| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1995;60:1716-1722
© 1995 The Society of Thoracic Surgeons
Departments of Anesthesiology, Neurology, Surgery, and The Irving Center for Clinical Research, Columbia University, New York, New York
Accepted for publication July 25, 1995.
 |
Abstract |
|---|
 TopFootnotes Abstract IntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Methods. Sixty-eight patients (55 for open chamber cardiac operations and 13 for noncardiac operations) were prospectively studied. Patients were evaluated preoperatively and postoperatively before hospital discharge using a complete neurologic examination and a battery of standard neuropsychometric tests, and at surgical follow-up with neuropsychometric tests only.
Results. Postoperative changes detected by neurologic examination consisted of the appearance of new primitive reflexes in all groups. No statistically significant differences in incidence were found. The neuropsychometric performance of group 1 patients was statistically different from that of patients in groups 2 and 3 only in the early follow-up period.
Conclusions. Elderly patients having open chamber cardiac operations exhibit significantly more cerebral dysfunction in the early postoperative period than those undergoing major noncardiac operations and younger patients after open chamber procedures. These changes do not persist into the late follow-up period.
|
Introduction |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
The term cerebral dysfunction refers to a functional measure of neurologic injury estimated by a neurologic examination and a battery of neuropsychometric tests administered before and after operation. The incidence of cerebral dysfunction after cardiac operation with CPB has been found to vary from 0% to 70% [5, 6, 8–13]. This large variation in incidence depends on differences in (1) surgical, anesthetic, and CPB techniques [14–18]; (2) standards and methods of evaluating neurologic outcome [6]; and (3) types of cardiac procedures included in the study populations [4].
Age is a consistent predictor of poor neurologic outcome after cardiac operation. It is thought to be important because the injury associated with cardiac operation becomes either more evident when superimposed on a baseline of previous injury such as gradual loss of cerebral function due to subclinical cerebral infarcts, or more severe as in patients with significant large vessel disease. The purpose of this study is to evaluate the significance of age as a predictor of cerebral injury in patients undergoing open chamber cardiac operations. It compares these patients with age-matched controls having major noncardiac operations and younger patients having cardiac operations.
|
Material and Methods |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Nine scores were generated from a battery of four neuropsychometric tests and the quantitative mini-mental status component of the neurologic examination [19, 20]: Halstead-Reiten trails parts A and B [19] (pp. 556–7), right and left index finger tapping test [19] (p. 529), grooved pegboard test [19] (p. 532) with dominant (dom) and nondominant (nond) hands, and long-term retrieval and consistency of long-term retrieval on the Buschke repetitive word list learning test [21]. These tests were selected for their ease of administration, reliability, patient acceptance, and capacity to demonstrate general neuropsychologic abilities including memory, attention, orientation, language, and motor functioning. Each of these tests is thought to examine a different cognitive domain.
Patients were divided into three groups according to age and type of operation. Group 1, patients aged 65 years or older, and group 2, patients aged less than 65 years, all had open-chamber cardiac operation. Group 3, the control group, consisted of elderly patients (age, 65 years or more) undergoing either repair of an abdominal aortic aneurysm or a noncardiac intrathoracic operation. Patients with evidence of preexisting neurologic or psychiatric illness were excluded. All clinical examinations were performed by the same board certified neurologist (E.J.H.). Patients were assessed using both neurologic and neuropsychometric tests 1 day preoperatively and 5 to 13 days after operation (early follow-up). Neuropsychometric tests were also performed 4 to 6 weeks after hospital discharge (Late follow-up). All examinations were performed at least 3 hours or more after any analgesic or sedative medication was administered.
Anesthesia
All patients were premedicated with oral diazepam (0.1 to 0.15 mg/kg) before operation. Patients about to undergo cardiac operation were also given scopolamine intramuscularly (0.3 mg). In addition to routine intraoperative monitors, all patients had intraarterial catheters for continuous recording of systemic blood pressure and intermittent sampling for blood gas analysis. Patients scheduled for cardiac operation or abdominal aortic aneurysm repair had a pulmonary artery catheter placed before anesthetic induction. In these individuals, general anesthesia and muscle relaxation were induced (fentanyl, 25 to 50 µg/kg; vecuronium, 0.2 mg/kg) and maintained using a high dose narcotic-relaxant technique (fentanyl, up to 100 µg/kg; metocurine as needed). Intravenous diazepam (0.2 to 0.8 mg/kg) was given for amnesia. In patients having intrathoracic operations induction was achieved with intravenous fentanyl (3 to 5 µg/kg), midazolam (0.03 to 0.07 mg/kg), and sodium thiopental (2 to 5 mg/kg), and maintenance with isoflurane in 100% oxygen. Intubation was facilitated with succinylcholine (1.5 mg/kg after precurarization). Individuals about to undergo either an abdominal aortic aneurysm repair or a noncardiac intrathoracic operation had epidural catheters placed before operation for postoperative pain management.
Operation
Almost all open chamber cardiac operations were performed by two senior surgeons (E.A.R., C.R.S.). The operation consisted of replacement of a mitral valve (15 patients), an aortic valve (21 patients), repair of atrial or ventricular septal defect (4 patients), or a combination of valve replacement or coronary artery bypass grafts (16 patients). The group of 13 patients undergoing major noncardiac operations (group 3) consisted of 10 patients having abdominal aortic aneurysm repair performed by one vascular surgeon (G.J.T.) and 3 patients having noncardiac intrathoracic operation performed by one thoracic surgeon (M.G.).
Cardiopulmonary Bypass
The bypass system used during open chamber cardiac operation consisted of a centrifugal arterial pump (Biomedicus, Minneapolis, MN), a hollow-fiber membrane oxygenator (Baxter Univox, Irvine, CA) and a 40-µm arterial line filter (Pall, Glen Cove, NY). During CPB, nonpulsatile flows of 2.4 L • min-1 • m-2 and alpha-stat management of pH during hypothermia to 28°C were used. In some cases, vasoactive medications such as phenylephrine or sodium nitroprusside were administered during CPB to maintain mean arterial pressure between 55 and 70 mm Hg. The coronary arteries were perfused with anterograde and retrograde cold blood cardioplegic solution consisting of blood to crystalloid in a ratio of 4:1. Air was evacuated using a variety of mechanical techniques-Trendelenberg positioning, ventricular and ascending aortic venting; however, carbon dioxide was not used to flood the operative field. In many cases the adequacy of these procedures was assessed using transesophageal echocardiography.
Statistical Measures
The groups were analyzed by paired Student's t test for significant differences in education, sex, height, weight, and duration of operation. In the cardiac operation groups, duration of CPB and duration of aortic cross-clamping were similarly compared.
Any new neurologic finding was considered of significance. The incidence of new neurologic findings was compared among groups using the 
2 test. Using ``event-rate'' comparisons, each individual's overall neuropsychometric performance was assessed by comparing preoperative and postoperative scores for each test. For an individual test to be considered abnormal there had to be at least a 25% decrease in postoperative performance compared with preoperative performance. The incidence of abnormal neuropsychometric tests (zero, only one, only two, only three, four or more) was analyzed by both 2 and Mann-Whitney rank sum tests. In addition, means and standard deviations for all neuropsychometric tests were calculated for each group. Repeated measures analysis of variance was applied to analyze all of these values.
|
Results |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
). The time to extubation was significantly longer in group 1 (Table 1). A higher percentage (21%) of the elderly patients (group 1) had coronary artery bypass grafting associated with valve replacement compared to the younger patients (4%) (group 2). The control group (group 3) consisted of 13 patients and was similar to the elderly patients having cardiac operation (group 1) in terms of age and education, and the incidence of diabetes mellitus (8%) and a history of previous cardiac operation (15%). However, these groups differed in terms of the height, weight, duration of operation, and time to extubation, and the incidence of hypertension (15%) (Table 1). Only 4 patients died, 2 each from groups 1 and 2. All deaths were of nonneurologic causes.
|
The follow-up examinations were performed at statistically comparable postoperative days for groups 1 and 2, and differed only between groups 1 and 3 in that the elderly control patients had earlier early follow-up examinations (8.4 versus 6 days, p < 0.01). Subtle deficits were detected by neurologic examination in 33.3% (11 of 33) of older patients (group 1), 9.1% (2 of 22) of younger patients (group 2), and 23.1% (3 of 13) of older control patients (group 3). The differences among the three groups were not statistically significant. A left hemiparesis developed in 1 patient each in groups 1 and 2 (6.1% and 4.5%, respectively); none in group 3. The rest of the neurologic findings consisted of the presence of grasp, snout, or palmomental reflexes that had not existed preoperatively (Table 2).
|
2 test p < 0.001) (Table 3). In the early follow-up period all group 1 patients had one or more neuropsychometric test abnormalities, but 91% declined on two, three, or four or more tests; whereas 64% of group 2 patients had one or more abnormal neuropsychometric tests, and 73% of these were abnormal on only one or two tests or showed no abnormalities (see Table 3). Similarly, the distribution of test abnormalities for groups 1 and 3 were significantly different at early follow-up (Mann-Whitney rank sum test p < 0.001; 2 test p = 0.008). When the three groups were compared at late follow-up, there were no statistically significant differences among them (see Table 3). There was no statistical correlation between patients with abnormal findings on the neurologic examination and neuropsychometric performance.
|
). This improvement occurred most dramatically for group 1, but was also noted in groups 2 and 3 (Fig 1). The group mean scores also reflect this improvement; no scores were statistically worse at late follow-up compared with preoperatively (Table 4). In fact, group 1 patients improved on one test beyond baseline (repetitive tapping: nond), and group 2 patients on three tests (trails A, trails B, and repetitive tapping: dom) (Table 4).
|
|
). The preoperative mean scores for all tests in groups 1 and 3 were statistically comparable (Table 4). At early follow-up, multivariate tests revealed significant overall differences in response to operation between groups 1 and 2 on the trails part A (F(2,27) = 4.26, p < 0.05) and part B (F(2,27) = 3.36, p < 0.05), and both dominant hand (F(2,26) = 3.97, p < 0.05) and nondominant hand (F(2,25) = 7.72, p < 0.005) for the grooved pegboard tests. Although group 2 scores improved from preoperative to late postoperative measurements on trails parts A and B, group 1 scores declined during the early postoperative phase and recovered to baseline by the late postoperative measurement (F(1,28) = 6.23, p < 0.05). On the grooved pegboard for both dominant and nondominant hands, younger patients (group 2) maintained preoperative performance levels after operation, whereas older patient performance dramatically declined during the early postoperative time and improved relative to baseline by the late postoperative test (F(1,27) = 5.5, p < 0.05 dominant hand; F(1,26) = 16.02, p < 0.0005 nondominant hand).
Comparing groups 1 and 3, the differences did not approach statistical significance for any of the tests except for the grooved pegboard with the dominant hand, where control patients (group 3) did not return to preoperative performance levels after operation (late follow-up), whereas elderly patient performance did (F(1,21) = 4.2, p = 0.054 dominant hand) (Table 4).
|
Comment |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
The results from a battery of neuropsychometric tests can be presented in a number of ways. First, ''group-rate'' analysis compares group means and standard deviations of performance. Using repeated measures analysis of variance, it is clear that elderly patients (group 1) performed more poorly at late follow-up compared with younger patients (group 2) in trails B, grooved pegboard for dominant and nondominant hands. In part these differences arose from the observation that group 2 patients improved their performance with subsequent examination (``learning'') and group 1 patients did not. Second, ''event-rate'' analysis compares preoperative and postoperative performance for an individual patient. Differences in performance can be quantitated; however, ``abnormality'' must be defined. Shaw and colleagues [5, 10, 22] defined a decrease in performance of one or more standard deviations as abnormal, and severity as the number of abnormal tests. Using this definition, patients who perform poorly preoperatively are unlikely to decrease their performance sufficiently to be considered abnormal, because they would have to be just about unable to perform any test postoperatively. To avoid this problem of a ``floor effect,'' we defined abnormality as a 25% decline in postoperative performance. In general this definition requires a decrease in performance between one and two standard deviations for almost all of the tests. Next, we determined a distribution of the number of abnormal tests versus the number of patients. Comparing these distributions for the three groups was less arbitrary than classifying severity of cerebral dysfunction as a function of the number of abnormal tests.
One might expect that an abnormal neurologic examination would predict poor neuropsychometric performance. However, there was no statistical correlation between the neurologic outcome and the performance measured by a battery of neuropsychometric tests. These two measures of neurologic function are complementary; the neurologic examination evaluates cerebral areas associated with primary motor, sensory, cerebellar, and language functions, whereas the neuropsychometric tests assess areas associated with subtle higher cortical functions.
Effect of Cardiac Operation on Cerebral Function in Elderly Patients
To answer this question we had to study first the effect of cardiac operation on two different age groups (young versus elderly), and then the effect of anesthesia on the same age group having different types of operation (cardiac versus noncardiac). The two cardiac operation groups were similar in the duration of operation, bypass, or aortic cross-clamp time, but differed by time to extubation (see Table 1). The longer time to extubation in group 1 may reflect the more significant physiologic impact of operation on the elderly patient. The two groups were statistically similar in the percentage of patients having new neurologic findings (see Table 2), and in the incidence of strokes compared to other studies [2–6, 23]. However, by event-rate analysis, these two groups had distinctly different distributions at early but not late follow-up in terms of the percentage of patients having abnormal neuropsychometric tests (see Table 3).
Although it is difficult to find a surgical control group that receives an identical anesthetic to patients undergoing cardiac operation, patients having repair of an abdominal aortic aneurysm have major vascular disease and undergo a major noncardiac operation that requires admission to the intensive care unit, and patients undergoing noncardiac intrathoracic operation are operated on the same anatomic body cavity. Our control group (group 3), although similar in age and education to group 1, clearly differed in surgical parameters, namely the duration of operation and time to extubation (see Table 1). The groups also differed in terms of the height and weight, for no apparent reason. Although we consider these differences to be irrelevant for cerebral dysfunction, we do not have any objective method to address this issue. The difference in weight may be due, in part, to the chronic nature of the cardiac disease in patients undergoing cardiac operation.
The differences in cardiac surgical procedures probably do not account for the differences in test performance between groups 1 and 2; elderly patients (group 1) had more coronary artery bypass grafting procedures than younger patients (group 2), 21% versus 4%, respectively. However, if all patients who had coronary bypass grafting are excluded, groups 1 and 2, and groups 1 and 3 still differ from each other with p = 0.001 and p < 0.001 using the Mann-Whitney rank sum test.
Significance of Neurologic Findings
Surprisingly enough, 23.1% of the controls (group 3) had new neurologic findings. This was statistically similar to elderly patients having cardiac operation (group 1) (see Table 2). Our results differ from those of Shaw and colleagues [5], who reported that new neurologic findings developed in 18% of control patients having major operation for peripheral vascular disease. However, their findings consisted of peripheral nerve injury, not the presence of primitive reflexes, as did ours. The mean age of patients in their control group was 57.4 years, compared with 74.2 years in our series. This age difference probably accounts for the difference in results in the two studies.
Primitive reflexes-grasp, palmomental, snout, and rooting-are normally present in young children and in a percentage of elderly patients (25% to 33%), and are suppressed with maturation of higher cortical areas, primarily frontal cortex. These areas are commonly referred to as silent but undoubtedly serve functions more subtle than the tests used to evaluate them. Therefore, the presence of these reflexes after cardiac operation indicates loss of function from these areas; for example, diffuse frontal lobe injury either from multiple small emboli or loss of cerebral perfusion would permit reexpression of these primitive reflexes.
Age as a Predictor of Cerebral Dysfunction
Elderly patients might more easily manifest a dramatic decline in performance because acute cerebral injury may be superimposed on a decreased functional baseline [19] (pp. 215–220). These patients also sustain brain injury more often because there is more particulate material in the form of atherosclerotic debris released with trauma to large vessels, which occurs during the course of cardiac operation.
Improvement in Cerebral Functioning From Early to Late Follow-up
All three groups improved their performance on neuropsychometric tests between early and late follow-up (Fig 1). Using group rate analysis, patients in groups 1, 2, and 3 approached their level of preoperative performance at late follow-up, some even improved (Table 4). One reason for this improvement may be that in fact no neurons have been destroyed, but rather that there has been a reversible impairment in neuronal function. Another explanation may be that although there may have been neuronal loss, the brain effectively compensated for its lost function through the same mechanisms that produce recovery after a stroke.
Conclusion
Elderly patients undergoing open chamber cardiac operation have significantly more cerebral dysfunction in the early postoperative period than elderly patients undergoing major noncardiac operation and younger patients undergoing open chamber procedures. These changes do not persist into the late follow-up period.
|
Acknowledgments |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This work was supported in part by grants to Dr Heyer from the CPMC Office of Clinical Trials, New York; and to Mr McMahon from the National Institutes of Health, Division of Research Resources, General Clinical Research Centers Program, 5 M01 RR00645.
|
Footnotes |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopFootnotesAbstractIntroductionMaterial and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. S. Savino and A. T. Cheung Cardiac Anesthesia Card. Surg. Adult, January 1, 2008; 3(2008): 281 - 314. [Full Text]
|
|
|
|
 |
|
 S. K. Bhudia, D. M. Cosgrove, R. I. Naugle, J. Rajeswaran, B.-K. Lam, E. Walton, J. Petrich, R. C. Palumbo, A. M. Gillinov, C. Apperson-Hansen, et al. Magnesium as a neuroprotectant in cardiac surgery: A randomized clinical trial J. Thorac. Cardiovasc. Surg., April 1, 2006; 131(4): 853 - 861. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Savino, T. F. Floyd, and A. T. Cheung Cardiac Anesthesia Card. Surg. Adult, January 1, 2003; 2(2003): 249 - 281. [Full Text]
|
|
|
|
|
|
M. A. Borger, C. M. Peniston, R. D. Weisel, M. Vasiliou, R. E. A. Green, and C. M. Feindel Neuropsychologic impairment after coronary bypass surgery: Effect of gaseous microemboli during perfusionist interventions J. Thorac. Cardiovasc. Surg., April 1, 2001; 121(4): 743 - 749. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Ali, M. Harmer, and R. Vaughan Serum S100 protein as a marker of cerebral damage during cardiac surgery Br. J. Anaesth., August 1, 2000; 85(2): 287 - 298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. P. Grichnik, A. J.J. Ijsselmuiden, T. A. D’Amico, D. H. Harpole Jr, W. D. White, J. A. Blumenthal, and M. F. Newman Cognitive decline after major noncardiac operations: a preliminary prospective study Ann. Thorac. Surg., November 1, 1999; 68(5): 1786 - 1791. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Butterworth Ann. Thorac. Surg., April 1, 1999; 67(4): 1124 - 1124. [Full Text] [PDF]
|
|
|
|
 |
|
 D. P.B. Janssen, L. Noyez, J. A.M. van Druten, S. H. Skotnicki, and L. K. Lacquet Predictors of neurological morbidity after coronary artery bypass surgery Eur. J. Cardiothorac. Surg., February 1, 1999; 15(2): 166 - 172. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Miyano, M. Inagaki, N. Hashimoto, T. Shishido, T. Kawada, Y. Miyake, and K. Sunagawa Regional cerebral blood flow during rewarming of cardiopulmonary bypass correlates with posthypothermic regional glucose use J. Thorac. Cardiovasc. Surg., September 1, 1998; 116(3): 503 - 507. [Abstract] [Full Text]
|
|
|
|
|
|
H. P. Grocott, N. D. Croughwell, D. W. Amory, W. D. White, J. L. Kirchner, and M. F. Newman Cerebral Emboli and Serum S100{beta} During Cardiac Operations Ann. Thorac. Surg., June 1, 1998; 65(6): 1645 - 1649. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. J. Heyer, D. C. Adams, E. Delphin, D. J. McMahon, S. D. Steneck, M. C. Oz, R. E. Michler, and E. A. Rose CEREBRAL DYSFUNCTION AFTER CORONARY ARTERY BYPASS GRAFTING DONE WITH MILD OR MODERATE HYPOTHERMIA J. Thorac. Cardiovasc. Surg., August 1, 1997; 114(2): 270 - 277. [Abstract] [Full Text]
|
|
|
|
|
|
D. P. Taggart, J. W. Mazel, K. Bhattacharya, N. Meston, S. J. Standing, J. D. S. Kay, R. Pillai, P. Johnssson, and S. Westaby Comparison of Serum S-100{beta} Levels During CABG and Intracardiac Operations Ann. Thorac. Surg., February 1, 1997; 63(2): 492 - 496. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
