity. Vapor pressure is of great concern, because even if the average vapor pressure of the liquid is quite low (i.e., 1.3 torr at STP for FC-43) if even a small percentage of the species present have a far higher vapor pressure, then that fraction of the liquid can turn into a gas and create long-lasting and mechanically disruptive bubbles in lung tissue under conditions of baro- and/or volu-trauma; and will create ‘sponge rubber lung’ syndrome due to stable intra-alveolar gas bubble formation by vaporizing between surfactant and the alveolar epithelium. This phenomenon, known as hyper-inflated non-collapsible lungs (HNCL) occurs even under the relatively non-traumatic conditions of PLV if the vapor pressure is low enough; as is the case with F-alkylfuran in FC-75.[309] Similarly, the unspecified and often uncharacterized other perflurocompounds (or even incompletely fluorinated compounds) may be chemically toxic to cells.[310],[311],[292]

 Some Perfluorchemicals Used in Liquid Ventilation Research

 Table 3: Physical properties of some PFCs that have been used for liquid assisted ventilation: 3-M Fluorinert Liquids ™, Rimar-101, perflurodecalin and perflurooctylbromide (PFOB, Perflubron™). Perflubron™ is the first completely defined PFC intended for medical applications. Sources: [312],[292] ,[313].

Figure 3-2: Chemical structure of perfluoro-tri-n-butylamine (FC-43).

For these reasons a completely chemically defined molecule, perflurooctylbromide F₃(CF₂)₇Br, Perflubron,™ LiquiVent™), was developed by Alliance Pharmaceuticals of San Diego, CA  for use in clinical trials of liquid ventilation (Figure 25, below). Unfortunately, Perflubron™, (Figure 3-3) with a molecular weight (MW) of 498.97, has a freezing point of +6.0ºC which makes it unsuitable for use in inducing ultraprofound hypothermia (0-5^oC) where the temperature of the ventilating liquid must be in the range of 2ºC to 4ºC for optimum efficacy.

Toxicology

Figure 3-3: Perflurooctylbromide (LiquiVent™)

The high stability, chemical inertness, and nearly total insolubility in both water and lipids of PFCs used in liquid assisted ventilation preclude their metabolism and limit their interaction with biomolecules. Despite 40-plus years of use in biomedicine little published work has been done on the toxicology of these compounds. Based on data from the available literature their toxicity can be divided into two categories: biophysical/biomechanical and immunological. When administered intravenously or intraperitoneally as neat (pure) chemicals injury results from the biophysical interaction with the animal rather than from biochemical interactions. Because PFCs are not miscible in water they form vascular emboli in the same way that injecting intravascular oil or air would.

PFCs with higher vapor pressures can form vascular gas emboli even if emulsified [314] and can lethally distend closed body viscuses such as the peritoneum, or cause perfluorocarbon vapor pneumothoraces (‘perflurothorax’).

PFCs of intermediate vapor pressure may accumulate in the lungs and be converted to vapor which is retained for weeks or months in the alveoli or lung parenchyma resulting in what Clark, et al., termed hyperinflated non-collapsible lungs (HNCL). [309] This phenomenon is noted at necropsy after IV administration of emulsified perflurodecalin containing blood substitutes [315] and after LAPC with intermediate vapor pressure PFCs such as FC-75 or FC-77.[272]  Schutt, et al., call this ‘pulmonary alveolar gas trapping’ and they propose that the phenomenon occurs as a result of PFC liquid or vapor migration through the pulmonary surfactant-liquid bridges where it forms stable, long lasting, PFC vapor micro-bubbles. They propose that these intra-alveolar micro-bubbles are part of the ‘normal pulmonary elimination of perfluorocarbon vapor’ from the body.[316]  While HNCL or ‘pulmonary alveolar gas trapping’ may not be clinically evident, and does not perturb blood gases or interfere with gas exchange, it does interfere with normal respiratory mechanics and can cause ‘stiff lungs’ in dogs following LAPC using FC-75 or FC-77.[161],[272] Stiff lungs increase the work of breathing until the vapor dissipates enough to relieve the acute tension in the alveoli.[272]  As such, the author believes this phenomenon should properly be classified as an adverse biophysical effect, rather than an acceptable or normal mechanism of PFC elimination. It is interesting to note that in a chemical model of lung injury using inhaled kerosene even brief PLV with FC-77 increased mortality and resulted in extensive gas trapping.[317]

Depending upon the emulsion size intravascular PFCs may be phagocytized by PMNL’s and macrophages and be deposited in the reticuloendothelial system where they may cause enzyme induction or mild inflammation from their space occupying, mechanical effects distorting normal tissue architecture. [318]  These changes are typically reversible as the PFC is eliminated via the lungs and the hepatomegaly and splenomegaly dissipate (~ 3-weeks).

The immunological effects of PFCs vary with the molecule, method of preparation and particle size (if administered intravascularly). Some PFCs appear to be directly cytotoxic to PMNLs and macrophages.[319] However, as a class, the PFCs seem to interfere with PMNL and macrophage chemotaxis, activation and de-granulation without inducing apoptosis or necrosis, by mechanisms that are not understood.[320], [321],[322]  Augustin, et al., have observed that PFCs alter the cytoskeleton of hepatic macrophages in a dose dependent manner that varies with the compound. [323]  Inhibition of PMNL and macrophage chemotaxis and respiratory bursts gives the PFCs moderately potent anti-inflammatory effects and by the same token makes them immunosuppressive.  While this effect is immunomodulatory and probably beneficial in ARDS [324],[325],[326], it also has the potential to impair pulmonary and systemic immune surveillance and presumably increase the risk of infection and neoplasm. FC-43 has been used to delay neutrophil mediated xenograft rejection [327],[328] and Perflubron™ has been demonstrated to inhibit neutrophil activation in the rat heart after 2-hours of cold ischemia and whole blood reperfusion.[329]

A recently discovered novel and unexpected effect of at least one PFC, Perflubron™ [326], is direct inhibition of oxidative damage in both cultured pulmonary artery endothelial cells exposed to hydrogen peroxide and in linoleic acid micelles subjected to varying concentrations of the azo initiator 2,2’-diazo-bis-(2-amidinopropane) dihydrochloride . This result is unexpected because it has previously been presumed that the antioxidant activity of PFCs was secondary to their immunomodulating and immunosuppressive effects. The protective effect of Perflubron™ in a non-biological system raises many questions about its basic pharmacology, and possibly about the chemistry and environmental interactions of the PFCs as a class, should this effect prove replicable with similar compounds.

Environmental Impact and Future Availability

The PFCs may be justifiably described as the penultimate atmospheric (greenhouse) poison. The PFCs, like water vapor and methane, both absorb and emit long wave (infrared) radiation; effectively trapping heat from the sun and warming the terrestrial surface and atmosphere.

Unlike CO₂ and methane, PFCs are not subject to biological cycling, are unaffected by electrochemical reactions, and do not dissociate in aqueous media. They are essentially already fully oxidized and are unaffected by standard oxidizing agents such as permanganates, chromates, and the like. As previously noted, degradation via oxidation occurs only at very high temperatures. Because of their inertness, they are similarly resistant to degradation by reduction, except under extreme conditions, requiring reducing agents such as metallic sodium. This leaves photochemical decomposition, primarily via hydroxyl radical (^.OH) mediated degradation, as the only means of terrestrial disposition. Both Cicerone [330] and Yi Tang [331] have shown that the reaction of  ^.OH  with the C3 and CF4 moieties is negligible under ground-state conditions, and that the lifespan of the molecules, once they enter the atmosphere, is likely in excess of 10,000 years. The heavier, higher MW and lower vapor pressure PFCs which are ideal for liquid assisted ventilation can be expected to remain in the lower reaches of the troposphere indefinitely, and thus not reach the upper atmosphere where, however slowly, they might be photo-degraded. In any event, the PFCs are so resistant to photo-degradation that the Flourinerts, and related compounds that are used as chemically stable cooling agents in photochemical reactors, as carrier solvents for photo-decomposition of other organic molecules, and are likewise classed as ‘radiation durable compounds’ for use in the photolithography industry.[332]

At present, the PFCs constitute an insignificant contribution to greenhouse gas emissions. However, widespread medical use could change this, and in any event, 3M, DuPont and other manufactures of industrial quantities of PFCs are aggressively encouraging the use of alternative compounds which they manufacture, principally the hydrofluoroethers [333] and the perfluorinated alkyl vinyl ethers. In 1982 Riess and Le Blanc estimated that if PFC-based blood substitutes came into wide use the quantities required would be in the ‘multi-thousand-tons-per-year range’ [292] all of which would end up in the atmosphere. Widespread use of PFC-facilitated LAPC and PLV could easily require a similar amount of product.

Extensive medical use of PFCs would seem to mandate associated efforts at recovery and recycling to minimize environmental contamination. However, this is not easy to do even in hospital under controlled conditions. Recovery of PFCs used emergently for LAPC (i.e., in-field induction of hypothermia in stroke, myocardial infarction, cardiac arrest) and as the O₂ carrying molecules in blood substitutes would seem to preclude effective recovery. The indefinite lifespan of these compounds makes their use akin to radiation exposure wherein the effect is cumulatively damaging, and ultimately lethal to the biosphere (as it exists now) as a consequence of their greenhouse effects and indefinite atmospheric lifespan.

The PFCs used in liquid assisted ventilation do not seem likely to accumulate or concentrate in biota. However, they do have comparatively long dwell times in patients when used clinically, and the exposure of health care workers to these volatile compounds would seem unavoidable. In light of these facts and the recent discovery that PFCs have direct radical quenching effects (with possible important environmental ramifications), as well as immunosuppressive properties, it seems reasonable to question the future large scale production, and thus the biomedical availability of these molecules.

Section 4:

History of Liquid

Assisted Ventilation and Implications for LAPC

History of Liquid Ventilation

Figure 4-1: An ultra-deep sea diver breathing PFC liquid in the 1989 motion picture ‘The Abyss’ Directed by James Cameron. [Photo courtesy 20^th Century Fox and Lightstorm Entertainment.]

As was the case with the first great rationalization of surgery and wound management by Pare’ [352], the creation of scientific nursing by Nightingale [353], and the development of fluid resuscitation and the first effective medical management of shock by Cannon [354] and Blalock [355], the impetus for the development of liquid ventilation was also initially warfare. Interest in the use of liquid as a breathing medium in mammals originated in the early 1960s in response to the U.S. Navy’s need to develop rescue systems for submariners that would allow them to transiently breathe saline or some other aqueous liquid. Mortality among submariners in World War I (WWI) and WW II was greater than in any other branch of military service. In WWII 22% of U.S. and 75% German submariners were killed in action.[356] With the advent of nuclear submarines in 1951, and the use of submarines to carry and deliver nuclear weapons, prolonged and complex missions while continuously submerged created the need (still largely unmet) to carry out rescue of submariners, and recovery of nuclear weapons and other strategically critical materiel, from extreme depths.

Johannes Arnold Klystra, M.D. is the Dutch pulmonologist and clinical researcher responsible for developing saline lavage of the lung as a treatment for advanced cystic fibrosis in 1958.[357]  Klysta’s interests extended well beyond clinical innovation in the management of lung diseases, and in the late 1950s this maverick physician approached the Dutch Navy to explore possible ways to allow deep ocean recovery of submariners as well as the development of liquid breathing systems that would allow divers to be free from the constraints imposed by gas breathing under conditions of high pressures (Figure 4-1):

“Man has tried for centuries to invade the oceans, perhaps driven by a subconscious nostalgia for atavistic weightlessness in the vast hydrosphere that covers more than 70 per cent of the earth, but gas in his lungs, compressed by a layer of water above, confines his activities to the shallow. Nitrogen, for instance, produces a progressively severe intoxication at depths greater than100 feet and usually incapacitates a diver by ‘rapture of the deep’ at no more than300 feet. Moreover, relatively large amounts of carrier gas dissolve in blood and tissues to be released as bubbles whenever the diver returns to the surface too rapidly. These hazards are all due to the compressibility of gases. The properties of water, on the other hand, hardly change at all with pressure, and I have observed mice with fluid filled airspaces move around in no apparent distress at a simulated depth of 3000 feet. If man were able to breathe oxygenated water instead of an oxygenated carrier gas, exploration of the oceans would no longer be limited by gas toxicity and decompression sickness.”[358]

The first mammal to survive liquid breathing was a mongrel dog named ‘Snibby.’  [158] Snibby was shaved, bathed, anesthetized, intubated, and submerged in a tub of buffered salt solution in a large hyperbaric chamber at 5 atmospheres of pressure while O₂ was bubbled through the saline bath. The dog breathed the liquid, which was held at a temperature of 32ºC, for 24 minutes. As Klystra noted in his published account of the experiment: “Snibby’s recovery was uneventful and he was adopted by the officers and crew of H. M. Cerberus to serve as a mascot aboard this submarine rescue vessel of the Royal Netherlands Navy.”[359]

In 1962 Klystra documented survival of mice breathing a balanced, buffered salt solution under 8 atmospheres of pressure at 20ºC for 18-hours.[158] Throughout the 1960s Klystra and his associates probed the limits of liquid breathing using aqueous solutions and they were the first to document the problem of profound hypercarbia as a fundamental limitation in tidal liquid breathing.[157]  Klystra was also the first to demonstrate survival of mammals following extreme hyperbaria using spontaneous liquid breathing of a buffered salt solution.[357]  A further testimony to the highly creative and innovative nature of Klystra’s work was his use of the selectively liquid lavaged lung lobe as a possible replacement for the kidney; in other words, as a mass exchanger for nitrogenous wastes and as an osmotically driven ultrafilter for removal of excess water in the setting of renal failure.[360]

One of the most remarkable things about this pioneering work is that saline and other aqueous solutions denude the alveoli of surfactant, the stiff molecular cage that supports the 3-dimensional alveolar structure and keeps the acinar airways open to ventilation. Removal of surfactant is a primary cause of serious pulmonary injury and a major pathophysiological mechanism in both acute lung injury (ALI) and the acute respiratory distress syndrome (ARDS). Through the present, saline lavage of the lungs remains a standard model for inducing pulmonary injury to simulate ALI and ARDS in experimental animals.[361],[362] The inadequate gas carrying capacity of aqueous solutions under normobaric conditions, and the inherently injurious nature of water-based ventilating media made clinical or undersea application of liquid breathing infeasible. Indeed, breathing of balanced salt solutions, even under hyperbaric conditions in the presence of 100% O₂ was only possible for extended periods of time if the animals were hypothermic. While O₂ delivery was adequate, CO₂ elimination was not, and hypercarbia and respiratory acidosis were lethal complications of liquid breathing under normothermic conditions.[363]

Figure 4-2: Chemical structure of polydimethyl(siloxane). (Image from Wikimedia Commons: http://en.wikipedia.org/wiki/Image:Silicone-3D-vdW.png.)

Shortly after the publication of Klystra’s pioneering work Leland C. Clark began looking for more suitable liquid breathing media. Cark’s initial efforts focused on using polyunsaturated vegetable oils such as corn and safflower oil. These oils proved injurious to lungs and Clark next evaluated the organosilanes octamethyltrisiloxane, dodecamethylpentasiloxane, decamethyltetrasiloxane, polydimethylcyclosiloxane, and polydimethylsiloxane. These compounds, commonly referred to as ‘silicone oils,’ were manufactured by the Dow Chemical Co. of Midland, MI in various chain lengths, and thus vapor pressures and physiochemical properties.[364]  The organosilanes are made from a Si-O backbone to which a variety of organic groups are attached to the silicon atoms via a Si-C bond. Polydimethylsiloxane is the most common of the commercially produced organosilanes and is a polymer with a backbone consisting of a repetition of the (CH₃)₂SiO unit (see Figure 4-2, above). The organosilanes proved less toxic than vegetable oils, and better able to dissolve O₂ and CO₂, but were still too toxic to be used for liquid ventilation.

The problem of an inert and non-injurious breathing medium capable of carrying enough dissolved O₂ to support life under normobaric conditions was finally solved by Leland C. Clark and Frank Gollin in 1966 with their report that a variety of fluorocarbons performed well as liquid breathing media without apparent injury to the lungs and with long-term survival of the animals.[163]

Figure 4-3: Dr.  Leland C. Clark, Jr., 1918–2005 (Photo courtesy of Richard Bindstadt – Blackstar)

The PFC identified by Clark and Gollin, a ~50/50 mixture of isomers of F-alkylfurans (FC75), was evaluated under conditions of spontaneous respiration with the animals submerged in the liquid.[309]  As proved the case with aqueous solutions, the PFCs delivered adequate amounts of O₂, but failed to allow for effective clearance of CO₂ resulting in lethal hypercarbia and acidosis. The vastly greater density and viscosity of PFCs relative to air or other gases limited the diffusion of CO₂ into the liquid. An added problem contributing to the hypercarbia observed in liquid ventilation was the vastly greater work of breathing (WOB) imposed by a liquid 1,000 times as dense as air and the increased time for exhalation in the absence of active (negative pressure) pumping of the PFC from the lungs.

Figure 4-4: Archetypical tidal liquid ventilation (TLV) system. PFC completely replaces gas in the lungs and is moved in and out of the lungs using mechanical (usually roller) pumps. After exhalation the PFC is passed through a filter, and then through an oxygenator-heat exchanger to be scrubbed of CO₂, oxygenated and warmed to body temperature before being re-infused into the lungs.

In 1970 Gordon D. Moskowitz and Thomas H. Shaffer (Figure 4-5) began work on a mechanical liquid ventilator (Figure 4-4) to overcome the problem of the increased inspiratory workload accompanying liquid breathing.[365]  During the next 5-years these investigators developed progressively more sophisticated demand-regulated liquid ventilators which also began to address the need for assisted exhalation. [366],[367],[368] During the 1980s development of a variety of tidal liquid ventilators was undertaken by Shaffer, et al., and applied to animals ranging from cats [300] to preterm and neonatal lambs [369], culminating in the first clinical application of tidal liquid ventilation (TLV) to human neonates in 1989.[370]

Figure 4-5:  Dr. Thomas H. Shaffer, MSE, PhD. (Photo Courtesy of Dr. Thomas Shaffer.)

Partial Liquid Ventilation (PLV)

From the beginning of liquid ventilation research with the work of Klystra in 1960, and continuing until the publication of the work of Furhman, et al. in 1991, only one kind of liquid ventilation existed; tidal or total liquid ventilation (TLV). As early as 1976 Shaffer, et al., had noticed that peak airway pressures were dramatically improved in pre-term lambs after they were returned to gas ventilation following 20 minutes of TLV.[368] The observation that lung mechanics and gas exchange remained transiently improved following TLV was extended by Shaffer, et al., in 1983 with the observations that pulmonary compliance and paO₂ were increased and paCO₂ was decreased during conventional gas ventilation following TLV to values lower than those that could be achieved during TLV.[300]

Bradley Furhman, an anesthesiologist and critical care physician at the University of Pittsburgh Medical School, noticed the enduring salutary effects of PLV following reinstitution of conventional gas ventilation in a model of infant respiratory distress syndrome (IRDS) using pre-term lambs, and he further noted that Shaffer, et al., had reported that these improvements in lung function did not occur following TLV in the healthy lung.[371]  Furhman hypothesized that these salutary effects of TLV might be due to possible surfactant-like and alveolar recruitment effects of the comparatively large amounts of PFC retained in the lungs following TLV, since it was well documented that even with aggressive efforts to remove PFC following TLV, a volume approximately equal to the animal’s functional residual capacity (FRC) of 30 ml/kg remained in the lungs until it was eventually eliminated by evaporation.[370]

Figure 4-6: When filled to Functional Residual Capacity (FRC) the PFC forms a meniscus in the endotracheal tube. As shown at left, FRC constitutes all the volume in the lungs and trachea at the end of exhalation. (Modified by the author from the original art at Wikimedia Commons: http://en.wikipedia.org/wiki/Image:3DScience_respiratory_labeled.jpg.)

Furhman, et al., tested this hypothesis by administering FC-77 (a perfluorinated butyl-tetrahydrofuran isomer mixture) via the endotracheal tube in a volume equal to the FRC of neonatal swine.[372]  As opposed to using TLV, conventional positive PPV was continued using the same parameters employed before the PFC was instilled into the lungs (Figure 4-6). This technique, christened partial liquid ventilation (PLV), proved as effective, or more effective, than TLV in decreasing airway resistance, as well as peak and mean airway pressures. PLV also seemingly abolished the need for PEEP in healthy lungs, while providing adequate gas exchange. This study not only established the effectiveness of PLV, it also posited (correctly) the mechanisms by which PLV was achieving restoration of gas exchange, increasing pulmonary compliance, and homogenizing ventilation in the lungs thus greatly reducing volutrauma (Figure 4-7, below).

Figure 4-7: Lungs from two rabbits subjected to a model of ischemia-reperfusion injury. The lung on the left (A) shows severe volutrauma to the upper lobe (‘baby lung’) with obvious consolidation in the ventral, dependent areas of the lung. The lung on the right (B) is from an animal treated with PLV and shows no evidence of injury and is homogenously ventilated with PFC and gas. Note that the letter A has been placed on an apical ‘baby lung.’

The investigators noted that due to its higher density than water the PFC rapidly flowed to the most dependent areas of the lungs and in so doing it filled collapsed alveoli and displaced serous transudate in flooded alveoli (Figure 4-9, below). With each gas breath, liquid from the large and medium caliber airways was admixed with ventilating O₂ under conditions of turbulent flow, thus oxygenating it and washing it of CO₂. During exhalation some, but not all of the PFC in the alveoli, flowed out into the larger airways where it was also admixed with both ventilating gas and already oxygenated PFC. During the next inspiration the alveoli were refilled with PFC that was oxygenated and cleansed of CO₂. Because PFC is retained in the lungs to FRC, and gas is present in the alveoli only as micro-bubbles, if at all, the alveoli never completely empty and thus are vastly more compliant to re-inflation.  As Furhman, et al., noted, the alveoli of the lungs appear to be ventilated almost exclusively with PFC throughout the ventilatory cycle with direct gas admixing occurring mostly in the larger airways (trachea, bronchi and bronchioles). Gas exchange between the blood and PFC most likely occurs due to direct PFC-alveolar membrane contact.

Figure 4-8: Lung volumes.

In addition to recruiting alveoli to liquid-mediated gas exchange, PLV also displaces alveolar transudate, mucus, and cellular debris by continuously lavaging the airways; macroscopic to microscopic.[373]  PLV is cytoprotective of Type II alveolar epithelial cells, decreases PMNL adhesion in pulmonary capillaries [326], reduces alveolar hemorrhage, and generally preserves alveolar ultrastructure in the setting of ALI [324] and ARDS.[326]

Also of great importance is that PLV is simple to implement; it does not require novel, complex tidal liquid ventilators with an oxygenator, heat exchanger, water trap and filters – all under complex computer control. Because there is no bulk movement of PFC over long distances of airways, the resistance to PFC flow is greatly reduced, effectively eliminating the constraint of only 5 to 7 breaths per minute in TLV.[374] Because the transit times and distances between the alveoli and the bronchioles are very small in PLV (where ventilation gas admixing and gas exchange is occurring) and because CO₂ is probably exchanged in micro-bubbles in the PFC which are far smaller than would be the case in a ‘sphere’ of PFC the diameter of the alveolus (~250µ), the problem of hypercarbia is also eliminated.[375]

Figure 4-9: A: Alveoli are flooded alveoli in ARDS or pulmonary edema. When PFC is, literally, poured down the endotracheal tube it flows under gravity (due to its ~1.8x density of water) to the most dependent areas of the lungs. B: In so doing it opens the collapsed alveoli and displaces edema fluid from them. (Modified from original art by Patrick J. Lynch, medical illustrator, and is from Wikimedia Commons.)

From 1991 to 2000, PLV was extensively investigated in a number of different animals employing a variety of models of lung injury; saline FTLV, oleic acid injury, smoke inhalation, prematurity, intestinal ischemia, lung transplantation injury, and pneumococcal pneumonia.[376], [377], [378],[379],[380],[381] These studies, with no notable exceptions, showed marked benefit for PLV in ALI and ARDS.

In 1993 Leach, et al., [164] carried out the first clinical trial of PLV in IRDS (Figure 4-10). This was followed by a number of clinical trials for ARDS in both children [382] and adults.[383]  These trials were sponsored by Alliance Pharmaceuticals, Inc. of San Diego, CA (Alliance) in an effort to obtain FDA approval for the use of Perflubron™ as the first gas exchange PFC in PLV.

In 2006, after 5 years of delay, the ‘definitive,’ Phase III, prospective, randomized clinical trial (RCT) of PLV in ARDS was published (LiquiVent™ study).[384]  For reasons that are only now being understood, this trial showed PLV (using Perflubron™ at a dose equal to FRC (30 ml/kg), and at a lower dose of 10 ml/kg, to yield a worse outcome than conventional PPV with increased overall mortality and increased days requiring mechanical ventilation: The 28-day mortality in the control group was 15%, versus 26.3% in the low-dose (p=0.06) and 19.1% in the high-dose (p = 0.39) PLV groups. There were more ventilator-free days in the control group (13.0 ± 9.3) compared with both the low-dose (7.4 ± 8.5; p=0.001) and high-dose (9.9 ± 9.1; p =0.043) groups. Most remarkably, there was a high incidence of barotrauma: 34% pneumothoraces in the phase II LiquiVent™ trial (20% in the control group) and 29% and 28% in the phase III LiquiVent™ trial (control, 9%). Pneumothoraces requiring the placement of chest tubes is an ominous complication of PPV in ALI and ARDS and is associated with increased mortality, duration of ventilator time and length of stay in the ICU. In view of the consistently diverse, positive and well conducted animal studies demonstrating unequivocal benefit, this result was surprising.

Figure 4-10: Initiation of PLV in a neonate with IRDS. The only novel piece of equipment used was a luer-loc one-valve interposed between the endotracheal tube and the 16 mm connector to the ventilator to facilitate intra-tracheal administration of Perflubron™ without having to disconnect the patient from the breathing circuit. (Photo courtesy of Alliance Pharmaceutical, Inc.)

The reasons for the failure of the Phase III LiquiVent ™ RCT are both complicated and subtle – and are still being debated today. [385] The likely reasons for the failure of the Phase III study have important implications, not only for the future of PLV in ALI and ARDS, but also for the optimum use of PLV and LAPC in emergency and critical care medicine. The reasons for PLV’s failure are rooted in difficulties and errors that have plagued translational research from animals to humans in many areas of medical research.[386]  What follows is a point-by-point evaluation of the possible causes of failure of the Phase III PLV trial as well as an analysis of the implications of these problems for LAPC.

 Unanticipated Effect of Lung Protective Ventilation Strategies

The Phase III trial was designed in 1997, begun in 1998, and completed in 2002. This was well before the first report of the efficacy of lung-protective ventilation in reducing mortality in ARDS and ALI was reported in 2000 [387] and 7 years before the first influential ARDSnet study was published.[388]  The criticality of minimizing barotrauma and volutrauma, even over maintaining gas exchange at optimum physiological levels, was thus not taken into consideration in the LiquiVent™ study design. This was especially significant because, due to subtle shortcomings in the design of most of the animal studies, it was not understood that PLV is a source of barotrauma and volutrauma; even when far less than full FRC-dosing is used under circumstances most like those encountered clinically (see ‘Failure to Establish a Dose-Response Curve,’ below).

While the LiquiVent™ study experimental group had a worse outcome than the control group, it should be noted that the absolute results were actually no better (or worse) than those reported in the previously cited ARDSnet studies validating lung protective ventilation (i.e., 15 to 26%).  This is especially worth noting because all 3 groups of the LiquiVent™ study patients were considerably sicker than the patients in the ARDSnet studies. The objective entry criteria for the LiquiVent™ study were an initial PaO₂/FiO₂ <200mmHg followed by a failed (PaO₂/FiO2 < 300 mmHg) response to a PEEP of ≥ 13 cm H₂O at a FiO₂ ≥ 0.5. By contrast, the ARDSnet patients only needed to have a PaO₂/FiO2 of < 300 mmHg with no requirements for PEEP or FiO₂ upon randomization. One possible reason for the superior outcome in the LiquiVent™ control group, compared to that seen in most other studies of ARDS at that time, is that Alliance selected only the very best centers of excellence in the management of ALI and ARDS to conduct the trials. By contrast, ARDSnet studies were also conducted on a contract basis by NIH at institutions that were more representative of the actual quality of care available at large metropolitan hospitals in the U.S. Alliance thus placed LiquiVent™, and consequently the entire field of PLV, on trial in a setting with the sickest patients receiving the best medical management for ARDS

Defective Translational Research Models

The animal models of ALI and ARDS used to evaluate PLV did not model the real-world course of lung injury in clinical illness. Researchers typically inflict an insult and then wait a uniform, and often unrealistically short time, for the injury to develop. By the time human patients in respiratory distress enter the ICU they have usually been ill for many hours, or even days, and as a consequence the degree of pulmonary compromise, and in particular, pulmonary edema, may be greater. Furthermore, animal models of lung injury are inherently more homogenous than is usually the case in human ARDS. Heterogeneity of injury is one of the hallmarks of both ALI and ARDS in humans, and heterogeneity means that normal or minimally injured areas of lung will be subjected to the same conditions as more severely injured areas (see discussion of varying requirements for PEEP depending upon the degree of injury individual alveoli under the heading, ‘The PFC Air Interface and Shear Effects in the Small Airways,’ below).

These observations have other important implications because loading to the theoretical FRC (30 ml/kg) in ill and often aged humans may not be possible, in the sense that the ‘normal’ or predicted volume of lung to be recruited may not be available. Alveoli can be filled with PFC only if there is enough room in the thorax to allow them to fill. If fluid in severely edematous lung parenchyma stubbornly resists relocation to the vascular compartment due to hypoalbuminemia and an interstitial pressure in excess of the hydrostatic force generated by the PFC, or if for other mechanical reasons there is not sufficient volume to accommodate a FRC-dose of PFC, then the result will be volutrauma and barotrauma to the less dependent lung. This possibility was noted by Cox, et al., as early as 1997 and was later raised in a study done by Lim et al., in 2000.[389]

Failure to Establish a  Dose-Response Curve

While as previously noted, a wide range of animal models, and models of injury were investigated with respect to PLV, there was little study to determine the optimum dose-response curve of either LiquiVent™ or other PFCs used in PLV. In hindsight this seems strange because in the experimental evaluation of any novel drug the first step is usually to establish a dose-response curve and thus to bound the ‘safe and effective dose.’ This was not done in PLV and those studies which documented injury from PLV in normal lungs at FRC dosing were arguably not given the attention they deserved.[390],[391]

Recently, the work of Dreyfuss and Ricard [392] has demonstrated that dosing to FRC in healthy rats actually causes alveolar capillary leak and induces lung injury. In a series of elegant studies they have examined the complex relationship between PFC dose, PEEP and Pplat. Their studies indicate that the probable ideal dose of PFC (or of Perflubron™ in this case) is ~ 3 ml/kg; 10% of FRC, and still only 30% of the 10 ml/kg dose used in the ‘low dose’ LiquiVent™ study.[393] Furthermore, these investigators have documented that FRC dosing with Perflubron™ causes gas trapping in the lungs, and that under these circumstances, paradoxically, PEEP is protective.[391]

 Gas Trapping and Selection of the Appropriate PFC

Perflubron’s™ comparatively low vapor pressure is similar to that of perflurodecalin and thus probably results in a lower spreading coefficient relative to most other PFCs used in the animal studies. This would likely result in slower dynamic flow during inspiration and in ‘plugging’ of medium caliber airways (with the previously noted effect of gas-trapping) due to high gas-fluid interfacial tension.[394],[395] These effects may contribute to barotrauma and volutrauma when Perflubron™ is administered to full FRC.[396],[397]

It now appears that if PLV is to have any chance at conventional clinical application in the West, clinical trials will have to start from scratch, quite possibly with a different PFC, or blend of PFCs, being used to achieve the pharmaco-physical properties required for the particular application – including possibly tailoring the medium to the size of the patient’s intermediate sized airways (which  are greatly different between neonates, children and adults) and to the particular application at hand. For instance, as Jeng, et al., (from Schaffer’s group) states:

“In terms of clinical application, the appropriate fluid for liquid assisted ventilation will depend on the clinical situation. For example, a more viscous fluid may be more appropriate for supporting the lung during extracorporeal membrane oxygenation during which the PFC application is aimed at preventing atelectasis and fluid flux across lung-at-rest conditions. In contrast, a less viscous fluid may be preferred for TLV during which tidal volumes of fluid are exchanged. For PLV, a fluid with low vapor pressure would reduce dosing requirements during the course of the treatment. Thus, the data presented herein further relate fluid physical properties with liquid ventilation applications.”[304]

The PFC Air Interface and Shear Effects in Small Airways

Figure 4-11: The shear- inducing effect of a single flooded alveolus on neighboring alveoli. Open alveoli (A) expand evenly in unison and experience no shear. After alveolar flooding and collapse (B) shear forces occur on the adjoining alveolar septa. PFC filled; partially filled and unfilled alveoli and other acinar structures will be subject to the same damaging shear forces as is the case when aqueous media are present in the acini.

Although the alveolar air-liquid interface is eliminated during PLV, giving PLV its PEEP and surfactant-like properties, a PFC-gas interface is created. The law of Laplace ( P = 2γ/r) describes the relationship between the pressure to stabilize an alveolus (P) and surface tension at the gas-PFC interface of an alveolus (γ) in relation to the radius of the alveolus r. [398]  In the normal lung, surfactant present at the gas-liquid interface lowers the air-liquid surface tension in lockstep with decreasing alveolar radius (to nearly 0 mN · m^-1 for low alveolar radii), thus keeping the ratio of γ/r of the alveolus constant and guaranteeing alveolar end expiratory stability at low pressures.[399],[400] By contrast, PFCs exhibit a constant gas-PFC surface tension for any alveolar radius. This means that at the end of exhalation, the alveolar radius will be quite small, while the surface tension remains unchanged, and therefore very high compared to when the alveolus is inflated. Thus, the alveoli will collapse unless they are supported by PEEP from the ventilating gas.[401]  The implication is that PFC-derived ‘liquid PEEP’ must always be balanced by precisely the right amount of ‘gas PEEP’ to prevent end-expiratory collapse of non-PFC-filled alveoli (Figure 4-11). This presents a formidable challenge in both the laboratory and the clinic.

Partial Liquid Ventilation and the Law of Laplace

Figure 4-12: In attempting to determine the extent to which PFC is likely to cause alveolar injury due to gas-liquid mediated shear forces, and variable distention of alveoli filled with PFC as opposed to gas, it is instructive to compare the dynamic behavior of PFC (red line) with surfactant (green) as well as with other liquids, such as pulmonary edema transudate (yellow) (which contains dissolved surfactant), and saline (blue line). While PFC exerts far less surface tension than saline, it still exerts a force at the air-PFC interface of ~20 dynes cm^-1, and, like saline, lacks the dynamic responsiveness to changes in surface area which are the unique property of surfactant and surfactant containing solutions.

In addition to the problem of end-tidal alveolar collapse, high shear forces at the alveolar membrane as a result of insufficient PEEP during PLV may cause alveolar rupture and consequently gas/PFC leak and the development of pneumo- and/or perflurothorces.[402]  The complexity and difficulty of this problem becomes clearer when consideration is given to the fact that the amount of ventilator-applied ‘gas PEEP’ will be a function of the pathological condition of each alveolus (which will vary widely in the same lung, let alone from patient to patient). This is so because the presence of a thin film of PFC in the alveolus will only be beneficial in alveoli where the native surfactant has failed to maintain alveolar patency (diameter).

In those compromised alveoli that are unable to reduce surface tension to the gas-PFC interfacial tension, the level of ‘gas PEEP’ required to keep them open at the end of exhalation will be lower during PLV than the level of ‘gas PEEP’ during conventional mechanical ventilation. Conversely, in alveoli with functioning surfactant, (i.e., able to reduce their surface tension to below that of the gas-PFC interface) there will be an interaction between PFC and the normal alveolar membrane fluid, leading to levels of ‘gas PEEP’ with PLV higher than those required to keep the alveoli open with ‘gas PEEP’ required in conventional ventilation alone (Figure 4-12). In the clinical milieu this implies careful titration of PFC dose during initiation and maintenance of PLV and equally careful titration of PFC ‘removal’ (i.e. evaporation) or weaning from PFC during the transition from PLV to gas-only PPV.[401]

Even assuming that the means can be developed to determine and control these parameters, the seemingly intractable problem of inhomogeneous gas distribution during tidal gas ventilation in PLV remains.  Gas will go preferentially to the least PFC liquid loaded and least dependent airways and this will likely produce over-distension and volutrauma much as happens in the edematous consolidated lung.  At a minimum it will be necessary combine fluid PEEP with pressure-controlled ventilation in a way such that the pressure in any alveolus does not exceed the pressure of the ventilating gas.[381]  Failure to prevent shear or alveolar hyperinflation will result not only in direct mechanical injury to the alveolar membrane and pulmonary capillaries, [403] but also in both local and systemic injury from pro-inflammatory cytokines whose release by alveolar epithelial cells is triggered by even modest shear stress or over-distension.[404],[405]

An additional source of shear injury in PLV and thus presumably LAPC is only now beginning to emerge. Research on VLI has predominately focused on the role of high inflation pressures and large tidal volumes.[404],[406],[407] However, ventilation at low lung volumes and pressures results in a different type of lung injury, in which airway instability leads to repetitive collapse and reopening of the terminal airways.[408]  This type of injury is relevant to LAPC and PLV because the air-PFC interface behaves similarly to the cyclically re-inflated collapsed airway. During reopening of collapsed airways a finger of air moves through the airway generating stresses on airway walls and injuring the airway epithelium.[409],[410],[411],[412]  This does not occur during normal tidal ventilation because pulmonary surfactant stabilizes the airways and prevents their collapse during exhalation.

Figure 4-13: The power of surface tension is perhaps most easily illustrated by meniscus formation at tripartite air-cylinder-liquid interface. When water wets a small diameter tube the liquid surface inside the tube forms a concave meniscus, which is a virtually spherical surface having the same radius, r, as the inside of the tube. The tube experiences a downward force of magnitude 2πrdσ.The gas-liquid interface under the influence of the tidal forces of ventilation generates enormous shear stress on the respiratory epithelium of small caliber airways. The ‘pull’ exerted by PFC on a capillary wall is approximately 1/5^th that of saline, or ~ 0.4 πrdσ; more than enough to cause endothelial cell injury during tidal ventilation. A metal paperclip floating atop a glass of water illustrates the power of surface tension.

However, in ARDS, and where bulk liquid mixed with gas is present in the small airways (including PFC) surfactant cannot act to protect the airway epithelium against the stress field exerted on the walls of these airways as bubbles move to and fro through the liquid inside them. Bilek, et al., have investigated this phenomenon in vitro and have modelled it computationally as a semi-infinite bubble progressing through a compliantly collapsed airway, as well as a bubble progressing through a rigid tube occluded by fluid. [413]  In this process, the airway walls are separated in a peeling motion as the bubble traverses the airway. This effect has been indirectly documented by experimental observation in vivo. [414],[415] Airway reopening induces large and rapid changes in normal and shear stress along the airway walls. These spatial and temporal gradients of stress exert dynamic, large, and potentially damaging stresses on the airway epithelium that do not occur under one-phase steady-flow conditions.[411], [416],[415] These forces are shown schematically in Figure 4-14. The shear stresses induced by bubble progression along both collapsed and liquid filled airways cause direct trauma due to substrate stretch-induced injuries of pulmonary epithelial cells as a result of  stretching of the plasma membrane causing small tears.[417],[404] Additionally, mechanical stresses from the fluid flow may stretch the plasma membrane either directly, or as a consequence of cellular deformation.

For a low profile, predominately flat region of a cell, the non-uniformly distributed load may regionally deform the membrane. In addition, the normal-stress difference could induce transient internal flows within the cell that exert hydrodynamic stresses on the intracellular surface of the cell membrane, which might injure the membrane by the same mechanisms as extracellular stresses, and additionally be disruptive to the cytoskeleton. Bilek, et al., also describe the effects of irregular airway topology on forces generated at the air-liquid interface. They note that bulges of as little as 2 μ into the lumen of an airway result in greatly amplified shear stresses. Such bulges are commonplace in healthy airway epithelium as a result of the protrusion of epithelial cell nuclei into the airway lumen. The smoothness of the pulmonary epithelium is greatly compromised in ARDS and pulmonary edema and this may be expected to exacerbate topologically mediated shear injury to the small airways. It is interesting to note that Bilek, et al., found that the addition of surfactant to their model systems abolished shear injury in both reopening and bubble- traversing, saline occluded models; perhaps as a result of moderating liquid film thickness over the surface of the airway epithelium thereby decreasing the flow resistance and stress amplification. To what extent this may be applicable in the setting of PLV or LAPC using PFCs is unclear, since presumably surfactant is only effective by being dissolved in the bulk liquid present in the airways (i.e., in the Bilkek, et al., model, saline).

Figure 4-14: Hypothetical stresses imparted on the epithelial cells of an airway during reopening. A: a collapsed compliant airway is forced open by a finger of air moving from left to right. A dynamic wave of stresses is imparted on the airway tissues as the bubble progresses. Circles show the cycle of stresses that an airway epithelial cell might experience during reopening. The cell far downstream is nominally stressed. As the bubble approaches, the cell is pulled up and toward the bubble. As the bubble passes, the cell is pushed away from the bubble. After the bubble has passed, the cell is pushed outward.

B: A fluid ‘occlusion’ in a rigid narrow channel is cleared by the progression of a finger of air moving from left to right. A dynamic wave of stresses is imparted on the pulmonary epithelial cells lining the channel wall. The circles show the cycle of stresses that the cells might experience during reopening. Far downstream, the cell is pushed forward and slightly out. As the bubble approaches, a sudden rise in pressure and a peak in shear stress occurs, pushing the cell forward and outward with much greater force. After the bubble has passed, the cell is pushed outward. Pressure gradients generated in the presence of an air-PFC interface can also be expected to create normal stress imbalances on the cell membrane over the length of the cell. (Illustration and accompanying text reproduced from Bilek, AK, Dee, KC, Gaver, DP, III. Mechanisms of surface-tension-induced epithelial cell damage in a model of pulmonary airway reopening. J Appl Physiol 94:770-783; 2003.)

 These problems may only (if ever) be solved when applying PLV to ALI and ARDS by eliminating tidal gas ventilation and replacing it with high frequency oscillating ventilation (HFOV) which yields uniform and far lower mean airway and P_t pressures and abolishes peak pressures associated with inspiration [418]; although this modality did not prove superior to either PLV or HFOV alone in animal studies – albeit using high (FRC) doses of PFC.[419]

In the case of LAPC, the short duration of FTLV as compared to that required for the treatment of ARDS using PLV may not cause sufficient shear injury to be of clinical concern, particularly in the absence of extensive preexisting pulmonary pathology. In the event that shear injury does prove to be a problem, it may be possible to use HFOV in combination with a more or less continuous process of PFC introduction and removal at or near the level of the carina.  Although, the extent to which HFOV would be effective in rapidly exchanging liquid, as opposed to gas, between the large and small airways is unknown.

The Best as the Enemy of the Good

Finally, another possible factor in the discrepancy between the human and animal trials of PLV was the very close control over the availability of Perflubron™ to investigators exerted by Alliance. In part, this tightly controlled and highly selective distribution of Perflubron™ to only a small number of carefully vetted researchers was driven by the high intrinsic cost of the molecule, and by the cultural and regulatory milieu currently present in the West. Gone are the days when pharmaceutical companies freely distributed putative new drugs for studies more or less upon request. The staggering cost of regulatory approval for a new drug, coupled with often irresponsible ‘research’ aimed at inciting the media frenzy that occurs whenever the toxicity or carcinogenicity of any novel or synthetic molecule (the so-called ‘cyclamate-effect’) is newly demonstrated (regardless of the lack of soundness of the experimental design), has understandably made drug developers cautious about to whom they entrust the evaluation of potentially multibillion dollar compounds. An unfortunate effect of this abundance of caution is that novel drugs are often protected from robust evaluation under more widely varying conditions that more closely approximated those seen in the real world.

Implications for SCA and LAPC

To a much greater degree than was the case with the LiquiVent™ trials (and PLV studies in general), the study being used to initially determine the feasibility of  LAPC for SCA using a TLV-type approach conducted by the author and his colleagues [272] suffers from the same defects that plagued the Alliance PLV studies. This study was conducted on healthy dogs – not on animals that were undergoing CPR with the associated very high peak and mean airway pressures. As was the case in the LiquiVent™ studies, this work preceded the ARDSnet data demonstrating the importance of low tidal volume ventilation and lung protective strategies in general, including minimizing peak and plateau airway pressures. At the time this study was published the authors were, like the LiquiVent™ investigators, unaware of the adverse effects of loading with PFC to FRC – although we certainly observed volutrauma and barotrauma – and became very sensitive to the need for controlling peak and mean airway pressures, as well as to a nuanced shaping of the flow-pressure curve. Indeed, the problem of automating the ideal flow-pressure algorithm has reportedly remained elusive, and ventilation using the technique of LAPC we reported is still least traumatically performed by hand.[420]

Figure 4-15 (left):  The first human cryopatient, Eleanor Williams, undergoing LAPC on 02 March, 2002; several 1,500 ml FTLVs of PFC chilled to ~4ºC were administered by the author using gravity delivery from a flexible 2-liter peritoneal dialysis bag (blue arrow) and then suctioned out. This patient experienced massive hemoptysis immediately after the start of CPS and before LAPC was initiated (red arrow). The patient hemorrhaged ~1,500 ml of blood in less than 5 minutes: underscoring the catastrophic nature of pulmonary bleeds in the setting of friable lungs and CPR. (Photo courtesy of the Alcor Life Extension Foundation.)

The recent insights into the reasons for the failure of PLV Phase III clinical trials suggest that to the extent it is possible to reduce the volume of gas breaths and of the FTLVs required to facilitate heat exchange (i.e., delivery of PFC to and retrieval of PFC from the lungs on a cyclical basis) this will likely reduce the severity of lung injury resulting from LAPC. The measured intrapulmonary pressure in patients undergoing TLV in the Phase III LiquiVent™ study were ~60 cmH₂0 and this resulted in a high incidence of air leak. Intrathoracic pressure in CPR is typically in the same range; 45 to 55 mmHg or ~61 to 75 cmH₂0.[253] The combination of LAPC at FRC with CPR is unknown territory and should be approached with caution; and hopefully also approached with additional studies in animal models of extended duration CPR with long term follow up of the animals.

References

1.         American-Heart-Association, Heart Disease and Stroke Statistics – 2008 Update. . 2008, American Heart Association: Dallas, Texas.

2.         de Vreede-Swagemakers, J., et al., Out-of-hospital cardiac arrest in the 1990′s: a population-based study in the Maastricht area on incidence, characteristics and survival. . J Am Coll Cardiol, 1997. 30: p. 1500-5.

3.         American-Heart-Association-and-National-Research-Council, Standards for cardiopulmonary resuscitation (CPR) and emergency cardiac care (ECC). . JJ Amer Med Assoc, 1974. 227(suppl): p. 833-68.

4.         Sakai, A., Sudden deaths among male employees: a six-year epidemiological survey. J Cardiol, 1990. 20: p. 957-61.

5.         Safranek, D., Eisenberg, MS, Larsen, MP., The epidemiology of cardiac arrest in young adults. Ann Emerg Med, 1992. 21: p. 1102-6.

6.         Viskin, S., Belhassen, B., Idiopathic ventricular fibrillation. American Heart Journal, 1990. 120: p. 661 – 671.

7.         Skogvoll, E., et al., Out-of-hospital cardiopulmonary resuscitation: a population-based Norwegian study of incidence and survival. . Eur J Emerg Med, 1999. 6: p. 323-30.

8.         Weale, F., The efficacy of cardiac massage. Lancet, 1960. 1: p. 990-96.

9.         Eisenberg, M., Cardiac Arrest and Resuscitation: A tale of 29 cities. Ann of Emer Med, 1990. 19: p. 179-86.

10.       Kentsch, M., et al., Early prediction of prognosis in out-of-hospital cardiac arrest. Intensive Care Med, 1990. 16: p. 378-83.

11.       Troiano, P., et al., The effect of bystander CPR on neurologic outcome in survivors of prehospital cardiac arrests. Resuscitation, 1989. 17: p. 91-8.

12.       Bossaert, L., Van Hoeyweghen, R., Bystander cardiopulmonary resuscitation (CPR) in out-of-hospital cardiac arrest. The Cerebral Resuscitation Study Group. . Resuscitation, 1989. 17(Suppl): p. S55-69; discussion S199-206.

13.       Stueven, H., et al. , Bystander/first responder CPR: ten years experience in a paramedic system. Ann Emerg Med, 1986. 15: p. 707-10.

14.       Lombardi, G., Gallagher, J, Gennis, P., Outcome of out-of-hospital cardiac arrest in New York City. The Pre- Hospital Arrest Survival Evaluation (PHASE) Study [see comments]. JAMA, 1994. 271: p. 678-83.

15.       McCarthy, M., Looking after your neighbors Seattle-style. Lancet, 1998. 351: p. 732.

16.       Hayward, M., Cardiopulmonary resuscitation: are practitioners being realistic? Br J Nurs, 1999. 8: p. 810-4.

17.       Bengtsson, M., A psychiatric-psychosocial investigation of patients who had survived circulatory arrest. Acta Psychiat Scan, 1969. 45: p. 327.

18.       Roewer, N., Kloss, T, Puschel, K., Long-term result and quality of life following preclinical cardiopulmonary resuscitation. Anasth Intensivther Notfallmed, 1985. 20: p. 244-50.

19.       de Vos, R., Quality of life after cardiopulmonary resuscitation. Resuscitation, 1997. 35: p. 231-6.

20.       Phillips, S., Resuscitation for cardiogenic shock with extracorporeal membrane oxygenation systems. Semin Thorac Cardiovasc Surg, 1994. 6: p. 131-5.

21.       Younger, J., et al., Extracorporeal resuscitation of cardiac arrest [see comments]. Acad Emerg Med, 1999. 6: p. 700-7.

22.       Matsuwaka, R., et al., Emergency percutaneous cardiopulmonary support for patients with cardiac arrest or severe cardiogenic shock. Nippon Kyobu Geka Gakkai Zasshi, 1996. 44: p. 2006-10.

23.       Myerburg, R., Clinical, electrophysiologic, and hemodynamic profiles of patients resuscitated from pre-hospital cardiac arrest. . Amer J Med, 1980. 68: p. 568.

24.       Safar, P., Abramson, NS, Angelos, M, et al., Emergency cardiopulmonary bypass for resuscitation from prolonged cardiac arrest. Am J Emerg Med, 1990. 8: p. 55-67.

25.       Peterson, M., et al., Outcome after cardiopulmonary resuscitation in a medical intensive care unit. Chest, 1991. 100: p. 168-74.

26.       Gener, J., et al., Immediate and 1-year survival after cardiopulmonary resuscitation at an intensive care unit. Med Clin (Barc), 1989. 93: p. 445-8.

27.       Rubertsson, S., et al., Blood flow and perfusion pressure during open-chest versus closed-chest cardiopulmonary resuscitation in pigs. Am J Emerg Med, 1984. 23: p. 568-571.

28.       Bircher, N., Safar, P., Open-chest CPR: An old method whose time has returned. Am J Emerg Med, 1984. 2: p. 568-71.

29.       McDonald, J., Systolic and mean arterial pressures during manual and mechanical CPR in humans. Ann Emerg Med, 1982. 11: p. 292-5.

30.       Ornato, J., et al., Measurement of ventilation during cardiopulmonary resuscitation. Crit Care Med., 1983. 1: p. 79-82.

31.       Kim, H., et al., Amelioration of impaired cerebral metabolism after severe acidotic ischemia by tirilazad post-treatment in dogs. . Stroke, 1996. 27: p. 114-21.

32.       Iwatsuki, N., et al., Hyperbaric oxygen combined with nicardipine administration accelerates neurologic recovery after cerebral ischemia in a canine model. . Crit Care Med, 1994. 22: p. 858-63.

33.       Cervantes, M., Moralı´, G, Letechipı´a-Vallejo, G., Melatonin and ischemia reperfusion injury of the brain. J. Pineal Res, 2008. 45: p. 1-7.

34.       Krep, H., Bernd W, Bottiger, BW, et al., Time course of circulatory and metabolic recovery of cat brain after cardiac arrest assessed by perfusion- and diffusion-weighted imaging and MR-spectroscopy. Resuscitation, 2003. 58: p. 337-348.

35.       DeGraba, T., Pettigrew, C., Why do neroprotectivedrugs work in animals but not in humans? Neurologic Clinics, 2000. 18: p. 475-493.

36.       Ginsberg, M., Adventures in the Pathophysiology of Brain Ischemia: Penumbra, Gene Expression, Neuroprotection. The 2002 Thomas Willis Lecture. Stroke, 2003. 34: p. 214-223.

37.       Cheng, J., Al-Khoury, L, Zivin, JA., Neuroprotection for Ischemic Stroke: Two Decades of Success and Failure. NeuroRX, 2004. 1: p. 36-45.

38.       Roine, R.O., et al., Nimodipine after resuscitation from out-of-hospital ventricular fibrillation. A placebo-controlled, double-blind, randomized trial. JAMA, 1990. 264(24): p. 3171-3177.

39.       Longstreth, W.T., Jr., et al., Randomized clinical trial of magnesium, diazepam, or both after out-of-hospital cardiac arrest. Neurology, 2002. 59(4): p. 506-514.

40.       Landau, W.M., et al., Randomized clinical trial of magnesium, diazepam, or both after out-of-hospital cardiac arrest. Neurology, 2003. 60(11): p. 1868-1869.

41.       Halstrom, A., Rea, TD, et al., Manual chest compression vs use of an automated chest compression device during resuscitation following out-of-hospital cardiac arrest: a randomized trial. JAMA, 2006. 295: p. 2620-8.

42.       Lafuente-Lafuente, C., Melero-Bascones, M., Active chest compression-decompression for cardiopulmonary resuscitation. Cochrane Database Syst Rev., 2004. (2):CD002751.

43.       Aung, K., Htay, T., Vasopressin for cardiac arrest: a systematic review and meta-analysis. Arch Intern Med, 2005. 10: p. 17-24.

44.       Callaham, M., Madsen, CD, Barton, CW, Saunders, CE, Pointer, J., A randomized clinical trial of high-dose epinephrine and norepinephrine vs standard-dose epinephrine in prehospital cardiac arrest. JAMA, 1992. 268: p. 2667-72.

45.       Rincon, F., Mayer, SA., Therapeutic hypothermia for brain injury after cardiac arrest. . Semin Neurol 2006. 26: p. 387-395.

46.       Bernard, S., Gray, TW, Buist, MD, et al., Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia. N Engl J Med, 2002. 346: p. 557-563.

47.       Liu, L. and M.A. Yenari, Therapeutic hypothermia: neuroprotective mechanisms. Front Biosci, 2007. 12: p. 816-25.

48.       The-Hypothermia-after-Cardiac-Arrest-Study-Group, Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med, 2002. 346: p. 549-556.

49.       Nolan, J., Morley, PT, Vanden Hoek, TL, Hickey, RW., Therapeutic Hypothermia After Cardiac Arrest: An Advisory Statement by the Advanced Life Support Task Force of the International Liaison Committee on Resuscitation. Circulation, 2003. 108: p. 118-121.

50.       Arrich, J., Clinical application of mild therapeutic hypothermia after cardiac arrest. Crit Care Med, 2007. 35(4): p. 1041-7.

51.       Sandroni, C., et al., In-hospital cardiac arrest: incidence, prognosis and possible measures to improve survival. Intensive Care Med, 2007. 33(2): p. 237-45.

52.       Safar, P., Resuscitation from clinical death: Pathophysiologic limits and therapeutic potentials. Crit Care Med, 1988. 16: p. 923-941.

53.       Safar, P., et al., Improved cerebral resuscitation from cardiac arrest in dogs with mild hypothermia plus blood flow promotion. Stroke, 1996. 27(1): p. 105-13.

54.       Polderman, K.H., Hypothermia and neurological outcome after cardiac arrest: state of the art. Eur J Anaesthesiol Suppl, 2008. 42: p. 23-30.

55.       Busto, R., Dietrich, WD, Globus, MY, et al., Small differences in intraischemic brain temperature critically determine the extent of ischemic neuronal injury. J Cereb Blood Flow Metab, 1987. 7: p. 729-738.

56.       Safar, P., Behringer, W., ed. Brain resuscitation after cardiac arrest. Textbook of neurointensive care., ed. A. Layon, Gabrielli, A, Friedman, WA (Eds). . 2001, WB Saunders: Philadelphia.

57.       Shimohata, T., H. Zhao, and G.K. Steinberg, Epsilon PKC may contribute to the protective effect of hypothermia in a rat focal cerebral ischemia model. Stroke, 2007. 38(2): p. 375-80.

58.       Zhao, H., G.K. Steinberg, and R.M. Sapolsky, General versus specific actions of mild-moderate hypothermia in attenuating cerebral ischemic damage. J Cereb Blood Flow Metab, 2007. 27(12): p. 1879-94.

59.       American-Heart-Association, AHA CPR Guidelines: Part 7.2: Management of Cardiac Arrest. Circulation, 2005. 112:IV-58-IV-66.

60.       Abella, B., Rhee, JW, Huang, KN, et al., Induced hypothermia is underused after resuscitation from cardiac arrest: A current practice survey. Resuscitation, 2005. 64: p. 181-186.

61.       Merchant, R., Soar, J, Skrifvars, MB, et al., Therapeutic hypothermia utilization among physicians after resuscitation from cardiac arrest. . Crit Care Med, 2006. 34: p. 1935-1940.

62.       Kennedy, J., Green, RS, Stenstrom, R., The use of induced hypothermia after cardiac arrest: A survey of Canadian emergency physicians. CJEM, 2008. 10: p. 125-130.

63.       Suffoletto, B., Salcido, DD, Menegazzi, JJ., Use of prehospital-induced hypothermia after out-of-hospital cardiac arrest: A Survey of the National Association of Emergency Medical Services Physicians. Prehosp Emerg Care, 2008. 12: p. 52-56.

64.       Alam, H., Just do it. Critical care medicine,, 2008. 36: p. 2456-7.

65.       Skulec, R., et al., Induction of mild hypothermia in cardiac arrest survivors presenting with cardiogenic shock syndrome. Acta Anaesthesiol Scand, 2008. 52(2): p. 188-94.

66.       Hay, A.W., et al., Therapeutic hypothermia in comatose patients after out-of-hospital cardiac arrest. Anaesthesia, 2008. 63(1): p. 15-9.

67.       Wolfrum, S., et al., Mild therapeutic hypothermia in patients after out-of-hospital cardiac arrest due to acute ST-segment elevation myocardial infarction undergoing immediate percutaneous coronary intervention. Crit Care Med, 2008. 36(6): p. 1780-6.

68.       Bernard, S., Therapeutic hypothermia after cardiac arrest: now a standard of care. Crit Care Med, 2006. 34(3): p. 923-4.

69.       Wolff, B., et al., Early achievement of mild therapeutic hypothermia and the neurologic outcome after cardiac arrest. Int J Cardiol, 2008.

70.       Nagel, S., et al., Therapeutic hypothermia in experimental models of focal and global cerebral ischemia and intracerebral hemorrhage. Expert Rev Neurother, 2008. 8(8): p. 1255-68.

71.       Nozari, A., et al., Critical time window for intra-arrest cooling with cold saline flush in a dog model of cardiopulmonary resuscitation. Circulation, 2006. 113(23): p. 2690-6.

72.       Reich, H., et al., Cardiac resuscitability with cardiopulmonary bypass after increasing ventricular fibrillation times in dogs. Ann Emerg Med, 1990. 19(8): p. 887-90.

73.       Angelos, M., P. Safar, and H. Reich, A comparison of cardiopulmonary resuscitation with cardiopulmonary bypass after prolonged cardiac arrest in dogs. Reperfusion pressures and neurologic recovery. Resuscitation, 1991. 21(2-3): p. 121-35.

74.       Radovsky, A., et al., Regional prevalence and distribution of ischemic neurons in dog brains 96 hours after cardiac arrest of 0 to 20 minutes. Stroke, 1995. 26(11): p. 2127-33; discussion 2133-4.

75.       Thoresen, M., Wyatt, J.  [see comments], Keeping a cool head, post-hypoxic hypothermia: an old idea revisited. Acta Paediatr 1997. 86: p. 1029-33.

76.       Nozari, A., et al., Mild hypothermia during prolonged cardiopulmonary cerebral resuscitation increases conscious survival in dogs. Crit Care Med, 2004. 32(10): p. 2110-6.

77.       Kuboyama, K., Safar, P, Radovsky, A, Tisherman, SA, Stezoski, SW, Alexander, H., Delay in cooling negates the beneficial effect of mild resuscitative cerebral hypothermia after cardiac arrest in dogs: a prospective, randomized study [see comments]. 21, 1993. Crit Care Med: p. 1348-58.

78.       Takata, K., Takeda, Y, Sato T, Nakatsuka, H, Yokoyama, M, Morita, K., Effects of hypothermia for a short period on histologic outcome and extracellular glutamate concentration during and after cardiac arrest in rats. . Crit Care Med, 2005. 33: p. 1340 -1345.

79.       Abella B, Z.D., Alvarado J, Hamann K, Vanden Hoek T, Becker L., Intra-arrest cooling improves outcomes in a murine cardiac arrest model. Circulation, 2004. 109: p. 2786 -2791.

80.       Dietrich, W., Busto, R, Alonso, O, et al. 993, Intraischemic but not postischemic brain hypothermia protects chronically following global forebrain ischemia in rats. J Cerebr Blood Flow  Metab, 1993. 13: p. 541-549.

81.       Maier, C.M., et al., Delayed induction and long-term effects of mild hypothermia in a focal model of transient cerebral ischemia: neurological outcome and infarct size. J Neurosurg, 2001. 94(1): p. 90-6.

82.       Marion, D., Leonov Y, Ginsberg, M, Katz, LM, Kochanek, PM, Lechleuthner, A, Nemoto, EM, Obrist, W, Safar, P, Sterz, F, Tisherman, SA, White, RJ, Xiao, F, Zar, H., Resuscitative hypothermia. Crit Care Med, 1996. 24: p. S81-9.

83.       Zivin, J., Editorial comment. Stroke, 1999. 30: p. 1899.

84.       Wake-County-Government, Wake County Emergency Medical Services. 2008, Wake County Government.

85.       Hinchey, P., et al., Out-of-hospital cardiac arrest survival after the sequential implementation of 2005 AHA guidelines for compressions, ventilations, and induced hypothermia., in SAEM Meeting. 2008: Washington, D.C. .

86.       Olson, J., Hinchey, P, Myers, B. (2006) Wake County EMS Cools Down: System begins induced hypothermia of ROSC patients. JEMS Volume,

87.       Bayegan, K., et al., Rapid non-invasive external cooling to induce mild therapeutic hypothermia in adult human-sized swine. Resuscitation, 2008. 76(2): p. 291-8.

88.       Kurz, A., Sessler, DI, Birnbauer, F, Illievich, U, Spiss, C., Thermoregulatory vasoconstriction impairs active core cooling. Anesthesiology 1995. 82: p. 870-6.

89.       Stolwijk, J., Hardy, JD., ed. Control of body temperature: Reactions to environmental agents. In Handbook of Physiology., ed. e. Douglas HK. . 1977, American Physiological Society: Bethesda. 45-69.

90.       Doufas, A., Consequences of inadvertent perioperative hypothermia. Best Pract Res Clin Anaesthesiol, 2003. 17: p. 535-49.

91.       Sheffield, C., Sessler, DI, Hunt, TK.   , Mild hypothermia during isoflurane anesthesia decreases resistance to E. coli dermal infection in guinea pigs. Acta Anaesthesiol Scand, 1994. 38: p. 201-5.

92.       Lee, S., Battistella, FD, Go, K., Hypothermia induces T-Cell production of immunosuppressive cytokines. J Surg Res, 2001 100(4): p. 150-153.

93.       Reed, R., Bracey, AW, Hudson, JD, Miller, TA, Fisher, RP. , Hypothermia and blood coagulation: dissociation between enzyme activity and clotting factor levels. Circ Shock, 1990. 32: p. 141-152.

94.       Nirula, R., Gentilello, LM., ed. Detrimental Effects of Hypothermia Therapeutic Hypothermia., ed. S. Tisherman, Sterz, F. 2007, Springer US: New York.

95.       Flores-Maldonado, A., Medina-Escobedo, CE, Rios-Rodriguez, HM, Fernandez-Dominguez, R. , Mild perioperative hypothermia and the risk of wound infection. Arch Med Res, 2001. 32: p. 227-231.

96.       Kurz, A., Sessler, DI, Lenhardt, R., Perioperative normothermia to reduce the incidence of surgical wound infection and shorten hospitalization. N Engl J Med, 1996. 334: p. 1209-15.

97.       Sheffield, C., Sessler, DI, Hopf, HW, Schroeder, M, Moayeri, A, Hunt, TK, West, JM. , Centrally and locally mediated thermoregulatory responses alter subcutaneous oxygen tension. Wouns Rep Reg 1997. 4: p. 339-45.

98.       Wenisch, C., Narzt, E, Sessler, DI. Parschalk, B, Lenhardt, R, Kurz, A, Graninger, W., Mild intraoperative hypothermia reduces production of reactive oxygen intermediates by polymorphonuclear leukocytes. Anesth Analg, 1996. 82: p. 810-6.

99.       Beilin, B., Shavit, Y, Razumovsky, J, et al. , Effects of mild perioperative hypothermia on cellular immune responses. Anesthesiology, 1998. 89: p. 1133-1140.

100.    Jyung, R., Mustoe, TA., Role of cytokines in wound repair. Clinical Application of Cytokines., ed. J. Oppenheim, Rossio, JL, Gearing, AJH, eds. 1993, New York: Oxford University Press.

101.    Baigrie, R., Lamont, PM, Kwiatkowski, D, Dallman, MG, Morris, PG., Systemic cytokine response after major surgery. Br J Surg, 1992. 79: p. 757-60.

102.    Qwarnstrom, E., Page, RC, Gillis, S, Dower, SK., Binding, internalization and intracellular localization of interleukin-1 beta in human diploid fibroblasts. B Biol Chem, 1988;. 263:: p. 8261-9.

103.    Peterson, K., Carson, S, Carney, N., Hypothermia treatment for traumatic brain injury: a systematic review and meta-analysis. J Neurotrauma, 2008. 25(1): p. 62-71.

104.    Hammersborga, S., Brekkea, KB, Haugena, OM, et al., Surface cooling versus core cooling: Comparative studies of microvascular fluid- and protein-shifts in a porcine model. Resuscitation, 2008. 79: p. 292-300.

105.    Merchant, R., Abella,BS, Peberdy, MA, Soar, J, et al., Therapeutic hypothermia after cardiac arrest: Unintentional overcooling is common using ice packs and conventional cooling blankets. Crit Care Med, 2006. 34[Suppl.]: p. S490-S494.

106.    Al-Senani, F., Graffagnino, C, Grotta, JC, et al., A prospective, multicenter pilot study to evaluate the feasibility and safety of using the CoolGard System and Icy catheter following cardiac arrest. . Resuscitation, 2004. 62(2): p. 143-50.

107.    Chen, R., Chien, S., Hemodynamic functions and blood viscosity in surface hypothermia. Am J Physiol Lung Cell Mol Physiol, 1978. 235: p. H136-43.

108.    Zarins, C., Skinner, DB., Hemodynamics and blood flow distribution following prolonged circulation at 5 degrees C. Am J Physiol., 1975. 229: p. 275-8.

109.    Hammersborga, S., Farstad, M, Haugen, O, Kvalheim, VL, Onarheim, H, Husby, P., Time course variations of haemodynamics, plasma volume and microvascular fluid exchange following surface cooling: an experimental approach to accidental hypothermia. Resuscitation, 2005. 65: p. 211-9.

110.    Lewis, T., Swelling of the human limbs in response to immersion in cold water. Clin Sci, 1939. 4.

111.    Farry, P., Prentice, NG, Hunter, AC, Wakelin, CA., Ice treatment of injured ligament: an experimental model. N Z Med J. , 1980. 91: p. 12-14.

112.    Matthew, C., Sils, IV, Bastille, AM., Tissue-specific extravasation of albumin-bound Evans blue in hypothermic and rewarmed rats. Can J Physiol Pharmacol, 2002. 80(3): p. 233-243.

113.    Kruuv, J., Glofcheski, D, Cheng, KH, Campbell, SD, Al-Qysi, HMA, Nolan, WT, Lepock, JR, Factors influencing survival and growth of mammalian cells exposed to hypothermia. I. Effects of temperature and membrane lipid perturbers. J Cell Physiol, 1983. 115: p. 179-185.

114.    Rule, G., Frim, J, Thompson, JE, Lepock, JR, Kruuv, J.  , Theeffect of membrane lipid perturbers on survival of mammalian cells to cold. Cryobiology 1978. 15: p. 408-414.

115.    Marsh, D., Lindell, SL, Fox, LE, Belzer, FO, Southard, JH. , Hypothermic preservation of hepatocytes. I. Role of cell swelling. Cryobiology, 1989. 26: p. 524-534.

116.    McAnulty, J., Ametani, MS, Southard, JH, Belzer, FO., Effect of hypothermia on intracellular Ca21 in rabbit renal tubules suspended in UW-gluconate preservation solution. . Cryobiology, 1996. 33: p. 196-204.

117.    Breton, S., Brown, D., Cold-induced microtubule disruption and relocalization of membrane proteins in kidney epithelial cells. J Am Soc Nephrol, 1998. 9: p. 155-166.

118.    Stefanovich, P.E., RM, Sheehan, SJ, Tompkins, RG, Yarmush, ML., Effects of hypothermia on the function, membrane integrity, and cytoskeletal structure of hepatocytes. Cryobiology, 1995. 23: p. 389-403.

119.    Zieger, M., Glofcheski, DJ, Lepock, JR, Kruuv, J. , Factors influencing survival of mammalian cells exposed to hypothermia IV. Effects of iron chelation. . Cryobiology, 1990. 27: p. 452-464.

120.    Rauen, R., DeGroot, H., Cold induced release of reactive oxygen species asa decisive mediator of hypotherrmiainjury to cultured liver cells. Free Rad Biol & Med, 1998. 24(7/8): p. 1316-1323.

121.    Hochachka, P., Defense strategies against hypoxia and hypothermia. Science 1986. 231: p. 234-241.

122.    Endrich, B., Hammersen F, Messmer, K., Microvascular ultrastructure in non-freezing cold injury. Res Exp Med, 1990. 190: p. 365-79.

123.    Verrey, F., Groscurth, P, Bolliger, UI., Cytoskeletal disruption in A6 kidney cells: Impact on endo-exocytosis and NaCI transport regulation by antidiuretic hormone. J Membr Bio, 1995. 145: p. 193-204.

124.    Hall, S., Komai, H, Reader, J, Haworth, SG. , Donor lung preservation: Effect of cold preservation fluids on cultured pulmonary endothelial cells. . Am J Physiol Lung Cell Mol Physiol, 1994. 267: p. L508-L5 I 7.

125.    Breton, S., Brown, D., Cold-induced microtubule disruption and relocalization of membrane proteins inkidney epithelial cells. J Am Soc Nephrol, 1998. 9: p. 155-66.

126.    Reed, C., Clark, D, editors, ed. Heat exchangers and hypothermia. Cardiopulmonary Perfusion. 1975, Texas Medical Press, Inc.: Houston TX. 272–8.

127.    Furuse, M., Ohta, T, Ikenaga, T, et al., Effects of intravascular perfusion of cooled crystalloid solution on cold-induced brain injury using an extracorporeal cooling-filtration system. Acta Neurochir, 2003. 145: p. 983-993.

128.    Behringer, W., et al., Veno-venous extracorporeal blood shunt cooling to induce mild hypothermia in dog experiments and review of cooling methods. Resuscitation, 2002. 54(1): p. 89-98.

129.    Overlie, P., Emergency Use of Cardiopulmonary Bypass. Journal of Interventional Cardiology, 1995. 8: p. 239-247.

130.    Kurusz, M., Zwischenberger, JB., Percutaneous cardiopulmonary bypass for cardiac emergencies. Perfusion, 2002. 17: p. 269-77.

131.    Holzer, M., Behringer, W, Janata, A, Bayegan, K, Schima, H, Deckert, Z, Losert, U, Laggner, AN, Sterz, F., Extracorporeal venovenous cooling for induction of mild hypothermia in human-sized swine. Crit Care Med, 2005. 33: p. 1346-50.

132.    Bernard, S., Buist, M, Monterio, O, Smith, K., Induced hypothermia using large volume, ice-cold intravenous fluid in comatose survivors of out-of-hospital cardiac arrest: a preliminary report. Resuscitation, 2003. 56: p. 9-13

133.    Bernard, S., Rosalion, A., Therapeutic hypothermia induced during cardiopulmonary resuscitation using large-volume, ice-cold intravenous fluid. Resuscitation, 2008. 76: p. 311 – 313.

134.    Kamarainen, A., et al., Prehospital induction of therapeutic hypothermia during CPR: a pilot study. Resuscitation, 2008. 76(3): p. 360-3.

135.    Haslam, D., James, WP. , Obesity. Lancet, 2005. 366: p. 1197-209.

136.    Feldschuh, J., Enson, Y., Prediction of the normal blood volume. Relation of blood volume to body habitus. Circulation, 1977. 56: p. 605 – 612.

137.    Plaisier, B.R., Thoracic lavage in accidental hypothermia with cardiac arrest–report of a case and review of the literature. Resuscitation, 2005. 66(1): p. 99-104.

138.    Xiao, F., P. Safar, and H. Alexander, Peritoneal cooling for mild cerebral hypothermia after cardiac arrest in dogs. Resuscitation, 1995. 30(1): p. 51-9.

139.    Darwin, M., Leaf, JD, Hixon, HL. Neuropreservation of Alcor patient A-1068. Cryonics  1986  [cited 7 2]; 17-32]. Available from: http://www.alcor.org/Library/html/casereport8504.html.

140.    Darwin, M. Cryopreservation Patient Case Report: Arlene Francis Fried, A-1049.  2006  [cited 2007 07 June]; Available from: http://www.alcor.org/Library/html/fried.html.

141.    Jones, T., Cryopreservation case report: the cryopreservation of patient A-2063. Cryonics, 2005. 26(1): p. 10-12.

142.    Bellamy, R., The causes of death in conventional land warfare: Implications for combat casualty research. Military Medicine, 1984. 149: p. 55-62.

143.    Bellamy, R., Death on the battlefield and the role of first aid. Military Medicine, 1987a. 152: p. 634-35.

144.    Bellamy, R., Death on the battlefield and the role of first aid. Military Medicine. 1987b. 152 p. 617-21.

145.    Bellamy, R., Trauma epidemiology of combat casualties. , in Presentation to Institute of Medicine Committee on Fluid Resuscitation for Combat Casualties. 1998: Washington, D.C.

146.    Bowen, T., Bellamy, RF, (eds). , Emergency war Surgery: Second United States Revision of the Emergency War Surgery Handbook. 1998, US Government printing Office: Washington, D.C.

147.    Wu, X., et al., Emergency preservation and resuscitation with profound hypothermia, oxygen, and glucose allows reliable neurological recovery after 3 h of cardiac arrest from rapid exsanguination in dogs. J Cereb Blood Flow Metab, 2008. 28(2): p. 302-11.

148.    Tisherman, S., Rodriguez, A, Safar, P., Therapeutic hypothermia in traumatology. Surg Clin North Am, 1999. 79: p. 1269-89.

149.    www.engineeringtoolbox.com, Dry air properties at temperatures ranging 175 – 1900 K – specific heat capacity, ratio of specific heats, dynamic viscosity, thermal conductivity, Prandtl number, density and kinematic viscosity. 2008.

150.    www.engineeringtoolbox.com., Thermal properties of water – density, freezing temperature, boiling temperature, latent heat of melting, latent heat of evaporation, critical temperature and more. 2008.

151.    Zikria, B., Ferrer, JM, Malm, JR., Pulmonary hypothermia in dogs. J Appl Physiol 1968. 24: p. 707-10.

152.    Harrison, M., Hysing, ES, Bo, G., Control of body temperature: use of respiratory tract as a heat exchanger. J Pediatr Surg, 1977. 12: p. 821-8.

153.    Powell, F., Hopkins, SR., Comparative physiology of lung complexity: implications for gas exchange. News in Physiological Science, 2004. 19: p. 55-60.

154.    Webb, P., ed. Cold exposure. In: The physiology and medicine of diving., ed. P. Bennett, Elliot, DM, eds. . 1975, Bellaire Tindall: London.

155.    Shaffer, T., et al., The effects of liquid ventilation on cardiopulmonary function in preterm and neonatal lambs. Pediatr Res, 1983. 4: p. 303-6.

156.    Klystra, J., Survival in air after breathing fluid. Lancet, 1962. 2: p. 1170.

157.    Klystra, J., et al., Hydraulic Compression of Mice to 166 Atmospheres. Science, 1967. 158: p. 793-94.

158.    Klystra, J., Lanphier, EH., Gas Exchange in Fluid Ventilated Dogs. Fed.Proc, 1964. 23: p. 469.

159.    Darwin, M., Liquid ventilation: A bypass on the way to bypass. CryoCare Report, 1996(#7): p. 11-16.

160.    Safar, P., Stezoski, W, Nemoto, EM,  Safar, P (P.I.). Amelioration of brain damage after 12 minutes’ cardiac arrest in dogs. Brain Resuscitation Clinical Trial I Study Group. Arch Neurol, 1976. 33/2: p. 91-95.

161.    Darwin, M., Russell, S, Rasch, C, O’Farrell, J, Harris, S., A novel method of rapidly inducing or treating hypothermia or hyperpyrexia, by means of ‘mixed-mode’ (gas and liquid) ventilation using perfluorochemicals., in In: Society of Critical Care Medicine 28th Educational and Scientific Symposium. 1999, Critical Care Medicine: San Francisco. p. A81.

162.    Taylor, R., et al., Design of the mammalian respiratory system. III. Scaling maximum aerobic capacity to body mass: Wild and domestic mammals. Respir Physiol, 1981. 44: p. 25-37.

163.    Clark, L., Gollin, F., Survival of Mammal Breathing Organic Liquid Equilibrated with Oxygen at Atmospheric Pressure. Science, 1966: p. 1755.

164.    Leach, C., et al., Perfluorocarbon associated gas exchange (partial liquid  ventilation) in respiratory syndrome: A prospective, randomized, controlled study. Crit Care Med, 1993. 21: p. 1270-8.

165.    Hirschl, R., Pranikoff, T, Gauger, P, et al., Liquid ventilation in adults, children and neonates. Lancet, 1995. 346: p. 1201-2.

166.    Varani, J.H., RB, Dame, M, Johnson, K., Perfluorocarbon protects lung epithelial cells from neutrophil-mediated injury in an in vitro model of liquid ventilation therapy. Shock, 1996. 6(5): p. 339-44.

167.    Hirschl, R., Tooley, R, Parent, AC, Johnson, K, Bartlett, RH., Improvement of gas exchange, pulmonary function, and lung injury with partial liquid ventilation. A study model in a setting of severe respiratory failure. Chest, 1995. 108(2): p. 500-8.

168.    Uihlein, A., Theye, RA, Dawson, B, et al., The use of profound hypothermia, extracorporeal circulation and total circulatory arrest for an intracranial aneurysm. Preliminary report with reports of cases. Mayo Clin Proc, 1960. 35: p. 567-576.

169.    Michenfelder, J.D., et al., Profound Hypothermia and Total Circulatory Arrest in Neurosurgery: Methods, Results and Physiologic Effects. Acta Neurochir Suppl, 1964. 20: p. SUPPL 13:159+.

170.    Woodhall, B., Sealy, WC, Hall, KD, et al., Craniotomy under conditions of quinidine-protected cardioplegia and profound hypothermia. Ann Surg, 1960. 152: p. 37-44.

171.    Du Bouchet, N., E. Prochiant, and B. Latscha, [Cardiac arrest induced by deep hypothermia. Its superiority over other technics.]. Anesth Analg (Paris), 1962. 19: p. 239-50.

172.    Vigouroux, R., et al., [Hypothermia and cardiac arrest. (Considerations on a case of recuperation after 80 minutes of ventricular fibrillation)]. Mars Chir, 1962. 14: p. 102-6.

173.    Benson, D.W., et al., The use of hypothermia after cardiac arrest. Anesth Analg, 1959. 38: p. 423-8.

174.    Feldman, E., B. Rubin, and S.N. Surks, Beneficial effects of hypothermia after cardiac arrest. J Am Med Assoc, 1960. 173: p. 499-501.

175.    Lore, J.M., S.G. Gordon, and E.W. Gordon, Successful use of hypothermia following cardiac arrest in twelve-day-old infant. N Y State J Med, 1960. 60: p. 278-9.

176.    Gravel, J.A., J.P. Dechene, and M. Beaulieu, [Hypothermia in the prevention of cerebral lesions following cardiac arrest.]. Laval Med, 1960. 29: p. 48-60.

177.    Barone, F., Feuerstein, GZ, White, RF., Brain cooling during transient focal ischemia provides complete neuroprotection. Neurosci Biobehav Rev, 1997. 21: p. 31-44.

178.    Bell, T.E., G.L. Kongable, and G.K. Steinberg, Mild hypothermia: an alternative to deep hypothermia for achieving neuroprotection. J Cardiovasc Nurs, 1998. 13(1): p. 34-44.

179.    Yenari, M.A. and H.S. Han, Influence of hypothermia on post-ischemic inflammation: role of nuclear factor kappa B (NFkappaB). Neurochem Int, 2006. 49(2): p. 164-9.

180.    Krieger, D.W. and M.A. Yenari, Therapeutic hypothermia for acute ischemic stroke: what do laboratory studies teach us? Stroke, 2004. 35(6): p. 1482-9.

181.    Corbett, D., Thornhill, J.  , Temperature modulation (hypothermic and hyperthermic conditions) and its influence on histological and behavioral outcomes following cerebral ischemia. Brain Pathol, 2000. 10: p. 145-52.

182.    Reed, C., Clark, D. , Heat exchangers and hypothermia. Cardiopulmonary Perfusion. 1975: Texas Medical Press, Inc. 272-8.

183.    Darwin, M., Cryopreservation of CryoCare Patient C-2150. CryoCare Report #6, January, 1999: p. 4-16.

184.    Notebook MPI: 3M Performance Fluids. 1995, St Paul MN: 3M Company.

185.    Giesbrecht, G., Bristow, GK., A second postcooling afterdrop: more evidence for a convective mechanism. J Appl Physiol, 1992. 73: p. 1253-8.

186.    Clark, L., Jr, Hoffmann, RE, Davis, SL., Response of the rabbit lung as a criterion of safety for fluorocarbon breathing and blood substitutes. . Biomater Artif Cells Immobilization Biotechnol, 1992. 20: p. 1085-99.

187.    Yannopoulos, D., Nadkarni, VM, McKnite, SH, Rao, A, Kruger, K, Metzger, J, Benditt, D, Lurie,KJ., Intrathoracic pressure regulator during continuous-chest-compression advanced cardiac resuscitation improves vital organ perfusion pressures in a porcine model of cardiac arrest. Circulation, 2005. 112: p. 803-81.

188.    Lurie, K.Z., T, McKnite, S., The use of an inspiratory impedance valve improves neurologically intact survival in a porcine model of ventricular fibrillation. Circulation, 2002. 105: p. 124 -129.

189.    Lurie, K., Zielinski, T, Voelckel, W, et al., Augmentation of ventricular preload during treatment of cardiovascular collapse and cardiac arrest. Crit Care Med 2002. 30: p. (Suppl):S162-5.

190.    Miyamoto, Y., Mikami, T., Maximum capacity of ventilation and efficiency of gas exchange during liquid breathing in guinea pigs. Jpn J Physiol 1976. 26: p. 603-18.

191.    Koen, P., Wolfson, MR, Shaffer, TH., Fluorocarbon ventilation: maximal expiratory flows and CO2 elimination. Pediatr Res, 1988. 24: p. 291-6.

192.    Matthews, W., Balzer, RH, Shelburne, JD, Pratt, PC, Kylstra, JA., Steady-state gas exchange in normothermic, anesthetized, liquid ventilated dogs. Undersea Biomed Res, 1978. 5: p. 341-54.

193.    Reece, W., ed. Respiration in mammals. In Duke’s Physiology of Domestic Animals, 10th Edition, ed. e. Swenson M. 1987, Comstock Publishing Associates: London. 235–7.

194.    Yu, C., Jimenez, O, Marcillo, AE, Weider, B, Bangerter, K, Dietrich, WD, Castro, S, Yezierski, RP., Beneficial effects of modest systemic hypothermia on locomotor function and histopathological damage following contusion-induced spinal cord injury in rats. J Neurosurg, 2000. 93(1 Suppl): p. 85-93.

195.    Ha, K., Kim, YH., Neuroprotective effect of moderate epidural hypothermia after spinal cord injury in rats. Spine, 2008. 33: p. 2059-65.

196.    Shibuya, S., Miyamoto, O, Janjua, NA, Itano, T, Mori, S, Norimatsu, H., Post-traumatic moderate systemic hypothermia reduces TUNEL positive cells following spinal cord injury in rat. Spinal Cord, 2004. 42: p. 29-34.

197.    Safar, P., Cerebral resuscitation (Letter). Ann Emerg Med, 1993;. 22: p. 759.

198.    Safar, P., On the Future of Reanimatology. Academic Emergency Medicine, 2000. 7(1): p. 75-89.

199.    Schmid-Elsaesser, R., Hungerhuber, E, Zausinger, S, Baethmann, A, Reulen, HJ., Combination drug therapy and mild hypothermia: a promising treatment strategy for reversible, focal cerebral ischemia. Stroke, 1999. 30: p. 1891-9.

200.    Rose, G., Cold weather and ischaemic heart disease. Br J Prev Soc Med 1966. 20: p. 97-100.

201.    Anderson, T., Rochard, C., Cold snaps, snowfall and sudden death from ischaemic heart disease. Can Med Assoc, 1979. 121: p. 1580-3.

202.    Epstein, S., Stampfer, M, Beiser, GD, Goldstein, RE, Braunwald, E., Effects of a reduction in environmental temperature on the circulatory response to exercise in man.  Implications concerning angina pectoris. N Engl J Med, 1969. 280: p. 7-11.

203.    Hayward, J., Holmes, WF, Gooden, BA., Cardiovascular responses in man to a stream of cold air. Cardiovasc Res 1976. 10: p. 691-6.

204.    Velasco, M., Gómez, J, Blanco, M, Rodriguez, I., The cold pressor test: pharmacological and therapeutic aspects. Am J Ther, 1997. 4: p. 34-8.

205.    Gondi, B., Nanda, NC., Evaluation of coronary artery disease by cold pressor two-dimensional echocardiography. Circulation, 1981. 64: p. Abstracts of 54th scientific sessions of American Heart Association, iv-14.

206.    Mudge, G., Grossman,W, Mills, RM, Lesch, M, Braunwald, E., Reflex increase in coronary vascular resistance in patients with ischaemic heart disease. . N Engl J Med, 1976. 295(1): p. 333-7.

207.    Raizner, A., Chahine, RA, Ishimori, T, Verani, MS, Zacca, N, Jamal, N, Miller, RR, Luchi, RJ., Provocation of coronary artery spasm by the cold pressor test. Hemodynamic, arteriographic and quantitative angiographic observations. Circulation, 1980. 62: p. 925-932.

208.    Hattenhauer, M., Neil, WA., The effect of cold air inhalation on angina pectoris and myocardial oxygen supply. Circulation, 1975. 51: p. 1053-8.

209.    Lassvik, C., Areskog, NH., Angina pectoris during inhalation of cold air reactions to exercise. Br Heart J, 1980. 43: p. 661-7.

210.    Dodds, P., Bellamy, CM, Muirhead, RA, Perry, RA., Vasoconstrictor peptides and cold intolerance in patients with stable angina pectoris. Heart, 1995. 73: p. 25-31.

211.    Maggiore, Q., Pizzarelli, F, Sisca, S, et al., Blood temperature and vascular stability during hemodialysis and hemofiltration. Trans Am Soc Artif Intern Organs, 1982. 28: p. 523-527.

212.    Pizzarelli, F., Sisca, S, Zoccali, C, et al., Blood temperature and vascular stability in hemofiltration. Int J Artif Organs, 1983. 6: p. 37-41.

213.    Sherman, R., Faustino, EF, Bernholc, AS, Eisinger, RP., Effect of variations in dialysate temperature on blood pressure during hemodialysis. Am J Kidney Dis, 1984. 4: p. 66-68.

214.    Kishimoto, T., Yamamoto, T, Shimuzu, G, et al., Cardiovascular stability in low temperature dialysis. Dial Transplant, 1986. 15: p. 329-333.

215.    Selby, N., McIntyre, CW., A systematic review of the clinical effects of reducing dialysate fluid temperature. Nephrol Dial Transplant, 2006. 21: p. 1883-1898.

216.    Mahida, B., Duler, F, Zasuwa, G, Fleig, G, Levin, NW., Effect of cooled dialysate on serum catecholamines and blood pressure stability. Trans Am Soc Artif Intern Organs, 1983. 29: p. 384-389.

217.    Pizzarelli, F., Fom cold to isothermic dialysis: a twenty-five year voyage. Nephrol Dial Transplant, 2007. 22: p. 1007-1012.

218.    Maggiore, Q., Enia, G, Catalano, C, Misefari, V, Mundo, A., Effect of blood cooling on cuprophan-induced anaphylotoxin generation. Kidney Int, 1987. 32: p. 908-911.

219.    Raja, R., Kramer, M, Alvis, R, Goldstein, S, DeLosAngeles, A., Effect of varying dialysate temperature on hemodialysis hypoxemia. Trans Am Soc Artif Organs, 1984. 30: p. 15-17.

220.    Behringer, W., Kittler, H., Sterz, F, Domanovits, H, Schoerkhuber, W, Holzer, M, Mullner, M, Laggner, AN., Cumulative epinephrine dose during cardiopulmonary resuscitation and neurologic outcome. Annals of Internal Medicine, 1998. 129: p. 450-456.

221.    Adgey, A., Johnston PW., Approaches to modern management of cardiac arrest. Heart, 1998. 80: p. 397-401.

222.    Rhee, B., Zahng, Y Boddicker, KA, Davies, LR, Kerber, RE., Effect of hypothermia on transthoracic defibrillation in a swine model . Resuscitation, 2005. 65: p. 79-85.

223.    Wira, C., et al., Resuscitation, 2006. 69: p. 509-16.

224.    Boddicker, K., Zang, Y, Zimmerman, B, Davies, LR, Kerber, RE., Hypothermia Improves Defibrillation Success and Resuscitation Outcomes From Ventricular Fibrillation. Circulation, 2005. 111: p. 3195-3201.

225.    Martin, D., Garcia, J, Valer, CR, Khuri, SF. , The effects of normothermic and hypothermic cardiopulmonary bypass on defibrillation energy requirements and transmyocardial impedance. Implications for implantable cardioverter-defibrillator implantation. The Journal of Thoracic and Cardiovascular Surgery, 1995. 109(5): p. 981-988.

226.    Ujhelyi, M., Sims, JJ, Dubin, SA, Vender, JM, Miller, WA., Defibrillation energy requirements and electrical heterogeneity during total body hypothermia. Crit Care Med, 2001. 29(5): p. 1006-1011.

227.    Rankin, A., Rae, AP., Cardiac arrhythmias during rewarming of patients with accidental hypothermia. Br Med J 1984. 289: p. 874-7.

228.    Covino, B., Charleston, DA, D’Amato, HE., Ventricular fibrillation in the hypothermic dog. . Am J Physiol Lung Cell Mol Physiol, 1954. 178: p. 148-54.

229.    Okada, M., The cardiac rhythm in accidental hypothermia. J Electrocardiol, 1984. 17: p. 123-8.

230.    Alexander, L., The treatment of shock from prolonged exposure to cold, especially in water., S.R.A. C.I.O.S. Target Medical Combined Intelligence Objectives Subcommittee. G-S Division, Editor. 1945, Office of the Publication Board, Department of Commerce, Washington D.C.

231.    Reuler, J., Hypothermia: Pathophysiology, clinical setting, and management. Ann Intern Med, 1978. 89: p. 519-527.

232.    Southwick, F., Dalglish, PH, Jr., Recovery after prolonged asystolic cardiac arrest in profound hypothermia: a case report and literature review. JAMA, 1980. 243: p. 1250-3.

233.    Mortensen, E., Berntsen, R, Tveita, T, Lathrop, DA, Refsum, H., Changes in ventricular fibrillation threshold during acute hypothermia. A model for future studies. J Basic Clin Physiol Pharmacol, 1993. 4: p. 313-9.

234.    Baumgardner, J., Baranov, D, Smith, DS, Zager, EL., The effectiveness of rapidly infused Intravenous fluids for inducing moderate hypothermia in neurosurgical patients. Anesth Analg, 1999. 89: p. 163-9.

235.    Rajek, A., Greif ,R ,Sessler, DI, Baumgardner, J, Laciny, S, Bastanmehr, H., Core Cooling by Central Venous Infusion of Ice-cold (4°C and 20°C) Fluid. Anesthesiology, 2000. 93.

236.    Virkkunen, I., Yli-Hankala, A, Silfvast, T., Induction of therapeutic hypothermia after cardiac arrest in prehospital patients using ice-cold Ringer’s solution: a pilot study. Resuscitation, 2004. 62: p. 299-302.

237.    Baba, A., Kim, YK, Zhang, H, Liu, M, Slutsky, AS., Perfluorocarbon blocks tumor necrosis factor-alpha-induced interleukin-8 release from alveolar epithelial cells in vitro. Crit Care Med, 2000. 28: p. 1113-18.

238.    Kliegel, A., Janata, A, Wandaller, C, et al., Cold infusions alone are effective for induction of therapeutic hypothermia but do not keep patients cool after cardiac arrest. Resuscitation, 2007. 73 p. 46-53.

239.    Nordmark, J., Rubertsson, M., Induction of mild hypothermia with infusion of cold (4 ◦C) fluid during ongoing experimental CPR. Resuscitation, 2005. 66: p. 357-365.

240.    Cohen, T., Tucker, KJ, Redberg, RF, Lurie, KG, Chin, MC, Dutton, JP, Scheinman, MM., Active compression-decompression resuscitation: a novel method of cardiopulmonary resuscitation. Am Heart J, 1992. 124: p. 1145-50.

241.    Cabrini, L., Beccaria, P, Landoni, G, Biondi-Zoccai, GG, Sheiban, I, Cristofolini, M, Fochi, O, Maj, G, Zangrillo, A., Impact of impedance threshold devices on cardiopulmonary resuscitation: a systematic review and meta-analysis of randomized controlled studies. Crit Care Med, 2008. 36: p. 1625-32.

242.    Lurie, K., Voelckel, W, Plaisance, P, et al., Use of an inspiratory impedance threshold valve during cardiopulmonary resuscitation: A progress report. Resuscitation, 2000. 44: p. 219-30.

243.    Plaisance, P., Lurie, K, Payen, D., Inspiratory impedance during active compression decompression cardiopulmonary resuscitation: a randomized evaluation in patients in cardiac arrest. . Circulation, 2000. 10: p. 989-994.

244.    Wolcke, B., Mauer, DK, Schoefmann, MF, Teichmann, H, Provo, TA, Lindner. KH, Dick WF, Aeppli D, Lurie KG., Comparison of standard cardiopulmonary resuscitation versus the combination of active compression-decompression cardiopulmonary resuscitation and an inspiratory threshold device for out-of-hospital cardiac arrest. Circulation, 2003. 108: p. 2201-2205.

245.    Jellinek, H., Krenn, H, Oczenski, W, et al., Influence of positive airway pressure on the pressure gradient for venous return in humans. J Appl Physiol, 2000. 88: p. 926-932.

246.    Aufderheide, T., Lurie, KG., Death by hyperventilation: A common and life-threatening problem during cardiopulmonary resuscitation. Crit Care Med, 2004. 32(No. 9 (Suppl.)): p. S345–S351.

247.    Aufderheide, T., Sigurdsson, G, Pirrallo, RG, et al. , Hyperventilation-induced hypotension during cardiopulmonary resuscitation. Circulation, 2004. 109: p. 1960-1965.

248.    Cheifetz, I., Craig, DM, Quick, G, et al., Increasing tidal volumes and pulmonary overdistention adversely affect pulmonary vascular mechanics and cardiac output in a pediatric swine model. Crit Care Med, 1998. 26: p. 710-716.

249.    Fewell JE, A.D., Carlson CJ, et al., Continuous positive-pressure ventilation decreases right and left ventricular end diastolic volumes in the dog. Circ Res, 1980. 46: p. 125-132.

250.    O’Neil, J., Deakin, CD., Do we hyperventilate cardiac arrest patients? Resuscitation, 2007. 73: p. 82-85.

251.    Guerci, A., Shi, AY, Levin, H, Tsitlik, J, Weisfeldt, ML, Chandra, N., Transmission of Intrathoracic Pressure to the Intracranial Space during Cardiopulmonary Resuscitation in Dogs. Circ Res, 1985. 56: p. 20-30.

252.    Srinivasana, V., Nadkarnia, VA, Yannopoulosb, D, Marinoa, BS, Sigurdssonc, G,  McKnitec, SH, Zookc, M, Bendittc, DG, Lurie, KG., Spontaneous gasping decreases intracranial pressure and improves cerebral perfusion in a pig model of ventricular fibrillation. Resuscitation, 2006. 69: p. 329-334.

253.    Chandra, N.C., et al., Observations of hemodynamics during human cardiopulmonary resuscitation. Crit Care Med, 1990. 18: p. 929-34.

254.    Weil, M., Rackow, EC, Trevino, R, Grundler, W, Falk, JL, Griffel, MI., Difference in acid-base state between venous and arterial blood during cardiopulmonary resuscitation. N Engl J Med, 1986. 315: p. 153-156.

255.    Sigurdsson, G., Yannopoulos, D, McKnite, S, et al., Lowering of intrathoracic pressure improves blood pressure and survival rates in a porcine model of hemorrhagic shock. Resuscitation, 2006. 68: p. 399-404.

256.    Rudikoff, M., Maughan, WL, Effron, M. Freund. P, Weisfeldt, ML., Mechanisms of blood flow during cardiopulmonary resuscitation. Circulation, 1980. 61: p. 345-352.

257.    Maier, G., Tyson, GS. Olsen, CO, Kerstein, KH, Davis, JW, Conn, A,. Sabiston, DW, Jr., Rankin, JS., The physiology of external cardiac massage: high-impulse cardiopulmonary resuscitation. Circulation 1984. 70: p. 86-101.

258.    Frenneaux, A., Haemodynamics of cardiac arrest and resuscitation. Curr Opin Crit Care, 2006. 12: p. 198-203.

259.    Kouwenhoven, W., Jude, JR, Knickerbocker, GG., Closed-chest cardiac massage. JAMA 1960. 173: p. 1064-1067.

260.    Redberg, R., Tucker, KJ, Cohen, TJ, Dutton, JP, Callaham, ML, Schiller, NB., Physiology of blood flow during cardiopulmonary resuscitation. A transesophageal echocardiographic study. Circulation 1993. 88: p. 534-542.

261.    Weisfeldt, M., Chandra, N., Physiology of cardiopulmonary resuscitation. Ann Rev Med, 1981. 32: p. 43-42.

262.    Niemann, J., Garner, D, Rosborough, J, Criley, JM., The mechanism of blood flow in closed chest cardiopulmonary resuscitation. Circulation, 1979. 59 & 60: p. I1-74.

263.    Niemann, J.R., JP, Hausknecht, M, Garner, D, Criley, JM., Pressure-synchronised cineangiography during experimental cardiopulmonary resuscitation. Circulation, 1981: p. 985-991.

264.    Haas, T., Voelcke,l WG; Wenzel, V,Antretter, H, Dessl, A, Lindner, K.H., Revisiting the cardiac versus thoracic pump mechanism during cardiopulmonary resuscitation. Resuscitation 2003. 58: p. 113-116.

265.    Werner, J., Greenem, HL, Janko, CK, Cobb, LA., Visualization of cardiac valve motion in man during external chest compression using two-dimensional echocardiography. Implications regarding the mechanism of blood flow. Circulation, 1981. 63: p. 1417-1421.

266.    Cohen, J., Chandra, N, Alderson,PO, van Aswegen, A, Tsitlik, JE, Weisfeldt, ML., Timing of pulmonary and systemic blood flow during intermittent high intrathoracic pressure cardiopulmonary resuscitation in the dog. . Am J Cardiol, 1982. 49: p. 1883-1889.

267.    Rich, S., Wix, HL, Shapiro, EH., Clinical assessment of heart chamber size and valve motion during cardiopulmonary resuscitation by two-dimensional echocardiography. Am Heart J 1981. 102: p. 368-373.

268.    Feneley, M., Maier, GW, Gaynor, JW, Gall, SA, Kisslo, JA, Davis, JW, Rankin, JS., Sequence of mitral valve motion and transmitral blood flow during manual cardiopulmonary resuscitation in dogs. Circulation, 1987. 76: p. 363-375.

269.    Deshmukh, H., Weil, MH, Rackow, EW, Trevino, R, Bisera, J., Echocardiographic observations during cardiopulmonary resuscitation: A preliminary report. J.Crit Care Med, 1985. 13: p. 904-906.

270.    Ma, M., Hwang, JJ, Lai, LP, et al., Transesophageal echocardiographic assessment of mitral valve position and pulmonary venous flow during cardiopulmonary resuscitation in humans. . Circulation 1995. 92: p. 854-861.

271.    Hong, S.B., et al., Physiologic characteristics of cold perfluorocarbon-induced hypothermia during partial liquid ventilation in normal rabbits. Anesth Analg, 2002. 94(1): p. 157-62, table of contents.

272.    Harris, S., et al., Rapid (0.5°C/min) minimally invasive induction of hypothermia using cold perfluorochemical lung lavage in dogs. Resuscitation, 2001. 50: p. 189-204.

273.    Ko, A.C., et al., Segmental hemodynamics during partial liquid ventilation in isolated rat lungs. Resuscitation, 2003. 57(1): p. 85-91.

274.    Beznak, M., Cardiac output during the development of cardiac hypertrophy in the rat. Circ Res, 1958. 6: p. 207-212.

275.    Tissier, R., et al., Total liquid ventilation provides ultra-fast cardioprotective cooling. J Am Coll Cardiol, 2007. 49(5): p. 601-5.

276.    Schwartz, L., Verbinski, SG, Vander Heide, RS, Reimer, KA., Epicardial temperature is a major predictor of myocardial infarct size in dogs. J Mol Cell Cardiol, 1997. 29: p. 1577- 83.

277.    Dixon, S., Whitbourn, RJ, Dae, MW, et al., Induction of mild systemic hypothermia with endovascular cooling during primary percutaneous coronary intervention for acute myocardial infarction. J Am Coll Cardiol 2002. 40: p. 1928 -34.

278.    Miki, T., Liu GS, Cohen, MV, Downey, JM., Mild hypothermia reduces infarct size in the beating rabbit heart: a practical intervention for acute myocardial infarction? Basic Res Cardiol, 1998. 93: p. 372- 83.

279.    Hale, S., Dave, RH, Kloner, RA., Regional hypothermia reduces myocardial necrosis even when instituted after the onset of ischemia. Basic Res Cardiol, 1997. 92: p. 351-7.

280.    Hale, S., Kloner, RA., Myocardial temperature reduction attenuates necrosis after prolonged ischemia in rabbits. Cardiovasc Res 1998;. 40: p. 502-7.

281.    Hale, S., Dae, MW, Kloner, RA., Hypothermia during reperfusion limits ‘no-reflow’ injury in a rabbit model of acute myocardial infarction. Cardiovasc Res, 2003. 59: p. 715-22.

282.    Koreny, M., et al., Effect of cooling after human cardiac arrest on myocardial infarct size. Resuscitation, 2008.

283.    Magid, D., Wang, Y, Herrin J, et al., Relationship between time of day, day of week, timeliness of reperfusion, and in-hospital mortality for patients with acute ST-segment elevation myocardial infarction. JAMA, 2005. 294: p. 803-12.

284.    Staffey, K.S., et al., Liquid ventilation with perfluorocarbons facilitates resumption of spontaneous circulation in a swine cardiac arrest model. Resuscitation, 2008. 78(1): p. 77-84.

285.    Gentilello, L., Cortes, V, Moujaes, S, Viamonte, M, Malinin, TL, Ho, C, Gomez, GA., Continuous arteriovenous rewarming: experimental results and thermodynamic model simulation of treatment for hypothermia. J Trauma, 1990. 30: p. 1436-49.

286.    Hoedemaekers, C., Ezzahti, M, Gerritsen, A,  van der Hoeven, JG., Comparison of cooling methods to induce and maintain normo- and hypothermia in intensive care unit patients: a prospective intervention study. Critical Care, 2007. 11: p. R91.

287.    Cox, C., Wolfson, MR, Shaffer, TH., Liquid ventilation: a comprehensive overview. Neonatal Netw 1996. 15: p. 31-43.

288.    Shiozaki, T., Hayakata, T, Taneda, M, Nakajima, Y, Hashiguchi, N, Fujimi, S, Nakamori, Y, Tanaka, H, Shimazu, T, Sugimoto, H,  Mild hypothermia study group in Japan., A multicenter prospective randomized controlled trial of the efficacy of mild hypothermia for severely head injured patients with low intracranial pressure. J Neurosurg, 2001. 94: p. 50-4.

289.    Faber, P., Garby. L., Fat content affects heat capacity: a study in mice. Acta Physiol Scand, 1995. 153: p. 185–7.

290.    Sargent , J., Seffl, RJ., Properties of perfluorinated liquids. Fed Proc, 1970. 29: p. 1699-1703.

291.    Simmons, J., Fluorine chemistry, vols I, II, and V., ed. J. Simmons. 1950, New York: New York: Academic Press.

292.    Riess, J.G., Le Blanc, M, Solubility and transport phenomenon in perflurochemicals relevant to blood substitution and other biomedical applications. Pure & Appl.Chem., 1982. 54(12): p. 2383-24O6.

293.    Banks, R., Smart, BE, Tatlow, JC, ed. Organofluorine Chemistry: Principles and Commercial Applications. Topics in Applied Chemistry., ed. R. Banks. 1994, Plenum Press: New York and London.

294.    Bondi, A., van der Waals volume and radii. J Phys Chem, 1964. 68: p. 441-450.

295.    Krafft, M., Reiss, JG., Highly fluorinated amphiphiles and colloidal systems, and their applications in the biomedical field. A contribution. Biochimie. 1998. 80: p. 489-514.

296.    Sargent, J.W., Seffl R J. , Properties of perfluorinated liquids. . Fed Proc 1970. 29: p. 1699-1703.

297.    Lowe, K., Perfluorocarbons as oxygen-transport fluids. Comp Biochem Physiol, 1987. 87: p. 825-38.

298.    Lowe, K., Fluorinated Blood Substitutes and Oxygen Carriers. J Fluorine Chem, 2001. 109: p. 59-65.

299.    Riess, J., Krafft, MP., Advanced fluorocarbon-based systems for oxygen and drug delivery, and diagnosis. Artif Cells Blood Substit Immobil Biotechnol, 1997. 25: p. 43-52.

300.    Schaffer, T., et al., Physiological effects of ventilation with liquid fluorocarbon at controlled temperatures. Undersea Biomed Res, 1984. 3: p. 287-98.

301.    Reiss, J., Oxygen Carriers (“Blood Substitutes”)-Raison d’Etre, Chemistry, and Some Physiology Blut ist ein ganz besondrer Saft. . Chem Rev, 2001. 101: p. 2797-2920.

302.    Geyer, R. Oxygen Carrying Colloidal Blood Substitutes. in Proc Internatl Symp. 1981. Munich: W. Zuckschwerdt Verlag.

303.    Riess, J., Overview of progress in the fluorocarbon approach to in vivo oxygen delivery. Biomater Artif Cells Immobilization Biotechnol, 1992. 20: p. 183-202.

304.    Jeng, M., et al., Perfluorochemical (PFC) Combinations for Acute Lung Injury: An In Vitro and In Vivo Study in Juvenile Rabbits. Pediatr Res, 2003. 53: p. 81-88.

305.    Nagase, S., Banks, RE., Preparation, properties, and industrial applications of organofluorine compounds., R. Banks, Editor. 1982, Ellis Horwood Ltd; Halsted Press: Chichester West Sussex and New York.

306.    Singh, R., Shreeve JM., Perfluoroalkylation of simple inorganic molecules: A one step route to novel perfluoroalkylated compounds. Chem Commun, 2002. 7: p. 1818 – 1819.

307.    Conte, L., Napoli, M, Gambaretto, GP., J. Fluorine Chem, 1985. 30: p. 89.

308.    Shaffer, T., et al., Liquid ventilation. Pediatr Pulmonol, 1992. 14: p. 102-9.

309.    Clark, L., Jr, Hoffmann, RE, Davis, SL., Response of the rabbit lung as a criterion of safety for fluorocarbon breathing and blood substitutes. Biomater Artif Cells Immobilization Biotechnol, 1992. 20(2-4): p. 1085-99.

310.    Lattes, A., Rico-Lattes, I., Microemulsions of perfluorinated and semi-fluorinated compounds. Artif Cells Blood Substit Immobil Biotechnol, 1994. 22(4): p. 1007-18.

311.    Chubb, C., Reversal of the endocrine toxicity of commercially produced perfluorochemical emulsion. Biol Reprod, 1985. 33: p. 854-58.

312.    3-M-Company, 3M Fluorinert liquids for electronic manufacturing: publication 4658 (HB)98-0212-3713-03. 2003: 3M Company, St, Paul, MN.

313.    Kaisers, U., K.P. Kelly, and T. Busch, Liquid ventilation. Br J Anaesth, 2003. 91(1): p. 143-51.

314.    Sass, D., Van Dyke, RA, Wood, EH, Johnson, SA, Didisheim, P., Gas embolism due to intravenous FC 80 liquid fluorocarbon. J Appl Physiol, 1976. 40(5): p. 745-51.

315.    Leakakos, T., Schutt, EG, Cavin, JC, Smith, D, Bradley, JD, Strnat, DA, del Balzo, U, Hazard, DY, Otto, S, Fields, TK, et al., Pulmonary gas trapping differences among animal species in response to intravenous infusion of perfluorocarbon emulsions. Artif Cells Blood Substit Immobil Biotechnol, 1994. 22: p. 1199-204.

316.    Schutt, E., Barber, P, Fields, T, Flaim, S, Horodniak, J, Keipert, P, Kinner, R, Kornbrust, L, Leakakos, T, Pelura, T, et al., Proposed mechanism of pulmonary gas trapping (PGT) following intravenous perfluorocarbon emulsion administration. Artif Cells Blood Substit Immobil Biotechnol, 1994. 22: p. 1205-14.

317.    Burns, M.J., et al., Enhanced mortality from perfluorocarbon administration in a rat model of kerosene aspiration. J Toxicol Clin Toxicol, 1999. 37: p. 855-9.

318.    Mason, K., Withers, HR, Steckel, RJ., Acute effects of a perfluorochemical oxygen carrier on normal tissues of the mouse. Radiat Res, 1985. 104: p. 387-94.

319.    Bucala, R., Kawakami, M, Cerami, A., Cytotoxicity of a perfluorocarbon blood substitute to macrophages in vitro. Science, 1983. 27: p. 965-7.

320.    Edwards, C., Lowe, KC, Röhlke, W, Geister, U, Reuter, P, Meinert , H., Effects of a novel perfluorocarbon emulsion on neutrophil chemiluminescence in human whole blood in vitro. Artif Cells Blood Substit Immobil Biotechnol, 1997. 25: p. 255-60.

321.    Flaim, S., Pharmacokinetics and side effects of perfluorocarbon-based blood substitutes. Artif Cells Blood Substit Immobil Biotechnol, 1994. 22: p. 1043-54.

322.    Wiedemann, H., Partial liquid ventilation for acute respiratory distress syndrome. Clin Chest Med,, 2000. 21: p. 543-54.

323.    Augustin, A., Spitznas, M, Koch, FH, Böker, T, Meller D, Lutz J., Systemic effects of different perfluorochemical agents. Graefes Arch Clin Exp Ophthalmol, 1995. 233: p. 48-51.

324.    van Eeden, S., Klut, ME, Leal, MA, Alexander, J, Zonis, Z, Skippen,  P., Partial liquid ventilation with perfluorocarbon in acute lung injury: light and transmission electron microscopy studies. Am J Respir Cell Mol Biol, 2000. 22: p. 441-50.

325.    Koch, T., et al., Perfluorohexane attenuates proinflammatory and procoagulatory response of activated monocytes and alveolar macrophages. Anesthesiology, 2001. 94(1): p. 101-9.

326.    Rotta, A., Steinhorn, DM., Partial liquid ventilation reduces pulmonary neutrophil accumulation in an experimental model of systemic endotoxemia and acute lung injury. Crit Care Med, 1998. 26(10): p. 1707-15.

327.    Wada, S., Kajihara, H, Murakami, H, Sueda, T, Matsuura ,Y., Effects of FC43 emulsion against hyperacute rejection in rodent discordant xenotransplantation. Heart Lung Transplant, 1995. 14: p. 968-72.

328.    Tanaka, M., [Mechanisms of rejection in guinea pig (GP)-to-rat liver xenotransplantation: improvement of recoloration of GP liver graft and effects of perfluorochemical (FC43) in GP-to-rat liver xenotransplantation][Article in Japanese]. Hokkaido Igaku Zasshi, 1999. 74: p. 441-55.

329.    Gale, S.C., Gorman, G D, Copeland, J G, McDonagh, P F., Perflubron emulsion prevents PMN activation and improves myocardial functional recovery after cold ischemia and reperfusion. J Surg Res, 2007. 138: p. 135-40.

330.    Cicerone, R.J., Atmospheric Carbon Tetrafluoride: A nearly Inert Gas. Science, 1979. 206: p. 59-60.

331.    Yi-Tang, Atmospheric Fate of Various Fluorochemicals. 1993, Massachusetts Institute of Technology Department of Chemistry: Boston.

332.    3M-Company, 3M Fluorinert liquids for electronic manufacturing: publication 4658 (HB)98-0212-3713-03. 2003.

333.    3-M-Company, Century of Innovation: The 3M Story. 2003.

334.    Renner, R., Piecing Together the Perfluroniated Puzzle. Analytical Chemistry News, 2003: p. 1A-40A.

335.    Houde, M., et al., Biological monitoring of polyfluoroalkyl substances: A review. Environ Sci Technol. 2006. 40: p. 3463-73.

336.    Kannan, K., et al., Perfluorooctanesulfonate and related fluorochemicals in human blood from several countries. Environ Sci Technol, 2004. 38(17): p. 4489-95.

337.    Guruge, K.S., et al., Perfluorinated organic compounds in human blood serum and seminal plasma: a study of urban and rural tea worker populations in Sri Lanka. J Environ Monit, 2005. 7: p. 371-7.

338.    Giesy, J.P. and K. Kannan, Global distribution of perfluorooctane sulfonate in wildlife. Environ Sci Technol, 2001. 35(7): p. 1339-42.

339.    Tomy, G., Wski, W, Halldorson, T, Helmm P, Stern, G, Tittlemier, S., Fluorinated organic compounds in an eastern arctic marine food web. Organohalogen Compounds, 2003: p. 323-327.

340.    Tao, L., et al., Perfluorooctanesulfonate and related fluorochemicals in albatrosses, elephant seals, penguins, and polar skuas from the Southern Ocean. Environ Sci Technol, 2006. 40(24): p. 7642-8.

341.    Brown, D., Mayer , CE. and 3M to pare Scotchgard products: one long-lasting compound is cited., in The Washington Post. May 17, 2000. 2000: Washington, D.C.

342.    Fromme, H., et al., Occurrence of perfluorinated substances in an adult German population in southern Bavaria. Int Arch Occup Environ Health, 2007. 80(4): p. 313-9.

343.    Calafat, A., Needham, LL, Kuklenyik, Z, Reidy, JA, Tully, JS, Aguilar-Villalobos, M, Naeher, LP., Perfluorinated chemicals in selected residents of the American continent. Chemosphere, 2006. 63(3): p. 490-6.

344.    Kudo, N., Kawashima, Y., Toxicity and toxicokinetics of perfluorooctanoic acid in humans and animals. J Toxicol Sci, 2003. 28(2): p. 49-57.

345.    Nakayama, S., Harada, K, Inoue, K, Sasaki, K, Seery, B, Saito, N, Koizumi, A., Distributions of perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) in Japan and their toxicities. Environ Sci, 2005. 12(6): p. 293-313.

346.    Karrman, A., et al., Levels of 12 perfluorinated chemicals in pooled australian serum, collected 2002-2003, in relation to age, gender, and region. . Environ Sci Technol, 2006. 40: p. 3742-8.

347.    Kudo, N. and Y. Kawashima, Toxicity and toxicokinetics of perfluorooctanoic acid in humans and animals. J Toxicol Sci, 2003. 28(2): p. 49-57.

348.    Yang, Q., et al., Potent suppression of the adaptive immune response in mice upon dietary exposure to the potent peroxisome proliferator, perfluorooctanoic acid. Int Immunopharmacol, 2002. 2(2-3): p. 389-97.

349.    Yang, Q., et al., Involvement of the peroxisome proliferator-activated receptor alpha in the immunomodulation caused by peroxisome proliferators in mice. Biochem Pharmacol, 2002. 63(10): p. 1893-900.

350.    Peden-Adams, M.M., et al., Suppression of humoral immunity in mice following exposure to perfluorooctane sulfonate. Toxicol Sci, 2008. 104(1): p. 144-54.

351.    Kannan, K., Perrotta, E, Thomas, NJ., Association between perfluorinated compounds and pathological conditions in southern sea otters. Environ Sci Technol, 2006. 40(16): p. 4943-8.

352.    Bagwell, C., Ambroise Pare and the renaissance of surgery. Surg Gynecol Obstet, 1981. 152: p. 350-5.

353.    Dosey, B., Florence Nightingale: Mystic, Visionary, Healer. 2000, New York: Lippincott Williams & Wilkins.

354.    Cannon, W., Traumatic Shock. 1923, New York: Appleton Company, Inc.

355.    Blalock, A., Principles of Surgical Care: Shock and Other Problems. 1940, St. Louis: Mosby.

356.    Parrish, T., The Submarine: A History. 2005, New York: Penguin.

357.    Klystra, J., Lavage of the lung. Acta Physiol Pharmacol Neerl, 1958. 7: p. 163-221.

358.    Klystra, J., Simplified technique of lavage of the lung. Acta Physiol Pharmacol Neerl, 1960. 9: p. 225-39.

359.    Klystra, J., Breathing of pressure oxygenated salt solutions. Chest, 1965. 47: p. 157.

360.    Klystra, J., et al., Lavage of the lung. II. A report on “long term” effects in dogs and suggestions concerning possible modifications of the original technique in order to improve its applicability for the treatment of chronic progressive and fatal diseases of the kidney in man. Acta Physiol Pharmacol Neerl., 1959. 8: p. 326-36.

361.    Costa, E., Amato, M., Maintenance of end-expiratory recruitment with increased respiratory rate after saline-lavage lung injury. J Appl Physiol, 2007. 102: p. 2414.

362.    Brederlsau, J., et al., Combination of arteriovenous extracorporeal lung assist and high-frequency oscillatory ventilation in a porcine model of lavage-induced acute lung injury: a randomized controlled trial. J Trauma, 2007. 62: p. 336-46; discussion 345-6.

363.    Klystra, J., Schoenfisch, X, Maximum expiratory flow and estimated CO2 elimination in liquid-ventilated dogs’ lungs. J. Appl Physiol 1973. 35: p. 117-121.

364.    Clark, L., Introduction to federation proceedings. Fed Proc, 1970. 29: p. 698.

365.    Moskowitz, G., A mechanical respirator for control of liquid breathing. Fed Proc, 1970. 29: p. 1751-2.

366.    Moskowitz, G., et al., Liquid breathing trials and animal studies with a demand-regulated breathing system. Med Instrum, 1975. 9: p. 28-33.

367.    Shaffer, T., Moskowitz, GD., An electromechanical demand regulated liquid breathing system. IEEE Trans Biomed Eng, 1975. 22: p. 24-8.

368.    Shaffer, T., Moskowitz, GD., Demand-controlled liquid ventilation of the lungs. J Appl Physiol, 1974. 36: p. 208-13.

369.    Schaffer, T., et al., The effects of liquid ventilation on cardiopulmonary function in pre-term lambs. Pediatr Res, 1983. 17: p. 303.

370.    Greenspan, J., et al., Liquid ventilation of a preterm baby. Lancet, 1989. 2: p. 1095.

371.    Shaffer, T.H., Moskowitz, G D., Demand-controlled liquid ventilation of the lungs. J Appl Physiol, 1974. 36: p. 208-13.

372.    Furhman, B., et al., Perfluorocarbon-associated gas exchange. Crit Care Med., 1991. 19: p. 712-22.

373.    Marraro, G., Bonati, M, Ferrari, A, Barzaghi, MM, Pagani, C, Bortolotti, A, Galbiati, A, Luchetti, M, Croce, A., Perfluorocarbon broncho-alveolar lavage and liquid ventilation versus saline broncho-alveolar lavage in adult guinea pig experimental model of meconium inhalation. Intensive Care Med, 1998. 24(5): p. 501-8.

374.    Bull, J., et al., Flow limitation in liquid-filled lungs: effects of liquid properties. J Biomech Eng, 2005. 127: p. 630-6.

375.    Suresh, V., Anderson, JC, Grotberg, JB, Hirschl, RB., A mathematical model of alveolar gas exchange in partial liquid ventilation. J Biomech Eng, 2005. 127(1): p. 46-59.

376.    Sekins, K., et al., Long-term partial liquid ventilation (PLV) with perflubron in the near-term baboon neonate. Artif Cells Blood Substit Immobil Biotechnol, 1994. 22: p. 1381-7.

377.    Foust, R., 3rd, et al., Liquid assisted ventilation: an alternative ventilatory strategy for acute meconium aspiration injury. Pediatr Pulmonol, 1996. 21: p. 316-22.

378.    Momoki, Y., et al., Experimental study in partial liquid ventilation for acute respiratory failure after ischemia reperfusion pulmonary injury in a rabbit model. Jpn J Thorac Cardiovasc Surg., 1998. 46: p. 65-70.

379.    Dickson, E., et al., Partial liquid ventilation with perfluorocarbon in the treatment of rats with lethal pneumococcal pneumonia. Anesthesiology, 1998. 88: p. 218-23.

380.    Younger, J., et al., Partial liquid ventilation protects lung during resuscitation from shock. Appl Physiol, 1997. 83: p. 666-70.

381.    Verbrugge, S., et al., Different ventilation strategies affect lung function but do not increase tumor necrosis factor-alpha and prostacyclin production in lavaged rat lungs in vivo. Anesthesiology, 1999. 91: p. 1834-43.

382.    Arnold, J., et al., Prospective, randomized comparison of high-frequency oscillatory ventilation and conventional mechanical ventilation in pediatric respiratory failure. Crit Care Med, 1994. 22: p. 1530-39.

383.    Hirschl, R., Liquid ventilation in the setting of respiratory failure. Asaio J, 1998. 44(3): p. 231-3.

384.    Kacmarekm RM, L., PT, et al., Partial liquid ventilation in adult patients with acute respiratory distress syndrome. Am J Respir Crit Care Med, 2006. 178: p. 882-89.

385.    Lemaire, F., Low- dose perfluorocarbon: A revival for partial liquid ventilation? Crit Care Med, 2007. 35: p. 662-63.

386.    Pound, P., et al., Where is the evidence that animal research benefits humans? BMJ, 2004. 328: p. 514.

387.    ARDSnet, Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volume as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Eng J Med, 2000. 342: p. 1301-08.

388.    ARDSnet, Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. NEM, 2000. 342: p. 1301-08.

389.    Lim, C.M., et al., An optimal dose of perfluorocarbon for respiratory mechanics in partial liquid ventilation for dependent lung-dominant acute lung injury. Chest, 2000. 117: p. 199-204.

390.    Tutuncu, A., et al., Intratracheal perfluorocarbon administration as an aid in the ventilatory management of respiratory distress syndrome. Anesthesiology, 1993. 79: p. 1083-93.

391.    Ricard, J., et al., Perflubron™ dosing affects ventilator-induced lung injury in rats with previous lung injury. Crit Care Med, 2007. 35: p. 561-67.

392.    Ricard, J.D., et al., Perflubron dosing affects ventilator-induced lung injury in rats with previous lung injury. Crit Care Med, 2007. 35(2): p. 561-7.

393.    Ricard, J., Dreyfuss, D, Laissy, J. P., et al., Dose-response effect of perfluorocarbon administration on lung microvascular permeability in rats. Am J Respir Crit Care Med, 2003. 168: p. 1378-82.

394.    Tredici, S., Tredici, F, Brant, DO, Hirschl, RB, Bull, JL., Effect of viscosity on instilled perfluorocarbon distribution in rabbit lungs. J Biomech Eng, 2006. 128: p. 857-61.

395.    Zheng, Y., et al., Effects of inertia and gravity on liquid plug splitting at a bifurcation. J Biomech Eng, 2006. 128(5): p. 707-16.

396.    Hirschl, R., et al., Prospective, randomized, controlled pilot study of partial liquid ventilation in adult acute respiratory distress syndrome. Crit Care Med, 2002. 165: p. 781-87.

397.    Wiedemann, H., et al., A multicenter, randomized, feasibilitystudy of two doses of perflubron administered for partial liquid ventilation in patients with acute respiratory distress syndrome. Abstr. Am J Respir Crit Care Med, 1999. 159:A80.

398.    Griese, M., Griese M. Pulmonary surfactant in health and human lung diseases: state of the art. Eur Respir J, 1999. 13: p. 1455-76.

399.    Rosenzweig, J., Jensen, OE., Capillary-elastic instabilities of liquid-lined lung airways. J Biomech Eng, 2002. 124(6): p. 650-55.

400.    Kamm, R., Schroter, RC., Is airway closure caused by a liquid film instability? Respir Physiol, 1989. 75(2): p. 141-56.

401.    Salman, N., Fuhrman, BP, Steinhorn, DM, et al., Prolonged studies of perfluorocarbon-associated gas exchange and of the resumption of conventional mechanical ventilation. Crit Care Med, 1995. 23: p. 919-24.

402.    Kaisers, U., et al., Partial liquid ventilation with small volumes of FC 3280 increases survival time in experimental ARDS. Eur Respir J 1997. 10.

403.    West, J., Tsukimoto K, Mathieu-Costello, O, et al., Stress failure in pulmonary capillaries. J Appl Physiol, 1991. 70: p. 1731-42.

404.    Dos Santos, C., Slutsky, AS., Invited review: Mechanisms of ventilator-induced lung injury: a perspective. J Appl Physiol, 2000. 89: p. 1645-1655.

405.    Chu, E., Whitehead, T, Slutsky, AS., Effects of cyclic opening and closing at low- and high-volume ventilation on bronchoalveolar lavage cytokines. Crit Care Med, 2004. 32: p. 168-74.

406.    Dreyfuss, D., Saumon, G, Ventilator-induced lung injury: Lessons from experimental studies. Am J Respir Crit Care Med, 1998. 157: p. 294-323.

407.    Hughes, J., Rosenzweig, DY, Kivitz, PB., Site of airway closure in excised dog lungs: histological demonstration. J Appl Physiol, 1970. 29: p. 340-344.

408.    Ghadiali, S.N. and D.P. Gaver, Biomechanics of liquid-epithelium interactions in pulmonary airways. Respir Physiol Neurobiol, 2008. 163(1-3): p. 232-43.

409.    Argiras, E., Blakeley, CR, Dunnill, MS, Otremski, S, Sykes, MK., High PEEP decreases hyaline membrane formation in surfactant deficient lungs. Br J Anaesth 1987. 59: p. 1278-1285.

410.    Gaver, D., Halpern, D, III, Jensen, O, Grotberg, JB., The steady motion of a semi-infinite bubble through a flexible-walled channel. J Fluid Mech, 1996. 319: p. 25-65.

411.    Mead, J., Takishima, T, Leith, D., Stress distribution in lungs: a model of pulmonary elasticity. J Appl Physiol, 1970. 28: p. 596-608.

412.    Yap, D., Gaver, D., The influence of surfactant on two-phase flow in a flexible-walled channel under bulk equilibrium conditions. Phys Fluids, 1998. 10: p. 1846-1863.

413.    Bilek, A., Dee, KC, Gaver, DP, III., Mechanisms of surface-tension-induced epithelial cell damage in a model of pulmonary airway reopening. J Appl Physiol, 2003. 94: p. 770-783.

414.    Naureckas, E., Dawson, CA, Gerber, BS, Gaver, DP, Gerber, HL III, Linehan, JH, Solway, J, and Samsel, RW., Airway reopening pressure in isolated rat lungs. J Appl Physiol, 1994. 76: p. 1372-1377.

415.    Yap, D., Liebkemann, W, Solway, J, Gaver, D., Influences of parenchymal tethering on the reopening of closed pulmonary airways. J Appl Physiol, 1994. 76: p. 2095-2105.

416.    Jensen, O., Horsburgh, MK, Halpern, D, Gaver, DP, III., The steady propagation of a bubble in a flexible-walled channel: asymptotic and computational models. Phys Fluids 2002. 14: p. 443-457.

417.    Vlahakis, N., Hubmayr, RD., Invited review: plasma membrane stress failure in alveolar epithelial cells. J Appl Physiol, 2000. 89: p. 2490-2497.

418.    Doctor, A., et al., High-frequency oscillatory ventilation of the perfluorocarbon-filled lung: preliminary results in an animal model of acute lung injury. Crit Care Med, 1999. 27(11): p. 2500-7.

419.    Rotta, A.T., et al., Combining lung-protective strategies in experimental acute lung injury: The impact of high-frequency partial liquid ventilation. Pediatr Crit Care Med, 2006. 7(6): p. 562-70.

420.    Harris, S.B., New Breakthrough in Rapid Cooling of Cryopreservation Patients after Cardiac Arrest, presented at the Suspended Animation, Inc. Advances in Cryopreservation Conference. 2007: 19 May, Ft. Lauderdale.

Appendix B: Explanation of the Mechanics of Blood Flow during Closed Chest CPR: The Thoracic Pump Theory

The following text, with the accompany references, is reproduced from Weisfeldt, M., Chandra, N., Physiology of cardiopulmonary resuscitation. Ann Rev Med, 1981. 32: p. 43-42.

“At least three general mechanisms are now thought to contribute to the generation of the extrathoracic arterial-venous pressure gradient observed during conventional CPR in the dog. These factors are (a) various venous valving mechanisms are operating, (b) peripheral venous capacitance greater than arterial capacitance, and (c) arterial resistance to collapse greater than venous resistance.

 The most easily understood of these mechanisms is that of a valving mechanism. Veins at the thoracic inlet and other veins leading from the brain appear to have anatomic valves that prevent retrograde flow of blood during increases in intrathoracic pressures (9, 10, 16, 18). The valves along the extrathoracic veins, and the one at the thoracic inlet appear to be important.

 The second factor contributing to the generation of the peripheral arterial-venous pressure gradient is that venous capacitance is greater than arterial. Clearly, if the same amount of blood were to move from the intrathoracic arterial and from the intrathoracic venous systems into the extrathoracic arterial and venous systems, arterial pressure would rise more than venous pressure because of the differences in extrathoracic arterial and venous capacitance.

 The third contributor to the peripheral arterial-venous pressure gradient is the difference in arterial and venous resistance to collapse. Venous structures readily collapse when inside pressures fall below surrounding pressures by even a small amount. Recently we showed that the in vivo carotid artery at the thoracic inlet exhibits considerable intrinsic resistance to collapse.

 This resistance to collapse can be increased in the presence of vasoconstrictor agents (23). Resistance to collapse of the arterial vessels would allow blood flow to continue toward the brain despite surrounding intrathoracic pressures which exceed intravascular pressures. The veins would readily collapse at the exit to the high pressure region, i.e. at the thoracic inlet (1, 13).

 Blood returns from the periphery to the central circulation between compression cycles. Extrathoracic venous pressure rises (20) when blood flows from arteries to veins during compression. Between compressions intrathoracic pressure falls to near atmospheric, and an extrathoracic-to intrathoracic venous pressure gradient appears, which leads to flow into the chest. Right heart and pulmonary blood flow is also diastolic, at least in part (7). With conventional CPR, fight heart compression may be a component of the mechanism for pulmonary flow.”

References

1. Brecher, G. A. 1952. Mechanism of venous flow under different degrees of aspiration. Am. J. Physiol. 169-423.

7. Cohen, J. M., Alderson, P. O., Van AswegenA, ., Chandra,N ., Tsitlik, J.,

Weisfeldt,M .L . 1979.T imingo f intrathoracic blood flow during resuscitation

with high intrathoracic pressure. Circulation. 59 & 60:II-19.

10. Franklin,K . J. 1927.V alvesin veins: an historical survey. In Proc. R. Soc. ed. Sect. History Med., pp. 4-6.

13. Holt, J. P. 1941. The collapse factor in the measurement of venous pressure.

Am. J. Physiol. 134:292.

16. MacKenzie, J. 1894. The venous and liver pulses, and the arhythmie (sic) contraction of the cardiac cavities, d. Pathol. Bacteriol. 2:113.

18. Niemann, J. T., Garner, D., Rosborough, J., Criley, J. M. 1979. The mechanism of blood flow in closed chest cardiopulmonary resuscitation. Circulation 59 & 60:I1-74.

21. Weale,F . E., Rothwell-JacksonR, . L. 1962. The efficiency of cardiac  massage. Lancet 1:990-92.

23. Yin, F. C. P., CohenJ,. M., Tsitlik, J., Weisfeldt, M. L. 1979. Arterial resistance

to collapse: A determinant of peripheral flow resulting from high intrathoracic

pressure. Circulation 59 & 60:I1-196.

Reproduced from the  Annual Reviews www.annualreviews.org/aronline

Annu. Rev. Med. 1981.32:435-442. Downloaded from: arjournals.annualreviews.org by PALCI on 10/25/08.

Experimental Studies to Determine the Effectiveness of LAPC under Laboratory Conditions

Experimental Studies to Determine the Effectiveness of LAPC under Laboratory Conditions

 [This section is an edited version of an article authored by Steven B. Harris, Michael G. Darwin, Sandra R. Russell, Joan M. O’Farrell, Mike Fletcher and Brian Wowk entitled, Rapid (0.5°C/min) minimally invasive induction of hypothermia using ~4ºC perfluorochemical lung lavage in dogs, which first appeared in Resuscitation, 2001. 50: p. 189-204.)]

 1. Introduction

The potential utility of profound and ultra-profound hypothermia (0-5^oC ) to arrest deleterious neurological changes has long been understood in both biology and medicine.[168],[169],[170],[171],[172] In 1959 Benson, et al., reported good outcome using profound hypothermia (10-22^oC ) as a treatment following cardiac arrest. [173] This work was followed up by a number of clinicians [168],[174],[175],[176] who also reported favourable results. However, due to coagulopathy, arrhythmias, and the increased incidence of pneumonia and sepsis associated with such deep and prolonged cooling, post arrest hypothermia failed to gain acceptance and was abandoned. It was not until the work of Safar, et al., [160],[53],[82] that the utility of mild therapeutic hypothermia (MTH) (DT = −2 to −3°C) as an active treatment for the post-resuscitation syndrome was rigorously demonstrated, and subsequently validated by others. [177],[178],[55],[179-181]  As noted in Section 1, while CPB offers the most rapid core cooling possible, it is logistically unsupportable as currently practiced. Additionally, CPB carries the added risks of anticoagulation, further activation of the immune-inflammatory cascade, RBC aggregation, and the danger of gas embolism, as chilled, nitrogen-saturated blood is rapidly re-warmed as it perfuses warm tissues.[182]

As was also previously noted, less invasive modalities with the potential for in-field application, such as surface cooling and lavage of body viscuses with a balanced salt solution, are only effective in achieving cooling rates in the range of 0.10–0.15°C/min. The seemingly straightforward  experimental technique of ‘tidal liquid ventilation’ (TLV) with chilled, oxygenated PFC uses the ~20 m² surface area of the lungs for heat exchange, but thus far has been no more effective in inducing hypothermia than surface cooling with ice bags or chilled water blankets.[155] After preliminary experiments demonstrated the technical adequacy of LAPC at achieving heat exchange in range of 0.25 to 0.35°C/min [183] a comprehensive study was undertaken by 21^st Century Medicine Inc., (21CM) and Critical Care Research, Inc., (CCR), beginning in 1999, to define and validate this cooling modality in a canine model. The goals of this research were to, a) demonstrate the fundamental safety and efficacy of the technique, b) determine the optimum cycle and volume of liquid and gas fractional tidal liquid ventilations (FTLVs), and c ) attempt to determine safe airway pressures and define liquid and gas ventilation strategies that minimized or eliminated baro- and volutrauma.

This technique, developed at 21CM/CCR, was initially called ‘mixed-mode liquid ventilation cooling’ and was later renamed ‘gas-liquid ventilation’ (GLV). However, neither of these names adequately describes the technique, and this author (Darwin) has chosen to the use the term liquid assisted pulmonary cooling (LAPC) instead. In previous studies where large fractional tidal liquid volumes and shorter cycles of FTLV were used, the performance of LAPC deteriorated towards that seen when TLV-cooling (or warming) was used. In practice however, certain significant differences remained and understanding these differences proved essential to optimization of the technique.

In LAPC the critical elements of gas ventilation are retained allowing for flexibility in selecting ventilation parameters independently for heat and gas-exchange, allowing for liquid-mediated heat-exchange to be easily undertaken using existing ventilation systems[1]. The combination of gas and liquid FTLVs may also play a role in the surprisingly good thermal efficiency of LAPC as compared with TLV.

The following study explored the performance of LAPC using a prototype automated FTLV device, and discussed the basic mechanics and intrinsic limitations of heat-exchange using FTLV.

2. Materials and methods

These experiments were approved by 21st Century Medicine, Inc.’s Institutional Animal Care and Use Committee and were in compliance with the Animal Welfare Act and the National Research Council’s Guide for the Care and Use of Laboratory Animals. Fifteen mongrel dogs weighing 13.8–25.7 kg were used (Table 1). Dogs were pre-medicated with I.M. acepromazine (1.0 mg/kg) and atropine (0.02 mg/kg) prior to induction of general anesthesia using sodium pentobarbital (30 mg/kg I.V., with maintenance dosing). Anesthetized dogs were intubated with a reinforced 10.0 mm I.D. (Willy Rusch AG, Kernen, Germany) endotracheal tube (E.T.), and ventilated on room air using a Bennett MA1 or Siemens Servo 900 C ventilator. Ventilator parameters, unless otherwise noted, were 12 gas-breaths/min, gas tidal-volume of 15 ml/kg, I:E ratio of 1:3, and a maximal positive inspiratory pressure (PIP) limit of 26 cm H₂O (2.5 kPa). Gas pressures were measured at the E.T. adapter. Gas minute-volume (V_g) was adjusted to maintain PaCO₂ between 35 and 40 torr. Animals were maintained at ~37.5°C prior to LAPC, using a temperature-controlled water blanket. Rectal and bilateral tympanic temperatures (Ttym) were monitored continuously using a type-T thermocouple system (Cole-Parmer, Vernon Hills, IL) with a response time constant (to) of 5 s.

Combination pressure, blood sampling, and temperature-probe catheters were constructed from rigid polyethylene pressure-monitoring catheters, threaded centrally with 0.05 in. O.D. Teflon™-sheathed type-T thermocouples (to=0.3 s, Physitemp Instruments, Clifton, NJ). In order to reduce the risk of catheter-associated clot formation, I.V. sodium heparin was given to adjust activated clotting times to 300–500 s, prior to central line placement. Femoral vessels were isolated surgically, and arterial and venous catheters placed and advanced to a level above the renal vessels, as confirmed by X-ray. During surgery, bupivacaine (0.5%) was infiltrated into wounds to mitigate post-operative pain.

In one dog (Trial I-2), a femorally-placed pulmonary artery thermodilution catheter replaced the venous combination catheter. Blood and ventilator pressures were acquired through a Hewlett Packard 78532-B monitor/transducer system.

Immediately prior to LAPC, dogs were assessed for adequacy of general anesthesia, and then given pancuronium bromide (2 mg) to inhibit shivering and spontaneous breathing. FIO₂ was increased to 100% and external temperature control discontinued. To serve as a cannula for both delivery and removal of PFC liquid, a 19-Fr. flat-wire reinforced Bio-Medicus® venous catheter (Medtronic, Eden Prairie, MN) was introduced through the suction port of the E.T. adapter, and advanced ~45 cm to approximately the level of the carina (Figures 2-1 and 2-2) as confirmed by X-ray. This cannula was connected to the LAPC apparatus described below. LAPC was performed using the PFC liquid ‘FC-75’ (3M Corporation, St. Paul, MN), a perfluorinated butyl-tetrahydrofuran isomer mixture.[184]

Figure 2-1 (right): PFC delivery and withdrawal catheter threaded through the endotracheal tube with the tip positioned at the level of the carina.

A two reservoir circuit (Figure 2-3) was used to deliver and remove PFC from the lungs (FTLV) via the cannula, in cycle periods of 37 s (Trial I) or 16 s (Trial II). During timed PFC infusions (t_in=20 s for Trial I, or 10 s for Trial II), PFC was pumped through the cannula by a continuously-operating Travenol CPB roller-pump (Sarns, Ann Arbor, MI). A bypass loop, open during suction, allowed the roller-pump to divert (recirculate) PFC flow back into the storage reservoir whenever flow was not directed by line clamps V1–V3 into the animal. PFC was pumped continuously through an in-line 0.2 μ ‘pre-bypass’ filter (Pall PP 3802, Pall, East Hills, NY), a primary heat-exchanger (Torpedo-T, Sarns, Ann Arbor, MI), and a combination silicone membrane oxygenator/heat-exchanger (SciMed II-SM35, SciMed Life Systems, Minneapolis, MN). The oxygenator was supplied with 5–6.5 l/min O₂ (maximal device design rate), and the reservoir PFC was allowed to circulate and equilibrate with heat-exchangers and O₂, before LAPC was initiated. The circuit tubing was constructed of S-50 HL TYGON® 3/8 and 1/2 in. I.D. class VI tubing (Norton/Performance Plastics, Akron, OH) with the exception of a length of silicone tubing (Masterflex® 96410-73, Barrant Co., Barrington, IL) used in the roller-pump head in order to allow flexibility at low temperatures. PFC suction was driven by a vacuum pump (model 107CAB18B, Thomas Compressors, Sheboygan, WI), and suction reservoir negative pressure was limited to −35 torr by a vacuum relief valve. Figure 2-2: Detail of LAPC PFC introduction and removal catheter and ET Tube and gas ventilator configuration. A Biomedicus CPB venous return catheter was threaded through the suction port of a standard 16 mm respiratory ET tube swivel connector.

Figure 2-3: The LAPC system. The LAPC system was connected to a catheter inserted into the suction port of the E.T. adapter. PFC flows were directed by manual or mechanical clamps at V1–3. During the suction phase, FC from the lungs was removed into a sealed ‘suction’ reservoir, for later addition to the primary circuit (via adjustment of V4 and V5), while ‘infusion ready’ PFC was re-circulated through a bypass loop. Negative pressure was limited by a vacuum relief valve (VrV). Photo (right) A) Suction Reservoir, B) Storage Reservoir, C) Solenoid Valves (V1-V3), D) SciMed Oxygenator & Heat Exchanger, E)Sarns Roller Pump, F) PFC Suction/Delivery Catheter, G) Pump Controller, H) Heat Exchanger Return Line with weighted water diffuser (yellow), I) Thermocouple Probes.

Figure 2-4: Gas ventilator and respiratory monitoring equipment used in the LAPC experiments; a) Novametrix CO₂SMO respiratory function monitor and capnograph, b) Siemens Servo 900 C ventilator, c) Korr Medical, Inc., automated device used to perform rapid-cycle LAPC, d) LAPC apparatus.

  Concurrent FC-75 FTLV and gas ventilation (LAPC) was performed for 18 min in Trials I and II (n=12). This time was chosen, on the basis of preliminary work (data not shown), to achieve rapid systemic-cooling of greater than 5°C. For Trials I and II, the PFC recirculation rate within the LAPC device (=PFC infusion rate, ˙V _inf), was set at 50 ml/kg per min rate, V_FTLV) was set at 50 ml/kg per min.

Immediately after a timed FTLV, PFC was removed as rapidly as possible. Infusion of PFC for the next FTLV began immediately after suction was discontinued. In LAPC experiments, PFC was chilled to ~4°C prior to FTLV (Table 1), whereas in normothermic (control) dogs, isothermic PFC was delivered to the dog within ~2°C of tympanic temperature (T_tym). The PFC inflow and outflow temperature was measured continuously by a thermocouple inserted into the PFC path at the base of the delivery/removal cannula. Temperature data was collected throughout LAPC, and for 22 min after LAPC was completed. Arterial blood gas (ABG) samples were taken from the femoral arterial line before the start of LAPC, and every 2 min during LAPC. Following the post-LAPC equilibration period, monitoring devices were removed and incisions closed.

Table 1:

Table 2:

2.3. Trial I (manually-controlled LAPC)

Trial I was designed to investigate the variability in the response of individual animals to LAPC and to investigate the physiological effects of the LAPC technique with and without cooling (i.e., ~4ºC PFC vs. isothermic PFC FTLV). Either isothermic (near-body temperature) or ~4ºC PFC FTLV was administered using a manually-controlled system (V1–V5 in Fig. 1 represent CPB tubing-occluders in this Trial). One FTLV (period t_c ~37 s) was composed of a timed FTLV (t_inf = 20 s), followed by PFC suction (t_s ~17 s). Suction was stopped when PFC liquid return became sparse, or gas pressure in the ventilator circuit fell below −5 cm H2O (−0.5 kPa). Five dogs received ~4ºC FTLV (Trial I-1–5), while two controls received the same protocol using isothermic FTLV (Trial I-6 and 7).

 2.4. Trial II (machine-controlled LAPC)

Trial II assessed the utility of using an automated device (custom manufactured by Korr Medical, Inc., Salt Lake City, UT) to perform rapid-cycle LAPC. Computer-controlled solenoid clamp-valve occlusion of circuit lines at V1–V3 allowed smaller FTLV volumes (V_FTLV) and smaller t_c. While t_inf was decreased to 10 s in Trial II, V_FTLV remained constant, and the effective PFC FTLV rate (V_FTLV) remained in the range of V_FTLVfor Trial I. Table 1 gives relevant trial parameters. In Trial II, suction of PFC from the lungs began immediately after infusion, and was automatically stopped whenever a ventilator circuit pressure of −5 cm H₂O was reached (t_s ~6 s, giving t_c ~16 s). Three dogs received ~4ºC PFC (Trial II-1–3), while two controls (Trial II-4 and 5) received isothermic PFC.

 2.5. Animals A, B and C

 Selected data from three dogs in an earlier method development series was used. These dogs had been prepared as above, then manually given 1, 15 and 21 FTLVs, respectively with ~4ºC PFC, at much slower rates than in Trials I and II (Table 1). Data from these animals allowed independent measurements of FTLV volume heat-contents and temperatures, and thus the heat capacities and heat transfer efficiencies, by a more thorough thermal accounting method (Table 2, Appendix A).

2.6. Data collection and correction, statistical methods, graphical display and presentation

 Temperature and pressure data were collected using a PCI E series data acquisition board and LabVIEW™ software (National Instruments, Austin, TX). Graphical analysis and display of temperature data, and curve fitting, was done using the software package Origin™ (Microcal Software, Northhampton, MA). Statistical comparison of Trial group values was done using GraphPad Prism (GraphPad Software, San Diego, CA). Group means are reported ± standard

Figure 2-5 (above): Body temperature changes observed during LAPC (Method of Trial I). In this illustrative experiment from Trial I (I-4), FTLVs of ~4ºC FC-75 were infused (~20 s) and removed (~17 s) from the lungs. LAPC was performed for 18 min (hatched bar), then stopped to allow thermal equilibration (22 min). Arterial temperature (˜Tart), central venous temperature (˜T_ven), tympanic temperature (˜T_tym), and rectal temperature (˜T_rec) are shown. Inset: Enlarged view of temperature changes recorded during the first two cycles of PFC infusion (gray bar) and removal (yellow bar).

 deviation (SD) except as otherwise noted. For each animal, the T_tym from whichever probe cooled most rapidly, was used (right probe in 12/15 dogs). In order to facilitate comparison of cooling rates between sites in the same animal, temperatures at all probe sites were corrected to the baseline aortic temperature (T_art), as measured immediately prior to the start of LAPC. For ease of description, LAPC-cooling is presented in terms of thermal-deficit (‘cold’) moving from the lungs into successive body compartments. A compartmental analysis of thermal transfer in this model, and a glossary of notation and equations used, is given in Appendix A.

 3. Results

 LAPC allowed FTLV of dogs during concurrent gas ventilation. Suction from the submerged catheter tip at the carina allowed collection of PFC even during forced gas inspiration. It was discovered that a long suction catheter was necessary to insure that adequate suction pressure could be used to withdraw PFC throughout the liquid removal sequence, without prolonged exposure of the gas filled portion of the airways to the negative pressure of the suction system/reservoir. Additional protection of the airways against excessive negative pressure during the relatively brief time after liquid no longer filled the suction line was provided by incorporating a negative pressure relief valve on the suction reservoir. Suction in this manner was efficient, although FTLV volume measurements showed that the lungs retained ~12 ml/kg PFC (approximately FRC) between FTLVs.

The PFC pump circulation/infusion rate (˙V _inf), measured volumetrically preceding and following LAPC, was stable to within 1% over the duration of LAPC, and was not significantly different between trials (P = 0.28). The V_FTLV, calculated as tinf  _inf/t_c, was 30.7±2.3 ml/kg per min (Trial I) and 36.4 ± 3.2 ml/kg per min (Trial II). The ˙V_FTLV was significantly (P = 0.023) larger in Trial II because machine-controlled suction made more efficient use of available non-infusion time, resulting in faster net PFC removal.

Figure 2-6: Thermal equilibration after LAPC. Mean Tart and T_ven values (Fig. 2) are shown for Trial I, dogs 1–5. To highlight equilibration changes, T_tart curve nadirs (n=5) were superimposed before calculation of means, and T_ven data (n = 4) for each dog was adjusted with its corresponding Tart curve. Incompatible T_ven data from a pulmonary artery thermodilution catheter in I-2 has been omitted. Inset: The sigmoidal mean (N = 5) T_art recovery during the first ~12 s after final LAPC. FTLV is approximated by linear fitting.

Figure 2-7: Body temperature changes during manual and mechanical LAPC (Trial I vs. Trial II). The relative rates of core body cooling in dogs undergoing 18 min (hatched region) of manual (Trial I, solid squares) or machine-driven (Trial II, open circles) ~4ºC LAPC, were assessed by comparing changes in group mean T_tym. Symbols represent the mean and SEM (n = 5 for manual, and 3 for machine groups).

 3.1. Thermal results of LAPC

3.1.1. Cooling time delay

 Figure 2-5 illustrates LAPC cooling in a representative dog (I-4) from Trial I. The T_art began to decrease 3–6 s after the start of each PFC FTLV. Since this delay included circulation delay from lungs to aorta, the transfer of thermal-deficit from newly-introduced PFC to pulmonary blood was very rapid. The venous temperature (T_ven) began to decrease 10.4 ± 6.9 s after T_art decline, representing the minimum systemic circulation time. Though exhibiting delay, damping, and broadening behavior (presumably due to peripheral heat-exchange and varying systemic blood-return path lengths), T_ven transients from FTLVs mirrored T_art transients. T_tym temperatures, presumably reflecting brain and viscera temperatures, were non-oscillatory. The T_tym did not begin to decrease until ~24 s after the start of LAPC. This decrease occurred in three phases: an initial phase lasting ~100 s, an exponential phase lasting for ~900 s, and a final linearly-decreasing phase lasting until the end of LAPC. Core cooling as measured by T_tym continued for about 120 s after the end of LAPC (Figure 2-5), then exhibited a marked rebound effect [185] with exponential dampening (t ~20 min, Figures 2-5–2-7). These phases of cooling and equilibration were consistent with a five-compartment thermal model, in which the three compartments representing animal tissues corresponded roughly with (1) the blood and vasculature; (2) the classical thermal core; and (3) the classical thermal periphery (Figure 2-8). Modelling equations and estimation of compartment sizes are given in the Appendix A.

3.1.2. Cooling rate

 Crude cooling rates were determined numerically from appropriate T vs. t graph segments. The mean cooling rate from LAPC initiation, or DT_tym/Dt, reached a maximum value in Trial I at −0.49±0.09°C/min (t=6.6 min). The differential cooling rate d (DT_tym)/ dt = dT_tym/dt reached a maximum (max) value of -0.59 ± 13°C/min at t ~100 s, near the end of the initial heat exchange development region. (This value is comparable to analytic d (DT_tym)/dt  (max) from (Eq. (1)) = DT_k/t_o  = −0.63°C/min). Corresponding cooling rates in Trial II were DT_tym / Dt (max) = −0.33 ± 0.02°C/min (at t = 7.3 min) and dT_tym / d_t (max) = −0.37 ± 0.06°C/min (at t = 100 s).

3.1.3. Mean cooling power

 The mean heat removal rate (cooling-power) P over the entire duration of LAPC, for each animal, was estimated from DT_e according to P = m C_m DT_e/t (total).

Here t (total) is the entire LAPC application time = ~1080 s. (Note: for this calculation, the more accurate Trial I mean C_m is used for all Trial II animals.) The mean cooling power of Trial I was 336 ± 60 watts, while that of Trial II (using the Trial I value of C_m) was 207 ± 49 watts (P = 0.02). Variation in animal size was the major source of intra-group variability.

Figure 2-8: Heat transfer among body compartments during LAPC. Heat transfer during LAPC in the dog may be modeled using 5 thermal compartments. Heat transfer between compartments (which is by blood circulation, except as noted) is shown in the box diagram as double-headed arrows. The pair of arrows connecting Compartments 2 and 3 represents the different processes of lung equilibration with (1) pulmonary artery flow; and (2) with the complete blood volume and selected viscera.

 3.2. Gas exchange

 ABG measurements demonstrated that infusion of ~4ºC PFC stabilized PaO₂ and PaCO₂ during LAPC. In contrast, LAPC using isothermic PFC failed to maintain baseline PaO₂ or PaCO₂ levels (Figure 2-9). In Trial II-4, hypercarbia during the first 13 min of isothermic LAPC was abolished by increasing the tidal volume from 15 to 25 ml/kg (final V_g = 375 ml/kg per min). In Trial II-5, V_g was pre-set to 375 ml/kg per min in an attempt to avoid hypercarbia, and no significant ABG changes were observed.

Figure 2-9: LAPC does not maintain normocarbia at isothermic temperatures without alteration of the gas ventilation parameters. Animals in Trials I and II underwent LAPC using either ~4ºC (˜) or isothermic (r£ ™¯) FC-75. Both arterial PaO₂ (Panel A) and PaCO₂ (Panel B) levels were affected by FTLV temperature. ~4ºC FTLV data from Trials I and II were very similar in magnitude, and therefore, have been combined (n=8). Isothermic LAPC is shown as four separate experiments (Trial I-6 and 7, and Trial II-4 and 5). Gas tidal volume was increased from 15 to 25 ml/kg in Trial II-4 at t=13 min, and at t=0 in Trial II-5, normalizing PaO₂ and PaCO₂ in both animals. Declining PaO₂ in Trial I-7 was due to inadvertent failure to pre-oxygenate PFC.

Figure 2-10: Effect of LAPC on VCO₂ and EtCO₂ during 20 minutes of ~4ºC FTLV. LAPC allows superior CO₂ removal due to gas ventilation because ^·V _PFC = no more than 30 mL/kg/min of liquid, allowing gas ventilation of at least 200 mL/kg/min, resulting in a maximum ^·VCO₂ removal rate of  >8 mL/kg/min or a minimum of 400% of basal metabolic rate.

3.3. Clinical observations and gross pathology

 With the exception of one dog, animals subjected to LAPC displayed mild tachypnea and increased expiratory sounds, but otherwise exhibited unremarkable recovery from anesthesia, including the ability to walk and drink. The exception was an eosinophilic animal (Trial II-1) which had normal oxygenation during LAPC, but developed severe hypoxemia shortly after LAPC.

Chest X-ray pre- and post-procedure showed no (comparatively) remarkable features. This dog was sacrificed at 9 h.

Necropsy revealed a mass of D. immitis (heart- worm) embolized into the pulmonary arterial circulation, possibly as a result of local chilling of the parasite mass due to LAPC (this animal had been heartworm seronegative). Necropsies performed on nine remaining Trial I and II animals sacrificed 24 h post-procedure revealed diffuse spongy, resilient hyperinflated non-collapsible lungs (HNCL) seen in animals exposed to a high-vapor pressure PFC at high PIP pressures.[186]  HNCL was most prominent in the anterior, least dependent areas of the lung lobes. This trapped intra-alveolar PFC was thought to be the cause of broncho-constriction and wheezing found in post-LAPC animals. There was also evidence gross, dependent-lung damage evidenced by pulmonary edema with consolidation in both isothermic and ~4^oC PFC-FTLV animals.

Other organ systems in this series were grossly normal. Two animals in Trial II (II-3 and II-4) were not sacrificed, and were held for long term evaluation. They were neurologically normal at 1 year post-LAPC.

3.4 Impact on Hemodynamics

 FTLV with both ~4ºC and isothermic PFC resulted in an almost immediate modest decrease central venous pressure (CVP) which persisted for the duration of the FTLV and recovered to pre-FTLV values at the conclusion of each FTLV cycle (Figure 2-11 – 2-13).

Figure 2-11: Impact of LAPC on HR.

In animals subjected to FTLV with ~4ºC PFC there was an immediate, transient reduction in heart rate (HR) and mean arterial blood pressure (MAP) and a corresponding increase CVP during the FTLV cycle. In animals undergoing ~4ºC FTLV these effects could be attributed to the acute, cyclical chilling of the coronary blood supply during each FTLV with chilled PFC. The temperature of the blood entering the coronary os was 10^o to 15^oC colder than systemic blood (as measured by pulmonary artery catheter), and this would be expected to have an immediate depressive effect on myocardial contractility due to the transient hypothermia the myocardium would experience as a consequence of perfusion with chilled blood. In fact, consonant with this interpretation, HR decreased steadily during ~4ºC FTLV, recovering progressively less after each FTLV cycle, as systemic hypothermia was induced (Figure 21).

Figure 2-12: Cyclical variation in CVP is response to FTLVs with isothermal PFC.

Figure 2-13: Effect of FTLV with ~4ºC PFC on MAP and CVP over the course of 6 minutes of LAPC.  MAP is transiently markedly depressed and CVP is concurrently increased in response to loading with ~4ºC PFC; this effect is reversed when the PFC load is suctioned from the lungs; although there is increasing depression of MAP in response to the induction of systemic hypothermia. Cardiac output (CO) was not measured in these studies and mathematical analysis of the MAP waveforms generated during isothermic FTLV were not done. Thus, the precise extent to which FTLV with isothermic PFC (i.e., without the thermal-metabolic effects of chilled blood on the myocardium as occurs in LAPC) impairs cardiac output or coronary perfusion pressure is unknown.

Interestingly, the cyclical increase and recovery of the CVP remained constant during both ~4ºC and isothermic FTLV. Initially, the effect of FTLV on CVP was thought to be due to compression of the thoracic vena cavae by the relatively dense PFC load in the lungs. It was hypothesized that this effect might be more pronounced in the dog due to the V-shape of the canine thorax with the cavae resting at the bottom of the thoracic ‘trough’ in a dependent position under the lungs. To test this hypothesis, isothermic FTLV was carried out with a dog in the prone position. Pronation had no effect on the transient, cyclical depression of HR and increase in the CVP associated with FTLV. It thus seems possible that loading of the lungs with dense PFC liquid results in increased pressure on the thoracic vasculature, in particular on the thoracic venous vasculature, in much the same way gas PEEP reduces thoracic venous capacitance and raises CVP; reducing right ventricular preload and right ventricular output (cardiac output). These effects would seem the most likely explanation for the reduction in MAP observed during maximal PFC loading during both ~4ºC and isothermic FTLV.

CO was not measured during these LAPC studies nor was mathematical analysis of the aortic pressure waveforms undertaken to determine with precision the degree to which FTLV depressed MAP. Crude analysis of MAP during isothermic FTLV suggests that cyclical PFC loading is responsible for ~15-20% reduction in MAP over baseline.  Mean CVP is increased ~35% over basal levels during FTLV. To what extent CO will be impacted as a result of FTLV with PFC during CPR will have to be determined experimentally (see discussion in 4.5.3. Overcoming Increased Intrathoracic Pressure and Preserving CO, below).

4. Discussion

 4.1. Apparent effect of temperature on gas exchange

 Isothermic LAPC in this model was surprisingly poor at removing CO₂, considering that the CO₂ carrying capacity in FC-75 decreases by only ~23% from 0 to 40°C (extrapolated from [184]). A useful observation was that pO₂ values decreased even in isothermic animals, indicating an influence on total ventilation and possibly also reflecting decreased CO as evidenced by the (average) decline in MAP and increase in systemic vascular resistance during LAPC.

Capnographic analysis of LAPC in Trials I and II (data not shown) indicated that isothermic LAPC had a much larger negative effect on pressure-limited total gas ventilation V_g, as compared to ~4ºC LAPC using the same technique and the same gas ventilator settings. Since LAPC at a ˙V_FTLVr of 30–36 ml/kg per min relies on gas ventilation V_g for ~50% of total alveolar ventilation, a differential loss of pressure-limited V_g with temperature appeared to be the basis of CO₂ retention in isothermic LAPC. The mechanism of the implied differential change in lung compliance may be related to the depressive effects of FTLV on CO, and presumably, on perfusion. Thus, gas ventilation adjustments similar to those in Trial II-4 and 5 may be required if LAPC is used as a re-warming technique, and it may additionally be necessary to abolish gas PEEP, or even apply continuous negative airway pressure [187],[188] to counteract the PEEP-like effects of PCF loading and to generally improve CO and coronary perfusion pressure (CPP) during CPR.[189]

4.2. Thermal transfer efficiency and kinetics

 The optimal LAPC cooling (or warming) protocol remains unknown. However, the finding that the thermal equilibration of non-dead space PFC and local pulmonary blood flow proceeds very rapidly (t_o <12 s) suggests that FTLV infusion times need to be no longer than this time scale. When PFC lung dwell times exceed this duration, the FTLV load is in place longer than is required to transfer the most labile part of its thermal potential to the pulmonary blood and parenchyma.

Figure 2-14: Relative insensitivity of PFC dwell time to heat exchange was demonstrated by progressively shortening the duration of FTLVs and increasing their frequency. Even at the maximum achievable rate of FTLV at ˙V_FTLV  rates of 30 ml/kg per min and ˙V_FTLV  rates of 50 ml/kg per min no deterioration in the efficiency of heat exchange was observed. This suggests that thermal equilibration at the level of the alveolus is practically instantaneous and that the shortest FTLV dwell time should be used for optimum heat exchange.

Since FTLV ventilation rates (˙V_FTLVr) in the present study are already at least a third of the maximal rates possible in TLV, it seems probable that PFC infusion rates and pressures, rather than heat transfer rates from PFC to lung, will be the fundamentally limiting factor to power transfer in LAPC. These observations suggest that, as least to ˙V_{FTLV r} rates of 30 ml/kg per min and ˙V_FTLV  rates of 50 ml/kg per min, the total cooling power (cooling rate) in LAPC will be greatest if no PFC dwell time is allowed, and all available time during the FTLV cycle is used to either introduce, or remove, PFC.

4.3. Question of diffusion dead space in LAPC

 Mammalian lungs depend on simple gas diffusion for CO₂ transport through the acinar airways during normal tidal ventilation. An intractable problem in experimental TLV has been that simple diffusion is not sufficient to similarly move CO₂ through liquid PFC at physiologic CO₂ partial pressure gradients. This limitation appears in TLV as a ‘CO₂ diffusion dead space’ which effectively lowers alveolar ventilation. In part due to such extra physiologic dead space, TLV of adult humans has been estimated to require liquid minute-volumes near 70 ml/kg per min.[190] This value is at the upper bound of realistically attainable liquid flow rates [191],[192] and leaves little leeway for treating hypercarbia, hypermetabolic states, or lung disease. Such difficulties are not a theoretical limitation in LAPC, however, since LAPC does not require high liquid flow rates for ventilation. In the most rapid-cooling LAPC protocol used in this trial, ˙V_FTLV was 31 ml/kg per min—a low baseline value which permitted the addition of 10 times this minute-volume of gas ventilation (see Fig. 4, Trial II-4, 5). Moreover, since normal gas minute volumes were required to maintain normocapnia in Trial I, there is as yet no evidence for any CO₂ diffusion limitations caused by intrapulmonary PFC in LAPC. Possible reasons for this are discussed below.

Thermal-diffusion limits in TLV have not been studied per se, but their presence is suggested by the results of Shaffer and co-workers.[155] In their cat TLV model using a ˙V _FTLV of 75 ml/kg per min, a decrease in PFC inspiration temperature from 20 to 10°C (increasing the thermal gradient by a factor of 1.6) increased the cooling rate from −0.13 to −0.15°C/min. This small rate change represented a significant loss of efficiency. By contrast, in the present LAPC study using PFC at 4°C, there was no evidence of a thermal-diffusion limit at rates up to 4 FTLVs/min. Notably, in Trial I, where 100% of the V_FTLVr, and 40% of the ˙V_FTLVr of the cat TLV model was used, cooling rates for LAPC were more than three times those reported for cats subjected to TLV at 4.5 liquid breaths/min at 10°C.[155]

The possible quantitative presence of a thermal diffusion limit for LAPC at 4 FTLVs/min may be evaluated using a modified version of the concept of gas-exchange dead space (V_D). The respiratory system of a dog undergoing LAPC heat-exchange may be considered, by analogy with gas exchange dead space (V_D), to also contain a ‘thermal exchange dead space’ (V_Dtherm). Each thermal FTLV volume ˙V_FTLV of PFC then also contains a V_Dtherm , which by definition does not participate in heat-exchange. Thus, cycle thermal transfer efficiency E_f may be expressed as (V_FTLVr − V_Dtherm)  / ˙V_FTLV, and any measured value of mean E_f may be expressed as an equivalent mean V_Dtherm = ˙V_FTLVr  (1− E_f). For Trial I (E_f = 0.6, Appendix A for calculation), mean V_Dtherm was then seen to be 7.5±1.6 ml/kg, and in Trial II (using Eq. (6) E_f  Value = 0.40), V_Dtherm was 5.3±0.8 ml/kg (P = 0.072). The absence of an increase in V_Dtherm in Trial II vs. Trial I indicated that the size of V_Dtherm in these LAPC protocols was non-dynamic at time-scales of one FTLV cycle, providing evidence against the presence of a ‘thermal diffusion dead space’ (analogous to a CO₂ diffusion dead space) at these FTLV rates. In absolute terms, it may be useful to compare calculated V_Dtherm in the LAPC dog model to the expected physiologic gas-exchange dead space, VDCA, which in healthy animals is close to the dog anatomic VD = ~6.5 ml/kg.[193]

In thermal diffusion, as in gas diffusion, diffusion physiologic dead space would be expected to significantly add to anatomical dead space. However, the sum of mechanical-VD in the LAPC circuit (~1.5 ml/kg) plus the anatomic VD for dogs is found to be more than the calculated V_Dtherm in either trial in this study, leaving little room for a large heat-diffusion contribution to V_Dtherm. For these reasons it is suggested that the loss of cooling power observed in Trial II was not due to heat diffusion limitations, but instead due to a loss of efficiency effect similar to that seen with low tidal volumes in ordinary gas ventilation. In these terms, low FTLV volumes in LAPC result in an increase in ‘thermal dead space ventilation’ at the expense of PFC flow involved in active heat-exchange, resulting in a larger ‘wasted’ FTLV V_Dtherm / ˙V_FTLV. VD in heat transfer (V_Dtherm) that is analogous to VD in gas-transfer; in as much as all dead space is ‘diffusion dead space’ at long-enough time-scales. However, some of the mechanisms for diffusion modification of V_Dtherm are unique. By contrast with gas molecules, heat diffuses rapidly through device tubing into the PFC in the LAPC circuit dead space, and heat also diffuses directly through the tracheal wall into the anatomic-VD. Thus, heat diffusion from dead space liquid at sufficiently slow FTLV rates might be expected to have a pronounced effect on E_f in LAPC, due to slow heat-diffusion reduction in V_Dtherm.

Some evidence for such a process was found, though at FTLV dwell times too long to be of interest for rapid cooling. At the relatively small tc of Trials I and II, the calculated V_Dtherm was found to be ~VDCA; but in animal B, with a much longer t_c of 7 min, the V_Dtherm was only 2.6 ml/kg. The limit of this process was reached in animal A, in which the V_Dtherm of a single retained ‘breath’ of highly-oxygenated PFC fell to nearly zero after 10 min. Disappearance of V_Dtherm by thermal equilibration, estimated from individual cycle E_f variations in animals B and C, was estimated to occur with a half-time of ~5 min (data not shown).

This process was slow enough to be neglected when the duration of FTLV intervals (t_c) was less than several minutes.

Thus, at the FTLV rates of Trials I and II, a full-sized V_Dtherm of ~6 ml/kg appeared, and accounted for significant loss of cooling power at low ˙V_FTLV (e.g. Trial II where V_FTLV was only 8.8 ml/kg). The characteristic size of V_Dtherm at all but the slowest FTLV rates (~1 FTLV per 5 min) implies that the only thermally-efficient solution for performing LAPC at faster rates is maintenance of [˙V_FTLVr  / VDCA] or [˙V_FTLVr / V_Dtherm] ratios >3, in order to avoid excessive ‘wasted’

V_Dtherm  ventilation. This requires a ˙V_FTLV of ~18 ml/kg in dogs. In humans, where the anatomic-VD is <3 ml/kg, less than half the value for dogs, both the V_Dtherm, and therefore, most-efficient ˙V_FTLV values, might also be expected to be correspondingly less. In any case, it is clear that rapid-cooling LAPC techniques cannot wait for the relatively slow thermal equilibration of PFC within the anatomical VD, since equilibration in the remaining non-VD parts of the lung is so rapid (i.e. less than Trial II t_c of 16 s).

4.3.2. Possible synergy of combined gas and liquid ventilation in assisting mass (CO₂) and heat transfer

 The absence of expected heat-diffusion and gas-diffusion limitations in LAPC suggests that some assistive process for both gas and heat transfer through PFC in the peripheral lung may occur in LAPC. The authors’ fluoroscopic observations (made with the non-brominated and relatively radiolucent FC-75) have been that each gas breath in PLV produces a flash of fine bubbles which spread uniformly throughout the lung. As compared to the more familiar behavior of water, the low surface tension of PFCs (15 dyne-cm for FC-75, about 1/5th that of water) lowers the energy barrier to producing small bubbles in forced gas/liquid flows. Such

bubbles moving within small airways may induce eddies and turbulence in laminar liquid flows at small scales, contributing significantly to heat and mass (CO2) transport though PFC liquid by means other than diffusion.

It is hypothesized that the lack of bubble-induced turbulence in TLV may account for the large diffusion-dead-space for heat and CO₂ which seems to be present in TLV at even low breathing rates – an effect which is apparently absent in both PLV and LAPC.

4.4. Potential development of clinical LAPC

 Rapid cooling of the CNS has now become a primary goal in the clinical management of the post resuscitation syndrome [49] and potentially in the acute management of spinal cord injury.[194],[195],[196] Based on their work, Safar, et al., have noted that clinical implementation of mild resuscitative hypothermia, which was highly effective in the dog model of SCA, will depend upon the development of truly rapid MTH.[197]  A recent editorial in Stroke [83] commented on the striking ability of the combination of  (33-35^oC ) [198] and pharmacological pre-treatment to ameliorate ischemic brain damage in the rat middle cerebral artery occlusion model, and  then addressed similar concerns: “A problem for use of this technique for acute stroke therapy is that the time required to induce hypothermia in patients is likely to be considerably longer than for rats. […]. To substantially increase the rate of hypothermia induction in humans, it will almost certainly be necessary to use some sort of invasive procedure, such as a heat-exchanger, to cool the circulation.”

The technique of LAPC may eliminate the need for such invasive measures. For example, in the cited trial [199], rats were cooled from 37 to 33°C (−4°C) over 40 min, using surface cooling with ice packs. By contrast, the present study demonstrates cooling of the canine body core and brain by −4°C in less than 10 min.

4.5. Challenges Ahead

 4.5.1. PFC Selection

Development of clinical LAPC awaits identification of suitable PFCs for various LAPC applications. For example, the pharmaceutical PFC perfluorooctylbromide (Perflubron™ Alliance Pharmaceuticals) would presumably not be suitable for rapid LAPC cooling due to its freezing point of +6°C, but might be useful for slower cooling or for LAPHE facilitated re-warming. Some industrial PFCs have pour-points low enough to make them potentially useful as LAPC rapid-cooling media; however, most of these agents also have unacceptably high vapor pressures at 37°C or are not chemically defined in terms of chain length or even precise chemical composition (see Section 3: Perfluorchemicals). PFCs with such high vapor pressures exacerbate barotrauma by causing HNCL.

High vapor pressure PFCs may also increase the danger of long lasting lung collapse as a result of PFC, secondary to pneumothoraces, entering the pleural space and vaporizing (‘perfluorothorax’). FC-75, (formerly FX-80) was the first PFC used in liquid ventilation [10], but its relatively high vapor pressure (31.5 torr) makes it an undesirable LAPC agent. Assuming that a PFC with the desired biophysical properties is identified and produced to medical standards, LAPC should be easily scalable to adult humans. For example, the viscosity of FC-75 is similar to water [11], and under standard suction a 19 Fr. An adult pulmonary toilet catheter will remove FC-75 at ~2 l/min. As in the system described, a LAPC system may interface with a conventional gas ventilator system via a simple liquid-carrying catheter which extends through the endotracheal tube adapter suction port.

 4.5.2. Coronary Perfusion during LAPC

The effect of LAPC on coronary perfusion in the setting of ROSC following cardiac arrest due to myocardial infarction (MI) or in the presence of coronary artery disease (CAD) is unknown but is a possible cause of concern. One possible adverse effect is the potential compromise of the coronary circulation in CAD due to perfusion of the heart with profoundly chilled blood (i.e., blood temperature  @10°C below systemic temperature) leaving the pulmonary circulation and entering the coronary os. One of the authors (Darwin) has observed the onset of severe, acute angina in two hemodialysis patients with stable angina who were inadvertently dialyzed using very cold dialysate (Q_B of 250 ml/min at a blood temp of ~10° to 15°C) which resolved only when the dialysate temperature was increased to 30°C or above.

It is well established that exercise in cold environments, with associated inhalation of cold air, can trigger angina and lead to cardiac arrest in patients with coronary artery disease.[200],[201]  There has been considerable debate as to whether the cause of angina in this setting is due to cold air inhalation per se, or to the effects of a cold environment. It has been suggested that exposure to a cold environment is the primary factor in inducing cold weather angina, presumably by increasing peripheral vascular resistance resulting in an increase in cardiac workload at any given level of exercise [202],[203] in the same way that the cold pressor test produces acutely increased afterload and thus increased left ventricular wall stress.[204],[205],[206],[207] Hattenhauer and Neill [208] studied 33 male patients with coronary artery disease,11 of whom gave a positive history of angina when exposed to cold winter air. Seventeen of these patients were subjected to inhalation of cold air at -20°C for 4 min, and this resulted in angina at rest in four of these patients.

Cold air inhalation also produced angina in four of the seven patients who were paced at a heart rate which was well below the threshold for angina at room temperature. Cold air inhalation did not significantly increase myocardial oxygen consumption, or alter coronary blood flow as determined by the xenon clearance method. In a separate arm of the Hattenhauer and Neill study, cold air inhalation for 90 s in 18 patients produced no detectable-constriction of coronary arteries visualized angiographically. The investigators concluded from these findings that cold air inhalation induced angina could not be explained by an increase in cardiac work and myocardial oxygen consumption.

This study also demonstrated that there was no evidence of large or intermediate coronary artery or coronary arteriole constriction in response to the inhalation of very cold air. As a consequence, the authors proposed that cold air inhalation induced angina might be the result of constriction of minute coronary collaterals, or to other vessels compromising blood flow to potentially ischemic regions of the diseased myocardium. The work of Lassvik and Aveskog [209], confirmed the effects of cold air inhalation in inducing angina and failed to demonstrate any decrease in workload at either the onset of angina or at maximal workload during inhalation of moderately cold air (-10°C) in a room at 20°C. Dodds, et al., [210] evaluated 12 male patients with stable angina inhaling cold air (-8.8°C) during exercise to investigate if the vasoconstrictor peptides endothelin-l (ET-1) and angiotensin-II (AT-lI) played a material role in the etiology of this phenomenon. They concluded that neither ET-1 nor ATII had any significant role in the pathophysiology of cold air inhalation induced angina.

In contrast to Hattenhauer and Neill, and Lassvik and Aveskog, these investigators documented decreased myocardial oxygen consumption during peak exercise in cold air inhalation and concluded that the cause of this was a centrally operating mechanism such as a reduction in coronary flow. These investigators noted that the response to cold air inhalation was biphasic, and posit that the initial response; earlier onset of angina during exercise while breathing cold air, was due to activation of cold receptors in the upper airways stimulating a systemic increase in peripheral vascular resistance; and thus the observed accompanying rise in blood pressure and increased cardiac workload (i.e., the cold pressor test response). The secondary response; reduction in total exercise time, which occurred at a significantly lower rate-pressure product compared with the same patients breathing ambient temperature air, was thought to be due to peripheral reflex responses. Thus, these investigators hypothesize that the early onset of angina during exercise is due to sympathetic stimulation from inhaled cold air, but as exercise continues, central mechanisms play an increasing role in the pathophysiology of cold air inhalation induced angina.

The implications of cold air induced angina for LAPC in the setting of coronary artery disease are troubling and certainly bear careful investigation in animal models of compromised myocardial circulation. Patients in all of these studies developed significant ST depression (³ 1 mm) concurrent with the onset of cold air inhalation induced angina, suggesting clinically significant myocardial ischemia. What is unclear is whether the concomitant profound reduction in myocardial temperature seen in LAPC (both transient and long-term) will be protective against any perturbation in myocardial perfusion induced as a result of the sympathetic or central effects of cold FTLV.

Such a protective effect is suggested by the well established finding that ‘cold’ (~34°C) hemodialysis (HD) is protective against both intra-dialytic hypotension and angina [211], [212],[213],[214], as a result of sympathetic stimulation from the return of ‘cool’ blood to the systemic circulation. [215],[216],[217] In addition to its protective effect on hemodynamics during aggressive ultrafiltration, cold HD also attenuates the hypoxemia leucopoenia, [218] and the production of Complement 5a induced by blood exposure to the dialyzer membrane.[218],[219] It should also be noted that catecholamine administration (mimicking profound sympathetic stimulation) in the form of epinephrine is still a mainstay treatment of ventricular fibrillation and aystole in SCA.[59, 220],[221]

Finally, it is essential to point out that mild and even moderate induced hypothermia not only do not interfere with defibrillation from cardiac arrest, but actually greatly facilitate conversion of VF to perfusing NSR.[222],[223],[224],[225],[226]

 4.5.3. Potential for Regional (Myocardial) Overcooling

Under the low flow conditions of CPR, the possibility exists that due to thermal compartmentalization myocardial temperature could be reduced to below the fibrillation threshold or to adversely affect contractility. In the laboratory and the clinical setting moderate (28-32 ^oC ) and deep (10-22 ^oC) hypothermia are known to induce both benign and malignant cardiac arrhythmias; [227],[228] and  ventricular fibrillation is the most common cause of death in accidental hypothermia. [229],[230] In addition to the arrythymogenic effect of deep hypothermia (J waves, prolonged PR, QRS, QT intervals, and atrial arrhythmias) there is evidence that defibrillation becomes increasingly problematic as myocardial temperature decreases below 30^oC.[231],[232],[233]

The temperature of blood leaving the pulmonary circulation under normal flow conditions (during spontaneous circulation) within 5 minutes of the start of LAPC can reach temperatures of 17^-20^oC (see Figure 2-14, FC Suction Reservoir Temperature; this temperature closely approximates the PA blood temperature during the FTLV cycle). This is benign under high flow conditions because heat is being rapidly and continuously transferred between body compartments (Figures 2-5 – 2-7). However, under low flow conditions, and especially in the presence of pharmacologically induced severe peripheral vasoconstriction, rapid, selective core over-cooling could occur.

Disproportionate cooling of the body core happens under normal conditions in humans given large volumes of intravenous fluid chilled to 4^oC, and this is the basis for cold IV fluid induced hypothermia for cardiac arrest.[234],[235],[236],[237] This surprisingly durable core cooling, well beyond that predicted on the basis of calculations for the whole body, occurs because the thermal distribution volume in humans given rapid  cold IV infusions turns out to be much lower than total body volume. The result is that chilled IV fluids are ~3 times more effective in inducing hypothermia than suggested by all-compartment equilibrium calculations.[234]  Several explanations have been offered for this apparent thermodynamic inconsistency; most plausibly that peripheral vasoconstriction and inherently slower kinetics of heat exchange between central and peripheral compartments act to keep core temperature below the all-compartment equilibrium for at least 60 min after the conclusion of the cold IV infusion [238],[239] which is long enough for external cooling to begin contributing to cooling and for endovascular cooling to be initiated under ideal conditions.

These findings provide additional reason for vigilance in avoiding excessive myocardial or core cooling during CPR and suggest that a surrogate for myocardial temperature be sought and that the use of warmer PFC FTLVs be explored; trading off rapidity of temperature systemic reduction against the danger of over-cooling the heart.

 4.5.4. Overcoming Increased Intrathoracic Pressure and Preserving CO

Following the development of active compression decompression CPR (ACD-CPR) by Cohen, et al., in 1992 [240] the critical importance of maintaining negative intrathoracic pressure during the decompression phase of the CPR duty cycle has become increasingly understood.[241],[242]  There is a rapidly growing body of both animal and clinical CPR research documenting improved survival and decreased neurological morbidity when the intrathoracic pressure is kept negative during the  decompression (release of chest compression) phase of CPR by the use of inspiratory impedance threshold devices and active ACD-CPR.[188],[243],[244],[241] Similarly, there is accumulating evidence that the increased intrathoracic pressure that results from excessive positive pressure ventilation (PPV) during CPR dramatically reduces CO and causes increased morbidity and mortality.[245],[246],[247],[248],[249],[250]

In 2004, Yannopoulos, et al., reported the development of a device which allows for the continuous application of negative intrathoracic pressure (Figure 2-15) by applying controlled suction to the airway.[187]  This device, called the intrathoracic pressure regulator (ITPR) allows PPV to be delivered as needed during ACD-CPR, while maintaining negative intrathoracic pressure at all times when PPV is not being administered. The device effectively transforms the thoracic cavity into a low negative pressure (vacuum) chamber; increasing venous return from the body and consequently increasing preload and cardiac output. The ITPR also markedly increases CPP while at the same time decreasing intracranial pressure (ICP). In CPR ICP is typically elevated from the basal value of 12-16 mm Hg to 22-30 mm Hg (as the result of pressure transmission by blood in non-valved veins and by transmission of intrathoracic pressure via the cerebrospinal fluid) further compromising already inadequate cerebral perfusion.[251] Reduction of ICP during CPR has been shown to improve both survival and neurological outcome in an animal model of CPR.[252]

Figure 2-15: Prototype ITPR (Advanced Circulatory Systems, Inc.) in position in a typical bag-vale resuscitator – ET tube set-up.

 The ITPR has been shown to dramatically improve gas exchange, hemodynamics, blood flow, vital organ perfusion, and short-term survival rates during VF cardiac arrest in a porcine model of SCA and CPR [187] The ITPR is able to not only overcome the high intrathoracic pressures. associated with CPR (45 to 55 mmHg or ~61 to 75 cmH₂0 [253]) but to both create and sustain negative intrathoracic pressure (determined indirectly by measuring the ET tube pressure) continuously during prolonged periods of ITPR-CPR, even in the presence of induced hypovolemia.[189]   In hemorrhaged (hypovolemic) pigs, Yannopoulos, et al., were able to sustain CPP at >15 mm Hg (the accepted threshold for successful defibrillation in human SCA) and the isovolemic VF animals in the study maintained CPP at >25 mm Hg throughout the full 15 minutes of ITPR-CPR. In both groups, ETCO₂ was consistently maintained >25 mm Hg, and the 1-hour survival was 100%, as contrasted with 10% in control animals receiving AHA standard CPR (P = 0.0001).

By comparison, after 3 minutes of conventional CPR the control animals had a mean coronary perfusion pressure of <15 mm Hg and all had developed pseudo-respiratory alkalosis indicative of the V/Q mismatch of standard CPR.[254]  Blood gases in VF animals were strikingly preserved during ITPR-CPR; paO₂, which was 96±2 mm Hg at baseline, was   214±12.37 mm Hg after 10 min and 198±6.75 mm Hg after 15 min of ITPR-CPR. These findings would seem to suggest that ITPR-CPR may be reducing or eliminating the pulmonary edema that accompanies CPR and the high intrathoracic (and thus pulmonary arterial and venous pressures) generated during CPR.  ITPR is similarly effective at improving both hemodynamics and survival in a swine model of severe hypovolemic hypotension.[255]

The use of an ITPR during LAPC offers the prospect of reducing or abolishing the undesirable hemodynamic effects of FTLV with PFC. These effects, while not clinically significant in the healthy animal with spontaneous circulation undergoing LAPC (and the ability to dynamically respond to alterations in preload, CVP and SVR), may be unacceptable in the setting of cardiac arrest and CPR.

While it is clear that FTLV with PFC reduces MAP and elevates the mean CVP,[2] the extent to which these effects will reduce CO during CPR is unknown and will necessarily require further experimentation in animal models of SCA and CPR that closely approximates those experienced in humans under clinical conditions. The mechanics of CO and perfusion in CPR are radically different than those that pertain under conditions of spontaneous circulation.[256],[257],[258] In the healthy beating heart, modest increases in CVP (~5-15 mm Hg) result in increased cardiac output via the Frank-Starling mechanism (Starling’s Law of the Heart). Increased CVP increases left ventricular diastolic end pressure (LVEDP) increasing ventricular volume and stretching the ventricular myofibrils. Myofibrillary stretch results in sarcomere extension thereby increasing the affinity of troponin C for calcium, causing a greater number of cross-bridges to form within the myofibrils; this increases the contractile force of the heart increasing CO. This biochemical mechanism is not operational during cardiac arrest and CPR. Indeed, as outlined in the discussion below, little of the mechanics of perfusion under normal physiological conditions pertain in the setting of CPR.

4.5.5. Consideration of the Mechanics of Blood Flow in CPR and Implications for LAPC

During CPR in humans ventricular volume is not a primary determinant of CO, at least not in an appreciable fraction of patients undergoing CPR.[258] Despite the fact that it has been forty-eight years since the invention of closed chest CPR – the mechanics of blood flow during CPR in humans have not been definitively established – and a great deal of controversy surrounds the subject. Broadly, three mechanisms of antegrade blood flow have been proposed in (human) CPR:

• The Direct Cardiac Compression (Cardiac Pump) Theory posits that mechanical compression of the ventricles between the sternum and vertebral column creates a pressure gradient between the ventricle and the aorta (or pulmonary artery in the case of the right ventricle); as a consequence the mitral and tricuspid valves are closed, and blood is moved antegrade out of the ventricle. The ventricle then refills during the decompression phase of CPR and the process is repeated with each compression cycle.[259]  The cardiac pump theory was most definitively challenged with the publication of case data documenting the ineffectiveness of CPR in patients with flail chest. This could be reversed only be restoring elastic recoil to the chest by binding it; indicating that it was the generation of a net negative intrathoracic pressure upon recoil of the chest during the decompression phase of CPR that was essential for blood flow.[260]
• The Thoracic Pump Theory asserts that chest compression increases intrathoracic pressure forcing blood to flow from the thoracic to the extrathoracic circulation[3]. Retrograde flow from the right heart to the systemic veins is prevented by the venous valves; with the heart serving only as a passive conduit, and having no function as a pump.[261], [262] In the thoracic pump paradigm blood circulates because increased intrathoracic pressure is transmitted more or less equally to all of the intrathoracic vascular structures [256] and the atrioventricular valves remain open during the compression phase of CPR.[263]  However, these intravascular pressures are not equally transmitted to the extrathoracic arterial and venous beds, thus creating an extrathoracic arterial-venous pressure gradient resulting in antegrade blood flow during the period of high thoracic pressure. During the decompression phase of CPR, blood flows into the lungs as a result of the extrathoracic venous-to-intrapulmonary pressure gradient.  Angiographic [256] and echocardiographic studies in dogs documented static ventricular volumes during CPR [262], and case reports of human patients with cardiac tamponade recovering successfully after closed chest CPR both supported the thoracic pump mechanism as the explanation for antegrade flow in CPR.[264]
• In the Lung Pump Theory, as in the thoracic pump theory, the increased intrathoracic pressure during the compression phase of CPR is similarly transmitted equally to all intrathoracic vascular structures and there is insignificant regurgitation of blood from the pulmonary artery into the right ventricle and vena cavae until the pulmonary valve closes. [265] Thus, blood under pressure within the pulmonary vasculature will flow out of the lungs via the left side of the heart.[263],[266] During the decompression phase of the cycle the intrathoracic pressure drops below that present in the extrathoracic vasculature and blood flows into the thorax via the vena cavae and aorta. The pulmonary vascular volume is replenished by blood flowing from the right side of the heart through the open tricuspid and pulmonary valves. [266], [267] Retrograde aortic flow is prevented by competent closure of the aortic valve.[268],[265]

Transthoracic echocardiographic studies reported in the early 1980s during CPR in humans supported the thoracic pump theory.[268],[265] However,more recent studies employing transesophageal echocardiography (TEE) have concluded that because the mitral valve was observed to close during the compression phase, and open during the decompression phase, and the left and right ventricular volumes decreased during the compression phase, the  mechanism of antegrade flow during CPR was consistent with the direct cardiac compression theory.[269],[263]

Perhaps the most likely explanation of these seemingly incompatible findings is that all three mechanisms are responsible for antegrade flow, but not in every patient. In other words, the mechanics of forward flow may differ from patient to patient.  In fact, in the 1981 paper in which they proposed the thoracic pump theory of antegrade blood flow in CPR, Weisfeldt and Chandra stated, “It is not essential in the human to think about these mechanisms in an exclusive fashion. Direct cardiac compression is useful when possible; when it is not potent enough to maintain cerebral perfusion, manipulation of intrathoracic pressures would likely have a favorable additive effect on carotid blood flow.”[261]

In support of this view is the study by Ma, et al., which evaluated 17 patients undergoing CPR using TEE to measure both pulmonary and trans-mitral flow.[270]  Five of the 17 patients demonstrated closure of the mitral valve during the compression phase of CPR, with associated mitral regurgitation and forward aortic flow occurring which is consistent with the cardiac pump theory.

In the remaining 12 patients, the mitral valve remained open during both the compression and decompression phases of CPR with maximal antegrade mitral flow occurring during the compression phase of CPR. Eight of these 12 patients demonstrated antegrade mitral flow during the compression phase that was accompanied by antegrade pulmonary vein flow; consistent with the classic ‘thoracic pump’ mechanism. The last 4 patients in this study evidenced retrograde pulmonary vein flow concurrent with antegrade mitral flow during the compression phase; which is most consistent with lung pump theory as the primary cause of antegrade flow, at least in these patients.

5.0 Review of the Literature on LAPC

From the foregoing, it should be easy to understand the difficulty of establishing by experiment, let alone extrapolating on any theoretical basis, what the hemodynamic impact of FTLV or LAPC will be under clinical conditions in humans undergoing CPR. Similarly, the effect of LAPC on regional myocardial blood flow, myocardial irritability, coronary electrophysiology, and susceptibility to defibrillation will require further laboratory and, likely, experimental clinical investigation as well. Since publication of the 21CM/CCR LAPC research in 2001 there have been 4 subsequent studies to evaluate various aspects of the utility and safety of LAPC for application in CPR, or as a therapy in MI.

The first of these studies was published by Hong, et al., in 2002 and announced, “Our study is the first to demonstrate an induction of hypothermia by adopting PLV (no need of an extracorporeal circuit) and 0°C PFC.” [271] failing to cite the previously published work of either Darwin, et al., or Harris, et al. [272] These investigators attempted to validate the utility of LAPC using FTLV to achieve rapid reduction in core temperature and also studied the physiological impact of the procedure. The cooling rate achieved appears to have been ~0.11°C/min and this slow rate was almost certainly an artifact of the small volume of PFC used for each intrapulmonary exchange (~20 ml) and the small number of exchange (10 – 12 exchanges over 38 minutes). In fact, the cooling rate was faster in the animals in this study that were cooled externally by repeated application of ice water slush (0.17°C/min).

The most interesting result of this investigation were that MAP, mean PA pressure, HR, and CO, CBC and lactate of the LAPC treated animals were not significantly different than in the surface cooled animals. One variable that was altered in the LAPC group was an increase in pulmonary vascular resistance which appeared immediately after liquid loading and did not begin to return to baseline until halfway through the LAPC period. It is also remarkable that 1 animal in both the LAPC and the surface cooled groups (n = 7 for both groups) developed pulmonary hypertension and lethal pulmonary edema. The authors speculate that the observed pulmonary hypertension could have been a result of pulmonary venous constriction and/or increased pulmonary microvascular blood sludging due to the profound local cooling of lung vasculature resulting from instillation of 0°C PFC. They also raise the possibility that the observed elevation in PAP may have resulted from the mechanical effects of the PFC load; but offer no explanation for this.

Otherwise, hemodynamics were unaffected by LAPC. The paCO₂ became progressively elevated during the hypothermic interval in the LAPC group, perhaps due to inadequate PEEP or inappropriate parameters of mechanical ventilation in the face of evaporating PFC (which was not replenished during the course of the experiments).

The second study to be published was by Ko, et al., in 2002.[273] The purpose of this study was to evaluate the effect of PLV on pulmonary blood flow under low blood flow conditions designed to simulate those encountered during CPR in order to further validate the use of LAPC as an adjunct to cardiopulmonary cerebral resuscitation. Isolated perfused rat lungs were subjected to 20 min of PLV with room temperature Perflubron™ and segmental (i.e. pre-capillary, capillary, and post-capillary) hemodynamics were studied at a perfusate flow rate of 6 ml/min (~5% normal cardiac output [274]). Lungs received either gas ventilation or 5 or 10 ml/kg PLV. Segmental pressures and vascular resistances were determined, as was transcapillary fluid flux. PLV at both the 5 and 10 ml/kg Perflubron™ dose produced no detectable changes in pulmonary blood flow or in transcapillary fluid flux and the investigators concluded that, “These data support further investigation of this technique as an adjunct to cardiopulmonary resuscitation.”

Figure 2-16: Dramatic reduction in myocardial infarction size in rabbits rapidly cooled to ~34 by the use of TLV LAPC: infarct volume was 4.0 ± 5% in the LAPC groups vs. 37.7 ± 1.3% in the normothermic gas ventilated group (p = 0.001). Redrawn from Tissier, et al.[275]

Unfortunately, this study did not explore the effect of instilling cold PFC into the lungs, nor did it employ FTLV. Because it was an ex vivo study it was not possible to determine the effects of PFC loading, cold or isothermic, on the sympathetic response, coronary blood flows or other physiological parameters of concern in LAPC.

In 2007 a study by Tissier, et al., [275] used a rabbit model of evolving MI to evaluate the efficacy of TLV LAPC administered during the ischemic interval in achieving rapid core cooling to reduce infarct size and provide myocardial protection [276],[277],[278] even after substantial delay  [279] and when induced during reperfusion. [280],[281] It is well established that intra- and post-ischemic myocardial hypothermia dramatically reduces infarct size in animal models of MI and this observation has recently been extended to humans in the clinic.[282]

These investigators note that in the US the time to coronary revascularization following infarction is 120 min in 41.5% of patients admitted during off-hours and that 27.7% of patients presenting during normal business hours still averaged delays to revascularization in excess of 120 min.[283]  Rapid induction of hypothermia in these patients could be effective in providing substantial myocardial salvage during the delay between presentation and revascularization. This study demonstrated a remarkable reduction in infarct size in the LAPC treated group; 4.0 ± 5% vs. 37.7 ± 1.3% in the normothermic, gas ventilated animals (p = 0.001). The cooling rate achieved with TLV was 1.32°C/min; 6.6°C in 5 minutes. The effectiveness of TLV LAPC is not consistent with results obtained in larger animals in this author’s experience. It may be that the relatively short distances between the large and small airways and the very small diameter of even the largest airways in the rabbit as compared with the dog or human, may have improved the efficiency of heat exchange.

The most recent study by Staffey, et al., [284] is more on-point and provides considerable reassurance that cold PFC loading of the lungs during CPR and subsequent resuscitation is not only not deleterious, but in fact is markedly beneficial. In this study swine were subjected to 11 min of VF and treated either with a static intrapulmonary infusion of PFC chilled to -12 C to ~75% of total lung volume (40 ml/kg), static infusion of isothermic PFC (33^oC) under the same conditions, TLV with -15^oC PFC at 6-cycles per minute during 10.5 minutes of the arrest interval, and a control group that was arrested with no intervention during the ischemic interval. The animal were then given AHA standard CPR with defibrillation and advanced cardiac life support (ACLS). The specific objectives of this study were to determine what the effects LAPC; tidal and static, administered during the period of cardiac arrest would be on hemodynamics, CPP, gas exchange, defibrillation, and hemodynamic stability during a 1 hour period of evaluation post ROSC. The endpoint was ROSC for 1 hour without ionotropic support.

The global objective of the study was to determine if LAPC could be used to rapidly induce cardiopulmonary hypothermia during the arrest period as opposed to cerebral or systemic hypothermia. Work by Rhee, et al., and Boddicker et al., also using swine models, had previously demonstrated that systemic hypothermia induced before resuscitation from cardiac arrest resulted in more rapid and consistent defibrillation from VF as well as earlier and more stable ROSC. Since the purpose of this study was to cool only the thoracic viscera to determine the effect of local cardiopulmonary hypothermia on resuscitation; ‘targeted cardiopulmonary intra-arrest moderate hypothermia (28-32^oC),‘ LAPC was not continued beyond the period of cardiac arrest.

Both static and TLV LAPC succeeded in reducing the pulmonary artery temperature to the desired temperature of 34^oC by 6 and 10 min post arrest, respectively. Eighty two percent of the animals in both LAPC groups were resuscitated successfully and survived the 1 hour evaluation period without pharmacological support, as contrasted with only 27% of the controls animals (p = 0.03).  Interestingly, 73% of the isothermic  TLV swine achieved and maintained ROSC as opposed to 27% of the swine in the control group (p = 0.09), suggesting that the presence of PFC in the lungs during either the arrest or resuscitation interval may improve the odds of successful defibrillation.

Figure 2-17: Results of the study by Staffey, et al., to evaluate the effect of cold and isothermic PFC infusion and TLV on resuscitability of swine after 11 min of electrically induced VF cardiac arrest. PFC loading, and in particular cold PFC loading or TLV improved 1 hour stable hemodynamic survival by 73% versus 27% in controls. Infusion of isothermic PFC to near vital capacity also showed a trend towards increased survival. Redrawn from Staffey, et al.,[284].

Chamberlain et al., have recently proposed that dilation of the right ventricle (RV) that occurs after the first ~5 min of cardiac arrest results in compression of the left ventricle (LV) within the confines of the pericardium. The effect of this would be to reduce myocyte stretch and reducing the contractile strength of the left ventricle in response to defibrillation (see discussion of the Frank-Starling curve above). They posit that an arrested heart that has a reduced contractile state, and in which left ventricular volume is reduced by the right ventricle over-distended with venous blood will not be able to initiate effective contraction even if coordinated electrical activity is restored. Distension of the RV occurs post-arrest due to centrally mediated sympathetic contraction of the vasculature, as well as movement of blood into the venous circulation as arterial and venous pressure equilibrate; this is a commonplace finding at both necropsy and autopsy in normovolemic subjects.

Unloading the RV prior to defibrillation results in an immediate improvement in successful ROSC independent of any known metabolic that might accrue from ~10 to ~15 sec of coronary perfusion that might also result. Chamberlain, et al., believes that it is the decompression of the left ventricle and restoration of a more physiologic morphology that facilitates or even enables defibrillation under these conditions. Instillation of a large volume of dense PFC may antagonize distension of the RV and prevent LV volume loss and compression of the LV to below the threshold required to established effective contractile activity in response to defibrillation. PFC to vital capacity (VC) should effectively preclude this post-arrest pooling of blood in the thoracic caval and pulmonary vessels and may act to preserve left ventricular morphology during prolonged cardiac arrest.

Especially encouraging findings from the Staffey, et al., study were the absence of any noticeable adverse hemodynamic impact from either isothermic PFC loading or from cold PFC loading or cold TLV. MAP, CVP, CPP, pH and blood gases were not statistically different between the four groups of animals in the study.

These four studies of LAPC aimed at answering questions bearing on the feasibility of LAPC in MI, SCA and CPR are very encouraging. They indicate that LAPC is being seriously considered as a potential therapeutic application in humans. That this research is being undertaken in independent academic and medical research both in the US and elsewhere is especially heartening and would seem to indicate that the enormous therapeutic potential of LAPC-induced hypothermia has been successfully communicated.

6. Conclusions

LAPC is capable of inducing hypothermia in a fraction of the time that it takes to prepare a patient for cooling via CPB. In addition, automated LAPC need not have the spatial and technical restrictions of the hospital setting. Although relatively simple methods of continuous arterio-venous shunt heat-exchange that do not require a blood pump or carry the most of the risks attendant to CPB have been described which might be potentially applicable in the field [285], these techniques also have the drawback of requiring skilled personnel for cannulation of a major artery and vein. Intracaval heat exchange catheters can reduce core temperature,  but only at rates of ~1.46 ± 0.42°C/h [286]; far too slowly to achieve the maximum benefit from post-reperfusion hypothermia

Figure 2-18: Large negative excursions in airway pressure occurred during suctioning of PFC at the end of most FTLVs when this operation was carried out manually (A). This occurred because when the PFC suction catheter was no longer filled with PFC, evacuation of gas from the airways occurred very rapidly owing to the much lower viscosity of gas compared to PFC. It was impossible for the operator to anticipate when the last of the bulk liquid would be removed, or to react rapidly enough when this occurred. Computerized sensing and control eliminated this potential source of baro-injury (B).

By contrast, LAPC may be a candidate for a much wider range of emergency field-uses in civilian and military settings since the primary technical skill required to initiate LAPC in the field is endotracheal intubation; a skill possessed by paramedical personnel throughout Canada, the U.S. and Europe. LAPC has also potential as a very rapid treatment for heatstroke and malignant hyperthermia. While  not the subject of this report, LAPHE clearly also holds promise for core rewarming in severe hypothermia, although an absolute maximal PFC temperature of 42°C would in theory limit the re-warming rate to about one-third of that possible in cooling.

Computer control of both gas and FTLVs is effective at eliminating excessive positive and negative airway pressures during LAPC (Figure 2-16) and computer control can easily be extended to encompass cooling rate, depth, and duration and can also be extended to control gas ventilation, as necessary.

Whether used inside or outside hospital, successfully implemented LAPC might more generally serve as a neuroprotective bridge [287] in order to gain time for more technically sophisticated supportive or definitive treatment (e.g. neurovascular thrombolysis or interventional thrombectomy, emergency CPB, spinal cord decompression or definitive management of hemorrhagic shock in trauma or surgery).

Although some modalities of liquid ventilation have been clinically evaluated [288], the safety parameters of rapid and cold liquid delivery to the lungs remain to be determined. As noted in this study, LAPC can cause pulmonary injury. The mechanism of such damage suggest by the location of lesions indicates that both barotrauma (dependent lung) and volu-trauma (nondependent lung) are the primary, if not the sole factors. It has been observed that LAPC causes little permanent lung injury in long term survival animals. Similar pathology seen in lungs exposed to either isothermic or ~4ºC LAPC in the present study (data not shown) suggest that thermal/chilling-injury per se is not the major insult. Although more subtle biochemical and immune problems secondary to hypothermia itself are suggested by reports from some longer duration studies of MTH (pneumonia and sepsis), it is not clear that the short duration of treatment necessary to achieve the benefit of post-resuscitative MTH will pose such problems. It is hypothesized that the pulmonary injury observed in LAPC may be reduced with better control of LAPC pressure and volume limits, and by use of PFC liquids having more physiologically suitable properties.

A significant, unresolved concern is the potential negative impact of LAPC on myocardial perfusion and the danger of overcooling the heart during the low flow conditions of CPR with possible adverse effects on achieving ROSC and maintaining a stable rhythm following defibrillation. As already noted, these questions can only be resolved with further study.

Acknowledgements

 The authors thank Saul Kent and William Faloon for support, and Casey Brechtel for helpful discussions. Several of the authors have applied for LAPC device patents. This trial was funded by a grant from the Life Extension Foundation (Hollywood, FL).

Appendix A

 (The abbreviations contained in this appendix are also reproduced in the table of abbreviations and acronyms at the beginning of this document.)

 A.1. Abbreviations and notation

 In the text, volumes (V) are given in ml/kg, and flows (dV/dt =V’= ˙V ) in ml/kg per min. Since all V and ˙V are expressed in mass-specific (per kg animal) terms, derived quantities DQ and Cm are automatically mass-specific. C_m and C_vf are given in calories / (g or ml) per °K for easy comparison with water.

PFC = perfluorochemical; hydrogen-free organic molecule in which most of the peripheral atoms are fluorine.

TLV = tidal liquid ventilation is a modality in which liquid completely fills the lungs and ventilator.

PLV = partial liquid ventilation is a modality in which all gas exchange is via gas ventilation, with ~1/2 FRC of PFC liquid present in the lungs to recruit dependent lung in ALI or ARDS.

LAPC = liquid assisted pulmonary cooling is a heat-exchange modality in which ventilation occurs via both gas and liquid ventilation proceeding concurrently.

T_tym = tympanic temperature

T_art = arterial temperature

T_ven = central venous temperature

T_rec = rectal temperature

DT_enet DT_tymp = resulting from LAPC, after equilibration at t=40 min

T_FTLV  = FTLV cycle infusion time

t_s = FTLV cycle suction time

t_c = FTLV cycle period (=tinftc +ts)

V_FTLV = single-cycle PFC FTLV infusion volume=tinf V_ inf

 V_S = single-cycle PFC FTLV suction-return volume

V_D  = ventilatory dead space (any type)

V_DCA = expected gas ventilation VD=sum of circuit (mechanical) VD plus anatomic VD

V_DTherm = thermal or heat-exchange VD (ml/kg, in reference to liquid PFC infusion)

˙V _inf = PFC infusion rate (set to _50 ml/kg per min in Trials I and II)

˙V_FTLV = effective PFC FTLV rate=LAPC liquid FTLV minute-ventilation (ml/kg per min) = V_FTLV/ t_c

˙V_g = gas minute-ventilation (ml/kg per min) m animal mass Ch heat capacity

 C_T = total heat capacity of the animal (= mC_m)

m = mean mass-specific heat capacity of the animal ( = DQ_T / DT_e)

C_vf = volume-specific heat capacity of FC-75 (mean of 0 and 25°C values) = 0.45 cal/ml per °K

DQT = total heat removed during LAPC (kJ/kg animal) = S DQ_c

DQc = heat removed during one FTLV cycle?????

E_f = mean cycle heat transfer efficiency=mean of [DQ_c / (theoretic DQ_c (max)] for all cycles in a single experiment

n = number of FTLV cycles in LAPC experiment

S = sum entire quantity following, for all cycles i = 1 through n

 T_inf   = PFC infusion temperature

T_S = PFC suction removal temperature (time-averaged PFC suction flow temperature)

T_SM= PFC mixed suction return-volume temperature (temperature of mixed VS)

 A.2. Thermal kinetics

 During LAPC cooling and equilibration, the blood and tympanic temperature changes in the animals were modelled by a simple five compartment model (Figure 2-8). During the initial ~100 s of LAPC (value used as empiric time mark), full development of heat-exchange behavior is established between the lungs, blood volume, and the thermal core of the animal, as suggested by the characteristic half-times for equilibration of these systems (see below).

Modelling of cooling during LAPC:

After the initial ~100 s of cooling, the data for tympanic DT(t) = DT_tym during LAPC in Trial I and II were modelled by a single time-constant exponential decline.

Mean DTtym data for each trial from times t=100 to 1080 s were fit using (Eq. (1)).

DT (t) = T₁₀₀ + DT_k [1− exp (−t / t_o) ],

DT(t), total T_tym change from baseline T_tym at start of LAPC; t, =time in seconds after empiric time mark, 100 s after start of LAPC; T₁₀₀, observed DT at empiric time mark, 100 s after start of LAPC; DT_k, observed temperature-interval constant, specific to each LAPC method; to, observed natural-base time-constant, in sec (t_o = halftime / ln 2).

Best-fit values for Trial I data were: T₁₀₀ = −0.52 ± 0.02°C; DT_k = −11.2 ± 0.02°C; and to = 1064 ~3 s. Trial II values were T₁₀₀ = −0.24 ± 0.02°C; DT_k = −8.14 ± 0.02°C; and t_o = 1107 ± 5 s. The relatively long time-constant associated with this thermal phase, which was similar in the two trials, presumably reflects the long time-constant (~700 s, see below) associated with heat transfer from the thermal core of the animal to the thermal periphery; thus the exponential phase represents full development of heat exchange between the LAPC cooling device and the entire animal. The final linear segments of cooling occurring after this phase, measured at −0.29°C/min (Trial I) and −0.21°C/min(Trial II), represent the final relatively simple state which exists after heat exchange equilibrium between cooling device and animal has been fully established.

Cooling in blood and tympanic sites during LAPC, and thermal evolution in these sites during equilibration phase after LAPC was discontinued, was in accordance with a five-compartment thermal model (Figure 2-8). In this model, the tissues of the animal are divided into three thermal compartments, corresponding loosely with the vascular system, the thermal core, and the thermal periphery.

Modelling of equilibration after LAPC:

Perfusion-driven convection is the major heat transfer mechanism in very rapid systemic cooling processes. This fact allowed T_art and T_ven changes to be used to quantify some features of heat transfer between body thermal compartments during the equilibration period after LAPC. The mean T_art curve in Trial I increased nearly linearly (R² = 0.9976) for 12 s after the end of LAPC, rising at a rate of 7.9°C/min. After this initial 12 s, T_art departed from linearity (Figure 2-6 inset), and was modelled by the sum of three exponential terms with respective time constants (t₀) of 12 ± 0.4, 102 ± 2 and 701 ± 8 s. These t₀ times differed to a large enough extent that their respective influences could be considered to be controlling over discrete time periods of about twice their value. Thus, the four equilibration phases seen after the end of LAPC lasted approximately 12, 24, 200, and 1400 s (23 min), respectively and represented 34, 14, 25, and 27% of the 5.1°C rise in T_art during equilibration after LAPC.

These data may be interpreted as follows: during each phase of the equilibration process, one or more thermal compartments in the animal equilibrated with the next-most closely-connected compartment (Figure 2-8). Afterwards, the newly captured compartment(s), as part of a larger unit bound together by blood mediated convection, equilibrated with the next-most closely connected compartment, and so on. The 12 s linear first equilibration phase (Figure 2-7, inset) most likely represents development of heat transfer from the lungs to local pulmonary blood flow. This phase was not associated with blood recirculation since it was seen as a rise in T_art but not T_ven. The second equilibration phase (duration ~24 s) was characterized by an increase in dT_ven / dt to the value of d_Tart  /dt, indicating that the lungs, blood-volume, and certain other well-perfused viscera, such as the kidneys, were now evolving into a single thermal system. Since the observed to for this phase was 12 s, less than the animal’s mean circulation time (= cardiac output / blood volume ~30 s), this process appeared to be driven by blood circulation via the most rapid paths (e.g. renal circulation). Such short paths for circulatory heat transfer were evident in the relatively small lag times (10.4 ± 6.9 s) noted between T_art and T_ven changes in these animals.

During the first two equilibration processes, the pulmonary circulation added thermal potential to the blood-volume more rapidly than it could be removed by the systemic circulation. By the end of the second equilibration phase, however, lung-to-blood heat transfer no longer dominated, and the gap between T_art and T_ven was set by the magnitude of heat transfer from the circulating blood volume to the tissues that comprise the ‘thermal core’ of the animals. In this third equilibration phase (duration ~200 s), the viscera and blood-volume, as a unit, equilibrated with the remainder of the ‘well-perfused’ tissues of the body (thermal core, comprising about 70% of the animal’s heat capacity). Heat capacities for thermal compartments are calculated below. The t_o for this process is seen most directly in the ~2 min. delay between maximal DT_ven and maximal DT_tym (Figure 2-7).

Finally, heat transfer within well-perfused tissues fell to a new minimum, and the T_art to T_ven gap decreased to a value set by the fourth equilibration phase (duration ~23 min) during which the well-perfused tissues equilibrated, as a unit, with a succession of the more poorly-perfused compartments, e.g. gut contents, fat, and other tissues comprising the thermal ‘periphery’ [12]. These processes could be consolidated into a single exponential term. Due to the long time-scale, heat transfer during phase four was probably partly conductive. Estimates of basal metabolism in the anesthetized, non-shivering dog (@90 J/kg per minute) indicate that as much as 0.6°C of warming per 20 min in this model may be due to metabolism.

A.3. Thermal accounting

 Heat transfer efficiency:

Although machine-LAPC allowed 2.3 times the FTLV frequency of the manual method, and resulted in a larger ˙V_FTLV by a factor of 1.2, the cooling magnitudes and rates for machine-LAPC significantly (P < 0.001) fell short of those obtained with manual-LAPC (Fig. 3). The strategy of increasing FTLV frequency (1/t_c) and decreasing ˙V_FTLV, in order to arrive at approximately the same FTLV rate (˙V_FTLV), therefore, significantly decreased the fraction of thermal potential which was transferred from each FTLV (= heat transfer efficiency, E_f). Unexpectedly, when the E_f for each of the 8 LAPC cooled animals of Trials I and II (Table 2) was calculated using (Eq. (2)), the value did not differ (P = 0.46) between trials; nor did total heat removed per kg (DQT), as calculated using (Eq. (3)), differ (P=0.14) between Trials. Both of these quantitative methods were therefore inaccurate for some dogs. Calculation of whole-animal mass-specific heat capacities C_m (= DQT / DT_e) suggested that the Trial II values of ~DQT and E_f principally were inaccurate, since the mean C_m value of 0.70 ± 0.1 cal/g per °K for Trial I was consistent with the C_m reported in the literature for mice and humans [289], whereas C_m values calculated for Trial II using (Eq. (2)) were unrealistic, being greater than the C_m of water.

E_f  (method 1) = 1/nS (T_S−T_inf) / (T_ven−T_inf),     (2)

DQ_T (method 1) = ˙V_FTLV C_vf ST_S−T_inf.                     (3)

To independently check the accuracy of Trial I values, we computed DQT and C_m for three dogs from an earlier study (Tables 1 and 2: dogs A, B, and C) that had been given ~4ºC PFC with FTLV times sufficiently long to allow the volumes and mixed-temperatures of suction-return liquid to be measured for each FTLV.

This allowed computation of DQ_T and E_f by a more detailed method (Method 2, Eqs. (4) and (5)), which used the extra thermal data (not available for Trials I and II) to more directly estimate FTLV heat transfer.

DQT (method 2)

= C_vf S V_FTLV (T_ven−T_inf) − V_S (T_ven−T_SM),                              (4)

When this was done, the mean Cm for dogs A, B, and C was found to be 0.68 ± 0.06 cal/g per °K, consistent with the Cm in Trial I (P=0.80). Method 1 (Eqs. (2) and (3)) required the assumptions that PFC suction volume equalled infusion volume, that suction flows remained constant, and that thermal hysteresis was negligible. These assumptions apparently held true at the larger ˙V_FTLV and t_c values of Trial I, but not for the smaller values of Trial II.

With this information, a new E_f for Trial II was estimated using method 3 (Eq. (6)), which employed an estimate for the heat required for the observed DT_e, vs. the total PFC thermal-deficit theoretically available.

This estimate required a presumed value of C_m. However, if the mean Trial II Cm was assumed to be the same as that of Trial I, then the true Trial II Ef could be calculated (Eq. (6)) to be 0.40 ± 0.06.

This value agreed with the rough estimation that since Trial II achieved only 73% of the DTe of Trial I, despite using 1.19 times more total PFC (Table 1), the

E_f in Trial II was expected to be about 73%/1.19 = 61% that of Trial I.

Thermal compartment size:

Heat removal for each cycle (DQ_c) in Trial I was calculated from the individual terms of Eq. (3), and individual-cycle mean cooling power calculated as DQ_c/t_c. The latter parameter was useful since thermal compartments in the dog are relatively isolated at short time scales (Figure 2-8), and thus the ratio of cooling-power to cooling rate (Eq. (7)) at a given probe site was expected to give the heat capacity (C_h) of the system of thermal compartments that were in equilibrium with each other, and with the site, at the time of the measurement:

C_h (Comp N), total heat capacity of thermal compartments N = 2 + 3, or N = 2 + 3 + 4.

Both t_c and dT/dt values were chosen at a time t, of interest when compartment system N has not yet equilibrated with slower half-time compartment(s). Thus, in

Trial I, near the end of FTLV cycle c1 (t = 30 s), the FTLV thermal-deficit had equilibrated within thermal Compartments 2 + 3 (PFC/viscera/blood-volume), but had not yet significantly reached Compartments 4 or 5.

If the cooling rate of T_ven at t = 30 s (−1.9 ± 0.8°C/min) was then taken as the cooling rate of the system of PFC/viscera/blood-volume, the C_h for this compartment system could be estimated from (Eq. (7)) as 20 ± 9% of C_T, the total heat capacity of the animal (C_T = m C_m). Subtracting the C_h contribution of lung PFC (=m ˙V_FTLV C_vf) allowed estimation of the remaining tissue C_h for Compartment 3 as ~19 ± 9% of C_T.

Similarly, the Ch of Compartments 2, 3, and 4 together, was estimated at t = ~140 s (cycle 4) as 71 ± 17% of C_T, corresponding to the classical whole-body ‘thermal core.’ The Compartment 5 C_h was then calculated to be the remainder (100−71%) = 29 ± 17% of C_T, corresponding to the classical ‘thermal periphery.’

Table of Contents

Introduction………………………………………………………………………………………………………. 6

Table of Abbreviations, Symbols and Acronyms………………………………………….. 9

Section I: Introducing Liquid Assisted Pulmonary Cooling………………………… 14

Liquid Assisted Pulmonary Cooling in Cardiopulmonary Cerebral Resuscitation, Introduction

………………………………………………………………………………………………………………….. 15

Hypothermia as an Active Therapeutic Agent

………………………………………………………………………………………………………………….. 17

The Benefits and Limits of ‘Delayed’ MTH: Real World Experience

………………………………………………………………………………………………………………….. 26

The Problem of Heat Exchange

………………………………………………………………………………………………………………….. 29

The Pathophysiology and Biophysical Limitations of External Cooling

………………………………………………………………………………………………………………….. 33

Consideration of Invasive Core Cooling Methods

………………………………………………………………………………………………………………….. 35

Exsanguinating Trauma Resulting Cardiac Arrest

………………………………………………………………………………………………………………….. 37

The Lungs as Heat Exchangers

………………………………………………………………………………………………………………….. 38

A Brief Précis of the History and Development of Liquid assisted Pulmonary Cooling (LAPC) for Induction of Hypothermia during CPR

………………………………………………………………………………………………………………….. 41

What LAPC Can Potentially Deliver

………………………………………………………………………………………………………………….. 46

Section 2: Experimental Studies to Determine the Effectiveness of LAPC under Laboratory Conditions ……………………………………………………………………………………………………… 51

1. Introduction

………………………………………………………………………………………………………………….. 52

2. Materials and methods

………………………………………………………………………………………………………………….. 53

2.3. Trial I (manually-controlled LAPC)

………………………………………………………………………………………………………………….. 59

2.4. Trial II (machine-controlled LAPC)

………………………………………………………………………………………………………………….. 59

2.5. Animals A, B and C

………………………………………………………………………………………………………………….. 59

2.6. Data collection and correction, statistical methods, graphical display and presentation

………………………………………………………………………………………………………………….. 59

3. Results

………………………………………………………………………………………………………………….. 60

3.1. Thermal results of LAPC

………………………………………………………………………………………………………………….. 62

3.1.1. Cooling time delay

………………………………………………………………………………………………………………….. 63

3.1.2. Cooling rate

………………………………………………………………………………………………………………….. 63

3.1.3. Mean cooling power

………………………………………………………………………………………………………………….. 63

3.2. Gas exchange

………………………………………………………………………………………………………………….. 64

3.3. Clinical observations and gross pathology

………………………………………………………………………………………………………………….. 66

3.4 Impact on Hemodynamics

………………………………………………………………………………………………………………….. 67

4. Discussion

………………………………………………………………………………………………………………….. 69

4.1. Apparent effect of temperature on gas exchange

………………………………………………………………………………………………………………….. 69

4.2. Thermal transfer efficiency and kinetics

………………………………………………………………………………………………………………….. 70

4.3. Question of diffusion dead space in LAPC

………………………………………………………………………………………………………………….. 71

4.3.2. Possible synergy of combined gas and liquid ventilation in assisting mass (CO₂) and heat transfer

………………………………………………………………………………………………………………….. 73

4.4. Potential development of clinical LAPC

………………………………………………………………………………………………………………….. 73

4.5. Challenges Ahead

………………………………………………………………………………………………………………….. 74

Acknowledgements…………………………………………………………………………………… 81

Appendix A……………………………………………………………………………………………….. 81

Section 3: Perflurochemicals………………………………………………………………………… 96

The Perflurochemicals

………………………………………………………………………………………………………………….. 97

Physical Chemistry and Synthesis

………………………………………………………………………………………………………………….. 97

Physical Properties

………………………………………………………………………………………………………………….. 98

Commercially Available PFCs

………………………………………………………………………………………………………………….. 99

Toxicology

………………………………………………………………………………………………………………… 102

Environmental Impact and Future Availability

………………………………………………………………………………………………………………… 104

Section 4: History of Liquid Assisted Ventilation and Implications for LAPC

………………………………………………………………………………………………………………………109

History of liquid Ventilation

………………………………………………………………………………………………………………… 110

Partial Liquid Ventilation (PLV)

………………………………………………………………………………………………………………… 114

Unanticipated Effect of Lung Protective Ventilation Strategies

………………………………………………………………………………………………………………… 119

Defective Translational Research Models

………………………………………………………………………………………………………………… 119

Failure to establish a Dose-Response Curve

………………………………………………………………………………………………………………… 120

Gas Trapping and Selection of Appropriate PFCs

………………………………………………………………………………………………………………… 120

The PFC-Air Interface and Shear Effects in Small Airways

………………………………………………………………………………………………………………… 121

PLV and the Law of Laplace

………………………………………………………………………………………………………………… 122

The Best as the Enemy of the Good

………………………………………………………………………………………………………………… 127

Possible Implications for SCA and LAPC

………………………………………………………………………………………………………………… 127

__________________________________________________________

This work is dedicated to David W. Crippen, MD, FCCM, who, literally, made it possible. The computers this work was written on, the myriad books and journals consulted, and the liaison with many of the professionals required for such work were all directly facilitated by Dave Crippen. But, these are the least important elements he contributed. By far the most critical ingredients were and are his faith and friendship. For these he has my unending thanks.

___________________________________________________________

 
Intellectual Property Rights: The intellectual property rights to the technology of liquid assisted pulmonary cooling (heat exchange) (LAPC or LAPHE) using combined gas and fractional tidal liquid ventilation are controlled by Critical Care Research, Inc., under US Patent 6,694,977, “Mixed-mode liquid ventilation gas and heat exchange.” Inquiries regarding access to this technology should be directed to Critical Care Research, Inc., 10743 Civic Center Drive, Rancho Cucamonga, CA 91730-3806. Telephone: (909)987-3883.

Liquid Assisted Pulmonary Cooling in Cardiopulmonary Cerebral Resuscitation

By Michael G. Darwin

Introduction

Liquid assisted pulmonary cooling (LAPC), or liquid assisted pulmonary heat exchange (LAPHE) will not likely be applied in the West any time in the foreseeable future for many reasons; not a few of them having to do with regulation and consumer perception of what constitutes acceptable risk versus benefit.

LAPC relies completely upon the unique properties of the perflurochemicals (PFCs). PFCs are other-worldly molecules; neither soluble in water or lipids, twice as dense as water, but much less viscous; and chemically so inert as to make gold seem a highly reactive metal by comparison. It is impossible to work with the PFCs suitable for introduction into the lungs of mammals without immediately appreciating how truly amazing they are. PFC chemistry exists in a world of its own outside of conventional chemistry, and in fact, conventional chemists refer to PFC chemists as ‘unnatural’ chemists; with the all the double entendre that moniker implies. PFCs do not occur in nature, and if you want to find a marker for intelligent, industrial life elsewhere in the universe, one approach might be to look for the spectra of PFCs in the atmosphere of the candidate planet; they occur only as a result of deliberate, intelligent industrial activity. Thus, they are what I call, ‘the thinking species’ molecule.’

There is an ancient Chinese curse: “May you live in interesting times and come to the attention of important people.” The medical-molecular equivalent is: “May you be synthesized in interesting times and come to the attention of important medical (and environmental) bureaucrats.” This perfectly describes the fate of the medically useful liquid PFCs. They are dramatic molecules and they are at once both therapeutically paradigm changing and physiologically outrageous. There is a strong emotional reaction – one part wonder and one part shock – at seeing a patient’s lungs filled up with liquid, any liquid. Bulk liquid does not belong in mammalian airways, and in every clinical instance where it occurs, or is introduced into the lungs – in pulmonary edema, drowning and inhalation of organic liquids – it is an unmitigated disaster. The ‘breathable’ PFCs thus attracted a lot of attention.

So, PFCs had this strike against them, and they came about as putative therapeutic agents in ‘interesting’ times in the most Chinese sense of the word. The explosion of technological innovation that has occurred since the Industrial Revolution has brought not only powerful advantages, but also often damaging and not infrequently frightening downsides. We live in an age when people, at least in the West, have become acutely sensitized to the dark side of technology and increasingly unwilling to accept risk unless it is precisely quantified, small, and will almost certainly not affect them.

The PFCs present some unfortunate challenges in this regard. First, while they are among the most chemically inert substances known to man, they are biologically active in fentogram quantities. Most of the medically useful PFCs abolish white blood cell chemotaxis in picogram to fentogram concentrations.  While their very inertness and lack of chemical reactivity make them indispensable as a liquid ventilating medium, these same properties mean that, should they escape the lungs due to pneumothorax (and enter the mediastinum or other closed body viscuses or capsules) they will likely remain there for the rest of the patient’s life. This may seem ample reason to eschew these molecules as therapeutic agents. However, I believe these facts must be taken in the context of how we use and abuse similar (and in some cases nearly identical) molecules both in medicine and in our daily lives.

Introducing liquid into patients’ lungs will never be a trivial matter, or something undertaken lightly. Such therapies and their enabling molecules will not be broadly used (or abused) in medicine. Indeed, their use is necessarily confined to the emergency medical and critical care setting in intubated and mechanically ventilated patients. Foreseeable off-label medical use of neat PFCs is virtually nil. And yet, these molecules will ‘seem unlikely to be approved for medical use in the West in foreseeable future.

All of the existing molecules are either unsuitable for use in the induction of hypothermia (such as Perflubron™ which freezes at +6 deg C) or are chemically heterogeneous, as I explain in this manuscript. It would thus be necessary to synthesize and purify new classes of molecules and then vet them individually for safety and efficacy. The cost of doing this is more than any Western drug company could bear when weighed against even the most optimistic projections of financial return on the investment. There is very little place for a drug that a patient will use once during a lifetime and that, if it works, will leave him more than able to sue all involved in its application to him should he develop some adverse effects, even if these occur only after decades of additional, healthy and productive life which would not have been possible otherwise.

I hasten to add that I do not think the PFCs will be free of long term side effects. Most of the chronic survival dogs in our studies that underwent PFC ventilation and/or cooling have since grown old and died. We saw nothing unusual; they died of heart failure, cancer and the usual things old dogs die of, and seemingly at the usual rates, and at the usual ages. But, our sample size was ‘microscopic;’ and I think it likely that large populations of people treated with PFCs may have a statistically significant increase in conditions related to the immunomodulating and immunosuppressive effects of these molecules; possibly more infections, more serious infections, and more neoplasms. My answer to that is, ‘so what?’ We accept adverse effects in oncology, chronic hemodialysis, chronic circulatory support (left ventricular assist devices) and solid organ transplantation that are, by comparison, the stuff of nightmares. We know and accept that children who are given radiation and chemotherapy have a good chance of surviving many childhood cancers, but very often at the expense of lowered IQs and a lifetime complicated by greatly increased risk of infection and of cancers unrelated to their original diagnosis. The liquid ventilating PFCs are benign compared to OTC aspirin which claims ~50,000 lives each year; mostly in treating nothing more life threatening than the pain of osteoarthritis, headaches and joint and muscle pain. And yet, the PFCs will not be approved for liquid ventilation or heat exchange applications in the foreseeable future; not here in the US, and not anywhere else in the West.

In the Russian Federation these very same molecules are approved and are being used parenterally in the form of any oxygen carrying ‘blood substitute’ called Perftoran.[1] It isn’t the best such oxygen therapeutic possible (Alliance’s Oxygent™ was far superior[2]), but it works and it is an acceptable risk; at least Russians physicians and regulators think so (and for what it is worth, I agree). The Chinese, vastly more Westernized than the Russians, have still to approve the first PFC-based blood substitute (or liquid ventilating media, for that matter). Nevertheless, it will be in one of these places, or someplace like them, not in the US, Australia or Europe, where LAPC/LAPHE is first clinically applied. So, the information present here will necessarily be of mostly theoretical interest. Hopefully it will find distribution in parts of the world where it will be of some practical value as well.

 Mike Darwin,

22 October, 2008

TABLE OF ABBREVIATIONS, SYMBOLS & ACRONYMS

Section 1:

Introducing Liquid Assisted Pulmonary Cooling

Liquid Assisted Pulmonary Cooling in Cardiopulmonary Cerebral Resuscitation

By Michael G. Darwin

Introduction

Each year in the United States there are ~450,00 deaths from myocardial infarction (MI) [1] (with 310,000of these deaths occurring before the patient reaches the hospital) as a result of a non-perfusing arrhythmia, principally ventricular fibrillation.[2] This mode of sudden cardiac arrest[3] (SCA) is also responsible for the majority of the 190,000 in-hospital deaths from MI, which typically occur within the first 24 hours following admission.[3]   Especially tragic is that 50% of these deaths occur in persons ~60 years of age or less.[4]  An estimated additional 20,000 incidents of SCA occur as a result of asphyxiation, drowning, electrocution, and genetic or developmental predisposition to lethal arrhythmias (Wolf-Parkinson’s White Syndrome, congenital thickening of the interventricular septum, and idiopathic arrhythmic disease) and other non-atherosclerosis causes. This latter category of SCA typically occurs in individuals whose mean age is less than 35.[5],[6]

At this time the principal treatments for SCA consist of initiation of manual, ‘bystander’ cardiopulmonary resuscitation, so-called Basic Cardiac Life Support (BCLS or BLS) followed by ‘definitive’ treatment of the arrhythmia beginning with defibrillation and the application of Advanced Cardiac Life Support (ACLS or ALS).[7]

 Figure 1-1 (right): Mortality from sudden cardiac arrest (SCA)in 2004  as a result of myocardial infarction compared to death from other ‘high profile’ causes of mortality in the US.

ACLS consists of the application of an algorithm of manual CPR, electrical defibrillation and pharmacologic therapy aimed at restoring a perfusing cardiac rhythm and adequate blood pressure and cardiac output to sustain life until definitive treatment of the underlying cause of the cardiac arrest can be achieved (e.g., coronary revascularization, implantation of an automatic defibrillator, or life-long anti-arrhythmic therapy).

 Figure 1-2 (right): Probability of survival as a function of time following cardiac arrest.[8]

As is shown in Figure 1-6 below, the time to survival without neurological deficit following cardiac arrest in the absence of BCLS declines rapidly following a sigmoid curve with survival without neurological deficit being ~80-90% following 1 minute of arrest time, and less than 10% following 9 minutes of arrest.[8] Put another way, 50% of patients will experience significant morbidity or death following 4 minutes of circulatory arrest (Figure 1-2).

What is not shown in this graph is that the effect of immediate bystander CPR on survival is negligible in most studies [9],[10] with the primary benefit being observed in patients who’s time from the initiation of BCLS to successful cardiac resuscitation was greater than 8 minutes.[11] There is evidence in the literature that morbidity is improved with prompt by-stander CPR [12] providing that EMS response is also rapid, although this remains controversial.[11],[13]  A corollary of this is that the overall survival rate following SCA, with or without serious neurological morbidity, ranges between 1% (New York City, NY) [14] to 17% (Seattle, WA).[15] The mean survival (defined as survival to discharge from the hospital) in the United States as a whole is generally agreed to be at best 15%  [16] with ~70% of these patients experiencing lasting neurological morbidity (ranging from ‘mild’ cognitive impairment to total incapacitation in the Persistent Vegetative State (PVS).[17],[18],[19]

The primary cause of non-survival in patients experiencing SCA is failed cardiac or cerebral resuscitation. Arguably, it is failed cerebral resuscitation, since most underlying causes of refractory cardiac arrest could be treated by ‘bridging’ supportive technologies such as emergency femoral-femoral cardiopulmonary bypass (CPB) until myocardial revascularization and hemodynamic stabilization were achieved.[20] When emergency CPB is applied to patients who are candidates for good neurological outcome, the survival rate is increased.[21],[22],[23],[24] However, these technologies are not typically used on patients who are unsuccessfully resuscitated (restoration of adequate cardiac rhythm and perfusion) because of the justified perception that irreversible brain damage would have occurred during the prolonged period of cardiac arrest or CPR/ACLS.[21]  Similarly, it is for this reason that most attempts to achieve cardiopulmonary resuscitation in hospitalized patients who are not hypothermic or intoxicated with sedative drug are terminated after 15 minutes.[25],[26]

Within medicine it is widely understood that ‘CPR doesn’t really work’ and that if the return of spontaneous circulation (ROSC) is not achieved within ~ 5 minutes of cardiac arrest, the chances for survival are slim, and the chances for survival absent neurological impairment are slimmer still.[8] The principal reasons that conventional CPR is not effective are that it fails to supply an adequate amount of flow at an adequate pressure. Cardiac output (CO) is typically ~1/3^rd of the at-rest requirement (~1.5 versus ~4.5 liters per minute), and mean arterial pressure (MAP) is typically 25 mm Hg to 45 mm Hg; well short of the 60 mmHg required to sustain cerebral viability.[27],[28]

The condition of the typical sudden cardiac arrest (SCA) patient and the circumstances under which he experiences cardiac arrest are far from the ideal of a patient who is a candidate for emergency cardiopulmonary bypass (CPB) in hospital. The typical SCA patient is middle aged or elderly, often suffering from one or more co-morbidities (diabetes, obesity, COPD, hypertension), and if subjected to prolonged CPR will invariably have impaired gas exchange due accumulation of fluid in both the parenchyma and the air-spaces of the lungs (pulmonary edema with alveolar flooding). This occurs because closed chest CPR quickly causes pulmonary edema.[29],[30]  As previously noted, even when the SCA patient is a ‘good’ candidate for salvage; someone who is relatively young and free of co-morbidities, CPR will likely prove futile due to cerebral ischemia-reperfusion injury and the post-resuscitation syndrome.

Over the past 25 years a vast number of therapeutic interventions have shown great promise in animal models of regional and global cerebral ischemia in the laboratory.[31],[32],[33],[34] In the last 6 years alone, over 1000 experimental papers and over 400 clinical articles on pharmacological neuroprotection have been published.[35],[36] However, with one exception, none of these interventions has been successfully applied clinically despite many attempts. [37],[38],[39],[40],[41],[42],[43],[44] The sole exception to this frustrating debacle has been the introduction of mild therapeutic hypothermia (MTH) as the standard of care for a select (and very small) minority of SCA patients.[45],[46],[47],[48],[49],[50],[51]

Hypothermia as an Active Therapeutic Agent

Since the demonstration by Safar, et al., of the neuro-salvaging effects of mild systemic hypothermia after prolonged cardiac arrest in dogs [52],[53] there has been an explosion of translational research which has lead to a transformation in our understanding and application of mild hypothermia.[54], [55] Once seen solely as a protective tool which conferred benefit by reducing metabolism, it has become clear that mild hypothermia (33°C–35°C) [56] has therapeutic effects which appear to be primarily anti-inflammatory and anti-apoptotic in nature, and which operate independently of hypothermia’s effect on metabolic rate.[57],[58] Table 1-1 reviews some of the known pro-inflammatory factors inhibited or moderated by mild therapeutic hypothermia (MTH) and documents the supporting literature.

 Table 1-1: Inhibition of Injury Cascades by Mild Therapeutic Hypothermia (MTH)

Reproduced with modifications from Zhao, H., Steinberg, GK, Sapolsky, RM., General versus specific actions of mild-moderate hypothermia in attenuating cerebral ischemic damage. J Cerebr Blood Flow Metab, 2007. 27: p. 1879-1894.

Table 1-1:  Intraischemic hypothermia delays or attenuates both ATP depletion (Ibayashi et al, 2000; Sutton et al,1991; Welsh et al, 1990) and anoxic depolarization (Bart et al, 1998; Nakashima and Todd, 1996; Takeda, et al, 2003), it also blocks glutamate release (Busto et al, 1989b; Patel et al, 1994; Winfree et al, 1996), suppresses inflammation (Kawai et al, 2000; Wang et al, 2002), maintains the integrity of the BBB (Dietrich et al, 1990; Huang et al, 1999; Kawanishi, 2003), reduces free radical production (Maier et al, 2002), inhibits protein kinase C translocation (Cardell et al, 1991; Shimohata et al, 2007a, b; Tohyama et al, 1998), inhibits matrix metalloproteinase expression (Hamann et al, 2004), and blocks both necrosis and apoptosis. Intraischemic hypothermia also preserves the base-excision repair pathway, which repairs oxidative damage (Luo et al, 2007). In addition to those cascades directly associated with neuronal injury, hypothermia further blocks astrocyte activity and inhibits white matter injury (Colbourne et al, 1997; Dempsey et al, 1987; Kimura et al, 2002). Similarly, postischemic hypothermia blocks free radical generation (Horiguchi et al, 2003), attenuates inflammation (Horstmann et al, 2003; Ohta et al, 2007), prevents BBB permeability (Preston and Webster, 2004), and suppresses caspase activities (Van Hemelrijck et al, 2005). Indeed, a browse through the literature gives an overwhelming impression that hypothermia seems to block every damaging event associated with necrosis or apoptosis. One reason for this impression of pan-inhibition may lie in the causality of ischemic damage.  For example, is the inflammatory response the cause of tissue damage or is it induced by brain injury? If it is the latter, then since hypothermia prevents tissue damage, it certainly also prevents the inflammatory response.

No doubt, the commonly cited ‘obstacles’ of lack of institutional protocols, lack of physician education about the benefits and guideline changes, as well as the inevitable inertia that accompanies any paradigm shift in treatment are playing a significant role in the failure of MTH to become the practiced standard of care for the post resuscitation syndrome.[64],[60] However, what is not being said, or considered, is that while MTH as currently practiced represents a large relative improvement in outcome, the benefits are still modest in absolute terms. Only a miniscule subgroup of SCA patients currently can benefit from MTH; and even in its best clinical implementation MTH still fails to rescue ~60% of that sub-group of SCA patients to whom it is applied.[65],[66],[67],[68],[69] This is in stark contrast to what can be achieved with MTH in ameliorating post-ischemic encephalopathy in the laboratory, where post-resuscitation MTH consistently provides rescue with stunning efficacy.[70],[71]

Figure 1-3: The impact of a delay of 10 min in inducing MHT is a dog model of cardiac arrest followed by 3 min of systemic ischemia, 7 minutes of mechanical CPR and 50 minutes of advanced life support. Hypothermia to 34^oC was induced beginning at 10 min post arrest in the early hypothermia group˜ and at 20 min post arrest in the delayed hypothermia group ¢.  In the early hypotherrmia group group, 5 of 7 surviving dogs were functionally normal (OPC 1 or 2), 1 had OPC 3, and 1 had OPC 4 (coma) at 96 hours of recovery. Histologically, 4 of 8 dogs in this group were normal (HDS 0), 1 had HDS 16, 1 had 22, and 1 had 98. The only surviving dog in the DH group was functionally normal at 96 hours (OPC 1, NDS 0) with an HDS of score of32 (mild injury) Due to early mortality only two other dogs in the delayed hypothermia group were evaluated histologically and their  HDS scores 38 and 45, respectively. Dogs in this study were scored by ‘overall performance categories’ (OPC; 1=normal, 2=moderate disability, 3=severe disability but conscious, 4=coma, and 5=death) Neurological function and  neurological deficit scores (NDS; 0% to 10%=normal, 100%=brain death). [72],[73] Histological damage scores were obtained by neuropathological examination of 19 distcrete brain regions for severity and extent of ischemic neuronal changes, infarcts, and edema.  A total brain histological damage score (HDS)  >40 represented moderate damage, and HDS >100 represented severe damag.[74] Redrawn from Nozari, A., et al., Critical time window for intra-arrest cooling with cold saline flush in a dog model of cardiopulmonary resuscitation. Circulation, 2006. 113(23): p. 2690-6.

The primary obstacle to realizing this bonanza in translation research has been the practical impossibility of achieving systemic cooling within the narrow therapeutic window demonstrated in animal models of SCA and resuscitation.[75],[67],[65],[66],[67],[68],[69] If the clinical outcome of MTH was even half that achievable in the laboratory, widespread application would likely have been rapid and uniform; there is rarely resistance to the ‘miraculous’ if it is simple, easy to understand, biophysically well characterized and highly cost-effective. MTH applied immediately post ROSC would be all of these things.[4]

Figure 1-4: The results of the Nozari, et al., [71] study on the effects of delayed MTH are presented graphically at right with the addition of historical controls from the literature treated similarly, but with no hypothermia (no survivors). This graphic illustrates the potency of truly rapid post arrest hypothermia in ncreasing survival.

The data in Figures 1-3 and 1-4 exemplify what is possible when MHT is induced within its optimum therapeutic window of 0-15 min post ROSC versus a delay of even 10 minutes. In this study by Nozari, et al., of PeterSafar’s group, [71] VF was electrically induced in 17 dogs all of whom were subjected to a period of 3 minutes of no flow beginning when the MAP dropped below 30 mm Hg, followed by 7 minutes of mechanical CPR and 50 minutes of advanced life support during which time VF was maintained and mechanical CPR was continued. Nine animals were treated with rapid (early) induction of MTH to 34^oC starting at 10  min post arrest (EH group) (concurrent with the start of ALS to simulate the time course of arrival of EMS paramedics) using a combination of cold IV saline and veno-venous heat exchange. Induction of hypothermia was not

Figure 1-5: Results of a study of 48 cardiac arrest patients treated with MTH via endovascular cooling. A strong correlation was found between rapidity of cooling and both neurological outcome and serum neuron specific enolase levels. Left: Time course of MTH among patients with good and poor neurological outcome. The curves indicate the course of mean body core temperature during MTH among patients with good (¢) and those with poor (p) outcome as well as in the entire (▬) patient group. Right: Correlation between time to coldest temperature (minutes) and the maximum NSE values (μg/L). Normal serum NSE is 9.6±0.7 μg/L. Redrawn from Wolff B, et al., Early achievement of mild therapeutic hypothermia and the neurologic outcome after cardiac arrest, Int J Cardiol (2008).

begun until 20 min post arrest in the delayed hypothermia group (DH group) which consisted of 8 dogs. Target core temperature was achieved at 6.0±2.7 minutes after the initiation of cooling (3.5 minutes after the start of veno-venous cooling) in both groups. The delay from arrest to reaching ~34^oC  was 16.6 min in the EH group and 25.4 minutes in the DH group.

After 60 minutes of VF, ROSC was achieved with cardiopulmonary bypass for 4 hours, and intensive care was given for 96 hours. In the early hypothermia group, 7 of 9 dogs survived to 96 hours, 5 with good neurological outcome. By contrast, in the delayed hypotherrmia group 7 of 8 dogs died of multiple organ failure within 37 hours (P=0.012);  3 animals in secondary VF that was resistant to CPR with antiarrhythmic treatment and repeated defibrillations. Only one dog in the EH group died, and that animal succumbed to single organ failure; pulmonary edema with hemoptysis. This study extends the previous work by this group documnting an optimum therapeutic window for MHT (in dogs) of ~10-15 min.[76],[76],[77] The therapeutic window of MTH after cardiac arrest has been demonstrated to be similarly short in other species.[78],[79],[80],[81]

The dramatic efficacy of MTH in the laboratory made quick converts of the pioneering researcher-clinicians who forged ahead with the application of MTH to SCA in the clinic precisely because it was dramatic; indeed it was as close to the miraculous as interventions in medicine come. The real barrier to translating that ‘miracle’ to everyday practice has been the seemingly intractable problem of achieving cooling over the same time course that has proven so effective in the research setting.

The problem is that the optimum therapeutic window for the treatment of cerebral ischemia-reperfusion injury appears to be in the range of 0 to 15 minutes post ROSC. One of the first follow-up studies on MTH carried out by Safar, et al., demonstrated that in a standardized model of cardiac arrest in dogs a delay in the application MTH of as little as15 min after ROSC abolished most of the benefit.[77],[82] While the work of Bernard, et al., [46] and that of the Hypothermia after Cardiac Arrest Study Group [48] demonstrated that delays in cooling of up to 2-3 hours post ROSC in humans still have sufficient clinical utility to justify the routine application of MTH in a selected group of SCA patients, this benefit is marginal when contrasted to that achievable in the laboratory when MTH is rapidly induced during the first 15 minutes after ROSC.[77] Thus, the optimum clinical benefit of MTH in ischemic and very likely traumatic, CNS injury requires the ability to achieve very rapid core cooling.[83]

Figure 1-6: Survival after cardiac arrest declines rapidly as a function of time to ROSC, exhibiting the sigmoidal curve shown at left, above (¢>¢), with essentially all patients failing to survive with normal mentation after arrest intervals of @ 10 min. Application of MTH (¢) within the window of @ 15 min offers the promise of squaring the survival curve in SCA of @ 10 min duration yielding a survival rate of ~65-70% with little or no neurological deficit. Application of deep (10-22^oC) or profound hypothermia (5-9^oC) (¢) may allow survival after intervals of as long as 1-2 hours of CPR. Graphic by M.G. Darwin

In the clinical arena the time to reach the target core temperature under ‘good’ circumstances is in the range of 3-4 hrs; not 10 to 30 min, as is the case in the laboratory. Even with such long delays in cooling the adverse effect of delay is still present. Wolf, et al., recently published a study of 49 out of hospital cardiac arrest patients who were treated with MTH (32.0-34.0°C; with a target temperature of 33.0°C) of 24 h duration using endovascular cooling.[69]  The study endpoints were neurological outcome on discharge from hospital and serum neuron specific enolase (NSE) levels (a sensitive and specific marker of neuroinjury) at 24 h intervals to 3 days (Figure 1-5).

Figure 1-7: The graph above shows the hypothesized relative effect on survival of effectively administered CPR started at 5 min post-arrest followed by defibrillation at 6 min and ACLS at 8 min post arrest (~30% survival). The light blue shaded area of this graph shows the expected improvement in survival if MTH is induced at the start of ACLS (8 min post-arrest) and target temperature is reached by 15 min post ROSC. The dark blue shaded area shows the potential of Emergency Preservation resuscitation (EPR) using moderate (10-22^oC ) or profound (5-9^oC) hypothermia to not only square the curve of survival with CPR, but to facilitate survival in patients who would otherwise not benefit from either BCLS or ACLS (i.e., refractory to defibrillation, hypovolemic, etc.). Graphic by M.G. Darwin

 As is the case in laboratory studies of MTH in cardiac arrest in dogs, Wolff, et al., found that neurological outcomes were binary, with no patients who survived experiencing moderate degrees of disability; patients either recovered well with no or /mild neurological impairment, or experienced severe disability (n = 1) coma or PVS (n = 6). Twenty-eight patients were discharged with a good outcome and a strong correlation was found between good outcome and the time interval from the start of cooling to the lowest temperature (p =.035) and a less robust correlation with the time to reach target temperature (p=.071). Similarly,

NSE levels were found to correlate well with the time required to reach the lowest temperature achieved in each patient (Figure 1-5).

Even with delays in the start of cooling that averaged 2.5 hrs; and a mean time to reach target core temperature of 6.8 hrs, additional injury accruing from slowness in cooling was still clinically and biologically apparent. Despite the homogeneity of the patients, their arrest times, and their course of treatment, ~60% of the patients in this study either did not survive, or were comatose or PVS.

The Benefits and Limits of ‘Delayed’ MTH: Real World Experience

To understand the benefits and limits of MTH when it is aggressively and competently implemented with currently available technology, it would be hard to find a better example than that of Wake County, SC. Wake County, is located in the northeast central region of North Carolina and is part of the Research Triangle metropolitan area, which consists of Raleigh, Durham, Chapel Hill, and surrounding urban and suburban areas. The area serviced by the Wake County Emergency Medical System (Wake EMS) has a population of 832,970 (as of 2007). The Wake County EMS operates 35 ambulances from 23 locations with 825 ALS personnel; ambulances are staffed with two paramedics 95% of the time and there is always one paramedic responding.[84] In 2002 the Wake EMS answered more than 50,000 medical requests for service. Based on the latest national data Wake County ranks third in the US for recovery from “survivable” cardiac arrests (primarily ventricular fibrillation-ventricular tachycardia). Nationally, the average survival rate is 17% for patients presenting with these arrhythmias.

 Figure 1-8: Improvement in overall survival of patients in cardiac arrest in response to the phased introduction of CPR per AHA 2005 Guidelines, use of an impedance threshold device (ITD) and in-field induction of MTH.[85]

 Beginning in January 2004 Wake EMS initiated a study to evaluate the efficacy of CPR as they then practiced it, and to evaluate the effectiveness of the impending change in the American Heart association (AHA) guidelines for CPR, the introduction of an impedance threshold valve (the ResQPod™) and Wake EMS’ planned implementation of the ILCOR Guidelines for post-arrest MTH.[85],[86] From January 2004 until April of 2005, Wake EMS personnel employed the then extant AHA guidelines, which mandated an emphasis on intubation and a 15:2 compression-to-ventilation ratio; with interruption of chest compressions for ventilation. This period constituted the baseline of the study, and data were collected per protocol; not gathered retrospectively.

During the baseline period survival to discharge from hospital was 2.4% for all patients given CPR and 12.1% for patients with ventricular fibrillation-ventricular tachycardia (VF-VT) arrhythmias. In April 2005 Wake EMS implemented continuous cardiac compression CPR with a 30:2 compression-to-ventilation ratio with emphasis on no, or very minimal, interruption of chest compressions.  After 12 months, the overall survival rate had risen to 4%; and had more than doubled to 21.8% for patients who presented in VF-VT.

In April 2006, Wake EMS added the use of an impedance threshold device (ITD) to improve cerebral and coronary perfusion during CPR. Introduction of the ITD resulted in an increase in overall survival to 4.5% and an increase in the survival of patients with VF-VT to 28.5%.

  Figure 1-9: Dramatic improvement in neurologically intact survival as a result of the phased introduction of CPR per AHA 2005 Guidelines, use of an impedance threshold device (ITD) and in-field induction of MTH.[85]

 The final phase of the investigative protocol began in October of 2006 when Wake EMS added in-field induction of MTH to the two previous interventions. MTH was induced using a combination of external cooling employing ammonium nitrate-water eutectic ‘instant cold packs’ applied to the axilla and groin, and  cold IV saline (1-2^oC, 30mL/kg to a maximum of 2 liters) given rapidly via two large bore catheters and/or intraosseous infusion. Criteria for induction of hypothermia were that the patient have ROSC and show no return of consciousness (Glasgow Coma Score (GCS) <8). Induction of hypothermia was initiated two to three minutes after ROSC. There was heavy emphasis on avoiding over-ventilation and on attempting to maintain end-tidal CO₂ (EtCO₂) at a minimum of 40 mm Hg.  Patients undergoing MHT were sedated with etomidate, paralyzed with vercuronium and given a titrated dopamine drip to maintain mean arterial pressure (MAP) between 90-100 mm Hg. The mean time to target temperature (34^oC) in this study was extraordinarily short: 68 minutes (95% CI 47 to 88); compared to the 2-3 hours typically required to induce MTH.

With the combination of continuous compression CPR, use of the ITD and prompt application of MTH, survival rates for the 12 months from October of 2006 to October 2007 had increase to 6.7% overall and 37.4% for patients with VF-VT. The odds of overall survival increased three-fold (95% CI 1.7 to 5.0) and the odds of survival for patients in VF-VT increased 4.3-fold (95% CI 2.2 to 8.6) from the beginning of the study (Figures 1-8 and 1-10).The probability of a good neurological outcome increased from 20% at baseline to 80% at the conclusion of the study (Figure 1-9). In a multivariate analysis, the odds ratios for survival for each phase of implementation were as follows:

Figure 1-10 (right): Multivariate odds for all factors in outcome evaluated during the Wake EMS study. MTH was by far the most powerful intervention. As in most previous studies of survival factors associated with CPR age and residence (home versus extended care or assisted living facility) had only modest impact on survival.[85]

• New CPR protocol: 2.13 (95% CI 1.12 to 4.04)
• Addition of impedance threshold device: 2.33 (95% CI 1.09 to 5.00)
• Addition of early hypothermia: 3.99 (95% CI 2.19 to 7.27)
• Patients who received bystander CPR 1.79-fold (95% CI 1.18 to 2.72) more likely to survive.

Figure 1-11 (right): The Engel-15 portable, (compressor-type) refrigerator/freezer has a 14 L capacity, weighs 11.5 kg and can maintain 12-13 liters of saline at 1-2^oC at ambient temperatures as high as 40^oC . It retails for ~$380 US. [Photo courtesy of Engel, Ltd., Australia]

Interestingly, all three elements of the Wake EMS protocol were implemented at a cost of less than $200 per patient. Due to budget constraints, Wake EMS chose simple, inexpensive commercial products for refrigeration of IV fluid and implementation of external cooling, as opposed to more costly products developed specifically for medical application, such as the EMCOOLS surface cooling system (Emergency Medical Cooling Systems, AG, Austria). Saline was kept at the requisite temperature of 1-2^oC with a compact, 12V operated, consumer travel refrigerator/freezer (Figure: 1-11, Engel-15: http://www.i-m-d.com/) and surface cooling was with generic ammonium nitrate cooling packs.

The Wake County EMS program is extraordinary in every way. It represents the best application of the best available technology by arguably some of the best medical and paramedical personnel in the US. The mean time to target temperature of 68 minutes is unprecedented in any clinical study employing MTH. Of the 359 patients who participated in the study (all comers) after MTH was in place; 25 survived. In the subgroup of 93 patients who presented with VF-VT; 34 survived, with 78% or 27 patients being discharged with a good neurological outcome. Put another way 92% of patients who presented under the most favourable circumstances (VF-VT), treated with the best currently available interventions, at the fastest rate of cooling so far reported, failed to survive or did so with profound neurological debility.

The primary difference between the survivors and the profoundly disabled or dead was the development of the post-resuscitation syndrome and the primary reason for this complication was not comorbidity, or delay in paramedical assistance, but rather delay in the rapidity of cooling which, if achieved within the first 15 min post ROSC, would have offered the prospect of neurologically intact survival in the range of 70-80% in patients presenting with VF-VT, and 30-40% in all comers.

These interventions, remarkable achievements that they are, do not escape from the harsh reality that the 400% increase in survival from cardiac arrest in Wake County, when expressed in absolute terms, means that the number of lives saved increased from ~5 to 25 – out of 395 SCA patients; a huge relative gain, but a comparatively small increase in the absolute number and percentage of lives saved, and minds salvaged. The true life saving potential of MTH remains elusive by virtue of its exceedingly small therapeutic window.

 The Problem of Heat Exchange

Because of this minute therapeutic window, there is a pressing need to achieve rapid and durable core cooling of patients during CPR by simple, easily accessible means.  External cooling is only effective at reducing core temperatures by 0.15 to 0.25ºC/min in the average patient undergoing CPR (Figure 1-12) and this is achieved only by complete immersion of patients in a stirred ice water bath.

The Efficacy of External Cooling in Four Cryopatients[5]

 Figure 1-12:  Comparison of the cooling rates of four cryopatients. Immediately following pronouncement of medico-legal death patients were given closed chest mechanical cardiopulmonary support and placed in a stirred ice water bath for induction of hypothermia. Epinephrine was administered as per ACLS guidelines; thus peripheral vasoconstriction would be expected to be comparable to that seen in the typical SCA patient undergoing cardiac resuscitation.

The most effective external cooling achieved by a commercial system using direct, whole body surface cooling employing circulation of ice water (ThermoSuit,™ Life Recovery Systems, Kinnelon, NJ)  is probably the work of Janata, et al., using human human-sized swine.[87]

They were able to achieve core cooling at a rate of 0.3^oC/min; however it is important to note that the animals in this study were not in cardiac arrest while undergoing CPR in the presence of profoundly peripherally vasoconstricting agents, such as epinephrine or vasopressin; as would usually be the case during ACLS in humans [59] and which is known to further slow surface cooling.[88]

 Figure 1-13: The Life Recovery Systems ThermoSuit™ employs direct ice water contact with the patient’s skin to achieve the maximum possible rate of cooling by external means. The system consists of an inflatable insulating and water containment patient enclosure inside of which the patient rests on a mat of Dacron bonded polyester ‘wool’ which acts to diffuse and film water pumped over the dorsal surface of the patient’s body. Water at 2-4^oC is thin-filmed over the ventral surface of the body by a thin, transparent blanket with many hundreds of small perforations through which water under pressure pours out and over the patient. Cold water is recirculated over crushed or cubed ice in an insulated reservoir containing a disposable liner and pumps. Cooling is computer controlled via a thermistor which can be placed in any desired anatomical location. All patient contact items are single-use and disposable (including, as previously mentioned, the pumps) http://www.life-recovery.com/.

The obvious problems with this system are its bulk (Figure 1-13), likely high cost, lack of ease in field deployment (again related to its bulk and weight) and the intrinsic physiological problems associated with the induction of hypothermia via external cooling. As extensively discussed in Section Two, the mammalian body consists of multiple thermal compartments transiently ‘isolated’ from each other by differences in blood flow and heat conductivity.[89] Broadly, these compartments can be classified as strongly and weakly circulated (perfused); corresponding to the body core and periphery. The core tissues receive ~63% of the resting cardiac output (CO) but constitute only ~19% of the total body mass. By contrast, the peripheral tissues receive ~37% of the basal CO and constitute ~81% of the body’s mass (Figure 1-14).

 Figure 1-14: The parenchymatous organs that comprise the visceral core of the body receive an aggregate of ~63% of resting the cardiac output while comprising only ~19% of the body mass. By contrast, the peripheral tissue mass which accounts for ~81% of body mass receive only ~19% of the cardiac output. External cooling profoundly chills peripheral tissues before significantly reducing core temperature. Values for organ and tissue masses were obtained from: IAEA. Compilation of anatomical, physiological and metabolic characteristics for a Reference Asian Man. Volumes 1 and 2. Report IAEA-TECDOC-1005, (Vienna, Austria: International Atomic Energy Agency) (1998), Boecker, BB. References values for Basic Human Anatomical and physiological characteristics for use in radiation protection. Radiation Protection Dosimetry.105(1–4): 571–574;2003, de la Grandmaison GL, Clairand I, Durigon M. Organ weight in 684 adult autopsies: new tables for a Caucasoid population. Forensic Sci Int. 2001 Jun 15;119(2):149-54 and Heymsfield SB, Gallagher D, Mayer L, Beetsch J, Pietrobelli A. Scaling of human body composition to stature: new insights into body mass index. Am J Clin Nutr 2007;86:82–91.Values for organ and tissue blood flows were obtained from: Williams, LR, Leggett, RW. Reference values for resting blood flow to organs of man. Clin Physiol Meas. 10:187-212;1989. Graphic by M.G. Darwin

 The Pathophysiology and Biophysical Limitations of External Cooling

The objective of MTH is to provide protection against ischemia-reperfusion injury to the brain, heart, kidneys and liver; the visceral organs that constitute the strongly circulated core of the body. The peripheral tissues (skin, skeletal muscle, connective tissues and bone) are at once much more resistant to ischemia and less well perfused.  External cooling rapidly chills the ischemia-resistant peripheral tissues cooling them profoundly, while failing to provide protection to the vulnerable parenchymatous organs in the body core. This is not only undesirable in terms of its inefficiency; it also poses a number of hazards and risks.[90] Hypothermia is therapeutic in ischemia-reperfusion because it down-regulates the immune-inflammatory response; a response that is vital for host defense, wound healing and hemostasis. Hypothermia, like any major medical intervention that perturbs fundamental physiological processes, carries with it serious risks, as well as benefits. In both animals and humans, hypothermia is markedly immunosuppressive [91],[92] and interferes with the both the biochemistry of the clotting cascade and the production of platelets and clotting proteins.[93],[94]

In humans perioperative minimal hypothermia (MinH) (36^oC) increases the rate of wound infections [95] and prolongs hospitalization. [96] These effects occur in part due to the regional thermoregulatory vasoconstriction MinH induces; which in turn leads to reduced oxygen delivery to injured  tissues, [97] inhibition of oxidative killing by neutrophils, [98] and reduced collagen deposition.[96]  MinH induces significant suppression of mitogenic responses to Concanavalin A (con A), phytohemagglutinin (PHA), and pokeweed mitogen (PWM) and these changes are known to persist for at least 48 h. The mitogens PHA and ConA activate T cells, whereas PWM stimulates both T and B cells, thus indicating that the suppressive effects of MinH involve a variety of lymphocyte subpopulations. Hypothermia of as little as 1^oC significantly inhibits production of  interlukins (IL-1б, IL-2, IL-6) and TNFα in post-surgical patients, [99] and this suppression of cytokine production persists for least 24 hours after even a brief post-operative hypothermic interval.[96] The inhibition of pro-inflammatory cytokine production by IL-1б and TNFα induce tissue factor which is critical to angiogenesis, collagen elaboration and fibroblast activation; all essential processes in wound repair and hemostasis.[100],[101],[102] Significantly, many of these of adverse effects of post-operative MinH can be prevented by maintaining normothermia in the perioperative period.[96]

In the Hypothermia After Cardiac Arrest Study Group, patients treated with MTH experienced twice the incidence of sepsis. This finding is consistent with other studies where MinH and MTH were found to double the rate of post operative wound infection.[96]  A recent meta-analysis of MTH for traumatic brain injury (TBI) found that the incidence of pneumonia was also doubled for patients undergoing MTH.[103]

External cooling causes greater perturbation in hemodynamics than does central cooling, [104] resulting in increased systemic vascular resistance and decreased cardiac index; a phenomenon observed in all of the patients in the Bernard, et al., study that employed MTH for post-arrest cerebral resuscitation.[46]  This is particularly undesirable in the setting of MI, CHF and cardiogenic shock. For these reasons, as is the case with any potent therapy, careful attention must be paid to the dose-response curve, and overshoot or excessive regional cooling must be minimized or avoided.[105]

In patients cooled with ice packs and non-feedback controlled cooling blankets, there have been persistent and extensive problems with overshoot. In one study, 63% of patients overshot to <32^oC, 28% to <31^oC, and 13% were inadvertently cooled <30^oC.[105] By contrast, patients in the European Resuscitation Council Hypothermia After Cardiac Arrest Registry who were cooled endovascularly had less overshoot (mean lowest temperature 32.9^oC, IQR; 32.6^oC to 33^oC) when compared to patients cooled by other methods (mean lowest temperature 32.4^oC, IQR: 31^oC to 32.9^oC).[106]

In part, the problem of overshoot in external cooling can be corrected by servo control of patient cooling. However, the fundamental problems of external cooling remain. Excessive cooling of poorly perfused peripheral tissues will inevitably result in ‘after-drop’ of core temperature as thermal equilibration occurs. This process is idiosyncratic and inherently difficult to model or predict. In large measure the speed and character of thermal equilibration between peripheral and central tissues will depend upon highly variable factors such as body morphology and composition, body surface area, cardiac output, regional blood flow distribution, and the administration of vasoactive medication such as ionotropes; with their attendant peripheral vasoconstriction.

When the tissues of non-hibernating (or unprepared hibernating mammals) are cooled to £20ºC a wide range of deleterious changes occur. The saturated fats which comprise cell membrane lipids undergo phase change, resulting in red and white blood cell rigidity; with accompanying inability to deform and pass through capillaries. Red cell aggregation also occurs and this, in association with reduced flow as a result of vasoconstriction, results in blood sludging and failure of the microcirculation.[107],[108]  Profound hypothermia, either local or systemic, results in hemoconcentration as a consequence of translocation of vascular water and electrolyte to the interstitial space.[109],[104],[110],[111],[112] This hemoconcentration further exacerbates regional ischemia in deeply chilled tissues.

Independent of injury from the freezing of water, moderate, profound or ultraprofound hypothermia is known to cause cellular damage which is referred to as ‘chilling injury.’ Chilling injury appears to be a multifactorial process in which alterations of membrane structure (reorganization of lamellar lipid sheets with lateral phase separation between regions of gel phase and regions of liquid crystal phase result in loss of membrane integrity), [113],[114] failure of ion pumping (with consequent disruption of cellular ionic homeostasis), [115],[116] depolymerisation of some elements of the cytoskeleton, [117],[118] generation of free radicals, [119],[120] and metabolic disruption due to selective inactivation of critical enzymes [121] all appear to play a role.

Cooling of tissues to 5ºC for as little as 1 hour has been shown to cause microvascular endothelial damage similar to that observed in ischemia-reperfusion injury; loss of endothelial cell tight junctions, infiltration of capillary and venule walls with leukocytes, and frank extravasation of red cells from injured vessels.[122] A possible reason for the similarity in the histological appearance of chilling injury with ischemia-reperfusion injury may be due to the fact that both types of injury appear to be caused, at least in part, by reactive oxygen species and by disruption of the cytoskeleton.

The molecular changes induced by moderate, profound and ultraprofound hypothermia may also directly compromise endothelial cell integrity. For example, chilling of several types of epithelial cells has been shown to result in disassembly (depolymerisation) of the intracellular microtubules resulting in compromise of the polarized membrane expression and function of some transport proteins in these cells.[123] These functions are slow to return to normal (@20 hr) and are associated with prolonged dysfunction of allografts that have undergone cold preservation storage.[124],[125]

Deep cooling of the peripheral tissues may also result in immunosuppresion in chilled limbs and skin, and possibly impaired hematopoiesis due to localized moderate hypothermia in bone marrow in the cranium, sternum, vertebrae, and to a lesser extent, in the pelvis. [In children, the long bones are the principal repository of hematocytoblasts, and this marrow would also be disproportionately chilled during external cooling.] In contrast to the anti-inflammatory effect of MTH, systemic hypothermia to £28ºC, either accidental or induced has been shown to increase levels of pro-inflammatory cytokines.

Core cooling is thus the gold standard for the induction of systemic hypothermia (mild, moderate, profound or ultraprofound) and external (peripheral) cooling should be used only where there is no other alternative for achieving truly rapid cooling, or maintaining it for the required 24-48 hours following induction.

 Consideration of Invasive Core Cooling Methods

Figure 1-15 (right): Typical (idealized) cooling and re-warming curve achievable with maximum extracorporeal (cardiopulmonary bypass) cooling.

Extracorporeal cooling via cardiopulmonary bypass (CPB) [126] or high flow veno-venous heat exchange [127],[128]  allows for cooling rates of 0.8ºC/min to 1.0ºC/min (Figure 1-15) However, it cannot be applied rapidly enough given existing logistic and regulatory constraints. CPB requires complex hardware and highly skilled personnel who must maintain their clinical reflexes by practicing perfusion on a regular (preferably daily or at least weekly) basis. Even in centers of excellence, with a highly skilled, rapid-response CPB team at the ready (including a well-practiced surgeon or cardiologist), the soonest CPB can be initiated after cardiac arrest is typically 15-20 minutes.[129],[130]

The use of emergency CPB applied in this time frame is largely confined to patients undergoing cardiac catheterization and/or revascularization (i.e., angioplasty or stent placement) who arrest in the cardiac catheterization lab. When such patients experience cardiac arrest that is refractory to treatment with drugs and defibrillation – they are usually otherwise healthy – they have normally functioning lungs, normal fluid balance and fluid distribution (are neither dehydrated nor edematous from fluid overload), and have failure of a only a single organ –  the heart. Even under such ‘ideal’ conditions, CPR is often inadequate to maintain brain viability during the brief interval between cardiac arrest and the start of CPB.

While veno-venous extracorporeal cooling is less technically demanding, it still requires skill-intensive vascular access under field conditions, and reported rates of cooling are modest; ~0.012ºC/min [131],[87]. Both of these techniques may require anticoagulation and certainly require coagulation monitoring; which again is a barrier to in-field application.

Administration of large volumes (40mL/kg) of chilled intravenous fluid has also been used to reduce core temperature in the field in patients undergoing CPR. However this method is extremely slow (~0.058ºC/min) [132],[133],[134], and is sharply constrained by the maximum volume of fluid that can be administered.

 Figure 1-16 (left): Cryopatient undergoing ECMO and blood washout in the home. Even under ideal circumstances cardiopulmonary bypass takes in excess of 30 minutes to initiate.

 The effectiveness of intravascular fluid administration in achieving durable core cooling is also a function of body composition. Obesity, which has become epidemic in the developed world, is now rapidly becoming a global problem with 1.1 billion adults worldwide classified as overweight,and 312 million of them as obese by the World Health Organization (WHO). [135] Vascular volume does not increase proportionally to increase in body weight; in fact, the vascular volume to body weight ratio falls toward an asymptotic value of approximately 45 ml/kg in the obese human.[136] Thus, the use of cold intravenous infusions will necessarily be even less effective in the obese than might be expected since the maximum volume of fluid that may be safely given does not increase linearly with body weight. Fat is also an extraordinarily good insulator, and obesity often unfavorably alters body surface to volume area. Both of these factors combine to dramatically reduce the speed and overall effectiveness of external cooling.

The limits of efficiency of core cooling that can be achieved by irrigating the peritoneal or pleural spaces with cold liquid are in the range of 0.05 to 0.3ºC/min [137],[138] and both of these techniques are invasive, require considerable technical skill, and carry with them risks of serious iatrogenic consequences.

As noted earlier, the ideal way to address these needs would be extracorporeal membrane oxygenation (ECMO) and cooling. However, the practicality of rapidly bringing this demanding technology to bear in the EMS setting is negligible. Surgery (or time-consuming percutaneous vascular access; at or beyond the optimum therapeutic window of |15 min post ROSC) is required to access the circulation and this cannot be accomplished under field conditions. Experience with emergent initiation of ECMO in patients presenting for experimental cryopreservation (Figure 1-16) are probably representative of the time required to achieve extracorporeal support under field conditions for patients with SCA. Such patients typically experience a delay of 60 to 80 minutes (and not infrequently longer) after the initiation of closed chest CPR before the application of CPB, even under optimum circumstances.[139],[140],[141]

Thus, what is needed for patients who have suffered prolonged (³5 min) cardiac arrest, or in whom ROSC cannot be effectively established, is the ability to rapidly induce hypothermia via core cooling simply and noninvasively.  In addition to moderating the injury cascade initiated by ischemia-reperfusion, mild (33-35^oC) or moderate (28-32^oC) hypothermia can indirectly act to improve gas exchange and hemodynamics by bringing the patient’s cerebral and systemic metabolic demands closer to those that can be delivered by CPR.

Exsanguinating Trauma Resulting in Cardiac Arrest

Closely related in pathophysiology to prolonged normothermic ischemia secondary to SCD is cardiac arrest secondary to exsanguinating trauma. It is estimated that ~20,000 US civilians a year die as result of hemorrhage from abdominal and thoracic injuries [36], or from poly-trauma. In developing nations this problem is even more severe as a disproportionate amount of trauma occurs in rural settings remote from tertiary care facilities; and with no helicopter or other airlift infrastructure available to shorten this interval.

Similarly, approximately 20% of the battlefield casualties who fail to reach tertiary care facilities die from intractable hemorrhage on the battlefield or during transport.[142],[143],[144],[145],[146] The US military, under the auspices of DARPA, has been funding a multimillion project to achieve ~30 minutes of battlefield ‘suspended animation’ using chilled, drug containing crystalloid solutions, to solve this major cause of war-related mortality [37] and preliminary studies have been positive.[147],[128],[147]

Cardiac arrest secondary to exsanguinating trauma offers a unique opportunity for intervention in the pathophysiological cascade of cardiac arrest. Because blood loss and deterioration of the patient to the agonal state occur over a time course of minutes, to an hour or longer, it is possible to begin induction of hypothermia before cardiac arrest occurs; potentially allowing for the opportunity to prevent many of the foreseeable irreversible pathological events which occur during ischemia as a result of cardiac arrest.

Beyond inhibition or moderation of the injury cascade attendant to ischemia-reperfusion injury, deep (15-27^oC) profound (5-14^oC) or ultra-profound (5-0^oC) hypothermia offers the prospect of acting  as a bridge to definitive therapy in cases of uncontrolled hemorrhage in both the military and civilian settings, as well as in cases of cardiac arrest that are refractory to prompt defibrillation and that require prolonged CPR until CPB can be initiated and coronary revascularization or application of long-term circulatory support (i.e., left ventricular assist device (LVAD) or total artificial heart (TAH)) can be implemented. This was the vision of one of the fathers of CPR and the discoverer of MTH, Dr. Peter Safar. Safer envisioned a time in the near tomorrow when patients who were not salvageable in the field would be placed into what he termed a ‘metabolic lock-box’ and preserved until definitive therapy could be applied in hospital. His formal name for this paradigm was ‘emergency preservation and resuscitation’ (EPR). [148]

As is the case with truly effective implementation of MTH in cardiac arrest or neurotrauma the barrier to EPR is safe, rapid and easily implemented core cooling. One possible solution to this problem is to use the lungs as a heat exchanger.

The Lungs as Heat Exchangers

Possessing the surface area of a tennis court, and being obligated, as they are, to accept and thin film all of the cardiac output over much of that surface area, the lungs commend themselves as extraordinarily suited to serve as heat exchangers, as well as mass exchangers. The principal obstacle to utilizing the lungs for this purpose is the extreme poverty of gas as a heat exchange medium. While the specific heat of room air (gas) is 1.4 J g⁻¹ K⁻¹ c_p at 23ºC [149] compared to 4.18 J g⁻¹ K⁻¹ c_p for water (liquid) at 25ºC [150], this is not the rate-limiting factor for heat transfer using air or oxygen (0.918 J g⁻¹ K⁻¹ c_p at 25ºC). While water possesses an extraordinarily high specific heat, this is atypical and is exceeded only by liquid ammonia (4.70 J g⁻¹ K⁻¹ c_p at 25ºC) and hydrogen (gas) (14.30 J g⁻¹ K⁻¹ c_p at 25ºC). The limitation of gases in facilitating heat exchange is their low density compared to liquids. Dry air has a density of ρ_SATP = 1.168 kg/m³ compared to 997.05 kg m^-3 (25.0°C) for liquid water. This is roughly a 1,000 to 1 difference in density and is the primary reason why air, oxygen and gases in general transfer heat at rates too low to be useful for rapidly reducing body temperature.[151],[152]

Yet another constraint on the use of gases for heat exchange in the lungs of most vertebrates is their tidal, bi-directional scheme of ventilation which results in admixture of inspired and expired gases and sharply constrains the maximum flow rate of gas per unit of time (minute volume) ‘through’ (i.e., in and out of) the lungs. By contrast, the lungs of birds employ a unidirectional, flow-through approach to ventilation (circulatory lungs) which is more efficient at gas exchange and allows for higher peak minute volumes.[153]

The final constraint on using gases is the necessity of keeping the temperature of the gas (or other heat exchange medium) at or above 0ºC in order to avoid freezing damage to the tissues or provoking broncho-constriction which interferes with both gas and heat exchange.[154]   This sharply limits the ΔT, which will also decay steadily as the temperature of the patient decreases towards 0ºC.

The only solution to the heat exchange limitation imposed by gases is to replace the tidal gaseous breath, or some fraction of it, with an appropriate liquid. This is a formidable challenge because the respiratory systems of air breathing animals have evolved complex and delicate systems to facilitate gas exchange; and these systems are incompatible with the presence of bulk liquid in the airways.

Figure 1-17: Anatomy of the acinus. Anatomically, alveoli are not independent structures but rather have common walls and are organized around and attached to the terminal bronchiole. This unit of structural organization is the acinus. The acinus is reinforced with both comparatively rigid (collagen) and elastic (elastin) fibers and the arrangement of the alveoli in a honeycomb-like manner within the acinar unit provides added structural stability and give the acinus emergent properties similar to those of foam. The structural features of the acinus, acting in concert with surfactant (which lowers alveolar surface tension providing additional architectural stability), is termed interdependence. Interdependent structural stabilization of the alveoli maintains alveolar volume and surface area fairly constant during ventilation under normal conditions. In lung injury, even the collapse of a single alveolus in an acinus causes shear stress – not only in the walls of the collapsed alveolus, but also in the walls of the adjacent alveoli (see Figure 4-11). The shear stress that develops under such conditions may exceed 140 cmH₂0, and the stabilizing interdependence of the acinus is lost, with the alveoli increasingly behaving as independent units that change volume dramatically with each tidal cycle.

The smallest gas exchange units of the mammalian respiratory system, the alveoli, are a mere ~125 µ in diameter and, absent molecular reinforcement in the form of surfactant, are inherently unstable and collapse. Surfactant has evolved to function as a semi-aqueous thin film lining the alveolus and other small acinar structures (such as the terminal bronchioles and the alveolar ducts) (Figure 1-17) and it is dissolved, damaged, or inactivated when exposed to water, and to other molecules typically found in body fluids such as proteins and lipid particles. Surfactant has both hydrophilic and lipophylic components which impose the seemingly impossible constraint that any liquid introduced into the lungs must be both hydro- and lipophobic, as well as virtually completely chemically inert.

 A Brief Précis of the History and Development of Liquid assisted Pulmonary Cooling (LAPC) for Induction of Hypothermia during CPR

So far, only one class of compounds, the perfluorocarbons (PFCs), meets these requirements. With the demonstration in 1966 by Clark and Gollin that mice could survive extended periods of ‘liquid breathing’ with the PFC FX-80 (FC-75; perflurotetrahydrofuran) the feasibility of using liquids as gas exchange media began to be explored. Despite the (in retrospect) painfully obvious implication that PFCs could be used in the lungs for heat exchange as well as gas exchange, the idea did not occur to the author until sometime between the winter of 1993 and the spring of 1994. This happened while rereading Guyton’s Textbook of Medical Physiology during the course of several transcontinental flights across the United States (US). The author had been aware of Leland Clark’s ‘fluorocarbon breathing mice’ since shortly after these experiments were carried out in 1966; they were the subject of extensive media coverage, including dramatic footage shown on science-oriented television programs of the time (such CBS’ ‘The 21^st Century’). Yet, it was not until re-reading Guyton’s chapter on respiratory physiology that the idea of using the lungs as a liquid-based heat exchanger became clear. Having spent literally 22-years obsessed with the problem of how to rapidly, non-invasively, and relatively simply, induce hypothermia in humans under field conditions; in hindsight it seems utterly obvious to use tidal liquid ventilation (TLV) employing a refrigerated PFC as a way of achieving rapid, homogenous, systemic cooling.

Figure 1-18: Mice can briefly tolerate breathing the PFC FC-77 but soon succumb to exhaustion from the work of breathing and hypothermia from the ambient temperature liquid.

An investigation of the literature disclosed the feasibility of TLV based on the work of Shaffer and his colleagues. [155]  Unfortunately, acquiring a PFC suitable for liquid assisted ventilation was very problematic. The only commercially available PFCs at that time were the 3M Fluorinet™ fluids and Rimar-101 (the latter manufactured by the Miteni Corporation in Milan, Italy). Neither company distributed their products through the conventional chemical houses[6]. The reason for this was that both 3M and Miteni specifically embargoed the sale of their PFCs for any research application involving biological systems – and in particular any application that involved liquid assisted ventilation (LAV) or O₂ carrying blood substitutes (oxygen therapeutics).

By January of 1995, the author and Respiratory Therapist Michael Fletcher had constructed the first TLV ventilator for research into inducing MTH during CPR, with the idea very much in mind that this technology had tremendous potential for broad medical application (post resuscitation syndrome, stroke, subarachnoid hemorrhage (SAH) and possibly traumatic brain and ischemic (and traumatic) spinal cord injury (TBI and SCI).

Figure 1-19: The first TLV apparatus for cardiopulmonary cerebral resuscitation-related research (left) used blood as the gas exchange medium. The device used a bubble oxygenator and heat exchanger and relied on gravity to provide ‘exhalation’ drainage from the lungs. The inspiratory roller pump and the timing and pressure controller functioned well, but blood proved to be too injurious to the lungs t serve as a gas exchange medium. These experiments made the need for an expiratory pump obvious. Photo: circa February, 1995.

Because of the unavailability of PFCs, and because of the prior liquid breathing research by Johannes Klystra  [156],[157],[158] which demonstrated that mammals could survive after breathing a balanced salt solution under hyperbaric conditions (where gas solubility in water is dramatically improved), we decided to use bovine blood as the gas exchange medium in a non-survival canine model of TLV. Blood, unlike saline, has the proven capacity to deliver and exchange an adequate amount of O₂ and CO₂under normobaric conditions. A review the literature (then and now) indicated that this had not been attempted, or apparently even proposed. These two initial experiments using bovine blood in an acute canine model of TLV demonstrated that blood was not suitable for use, and that TLV was a demanding technique that required an exhalation pump in addition to an inspiration pump. The prototype device taught us a great deal, and in fact, the control unit from that first TLV ventilator was used until 2000 to conduct much of the early 21^st Century Medicine, Inc., and Critical Care Research, Inc., work on liquid assisted pulmonary cooling (LAPC).

 Figure 1-16: Prototype tidal liquid ventilator using the ‘sweep flow’ (continuous liquid delivery/removal to the carina) technique circa 1995, and the active compression-decompression, high impulse Thumper™ mechanical CPR device (Figure 1-17) used to facilitate tidal movement of PFC between large and small airways (below). [Photo (left) courtesy of Charles Platt, 1995.]

By March of 1995 several PFCs had become available in sufficient quantity for use in a canine model of TLV. A continuous flow approach was being used to deliver chilled PFC to the carina, with tidal movement of liquid in and out of the small airways being achieved by active compression and decompression of the chest wall using the suction cup of an Ambu CardioPump™ on a specially constructed, pneumatically driven Thumper™ CPR device. This technique was christened ‘sweep-flow liquid ventilation.’ [159]

Figure 1-17: Custom fabricated Michigan Instruments, Inc., Thumper™ active compression decompression cardiopulmonary resuscitator

Dynamically sensing and controlling the amount of PFC in the animals lungs proved problematic with this system, and despite a very high predicted minute alveolar ventilation rate (with 80 to 100 liquid ‘breaths’ per minute from chest compression-decompressions), and minimal anatomical dead-space (the tip of the liquid delivery catheter was at the level of the carina), hypercarbia developed after the first 10 to 15 minutes of ventilation.

While survival of animals following TLV using PFCs was 100%, with a low incidence of adverse effects (providing TLV was discontinued prior to lethal levels of hypoxemia), the logistics of applying TLV to a field-setting under emergency conditions was deemed insurmountable. This was particularly true in the mid-1990s due to limitations on both portable computing power and the lack of adequately miniaturized and lightweight sensing and servo-control technology.

During the fall of 1996, the author learned of the development and initial clinical application of a new technique of liquid assisted ventilation (LAV) called partial liquid ventilation (PLV) which was being used to treat respiratory distress syndrome in (primarily) premature infants. This work was rapidly extended to include adults, and in the Spring of 1996 the author attended a three-day conference and workshop: the ‘Basics of Liquid Ventilation Management of Severe ARDS, ECMO, Liquid Breathing and PCIRV Conference’ in Ann Arbor, MI.[7]  In contrast to TLV, PLV was effected by simply adding PFC to the patient’s respiratory system until it was filled to functional residual capacity (FRC, ~30 ml/kg) while continuing conventional mechanical ventilation with gas; PLV was the essence of simplicity PFC was added to the endotracheal tube with a syringe until the desired volume was given or a meniscus of PFC appeared at the desired level in the ET tube. What is more, as the author had already observed while rounding at the University of Michigan Medical Center’s ICU, due to their unique physical properties, PFCs had the ability to rapidly reverse the hypoxemia and hypercarbia that accompany respiratory distress and its attendant pulmonary edema. Thus, PLV offered the added promise of restoring adequate gas exchange in fulminating pulmonary edema, as well as serving as a safe and effective heat exchange medium.

Early in 1997, due to the technical and logistic difficulties imposed by TLV, the author began to investigate the use of multiple lavages of the lungs in dogs using PFC chilled to ~4ºC. In this technique the animals’ lungs were filled with ~4 ºC PFC to vital capacity (VC, 60-70 ml/kg) following which the PFC was suctioned out and this process was repeated with additional loads of ~4ºC PFC until the desired core temperature was reached.[159]  While technically straightforward and very simple in application, this approach required large amounts of chilled PFC and had the obvious potential to interfere with gas exchange under conditions of hyper-metabolism, oxygen debt, or hypercarbia. Additionally, the time required for loading and suctioning out 60-70 ml/kg of PFC did not appear to be consonant with maximizing heat exchange, given that much of the PFC load would necessarily remain in the large and intermediate sized airways, where heat exchange is very slow compared to in the alveoli.

At this point it became apparent that a practical, field-applicable method of inducing hypothermia by PFC lung lavage would involve a hybrid approach, wherein both liquid and gas tidal ventilation would have to proceed simultaneously, or on a rapidly alternating basis. This work was largely set aside at that time (1997) due to a research commitment to develop a multi-modal method of dealing with ischemia-reperfusion injury attendant to prolonged cardiac arrest. This model involved the use of multiple drugs, as well as the rapid (~15 minutes) induction of mild post-insult hypothermia using the Safar, et al., [160] model canine model of cardiac arrest and resuscitation, which employed CPB for the induction of hypothermia.

 Figure 1-18 (above): The alveolar capillary unit is the fundamental repeating structural unit of the lung. Above is a schematic rendering of an alveolus as it would appear magnified ~3,000 times. The polyhedral alveolar air spaces (AAS) are separated from each other by the alveolar septa which are ~15µ thick. The Type I epithelial cells overlay the capillaries (C) and constitute the air side of the airway-blood barrier (ABB). The capillaries are the yellow spaces with the red blood cells (RBCs) inside them. The Type II epithelial cells are heavily vesiculated and secrete surfactant which maintains the structural integrity of the alveoli and prevents them from collapsing during exhalation. The total thickness of the ABB is only ~0.5µ which means that heat exchange between blood and PFC at the alveolar wall is extremely rapid – the rate limiting factor on heat exchange is, in fact, the rate at which PFC can be moved into and out of the alveolus.

With the success of this post-resuscitation hypothermia and multi-drug research cerebral resuscitation model in 1998, it became important to demonstrate that very rapid, in-field induction of hypothermia was technically and logistically possible. A decision was made to return to the LAPC research in order to validate the feasibility of this approach for immediate, simple, post-arrest induction of MTH (~34 ºC).The first results of the work were presented in1999 at the 28th Educational and Scientific Symposium of the Society of Critical care Medicine in San Francisco, CA. [161]  At the time, it was anticipated that completion of the research would take approximately 6 months.

The team that began this work consisted of the author, Steven B. Harris, M.D., Sandra Russell, BS, Michael Fletcher, R.T., Joan O’Farrell and Brian Wowk, Ph.D.  This project proved more formidable than initially anticipated and ultimately consumed over 2 years of full-time effort in order to develop the core elements of a workable technique, and achieve a basic understanding of the physiology and pathophysiology of fractional tidal liquid assisted pulmonary cooling (LAPC); the technique that was ultimately developed. This team worked together seamlessly to establish the effective volumes of PFC to be loaded and unloaded (i.e., fractional liquid tidal volumes), optimum dwell times for PFC, tolerable peak and mean airway pressures for ventilation during liquid cycling, workable gas ventilation strategies, and the basic mechanics of both heat and gas exchange using this technique.

 What LAPC Can Potentially Deliver

Properly implemented, LAPC would offer the following advantages:

• Ease of application, requiring far less highly skilled personnel than are needed for CPB,
• Technically less demanding, requiring fewer total personnel than CPB,
• Effective at achieving a rate of heat exchange in the brain comparable to or better than that achievable with CPB,
• Effective at achieving good gas exchange even with patients with severe lung disease/injury (pulmonary edema, acute lung injury (ALI) and acute respiratory distress syndromes (ARDS),
• Relatively inexpensive to use,
• Improves oxygenation and ventilation,
• Reduces interfacial surface tensions (i.e., between liquid and gases and liquids, gases and alveolar membranes),
• Reduces ventilation pressures,
• Recruits alveoli,
• Redistributes pulmonary blood flow and mitigates V/Q mismatch,
• Mitigates barotraumas and volutrauma,
• Lavages airway and alveolar debris,
• Reduces inflammation and lung injury.
• Allow CPR  to serve as a viable bridge to CPB and eventual definitive therapy to restore spontaneous circulation or supplement or replace it as necessary (i.e., LVAD or TAH)

The physiological considerations underlying LAPC are these:

1)    All of the blood that flows out from the heart to the body flows through the lungs first, where it is oxygenated and carbon dioxide (CO₂) is removed. In order to facilitate gas exchange, the architecture of the lungs has evolved to provide an enormous surface area combined with an ultra-microscopic barrier between blood and air. The alveoli, the gas exchange units of the lungs, are polyhedral compartments ~250 µ in diameter which branch out from the terminal bronchioles (which are about twice the diameter of an alveolus; 250µ) (Figure 1-18, above). Each adult human lung contains ~300 x 10⁶ alveoli with an aggregate surface area of 140 m² [162] The alveolar epithelium is comprised of two types of cells; Type I epithelial cells that constitute the alveolar side of the air-blood barrier (ABB) and Type II epithelial cells which are intercalated in the epithelial lining and secrete surfactant. The Type I epithelial cells are so thinned out where they cover the capillary endothelial cells (which are even thinner) that the composite thickness of both cell layers and the basal lamina is ~0.5µ. These unique anatomical features of the lungs, combined with the physiological fact that they process all of the cardiac output, makes them ideal as a heat exchanger.

2) As previously noted, air, oxygen and other gases make extremely poor heat exchange media since they are roughly a thousand times less dense than water, they will remove heat at only roughly one thousandth the rate.

3) Mammals can survive briefly while spontaneously breathing perfluorochemicals (PFCs) and the usual cause of death when small mammals, such as mice, are left immersed in PFCs is hypothermia due to loss of heat from the large surface area of the lungs to the room temperature PFC in which they are immersed.[163] The severely injured lungs of human patients with pulmonary edema and/or acute lung injury (ALI) or acute respiratory distress syndrome (ARDS) not only tolerate extended (days to weeks) partial filling with PFCs of the appropriate vapor pressure, density, and viscosity, their ability to carry out gas exchange is improved, and there is evidence that long-term damage from ARDS may be reduced.[164],[165],[166],[167]

This technique of non-invasive in-field induction of hypothermia using the lungs is called liquid assisted pulmonary heat exchange (LAPHE), or liquid assisted pulmonary cooling (LAPC) when used for inducing hypothermia.

FOOTNOTES