Encephalopathy and Stroke After Coronary Artery Bypass Grafting Incidence, Consequences, and Prediction

ORIGINAL CONTRIBUTION Encephalopathy and Stroke After Coronary Artery Bypass Grafting Incidence, Consequences, and Prediction

Guy M. McKhann, MD; Maura A. Grega, MSN; Louis M. Borowicz, Jr, MS; Michon Bechamps, MHS; Ola A. Selnes, PhD; William A. Baumgartner, MD; Richard M. Royall, PhD

Background: In contrast to perioperative stroke, much Results: The incidence of encephalopathy was 6.9% and less attention has been paid to those with evidence of dif- of stroke, 2.7%. For patients without either of these out- fuse brain encephalopathy, presenting as delirium, con- comes, the average length of stay in the hospital was 6.6 fusion, coma, and seizures in the immediate postopera- days and the mortality was 1.4%. In contrast, patients with tive period. encephalopathy had a length of stay of 15.2 days and a mortality of 7.5%, and those with stroke, a length of stay of 17.5 Objective: To determine the incidence, consequences, days and a mortality of 22.0%. Predictive models were de- and predictive factors for encephalopathy and stroke fol- veloped for encephalopathy involving 5 preoperative fac- lowing coronary artery bypass grafting. tors (age, past stroke, carotid bruit, hypertension, and diabetes) and 1 perioperative factor (time on cardiopulmonary Methods: In a prospective evaluation of 2711 patients bypass). The model for stroke involved only 3 preopera- operated on between January 1, 1997, and December tive risk factors (past stroke, hypertension, and diabetes). 31, 2000, preoperative risk factors were obtained before surgery and postoperative outcomes, encephalopathy Conclusions: Encephalopathy or stroke is associated with and stroke, were determined on a daily basis. All significant increases in length of stay and mortality after strokes were confirmed by neurologic consultation and, coronary artery bypass grafting. Patients at higher risk in most instances, by imaging. Logistic regression for these outcomes can be identified before surgery. analyses were performed to determine risk factors for these outcomes. Arch Neurol. 2002;59:1422-1428

ONSIDERABLE attention has In this report, based on the prospec- been paid to the inci- tive study of 2711 patients undergoing dence, consequences, and CABG, we describe the incidence, the im- prediction of stroke asso- mediate consequences in terms of LOS and ciated with coronary ar- mortality, and predictive factors for each teryC bypass grafting (CABG). In contrast, of these 2 postoperative outcomes, en- From the Department of much less attention has been paid to pa- cephalopathy and stroke, as well as the ag- Neurology (Drs McKhann and tients with evidence of diffuse encepha- gregate of patients having either out- Selnes), Division of Cardiac lopathy presenting as delirium, confu- come. Surgery, Department of Surgery sion, coma, and seizures in the immediate (Ms Grega and postoperative period. Only a few studies Dr Baumgartner), and PATIENTS AND METHODS Department of Neuroscience quantify the risk of encephalopathy after 1,2 (Dr McKhann), Johns Hopkins CABG. This is somewhat surprising be- PATIENTS School of Medicine; the Zanvyl cause encephalopathy is more common Krieger Mind/Brain Institute than stroke and, as we demonstrate in this Between January 1, 1997, and December 31, (Dr McKhann and study, is also associated with an in- 2000, 2711 patients underwent isolated CABG Mr Borowicz); and the creased length of hospital stay (LOS) and at The Johns Hopkins Medical Institutions, Bal- Department of Biostatistics, a greater mortality. Furthermore, it is not timore, Md. There were 10 intraoperative The Johns Hopkins University known whether the presence of postop- deaths, and 2701 patients had data allowing de- Bloomberg School of Public erative encephalopathy is predictive of cog- terminations of postoperative encephalopa- Health (Dr Royall), The Johns thy and stroke. Complete information about nitive declines months or years after Hopkins University, Baltimore, 3,4 variables used in the analysis for encephalopa- Md; and the MD Program, CABG. The ability to predict, before thy was available in 2669 patients and for stroke Eastern Virginia Medical surgery, those likely to have encephalo- in 2668. Patients with concomitant cardiac pro- School, Norfolk pathy could improve the care of these cedures, such as valve replacement, repair of (Ms Bechamps). patients. a congenital defect, or carotid endarterec-

(REPRINTED) ARCH NEUROL / VOL 59, SEP 2002 WWW.ARCHNEUROL.COM 1422

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 tomy, were excluded, as were patients undergoing surgery with- Stroke out a bypass pump. The study was approved by the institu- tional review board of The Johns Hopkins Medical Institutions. A standard clinical examination was used to determine the presence of any neurologic deficit. When patients were sus- PREOPERATIVE AND OPERATIVE VARIABLES pected of having a focal deficit, a neurology consultation was obtained. Stroke was defined as any persistent focal neuro- The following data were collected before surgery by members logic deficit lasting 24 hours or more. All stroke outcomes in of the study team (M.A.G. and L.M.B.): (1) history of a previ- this study were diagnosed by a neurologist and, in most cases, ous stroke, (2) history of hypertension, (3) history of diabetes were localized by imaging studies, either head computed mellitus, (4) presence of carotid bruit, and (5) age. With the tomography or magnetic resonance imaging. Of note, in this use of this information, probabilities of encephalopathy, stroke, study the diagnoses of stroke and encephalopathy were mutu- and the aggregate of either were developed. The only opera- ally exclusive. tive variable analyzed was time on cardiopulmonary bypass (CPB). We did not analyze aortic cross-clamping time as a sepa- STATISTICAL ANALYSES rate variable, because in our data this variable was closely associated with CPB. Prediction of Encephalopathy

SURGICAL PROCEDURES, ANESTHESIA, Logistic regression analyses were performed to measure the risk AND PERFUSION of postoperative encephalopathy. Two models were developed: one that used only those 5 variables available preopera- Eight cardiac surgeons were involved in this study. Although tively (age, history of previous stroke, hypertension, diabetes there were some individual variations in technique, most pro- mellitus, and the presence of a carotid bruit) and a second that cedures were standardized. All patients underwent median ster- used these 5 preoperative variables plus CBP time. notomy. Anesthetic technique was standardized and con- sisted of low- to intermediate-dose narcotics, inhalation agents, Prediction of Stroke and paralytics. Cardiopulmonary bypass was carried out with nonpulsatile flow (in most cases), alpha-stat pH blood gas man- In the first set of analyses presented herein, a logistic regres- agement, antegrade crystalloid cardioplegia (in most cases) with sion model developed previously for the prediction of stroke topical hypothermia, moderate systemic hypothermia (28°C- using variables available preoperatively was applied to the cur- 32°C), and pump flow rates to achieve a mean arterial pres- rent patient population. This previously derived model sure of 60 to 80 mm Hg. The placement of the aortic cross- included 5 variables (age, history of previous stroke, hyperten- clamp varied. Some surgeons used the single clamp technique sion, diabetes mellitus, and the presence of a carotid bruit) that exclusively, while others used this technique only in certain had been selected from a total of 42 variables (26 demographic clinical situations. This variable was not collected as part of this and medical history variables and 16 intraoperative variables) study. by a series of univariate and multivariate analyses. Two stroke risk models were presented in a previously pub- 6 OUTCOME VARIABLES lished article, one based on the 5 preoperative variables indicated above and another that included these preoperative vari- Postoperatively, members of the study team followed up pa- ables as well as the operative variable CPB time. With the use tients in the cardiac surgery care units daily, in collaboration of these 5 preoperative variables and the CPB time variable, a with the medical and nursing staff. revised model was created for the new population. In this analysis, only 3 variables were significant: age, hypertension, and his- LOS and Death tory of previous stroke. Thus, a simplified model for the prediction of stroke was developed. Postoperative LOS referred to the number of days the patient remained in the hospital after surgery either before discharge Prediction of Encephalopathy or Stroke or before death. Mortality was defined as the deaths of patients that occurred during their postoperative hospital stay. The final logistic regression used the 5 preoperative variables indicated above for prediction of the aggregate outcome “en- Encephalopathy cephalopathy or stroke.” Time on CPB was also combined with these preoperative variables in a second analysis for this out- The diagnosis of encephalopathy was made by 2 members of come. the study team (M.A.G. and L.M.B.), who were not aware of the patient’s risk factors at the time of evaluation. We included in the encephalopathy group patients having RESULTS delirium, coma, or seizures at any time during the postoperative stay, after the first 24 hours following surgery. Neither the duration nor severity of encephalopathy was included in The average age of the patients was 64 years, and the popu- the analysis. lation included 73% men. The incidence of hyperten- The criteria for delirium were similar to those outlined in sion was 69%; diabetes, 31%; carotid bruit, 10%; and pre- Diagnostic and Statistical Manual of Mental Disorders, Fourth Edi- 5 vious stroke, 6%. Encephalopathy occurred in 186 patients tion and included any of the following: episodes of confusion, in the population (an incidence of 6.9%), and there were agitation, and/or combativeness; alterations and fluctuations in levels of consciousness; acute problems with cognition, includ- 72 episodes of stroke (an incidence of 2.7%). Both of these ing memory; and changes in perception, including hallucina- outcomes, as well as the aggregate of those having ei- tions. The presence of delirium, coma, or seizures was docu- ther, were associated with longer LOS in the hospital and mented by the nursing or medical staff, and the documentation with greater mortality than in those without these out- was collected and evaluated daily by study observers. comes (Table 1).

(REPRINTED) ARCH NEUROL / VOL 59, SEP 2002 WWW.ARCHNEUROL.COM 1423

Patients No. of Patients Length of Stay, Mean, d No. of Deaths Mortality, % Without stroke or encephalopathy 2443 6.6 35 1.4 With stroke 72 17.5* 16 22.2* With encephalopathy 186 15.2* 14 7.5* With either 258 15.8* 30 11.6*

*PϽ.001 vs without stroke or encephalopathy.

Table 2. Prediction of Encephalopathy*

Variable Odds Ratio SE P Value 95% Confidence Interval Past stroke 2.15 0.52 .002 1.34-3.44 CB 1.57 0.34 .04 1.03-2.40 HTN 1.48 0.29 .05 1.01-2.16 DM 1.48 0.24 .02 1.08-2.04 Age 1 1.48 0.28 .04 1.02-2.13 Age 2 2.84 0.56 .001 1.93-4.18

*CB indicates carotid bruit; HTN, hypertension; and DM, diabetes mellitus. For age 1, patients at least 65 but younger than 75 years, the odds in favor of encephalopathy are estimated to be 1.48 times the odds for those younger than 65 years. For age 2, patients at least 75 years of age, the odds in favor of encephalopathy are estimated to be 2.84 times the odds for those younger than 65 years.

(Table 2). With these 5 factors, the probability of post- –Stroke operative encephalopathy for an individual patient can be calculated, as shown in Figure 1. – CB + CB PREDICTION OF STROKE – HTN + HTN – HTN + HTN When preoperative variables were used to predict the pres- Age, y– DM + DM – DM + DM – DM + DM – DM + DM ence of stroke in this current population with the pre- <65 0.03 0.04 0.04 0.06 0.04 0.06 0.06 0.09 vious model,6 more than twice as many strokes were pre- 65-75 0.04 0.06 0.06 0.09 0.06 0.09 0.09 0.13 dicted than actually occurred (161 vs 71). Moreover, this >75 0.08 0.11 0.11 0.15 0.11 0.16 0.16 0.22 lack of accuracy was not uniform across levels of risk. +Stroke In the high-risk group, the model predicted almost 3 times the number of observed strokes (141 vs 50, respectively). – CB + CB In the low-risk group, the model predicted fewer strokes than occurred (8 vs 13, respectively). As a result of the – HTN + HTN – HTN + HTN unreliable predictions produced by the model from 1994, we developed a logistic regression model applying the Age, y– DM + DM – DM + DM – DM + DM – DM + DM same 5 risk factors to the current population. Two of the variables (presence of carotid bruit and history of dia- <65 0.06 0.08 0.08 0.12 0.09 0.13 0.13 0.18 65-75 0.09 0.12 0.12 0.17 0.13 0.18 0.18 0.24 betes) were no longer significant. Thus, a simpler model, >75 0.15 0.21 0.21 0.21 0.29 0.28 0.29 0.38 using only 3 variables (age, history of previous stroke, and history of hypertension), was developed. Table 3 Figure 1. Probability of postoperative encephalopathy. To determine a presents the odds ratios, significance levels, and confi- patient’s risk of developing encephalopathy, ask the following questions and dence intervals for the 3 risk factors in this new model. then proceed to the correct part of the flowchart: has the patient had a previous stroke, yes or no; does the patient have a carotid bruit (CB), yes or With this simpler model, the risk of stroke for an indi- no; does the patient have a diagnosis of hypertension (HTN), yes or no; does vidual is indicated in Figure 2. the patient have a diagnosis of diabetes mellitus (DM), yes or no. Next, determine the patient’s age and proceed to the correct row of age groupings to see the probability for that patient. PREDICTION OF ENCEPHALOPATHY OR STROKE When taken together, 258 patients had either stroke or en- PREDICTION OF POSTOPERATIVE cephalopathy (incidence of 9.6%). We therefore exam- ENCEPHALOPATHY ined the ability of the logistic regression model to predict the development of either stroke or encephalopathy after When the 5 preoperative variables specified above were CABG. Similar to the model predicting encephalopathy, all used in a logistic regression to predict the development 5 preoperative variables were significant (Table 4). Time of post-CABG encephalopathy, all were significant on CPB was also significant for this outcome.

(REPRINTED) ARCH NEUROL / VOL 59, SEP 2002 WWW.ARCHNEUROL.COM 1424

Variable Odds Ratio SE P Value 95% Confidence Interval Past stroke 2.11 0.75 .04 1.05-4.23 Hypertension 1.97 0.63 .04 1.05-3.69 Age 1 2.39 0.76 .01 1.28-4.45 Age 2 5.02 1.61 .001 2.68-9.40

*For age 1, patients at least 65 but younger than 75 years, the odds in favor of stroke are estimated to be 2.39 times the odds for those younger than 65 years. For age 2, patients at least 75 years of age, the odds in favor of stroke are estimated to be 5.02 times the odds for those younger than 65 years.

ASSOCIATION OF TIME ON CPB WITH ENCEPHALOPATHY AND STROKE –Stroke +Stroke

Age, y– HTN + HTN – HTN + HTN Time on CPB was strongly associated with the prediction of postoperative encephalopathy (P=.006). As in- <65 0.01 0.01 0.02 0.03 65-75 0.02 0.03 0.04 0.07 dicated in Table 5, for patients without multiple risk >75 0.04 0.07 0.07 0.13 factors, there was a 50% increase in the probability of encephalopathy and stroke for each 30 minutes of addi- Figure 2. Probability of postoperative stroke. To determine a patient’s risk of tional CPB time. For patients with multiple risk factors, developing a stroke, ask the following questions and then proceed to the correct part of the flowchart: has the patient had a previous stroke, yes or the increase in probability was 30% for each additional no; does the patient have a diagnosis of hypertension (HTN), yes or no. Next, 30 minutes of CPB time. determine the patient’s age and proceed to the correct row of age groupings In our previous model for prediction of stroke, the to see the probability for that patient. only significant operative variable was time on CPB. When this variable was added to the current model for the pre- tween bypass time and the odds in favor of postoperative diction of stroke, the variable was not significant (P=.86). encephalopathy. An additional hour on CPB was asso- Thus, we did not include bypass time in the current model ciated with an approximate doubling of the probability for stroke prediction. of postoperative encephalopathy. These findings are con- sistent with previous studies that suggested that history COMMENT of a stroke and time on CPB are risk factors for encephalopathy.1,2 In this study we have demonstrated that both postop- The relationship between bypass time and encepha- erative encephalopathy and stroke are associated with lopathy may reflect a longer exposure to anesthesia. On significantly longer LOS and greater mortality. Mortal- the basis of autopsy studies, however, longer bypass time ity after encephalopathy is 5 times that for patients is also associated with an increased embolic load, with a without encephalopathy or stroke (7.5% vs 1.4%). Post- 90% increase in embolic load for each 1-hour increase operative stroke has received considerable attention in in bypass time.7 This diffuse showering of embolic ma- previous studies, and the effect of a stroke on postop- terial into subcortical and cortical areas, either by itself erative morbidity and mortality is well documented, or in the setting of variable degrees of hypoperfusion, may whereas the morbidity and mortality associated with account for the increasing encephalopathy.7,8 postoperative encephalopathy are not. This may be due Postoperative delirium after noncardiac surgical pro- to a tendency to ascribe this condition to age, anesthe- cedures has been reported with remarkable variations in sia, fever, infections, medications, preoperative cogni- incidence (0% to 73.5%).9 However, in recent prospective status, or a combination of these factors. Our find- tive studies of patients undergoing a variety of noncar- ings suggest, however, that there are specific characteristics diac surgical procedures, the incidence of delirium ranged of patients undergoing CABG that allow one to predict from 9% to 11.4%.10-13 The literature regarding encepha- those at risk for encephalopathy: the patient’s age, cere- lopathy is confusing, in part because different terminol- brovascular status, and time on the bypass pump. Iden- ogy may be used for the same conditions. In some re- tification of this subpopulation sets the stage for study- ports patients with stroke are included in the definition ing the mechanisms of encephalopathy after CABG of delirium. Other authors have reported on these pa- and possible approaches to either prevent or treat this tients by using descriptions such as “neurologic compli- outcome. cations,”14 “cognitive disorders,”1 and patients with both coma and stroke.15 In this report we have considered en- PREDICTION OF POSTOPERATIVE cephalopathy and stroke as separate outcomes. ENCEPHALOPATHY In attempting to define factors that could be predictive of postoperative encephalopathy, we did not In the prediction of encephalopathy, all 5 preoperative analyze a wide range of preoperative and operative risk factors (increasing age, history of stroke, hypertension, factors for postoperative encephalopathy (eg, medical diabetes, and the presence of a carotid bruit) were sig- history, physical findings, operative factors other than nificant. When these preoperative variables were com- bypass time, or postoperative factors) but instead bined with time on CPB, there was a relationship be- focused on the risk factors for cardiovascular disease. In

(REPRINTED) ARCH NEUROL / VOL 59, SEP 2002 WWW.ARCHNEUROL.COM 1425

Variable Odds Ratio SE P Value 95% Confidence Interval Past stroke 2.28 0.48 .001 1.50-3.45 CB 1.57 0.30 .02 1.09-2.28 HTN 1.63 0.28 .004 1.17-2.29 DM 1.41 0.20 .02 1.07-1.86 Age 1 1.70 0.28 .002 1.23-2.35 Age 2 3.61 0.62 .001 2.57-5.06

*CB indicates carotid bruit; HTN, hypertension; and DM, diabetes mellitus. For age 1, patients at least 65 but younger than 75 years, the odds in favor of encephalopathy or stroke are estimated to be 1.70 times the odds for those younger than 65 years. For age 2, patients at least 75 years of age, the odds in favor of encephalopathy or stroke are estimated to be 3.61 times the odds for those younger than 65 years.

PREDICTION OF STROKE OR ENCEPHALOPATHY Table 5. Risk of Encephalopathy and Effects of Cardiopulmonary Bypass Time In addition to separate predictive models for stroke and encephalopathy, a predictive model for an aggregate of Probability of Encephalopathy* Cardiopulmonary these outcomes was developed. This analysis demon- Bypass Time, min Low Risk High Risk strated that the same cardiovascular factors, with or with- 60 0.02 0.31 out time on CPB, can be used to predict this aggregate. 120 0.05 0.50 As discussed in the next subsection, this aggregate may, at least in part, represent a continuum of vascular insult *Low risk indicates younger age without cardiovascular risk factors; to the brain. high risk, older age with multiple cardiovascular risk factors. MECHANISMS OF ENCEPHALOPATHY AND STROKE addition, we did not have the data to include 2 cardiovascular risk factors: the arteriosclerotic state of the The mechanisms underlying postoperative encephalopa- aorta, as indicated by transesophageal echocardiogra- thy are not entirely clear. There are many possible causes, phy,16 and the patient’s preoperative cardiac function.17 including medications, cardiac failure, infections, and re- A more comprehensive approach might result in a dif- nal failure.9,10 However, as stated already, in this rela- ferent predictive model.9 tively homogeneous population, medications were used However, in a single hospital setting, the preopera- in a standardized way, and the incidence of infection, re- tive status, surgical procedures, anesthesia, and postop- nal failure, and cardiac failure was quite low. Regardless erative medications are relatively standardized for a given of these possible precipitating factors, the presence of vas- patient population. Nevertheless, postoperative encepha- cular risk factors predicted postoperative encephalopa- lopathy is still a common outcome among the elderly af- thy. Diffusion-weighted imaging (DWI) studies indi- ter major surgical procedures other than CABG.10-13 This cate multiple small vascular lesions in patients after CABG, suggests that there are factors other than those we have some of whom have encephalopathy,24 supporting the role listed that are important in the determination of which of these factors. In patients with strokes after CABG, DWI patients will have encephalopathy. Because cardiovas- also shows numerous small lesions in areas of the brain cular risk factors predicted post-CABG encephalopa- that are in addition to those involved in the major ische- thy, these factors may also be predictive of postopera- mic stroke. Thus, it is possible that, at least in some pa- tive encephalopathy in patients undergoing other forms tients, postoperative encephalopathy and stroke after of surgery. CABG represent a continuum of vascular insult to the brain. On the basis of current data, it is clear that pa- PREDICTION OF STROKE tients with risk factors for cerebrovascular disease are at greater risk of these outcomes and associated adverse se- In this study, a model for prediction of stroke after quelae, including increased LOS and death. CABG developed 8 to 10 years ago6 was no longer valid. It has been generally assumed that perioperative When applied to a current population, this previous stroke is caused by embolic material deposited in larger model overpredicted the occurrence of strokes, particu- vessels.25 This hypothesis has been supported by stud- larly in the high-risk group, and underpredicted in the ies with DWI, which indicates new areas of ischemia in low-risk group. With the use of a more recent surgical the postoperative period.24 These areas of new ischemia population, a simpler model that used only 3 variables are often not detectable with conventional magnetic reso- (increasing age, history of previous stroke, and hyper- nance imaging techniques, because conventional mag- tension) was highly significant. This finding is in agree- netic resonance imaging cannot readily distinguish be- ment with those of other studies of predictive factors for tween new and preexisting small vascular lesions. The stroke.6,18-23 In all of these studies, including our own, prevalence of silent infarcts in populations with coro- the predictive factors are evidence of cerebral and/or nary artery disease is high, with some studies reporting peripheral vascular disease. a 30% rate.26 In candidates for CABG, the prevalence of

(REPRINTED) ARCH NEUROL / VOL 59, SEP 2002 WWW.ARCHNEUROL.COM 1426

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 the predicted stroke rate may be twice as high or more. –Stroke Including patients at risk for encephalopathy in such a study would substantially increase the risks of adverse – CB + CB outcomes and thus reduce the numbers of patients required to show a definite effect. We base this suggestion – HTN + HTN – HTN + HTN on the demonstration that a predictive model for this combined outcome involves cardiovascular risk factors. If cor- Age, y– DM + DM – DM + DM – DM + DM – DM + DM rect, the probabilities of an adverse outcome reach re- <65 0.03 0.05 0.05 0.07 0.05 0.07 0.08 0.11 markably high levels (Figure 3). 65-75 0.08 0.08 0.09 0.12 0.09 0.11 0.13 0.18 Finally, we strongly suggest that encephalopathy af- >75 0.11 0.15 0.17 0.22 0.16 0.22 0.24 0.31 ter CABG should not be considered an inevitable out- +Stroke come for elderly patients. The consequences in terms of morbidity and mortality are too high. Careful prospec- – CB + CB tive studies, including postoperative DWI, are indicated to determine which patients have encephalopathy re- – HTN + HTN – HTN + HTN lated to vascular insults. As is being done with delirium after other surgical procedures, attempts to understand Age, y– DM + DM – DM + DM – DM + DM – DM + DM this phenomenon and to modify its outcome are war- ranted. <65 0.07 0.10 0.11 0.15 0.11 0.15 0.17 0.22 65-75 0.12 0.16 0.18 0.23 0.17 0.23 0.25 0.33 >75 0.22 0.28 0.32 0.39 0.42 0.39 0.42 0.51 Accepted for publication May 2, 2002. Author contributions: Study concept and design (Drs Figure 3. Probability of either postoperative stroke or encephalopathy. To McKhann, Selnes, and Royall; Mss Grega and Bechamps; determine a patient’s risk of developing either stroke or encephalopathy, ask and Mr Borowicz); acquisition of data (Ms Grega, Mr the following questions and then proceed to the correct part of the flowchart: Borowicz, and Dr Baumgartner); statistical analysis has the patient had a previous stroke, yes or no; does the patient have a carotid bruit (CB), yes or no; does the patient have a diagnosis of (Dr Royall); analysis and interpretation of data (Drs hypertension (HTN), yes or no; does the patient have a diagnosis of diabetes McKhann, Selnes, Baumgartner, and Royall; Mss Grega mellitus (DM), yes or no. Next, determine the patient’s age and proceed to and Bechamps; and Mr Borowicz); drafting of the manu- the correct row of age groupings to see the probability for that patient. script (Drs McKhann, Selnes, Baumgartner, and Royall; Mss Grega and Bechamps; and Mr Borowicz); critical re- silent infarcts is likely to be even higher. One recent study27 vision of the manuscript for important intellectual content from Japan reported that 50% of patients undergoing (Drs McKhann, Selnes, Baumgartner, and Royall; Mss CABG had one or more lacunar infarcts that are shown Grega and Bechamps; and Mr Borowicz); obtaining fund- on their preoperative magnetic resonance images.27 To ing (Drs McKhann and Baumgartner); administrative, tech- our knowledge, careful preoperative and postoperative nical, and material support (Drs McKhann and Baumgart- magnetic resonance imaging and DWI of patients at risk ner); and study supervision (Drs McKhann and Selnes). for encephalopathy have not been performed. This study was supported by grant 5 R01 NS35610-05 The ability to predict outcomes associated with from the National Institutes of Health, Bethesda, Md, and surgery has important implications. First, the added by the Charles A. Dana Foundation, New York, NY. information available to patients, their families, and We thank Robert Wityk, MD, and Pamela Talalay, PhD, physicians assists with decisions concerning possible for helpful comments on analysis, interpretation, and pre- therapies. For patients at higher risk of stroke, inter- sentation. ventional cardiology procedures, modification of exist- Corresponding author and reprints: Guy M. McKhann, ing CABG procedures, and alternative surgical proce- MD, Zanvyl Krieger Mind/Brain Institute, The Johns Hop- dures, such as surgery without a bypass pump, are kins University, 338 Krieger Hall, 3400 N Charles St, Balti- possible options. more, MD 21218-2685 (e-mail: [email protected]). Our studies indicate that postoperative encephalopathy is associated with increased LOS and mortality. REFERENCES Several studies have indicated that identifying patients at risk of delirium and instigating specific measures can 28-30 1. van der Mast RC, van den Broek WW, Fekkes D, Pepplinkhuizen L, Habbema JD. improve these outcomes. Notably, in all of these mod- Incidence of and preoperative predictors for delirium after cardiac surgery. J Psy- els the important preoperative factors can be easily de- chosom Res. 1999;46:479-483. termined by a physician while counseling patients and 2. Rolfson DB, McElhanay JE, Rockwood K, et al. Incidence and risk factors for de- their families about surgery, thus permitting more edu- lirium and other adverse outcomes in older adults after coronary artery bypass cated decisions. graft surgery. Can J Cardiol. 1999;15:771-776. 3. Selnes OA, Royall RM, Grega MA, Borowicz LM Jr, Quaskey S, McKhann GM. A second implication is concerned with identifying Cognitive changes 5 years after coronary artery bypass grafting: is there evi- patients at higher risk of adverse outcomes so that they dence of late decline? Arch Neurol. 2001;58:598-604. could be included in studies of the effects of neuropro- 4. Newman MF, Kirchner JL, Phillips-Bute B, et al. Longitudinal assessment of neu- tective agents or practices. For example, with a stroke rocognitive function after coronary-artery bypass surgery. N Engl J Med. 2001; 344:395-402. rate of 2% to 3% in the overall CABG population, very 5. American Psychiatric Association. Diagnostic and Statistical Manual of Mental large numbers of patients would be required to demon- Disorders, Fourth Edition. Washington, DC: American Psychiatric Association; strate any efficacy, whereas in a population at greater risk, 1994.

(REPRINTED) ARCH NEUROL / VOL 59, SEP 2002 WWW.ARCHNEUROL.COM 1427

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 10/01/2021 6. McKhann GM, Goldsborough MA, Borowicz LM Jr, et al. Predictors of stroke risk 19. Taggart DP, Westaby S. Neurological and cognitive disorders after coronary ar- in coronary artery bypass patients. Ann Thorac Surg. 1997;63:516-521. tery bypass grafting. Curr Opin Cardiol. 2001;16:271-276. 7. Brown WR, Moody DM, Challa VR, Stump DA, Hammon JW. Longer duration of 20. Herlitz J, Wognsen GB, Haglid M, et al. Risk indicators for cerebrovascular com- cardiopulmonary bypass is associated with greater numbers of cerebral micro- plications after coronary artery bypass grafting. J Thorac Cardiovasc Surg. 1998; emboli. Stroke. 2000;31:707-713. 46:20-24. 8. Caplan LR, Hennerici M. Impaired clearance of emboli (washout) is an impor- 21. John R, Choudhri AF, Weinberg AD, et al. Multicenter review of preoperative risk tant link between hypoperfusion, embolism, and ischemic stroke. Arch Neurol. factors for stroke after coronary artery bypass grafting. Ann Thorac Surg. 2000; 1998;55:1475-1482. 69:30-35. 9. van der Mast RC. Postoperative delirium. Dement Geriatr Cogn Disord. 1999; 22. Newman MF, Wolman R, Kanchuger M, et al, for the Multicenter Study of Peri- 10:401-405. operative Ischemia (McSPI) Research Group. Multicenter preoperative stroke in- 10. Litaker D, Locala J, Franco K, Bronson DL, Tannous Z. Preoperative risk factors dex for patients undergoing coronary artery bypass graft surgery. Circulation. for postoperative delirium. Gen Hosp Psychiatry. 2001;23:84-89. 1996;94:1174-1180. 11. Kaneko T, Takahashi S, Naka T, Hirooka Y, Inoue Y, Kaibara N. Postoperative 23. Stamou SC, Hill PC, Dangas G, et al. Stroke after coronary artery bypass: inci- delirium following gastrointestinal surgery in elderly patients. Surg Today. 1997; dence, predictors, and clinical outcome. Stroke. 2001;32:1508-1513. 27:107-111. 24. Wityk RJ, Goldsborough MA, Hillis A, et al. Diffusion- and perfusion-weighted 12. Brauer C, Morrison RS, Silberzweig SB, Siu AL. The cause of delirium in pa- brain magnetic resonance imaging in patients with neurologic complications af- tients with hip fracture. Arch Intern Med. 2000;160:1856-1860. ter cardiac surgery. Arch Neurol. 2001;58:571-576. 13. Marcantonio ER, Goldman L, Mangione CM, et al. A clinical prediction rule for 25. Borger MA, Ivanov J, Weisel RD, Rao V, Peniston CM. Stroke during coronary delirium after elective noncardiac surgery. JAMA. 1994;271:134-139. bypass surgery: principal role of cerebral macroemboli. Eur J Cardiothorac Surg. 14. Libman RB, Wirkowski E, Neystat M, Barr W, Gebb S, Graver M. Stroke associ- 2001;19:627-632. ated with cardiac surgery: determinants, timing, and stroke subtypes. Arch Neu- 26. Bernick C, Kuller L, Dulberg C, et al, for the Cardiovascular Health Study Col- rol. 1997;54:83-87. laborative Research Group. Silent MRI infarcts and the risk of future stroke: the 15. Smith MH, Wagenknecht LE, Legault C, et al. Age and other risk factors for neu- Cardiovascular Health Study. Neurology. 2001;57:1222-1229. ropsychologic decline in patients undergoing coronary artery bypass graft sur- 27. Goto T, Baba T, Honma K, et al. Magnetic resonance imaging findings and post- gery. J Cardiothorac Vasc Anesth. 2000;14:428-432. operative neurologic dysfunction in elderly patients undergoing coronary artery 16. Wareing TH, Davila-Roman VG, Barzilai B, Murphy SF, Kouchoukos NT. Man- bypass grafting. Ann Thorac Surg. 2001;72:137-142. agement of the severely atherosclerotic ascending aorta during cardiac opera- 28. Marcantonio ER, Flacker JM, Wright RJ, Resnick NM. Reducing delirium after tions: a strategy for detection and treatment. J Thorac Cardiovasc Surg. 1992; hip fracture: a randomized trial. J Am Geriatr Soc. 2001;49:516-522. 10:453-462. 29. Inouye SK. Prevention of delirium in hospitalized older patients: risk factors and 17. Caplan LR. Protecting the brains of patients after heart surgery. Arch Neurol. 2001; targeted intervention strategies. Ann Med. 2000;32:257-263. 58:549-550. 30. Rizzo JA, Bogardus ST Jr, Leo-Summers L, Williams CS, Acampora D, Inouye 18. Gardner TJ, Horneffer PJ, Manolio TA, et al. Stroke following coronary artery by- SK. Multicomponent targeted intervention to prevent delirium in hospitalized older pass grafting: a ten-year study. Ann Thorac Surg. 1985;40:574-581. patients: what is the economic value? Med Care. 2001;39:740-752.

CME Announcement CME Announcement

CME Hiatus: July Through December 2002. CME from JAMA/ARCHIVES will be suspended between July and December 2002. Beginning in early 2003, we will offer a new online CME program that will provide many enhancements: • Article-specific questions • Hypertext links from questions to the relevant content • Online CME questionnaire • Printable CME certificates and ability to access total CME credits We apologize for the interruption in CME and hope that you will enjoy the improved online features that will be available in early 2003.

(REPRINTED) ARCH NEUROL / VOL 59, SEP 2002 WWW.ARCHNEUROL.COM 1428