| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Departments of Anesthesiology and Cardiothoracic Surgery, Wake Forest University School of Medicine, Winston-Salem, North Carolina
Address correspondence to John Butterworth, MD, Department of Anesthesiology, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157–1009. Address e-mail to jbutter{at}wfubmc.edu
Neurological and neuropsychological complications of cardiac surgery remain vexing problems. As the risk-adjusted incidences of mortality and of cardiac morbidity have declined, an increasingly sick and aged population is being offered cardiac surgical procedures. These patients are at greater risk than younger patients for perioperative strokes and neurobehavioral deficits, and such complications now represent an increasingly prevalent cause of postoperative disability (1,2).
What causes these neurological and neurobehavioral complications? The prevailing opinions are that they result from hypoperfusion, embolism, inflammation, or combinations of the three (3,4). The key question for investigators is whether anything can be done, either by modifying surgical techniques or perfusion devices or administering neuroprotective drugs, that will ameliorate the damaging processes and significantly reduce the risk of adverse neurobehavioral outcomes.
Many drugs have been studied as neuroprotective during cardiac surgery, including thiopental (5), remacemide (6), nimodipine (7,8), clomethiazole (9), prostacyclin (10), lidocaine (11), and GM1 ganglioside (12). Nevertheless, no study results have provided sufficient evidence to induce a change in standard clinical practice. In this issue of Anesthesia & Analgesia, Wang et al. (13) report that lidocaine infused, at doses often used to inhibit ventricular arrhythmias, will decrease the short-term risk of neurobehavioral deficits in patients undergoing coronary surgery with cardiopulmonary bypass. Given that Mitchell et al. (11) have previously shown benefit when lidocaine was similarly infused to patients undergoing cardiac valve surgery, do these new results indicate that the time has come to routinely administer lidocaine to all patients undergoing cardiopulmonary bypass?
First, what does it mean to have impaired neurobehavioral performance? After cardiac surgery, impaired performance on neurobehavioral testing occurs much more often than stroke (13,14) and is particularly prevalent within the first 2 wk after surgery, as exemplified by a 40% incidence observed by Wang et al. in their control group at 9 days after surgery. Currently, stroke occurs in between 2% and 5% of elective coronary bypass procedures (1,2). Achieving adequate statistical power to detect a major reduction in stroke rate would thus require enormous sample sizes. Making the reasonable assumption that strokes and neurobehavioral deficits result from similar brain injury mechanisms, many studies of neuroprotection have focused on neurobehavioral testing as providing an outcome measure with more comprehensive assessment of brain function and greater statistical power than stroke. Neurobehavioral testing at 1–3 mo has been accepted by the Food and Drug Administration and European drug registration agencies as a suitable outcome assessment of efficacy of putative neuroprotective drugs.
Nevertheless, some will argue that in the absence of differences in hard findings such as stroke, a difference in neurobehavioral outcome, such as that reported here by Wang et al. (13), cannot be regarded as a definitive measure of drug efficacy. We disagree. The difference between a persisting postoperative deficit detected by a neurobehavioral test and a postoperative stroke needs to be recognized as semantic. Both are the result of persisting brain damage, and this brain damage can often be defined anatomically using magnetic resonance imaging. The difference: one (stroke) has an obvious motor manifestation, and the other, often involving brain regions that have only limited influence on motor function, which may only be identified with specialized neurobehavioral testing, has an anatomical but a limited pathophysiological basis.
Controversy persists among those who study outcomes after cardiac surgery as to the best way to analyze and report the results of neurobehavioral testing. Some authors have compared the incidence of deficits, often insisting that patients must have greater than a threshold decline in two or more tests to qualify as having a deficit (14). We favor this technique, and this was what Wang et al. did in the present study (13). Others have compared the group mean scores on individual tests or groups of tests (6). This mean score approach (individual tests or groups of tests) is most powerful if one assumes that the distribution in test score changes will be normal, varying gradually from no change to substantial changes. If, however, changes in scores occur in an all-or-none fashion, as might be plausible after ischemic events, the distribution of changes in scores may reasonably be bimodally distributed, resulting in patients either experiencing a deficit or not. Under these circumstances the incidence approach could be more powerful than the mean approach, where individual deficits can be debited out in calculating overall means. The incidence approach introduces concerns about interactions among tests and makes it difficult to define and explain the primary outcome variable.
Controversy also persists as to the best time point at which to conduct postoperative assessments. When neurobehavioral testing is performed too soon after surgery (e.g., in the first few postoperative days), residual effects of sedative-hypnotic and opioid drugs, sleep deprivation, and fatigue may artificially decrease the performance. Thus, many deficits observed in the first few days after surgery would no longer be present during subsequent examinations (15). However, delaying assessment to 6 mo or later increases the likelihood that observed deficits will be the result of later events rather than the result of surgery. Note that in this study, assessments were performed at 9 days, leaving open the possibility that some of the deficits might not have proved to be persisting. Until neurobehavioral impairment persists for at least 1 mo, one cannot assume that it is a permanent finding. This is one concern we have about the between-group differences reported by Wang et al. (13).
In the previous study by Mitchell et al. (11), lidocaine improved outcome in patients undergoing aortic and mitral valve surgery. Wang et al. assumed that neurologic complications occur less often in closed chamber cardiac procedures (e.g., coronary bypass) than in open chamber procedures (e.g., valve surgery) and that the effects of lidocaine might differ in the two different circumstances (13). We think and have presented data that neurologic outcome is similar after valve or coronary surgery (15). However, it is reasonable to question whether the mechanisms of brain injury might be different with the two types of surgery. Patients undergoing valve surgery have many more (presumably air) emboli than patients undergoing coronary artery surgery. Patients undergoing coronary artery surgery, because of their underlying atherosclerotic disease, might be expected to have a greater risk of atheromatous emboli after aortic manipulation. The long-term sequelae of air versus atheromatous emboli might well be different.
There are differences between the two lidocaine studies. Mitchell et al. (11) found significantly fewer patients in the lidocaine-treated group than in the placebo group with a decline in one or more neurobehavioral test at 10 days, 10 wk, and 6 mo after surgery. But, there was no significant between-group difference when adverse neurobehavioral outcome was defined by a decline in two or more tests. Wang et al. tested for a decline in two or more tests but only at 9 days (13). Because a multiple clamping technique was used, the incidence of ascending aortic arthrosclerosis, as assessed by epi-aortic echocardiography, could define differences between the two groups. Later follow-up data were not provided. Mitchell et al. (11) presented data from a study of only 65 patients; Wang et al. (13) studied 118, making these two studies hard to compare.
Although there are limited experimental data (16) to suggest that lidocaine has a neuroprotectant effect in focal neurologic injury, the mechanism has not been elucidated and is not intuitive. It would seem that additional investigation at the cellular level will be required to define protectant mechanisms and establish effective intracellular neuroprotective concentrations of lidocaine.
What should we conclude? There is clear evidence suggesting that lidocaine may be neuroprotective in patients undergoing cardiac surgery with cardiopulmonary bypass. The study in this issue identifies an improvement only on the ninth postoperative day. Whereas a definitive answer will await larger randomized trials, it is tempting to hope that lidocaine, an inexpensive, widely available, and relatively safe compound, will prove effective in reducing adverse neurobehavioral outcomes, even if the effect is only transient or small.
References
This article has been cited by other articles:
|
|
 |
|
  I. Szalma, A. Kiss, L. Kardos, G. Horvath, E. Nyitrai, Z. Tordai, and L. Csiba Piracetam prevents cognitive decline in coronary artery bypass: a randomized trial versus placebo. Ann. Thorac. Surg., October 1, 2006; 82(4): 1430 - 1435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Culley, G. Crosby, J. Butterworth, and J. W. Hammon Mechanisms of Postoperative Neurobehavioral Deficits and Stroke May Differ * Response Anesth. Analg., August 1, 2003; 97(2): 601 - 601. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
