Neuropsychologic change after cardiac surgery: A critical review

Neuropsychologic Change After Cardiac Surgery: A Critical Review Louis M. Borowicz, MS, Maura A. Goldsborough, RN, Ola A. Seines, PhD, and Guy M. McKhann, MD Studies that have examined neuropsychologic change after cardiac surgery address three main issues: (1) the incidence of cognitive change; (2) the identification of factors that put patients at higher risk; and (3) the evaluation of interventions to prevent these complications. This review attempts to bring together concerns associated with various study designs and to integrate the conclusions from these studies. Thirty-five studies have been examined in this review. Some of the difficulties encountered when quantifying the degree of cognitive change are related to study design, patient sampling, and deficit definition. Additionally, changing patient populations have influenced results reported from different health care settings. Increasing age and longer cardiopulmonary bypass times have been correlated with

S EARLY AS the 1920s, when visionaries speculated A about the possibilities of cardiac surgery, there was concern about the potential for brain injury that could result from cardiac procedures) Decades of work led to the first successful operation within the open human heart by Dr F. John Lewis on September 2, 1952 at the University of Minnesota Hospital. The first successful application of the cardiopulmonary bypass machine 2 years later made many types of cardiac surgery eventually possible, including coronary artery bypass grafting (CABG), valve replacement, and heart transplantation. Introduction of cardiopulmonary bypass, however, also resulted in adverse neurologic outcomes including coma, delirium, cognitive impairment, and stroke. The decades that followed were times of innovation, refinement, and expansion that saw cardiac surgery become a routine surgical practice and its associated mortality decrease dramatically,2,3 but various degrees of neurologic injury are still of concern.

SPECTRUM OF NEUROLOGIC INJURY

Neurologic complications after cardiac surgery constitute a broad spectrum of injury. The most serious injury is observed in patients who remain comatose or never regain meaningful neurologic function. Stroke is another devastating complication that can leave patients permanently disabled. The incidence of stroke ranges from 0.8% to 5.2%. 4-9 Much less information is available on the incidence of postoperative encephalopathy, delirium, and confusion, but values as high as 10% have been cited. 6 Patients may experience &sorientation, slowness in awakening from anesthesia, and agitation requiring restraint. These complications may arise from a variety of metabolic alterations during the postoperative period and merit further in-depth investigation. Finally, changes in cognitive status, although at times subtle, appear prevalent in this patzent population. Neuropsychologic testing of patients before and after cardiac surgery has documented important findings. The focus of thzs report is to review these studies, trace their evolution, and suggest ways to improve study design.

cognitive decline in a number of studies. Filtration devices and blood gas management techniques have decreased but not eliminated the number of patients who have cognitive decline. Cognitive change exists following cardiac procedures. Identification of a subgroup of patients at high risk for cognitive change has been difficult, possibly due to issues of study design. Design of future studies, which may include intraoperative or pharmacologic interventions, is dependent on identification of this high-risk group.

Copyright © 1996 by W.B. Saunders Company KEY WORDS: cardiac surgery, coronary artery bypass grafting, neuropsychology

NEUROPSYCHOLOGY

Neuropsychology is the applied science of behavioral expression of brain dysfunction. Neuropsychologic testing typically includes measures of memory, attention, visuoconstructional ability, and motor and psychomotor speed. The sensitivity and precision of these neuropsychologic measures make them well suited for identification, assessment, and investigation of cognitive complications of many neurologic diseases and brain injuries. Neuropsychologlc findings can be crucial for diagnosis in many conditions where computed tomography scans and other laboratory data are not useful or practical. Furthermore, researchers have used neuropsychologic tests in combination with medical history, physical examination, and operative variables to determine the incidence of cognitive change, as well as factors that increase risk. Successive neuropsychologic examinations can provide reliable indications of cognitwe change and how rapidly change occurs. Repeated testing can measure cognitive changes after surgical procedures or medical treatment. 1° Thus, it is pertinent to use neuropsychologic tests to determine the impact of cardiac surgery on the brain. Patients may not have focal neurologic deficits on examination, but neuropsychologic testing may show clinically significant cognitive changes that can profoundly affect outcome.ll, 12 Neuropsychologic evaluations, the subject of this review, have been used in the cardiac surgical population since the late 1950s.13 In one of the first published reports, Gilman 14 discussed 35 patients undergoing cardiac operations with cardiopulmonary bypass and hypothermia. Twelve patients From the Zanvyl KriegerMmd/Bram lnstttute, the Department of Neurology, Dtvtston of Neuropsychology, and the Department of Surgery, Dtvtston of Cardtac Surgery, Johns Hopkms Umverstty, Balttmore, MD Supported m part by The Research Network on SuccessfulAgmg of the John D and Cathenne T. MacArthurFoundatton; The CharlesA. Dana Foundatton, and TheAIREN Foundatton. Address repnnt requests to Louts M Borowtcz, MS, Johns Hopkms Hospital, Meyer2-222, 600 N Wolfe St, Balttmore, MD 21287-7222 Copyright © 1996 by W.B. Saunders Company 1053-0770/96/1001-001553 00/0

Journal of Cardlothoractc and Vascular Anesthes/a, Vol 1O,No 1 (January), 1996 pp 105-112

105

106

BOROWlCZ ET AL

had signs of cerebral disturbance; 6 of these 12 had disorders of intellectual function, including reading and calculation. Many studies have built upon these early results in an attempt to understand postoperative cognitive complications. PREVIOUS STUDIES

This review is based on 35 studies between 1980 and 19946,15-48(Table 1) that are fundamental to understanding issues of neuropsychologic change associated with cardiac surgery. This body of literature strongly supports the occurrence of cognitive decline, both short-term and longterm, after cardiac surgery. The incidence of short-term decline (studied less than 2 weeks postoperatively) ranges

from 26% to 79%, 20,23,24,29whereas for long-term decline (studied more than 1 month postoperatively), it ranges from 0% to 37%. 23,24The source of this variability requires further definition. Despite the large number of neuropsychologic studies, few have identified variables that increase the risk for cognitive change. Increasing age is the most consistently reported factor related to cognitive decline after cardiac surgery. 26,29,31,42 Prolonged cardiopulmonary bypass time has also been noted in many studies. 21,23,29,35,42However, age and cardiopulmonary bypass time may be surrogates for other physiologic characteristics of individual patients and intraoperative factors. 49 Consensus has not been reached on other premorbid risk

Table 1. Demographic and Outcome Measures in Neuropsychologic Studies in Cardiac Surgery Author

Incidence studies Sotanlemi TM Breuer 6 Hikkinen TM Calabrese 17 Jackson TM Klonoff 19 Stump20 Control studies EIsass 21 Raymond 22 Newman 23 Shaw 24 Hammeke 25 Townes 26 Blumentha127 O'Bnen 2s Correlabonal studies Kolkka z9 Fish 3° Savageau 3~ Jenkins 32 Aberg 33 Folks ~ Sotanleml 3s Nevln 36 Toner 3~ Intervent~onal studies Ellis 38 Slogoff 3° Garvey4° Aris 41 Nussmeler 42 Fish 43 Forsman 44 Grote 4s Wong 46 Sellman 47 Pugsley4s

No Patients (No Controls)

%Male

Mean Age (yrs)

63 86 75 100 NR 89 NR

44 NR 52 NR NR 55 NR

NR NR NR 10 NR 10 8

Other Other Other Anychange NR Any change Other

79 90 93 89 100 86 74 100

48 56 55 53 60 59 t 58

1 12 10 10 12 14 4 6

Any change Group comparison SD method SD method SD method Group comparison Group comparison Group comparison

204 20 227 318 94 58 49 65 15

NR 100 81 84 NR 76 64 NR 73

56 NR 54 54 NR 62 46 NR 59

NR 11 2 3 3 4 7 9 12

Other Any change SD method SD method Any change Group companson Any change SD method SD method

30 204 46 100 182 100 35 29 34 54 105

100 87 NR NR 67 100 77 72 NR 100 NR

55 55 NR 1" 1" 1 1" 1" 1. 59 1.

3 1 1 7 1 6 9 5 7 8 9

Other Other SD method Group comparison Group comparison Group compartson Group comparison SD method Group comparison SD method SD method

Year

Surgery

1980 1983 1985 1987 1989 1989 1990

Valve CABG All CABG All CABG CABG

100 421 364 59 57 135 27

1984 1984 1987 1987 1988 1989 1991 1992

All CAB CABG CABG CABG All All All

19 31 67 312 46 90 31 20

1980 1982 1982 1983 1984 1986 1986 1989 1994

All CABG All CABG All CABG Valve CABG CABG

1980 1982 1983 1986 1986 1987 1990 1992 1992 1992 1994

CABG All All All All CABG All CABG CABG CABG CABG

(17) (16) (24) (50) (40) (47) (8) (20)

No of T e s t s

DeficttDefmltmn*

/

Abbreviation NR, not reported *Deficit definition: other--the study either dtd not clearly identify how outcomes were measured or other deficit definitions were used; any change--a study defined cognitive change as any amount of change on any test, group companson--a study compared differences in cognttive test scores between two groups; SD method--a study used the group standard deviation of a test score to define deficit 1" = mean ages reported by 9rou p

NEUROPSYCHOLOGIC CHANGE AFTER CARDIAC SURGERY

factors or issues such as intraoperative blood pressure and perfusion flow rate. 29,38Several studies have demonstrated a lower incidence of cognitive decline related to oxygenator type, 47 u s e of arterial filters, 40,41 and blood gas management. 5° Some investigators have used neuropsychologic change and cerebral monitoring to identify possible mechanisms for mjury. 4s Further clarification may require improved study design as addressed next.

107

Table 3 Stroke and Mortahty Outcomes for CABG Patients Studied from February 1992 Through March 1993 at the Johns Hopkins Hospital No Patients

CVA Rate

Mortahty Rate

Cognltwelytested patients Untested patients

176 280

2 8% 7 1%

1 1% 6 1%

Totals

456

5 7%

4 2%

Changing Pauent Populattons As m many areas in which techmques have undergone major improvements, the population of patients undergomg CABG has changed dramatically since neuropsychologic studies began. At this institution, the average age of patients undergoing CABG increased from 52 years in 19748 to 64 years in 1994; this trend continues, and the mean age of CABG patients currently is 65. Moreover, about one third of patients are over 70. It Is believed that increasing age is a risk factor for cognitive decline, particularly in patients older than 70 51 Consequently, past studies that included only younger patients and studies from institutions that serve an older population can be expected to show different results. Thus, some studies may no longer be valid for comparison with current populations. In addition to being older, patients now present with more chronic disease risk factors. In Table 2, the current population of patients at Johns Hopkins Hospital is compared with patients studied by Coffey et al at Duke Unwersity from 1975 to 1982. 7 It can be seen that patients are now older and possibly slcker. It is unclear to what extent chronic illnesses affect neuropsychologlc testing, but it has been shown that these factors are associated with serious neurologic outcomes such as stroke. 79 Cardiac surgery populations from small community hospitals, or from countries with different health care standards, may vary In this manner. In summary, this overview highhghts some of the problems in quantifying the degree of cognitive change after cardiac surgery, making it difficult to assign cause and effect. In the following sections, issues of study design and suggestions for future studies are considered. ISSUES OF STUDY DESIGN

A variety of study designs are applicable to research m cardiac surgical populations (Table 1). Incidence studies attempt to determme the frequency of the problem in a given population. 6,15-28 Correlational studies examine the relationship between cognitive change and variables responsible for change. 2937 Interventlonal studies acknowledge Table 2 Chronic Disease Risk Factors m CABG Patients From Duke Unwersity Medical Center in 19837 and Johns Hopkins Hospital ,n 1995 Risk Factors

1983

1995

Hypertension Past stroke Carotid bru~t Diabetes melhtus

38% 2% 6% 10%

65% 6% 17% 29%

that changes exist and intervene to prevent further complications. 38"~8In all these situations, the addition of control groups is important to examine relationships between the populations in question and other surgical and nonsurglcal groups. All study designs have inherent flaws that can affect the clarification of specific research questions. These problems are reviewed next.

PaUent Samphng One of the most significant limitations in design of neuropsychologic studies of cardiac surgical patients is inadvertent biasing of the study population. Bias may occur for several reasons. These include length of the neuropsychologic test battery and length of interview process, urgency of surgery, and intentional or unintentional patient exclusion. Intentional exclusion criteria refer to issues such as age restrictions or past medical history. Unintentional exclusion of patients may occur because of patient unavailabihty or patients who are non-native English speaking. Bias may also occur because patients refuse testing for reasons that include preoperative anxiety, reluctance to participate, lack of interest, or inability to return for follow-up testing. Because these refusals may be random, a systematic bias may not exist. However, a large group of patients undergoing cardiac surgery may not be represented in the reports of neuropsychologic outcomes. Studies seldom provide outcome information on patients who undergo surgery but are not tested during the study p e r i o d . 23-25,32 In an ongoing study at this institution, detailed information was collected on all 456 patients who underwent CABG during the study period. Of 456 patients, 39% completed neuropsychologlc testing, whereas 61% refused or could not be approached for neuropsychologic testing. This distribution is shown in Table 3. These comparisons are tmportant because the stroke rate for patients who did not undergo neuropsychologic testing was two and a half times greater and mortality rate six times greater than those who completed cognmve testing. Clearly, the population of patients who received neuropsychologlc testing had a different neurologic outcome. Thus, the authors know that their patient sample is biased, and the actual incidence of neuropsychologac decline may be much greater, but this is unknown. Sample sizes from studies on the incidence of stroke are much larger 7-9 than sample sizes from neuropsychologic studies. The majority of neuropsychologlc studies in CABG patients have sample sizes between 19 and 90 patients (see Table 1). Problems encountered when designing studies are reflected in these small sample sizes. They may be insuffi-

108

cient to make generalizable conclusions about the true incidence of cognitive decline. The authors suggest, when attempting to identify the true incidence of decline, that minimal exclusion criteria should be used. Some studies have excluded patients with previous cerebrovascular disease, 28 and patients over a certain age level. ~9Because these patients do undergo cardiac surgery, they should be represented in the study population.

Control Groups Control groups are essential to any study of incidence. It has not been clearly established that the occurrence of cogmtive change is unique to cardiac patients compared with other patients undergoing surgical procedures of similar duration. 21,23 Therefore, the ideal study design would include two control groups, one surgical and one nonsurgical. 51 To date, only two studies have incorporated this study design. 25,5° Surgical control groups should have similar periods of general anesthesia and adequate numbers of patients to sample. Previous studies have used control groups undergoing thoracic and peripheral vascular surgery not requiring cardiopulmonary bypass. 2~-25Because it is unknown whether men and women respond in a similar fashion to surgery, the surgical sample should have a gender distribution comparable with cardiac surgical patients. One recent abstract 52 described a group of patients undergoing total knee replacement as controls. However, there were differences in gender distribution for this group as the majority of knee replacement patients were women, and the majority of cardiac patients were men. Surgical control groups should have similar age and education distributions because these variables have important influences on neuropsychologic test scores. In a nonsurgical control group, education and age become even more important. These control groups may sample from many populations, but the main purpose is to examine the effects of repeated neuropsychologic testing. Educational level and age significantly affect performance on most neuropsychologic tests, 53,54 and it is not appropriate to compare unadjusted scores of patients with college education to those with high school education. The effect of education has not been systematically reported in cardiac studies. Only 7 of the 35 studies in Table 1 reported the patients' level of education. 19,22,25-27,38,43 It is critical to consider covariates that may significantly influence performance when selecting a control group, since there are limitations to post-hoc statistical methods to compensate for demographic differences in study groups.

BOROWICZ ET AL

The authors' research suggests that patients were more likely to refuse testing at one year if their neuropsychologic scores had declined at one month. Patients were also more likely to refuse testing at subsequent time points if their preoperative neuropsychologic scores were below the average for the group. The majority of patients in this study who refused to complete follow-up testing at one month, also refused testing at one year. Other studies have reported similar refusal rates 15,24,38or have not provided information about patients who refused follow-up testing, i9,31,34,36

Neuropsychologic Test Battenes Few studies have used the same neuropsychologic tests. This is not surprising since there are approximately 350 different neuropsychologic tests 1° that can be administered individually or in combination as a test battery. Many neuropsychologic tests overlap in their measurement of cognitive functions (domains). Eighty percent of the studies have measured memory. Other domains measured include attention, psychomotor speed, motor speed, or language. Domains in which 50% of patients reportedly declined include memory, psychomotor speed, and attention. An attempt is being made to establish a standard, more focused, and shorter core group of neuropsychologic tests for use in cardiac surgery studies, 56 so that study results can be compared confidently. Cognitive test batteries vary in the amount of time required to administer. Some studies have used extensive test batteries that may take up to 2 hours to complete. 22The length of these interviews will influence which patients agree to study participation, since only the least impaired patients may be able to complete a lengthy test battery. The effects of mood should be considered because the occurrence of postoperative depression in these patients has been widely reported, s7 There has been concern that depression may mask cognitive changes in the postoperative period. Several investigators have included depression indices in neuropsychologic test batteries. ~7,18,22,26,34,37 In general, preoperative, postoperative, or change in depression does not correlate with neuropsychologic test decline. One study showed a relationship between patients' subjective complaints of cognitive impairment and depression scores but no relationship with actual neuropsychologic test decline 58. Anxiety in the preoperative period may influence test scores. Past studies have included measures of anxiety, 17'18'22'26'32'34'37 but anxiety issues did not appear to affect neuropsychologic outcomes for the group. These issues merit further investigation.

Patient Comphance

Testing Intervals

There are valid reasons why subjects may be unable to complete the research protocol: incapacitation, unscheduled relocation, or death. In neuropsychologic research, like other research requiring extensive lntervaew time, systematic bias may occur when patients who are most cognitively impaired choose not to complete follow-up testing. 55

Studies are usually designed to test patients before surgery, 3 to 7 days after surgery, 1 to 2 months postoperatively, and 6 months to 1 year postoperatively. 49 Each of these testing intervals has certain limitations as indicated next. Preoperative testing is often performed the day before surgery, which is a time of emotional turmoil for many

NEUROPSYCHOLOGIC CHANGE AFTER CARDIAC SURGERY

patients and families. Therefore, patients may be resistant or hesitant when asked to complete a 1- to 2-hour interview. Also, because of the increase in same day admissions, the time available to contact and interview patients preoperatively is decreasing. Indeed, in some instances, patients go directly to emergency surgery from the cardiac catheterization laboratory without an opportunity to complete a neuropsychologic assessment. Neuropsychologic test results obtained while patients are still hospitalized after surgery and those obtained 1 month after surgery may not be representative of cognitive decline resulting from surgery. Typically, patients receive moderate doses of narcotics for analgesia and pain relief that may affect neuropsychologic test performance, a° In studies I6,21 where patients were tested one day postoperatively, the majority were found to have a change in neuropsychologic status. It is unclear whether these results indicate a change in cognitive status or the effect of anesthesia and analgesia. Neuropsychologic testing at one year postoperatively may be influenced by the effects of other diseases on cognitive functioning. This is especially true in older populations where the incidence of Alzheimer's disease and other dementias is increased. For persons older than 70, the incidence of dementia and Alzheimer's disease is approximately 10% and increases yearly. 59,6°Because the intervals for long-term testing may differ, this could account for some of the variability seen in results.

Baseline Testing A requirement for measuring change is the ability to obtain a reliable baseline neuropsychologic assessment. Factors that may influence baseline performance include depression, anxiety, medical factors (fatigue, malaise, sleep deprivation, and pain), medications with central nervous system effects, motivational factors, and environmental factors (testing location). Since most of these factors should improve at follow-up testing, performance would be expected to improve, thereby potentially masking cognitive change induced by surgery, and thus biasing toward a null finding. These factors should be considered in the analysis of neuropsychologic change.

109

The degree of cognitive decline for a population may depend on the number of tests in a battery. It is important to know the number of tests because some investigators define change as any decline on any test. 17,19,21,30,33,35This definition is likely to include decline resulting from other extraneous factors and may not capture only those patients with clinically important cognitive changes. Numerous interventional and control group studies have not specifically defined deficit but instead have compared differences in scores between g r o u p s . 22'26"28,41"44,46 Again, this does not allow for comparison of an individual's test scores, and it does result in different methods of reporting a deficit. For example, Townes et a126 reported cognitive deficits in a patient population in two ways. In the first, they compared mean scores of a cardiac surgical group preoperatively and 7 months postoperatively and found no evidence of impairment for the group. In the second analysis, they used individual change scores and identified a substantial number of patients who declined from the baseline scores. Thus, analyzing mean scores can obscure a subgroup of individuals who experience a decline in neuropsychologic tests that may be related to surgical procedures. When standardizing a definition of cognitive change, consideration must be given to practice effect; that is the improvement in score that occurs with repeated exposure to a neuropsychologic test. 6i It is expected that postoperative scores will be higher than preoperative scores as a result of familiarity with the test. Furthermore, the factors that suppress baseline performance (poor testing conditions, high anxiety) are absent or minimal at follow-up testing. Expectation of improvement after surgery should be incorporated into change definition. In summary, issues to consider when defining cognitive change are complex. Various methods to analyze and define decline result in differing groups of patients with reported deficits. It is important to compare an individual's preoperative and postoperative test scores rather than the group as a whole, as this allows each patient to serve as his/her own control. However, this methodology introduces numerous statistical challenges. Compromise among various methods should take into account the issues of both baseline performance and practice effect.

Definition of Cognittve Change Another difficulty with comparison of neuropsychologic results in cardiac surgical populations is lack of a consistent definition of cognitive change (see Table 1). Investigators have used a variety of terms to describe neuropsychologic change, including decline, impairment, injury, disorder, abnormality, or dysfunction. It is unknown at what point changes in cognition become clinically important. Significant cognitive change has been defined as greater than one standard deviation decline from the preoperative score on two or more tests in the neuropsychologic test battery. 23 The use of a one or more standard deviation change has the disadvantage of comparing an individual's change with the group norm. This technique creates problems when analyzing scores that are low relative to the mean of the group at baseline. These scores would require a proportionately larger decrease to be classified as a decline.

S U M M A R Y AND RECOMMENDATIONS

Cognitive changes are part of the spectrum of neurologic injury suffered by some cardiac surgical patients. Issues of study design, sampling, and outcome definitions have comphcated interpretation of the reported incidence of cognitive change in these patients. These issues must be clarified in order to identify patients at high risk for cognitive complications. Design of future studies should focus on patients at high risk for cognitive decline for whom modification of surgical management may be important. A number of pharmacologic agents have the potential to minimize neuronal damage and decrease the occurrence of cognitive change. As such, the cardiac surgical patient is an ideal subject for studies of this type of intervention to improve patient outcome.

110

BOROWlCZ ET AL

REFERENCES

1 Treasure T: Overview, in Smith PL, Taylor KM (eds): Cardmc Surgery and the Brain London, Edward Arnold, 1993, pp 267-272 2. Llllehei CW. Historical development of cardiopulmonary bypass, m Gravlee GP, Davis RF, Utley JR (eds): Car&opulmonary Bypass. Principles and Practice. Baltimore, MD, Wdliams & Wilkins, 1993, pp 1-26 3. Taylor KM: Cardiac surgery and the brain: An mtroducUon, in Smith PL, Taylor KM (eds): Cardmc Surgery and the Brain London, UK, Edward Arnold, 1993, pp 1-14 4. Loop FD, Cosgrove DM, Lytle BW, et al: An 11 year evolution of coronary artery surgery (1967-1978). Ann Surg 190(4): 444-455, 1979 5. Turnipseed WD, Berkoff HA, Belzer FO: Postoperative stroke in cardiac and peripheral vascular disease. Ann Surg 192(3):365-368, 1980 6. Breuer AC, Furlan AJ, Hanson MR, et al: Central nervous system comphcatlons of coronary artery bypass graft surgery' Prospective analysis of 421 patients. Stroke 14(5):682-687, 1983 7 Coffey CE, Massey EW, Roberts KB, et al: Natural history of cerebral complications of coronary artery bypass graft surgery Neurology (Cleveland). 33' 1416-142L 1983 8. Gardner TJ, Horneffer PJ, Manolio TA, et al: Stroke followmg coronary artery bypass grafting: A ten-year study. Ann Thorac Surg 40(6)'574-581, 1985 9. Chnstakis GT, Ivanov J, Weisel RD, et al: The changing pattern of coronary artery bypass surgery. Circ 80:I-151-I-161, 1989(suppl 1) o 10. Lezak MD: Neuropsychological assessment. New York, NY, Oxford Unwersity Press, 1976 11. Bashein G, Bledsoe SW, Townes BD, et al: Tools for assessing central nervous system injury in the cardmc surgery patient, m Hdberman M (ed): Brain InJury and Protection During Heart Surgery. Boston, MA, Martinus Nijhoff, 1988, pp 109-135 12. Callahan CM, Hendrie HC, Tierney WM: Documentation and evaluation of cognitive impairment in elderly primary care patients. Ann Intern Med 122:422-429, 1995 13. Priest WS, Zaks MS, Yacorzynski GK, et al: The neurologic, psychiatric and psychologic aspects of cardiac surgery. Med Chn North Am 41:155-169, 1957 14. Gilman S: Cerebral &sorders after open-heart operations. NEJM 272(10)'489-498, 1965 15. Sotaniemi KA' Brain damage and neurological outcome after open-heart surgery. J Neurol Neurosurg Psych 43:127-135, 1980 16. Helkkmen L. Climcally significant neurological disorder following open heart surgery Thorac Cardlovasc Surgeon 33:201206, 1985 17. Calabrese JR, Skwerer RG, Gilledge AD, et al: Incidence of postoperatwe dehrium following myocardial revascularlzatlon: A prospectwe study. Cleve Chn J Med 54:29-32, 1987 18. Jackson M' Brain damage and open-heart surgery. Lancet 1096-1097, 1989 19 Klonoff H, Clark C, Kavanagh-Gray D, et al' Two-year follow-up study of coronary bypass surgery. Psychologic status, employment status, and quality of life. J Thorac Cardlovasc Surg 97'78-85, 1989 20 Stump DA, Newman SP, Coker LH, et al: Persistence of neuro-psychologlcal deficits following CABG. Anesth 73(3).A113, 1990 21 Elsass P, Henrlksen L: Acute cerebral dysfunction after open-heart surgery: A reaction-time study. Scand J Thor Cardiovasc Surg 18"161-165, 1984

22. Raymond M, Conklin C, Schaeffer J, et al" Coping with transient intellectual dysfunction after coronary bypass surgery. Heart & Lung 13(5):531-539, 1984 23. Newman S, Smxth P, Treasure T, et al: Acute neuropsychological consequences of coronary artery bypass surgery. Curr Psych Res & Rev 6(2):115-124, 1987 24. Shaw PJ, Bates D, Cartlidge NEF, et al: Neurologlc and neuropsychological morbidity following major surgery. Comparison of coronary artery bypass and peripheral vascular surgery. Stroke 18'700-707, 1987 25. Hammeke TA, Hastings JE: Neuropsychologlc alterations after car&ac operation. J Thorac Cardlovasc Surg 96.326-331, 1988 26. Townes BD, Bashein G, Hornbeln TF, et al. Neurobehavioral outcomes m car&ac operations' A prospective controlled study. J Thorac Cardiovasc Surg 98.774-782, 1989 27. Blumenthal JA, Madden D J, Burker EJ, et al: A preliminary study of the effects of cardiac procedures on cognitive performance. Int J Psychosom 38(1-4).13-16, 1991 28. O'Brlen DJ, Bauer RM, Yarandl H, et al: Patient memory before and after car&ac operations. J Thorac Car&ovasc Surg 104:1116-1124, 1992 29. Kolkka R, Hllberman M: Neurologic dysfunction following cardiac operation with low-flow, low-pressure car&opulmonary bypass. J Thorac Cardiovasc Surg 79:432-437, 1980 30. Fish KJ, Helms K, Sarnquist FH, et al' Neuropsychological dysfunction after coronary artery surgery. Anesthesiology 57(3): A55, 1982 31. Savageau JA, Stanton B, Jenkins CD, et al: Neuropsychologlcal dysfunction following electwe cardiac operation: Early assessment. J Thorac Cardiovasc Surg 84:585-594, 1982 32 Jenkins CD, Stanton B, Savageau JA, et al: Coronary artery bypass surgery' Physical, psychological, social, and economic outcomes six months later. JAMA 250:782-788, 1983 33. Aberg T, Ronquist G, TydOn H, et al: Adverse effects on the brain m cardiac operations as assessed by biochemical, psychometric, and radlologlc methods J Thorac Car&ovasc Surg 87(1).99105, 1984 34. Folks DG, Franceschlnl J, Sokol RS, et al: Coronary artery bypass surgery in older patients: Psychiatric morbidity. Southern Med J 79(3):303-305, 1986 35. Sotaniemi KA, Mononen H, Hokkanen TE' Long-term cerebral outcome after open-heart surgery: A five-year neuropsychologlcal follow-up study. Stroke 17(3):410-416, 1986 36. Nevin M, Colchester ACF, Adams S, et al. Pre&ctlon of neurological damage after cardiopulmonary bypass surgery. Use of cerebral function analyzing momtor. Anaesthesia 44'725-729, 1989 37. Toner I, Peden CJ, Hamld K, et al: Magnetic resonance xmagmg and neuropsychological changes after coronary artery bypass graft surgery' Preliminary findmgs J Neurosurg Anesth 6(3):163-169, 1994 38. Elhs R J, Wismewskl A, Potts R, et al' Reduction of flow rate and arterial pressure at moderate hypothermia does not result in cerebral dysfunction. J Thorac Cardiovasc Surg 79'173-180, 1980 39. Slogoff S, Glrgis KZ, Keats AS: Etiologic factors in neuropsychiatnc complications assocmted with cardlopulmonary bypass. Anesthesia & Analgesia 61(11):903-911, 1982 40. Garvey JW, Wdlner A, Wolpowltz A, et al: The effect of arterial filtration during open heart surgery on cerebral function. Clrc 68:11125-128, 1983(suppl 2) 41. Arls A, Solanes H, C~mara ML, et al: Arterial hne filtration during cardiopulmonary bypass; Neurologic, neuropsychologlc,

NEUROPSYCHOLOGIC CHANGE AFTER CARDIAC SURGERY

and hematologic studies. J Thorac Cardiovasc Surg 91:526-533, 1986 42. Nussmeier NA, Arlund C, SlogoffS: Neuropsychiatric complications after cardlopulmonary bypass: Cerebral protection by a barbiturate. Anesthes 64(2):165-170, 1986 43. Fish KJ, Helms KN, Sarnquist FH, et al: A prospective, randomized study of the effects of prostacyclin on neuropsychologic dysfunction after coronary artery operation. J Thorac Cardiovasc Surg 93:609-615, 1987 44. Forsman M, Olsnes BT, Semb G, et al: Effects of mmodipme on cerebral blood flow and neuropsychological outcome after cardiac surgery Brit J Anaesth 65'517-520, 1990 45. Grote CL, Shanahan PT, Salmon P, et al: Cognitiveoutcome after cardiac operations: Relationship to intraoperative computerized electroencephalographic data. J Thorac Cardiovasc Surg 104:1405-1409, 1992 46. Wong BI, McLean RF, Naylor CD, et al: Central nervous system dysfunction after warm or hypothermlc cardiopulmonary bypass. Lancet 339:1383-1384, 1992 47. Sellman M, Holm L, Ivert T, ¢.t al: A randomized study of neuropsychological function in patients undergoing coronary bypass surgery. Thorac Cardlovasc Surgeon 41:349-354, 1993 48. Pugsley W, Khnger L, Paschahs C, et al: The impact of microemboh during cardiopulmonary bypass on neuropsychological functioning. Stroke 25:1393-1399, 1994 49. Benedict RHB. Cogmtwe function after open-heart surgery: Are postoperative neuropsychologicaldeficits caused by cardiopulmonary bypass9 NeuropsychologyReview 4(3):223-255, 1994 50. Murkm JM, Martzke JS, Buchan AM, et al' A randomized study of the influence of perfusion technique and pH management strategy in 316 patients undergoing coronary artery bypass surgery: (Part 2) Neurologlc and cognitive outcomes. J Thorac Cardlovasc Surg 110.349-362, 1995 51 Butters N: Neuropsychologic issues related to coronary artery bypass grafting (CABG), m McKhann GM, Gibbs CJ (eds)' New Issues in Neurosciences: Basic and Clinical Approaches, vol 4. New York, NY, Thieme Medical Publishers, 1992/93, pp 338-341

111

52. Wllliams-Russo PG, Szatrowskl TP, Mattis S, et al: Postoperative cognitwe deterioration in cardiac vs. non-cardiac surgical patients. Perfusion (in press). 53. Heaton RK, Grant I, Matthews C: Differences m neuropsychologicaltest performance associated with age, education and sex, in Grant I, Adams KM (eds): Neuropsychological Assessment of Neuropsychiatric Disorders. New York, NY, Oxford Unwersity Press, 1986 54. Drebing CE, Van Gorp WG, Stuck AE, et al: Early detection of cognitive decline in higher cognitively functioning older adults: Sensitwityand specificityof a neuropsychologicalscreening battery. Neuropsychology8(1):31-37, 1994 55. Paulson G: What do patients with postoperatwe data tell us? The problem of fatal outcome and refusing postopertive testing: Psychiatric aspects, in Wlllner AE, Rodewald G (eds): Impact of Cardiac Surgery on the Quahty of Life. New York, NY, Plenum Press, 1990, pp 53-59 56. Murkin JM, Newman S, Stump D, et al: Statements of consensus on assessment of neurobehavioral outcomes after cardiac surgery. Ann Thorac Surg 59'1289-1295, 1995 57. Hails KC, Flnk PJ: Psychiatric problems after open heart surgery, m Kotler MN, Alfieri A, (eds): Cardiac and Noncar&ac Complications of Open Heart Surgery: Prevention, Diagnosis, and Treatment. Mount Kisco, NY, Futura Publishing, 1992, pp 299-309 58. Newman S, Klinger L, Venn G, et al: Subjectwe reports of cognition in relation to assessed cognitive performance followmg coronary artery bypass surgery J Psychosom Res 33:227-233, 1989 59. Evans DA, FunkenstelnHH, Albert MS, et al: Prevalence of Alzheimer's disease in a community population of older persons: Higher than previouslyreported. JAMA 262:2551-2556, 1989 60 Bachman DL, Wolf PA, Linn RT, et al. Incidence of dementia and probable Alzheimer's disease in a general population: The Framingham Study. Neurology 43:515-519, 1993 61. Mitrushina M, Satz P' Effect of repeated administration of a neuropsychologlcalbattery in the elderly. J Clln Psycho147(6):790801, 1991

Questions and Answers 1. What is the incidence of short-term cognitive decline after cardiac surgery? (a) < 10% (b) 10%-25% (c) 26%-79% (d) >80% 2. Interpretation of cognitive results is influenced by which of the following problems? (a) study design issues (b) changing patient populations (c) definition of deficit (d) all of the above 3. What two items are most frequently associated with cognitive change? (a) intraoperative blood pressure and perfusion flow rate (b) age & cardiopulmonary bypass time

(c) hypertension & diabetes (d) none of the above 4. What affects baseline performance? (a) anxiety (b) pain (c) testing location (d) all of the above 5. Why would you expect scores of neuropsychologic tests to increase when given at repeated intervals? (a) practice effect (b) improved testing conditions (c) a & b (d) none of the above 6. Which method of deficit definition examines individualonly patient data? (a) standard deviation (b) group comparison

12

BOROWICZ ET AL

(c) mean scores for groups (d) none of the above 7. How does the use of a surgical and nonsurglcal control group benefit the study of cognitive change in this patient population? (a) controls for age and medical history (b) controls for anesthesia and age/education (c) controls for age and gender (d) controls for medical history and anesthesia

Answers 1. C

2. 3. 4. 5. 6. 7.

d b d c a b