|
|
||||||||
J Thorac Cardiovasc Surg 2007;133:1206-1211
© 2007 The American Association for Thoracic Surgery
Surgery for Acquired Cardiovascular Disease |
a Department of Anesthesiology, University of Ottawa, Ottawa, Canada
b Department of Surgery, University of Ottawa, Ottawa, Canada
c Department of Epidemiology, University of Ottawa, Ottawa, Canada.
Received for publication May 2, 2006; revisions received July 4, 2006; accepted for publication September 11, 2006. * Address for reprints: H. J. Nathan, MD, University of Ottawa Heart Institute, 40 Ruskin St, Ottawa, Ontario K1Y 4W7, Canada. (Email: hnathan{at}ottawaheart.ca).
| Abstract |
|---|
|
|
|---|
Methods: Patients were cooled to 32°C during aortic crossclamping and then randomized to rewarming to either 34°C or 37°C, with no further rewarming until arrival in intensive care unit. Cognitive function was measured preoperatively and 1 week and 5 years postoperatively with a battery of 11 psychometric tests interrogating verbal memory, attention, and psychomotor speed and dexterity.
Results: Patients who had greater cognitive decline 1 week after surgery showed poorer performance 5 years later. The magnitude of cognitive decline over 5 years was modest. The incidence of deficits defined as a 1 standard deviation [SD] decline in at least 1 of 3 factors was not different between temperature groups. Fewer patients in the hypothermic group had deficits that persisted over the 5 years, but this difference did not attain statistical significance (RR = 0.64, P = .16).
Conclusions: The effect of surgery on cognitive function observed early after surgery is an important predictor of cognitive performance 5 years later. Although there was evidence of a neuroprotective effect of mild hypothermia early after surgery in the original cohort, the results after 5 years were inconclusive. In general, the magnitude of cognitive changes over 5 years was modest. We believe that further trials investigating the efficacy of mild hypothermia in patients having cardiac surgery are warranted.
| Introduction |
|---|
|
|
|---|
| Materials and Methods |
|---|
|
|
|---|
The protocol is given in detail in the original publication.5
In brief, patients underwent CABG with CPB using membrane oxygenators and 43-µm arterial filters. All were cooled to 32°C (nasopharyngeal temperature) during application of the aortic crossclamp. When rewarming commenced, a sealed opaque envelope containing the treatment assignment was opened. The perfusionists then rewarmed the patients to a nasopharyngeal temperature of either 34°C or 37°C, taking care not to raise the temperature of the blood leaving the pump-oxygenator above 37.5°C. The patients temperature was held constant until separation from CPB. Upon arrival in the intensive care unit, warming blankets were applied and all patients reached 36°C within 5 hours postoperatively.
Measurement of Neurocognitive Function and Quality of Life
Learning efficiency and memory consolidation were evaluated with a verbal list–learning procedure (Buschke Selective Reminding administration and scoring). Alternate forms were used to reduce practice effects. Attention span was evaluated with the Wechsler Adult Intelligence Scale—Revised Digit Span. Psychomotor speed and dexterity were measured by Trails A and B, Grooved Pegboard, and the Symbol Digit Modalities Test (oral administration). From these tests we calculated the following measures: (1) Buschke Total Learning Free Recall; (2) Buschke consistent long-term retrieval; (3) Buschke long-term retrieval; (4) Buschke long-term storage; (5) Buschke Delayed Recall; (6) Digit Span Forward; (7) Digit Span Backward; (8) Trails A; (9) Trails B (maximum score = 300 seconds); (10) Grooved Pegboard (dominant hand, maximum score = 300 seconds); and (11) Symbol Digit Modalities Test. Patients were tested within the 4 weeks before surgery and then approximately 1 week, 3 months (not reported here), and 5 years after surgery. Both the psychometrist and the patient were unaware of the treatment assignment.
Statistical Methods
We analyzed the psychometric test results both as continuous outcomes and as dichotomous outcomes. For each of the 11 psychometric test scores, analysis of covariance was used to assess the effect of temperature assignment and surgery. We used the 5-year score as the dependent variable, the baseline score and the 1-week postoperative score (effect of surgery) as covariates, and the treatment group as a fixed independent variable. For all tests a higher score indicates improved performance except for the timed tests of speed and dexterity (Trails A, Trails B, and Pegtime), where increased score indicates poorer performance. To facilitate the categorical analysis, we combined the 11 psychometric tests into 3 cognitive domains using factor analysis with orthogonal rotation as described by Newman and colleagues.6
The scoring coefficients used to generate the domain scores for the 3 visits were determined from the baseline scores of the entire 223 patients who were enrolled in the original trial. This method reduced the 11 scores into 3 variables that are uncorrelated. The 3 factors accounted for 80% of the variance present in the test battery and, in composition, resembled the 3 domains presented in our original publication: verbal memory, psychomotor speed and dexterity, and attention. The scores were adjusted so that an increase in score always indicates better performance. A composite score, intended to represent overall cognitive performance, was formed by summing the three individual factors. A patient was deemed to have had a cognitive deficit if one or more factor scores decreased by at least 1 SD. The incidence of deficits was compared between groups by an uncorrected
2 test.
| Results |
|---|
|
|
|---|
The preoperative characteristics of the patients are presented in Table 1. There was a trend for patients with poor left ventricular function (P = .091) and New York Heart Association class III and IV (P = .058) to be lost to follow-up; however, there was no imbalance across temperature groups. Intraoperative and postoperative characteristics (Table 2) were similar in the two temperature groups except for the design variable, temperature.
|
|
|
Patients exhibiting a decrease of at least 1 SD in one or more of the three factors at the 1-week postoperative visit were deemed to have a cognitive deficit. In the original cohort of 194 patients who were tested both preoperatively and 1 week postoperatively, the incidence of deficits was 46 (46%) of 100 in the normothermic group and 30 (32%) of 94 in the hypothermic group (RR 0.69, P = .045), suggesting a neuroprotective effect of mild hypothermia very similar to that determined in our earlier report using different criteria.5
In the 131 patients who completed 5-year testing, the incidence of deficits 1 week postoperatively was 30 (45%) of 66 in the normothermic group and 22 (34%) of 65 in the hypothermic group (P = .175). After 5 years, the incidences were 29 (44%) of 66 and 27 (42%) of 65 in the normothermic and hypothermic groups, respectively (P = .781).
To estimate the magnitude of cognitive decline experienced by the patients over the 5-year period, we calculated the mean change scores for each test and tested whether the change was significantly different from 0. We did the same selecting only patients who had a decline of 1 SD or more 1 week postoperatively in any of the 3 factors to quantitate the change in patients who showed an adverse effect of surgery. Standardized scores (z-scores) are presented to provide a common scale for comparisons between the different tests (Table 4). Even in the group defined as having early postoperative deficits, no test decreased by as much as 1 SD over the 5 years.
|
Last, we wished to determine whether the effect of temperature assignment on the number of patients who had deficits 1 week postoperatively persisted over the 5 years: that is, we tested whether there was a difference between temperature groups in the proportion of patients demonstrating deficits at both 1 week and 5 years. For completeness, we present 3 different criteria for defining deficits. Using the criterion of 1 SD decrease in at least 1 factor, the incidence was 19 (29%) of 66 in the normothermic group and 12 (18%) of 65 in the hypothermic group; the RR associated with hypothermia was 0.64 (95% confidence interval [CI], 0.34–1.21, P = .164). Using the criterion of 1 SD decrease in factor 2 (the most sensitive to surgery, see Table 3), the incidence was 10 (15%) of 66 in the normothermic group and 4 (6%) of 65 in the hypothermic group; the RR was 0.41 (CI, 0.13–1.23, P = .096). Using the criterion of 1 SD decrease in the composite cognitive index, the incidence was 12 (18%) of 66 in the normothermic group and 5 (8%) of 65 in the hypothermic group; the RR was 0.42 (CI, 0.16–1.13, P = .074). More patients in the hypothermic group who did not have deficits 1 week postoperatively demonstrated new deficits when tested 5 years later. For this reason, there was a similar proportion of patients with deficits in each group after 5 years.
| Discussion |
|---|
|
|
|---|
We found a highly significant and consistent effect of early postoperative decline on long-term cognitive outcome. The fact that we could demonstrate this on 9 of 11 raw test scores makes the present results unambiguous and confirms similar findings by others.6,7
This suggests that an intervention that reduces early decline would be expected to have long-lasting beneficial effects.
We, along with Newman6
and Selnes8
and their associates, observed only small decrements in cognitive function 5 years after CABG. Even the cohort defined as having had a significant decline 1 week after surgery demonstrated only modest decline. Longitudinal studies of healthy elderly persons9,10
show no change in cognitive function when tested yearly over periods longer than 5 years, in part because of practice effects. It would be expected that our patients with advanced atherosclerotic disease and less practice effect after 5 years, even without surgery, would have more decline than healthy subjects. Selnes and coworkers11
have reported the longest follow-up study (3 years) comparing CABG patients and nonsurgical controls with atherosclerosis and found few differences between the groups. However, group means reported in the above studies combine the results of individuals whose scores improve or stay the same with a smaller number of individuals who show a decline that may be important. This effect is better described with a categorical approach,12
as has also been done in the present study.
More patients in the hypothermic group had new deficits late after surgery, when the treatment strategy could have no effect. A reduction in the proportion of patients who had deficits 1 week postoperatively that persisted over the 5 years may be a more sensitive measure of a long-lasting benefit of hypothermia. The observed reduction in incidence of persistent deficits with hypothermia, although not statistically significant, should, we believe, encourage further testing of this neuroprotective strategy for this population of patients.
Several studies have found no benefit of hypothermic CPB on early postoperative neuropsychologic outcome of CABG.13-16
Two studies were suggestive but not conclusive of a beneficial effect.17,18
A common characteristic of these studies is rewarming of all patients before separation from CPB, often heating blood to temperatures exceeding 37°C. Thus, the duration of exposure to a beneficial effect of hypothermia was brief and these patients were likely exposed to the deleterious effects of cerebral hyperthermia during rewarming.19
The importance of extending the period of hypothermia as long as possible after injury is well established in the laboratory20
and is a feature of positive clinical studies.4
Although our trial maintained a temperature gradient between groups longer than other trials, this gradient did diminish after separation from CPB. Also, the normothermic group became mildly hypothermic. It is possible that if the groups could have been kept at 34°C and 37°C longer, the treatment may have been more efficacious. We are currently testing this hypothesis. We21
have recently demonstrated that extending hypothermia into the early postoperative period is safe for these patients.
Loss to follow-up is a limitation of this study. However, retention was better than in similar studies,6-8
and the data collected on the patients lost to follow-up show no evidence of a selective process that may have biased the study sample. Both groups of patients experienced rewarming after a period of hypothermia during aortic crossclamping. The 34°C group was rewarmed less. Although care was taken not to exceed perfusion (oxygenator outlet) temperatures of greater than 37.5°C, it is possible that rewarming was a greater stress to the 37°C group.
Our test battery contained the core tests recommended by a consensus committee22
and examined a broad range of brain functions. The use of cutoffs to categorize patients as having deficits is recommended by some6,12
but avoided by others.7,11,23
Comparing group means is appropriate if one assumes that all patients in a group are a sample of the same population, that is, had a similar degree of injury during surgery. The categorical approach assumes that patients showing a decline greater than the cutoff have had a significant injury whereas others have not. Both approaches help to describe the data, and we have chosen to present both to the reader.
| Conclusions |
|---|
|
|
|---|
| See related editorial on page 1133.
|
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
K. Kellermann and B. Jungwirth Avoiding Stroke During Cardiac Surgery Seminars in Cardiothoracic and Vascular Anesthesia, June 1, 2010; 14(2): 95 - 101. [Abstract] [PDF] |
||||
![]() |
P. J. Tully, R. A. Baker, J. L. Knight, D. A. Turnbull, and H. R. Winefield Neuropsychological Function 5 Years after Cardiac Surgery and the Effect of Psychological Distress Arch Clin Neuropsychol, December 1, 2009; 24(8): 741 - 751. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. J. Cook CON: Temperature Regimens and Neuroprotection During Cardiopulmonary Bypass: Does Rewarming Rate Matter? Anesth. Analg., December 1, 2009; 109(6): 1733 - 1737. [Full Text] [PDF] |
||||
![]() |
H. P. Grocott PRO: Temperature Regimens and Neuroprotection During Cardiopulmonary Bypass: Does Rewarming Rate Matter? Anesth. Analg., December 1, 2009; 109(6): 1738 - 1740. [Full Text] [PDF] |
||||
![]() |
A. M. Grigore, C. F. Murray, H. Ramakrishna, and G. Djaiani A Core Review of Temperature Regimens and Neuroprotection During Cardiopulmonary Bypass: Does Rewarming Rate Matter? Anesth. Analg., December 1, 2009; 109(6): 1741 - 1751. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. A. Selnes, M. A. Grega, M. M. Bailey, L. D. Pham, S. L. Zeger, W. A. Baumgartner, and G. M. McKhann Do Management Strategies for Coronary Artery Disease Influence 6-Year Cognitive Outcomes? Ann. Thorac. Surg., August 1, 2009; 88(2): 445 - 454. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Bainbridge What Was Hot and What Was Not in 2007?: A Literature Review Seminars in Cardiothoracic and Vascular Anesthesia, June 1, 2009; 13(2): 78 - 80. [Abstract] [PDF] |
||||
![]() |
G. S. Murphy, E. A. Hessel II, and R. C. Groom Optimal Perfusion During Cardiopulmonary Bypass: An Evidence-Based Approach Anesth. Analg., May 1, 2009; 108(5): 1394 - 1417. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Sahu, S. Chauhan, U. Kiran, A. Bisoi, L. Ramakrishnan, and A. Nehra Neuropsychological function in children with cyanotic heart disease undergoing corrective cardiac surgery: effect of two different rewarming strategies Eur J Cardiothorac Surg, March 1, 2009; 35(3): 505 - 510. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Boodhwani, F. Rubens, D. Wozny, R. Rodriguez, and H. J. Nathan Effects of sustained mild hypothermia on neurocognitive function after coronary artery bypass surgery: A randomized, double-blind study. J. Thorac. Cardiovasc. Surg., December 1, 2007; 134(6): 1443 - 1452.e1. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. W. Hammon Jr and D. A. Stump Five-year cognitive outcomes: Surgical effects or natural progression of vascular disease J. Thorac. Cardiovasc. Surg., May 1, 2007; 133(5): 1133 - 1134. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |