Hypothermia and Rewarming by Peritoneal Dialysis and Temperature-Controlled Inhalate Donald R. Sperling, M.D., and Anthony V. Beran, B.S. ABSTRACT In 12 rabbitshypothermia and rewarming were induced with temperature-controlled circulating peritoneal dialysis in combination with temperature-controlled hypoxic and hypercapnic gas mixtures. The average cooling time necessary for the esophageal temperature to decrease from 37.7"f 0.7 to 20.6" 5 1.O"Cwas 81 5 34 minutes with a range of 41 to 150 minutes. The average warming time for esophageal temperature to increase from 20.6" 5 1.0"C to 35.2"f 1.8"Cwas 90 f 35 minutes. Time of cooling was related to the proportions of inspired carbon dioxide and oxygen. In contrast to surface and bypass methods, esophageal and muscular temperatures agreed very closely, suggesting an absence of regional temperature gradients.
Deep hypothermia with complete circulatory arrest induced by surface cooling followed by cardiopulmonary bypass has recently been clinically accepted as a procedure to be applied during the surgical correction of congenital heart lesions in infants [ 2 , 111. Surface cooling cannot be easily controlled and requires a prolonged period to decrease body temperature. Hypothermia induced by cardiopulmonary bypass may be too rapid to allow complete equilibration between internal organs, muscle, and skin temperatures at the time of complete circulatory arrest. Temperature gradients between the skin and various organs, in addition to complications produced by cardiopulmonary bypass itself, make a search for a better method justifiable. This study was undertaken to evaluate the use of temperature-controlled circulating peritoneal dialysis in combination with breathing of cold or warm hypoxic and hypercapnic gas mixtures From the Department of Pediatrics, Division of Cardiology, University of California, Irvine, California College of Medicine, Irvine, CA. Accepted for publication May 9, 1975. Address reprint requests to Dr. Sperling, Department of Pediatrics (Division of Cardiology),University of California, Irvine, California College of Medicine, Irvine, CA 92717.
for the induction of total-body hypothermia and rewarming.
Material and Methods Experimental Procedure Twelve male New Zealand rabbits weighing 2.5 to 3.5 kg were used in this study. Sodium pentobarbital (25 mg per kilogram of body weight) was given intravenously for anesthesia. The trachea was surgically exposed and a Y-type tracheostomy tube was inserted and secured. A pneumotachygraph was placed in series between this tube and a respirator. Respiratory oxygen and carbon dioxide were measured at this point. An arterial catheter was passed from the femoral artery into the descending aorta. A venous catheter was placed in the right atrium retrograde from the femoral vein. The peritoneum was entered in two areas and dialysate heads were placed as described below. The incisions were sutured and sealed with collodion to prevent leakage. Temperature probes were placed in the esophagus, in the peritoneum, and intermuscularly in the axilla. To minimize interference with respiration by the dialysate, the animal was positioned on a surgical table inclined 20 degrees and was connected to a respirator (Harvard 607). Tidal volumes were adjusted to produce hypenrentilation. Dextran (10 mglkg) was administered intravenously [lo]. Control data for all physiological variables described below were obtained at this point. Two hundred milliliters of cold dialysate (+4"C) was then introduced into the peritoneal cavity, and recirculating dialysis was initiated. The oxygen, nitrogen, and carbon dioxide concentrations of the inhalate were adjusted to produce a desired gas mixture, similar to the method described by Miller [ 7 ] . These concentrations were continuously monitored at the tracheostomy tube. Dialysis was continued until an esophageal temperature of approximately 20°C was reached and maintained for 15 minutes. At the begin-
152 The Annals of Thoracic Surgery Vol 21 No 2 February 1976
ning of rewarming a second dose of Dextran (10 mglkg) was administered. Adrenaline (0.02 mglkg) was given if the systolic blood pressure decreased below 30 mm Hg. Blood gas samples were obtained every 15 minutes during cooling and rewarming while other physiological variables were continuously monitored. The animals were rewarmed to approximately 35°C and then killed.
During the cooling period, the temperature of the dialysate was maintained at 4°C by setting the adjustable thermometer of the constanttemperature circulator to 4°C while the immersion cooler operated continuously. When the temperature of the dialysate decreased below that level, the heating coils of the constanttemperature circulator were activated to bring the temperature back to 4°C. During the rewarming period the dialysate temperature was servocontrolled at 39°C by the constanttemperature circulator alone. This method provides for very precise control of dialysate tempera ture.
Peritoneal Dialysis Temperature Control System The system assembled for induction of hypothermia and rewarming by peritoneal dialysis is shown in Figure 1. Polyionic dialysate (Travenol, Dianol 1.5%) was adjusted to the exRespiratory Gas Temperature Control System tracellular electrolyte composition of the rabbit by adding potassium chloride, calcium gluco- During cooling, respiratory gas mixtures from nate, and sodium bicarbonate [l]. The afferent the respirator were forced through an ice-filled and efferent limbs of the system were made of Plexiglas tube (1m long, 3.8 cm diameter) conpolyethylene tubing (3.5 mm bore), heat insu- nected to the tracheostomy tube. During related with refrigeration tape and connected to warming, the Plexiglas tube was removed and specially designed intraabdominal heads that humidified gas mixtures from the respirator facilitated dialysate flow. The afferent dialysate were passed through a warming coil placed head was placed superiorly in the peritoneal cav- around the inspiratory tube of the respirator and ity and the efferent head was positioned deep in connected to the tracheostomy tube. During the pelvis. Afferent and efferent dialysate was cooling the temperature of the inhaled gas mixcirculated by a double-head, variable-speed tures was kept at 8°C. During rewarming it was pump (Cole-Parmer 70200), and flow was con- maintained at 39°C. trolled by adjustable valves. Central venous pressure and abdominal distention were used as Physiological Measurements guidelines for the amount of flow. Dialysate flow Arterial (AP) and venous (VP) blood pressures was usually maintained at 100 ml per minute, were continuously monitored by means of presand temperature was controlled by interaction of sure transducers (Statham P23Db). Heart rate the constant-temperature circulator (Haake FJ) (HR) was determined from the arterial pressure tracings. The percentage of inspiratory carbon and the immersion cooler (PSC-KR50). dioxide ( F I ~ Q )was monitored by a carbon dioxide analyzer (NV Godart, type 146) and the Fig 1 . The system used for cooling and rewarming percentage of inspiratory oxygen (FI-) by an during peritoneal dialysis. oxygen analyzer (Westinghouse 211). Arterial and venous Po2, Pco,, and pH were measured by a blood gas and pH analyzer (Instrumentation Laboratories, IL 113).Rigid calibration procedures were implemented [31. Corrections were made in Po,, Pco2,and pH to compensate for the difference between body temperature and sample temperature at the time of determination. Esophageal (ET), peritoneal (PT), and intermuscular axillary (IMAT) temperatures were measured by conventional methods. Oxygen saturation (So,) was determined by a microoximeter
153 Sperling and Beran: Hypothermia and Rewarming
(American Optical). All dynamic variables were recorded on an eight-channel chart recorder (Sanbom 358-100). Individual physiological variables were plotted as a function of time and cross-correlated.
Results The average cooling time necessary for the ET to decrease from 37.7" f 0.7"C to 20.6" f 1.O"C was 81 k 34 minutes. The average warming time required for i t to increase from 20.6" k 1.O"C to 35.2" +_ 13°C was 90 +_ 35 minutes. Esophageal and intermuscular axillary temperature changes paralleled each other closely during cooling and rewarming, the maximal difference during cooling being 1.3"C. Toward the end of warming ET was maintained at a higher level, the maximal difference being 1.6"C (Fig 2). Maximum cooling occurred within the first 45 minutes, during which time the ET decreased from 37.7" to 23.2"C, or 0.32"C per minute. Maximal warming occurred in the first 75 minutes, during which ET increased from 20.6" to 34.2"C, or 0.18"C per minute. All 12 animals survived the experiment without the use of cardiac massage or extracorporeal circulatory support. Fig2. Changes in peritoneal (PT), esophageal (ET), and intermuscular axillary (IMAT) temperatures as a function of time during cooling and rewarming.
TIM N I MW I
The means and standard deviations for AP (systolic and diastolic), VP, HR, PT, IMAT, ET, and arterial and venous P q , P c q , pH, and S q are presented in Table 1. Systolic and diastolic blood pressure decreased steadily from 114/79to 27/14 mm Hg at the end of cooling. During the rewarming period both systolic and diastolic pressures steadily increased and stabilized at 87/56 mm Hg 105 minutes following the beginning of rewarming. Venous pressure rose during the initial 45 minutes of cooling. This increase was then maintained above the control level until rewarming was begun. During rewarming it fluctuated but remained higher than the control value. Heart rate decreased from 227 to 42 beats per minute during cooling, and at the end of rewarming it had increased to 205 beats per minute. As a result of manipulation in F I Oand ~ FICO,, the arterial Po, decreased from 67 to 29 mm Hg during cooling and returned to nearly 70 mm Hg 105 minutes following initiation of rewarming. Arterial Pco2 and pH during cooling and rewarming were maintained at hypocapnic and alkalotic levels. Saturation was kept above 90% during cooling. During rewarming, after an initial decrease, oxygen saturation stabilized at the 89% level 60 minutes following the beginning of rewarming. Venous Po, fell progressively with the decrease .in temperature and returned to the control level following rewarming. Venous Pcop and pH were maintained during cooling and the initial rewarming phase. However, toward the end of rewarming the venous Pco2 increased and pH decreased. Venous oxygen saturation declined during cooling and returned to the control level during rewarming. Although the animals were breathing hypoxic and hypercapnic gas mixtures, arterial saturation stayed within the normal range and arterial PCO, remained decreased. The relationship between Fro,, arterial saturation, and Po, is shown in Figure 3A and the relationship between F I ~ ~ , and arterial PCO, in Figure 3B. The effect of hypercapnic and hypoxic gas mixtures upon the length of time necessary to reduce ET from 37.7" to 20.6"C is shown in Figure 4. With decreasing FIQ the time of cooling is decreased. The extreme values plotted are at 150 minutes while breathing room air and at 41 minutes while
154 The Annals of Thoracic Surgery Vol 21 No 2 February 1976
Table 1 . Changes in Physiological Variables during Cooling and Rewarming Variable
AP syst (mmHg) fiP diast (mm Hg) \’P (mm Hg) HR (beatsimin) PT (“C)
IMAT (“C) ET (“C) Artenal
Venous PO* (rnmHg) PCO,
(mmHg) PH Sat (%)
Control: 0 time 15
Cooling (min) 60 75
78 f20 55 f 19 4.7 f 2.0 105 f 46 12.9 f 3.2 28.1 % 2.5 27.3 2.5
61 f 29 45 t 22 5.8 f 2.7 81 f 27 10.9 f 0.7 24.3 f 2.9 23.2 1.7
58 f 27 40 f 23 4.7 f 2.4 67 f 46 10.3 f 1.4 23.5 f 2.2 22.7 f 1.4
63 f35 46 f 29 4.7 t 1.3 57 f 18 11.4 f 3.1 22.6 f 1.0 21.5 f 0.7
41 f 28 30 f 23 5.4 f 4.4 63 f 45 9.7 f 2.0 21.9 f 1.9 20.6 f 1.0
43 f 12 20 f4 4.0 t 2.3 55 f 60 8.2 f 1.5 22.2 f 0.3 21.5 f 1.0
31 f 20 24
30 f 12 26 f 9 7.540 f 0.16 92.7 f 6.4
30 f 20 23 f 8 7.505 f 0.06 91.0 f 8.8
40 t4 22 f 4 7.455 f 1.3 90.5 f 0.7
12 f 7 29
9 f 8 28 +4 7.477 f 0.04 38.7 f 30.8
... ... ... .,. .., ,.. . ,. .. .
114 f 22 79 f 15 3.5 f 1.6 227 f 39 37.6 f 1.0 36.8 f 0.8 37.7 t 0.7
87 f21 65 f 22 5.0 f 2.0 142 f 35 16.3 f 2.8 30.6 f 2.2 29.6 f 2.4
67 f 14 27 2 6 7.565 f 0.08 93.1 f 3.5
60 f 19 21 f 6 7.616 f 0.08 97.2 f 3.1
48 f 24 24 f10 7.561 f 0.16 94.0 k 5.3
7.573 f 0.17 90.4 f 7.2
33 f 16 28 f 9 7.471 f 0.11 91.3 f 8.5
26 +5 37 +5 7.498 f 0.09 47.1 f 13.5
15 f 4 27 f 6 7.592 f 0.04 43.7 f 13.9
17 f 5 30 f 6 7.491 f 0.10 50.5 f 11.1
12 f 5 26 f 4 7.523 f 0.14 46.7 f 16.4
13 f 7 29 f 6 7.473 f 0.11 51.6 t 26.4
7.517 f 0.16 51.3 f 19.1
Warming (min) 45 60
27 f 12 14 f1 3.8 f 2.1 42 f 54 7.8 f 0.2 20.8 f 0.3 20.6 t 1.0
41 f 24 25 t 17 4.9 t 2.5 72 k 39 34.5 f 5.1 22.9 f 2.7 22.7 f 2.5
54 f 31 31 f 22 6.1 f 2.3 110 f 57 36.6 f 4.1 27.2 f 3.1 27.3 f 3.4
70 f29 45 f 24 4.5 k 1.4 154 f 43 35.3 f 4.5 29.7 f 4.1 29.8 f 4.2
29 f 11 29 fl 7.370 0.04 90.0 f 1.4
22 f 17 25 +lo 7.501 f 0.18 76.7 f 14.8
32 f 17 31 f13 7.472 f 0.11 82.2 f 13.0
.. ... ._. ... .. .
10 f 6 35 +8 7.425 0.15 34.0 f 21.0
14 f 8 40 +I3 7.395 f 0.13 35.2 f 25.2
46 f 28
f 2.4 167 f 41 39.0 f 2.9 31.7 f 3.8 31.9 3.4
59 f 21 4.9 t 1.7 192 f 26 38.7 f 2.9 33.4 f 2.8 34.2 t 2.6
87 f 16 56 t 13 4.1 f 1.6 205 f 41 38.2 f 1.6 33.6 f 3.1 35.2 f 1.8
44 f 23 31 f15 7.523 f 0.14 85.1 f 14.4
54 f 21 26 f 9 7.507 f 0.19 89.2 f 6.8
63 f 22 31 f 7 7.390 f 0.22 86.2 f 15.9
69 f 23 34 f17 7.354 f 0.13 89 2 f 9.8
20 f 9 44 f18 7.430 t 0.09 42.8 f 6.6
22 f 7 48 220 7.382 f 0.20 45.1 f 22.2
28 +9 53 f20 7.352 f 0.18 49.6 f 20.4
29 f 7 50 f17 7.296 t 0.14 41.5 f 15.2
A.P = arterial uressure: VP = venous uressure: HR = heart rate; I T= peritoneal temperature; IMAT = intermuscular axillary temperature; ET = esophageal temperature.
breathing 8.2% oxygen. With increasing F I C the time of cooling is decreased from 150 minutes while breathing 0% carbon dioxide to 41 minutes while breathing approximately 3yo carbon dioxide. The effect could be interpreted only as a trend, since each point represents 1animal. Mean percentage changes in AP, HR, and arterial Po, and So, as a function of ET are shown in Figure 5: AP, HR, and Po, fell with decreasing temperature; SO, remained relatively constant. Comment Heat is normally lost or gained by the body through three physical processes: radiation, conduction, and vaporization of water [61. The relative importance of each, as well as the rate of change, varies with the temperature gradient and environmental conditions. In order to maintain temperature and water balance during respiration, the upper respiratory tract must condition air during inspiration, and it must conserve heat and water during expiration. This is accomplished by the process of heat loss or gain by turbulent convection and
~ water evaporation. For example, in an environment of 25°C and 40% relative humidity, during inspiration heat is transferred to the inspired air by turbulent convection and water is transferred by evaporation from the mucosa until equilibrium is established. Approximately 34 mg of water is added to 1 liter of air before it reaches the alveolar environment, decreasing the temperature of the mucosa to 31°C. During expiration, turbulent convection facilitates heat transfer from the warmer alveolar air to the cooler mucosa, simultaneously condensing and transferring the water to the mucosa. An adult male at rest in this environment loses 250 ml of water and 350 Cal of heat in his expired air per day . If the temperature and humidity of the climate are decreased, the total loss of heat and water in expired air is increased. In addition, a patient with a tracheostomy may eliminate as much as 4,200 ml of water per day [171. Therefore, bypass of the upper respiratory tract would increase heat and water loss. In our experiments temperature and humidity of the inspired gases were decreased and the
155 Sperling and Beran: Hypothermia and Rewarming
Fig 4 . Time of cooling as a function of ( A ) mean Fie, and ( B ) mean F I ~ ~(Data , . were not available for I experiment.)
Fig 3 . ( A ) Changes in inspired oxygen ( F I ~ , ) , arterial P o e (Pao,), and arterial oxygen saturation (Sao,) as a function of time during cooling. ( B ) Changes in inspired carbon dioxide ( F I ~ ~and , ) arterial Pco, (Paco,) as a function of time during cooling.
upper respiratory tract was bypassed, leading to increased expiratory heat loss by both turbulent convection and evaporation. Peritoneal dialysis is an effective means for the removal and delivery of body fluids and solutes . It has also been used for treatment of hydrops fetalis  and neonatal acidosis . Our work with rabbits demonstrated that in addition to fluid and electrolyte maintenance and
detoxification, arterial P q can be increased, P c q decreased, and pH and body temperature maintained by peritoneal dialysis [31. Under normal conditions the blood flow of the mesentery is estimated to be 10 to 15% of the cardiac output . Due to the mesenteric microcirculation and direct contact with peritoneal contents, this large amount of blood can equilibrate with dialysate temperatures almost instantaneously. In addition, the temperature of blood in the portal and renal circulation, in the descending aorta, and in the inferior vena cava will also be affected by conduction. It has previously been reported that optimal liver tissue perfusion during cooling is accomplished when a reduction in ambient oxygen and an increase in ambient carbon dioxide are initiated simultaneously with hypothermia . In addition, optimal maternal and fetal brain
156 The Annals of Thoracic Surgery Vol 21 No 2 February 1976
producing hypothermia. Indeed, our results support this assumption. Comparative data obtained by other investigators using different methods of producing hypothermia and rewarming are shown in Table 2. Compared to these data, our method produces an overall rate of cooling of 0.32"Clmin between 37.7" and 23.2"C. However, there is a more rapid initial cooling phase of 0.54"Clmin between 37.7" and 29.6"C. This is follawed by a slower rate of 0.21"Clmin between 29.6" and 23.2"C and of 0.03"Clmin between 23.2" and 20.6"C. Using our Fig5. Percentage changes in arterial systolic pressure method, rewarming is produced rapidly at a rate (AP), heart rate (HR), arterial P o , (Pao,), and of 0.18"Clmin from 20.6" to 34.2"C. Previously arterial saturation (Sao2) as a function of esophageal reported results for a similarrange of cooling iempera ture. were O.2l"Clmin using peritoneal dialysis  oxygenation is accomplished when the mother and 0.09"Clmin using surface cooling and hyperbreathes 10% oxygen and 10% carbon dioxide capnic gas mixtures [ill. In contrast to surface [91. This increase in tissue perfusion produced and bypass cooling methods, there was close by hypoxia and hypercapnia facilitates and agreement between esophageal and muscular equalizes heat dissipation from different body temperatures during cooling and rewarming, suggesting an absence of regional temperature regions. Therefore it seemed appropriate to assume differences; however, this should be further verthat the combination of these three individual ified by measuring temperatures of different ormodalities would provide an effective means of gans and of many muscle points.
Table 2 . Comparative Data Obtained by Other lnvestigators Using Different Methods for Induction of Hypothermia and Rewarming Cooling
Temp Drop Per Min ("C)
Surface Bypass Peritoneal dialysis
37-26 26-22 35-20
150 3-5 72
0.07 1.0 0.21
Barratt-Boyes et a1 [21 Trinkle et a1 [161 Sealy et a1 [151 Rhode et a1 I141 Method 1 Rhode et a1  Method 2 Rhode et a1 [141 Method 3 Mori et a1 [111 Patton and Doolittle [131
Temp Rise Per Min ("C)
Surface Bypass Peritoneal dialysis and surface Surface Surface
Peritoneal 37.2-32 dialysis Surface 37-24 Bypass 24-18 (hypercapnia) Surface 36-25
Peritoneal 32.4-37.8 dialysis Bypass & surface 18-35 (hypercapnia) 35-37 Surface Peritoneal dialysis 25-36 Surface 25-36
32-36 20-32 20-35
60 20 84
0.07 0.6 0.18
157 Sperling and Beran: Hypothermia and Rewarming
References 1. Altman PL, Gibson JF, Wang CC: Handbook of Respiration. Philadelphia, Saunders, 1969, p 93 2. Barratt-Boyes BG, Simpson M, Neutze JM: Intracardiac surgery in neonates and infants using deep hypothermia with surface cooling and limited cardiopulmonary bypass. Circulation 43 (Suppl 1):25, 1971 3. Beran AV, Taylor WF: Peritoneal dialysis for the support of respiratory insufficiency in rabbits. Clin Sci 43:695, 1972 4. Boda D, Muranyi L, Altorjay I, et al: Peritoneal dialysis in the treatment of hyaline membrane disease of newborn premature infants. Acta Paediat Scand 60:90, 1971 5. Bums GP, Schenk WG Jr: Effect of digestion and exercise on intestinal blood flow and cardiac output: an experimental study in the conscious dog. Arch Surg 98790, 1969 6. Fulton JR: A Textbook of Physiology. Philadelphia, Saunders, 1949, p 1079 7. Miller JA Jr, Kessler M: Tissue Po, levels in the liver of warm and cold rats artificially respired with different mixtures of 0, and COP: oxygen transport to tissue. Adv Exp Med Biol 37A:361, 1973 8. Miller RB, Tassistro CT: Current concepts: peritoneal dialysis. N Engl J Med 281:945, 1969
9. Misrahy GA, Beran AV, Hardwick DF: Fetal and neonatal brain oxygen. Am J Physiol203:160,1962 10. Mohri H, Hessel EA, Nelson RJ: Use of Rheomacrodex and hyperventilation in prolonged circulatory arrest under deep hypothermia induced by surface cooling. Am J Surg 112:241, 1966 11. Mori A, Muraoka R, Yokota Y, et al: Deep hypothermia combined with cardiopulmonary bypass for cardiac surgery in neonates and infants. J Thorac Cardiovasc Surg 64:422, 1972 12. Nathan E: Severe hydrops foetalis treated with peritoneal dialysis and positive-pressure ventilation. Lancet 1:1393, 1968 13. Patton JF, Doolittle WH: Core rewarming by dialysis following induced hypothermia in the dog. J Appl Physiol33:800, 1972 14. Rhode CM, Jennings WD Jr, Pittman W: Experimental total body hypothermia: a comparison of three methods of production. South Med J 63:1353, 1970 15. Sealy WC, Brown IW Jr, Young WG Jr: A report on the use of both extracorporeal circulation and hypothermia for open heart surgery. Ann Surg 147:603, 1958 16. Trinkle JK, Franz JL, Furman RW: Circulatory arrest during deep hypothermia induced by peritoneal dialysis. Arch Surg 103:648, 1971 17. Walker JEC, Wells RE: Heat and water exchange in the respiratory tract. Am J Med 30:259, 1961