|
|
||||||||
Eur J Cardiothorac Surg 2001;19:594-600
© 2001 Elsevier Science NL
a Department of Anesthesiology, Mount Sinai School of Medicine, New York, NY 10029, USA
b Department of Cardiothoracic Surgery, Mount Sinai School of Medicine, New York, NY 10029, USA
c Department of Biomathematical Sciences, Mount Sinai School of Medicine, New York, NY 10029, USA
Received 2 October 2000; received in revised form 20 February 2001; accepted 22 February 2001.
Corresponding author. Department of Anesthesiology, Box 1010, Mount Sinai School of Medicine, One Gustave L. Levy Place, New York, NY 10029-6574, USA. Tel.: +1-212-241-7467; fax: +1-212-241-1847
e-mail: david.reich{at}mssm.edu
| Abstract |
|---|
|
|
|---|
2 tests, MantelHaenszel tests and multiple logistic regression. P<0.05 was considered significant. Results: Memory dysfunction and NNO had strong associations with RCP. This effect remained significant when controlling separately for age and cerebral ischemia time. Conclusions: The effects of RCP are difficult to distinguish from those of age and prolonged cerebral ischemia time, because complex thoracic aortic repairs are associated with advanced age, prolonged cerebral ischemia and use of RCP. Despite this limitation, these preliminary data indicated that RCP had no beneficial effect (and most likely a negative effect) upon cognitive outcome.
Key Words: Aortic aneurysm, Thoracic Retrograde cerebral perfusion Thoracic surgery Neuropsychology Outcome study Hypothermic circulatory arrest
| 1. Introduction |
|---|
|
|
|---|
Neuropsychological testing is a sensitive method of measuring subtle neurological injury following surgical interventions, but has not been used to evaluate the effects of RCP. Our group previously reported that HCA >25 min in duration was associated with decreases in memory and fine motor function [4]. In the current investigation, we examined a cohort of patients undergoing surgery requiring HCA with or without RCP in order to determine whether RCP is a predictor of neuropsychological outcome.
| 2. Materials and methods |
|---|
|
|
|---|
The surgical technique for induction of profound hypothermia and use of circulatory arrest were constant throughout the duration of the study, and have been described in detail previously [5]. Briefly, central cooling on cardiopulmonary bypass was carried out using alpha-stat blood gas management to produce profound total body hypothermia to a core temperature of 1215°C, as measured in the esophagus. A minimum duration of 30 min is usually required for cooling thorough enough to prevent upward drift of body temperature during prolonged HCA. We used jugular bulb oxyhemoglobin saturation as an indicator of cerebral cooling and concomitant cerebral metabolic suppression. Jugular bulb oxyhemoglobin saturation was monitored every 5 min during the cooling phase prior to HCA, and cooling was continued until saturation exceeded 95% in most cases, or had plateaued above 90%. During HCA, the head was packed in ice to prevent warming of the central nervous system.
RCP was carried out by perfusion of the superior vena caval cannula with arterial blood via an arterial venous limb inserted into the perfusion circuit. Briefly, the venous cannula was perfused with oxygenated blood to a jugular bulb pressure of 1520 mmHg. No specific flow range limits were imposed. In certain patients, the period of RCP was brief to wash atheroembolic debris from the brachiocephalic vessels. In the remainder, RCP was administered continuously throughout the period during which anterograde flow was interrupted.
Upon reinstitution of anterograde cerebral perfusion and completion of the aortic repair, gradual warming was carried out by means of cardiopulmonary bypass, limiting the gradient between blood and body temperature to less than 10°C, with a maximum blood temperature of 37°C. A warming blanket was also utilized. Central warming was usually discontinued at an esophageal temperature of 3537°C, and a rectal or bladder temperature of 3035°C.
All patients were given 30 mg/kg of methylprednisolone prior to HCA. Glucocorticoids were continued in tapering doses for 48 h in those patients with an interval of HCA exceeding 30 min. Recovery occurred in the intensive care unit, and patients were evaluated for gross neurological deficits upon emergence from anesthesia and throughout the intensive care unit stay.
Demographic and perioperative data collected included: age, gender, smoking history, history of hypertension, previous neurological history, aortic atherosclerosis at operation, cardiopulmonary bypass time, minimum esophageal temperature, minimum bladder temperature, myocardial ischemia time, and cerebral ischemia time (defined as the cumulative length of interruption of anterograde cerebral perfusion). Postoperatively, the length of stay in the hospital, the presence of focal neurological deficits, and reasons why postoperative neuropsychological testing could not be completed were recorded.
2.1. Neuropsychological evaluation
All patients underwent neuropsychological evaluations by a psychologist trained in these techniques who was blinded to the clinical aspects of the surgical procedure. In order to assure that patients were not globally impaired to the extent that neuropsychological testing would be invalid, a brief orientation screen was administered (Orientation subtest of the Wechsler Memory Scale Revised, The Psychological Corporation, San Antonio, TX). No patients were excluded on this basis. The neuropsychological battery consisted of eight tests, with the data consolidated into five domains:
2.2. Data analysis
Data from all 94 patients were used to establish baseline values for each test and to assess postoperative neurologic deficits. Only those patients who also completed at least one postoperative test were included in the statistical analyses of neuropsychological outcomes.
Demographic and perioperative data were compared between the patients with and without postoperative neuropsychological testing using Wilcoxon two-sample tests for continuous data and
2 tests for binary data. The rates of postoperative neurological deficits for RCP and non-RCP patients were compared using Fisher's Exact test.
The initial step in the neuropsychological data analysis was the consolidation of the raw test data into composite scores for each cognitive domain for each patient at each testing interval. Composite scores were computed by standardizing each neuropsychological test measure in relation to the corresponding mean and standard deviation of the scores of the baseline sample. The total number of patients comprising the baseline sample was 94, but the number completing each individual test varied slightly because some patients did not complete all tests (range 8294). For each patient in the current study, the standardized score (Z-score) for each test was calculated by subtracting the baseline sample mean from the individual raw test score, and dividing this difference by the baseline sample standard deviation. Composite scores for each patient were defined as the simple numerical average Z-score for all tests (for which data were available) within each cognitive domain. For example, a patient Z-score of +1.0 represents a test score that was one standard deviation above average compared with all patients preoperatively. If a domain consisted of two tests and a patient had a Z-score of 1.0 in one test and 0.5 in the other, then the domain composite score was +0.75 for that patient at that time.
The neuropsychological outcomes were analyzed separately for each domain, first as continuous variables in terms of their composite Z-scores, and then as binary variables, which were defined as positive for the occurrence of neuropsychological deficit if the postoperative Z-score was less than the corresponding preoperative Z-score. In addition, patients with data on three or more domains were classified as experiencing an overall negative neuropsychological outcome (NNO) if they demonstrated a postoperative decrease in composite Z-scores in two or more domains.
The univariate relationship of RCP to the occurrence of deficits in each of the neuropsychological domains and to NNO was analyzed by Wilcoxon two-sample tests for continuous outcomes and
2 tests for binary data. The univariate relationships of age and cerebral ischemia time to deficits in each of the neuropsychological domains and to NNO were analyzed by
2 tests of linear trend, where ages and cerebral ischemia times were each grouped by quartiles.
Since RCP was used more often in older patients and in patients with longer cerebral ischemia times we controlled for age and for time in the tests of association between RCP and the occurrence of neuropsychological deficits. This was done by stratifying the data by quartiles of age and then by quartiles of cerebral ischemia time, and using MantelHaenszel tests to compare the risks of neuropsychological deficit in each domain and for NNO within each quartile between the patients who received RCP and those who did not. Additionally, we used multiple logistic regression analyses to check the magnitude and direction of the findings of the MantelHaenszel tests, controlling jointly for age and for cerebral ischemia. These were conducted using an exact analysis (LogXact, Cytel Corp., Cambridge MA).
| 3. Results |
|---|
|
|
|---|
Demographic and perioperative data comparing patients with postoperative neuropsychological data with patients lost to follow-up are summarized in Table 1. Patients lost to follow-up had higher rates of hypertension (P=0.042), aortic atherosclerosis (P=0.035), and longer cerebral ischemia times (P=0.004). It should also be noted that there were non-significant trends for these patients to be older and have more tobacco use. Overall, the patients who had complete neuropsychological data seem to be a slightly healthier subset.
|
The association of neuropsychological outcome and RCP was screened initially in univariate analyses of the continuous neuropsychological data (composite Z-scores). Postoperative decreases were significantly associated with RCP for the domains attention (P=0.0086), memory (P=0.01), and processing speed (P=0.01), but not with fine motor dysfunction (P=0.165) or executive function (P=0.23). These results were mirrored in univariate analyses using binary variables (any postoperative decrease in composite Z-scores). RCP remained significantly associated with deficits in attention (P=0.011), memory (P=0.008), and processing speed (P=0.049), but not with fine motor dysfunction (P=0.091) or executive function (P=0.70). RCP was also a significant univariate predictor of NNO (P=0.002). Given the concordance between the results of univariate analyses for continuous and binary outcome data, the remainder of the analyses address binary outcomes only.
RCP was more likely to be employed in older patients and in procedures with long cerebral ischemia times (Tables 2 and 3). Therefore, the univariate analyses described above may, in part, reflect an association between patient age as well as cerebral ischemia time with neuropsychological dysfunction following HCA, which we noted in previous work [4]. This led us to test the influence of these variables on neuropsychological outcome regardless of RCP in these data. Age was a significant univariate predictor of the occurrence of deficits in memory (P=0.002), fine motor function (P=0.015), and NNO (P=0.005). Cerebral ischemia time was also a significant predictor of NNO (P=0.022), but was a weaker predictor of specific deficits in memory (P=0.06) and fine motor function (P=0.121).
|
|
|
|
| 4. Discussion |
|---|
|
|
|---|
The previous literature regarding RCP does not address neuropsychological outcomes, which are the most sensitive measures of brain function. The existing data addressing gross neurological outcome in humans and behavioral outcomes in animal models are contradictory. In order to illustrate the controversial nature of these data, the following paragraphs review this literature.
4.1. Clinical outcome in thoracic aortic surgery
In some case series, RCP duration was not found to be a predictor of death or neurological morbidity [811], whereas in others it was a predictor [1214]. Clinical outcome studies comparing RCP versus HCA have also yielded mixed results. RCP has been found to be associated with neurological morbidity rates that are either similar to those associated with HCA [15,16] or lower [17,18], especially in older patients [19]. In studies which included selective (anterograde) cerebral perfusion patients, RCP patients had similar outcome in two studies [20,21], and worse outcome in one [22].
4.2. Neurological outcome in laboratory investigations
In a study by Juvonen et al. [1], embolization resulted in poor behavioral recovery with or without RCP, despite the effective washout of embolized microspheres. Even in the absence of embolization, mild and transient behavioral impairment occurred following RCP with inferior vena caval occlusion. In another study, Juvonen et al. [23] reported that behavioral outcome was best in RCP without inferior vena caval occlusion, compared to HCA or RCP with inferior vena caval occlusion. In the report by Midulla et al. [24], RCP and HCA with ice-packing of the head had similar outcomes that were significantly improved compared with HCA without ice-packing of the head. Safi et al. [19] also reported that RCP was better than HCA, but did not study HCA with ice-packing of the head. Yerlioglu et al. [25] reported that RCP and anterograde perfusion resulted in complete behavioral recovery, but that embolization resulted in poorer recovery, especially with RCP of >40 mmHg.
4.3. Limitations of the current study
The decision to use RCP in the clinical cohort that served as our data set was made at the surgeon's discretion. There was, therefore, non-random application of RCP, and variability in the flows, duration, and maximal pressure achieved. Furthermore, RCP was used primarily in older patients and patients with more complex repairs and longer cerebral ischemia times, although our data indicate that the influence of RCP was independent of these factors. The postoperative surgical follow-up visit occurred relatively early in the postoperative period (median of 62 days). This may have been too early for a final outcome measurement in some patients.
This imbalance in the data distribution and the small sample size complicated our attempts to evaluate the effects of RCP on memory dysfunction and NNO, independent of the joint effects of cerebral ischemia time and age. Whereas the MantelHaenszel tables demonstrate significant associations between RCP and adverse neuropsychological outcome, controlling for age and cerebral ischemia time separately, the small number of adverse outcomes limited the use of multiple logistic regressions and our ability to determine the independence of the effects of RCP from age and cerebral ischemia time jointly. Nevertheless, results of this analysis suggest an independent and negative influence of RCP on these outcomes.
4.4. Conclusions
Despite the limitations of the current study, the data provide preliminary evidence that RCP does not improve brain function following thoracic aortic surgery and potentially has harmful effects. Considerable uncertainty remains regarding the appropriateness of RCP, as there is controversy regarding its ability to provide cerebral protection and growing evidence that it may be injurious. Specific aspects of the RCP technique that require further investigation include: the optimal temperature, flow, and perfusion pressure; the cannulation method; the duration of perfusion (e.g. brief flush versus continuous flow); and the use of pharmacological adjuncts. Prospective randomized trials will be necessary to resolve these issues.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. Pretre and M. I. Turina Deep Hypothermic Circulatory Arrest Card. Surg. Adult, January 1, 2008; 3(2008): 431 - 442. [Full Text] |
||||
![]() |
I. Dorotta, P. Kimball-Jones, and R. Applegate II Deep hypothermia and circulatory arrest in adults. Seminars in Cardiothoracic and Vascular Anesthesia, March 1, 2007; 11(1): 66 - 76. [Abstract] [PDF] |
||||
![]() |
T. Miyairi, S. Takamoto, Y. Kotsuka, A. Takeuchi, K. Yamanaka, and H. Sato Comparison of neurocognitive results after coronary artery bypass grafting and thoracic aortic surgery using retrograde cerebral perfusion Eur. J. Cardiothorac. Surg., July 1, 2005; 28(1): 97 - 101. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Miyairi, S. Takamoto, Y. Kotsuka, A. Takeuchi, K. Yamanaka, and H. Sato Neurocognitive outcome after retrograde cerebral perfusion Ann. Thorac. Surg., May 1, 2004; 77(5): 1630 - 1634. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. L. Reich, L. M. Horn, S. Hossain, and S. Uysal Using jugular bulb oxyhemoglobin saturation to guide onset of deep hypothermic circulatory arrest does not affect post-operative neuropsychological function Eur. J. Cardiothorac. Surg., March 1, 2004; 25(3): 401 - 406. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Kazui Editorial comment: Post-operative neuropsychological function unaffected by SjO2 monitoring in DHCA Eur. J. Cardiothorac. Surg., March 1, 2004; 25(3): 406 - 408. [Full Text] [PDF] |
||||
![]() |
C. Hagl, N. Khaladj, M. Karck, K. Kallenbach, R. Leyh, M. Winterhalter, and A. Haverich Hypothermic circulatory arrest during ascending and aortic arch surgery: the theoretical impact of different cerebral perfusion techniques and other methods of cerebral protection Eur. J. Cardiothorac. Surg., September 1, 2003; 24(3): 371 - 378. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Murkin Retrograde cerebral perfusion: more risk than benefit? J. Thorac. Cardiovasc. Surg., September 1, 2003; 126(3): 631 - 633. [Full Text] [PDF] |
||||
![]() |
D.K. Harrington, M. Bonser, A. Moss, M.T.E. Heafield, M.J. Riddoch, and R.S. Bonser Neuropsychometric outcome following aortic arch surgery: a prospective randomized trial of retrograde cerebral perfusion J. Thorac. Cardiovasc. Surg., September 1, 2003; 126(3): 638 - 644. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. F. Duebener, I. Hagino, K. Schmitt, T. Sakamoto, C. Stamm, D. Zurakowski, H.-J. Schafers, and R. A. Jonas Direct visualization of minimal cerebral capillary flow during retrograde cerebral perfusion: an intravital fluorescence microscopy study in pigs Ann. Thorac. Surg., April 1, 2003; 75(4): 1288 - 1293. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Pretre and M. I. Turina Deep Hypothermic Circulatory Arrest Card. Surg. Adult, January 1, 2003; 2(2003): 401 - 412. [Full Text] |
||||
![]() |
D. Spielvogel, M. N. Mathur, and R. B. Griepp Aneurysms of the Aortic Arch Card. Surg. Adult, January 1, 2003; 2(2003): 1149 - 1168. [Full Text] |
||||
![]() |
W. A. L. Soong, S. Uysal, and D. L. Reich Cerebral Protection During Surgery of the Aortic Arch Seminars in Cardiothoracic and Vascular Anesthesia, November 1, 2001; 5(4): 286 - 292. [Abstract] [PDF] |
||||
![]() |
C. Baufreton, P. Binuani, F. Etcharry-Bouyx, and J. L. de Brux Long-term neuropsychological outcome after retrograde cerebral perfusion Eur. J. Cardiothorac. Surg., October 1, 2001; 20(4): 891 - 892. [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 |