The effects of deep hypothermic cardiopulmonary bypass and total circulatory arrest on cerebral blood flow in infants and children

J

THoRAc CARDIOVASC SURG

1989;97:737-45

The effects of deep hypothermic cardiopulmonary bypass and total circulatory arrest on cerebral blood flow in infants and children Cardiopulmonary bypass management in infants and children involves extensive alterations in temperature, hemodilution, and perfusion pressure, with occasional periods of circulatory arrest. Despite the use of these biologic extremes of temperature and perfusion, their effects on cerebral blood flow are unknown. This study was designed to examine the relationship of mean arterial pressure and nasopharyngeal temperature to cerebral blood flow during deep hypothermic cardiopulmonary bypass (18 0 to 22 0 C) with and without periods of total circulatory arrest. Cerebral blood flow was measured before, during, and after deep hypothermic cardiopulmonary bypass using xenon clearance techniques in 25 children, aged 2 days to 60 months. Fourteen patients underwent repair with circulatory arrest. There was a highly significant correlation of cerebral blood flow with temperature during cardiopulmonary bypass (p = 0.007). During deep hypothermic bypass there was a significant association between cerebral blood flow and mean arterial pressure (p = 0.027). In infants undergoing repair with deep hypothermia alone, cerebral blood flow returned to prebypass levels in the rewarming phase of bypass. However, in patients undergoing repair with circulatory arrest, no significant increase in cerebral blood flow during rewarming or even after bypass was observed (p = 0.01). These data show that deep hypothermic cardiopulmonary bypass significantly decreases cerebral blood flow because of temperature reduction. Under conditions of deep hypothermia, cerebral pressure-flow autoregulation is lost. This study also demonstrates that cerebral reperfusion after deep hypothermia is impaired if the patient is exposed to a period of total circulatory arrest.

William J. Greeley, MD, Ross M. Ungerleider, MD, L. Richard Smith, PhD, and 1. G. Reves, MD, Durham, s.c.

SystemiC pressure-cerebral blood flow autoregulation and cerebral vascular responses to arterial carbon dioxide tension (Paco.) are maintained in adults during moderate hypothermic cardiopulmonary bypass (CPB) (26° to 28° C).I-4 CPB management techniques in infants and children involve more extensive alterations in temperature, hemodilution, perfusion pressure, and pump flow rates. Indeed, the use of deep hypothermic CPB at temperatures of 18° to 20° C, occasionally with periods of total circulatory arrest (TCA), are commonly used in infants and children undergoing repair of congenital heart defects. Despite the use of these

From the Departments of Anesthesiology and Surgery, The Duke Heart Center, Durham, N.C. Read at the Fourteenth Annual Meeting of The Western Thoracic Surgical Association in Hawaii, June 22-25, 1988. Address for reprints: William J. Greeley, MD, Box 3046, Duke University Medical Center, Durham, NC 27710.

biologic extremes of temperature and perfusion, their effects on cerebral blood flow are unknown. Since deep hypothermic CPB and TCA may be associated with permanent and transient central nervous system dysfunction in infants and children.>" the dynamics of cerebral blood flow under these conditions should be elucidated. This study was designed to examine the relationship of mean arterial pressure and nasopharyngeal temperature to cerebral blood flow during deep hypothermic CPB with and without periods of TCA. Patients and methods Patients. After institutional review board approval and informed parental consent, 25 infants and children undergoing repair of congenital heart defects with the use of deep hypothermic CPB were studied. Ages ranged from 2 days to 60 months. In 14 patients the repair was done with deep hypothermic CPB and TCA (group A) and in 11 patients, with deep hypothermic CPB without TCA (group B). Anesthetic management was identical in all patients and consisted of fentanyl, oxygen, pancuronium, and controlled ventilation.

737

The Journal of Thoracic and Cardiovascular Surgery

7 3 8 Greeley et al.

-

40 c: 35 'E 30 E 0) 25 0 0 20 E 15 LL m 10 o 5 0

•

--...

40 35

• Temp

•.......•

CBF

.'.

' .'

r·············I········· II

30 25 20 15 10 5 0

]

III

VI

-

-... o

-... CD

::J

as CD a.

E ~

Stages Fig. 1. Cerebral blood flow (CBF) and nasopharyngeal temperature during stages II, III, and IV of CPB. Data are shown as mean ± standard deviation. Note the parallel relationship of CBF and temperature with time.

-'E

-c:

E 0)

0 0

T"""

E

u. rn

o

35

= O.494x R = 0.69 P = 0.0002 Y

30 25

- 1.806

20 10 5 0

• •

•

15

• 5

10

15

20

•

• ••

• 25

•

30

35

40

45

50

Arterial Pressure ( mmHg ) Fig. 2. The effect of perfusion pressure on cerebral blood flow (CBF) during deep hypothermic CPB (18° to 22° C, stage II). There was a significant association between arterial pressure and CBF.

No inhalation, vasoactive, or cerebral protective agents were used before cerebral blood flow determinations. CPO management. During CPB, nonpulsatile pump flow with a membrane oxygenator was maintained at the rate of 100 nil/kg/min. The pump-oxygenator system was primed with lactated Ringer's solution, albumin, mannitol, and packed red blood cells to achieve a hematocrit value of 20% ± 2% during CPO. All patients were cooled with the perfusate to deep hypothermic conditions (18° to 22° C). Blood gas management during CPB was directed at maintaining a pH of 7.35 to 7.40 and a Pco, of 35 to 40 mm Hg, uncorrected for temperature. Arterial oxygen tension (Po 2) was maintained between 100 and 200 mm Hg. Blood gases were maintained according to the principles of alpha-stat management because it maintains a constant buffering capacity of the alphaimidazole ring of the histidine amino acid moiety of hemoglobin during hypothermic conditions.I I No interventions were made to control arterial blood pressure during CPB. In both groups of patients, cerebral blood flow determinations at each

measurement interval during bypass were made over a narrow range of hematocrit, temperature, pump flow rate, Paco, and over a wide range of mean arterial pressures. Cerebral blood flow methodology. Regional cerebral blood flow was measured by xenon clearance methods.12• 14 Two extracranial cadmium telluride gamma emission detectors were placed over the right and left temporal lobes to detect the radioactive decay from the brain after injection into the arterial line of the pump-oxygenator of 1.5 mCi of radioactive xenon dissolved in 2 ml of sterile saline. The prebypass and postbypass determinations were made from injections in the ascending aorta by the surgeon. The cerebral blood flow determinations used a modification of the initial slope index method as originally described by Olesen, Paulson, and Lassen.15 According to this formula, cerebral blood flow = (slope) (~) (100), where slope = the natural logarithm of 133Xe clearance curve after the peak of the curve; ~ = tissueblood partition coefficient for xenon, corrected for temperature and hematocrit by the method of Chen and associates";

Volume 97 Number 5

Cerebral blood flow

May 1989

739

Table I. Demographic data Pt. No.

Age

TCA-group A (N = 14) I 3 days 2 10 rna 3 5 rna 4 5 rna 5 6 rna 6 2 rna 7 3.9 yr 8 8 days 9 1.9 yr 9 rna 10 II 6 rna 12 18 days 13 7 days 14 6 rna 9 ± 12 rna DHCPB-group B (N = II) 15 8 rna 16 2.9 yr 17 18 19 20 21 22 23 24 25

4.9 yr 9 rna 4.8 yr 1.9 yr 5 rna 21 days 6 rna 2 days 2 days 19 ± 22 rna

Weight (kg)

Diagnosis

Operation

VSD,CoA AVC AVC AVC MR DORV HAo, CoA TAPVR AS, HAo AVC AVC, PDA lAo, VSD TAPVR AVC

Arch repair, VSD closure Total repair Total repair Total repair MV repair Senning, VSD closure Aortic arch repair Total repair Valvotomy, arch repair Total repair Total repair VSD closure, arch repair Total repair Total repair

6.9 14

TOF L-TGA, VSD

18 8.1 18 10 7.1 3.2 6.1 3.3 3.3 8.9 ± 5.5

AS, AI TOF AVC; MR PA, VSD DCRV,VSD TOF VSD,ASD TOF TGA

Total repair VSD closure, PAB removal Konno procedure Total repair MV repair Rastelli Total repair Total repair Total repair Total repair Arterial switch

3.3 6.1 5.1 5.2 6.1 3.4 20 3.5 12 6.8 6.2 3.1 3.9 3.9 6.3 ± 4.6

CPS (min)

TCA (min)

90 III 115 102 120 120 90 141 95 119 97 91 82 83 104±17*

26 25 47 48 10 67 20 21 19 35 46 26 29 28 32 ± 15

93 65 170 82 80 118 95 98 68 85 170 102 ± 36

Mean ± standard deviation. TCA, Total circulatory arrest; DHCPB, deep hypothermic cardiopulmonary bypass; YSD, ventricular septal defect; CoA, coarctation; A YC, atrioventricular canal defect; MR, mitral regurgitation; DORY, double-outlet right ventricle; HAo, hypoplastic aortic arch; TAPYR, total anomalous pulmonary venous return: AS, aortic stenosis. PDA. patent ductus arteriosus; lAo, interrupted aortic arch; TOF, tetralogy of Fallot; TGA, transposition of the great arteries; AI, aortic insufficiency; PA. pulmonary atresia; DCRY. double-chambered right ventricle; ASD, atrial septal defect; MY, mitral valve; PAB. pulmonary artery band. 'Includes time of TCA.

and 100 converts ml/gm-l/min- 1 to ml/100 grn/rnin". Cerebral blood flow was calculated separately for each individual determination. Cerebral blood flow was calculated as the average value determined at the two probe locations. The need to correct clearance curves for residual xenon was eliminated by beginning subsequent measurements when reactivity had returned to baseline values (approximately 15 minutes after injection of xenon). . Five cerebral blood flow determinations were made in both groups at three predefined intervals during the operation: (I) before CPB (stage I), (2) during CPB at three stages (II = cold, III = cold, IV = rewarmed), and (3) after CPB (stage V). For group A, the CPB determinations (II to IV) were made at stable hypothermic conditions at 18° C before and after circulatory arrest and when the patient was rewarmed. In group B equivalent CPB determinations (II to IV) were made at stable hypothermic conditions between 18° and 22" C at 5 and 25 minutes and when the patient was rewarmed. During each cerebral blood flow measurement, nasopharyngeal temperature, pump flow rate, mean arterial pressure, Paco., and hematocrit value were recorded. To

examine the relationship of temperature and mean arterial pressure to cerebral blood flow during deep hypothermic CPB (I8° to 22°C), data from the stage II determinations were pooled because CPB conditions were similar and more observations could be examined. These same relationships were also examined at normothermic CPB during stage IV determinations for group B, providing a comparison of study variables for deep hypothermic and normothermic CPB conditions. To assess the effect of TCA on cerebral blood flow, group A (TCA) was compared to group B (deep hypothermic CPB without TCA), where conditions were identical excepting the use of TCA. Analysis. Paired data for groups A and B were compared for differences by two-tailed t tests. Because several variables were considered-time, hematocrit value, arterial Pco., nasopharyngeal temperature, mean arterial pressure, pump flow rate, and regional cerebral blood flow for each patientpatient-to-patient variables were serially removed. The variables that did not significantly (p < 0.05) affect regional cerebral blood flow were removed by a stepwise linear regression technique.

740

The Journal of Thoracic and Cardiovascular Surgery

Greeley et al.

Table II. Summary of data for group A (DHCPB + TCA) and group B (DHCPB only) Stages

Study variables

Group A B A B A B A B A B A B

CBF (ml/IOO

gm/rnin) Temperature (0C) Arterial pressure (mm Hg) Paco, (mm Hg) Pump flow rate (ml/rnin) Hematocrit value (%)

During CPB

Before CPB: Stage I 28 27 36 36 56 59 37 39

± ± ± ± ± ± ± ±

12 13 1 I 18 10 7 10

32 ± 4 34 ± 7

Stage II 11 ± 15 ± 18 ± 21 ± 32 ± 30 ± 36 ± 32 ± 732 ± 683 ± 17 ± 20 ±

9 8 2* 2 10 13 6 6 448 305 3 4

Stage III 17 15 22 22 40 41 35 30 739 654 18 20

± ± ± ± ± ± ± ± ± ± ± ±

12 9 5 2 13 18 8 5 407 276 3 4

Stage IV 15 ± 25 ± 36 ± 36 ± 50 ± 43 ± 32 ± 33 ± 841 ± 909 ± 21 ± 20 ±

11* 14 I I 18 20 5 3 430 388 3 3

After CPS: Stage V 17 34 37 37 62 60 34 34

± ± ± ± ± ± ± ±

10* 14 I I II II 5 7

31 ± 6 30 ± 3

Results are expressed as mean values ± standard deviation. DHCPB, Deep hypothermic cardiopulmonary bypass; TCA, total-circulatory arrest; CBF, cerebral blood

flow.

'Group A versus group B, p

< om.

Results A total of 125 cerebral blood flow measurements were made in 25 patients. Demographic data for both groups are shown in Table I. There was no significant difference between groups with respect to age, starting hematocrit value, total CPB duration, or initial cerebral blood flow. Fig. 1 shows the parallel relationship of regional cerebral blood flow and nasopharyngeal temperature with time. There was a highly significant correlation of regional cerebral blood flow with nasopharyngeal temperature during CPB (p = 0.007). The effect of perfusion pressure on cerebral blood flow during deep hypothermic CPB (stage II) for both groups is shown in Fig. 2. There was a highly significant association between cerebral blood flow and mean arterial pressure during deep hypothermic CPB (p = 0.0002) but not during normothermic CPB (stage IV) (p = NS*). Table II contains the study variables at the five stages of measurement for groups A and B. There were no differences between groups at these study stages for Paco., pump flow rate, mean arterial pressure, hematocrit value, or temperature excepting the statistically significant, but clinically insignificant difference in temperature during stage II. However, there was a significant difference in cerebral blood flow between groups A and B at stages IV and V (p = 0.01). Fig. 3 demonstrates the relationship of regional cerebral blood flow with time for groups A and B. As can be seen, both groups A and B showed a significant decrease in *NS

= Not

significant.

cerebral blood flow at deep hypothermic conditions compared to baseline, prebypass levels. In group B, the cerebral blood flow returned to baseline levels in the rewarming phase of CPB. Group A patients, however, in whom repair was done with deep hypothermia and TCA, had no significant increase in cerebral blood flow during rewarming after TCA or even after being weaned from CPB. There was a significant difference in cerebral blood flow between groups A and B during stages IV and V (p = 0.01), compared to stages I, II, and III, during which there was no statistical difference in cerebral blood flow between groups. Fig. 4 shows the individual and mean cerebral blood flow measurements before CPB (stage I) and after CPB (stage V) for both groups. There was a significant decrement in cerebral blood flow after bypass compared to prebypass levels for group A (p = 0.013). All patients survived their operations and were followed for short-term neurologic complications. There was no apparent neurologic sequelae in any patient, excepting one patient who had choreoathetoid movements that resolved after 3 months. This patient underwent atrioventricular canal repair, performed with deep hypothermic CPB and TCA (patient 14 in Table I), and the cerebral blood flow measurements were similar to those of the TCA group characteristics and not unusual in any manner. Discussion Autoregulation of cerebral blood flow is closely coupled to metabolism and determined by local metabolic needs.v'? In the normal, awake individual, cerebral

Volume 97 Number 5

Cerebral blood flow

May 1989

-

-s --.... c:::

'E

50

1II OHCPB + TCA-GRP A •

OHCPB-GRP B

Mean Values + SO

40

"GRP A vs B ""Stage II vs I

E

o

E

741

30

P

< 0.03

20

LL

m 10

o

o

I

Pre-CPB

II III IV L--CPB--J Stages

V

Post-CPB

Fig. 3. Cerebral blood flow (CBF) measurements in the two groups of patients during all five study intervals. There was a significant difference in CBF between groups at stages IV and V.

-

-a,

60

.5 50

E

gE

40 30

- 20 u,

m

o

I

P < 0.05

=--~::....-:::::-..-.-

10

OL.---'----------'-----'-------'----' I

V

L - DHCPB---l

V

'-----TCA--J

Fig. 4. Individual patient cerebral blood flow (CBF) determinations and mean values for groups A and B before and after CPB. Note the significant decrement in CBF after bypass compared to prebypass levels for the TCA group (A). blood flow is maintained constant over a wide range of arterial pressures (60 to 150 mm Hg) and is responsive to the Paco 2• 17 Several recent studies in adult patients have demonstrated the preservation of pressure-flow autoregulation and cerebrovascular response to Paco, during moderate hypothermic CPB (26° to 28° C) and with alpha-stat blood gas management (Paco, 38 to 42).1-3 These studies demonstrate that cerebral blood flow correlates well with nasopharyngeal temperature and is maintained constant through a wide range of perfusion pressures. Some of these studies have also documented the relationship between Paco, and cerebral blood flow, where increases in Paco, cause increases in cerebral blood flow.':' In this study and those of others, 133Xe clearance is used to determine cerebral blood flOW. 1-4. 18 There are

many laboratory methods to determine cerebral blood flow, but clinically, and especially in the operating room.P'Xe clearance has been reliable, reproducible, and accurate. The method is best used to observe changes rather than to quantitate specific regional flow in a particular area. We have elected to use a single bilateral gamma ray detector system since it has been shown by the very low standard deviations of mean values among 16 detectors at different regional sites that cerebral blood flow observations from a single detector provide as much information as multiple detectors during CPB. 2 In other words, although we were measuring regional cerebral blood flow in two areas of the brain, these changes do reflect global alterations in cerebral blood flow with this technique. The cerebral blood flow determinations at baseline in this study are

742

Greeley et al.

similar to those measurements determined in other published studies using 133Xe clearance and other methods in infants and children." Infants and children undergoing repair of congenital heart defects are exposed to more extreme conditions of CPB management. Specifically, these repairs are often performed under conditions of deep hypothermia (18° to 22° C), with or without a period of TCA. This present study investigated the effects of these extreme biologic conditions on cerebral blood flow with an attempt to account for the impact of variables such as temperature and perfusion pressure during deep hypothermia. A principal finding in this study is the significant decrease in cerebral blood flow during deep hypothermic CPB. The changes in cerebral blood flow during CPB are directly related to changes in temperature and presumably reflect reduced cerebral metabolism, which has been shown to be true in adults.' The prebypass cerebral blood flow levels in our patients are similar to published data for infants and children in other clinical situations." The higher prebypass cerebral blood flow levels (compared with our published adult data) in this present study are compatible with the known progressive age-related decreases in cerebral blood flow.4, 2o The larger decrements in cerebral blood flow during bypass in our patients compared with our adult studies are most likely related to the lower temperatures that are used with deep hypothermic CPB in infants and children. A new finding in this investigation is that deep hypothermia (unlike moderate) abolishes pressure-flow autoregulation. There is a significant association between cerebral blood flow and mean arterial pressure during deep hypothermic conditions (18 ° to 22 ° C), which suggests pressure-dependent cerebral blood flow at these low temperatures. However, during normothermic CPB (stage IV) there was no correlation between cerebral blood flow and mean arterial pressure, which indicates the presence of autoregulation. These observations suggest a loss of autoregulation of cerebral blood flow during deep hypothermia when mean arterial pressure varied widely (mean = 32 ± 10 rom Hg, range 15 to 65) and a return to a pressure-flow autoregulatory state during rewarming. The loss of autoregulation of cerebral blood flow in our study is most likely due to the deep hypothermic conditions. In a previous study, we have demonstrated a similar loss of cerebral autoregulation in infants and children during deep hypothermic CPB (18° to 22°) and maintenance of cerebral pressure-flow autoregulation during moderate hypothermic CPB (25° to 32°) within the same range of perfusion pressures. 21 An alternative explanation is the loss of this cerebral pressure-flow response resulting from low per-

The Journal of Thoracic and Cardiovascular Surgery

fusion pressures outside the range of cerebral autoregulation. In a recent experimental study in animals, a similar loss of cerebral blood flow autoregulation at low perfusion pressure «40 rom Hg) during deep hypothermic CPB conditions (20° C) was observed, but at higher perfusion pressures cerebral autoregulatory response could be detected. 22 Although current techniques of deep hypothermic CPB with TCA may have no apparent long-term neurologic or developmental sequelae, little information is available with regard to the critical aspects of this technique of CPB management. Several reports have documented the association of TCA with transient central nervous system,dysfunction. More subtle longterm neuropsychologic abnormalities have been suggested but not proved.":" The data from our study suggest that one possible mechanism causing transient cerebral dysfunction is abnormal reperfusion after TCA. Norwood, Norwood, and Castaneda," in a canine model, examined the effect of deep hypothermic circulatory arrest on cerebral reperfusion using a carbon black infusion technique. In that study, cerebral hypoperfusion, or no-reflow, was observed after normothermic circulatory arrest. This no-reflow phenomenon was abolished by deep hypothermia. In our study, unlike the Norwood study in animals," cerebral reperfusion after deep hypothermia was significantly impaired if the patient was exposed to a period of TCA. That this observation is related to the effects of TCA is strongly suggested by the fact that cerebral reperfusion was normal in a group of patients (group B) who were similar in all respects except for the exposure to TCA. Flow rates were always maintained at the calculated rate of normal for each patient, and the physiologic values of pH, Paco., and arterial P0 2 supported the adequacy of perfusion. There were no demonstrable differences in perfusion pressures or other variables to explain this difference in cerebral reperfusion between groups. We believe one possible explanation of the failure to increase cerebral blood flow after TCA is simply an appropriate cerebrovascular response to the markedly depressed metabolism of the rewarmed brain after circulatory arrest, which is evidenced by decreased cerebral activity on the electroencephalograms usually seen during and after rewarming." In other words, metabolic autoregulation, defined as coupling of flow and metabolism, is preserved after TCA where reduced CBF is related to reduced cerebral metabolic rate, and where brain oxygen supply jdemand ratio is maintained. Since oxygen consumption was not measured and cerebral metabolic rate for oxgyen was not determined, our explanation of balanced flowjmetabolism was not verified by this study and therefore remains speculative.

Volume 97 Number 5 May 1989

There was no association between the duration of TCA and changes in cerebral blood flow, but the number of observations limits a definitive conclusion. Of interest, the two patients in group A (TCA) who had a return of cerebral blood flow to baseline after CPB, as shown in Fig. 4, had the shortest TCA times (patients 5 and 9 in Table I). Whether impaired cerebral reperfusion after deep hypothermic CPB with TCA is related to a depressed cerebral metabolic rate and can be altered by changing flow rates, perfusion pressure, or administering cerebral protective agents, or whether it is related in severity to the duration of the TCA period, will be investigated by future studies. The clinical importance of these observations remains to be determined, since only one of these patients exhibited any signs of neurologic dysfunction in the postoperative period (transient choreoathetoid movements); however, the understanding of how cerebral reperfusion is affected by TCA and how those changes can be best controlled has far-reaching implications. In conclusion, changes in cerebral blood flow during deep hypothermic CPB are primarily due to changes in temperature where reduced metabolic rate reduces cerebral blood flow. Preservation of pressure-eerebral blood flow autoregulation is lost under deep hypothermic CPB conditions. Whether this finding is related to the direct effects of deep hypothermia or related to the perfusion pressures attendant with the use of deep hypothermic CPB which are outside the autoregulatory range remains to be determined. Our study also suggests that cerebral reperfusion is impaired after TCA. Further studies examining the relationship of cerebral blood flow and metabolism to deep hypothermic CPB with TCA, especially with respect to how these changes may impact on long-term neuropsychologic development, are warranted.

1.

2.

3.

4.

REFERENCES Govier AV, Reves JG, McKay RD, et al. Factors and their influence on regional cerebral blood flow during nonpulsatile cardiopulmonary bypass. Ann Thorac Surg 1984;38:592-600. Prough DS, Stump DA, Roy RC, et al. Response of cerebral blood flow to changes in carbon dioxide tension during hypothermic cardiopulmonary bypass. Anesthesiology 1986;64:576-81. Murkin JM, Farrar JK, Tweed WA, McKenzie FN, Guirauden G. Cerebral autoregulation and flow / metabolism coupling during cardiopulmonary bypass: The influence of PaCO z. Anesth Analg 1987;66:825-32. Brusino FG, Reves JG, Smith LR, Prough DS, Stump DA. McIntyre RW. The effect of age on cerebral blood

Cerebral blood flow

74 3

flow during hypothermic cardiopulmonary bypass. J THORAC CARDIOVASC SURG [in press]. 5. Clarkson PM, Barton MB, MacArthur A, et al. Developmental progress after cardiac surgery in infancy using hypothermia and circulatory arrest. Circulation 1980; 62:855-61. 6. Kramer RS, Sanders AP, Lesage AM, Woodhall B, Sealy We. The effect of profound hypothermia on preservation of cerebral ATP content during circulatory arrest. J THORAC CARDIOVASC SURG 1968;56(5):699-709. 7. Cohen ME, Olszowka JS, Subramanian S. Electroencephalographic and neurological correlates of deep hypothermia and circulatory arrest in infants. Ann Thorac Surg 1977;23:238-44. 8. Messmer BJ, Schallberger U, Gattiker R, Senning A. Psychomotor and intellectual development after deep hypothermia and circulatory arrest in early infancy. J THORAC CARDIOVASC SURG 1976;72:495-502. 9. Stevenson JG, Stone EF, Dillard DH, Morgan Be. Intellectual development of children subjected to prolonged circulatory arrest during hypothermic open heart surgery in infancy. Circulation 1974;49,50(Pt 2): 1154-9. 10. Hicks RG, Poole JL. Electroencephalographic changes with hypothermia and cardiopulmonary bypass in children. J THORAC CARDIOVASC SURG 1981;81:781-6. 11. Swan H. The importance of acid-base management for cardiac and cerebral preservation during open heart operations. Surg Gynecol Obstet 1984;158:391-414. 12. Obrist WD, Wilkinson WE. Stability and sensitivity of CBF indices of the noninvasive 13JXe method. In: Hartment A, Hoyer S, eds. Cerebral blood flow and metabolism measurement. New York: Springer-Verlag, 1985: 30-6. 13. Waltz AG, Wanek AR, Anderson RE. Comparison of analytic methods for calculation of cerebral blood flow after intracarotid injection of 13JXe. J Nucl Med 1971; 13:66-72. 14. Fan FC, Chen RYZ, Schuessler GB, Chien S. Comparison between the 13JXe clearance method and the microsphere technique in cerebral blood flow determinations in the dog. Circ Res 1979;44:653-9. 15. Olesen J, Paulson OB, Lassen NA. Regional cerebral blood flow in man determined by the initial slope of the clearance of intraarterially injected 13JXe. Stroke 1971; 2:519-25. 16. Chen RYZ, Fan FC, Kim S, Jan KM, Usami S, Chien S. Tissue-blood partition coefficient for xenon: temperature and hematocrit dependence. J Appl Physiol 1980;49: 17883. 17. Strandgaard S, Paulson OB. Cerebral autoregulation. Stroke 1984;15:413-6. 18. Henrikson L. Brain luxury perfusionduring cardiopulmonary bypass in humans: A study of the cerebral blood flow response to changes in COz, O 2, and blood pressure. J Cereb Blood Flow Metab 1986;6:366-78. 19. Younkin DP, Reivich M, Jaggi J, Obrist W, DelivoriaPapadopoulos M. Noninvasive method of estimating

The Journal of Thoracic and Cardiovascular Surgery

7 4 4 Greeley et al.

human newborn regional cerebral blood flow. J Cereb Blood Flow Metab 1982;2:415-20. 20. Globus M, Melamed E. Progressive age-related decrease in regional cerebral blood flow in healthy subject. Israel J Med Sci 1985;21:662-5. 21. Greeley WJ, Ungerleider RM, Smith LR, Reves JG. Cardiopulmonary bypass alters cerebral blood flow in infants and children during and after cardiovascular surgery. Circulation 1988;78(Pt 4):II356. 22. Tanaka J, Shiki K, Asou T, et al. Cerebral autoregulation during deep hypothermic nonpulsatile cardiopulmonary bypass with selective cerebral perfusion in dogs. J THORAC CARDIOVASC SURG 1988;95:124-32. 23. Henriksen L, Hjelms E, Lindeburgh T. Brain hyperperfusion during cardiac operations. J THORAC CARDIOVASC SURG 1983;86:202-8. 24. Ehyai A, Fenichel GM, Bender HW. Incidence and prognosis of seizures in infants after cardiac surgery with profound hypothermia and circulatory arrest. JAMA 1984;252-3165-7. 25. Tharion J, Johnson DC, Celermajer JM, et al. Profound hypothermia with circulatory arrest: nine years' clinical experience. J THORAC CARDIOVASC SURG 1982;84:66-72. 26. Weiss M, Weiss J, Cotton J, et al. A study of the electroencephalogram during surgery with deep hypothermia and circulatory arrest in infants. J THORAC CARDIOVASC SURG 1975;70:316-29. 27. Blackwood MJA, Haka-Ikse K, Steward DJ. Developmental outcome in children undergoing surgery with profound hypothermia. Anesthesiology 1986;65:437-40. 28. Haka-I1se K, Blackwood MJA, Steward DJ. Psychomotor development of infants and children after profound hypothermia during surgery for congenital heart disease. Dev Med Child Neurol 1978;20:62-70. 29. Dickinson DF, Sambrooks JE. Intellectual performance in children after circulatory arrest with profound hypothermia in infancy. Arch Dis Child 1979;54:1-6. 30. Wells FC, Coghill S, Caplan HL, Lincoln C. Duration of circulatory arrest does influence the psychological development of children after cardiac operation in early life. J THORAC CARDIOVASC SURG 1983;86:823-31. 31. Molina JE, Einzig S, Mastri AR, et al. Brain damage in profound hypothermia: perfusion versus circulatory arrest. J THoRAe; CARDIOVASC SURG 1984;87:596-604. 32. Newburger JW, Silbert AR, Buckley LP, Fyler DC. Cognitive function and age at repair of transposition of the great arteries in children. N Engl J Med 1984; 310:1495-9. 33. Treasure T, Naftel DC, Conger KA, Garcia JH, Kirklin JW, Blackstone EH. The effect of hypothermic circulatory arrest time on cerebral function, morphology, and chemistry: an experimental study. J THORAC CARDIOVASC SURG 1983;86:761-70. 34. Ellis RJ, Wisniewski A, Potts R, Calhoun C, Loucks P, Wells MR. Reduction of flow rate and arterial pressure at moderate hypothermia does not result in cerebral dysfunction, J THORAC CARDIOVASC SURG 1980;79:173-80.

35. Lundar T, Froysaker T, Nornes H. Cerebral damage following open-heart surgery in deep hypothermia and circulatory arrest. Scand J Thorac Cardiovasc Surg 1983; 17:237-42. 36. Muraoka R, Yokota M, Aoshima M, et al. Subclinical changes in brain morphology following cardiac operations as reflected by computed tomographic scans of the brain. J THORAC CARDIOVASC SURG 1981;81:364-9. 37. Norwood WI, Norwood CR, Castaneda AR: Cerebral anoxia: effect of deep hypothermia and pH. Surgery 1979;86:203-9. 38. Weiss M, Weiss J, Cotton J, et al. A study of the electroencephalogram during surgery with deep hypothermia and circulatory arrest in infants. J THORAC CARDIOVASC SURG 1975;70:31,6-29.

Discussion Dr. Edward S. Yee (Johnson City, Tenn.). Philosophically speaking, I have very little disagreement with your findings. I believe that some perfusion, even at low flow, probably affords better organ preservation during hypothermic bypass. However, the clinical use of TCA with carefully planned operative management has demonstrated excellent clinical results. By adhering to the objectives of achieving core temperatures in the 170 _ to 20 0-degree range, adequate hemodilution of hematocrit to around 20% ± 2%, and, more important, by minimizing the TCA period below the 45- to 6D-minute time frame, one can achieve excellent results, as witnessed in group A patients. These guidelines of using TCA have allowed delicate repairs in extremely small and young infants, in whom often the arterial and, more commonly, the atrial cannulas limit the operative exposure and technical repair. Elimination of the cannulas offered an unhindered, asanguineous operative field during TCA. Analogously, in adults the techniques of TCA have allowed important surgical repair in cases in which arterial perfusion cannot be maintained, such as the transverse aortic arch or performing hypothermic arrest for intracranial cerebral repair of giant aneurysm or arteriovenous malformation. For these reasons, I would like to ask three sets of questions regarding the methodology, the data analysis, and the clinical implications of your results. First, the arrived data or methodology evolved centered around two sets of clearance study during stage IV rewarming and stage V off bypass. Were these two time marks truly comparable, i.e., the 5- and 25-minute time part for group B as compared to arrested group A? It seems to me, in looking at the graphs, that group B patients were rewarmed longer. Dr. Greeley. The cerebral blood flow determination stages (I to V) were event-oriented (pre-CPB, pre-TCA, and postTCA) rather than time-oriented. We attempted to associate known intraoperative changes in CPB management with alterations in cerebral blood flow. There was no difference between the two groups at stages III, IV, and V with respect to temperature, arterial Pco., mean arterial pressure, hematocrit value, or pump flow rate. Therefore, although the stages may not appear to be comparable, the variables that most influence cerebral blood flow were similar and allowed group comparison. Furthermore, that the observed differences between groups for cerebral blood flow at stages IV and V are related to the effects ofTCA is strongly suggested by the similar study

Volume 97 Number 5 May 1989

conditions of both groups except for the exposure to TCA for group A. Dr. Vee. Could the head size, patient's body size compared to the xenon probe, and the methodology of examining the xenon clearance (i.e., the linear analysis as compared to exponential decay or kinetic analysis) account for some of these measured differences? I raise this question because you cited a dozen or so references in which 133Xe was used; this is the first study in which no flow intervention had been applied. Could this be artifactual? Dr. Greeley. Xenon clearance methodology for the measurement of cerebral blood flow is an effective, reproducible, convenient technique to measure cerebral blood flow in the operating room. This technique of cerebral blood flow determination has been shown to be reliable when compared to traditional methods of cerebral blood flow determinations, such as the Kety-Schmidt nitrous oxide washout technique. The results obtained in our study are similar to the cerebral blood flow measurements determined by xenon clearance and reportedin neonates and infants during other pathologic states, such as newborn asphyxial injury. There are unpublished data by Kirklin and associates in gerbils demonstrating a very similar no-reperfusion trend after TCA (data presented at First World Congress of Pediatric Cardiac Surgery, Bergamo, Italy, 1988). In their experiment they exposed gerbils to deep hypothermic bypass conditions with and without circulatory arrest. In the arrest group they observed a lack of reperfusion after TCA. Dr. Vee. One of the references" cited states that recirculation from extracerebral sources could account for some of the washout and for the important variables you have outlined in terms of cerebral blood flow, temperature, mean arterial pressure, and carbon dioxide. I was somewhat surprised that, despite your multiple regression analysis, there was no difference between the two age groups. I think that neither age nor hematocrit was analyzed. Would you care to speculate what accounts for your findings? Was it a source of micro air embolism that occurs during TCA and blocks some of the channels for reperfusion? You mentioned that some animal work or future work is in progress. Dr. Greeley. At this point we are unable to explain the lack of reperfusion after TCA. This phenomenon of no reperfusion after prolonged ischemia has been shown by numerous groups to occur under normothermic conditions. Norwood and associates have shown experimentally in animals that deep hypothermia abolishes this "no-reflew" after TCA. The previously mentioned work by Kirklin's group and the findings in our study show that deep hypothermia 'with rCA impairs perfusion. We do not understand the mechanism for this finding. Whether it is due to intraluminary collapse and an inability to reopen the arterioles and capillaries is one possibility. More important, the effects are probably related to a change in the

Cerebral blood flow

745

local milieu near the cerebral arterioles where metabolites and vasoactive substances may inhibit reperfusion. Dr. Vee. Finally, given these excellent clinical results, would you now advocate delay and palliation, to avoid TCA, for most congenital heart lesions, where often timing is critical? Dr. Greeley. No, I think the technique of TCA, as outlined in your introductory remarks, is an important, established method for repair of complex congenital heart defects in small infants and children. In no way do I mean to imply that this technology should not be used. I believe the findings of a lack of reperfusion after TCA are interesting and need further investigation. In the 14 patients who underwent TCA in this study, there were no long-term neurologic sequelae despite the lack of reperfusion. Many large series have shown that there are no long-term neurologic sequelae after TCA. I suspect that there is a delayed reperfusion which is equally matched by reduced cerebral metabolism after TCA. There probably is no imbalance of oxygen supply and demand, which would suggest no ischemic injury. We are currently examining cerebral blood flow changes and long-term neuropsychologic outcome. I wish to reemphasize two aspects of this study, First, under deep hypothermic conditions there apparently is a loss at low perfusion pressures of cerebral blood flow autoregulation for mean arterial pressure. This has been shown experimentally by Fox and associates (1984;87:658-64) and, recently, by Tanaka and associates (1988;95:124-32), both in reports from this JOURNAL. Our observation of a loss of cerebral blood flow autoregulation during deep hypothermia is in contrast to our study in adult patients and our study in children during mild hypothermic or normothermic conditions, for cerebral blood flow is independent of arterial pressure in the latter groups. The implication of this finding in our study is that there is a vasoparesis under deep hypothermic conditions, during which cerebral blood flow is very dependent on perfusion technology (i.e., arterial pressure and perhaps pump flow rate). That is, the patient is dependent on pump flow rate and perfusion pressure, unlike normothermic or moderate hypothermic conditions. The second implication is that there is impaired reperfusion after TCA. As mentioned previously, this has also been shown experimentally by Kirklin and associates. Clinically, this correlates with the abnormal electroencephalograms that one sees after TCA in these children. With time, all these changes are reversible. I suspect, therefore, that there is a parallel reduction in cerebral metabolism for oxygen after TCA, so that there is no imbalance of oxygen supply/demand causing injury. All the patients in our series survived their operations and did well, with no evidence of long-term neurologic sequelae. It is clear that future studies are necessary to further elucidate the mechanisms as well as the causes of these cerebral blood flow changes.