Neurocognitive Function in Patients Undergoing Coronary Artery Bypass Graft Surgery With Cardiopulmonary Bypass: The Effect of Two Different Rewarming Strategies

Neurocognitive Function in Patients Undergoing Coronary Artery Bypass Graft Surgery With Cardiopulmonary Bypass: The Effect of Two Different Rewarming Strategies Bikash Sahu, MD, DNB, PDCC,* Sandeep Chauhan, MD, PDCC,* Usha Kiran, MD,* Akshay Bisoi, MCh,† Ramakrishnan Lakshmy, MD,‡ Thiruvenkadam Selvaraj, MD, DNB, MNAMS,* and Ashima Nehra, PhD§ Objective: Hypothermia followed by rewarming during cardiopulmonary bypass can lead to cerebral hyperthermia, which has been implicated as 1 of the causes for postoperative deterioration in neurocognitive function in patients undergoing coronary revascularization. Hence, the authors studied the effects of 2 different rewarming strategies on postoperative neurocognitive function in adult patients undergoing coronary artery bypass graft surgery with the aid of cardiopulmonary bypass. Design: This was a randomized clinical trial. Setting: A cardiothoracic center of a tertiary level referral, teaching hospital. Participants: A total of 80 adult patients aged 45 to 70 years undergoing elective primary isolated coronary artery bypass graft surgery with cardiopulmonary bypass under moderate hypothermia at 30°C were included in this study. Interventions: The patients were randomly allocated into 2 groups of 40 each. In group A, patients were rewarmed to a nasopharyngeal temperature of 37°C; whereas, in group B, patients were rewarmed to a nasopharyngeal temperature of 33°C before weaning off bypass. The anesthetic and bypass management were standardized for both groups. Measurements: All patients were assessed for neurocognitive function preoperatively and on the fifth postoperative day using the Post Graduate Institute Memory Scale. The amount of blood loss and need for blood and blood product

transfusion postoperatively, the need for pacing, increased inotrope or vasodilator use, and time to extubation were also noted. Serum S100␤ levels were measured after anesthetic induction and at 24 hours postoperatively. The jugular venous oxygen saturation and oxygen tension were noted at 30°C and at the end of full rewarming (ie, at 37°C or 33°C, respectively, in the 2 groups). Results: There was a significant deterioration in neurocognitive function postoperatively as compared with preoperative function in patients of group A (37°C). This was associated with higher S100␤ levels 24 hours postoperatively in group A (37°C) compared with group B (33°C) patients. Also, there was a significant decrease in jugular venous oxygen saturation in group A (37°C) as compared with group B (33°C) at the end of rewarming. The time to extubation was longer in group B (33°C). No significant differences were noted in the amount of postoperative blood loss, blood and blood product use, inotrope or vasodilator use, and the need for pacing. Conclusion: Weaning from CPB at 33°C may be a simple and useful strategy to lower the postoperative impairment of neurocognitive function and may be used as a tool to decrease morbidity after coronary revascularization. © 2009 Elsevier Inc. All rights reserved.

T

temic end-organ protection. The nasopharyngeal temperature used for core temperature monitoring during bypass underestimates the brain temperature during rewarming. Thus, rewarming at the end of cardiac surgery can lead to cerebral hyperthermia. There is evidence that cerebral blood flow may be insufficient relative to the rising cerebral oxygen demand during and after rewarming from hypothermic CPB.3,4 Thus, the imbalance between cerebral oxygen supply and demand after rewarming may contribute to neurologic and neurocognitive injury. Although the anatomic and functional extent of embolic stroke can be determined through physical assessment and imaging techniques, there are inherent difficulties with the detection of subclinical problems such as neurocognitive dysfunction.5 The authors hypothesized that the risk of cerebral hyperthermia can be reduced by weaning patients from CPB at 33°C. The slower rate of rewarming, lower peak temperatures at the time of weaning from CPB, and sustained mild hypothermia postoperatively can reduce neurocognitive dysfunction postoperatively. Hence, the authors undertook this study to compare the effects of 2 different rewarming strategies on postoperative neurocognitive function in patients undergoing CABG surgery with the aid of CPB.

HE USE OF CARDIOPULMONARY BYPASS (CPB) for coronary artery bypass graft (CABG) surgery is associated with postoperative neurocognitive complications such as deficits in memory, attention, concentration, and learning.1 The etiology of this postoperative cognitive dysfunction is multifactorial.1 The impairment in neurocognitive and neuropsychologic performance occurs in 30% to 60% of cardiac surgery patients in the first postoperative week (5-8 days). Although neuropsychologic impairment tends to improve with time, 10% to 30% of cardiac surgery patients have measurable deterioration of their cognitive status months to years after undergoing surgery.2 Despite improvements in surgical and anesthetic techniques, the frequent incidence of postoperative neurocognitive dysfunction remains a concern because of its associated morbidity, impaired quality of life, and increased perioperative cost. Hypothermia has been used conventionally to improve myocardial and cerebral tolerance to ischemia and to provide sys-

From the Departments of *Cardiac Anesthesiology, †Cardiothoracic and Vascular Surgery, ‡Cardiac Biochemistry, and §Clinical Psychology, All India Institute of Medical Sciences, New Delhi, India. Address reprint requests to Bikash Sahu, MD, DNB, PDCC, Department of Cardiac Anesthesia, CN Centre, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110029, India. E-mail: drbikash_ [email protected] © 2009 Elsevier Inc. All rights reserved. 1053-0770/09/2301-0003$36.00/0 doi:10.1053/j.jvca.2008.07.010 14

KEY WORDS: rewarming, neurocognitive function, jugular venous oxygen saturation, S100␤

MATERIALS AND METHODS This study was conducted at the authors’ cardiothoracic center after obtaining approval of the hospital ethics committee and informed consent from the patients. The sample size was calculated by using STATA-9 software (STATA, College Station, TX) and calculated

Journal of Cardiothoracic and Vascular Anesthesia, Vol 23, No 1 (February), 2009: pp 14-21

NEUROCOGNITIVE FUNCTION AND CABG SURGERY

power ⬎85% with an ␣ ⫽ 0.05 (level of significance) if the 2 groups contained 40 patients each. Thus, a total of 80 adult patients aged 45 to 70 years undergoing elective primary isolated CABG surgery with CPB under moderate hypothermia at 30°C were included in this study. The patients were randomly allocated into 2 groups of 40 each using Fisher random number tables. In group A, patients were rewarmed to a nasopharyngeal temperature of 37°C, whereas group B patients were rewarmed to a nasopharyngeal temperature of 33°C before weaning them from bypass. The following patients were excluded from the study: those with pre-existing cerebrovascular disease, psychiatric illness, psychotropic drug abuse, alcoholism, renal disease (Se creatinine ⬎2 mg/dL), left ventricular ejection fraction ⬍40%, pre-existing bleeding disorders, preoperative rhythm disturbances, patients undergoing concomitant aneurysmectomy, ventricular septal defect closure, valvular repairs/ replacements and other vascular procedures, redo CABG surgery, and emergency CABG surgery. The Post Graduate Institute (PGI) memory scale used for neurocognitive assessment required patients to be able to read the questionnaire and write the answers down; hence, patients who could not read or write were excluded from this study. This test can be used in patients with a minimum education more than 7th grade; hence, patients with less than a 7th-grade education were excluded from this study. All patients were premedicated with morphine, 0.1 mg/kg, and promethazine, 0.5 mg/kg, intramuscularly, 30 to 45 minutes before the induction of anesthesia. Anesthesia was induced in all patients with intravenous thiopental (3-5 mg/kg), fentanyl (2-3 ␮g/kg), midazolam (1-2 mg), and the trachea was intubated after intravenous vecuronium (0.1 mg/kg). Anesthesia was maintained with O2 in air (50%), sevoflurane, and supplemental doses of intravenous fentanyl, midazolam, and vecuronium administered before skin incision, sternotomy, aortic cannulation, on bypass, at the start of rewarming, and after weaning from bypass. For the purpose of jugular venous cannulation, the distance between the proposed point of insertion to the ipsilateral mastoid process was measured, and a Secalon T 16-G cannula (2.0/160 mm; Ohmeda, Swindon, UK) was passed cephalad into the internal jugular vein to measure the jugular venous oxygen saturation and tension. The position of the cannula was confirmed postoperatively by x-ray (anteroposterior view, the tip of the cannula lies cranial to a line connecting the tips of the mastoid process). All patients underwent elective primary isolated CABG surgery with CPB under moderate hypothermia at 30°C. The CPB equipment consisted of a membrane oxygenator with a hard shell venous reservoir (Capiox SX18, Terumo), roller pump, arterial filter (Affinity, Medtronic), blood cardioplegia delivery system (Dideco), and 1/2-inch ⫻ 3/8th inch circuit tubings. The circuit was primed with Ringer’s lactate (15 mL/ kg), hydroxyethyl starch (Voluven 6% 130/0.4 [Fresenius Kabi, Friedberg, Germany], 10 mL/kg), mannitol (20%, 5 mL/kg), heparin, and sodium bicarbonate (1 mL/kg, 7.5% w/v). The hematocrit was maintained ⱖ25% on bypass with the addition of packed red blood cells if required. Systemic anticoagulation was achieved with intravenous heparin, 400 U/kg. The pump flow rates were maintained at 2.2 to 2.6 L/min/ m2. The perfusion pressures were maintained at 50 to 80 mmHg. If the pressure was below 50 mmHg, diluted phenylephrine (50-100 ␮g) was given. For pressures above 80 mmHg, boluses of fentanyl, midazolam, and vecuronium were given. If these failed to lower pressures, nitroglycerin infusion at 0.5 to 1 ␮g/kg/min was added on bypass; ␣-stat strategy was used for blood gas management on bypass. The blood sugar was maintained between 100 to 200 mg/dL on bypass with the addition of insulin if required. No antifibrinolytics were used in any of the patients. The core temperature was monitored by using a temperature probe in the nasopharynx. The peripheral temperature was simultaneously mon-

15

itored on the dorsum of the great toe, and a core periphery gradient of ⱕ1°C was maintained at all times. All patients were cooled to a nasopharyngeal temperature of 30°C during the surgery. Rewarming was initiated at the beginning of the last distal graft anastomosis. The patients were rewarmed to either 37°C (group A) or 33°C (group B) depending on the study group. TCM (temperature controlling machine; Sarns 3M Health Care, Ann Arbor, MI) was used for cooling and rewarming. The gradient between the water bath and the blood temperature was kept between 2° and 4°C. The aortic cross-clamp was released at the completion of the last distal anastomosis, and, thereafter, proximal anastomoses were done using an aortic side-biting clamp. Once the nasopharyngeal temperature reached 33°C in group B patients, active blood rewarming was stopped, and temperature drift was prevented by maintaining the temperature of the water bath at 33°C and by the use of a warm-water mattress with temperature maintained at 33°C until the completion of surgery. In group A patients, the rewarming was continued until a nasopharyngeal temperature of 37°C was reached, and, thereafter, active blood rewarming was stopped and temperature drift was prevented by maintaining the temperature of the water bath at 37°C and by the use of a warm-water mattress with the temperature at 37°C until the completion of surgery. The overshoot of temperature during rewarming on CPB was assiduously avoided. The mean nasopharyngeal temperatures at weaning from CPB were 37°C and 33°C, respectively, in groups A and B. All patients were weaned from CPB using intravenous infusions of dopamine, 5 ␮g/kg/min, and/or nitroglycerin, 0.5 ␮g/kg/min. Patients requiring intravenous infusions of inotropes or vasodilators greater than the previously mentioned doses for successful weaning from CPB were classified as “increased need of inotropes or vasodilators.” Anticoagulation was reversed by using intravenous protamine, 6 mg/kg. The jugular venous oxygen saturation (SjvO2) was measured at 2 time points: one at the lowest temperature on bypass (30°C) and other at the end of full rewarming (37°C or 33°C, respectively) in the 2 groups. Serum S100␤ levels were measured after anesthetic induction and at 24 hours postoperatively using CanAg S100BB EIA kit (CanAg Diagnostics AB, Gothenburg, Sweden). The CanAg S100BB EIA is a solid-phase noncompetitive assay based on the direct sandwich technique for optimal clinical sensitivity and specificity for determination of the S100BB isoform. The assay is based on an antibody specific for the S100BB dimer as catcher and horseradish peroxidase–labeled monoclonal antibody-specific for S100B for detection. The neurocognitive functions were assessed on the day before surgery and on the 5th postoperative day using the PGI Memory Scale.6,7 The neuropsychologist performing these tests as well as the intensive care unit (ICU) caregivers were blinded to the 2 groups. The same temperature-monitoring sites (nasopharynx and dorsum of great toe) were used postoperatively in the ICU. All the patients were allowed to rewarm passively and spontaneously in the ICU postoperatively; no active rewarming measures were used. All the patients were kept sedated and paralyzed by using intermittent doses of intravenous fentanyl, midazolam, and vecuronium on mechanical ventilation postoperatively. The dopamine and nitroglycerin infusions were continued postoperatively in the ICU. Once routine ventilatory weaning criteria were met, the patients were weaned from ventilatory support and extubated. The amount of blood loss, blood and blood product transfusion postoperatively, the need for pacing, the need of inotropes or vasodilators, and the time to extubation were noted. Thereafter, the dopamine and nitroglycerin infusions were tapered off and stopped based on hemodynamics. The Post Graduate Institute of Medical Education & Research, Chandigarh, India, memory scale is the Indian adaptation of the internationally acclaimed Wechsler Adult Intelligence Scale. The socioeconomic and educational backgrounds are important factors affecting neurocognitive function. Because the Indian patients belong to different

16

SAHU ET AL

Table 1. Demographic Data Variables

Group A (37°C) (n ⫽ 40)

Group B (33°C) (n ⫽ 40)

p Value

Age (y) Sex (M:F ratio) Educational level (y) Diabetes (no. of pts) No of grafts CPB (min) AoXCl (min)

57.3 ⫾ 8 35:5 12.5 ⫾ 2.2 6 3.7 ⫾ 0.7 67.3 ⫾ 25.9 39.3 ⫾ 14.7

56 ⫾ 6.9 33:7 12 ⫾ 2 9 3.7 ⫾ 0.7 59.2 ⫾ 21.3 36.6 ⫾ 13.9

0.44 0.53 0.8 0.3 0.87 0.13 0.39

NOTE. Continuous data expressed as mean ⫾ standard deviation. A t test was used for comparison of continuous variables, and a chisquare test was used for categoric variables; p ⬍ 0.05 statistically significant. Abbreviation: AoXCl, aortic cross-clamp.

social, economic, and educational background compared with the Western population, the authors used this test meant for the Indian population instead of the tests (eg, Wechsler Adult Intelligence Scale) meant for the Western population quoted in previous studies. The PGI memory scale consists of a questionnaire in Hindi and English, which the patient is supposed to read and answer. The questionnaire includes a battery of 11 tests covering a variety of cognitive domains (eg, remote memory, recent memory, mental balance, attention and concentration, delayed recall, immediate recall, retention, visual retention, and recognition). The test took around 60 minutes to be performed. Based on the answers given by the patient, the “raw scores” are calculated. The nomograms for different age groups and educational levels derived from an Indian population are available along with the questionnaire. These nomograms give the mean reference scores along with the standard deviation for individuals of different age groups and educational levels. From the raw score, the reference score obtained from the nomogram is subtracted and the difference divided by the standard deviation of the population of the reference value to get the “composite score or z score.” The preoperative and postoperative composite z scores were then used for intergroup and intragroup comparison and statistical analysis using the following formula: Composite z score ⫽ (raw score ⫺ reference score) ⁄ standard deviation of reference population The results were analyzed using STATA-9 software. The Student t test and paired t tests were used to compare normally distributed data. The Mann-Whitney U test and Wilcoxon signed-rank test were used to compare nonnormally distributed data. The categoric measures were compared by using a chi-square test; p ⬍ 0.05 was used to express statistical significance. RESULTS

The patients’ demographic data are listed in Table 1. Both the groups were comparable with regard to age, sex, education, number of diabetics, number of vessels grafted, CPB time, and aortic cross-clamp time. The outcome variables are listed in Table 2. There was no significant difference in the need for pacing while weaning from CPB between the 2 groups. The amount of blood loss postoperatively in group A (37°C) was comparable to that in group B (33°C). The amount of blood and blood products transfused postoperatively were also comparable between the 2 groups. There was no significant difference in the need for inotropes and dilators between the 2 groups.

The nasopharyngeal temperatures on arrival in the ICU were 36.5° ⫾ 0.5°C in group A (37°C) and 32.5° ⫾ 0.5°C in group B (33°C). The time for rewarming was the time needed for the nasopharyngeal temperature to reach 37°C postoperatively in both the groups after shifting the patient to the ICU in the absence of active rewarming measures. The patients in group B (33°C) required significantly greater time to rewarm to 37°C postoperatively as compared with patients in group A (37°C). The time to extubation from admission to the ICU was significantly prolonged in group B (33°C) as compared with group A (37°C). The baseline serum S100␤ levels measured just after anesthetic induction were comparable in the 2 groups. Significantly higher levels of S100␤ were found postoperatively after 24 hours in group A (37°C) when compared with group B (33°C). The SjvO2 values were comparable at 30°C between the 2 groups but a significantly lower jugular venous oxygen saturation was observed in group A (37°C) as compared with group B (33°C) at the end of rewarming. The neurocognitive functions tested along with the preoperative and postoperative scores are summarized in Table 3. On intragroup comparison, the patients in group A (37°C) showed significant postoperative deterioration in 8 of the 11 functions tested, whereas the patients in group B (33°C) showed significant postoperative deterioration in only 5 of the 11 functions tested (Table 3). When intergroup comparisons were done, the preoperative scores except for attention and concentration were comparable between both the groups (Table 4).

Table 2. Outcome Parameters Variables

Group A (37°C) (n ⫽ 40)

Group B (33°C) (n ⫽ 40)

Need for pacing (no. of patients) 4 (10%) 9 (23%) Postoperative blood loss (mL) 270 (210-350) 350 (150-750) Postoperative blood products use (mL) 450 (150-1,100) 590 (110-1,600) Increased inotrope (no. of patients) 0 2 (5%) Increased dilators (no. of patients) 0 1 (2.5%) Temp on reaching ICU (°C) 36.5 ⫾ 0.5 32.5 ⫾ 0.5 Time for rewarming (h) 2.8 ⫾ 1.4 5.9 ⫾ 2.5 Time to extubation (h) 13 ⫾ 3.6 18 ⫾ 3.1 ICU stay (d) 1 1 Hospital stay (d) 6 6 Mortality 0 0 S100␤ (ng/L) Postinduction 62 (0-620) 66.5 (0-215) Postop (24 h) 60 (10-135) 32 (5-85) SjvO2 (%) 30°C 79.1 ⫾ 9 80.4 ⫾ 7.3 End of rewarming 72.2 ⫾ 11 78.1 ⫾ 7.3

p Value

0.12 0.08 0.06 0.12 0.31 0.001 0.001 0.001 — — — 0.33 0.001 0.46 0.006

NOTE. Continuous data expressed as mean ⫾ standard deviation and median (range); t tests/Mann-Whitney U tests were used for the comparison of continuous variables depending on normality, and a chi-square test was used for categoric variables; p ⬍ 0.05 statistically significant.

NEUROCOGNITIVE FUNCTION AND CABG SURGERY

17

Table 3. Neurocognitive Tests (PGI Memory Scale): Intragroup Comparison Group A (37°C)

Group B (33°C)

Function Tested

Preoperative Score

Postoperative Score

p Value

Preoperative Score

Postoperative Score

p Value

Remote memory Recent memory Mental balance Attention and concentration Delayed recall Immediate recall Retention for similar pairs Retention for dissimilar pairs Visual Retention Recognition Total

0.2 (0, 0.7) 0.3 (⫺3.3, 0.4) 0.6 (⫺1, 1) 0.3 (⫺1.7, 1.8) 1.4 (⫺2.3, 1.5) 1.9 (1.8, 2.5) 0.8 (⫺0.7, 1.1) ⫺0.9 (⫺3.2, 0.1) 0.7 (⫺2.1, 1.8) 0.8 (0.1, 1.8) 0.9 (⫺1, 1.4)

0.2 (⫺3.8, 0.7) 0.3 (⫺3.9, 0.4) ⫺0.4 (⫺14.4, 1) ⫺0.6 (⫺1.7, 1.8) 0 (⫺2.8, 1.4) 1.9 (1.4, 2.5) ⫺0.3 (⫺1.9, 1.1) ⫺1.8 (⫺3.3, ⫺0.1) 0.4 (⫺2.5, 2.2) 0.3 (⫺0.6, 1.5) ⫺0.4 (⫺1.9, 1.2)

0.16 0.11 0.001 0.001 0.001 0.3 0.001 0.001 0.001 0.001 0.001

0 (0, 0.7) 0.3 (0.3, 0.4) 0.5 (⫺0.7, 1) ⫺0.2 (⫺1.7, 1.5) 1.4 (⫺2.3, 1.6) 2.4 (0.8, 2.4) 0.7 (⫺0.7, 1.1) ⫺1.2 (⫺2.5, ⫺0.2) 0.7 (0.1, 1.5) 0.8 (0.5, 1.9) 0.6 (⫺0.9, 1.2)

0 (0, 0.7) 0.3 (⫺3.3, 0.4) 0 (⫺1.5, 1) ⫺0.7 (⫺1.7, 0.4) 1.4 (⫺2.3, 1.6) 2.4 (0.8, 2.4) 0.7 (0.7, 1.1) ⫺1.5 (⫺2.8, ⫺0.5) 0.5 (⫺0.4, 1.5) 0.8 (0.1, 1.9) 0.3 (⫺1.2, 0.9)

— 0.16 0.001 0.001 0.45 — 0.32 0.001 0.001 0.32 0.001

NOTE. All scores are the composite z scores ⫽ ([raw score ⫺ reference score]/standard deviation]. Continuous data expressed as median (range). A Wilcoxon signed-rank test was used for comparison of continuous variables; p ⬍ 0.05 is statistically significant. Lower scores indicate deterioration in neurocognitive function.

However, the postoperative neurocognitive scores were higher in 6 of 11 functions in group B (33°C) (Table 4). When the authors compared the difference in preoperative and postoperative neurocognitive scores between the 2 groups, higher differences in 8 of 11 scores in group A were obtained (37°C) (Table 5). There was no postoperative mortality. There were no differences in the duration of ICU stay and hospital stay. There were no cases of hemodynamic instability or re-exploration for bleeding. All the patients remained hemodynamically stable pre- and post-CPB and in the ICU. DISCUSSION

Core body cooling during CPB using the heat exchanger incorporated into the membrane oxygenator is used for systemic hypothermia, which helps in organ protection. This systemic hypothermia necessitates rewarming of the patient on completion of surgery. Because of concerns regarding impaired coagulation, rhythm disturbances, and reduced myocardial contractility, targeted intraoperative rewarming is commonly practiced at the end of CPB. This rewarming at the conclusion of

surgery on CPB may lead to intraoperative cerebral hyperthermia.8 Cerebral hyperthermia occurs during rewarming from hypothermic CPB as a result of high cerebral blood flow and the proximity of carotid origins to the aortic cannulation site. These high brain temperatures are not correctly reflected by the nasopharyngeal temperature and may correlate with neurologic morbidity.9 Normally, the brain temperature is 0.5° to 1°C more than the core temperature. Within the brain, the temperature of deep regions (eg, caudate) is 0.5° to 1°C more than that of the superficial regions (eg, neocortex). These temperature differentials within the brain may increase by several degrees during cooling and rewarming on CPB, by up to 5°C in some patients. Because a brain temperature difference of 3° to 4°C influences neurologic outcome after ischemia, it is hypothesized that patients with large temperature differences are at greater risk of neurologic impairment either from insufficient brain cooling or excessive brain rewarming.10 In a recent study,11-14 nasopharyngeal temperature monitoring was compared with continuous jugular venous bulb tem-

Table 4. Neurocognitive Tests (PGI Memory Scale): Intergroup Comparison Preoperative Scores

Postoperative Scores

Function Tested

Group A (37°C)

Group B (33°C)

p Value

Group A (37°C)

Group B (33°C)

p Value

Remote memory Recent memory Mental balance Attention & Concentration Delayed recall Immediate recall Retention for similar pairs Retention for dissimilar pairs Visual Retention Recognition Total

0.2 (0, 0.7) 0.3 (⫺3.3, 0.4) 0.6 (⫺1, 1) 0.3 (⫺1.7, 1.8) 1.4 (⫺2.3, 1.5) 1.9 (1.8, 2.5) 0.8 (⫺0.7, 1.1) ⫺0.9 (⫺3.2, 0.1) 0.7 (⫺2.1, 1.8) 0.8 (0.1, 1.8) 0.9 (⫺1, 1.4)

0 (0, 0.7) 0.3 (0.3, 0.4) 0.5 (⫺0.7, 1) ⫺0.2 (⫺1.7, 1.5) 1.4 (⫺2.3, 1.6) 2.4 (0.8, 2.4) 0.7 (⫺0.7, 1.1) ⫺1.2 (⫺2.5, ⫺0.2) 0.7 (0.1, 1.5) 0.8 (0.5, 1.9) 0.6 (⫺0.9, 1.2)

0.12 0.6 0.17 0.007 0.26 0.14 0.39 0.20 0.33 0.12 0.05

0.2 (⫺3.8, 0.7) 0.3 (⫺3.9, 0.4) ⫺0.4 (⫺14.4, 1) ⫺0.6 (⫺1.7, 1.8) 0 (⫺2.8, 1.4) 1.9 (1.4, 2.5) ⫺0.3 (⫺1.9, 1.1) ⫺1.8 (⫺3.3, ⫺0.1) 0.4 (⫺2.5, 2.2) 0.3 (⫺0.6, 1.5) ⫺0.4 (⫺1.9, 1.2)

0 (0, 0.7) 0.3 (⫺3.3, 0.4) 0 (⫺1.5, 1) ⫺0.7 (⫺1.7, 0.4) 1.4 (⫺2.3, 1.6) 2.4 (0.8, 2.4) 0.7 (0.7, 1.1) ⫺1.5 (⫺2.8, ⫺.5) 0.5 (⫺0.4, 1.5) 0.8 (0.1, 1.9) 0.3 (⫺1.2, 0.9)

0.36 0.08 0.017 0.31 0.001 0.08 0.001 0.02 0.19 0.001 0.001

NOTE. All scores are the composite z cores ⫽ ([raw score ⫺ reference score]/standard deviation). Continuous data expressed as median (range). A Mann-Whitney U test used for comparison of continuous variables; p ⬍ 0.05 is statistically significant. Lower scores indicate greater deterioration in neurocognitive function.

18

SAHU ET AL

Table 5. Neurocognitive Tests (PGI Memory Scale): Intergroup Comparison (Preoperative ⴚ Postoperative) Scores Function Tested

Remote memory Recent memory Mental balance Attention and concentration Delayed recall Immediate recall Retention for similar pairs Retention for dissimilar pairs Visual retention Recognition Total

Group A (37°C)

Group B (33°C)

p Value

0 (0, 2) 0 (0) 0 (⫺3.6, 4.2) 0 (0, 3.6) 0.9 (⫺1.5, 13.7) 0.5 (⫺0.4, 0.8)

0.15 0.06 0.001

0.9 (0, 1.6) 0.7 (0, 3.6) 0 (⫺0.01, 1) 0.6 (⫺1.5, 1.7)

0.5 (0, 1.2) 0 (⫺1.4, 1.4) 0 (0) 0 (⫺1.5, 0)

0.001 0.001 0.31 0.001

0.7 (⫺0.9, 1.6) 0.4 (⫺0.4, 1.4) 0.7 (0, 1) 1 (0, 2.1)

0.3 (0, 0.9) 0 (0, 0.5) 0 (0, 0.7) 0.3 (0, 0.6)

0.001 0.001 0.001 0.001

NOTE. All scores are the differences between preoperative and postoperative composite z scores. Higher differences in scores indicate greater neurocognitive dysfunction. Continuous data are expressed as median (range). A Mann-Whitney U test was used for comparison of continuous variables; p ⬍ 0.05 statistically significant.

perature monitoring. The difference between the 2 was most pronounced during rewarming when the jugular venous temperature was found to be 3° to 4°C higher than the corresponding nasopharyngeal temperature. The authors concluded that nasopharyngeal temperature monitoring underestimated the brain temperature during rewarming and speculated that as a result of this difference in the monitored and actual brain temperatures, the brain may be at increased risk of neurologic injury. Busto et al15 have shown that as little as 1.5°C of hyperthermia (to 39°C) during a period of focal ischemia dramatically increases the size of cerebral infarction in rats. Dietrich et al16 showed in rats that mildly increasing the intraischemic brain temperature by 2°C converted cell injury to frank infarction in susceptible regions, increasing the morbidity and mortality postoperatively. Reith et al17 examined the relationship between temperature and infarct size, stroke severity, and mortality. They showed a relative risk increase of 2.2 for each 1°C increase in body temperature. There are several mechanisms by which hyperthermia can adversely affect the brain. Hyperthermia accentuates release of excitotoxic neurotransmitters (glutamate),18 increases oxygenderived free radical production,19 increases blood-brain barrier permeability20 and intracellular acidosis, and delays neuronal metabolic recovery.21 Grigore et al22 studied 165 patients undergoing elective CABG surgery. All patients were cooled to 30°C on CPB. The patients were divided into 2 groups while rewarming after the completion of surgery. During rewarming, a 2°C difference between nasopharyngeal and CPB perfusate temperatures was maintained for patients in the slow rewarming group, whereas the subjects in the control group were conventionally rewarmed at a nasopharyngeal-CPB perfusate temperature gradient of 4° to 6°C. The patients who were rewarmed slowly achieved lower peak temperatures on CPB and had lesser degrees of neurocognitive impairment 6 weeks after surgery (assessed by Wechsler Adult Intelligence Scale, Randt Memory Scale, and

Trail-B test). The patients who were rewarmed conventionally had higher overall maximum postoperative temperatures. The authors proposed an improvement in cerebral blood flow and cerebral metabolic rate balance and a decreased incidence of cerebral hyperthermia to be the mechanism for this neuroprotection in the slow rewarming group. They also proposed that because diabetic patients have impaired cerebral autoregulation during CPB, they may especially benefit from the use of slow rewarming rates. In the present study, despite a larger number of diabetics in group B (33°C), lesser neurocognitive dysfunction occurred in that group. Because the authors used slow rewarming rates in both groups, the higher peak rewarming temperature in group A (37°C) may have been responsible for the greater neurocognitive dysfunction seen in that group rather than the rate of rewarming. Mora et al23 studied 138 patients undergoing CABG surgery and randomly assigned them into 2 groups. One group was cooled to 28°C, whereas the other was actively warmed to ⱖ35°C throughout CPB. They suggested that warming patients during CPB to maintain the systemic temperature at ⱖ35°C increases the risk of neurologic deficits after CABG surgery but did not substantially influence neurocognitive performance (assessed by Grooved Pegboard test, Wechsler Adult Intelligence Scale, and Wechsler Memory Scale). In a study by Nathan et al24 in 223 patients undergoing CABG surgery, all patients were cooled to 32°C on CPB and then randomly assigned to rewarming to nasopharyngeal temperatures of either 37°C (control) or 34°C (hypothermic), respectively. These temperatures were kept constant until separation from CPB. On arrival in the ICU, warming blankets were applied, and all patients reached nasopharyngeal temperatures of 36°C within 5 hours postoperatively. The neurocognitive assessment was done preoperatively and at 1 week and 3 months postoperatively using the Wechsler Adult Intelligence Scale, Trail A and B tests, and Buschke Selective Reminding Scoring tests. The incidence of neurocognitive deficits 1 week after surgery was 62% in the control group and 48% in the hypothermic group. At 3 months, the hypothermic group still performed better on 1 test. There was no difference in morbidity or mortality. The authors concluded that mild hypothermia during and after CPB leads to a significant reduction in the number of patients with neurocognitive deficits as well as magnitude of deterioration of mean scores. They also suggested that these beneficial effects of hypothermia were not associated with any increases in postoperative bleeding, morbidity, mortality, or length of ICU or hospital stay. They followed up the patients 5 years postoperatively. The analysis of patients who had persistent cognitive deficits (1 week postoperatively as well as 5 years later) was suggestive but not conclusive of a longterm beneficial effect of mild hypothermia.25 Boodhwani et al26 studied 267 patients undergoing nonurgent CABG surgery. These patients were either actively cooled to 34°C in 1 group or actively warmed to 37°C from the beginning of surgery using thermal pads connected to a water circulating thermal control system. The nasopharyngeal temperature was kept as close to either 34°C or 37°C throughout the intraoperative period until arrival to the ICU. In the ICU, forced-air warming blankets were used. The authors concluded that the incidence of postoperative neurocognitive deficits (as

NEUROCOGNITIVE FUNCTION AND CABG SURGERY

assessed by the Wechsler Adult Intelligence Scale, Rey Auditory Verbal Learning Test, and Grooved Pegboard Test) were similar between the hypothermic and normothermic groups. Also, the mild hypothermia sustained throughout the intraoperative period did not result in any major adverse events. In patients undergoing CABG surgery, sustained and mild intraoperative hypothermia (34°C) is safe but does not reduce the incidence of neurocognitive deficits. In the absence of cerebral hyperthermia and rewarming, there is no difference in outcome between sustained mild hypothermia and normothermia. The authors also suggested that rewarming (32°C-37°C during a period of 10-15 minutes) even in the absence of hyperthermia was likely responsible for worse neurocognitive outcome in the normothermic group in the study by Nathan et al.24 In the present study, the rewarming was done at a slow rate. The gradient between the water bath and the blood temperature was kept between 2° and 4°C. There was a difference of approximately 28 minutes between CPB time and aortic crossclamp release time in group A (37°C). This was the time taken to rewarm patients in this group from 30°C to 37°C. The peak nasopharyngeal temperatures at weaning from CPB were 37°C and 33°C, respectively, in groups A and B. There was no overshoot of temperature. The authors did not actively rewarm the patients in the ICU. All the patients were allowed to passively rewarm in the ICU. The patients in group B (33°C) had higher neurocognitive scores than patients in group A (37°C). This was associated with higher SjvO2 at the end of rewarming and lower serum S100␤ values 24 hours postoperatively in group B (rewarmed to 33°C) compared with group A (37°C). However, there was no increase in the need for inotropes or vasodilators or in the need for pacing or the amount of postoperative blood loss or blood product usage in group B (33°C), thereby negating the concern of hypothermia increasing the need for pacing, inotropes, or postoperative blood loss. Thus, the authors concluded that the slower rate of rewarming, lower peak temperatures at the time of weaning from bypass, and sustained mild hypothermia postoperatively can reduce neurocognitive dysfunction postoperatively. Plourde et al27 and Engelmann et al28 found no benefit of different degrees of hypothermia during CPB on neurocognitive outcome of CABG surgery. A common characteristic of both these studies is rewarming of all patients before separation from CPB, often to temperatures exceeding 37°C. Thus, the duration of exposure to a beneficial effect of hypothermia was brief, and these patients were still exposed to the possibly deleterious effects of cerebral hyperthermia during rewarming. In contrast, the authors avoided raising the nasopharyngeal temperature ⬎33°C during rewarming in group B (33°C). This period of hypothermia was extended into the postoperative period. This probably explained the better neurocognitive scores in group B (33°C) in the present study. Chen et al,29 investigating 11 patients, found that SjvO2 was directly related to rewarming speed and it inversely correlated with nasopharyngeal temperature change, and they concluded that the magnitude and speed of temperature change are major determinants of cerebral blood flow and cerebral metabolic rate balance during rewarming.

19

Von Knobelsdorff et al30,31 noticed that during cooling and stable hypothermia the ratio of SjvO2 to the middle cerebral artery mean blood flow velocity was comparable to that before CPB. During rewarming, despite an increase in the mean blood flow velocity by 65%, SjvO2 decreased by 25%, indicating a mismatch between cerebral blood flow and cerebral metabolic rate for oxygen. Nakajima et al4 and Van der Linden et al32 both noted that jugular venous desaturation was greater with longer rewarming rates. A decrease in SjvO2 with rewarming has been associated with poorer neurologic outcome. Croughwell et al33,34 reported an association between the magnitude of rewarming-induced jugular venous hemoglobin desaturation and the severity of postoperative neuropsychologic dysfunction. Croughwell et al,33 in their study on 133 patients undergoing CPB, showed that 23% of patients undergoing uneventful cardiac surgical procedures during extracorporeal circulation had significant jugular venous blood desaturation defined as SjvO2 ⱕ50% or PjvO2 ⱕ25mmHg during the rewarming phase of nonpulsatile CPB. If it is presumed that cerebral blood flow is relatively constant during CPB, then cerebral desaturation may reflect a rise in cerebral oxygen demand during rewarming. SjvO2 is considered to be a general index of adequacy of matching of cerebral oxygen supply with demand. The present study is consistent with this finding. The group A patients were rewarmed to higher peak temperatures (37°C). This was associated with a greater mismatch of cerebral oxygen supply with demand, thereby leading to a greater degree of fall in jugular venous oxygen saturation in patients of group A (37°C), which was associated with greater neurocognitive deterioration in group A (37°C) patients as compared with group B (33°C) patients. To date, the diagnosis of cerebral injury has relied on clinical neurologic examination, computed tomography scans, and magnetic resonance imaging. However, these methods may not be available for patients soon after cardiac surgery. These patients may be unconscious, hemodynamically unstable, sedated, or on a ventilator and therefore may be unable to cooperate in the clinical investigation. Hence, a biochemical marker to assist in the detection of cerebral injury is potentially useful. Normally, the S100␤ protein is not present in the circulation. It is released from the glial cells either in response to stress, CPB, or after neurologic injury. If released in response to stress, S100␤ levels decrease to 0 within 24 hours. S100␤ is eliminated through glomerular filtration and degradation in the proximal tubule. S100␤ has a biologic t1/2 of 2 hours. Because the t1/2 of S100␤ is short, the serum concentration must be maintained by persistent release.35 Persistence of S100␤ in the circulation beyond 24 hours is indicative of central nervous system injury.36 Serum S100␤ levels may reflect the extent of cerebral involvement and may be a valuable prognostic indicator.36 Fazio et al37 sampled systemic and pericardial cavity blood before, during, and after CPB from 5 patients undergoing CABG surgery. They found that an increase in S100␤ blood levels observed during CPB occurs because of the presence of contaminants, especially non-S100␤ proteins (eg, haptoglobulin I precursor, apolipoprotein A-I precursor, complement factor B precursor, and complement C3 precursor presumably released by extra central nervous system tissue). Late increases

20

SAHU ET AL

in S100␤ after cardiac surgery, long after the chest is closed and cardiotomy influence has dissipated (ⱖ24 hours), may have a potential diagnostic value for cerebral injury. The authors measured S100␤ levels 24 hours postoperatively in the patients to rule out this contamination from non-S100␤ proteins. In the present study, higher serum S100␤ levels in group A (37°C) postoperatively at 24 hours compared with group B (33°C) coincided with poorer neurocognitive scores postoperatively in group A patients who were rewarmed to 37°C. The time taken for extubation was longer in group B (33°C) as compared with group A (37°C) patients (p ⬍ 0.001). This can be explained by the greater time needed for the body to rewarm (peripheral body temperature ⱖ35°C) in the absence of active rewarming measures in the ICU before the patient could be extubated. However, the ICU stay and the hospital stay were the same in both groups. There was no increase in the need for inotropes or vasodilators or in the need for pacing or the amount of postoperative blood loss or blood product usage in group B (33°C), thereby negating the concern of hypothermia increasing the need for pacing, inotropes, or postoperative blood loss. There is a trend toward numerically greater blood loss and need for transfusion in group B (33°C). However, the difference was not found to be statistically or clinically significant. LIMITATIONS

This was a short-term study. If the patients were followed up over a longer time period, probably better information about the

long-term extent and the degree of neurocognitive damage could have been obtained. The patients with a high risk of intraoperative neurologic insults (eg, those with prior cerebrovascular disease, cardiac dysfunction, and recent myocardial infarction) were not included in this study; thus, the impact of this rewarming strategy could not be assessed in these high-risk groups. The sample size may have been inadequate to detect adverse effects of hypothermia (eg, postoperative blood loss, blood product transfusion, inotrope/dilator use, and need for pacing), and larger studies may be needed to do so. CONCLUSION

Neurocognitive impairment can affect 30% to 60% of patients undergoing CABG surgery with CPB. Considering that cerebral hyperthermia may be an important and preventable etiologic factor of cerebral injury during CPB, the prevention of hyperthermia should be of utmost importance. A simple strategy of weaning the patients from bypass at 33°C, slower rate of rewarming, and sustained mild hypothermia in the postoperative period can be helpful in decreasing the incidence and severity of cognitive dysfunction, thereby reducing the morbidity and improving the quality of life postoperatively in patients undergoing CABG surgery. ACKNOWLEDGMENT The authors wish to acknowledge the statistical analysis done by Dr Bhupinder Kumar, PhD, Department of Biostatistics, All India Institute of Medical Sciences, New Delhi, India.

REFERENCES 1. Grocott HP, Mackensen GB, Grigore AM, et al: Postoperative hyperthermia is associated with cognitive dysfunction after coronary artery bypass graft surgery. Stroke 33:537-541, 2002 2. Mclean RF, Wong MD, Naylor CD, et al: Cardiopulmonary bypass, temperature and central nervous system dysfunction. Circulation 90:II-250-II-55, 1994 3. Croughwell ND, Frasco P, Blumenthal JA, et al: Warming during cardiopulmonary bypass is associated with jugular bulb desaturation. Ann Thorac Surg 53:827-832, 1992 4. Nakajima T, Kuro M, Hayashi Y, et al: Clinical evaluation of cerebral oxygen balance during cardiopulmonary bypass: Online continuous monitoring of jugular venous oxygen saturation. Anesth Analg 74:630-635, 1992 5. Westaby S, Andrew JP, Blomquist S, et al: Serum s100␤ protein: A potential marker for cerebral events during CPB. Ann Thorac Surg 6:88-92, 1996 6. Fletcher JM, Levin HS, Satz P: Neuropsychological and intellectual assessment in children, in Kaplan HI, Sadock BI (eds): Comprehensive Textbook of Psychiatry (ed 5). Baltimore, MD, Williams and Wilkins, 1989, pp 513-525 7. Kohli A, Manreet K, Manju M, et al: Neuropsychological functioning in specific learning disorders—Reading, writing and mixed groups. J Indian Assoc Child Adolesc Ment Health 2:112-115, 2006 8. Clark JA, Bar-Yosef S, Anderson A, et al: Postoperative hyperthermia following off-pump versus on-pump coronary artery bypass surgery. J Cardiothorac Vasc Anesth 19:426-429, 2005 9. Cook DJ, Orszulak TA, Daly RC, et al: Cerebral hyperthermia during cardiopulmonary bypass in adults. J Thorac Cardiovasc Surg 111:268-269, 1996

10. Stone JG, Young WL, Smith CR, et al: Do standard monitoring sites reflect the true brain temperature when profound hypothermia is rapidly induced and reversed? Anesthesiology 82:344-351, 1995 11. Arrowsmith JE, Grocott HP, Reves JG, et al: Central nervous system complications of cardiac surgery. Br J Anaesth 84:378-393, 2000 12. Cheng WP, Marino MR, Romanogli A, et al: Temperature measurement during CPB. Anesthesiology 93:A384, 2000 13. Nussmeier NA, Li S, Strickler AG, et al: Temperature measurement during CPB. Anesth Analg 93:SCA35, 2002 14. Grocott HP, Newman MF, Croughwell NA, et al: Continuous jugular venous versus nasopharyngeal temperature monitoring during hypothermic CPB for cardiac surgery. J Clin Anesth 9:312-316, 1997 15. Busto R, Dietrich W, Globus MT, et al: The importance of brain temperature in cerebral ischemic injury. Stroke 20:1113-1114, 1989 16. Dietrich WD, Busto R, Valdes I, et al: Effect of normothermic vs mild hypothermic forebrain ischemia in rats. Stroke 21:1318-1325, 1990 17. Reith J, Jorgensen H, Pedersen P, et al: Body temperature in acute stroke: Relation to stroke severity, infarct size, mortality and outcome. Lancet 347:422-425, 1996 18. Sternau L, Globus MT, Dietrich W, et al: Ischemia-induced neurotransmitter release: effects of mild intraischemic hyperthermia, in Globus MT, Dietrich W (eds): The Role of Neurotransmitter in Brain Injury. New York, NY, Plenum Press, 1992, pp 33-38 19. Globus M, Busto R, Lin B, et al: Detection of free radical activity during transient global ischemia and recirculation: Effects of intraischemic brain temperature modulation. J Neurochem 65:12501256, 1995

NEUROCOGNITIVE FUNCTION AND CABG SURGERY

20. Dietrich W, Halley M, Valdes I, et al: Interrelationship between increased vascular permeability and acute neuronal damage following temperature-controlled brain ischemia in rats. Acta Neuropathol 81: 615-625, 1991 21. Chopp M, Welch KM, Tidwell CD, et al: Effect of mild hyperthermia on recovery of metabolic function after global cerebral ischemia in rats. Stroke 19:1521-1525, 1998 22. Grigore AM, Grocott HP, Mathew JP, et al: The rewarming rate and increased peak temperature alter neurocognitive outcome after cardiac surgery. Anesth Analg 94:4-10, 2002 23. Mora CT, Henson MB, Weintraub WS, et al: Effect of temperature management during cardiopulmonary bypass on neurologic and neuropsychological outcome in patients undergoing coronary revascularisation, J Thorac Cardiovasc Surg 112:514-522, 1996 24. Nathan JH, Wells GA, Munson JL, et al: Neuroprotective effects of mild hypothermia in patients undergoing coronary artery surgery with cardiopulmonary bypass. Circulation 104:I-85-I-91, 2001 (suppl I) 25. Nathan HJ, Rodriguez R, Wozny D, et al: Neuroprotective effect of mild hypothermia in patients undergoing coronary artery surgery with cardiopulmonary bypass: Five-year follow-up of a randomized trial. J Thorac Cardiovasc Surg 133:1206-1211, 2007 26. Boodhwani M, Rubens F, Wozny D, et al: Effects of sustained mild hypothermia on neurocognitive function after coronary artery bypass surgery: A randomized, double- blind study. J Thorac Cardiovasc Surg 134:1443-1452, 2007 27. Plourde G, Leduc AS, Morin JE, et al: Temperature during cardiopulmonary bypass for coronary artery operations does not influence postoperative cognitive function: A prospective, randomized trial. J Thorac Cardiovasc Surg 114:123-128, 1997 28. Engelman RM, Pleet AB, Rousou JA, et al: Influence of cardiopulmonary bypass perfusion temperature on neurologic and hemato-

21

logic function after coronary artery bypass grafting. Ann Thorac Surg 67:1547-1556, 1999 29. Chen CS, Len BK, Liu K: Detection of cerebral desaturation during CPB by cerebral oximetry. Acta Anaesthesiol Sin 34:173-178, 1996 30. Van Knobelsdorff G, Tonner PH, Havel F, et al: Prolonged rewarming after hypothermic cardiopulmonary bypass does not attenuate reduction of jugular bulb oxygen saturation. J Cardiothorac Vasc Anesth 11:689-693, 1997 31. Von Knobelsdorff G, Hanel F, Werner C, et al: Jugular bulb oxygen saturation and middle cerebral blood flow velocity during cardiopulmonary bypass. J Neurosurg Anesthesiol 9:128-133, 1997 32. Van der Linden J, Ekroth R, Linole C, et al: Is cerebral blood flow metabolic mismatch during rewarming a risk factor after profound hypothermic procedure in small children? Eur J Cardiothorac Surg 3:209-215, 1989 33. Croughwell N, Smith LR, Quill T, et al: The effect of temperature on cerebral metabolism and blood flow in adults during cardiopulmonary bypass. J Thorac Cardiovasc Surg 103:549-554, 1992 34. Croughwell ND, Newman MF, Blumenthal JA, et al: Jugular bulb saturation and cognitive dysfunction after cardiopulmonary bypass. Ann Thorac Surg 58:1702-1708, 1994 35. Jonsson H, Blomquist S, Westaby S, et al: Significance of serum s100␤ release after coronary artery bypass grafting. Ann Thorac Surg 65:1639-1644, 1998 36. Shaaban MA, Harmer M, Vaughan R: Serum s100 protein as a marker of cerebral damage during cardiac surgery. Br J Anaesth 85: 287-298, 2000 37. Fazio V, Bhudia SK, Maricha N, et al: Peripheral detection of s100␤ during cardiothoracic surgery: What are we really measuring? Ann Thorac Surg 78:46-53, 2004