Issues Associated with the Identification of Cognitive Change Following Coronary Artery Bypass Grafting

Abstract

Objective: Coronary artery bypass grafting (CABG) is a surgical procedure used to treat individuals with ischaemic heart disease and to relieve angina. Disruption to the central nervous system (CNS) has frequently been reported by patients who have undergone CABG.

Method: The following paper is a review of the literature that has examined the effects of CABG on the CNS.

Results and Conclusions: It becomes apparent that issues about the incidence and severity of post-CABG cognitive decline are still unresolved. First, the cause of post-CABG CNS change has not yet been established, although the presence of changes to brain microvasculature as a result of the presence of microemboli appears to be a likely factor. Second, while some studies have reported high rates of poor performance on neuropsychological tests postoperatively, these reports are often subject to confounds such as variability in postoperative testing intervals, the definition of decline and the neuropsychological test batteries used. Finally, improvements in surgical techniques and changes in patient characteristics may have changed the real nature and prevalence of post-CABG cognitive decline. The review finishes with a series of recommendations for the neuropsychologlical study of CABG.

Keywords

cognitive decline coronary artery bypass grafting microemboli neuropsychological testing

Coronary artery bypass grafting (CABG) is a surgical procedure used to treat individuals with ischaemic heart disease and to relieve angina [1]. The number of patients undergoing CABG has risen dramatically in recent years, to the extent that it is now the most commonly performed major surgical procedure in the USA [2]. Furthermore, it has been estimated that each year over 800 000 myocardial revascularisation procedures are conducted worldwide [3,4]. This is perhaps in part attributable to advances in cardiac surgery which have succeeded in reducing the risk of mortality associated with the procedure [5].

Coronary artery bypass grafting, mortality and morbidity

As the rate of mortality has been reduced, issues of morbidity have become increasingly more important, particularly when evaluating the relative risks and benefits of the procedure for the individual. Furthermore, issues of morbidity are now of greater importance as they have an essential role in the quest to improve surgical outcomes and the quality of life of patients.

One major type of morbidity following CABG is central nervous system (CNS) dysfunction [6–8]. Of all the adverse neurological outcomes that may be incurred postoperatively, stroke is one of the most serious. However, due to technological and surgical improvements the incidence of stroke is now reported to be as low as between 0.8 and 5.8% [9]. Consequently, the rate of post-CABG stroke is no longer a sufficient index of CNS dysfunction. Neuropsychological research suggests, however, that a considerable proportion of all patients who undergo CABG sustain some degree of cerebral damage and that this manifests as mild cognitive impairment. Although these cognitive deficits rarely disturb activities of daily living, they are still considered cause for concern. Therefore, it is these less severe forms of neurologic injury which are now targeted for reduction in what has been described as an age of quality improvement [2,10,11].

Cognitive impairment following coronary artery bypass grafting

Neuropsychological tests are valuable tools in the assessment of brain dysfunction as they provide a method of systematically and quantitatively studying the behavioural expressions of this dysfunction [12]. As there is now only a low risk of stroke following CABG, milder forms of cerebral damage have become a greater focus of concern. Consequently neuropsychological assessment has become more important within the domain of cardiac surgery. The advantage of neuropsychological tests is that they are capable of detecting subtle changes in cognitive function. In comparison, conventional neurological assessment techniques, such as the Mini-Mental State Examination, are less sensitive and therefore less able to detect subtle CNS changes [10,13]. In addition, neurological assessment techniques do not lend themselves as readily to quantitative analysis [14].

A large body of research exists supporting the suggestion that cognitive decline occurs after CABG (see Table 1 for a summary). Cognitive decline has been observed by many researchers using batteries of neuropsychological tests, usually administered to patients before and after surgery. A patient's pre- and postoperative scores are then compared. In this way, intersubject variability is minimised as the subjects act as their own controls. While cognitive deficits have been consistently reported in the immediate postoperative period, some researchers have re-administered test batteries in the immediate postoperative period, typically within 5–10 days of surgery [7,9,15–20]. Others have reassessed patients over longer time periods ranging from 1 month [9] to 5 years after surgery [21,22].

Table 1.

Summary table of neuropsychological studies that have assessed cognitive function following coronary artery bypass grafting (CABG) using neuropsychological tests (chronological order)

A wide variety of neuropsychological tests have been used to assess the nature, incidence and severity of cognitive deficits following CABG. Table 2 summarises the various neuropsychological tests used in the various studies to date. From this table, there appear to be three major types of tests that are sensitive to changes in cognitive function following CABG, either by themselves or as a part of broader test batteries. These tests assess aspects of learning and memory (both verbal and non-verbal), complex attention and psychomotor function. Deficits in memory, attention and psychomotor skills accord well with the perceived cognitive changes and suggest that specific central nervous system networks may be at increased risk for damage following CABG. Consistent with this observation, Murkin et al [23] recently recommended that future neuropsychological research in this area should include the Rey Auditory Verbal Learning Test (RAVLT), the Trail Making Test (parts A and B), and the Grooved Pegboard Test. They stated that these tests have been found consistently to be sensitive to subtle cognitive changes following CABG. Therefore the use of a common battery of sensitive tests will allow direct comparison between studies on the incidence of CABG-related cognitive decline. The sensitivity of these tests may also reflect their popularity in longitudinal studies due to their ease of administration, possession of alternate forms and wide range of possible scores. However, the majority of these tests can be considered as requiring multiple cognitive processes for successful performance. Therefore, disruption to different cognitive systems may present as the same performance deficit.

Table 2.

Summary of the neuropsychological tests used in coronary artery bypass grafting (CABG) studies and their sensitivity to any CABG-related cognitive changes

This poor neuropsychological validity means that such tests cannot be used to help localise the major site for central nervous system disruption following CABG. There are also some interesting omissions from neuropsychological studies of CABG. For example, few studies have investigated language, visuo-constructional or executive function in these patients. Given that large parts of the cortex are involved in the processing of these functions, they should be disrupted very easily by small CABG-related insults. Thus, there appears to be no good evidence for any specific CABG-related neuro-psychological syndrome.

Suggested aetiology of cognitive impairment

Although the cause of cognitive decline after CABG is likely to be multifactorial [24], two processes appear preferentially involved. The first process centres around the suggestion that cerebral damage arises as a consequence of inadequate cerebral perfusion during surgery which leads, in turn, to ischaemic cell injury. This may occur for a number of reasons, such as if the patient has cerebrovascular occlusive disease or hypotension. Other factors such as the incorrect placement of the cannula and low flow have also been implicated [25,26]. The second process that has been identified as a possible cause of postoperative cerebral damage is linked to the accumulation of microemboli in the small blood vessels of the brain. Microemboli are small articles of debris that are usually of gaseous or particulate composition. Air microemboli may be formed when air bubbles, generated by the cardiopulmonary bypass (CPB) machine, become trapped in the blood [18,27–30]. The origins of particulate microemboli are commonly the fatty plaques (atheroma) which line the arterial walls of individuals with atherosclerosis [6,25,31]. During CABG the aorta is subjected to a series of manipulations, such as clamping and cannulation which may dislodge microemboli from atheromatous material which has accumulated on the walls of the aorta [32]. Once free, microemboli are readily transported via the blood around the body. Cerebral damage may occur when microemboli travel up the carotid arteries and lodge in the small blood vessels of the brain. As a result the blood flow to surrounding regions may be obstructed and ischaemic cell injury may occur [25].

The microemboli theory of post-CABG cognitive decline is supported by the development of techniques such as transcranial Doppler ultrasonography (TCD) and magnetic resonance imaging (MRI) to assess the consequences of CABG. This research supports the hypothesis that microemboli dislodged during CABG may contribute to post-CABG cognitive decline [16,18,33]. As a result the focus of research has shifted away from the hypoperfusion theory [6].

Evidence from transcranial Doppler ultrasonography studies

Transcranial Doppler ultrasonography enables researchers to non-invasively detect and quantify the number of microemboli passing through cerebral arteries. Since its inception, many researchers have used TCD to detect microemboli in cerebral circulation at various stages of CABG. Although there is some disagreement, perhaps due to the use of different CABG surgical techniques, there is a degree of consensus that the highest counts of microemboli occur during clamp removal, cannulation, cardiac manipulation and at the inception of bypass [6,16,18,30,33–36]. A positive correlation has also been found between microembolic load during CABG and the occurrence of cognitive decline. Patients with higher counts of microemboli during surgery have also been found to exhibit more signs of cognitive decline prior to discharge from hospital and 8 weeks after surgery [16,18,37]. The suggestion that there may be a link between microemboli and cognitive decline is also supported by the finding that when an arterial line filter was in place during surgery to lower the number of microemboli in the blood, the incidence of cognitive decline was reduced [18].

However, one study [24] failed to detect a correlation between cognitive decline, changes in global glucose metabolism and the absolute number of TCD signals. One of the limitations of TCD, which may account for these findings is TCD may only detect around 0.1–1.0% of all microemboli dislogged during CABG. The majority of microemboli may be simply too small to detect and therefore, TCD may only provide a relative measure of the microembolic load [38,39]. Furthermore the main determinant of microemboli-related cognitive decline may be the CNS location in which such emoboli are lodged rather than the absolute embolic load delivered to the brain. In addition, the impact of a lesion on behaviour may not be directly proportional to the amount of tissue damaged [10,24].

Evidence from magnetic resonance imaging studies

Small ischaemic lesions have been detected in the brains of patients after CABG using MRI. These lesions are thought to have been caused by microemboli and researchers have sought to establish whether these morphological changes correlate with the occurrence of cognitive decline [40,41]. The results however, have been mixed, and although Toner et al. [40,41] observed a correlation, Schmidt et al. [42] did not, as no ischaemic lesions were detected. In contrast, Vanninen et al. [2] did detect new lesions, yet there was no evidence of cognitive decline or neurological changes in these patients.

The issue of whether or not there is a correlation between cognitive decline and ischaemic lesions is also complicated by technological advances over time which have improved the spatial resolution of MRI. Researchers today are capable of detecting smaller lesions than they were 5 or 10 years ago. Given that lesions induced by microemboli are likely to be only very small, the resolving capacity of the MRI machine is critically important in their detection. The implication of these technological advances is that it is difficult to compare studies, and it is possible that Schmidt et al. [42] were unable to detect any ischaemic lesions simply because the imaging technology used may have lacked spatial resolving capacity.

On the basis of TCD and MRI evidence it seems probable that microemboli play a major causal role in cognitive decline following CABG. However, as Vanninen et al. [2] was unable to observe any statistically significant deteriorations in cognitive performance following CABG, despite finding evidence of new ischaemic lesions, this serves as a reminder that the clinical significance of microemboli has still not been established definitively. Similarly, in a recent paper, Jacobs et al. [24] found that alterations in cognitive function and cerebral glucose metabolism were not solely related to microembolic events. It was concluded that the processes that contribute to postoperative CNS dysfunction are likely to be complex and multifarious, and the influence of other factors such as postoperative brain swelling, the effects of anaesthesia and differences in vascular anatomy should not be dismissed entirely [24].

Patterns of cerebral damage

According to the microemboli theory, a diffuse pattern of cerebral damage is expected following CABG, as cerebral circulation distributes microemboli widely and relatively randomly throughout the brain. Results from neuropsychological research support this suggestion as each individual appears to display a unique pattern of cognitive decline following CABG and there have been no broad neuropsychological syndromes identified on the basis of large studies (see Table 2). The incidence of cognitive decline has also been found to be reduced several months following surgery (see Table 1). This finding is consistent with the microemboli theory of cognitive decline, as it is proposed that the plastic nature of the brain enables recovery from microembolic damage [43].

Imaging studies have also revealed that ischaemic lesions are diffusely distributed throughout the brain, and that patterns of cerebral damage differ between individuals [24,41]. There is also evidence to suggest that damage may occur more frequently in the right hemisphere due to the anatomical arrangement of the cerebral vascular system. Neuropsychological test results, however, have not reflected this asymmetry [10,24]. Finally, as predicted by the microemboli theory, any cerebral damage sustained following CABG does not appear to be specific to the cortex or subcortex. For example, Moody et al. [38] observed more lesions in the cortical and deep nuclear grey matter of the brain, whereas Vanninen et al. [2] detected more lesions in the white matter.

Incidence of cognitive impairment following coronary artery bypass grafting

While there is a general consensus that patients experience cognitive decline following CABG, agreement is lacking regarding the actual percentage impaired. Some researchers have reported cognitive deficits in nearly 80% of patients [19,37], others claim to have detected deficits in 60% [18] and others in only around 30% of patients [27,44]. By long-term follow-up assessments, the incidence of cognitive decline is generally found to be much lower (Table 1). As in the short-term follow-up interval, there is also a lack of consensus regarding the number who recover and the number who remain impaired at long-term follow-up. Some researchers report there are no lasting detrimental effects of CABG [1,7,20,45,46]. Others disagree, for example Pugsley et al. [18] observed 17% of patients exhibited signs of cognitive decline at 8 weeks. Savageau et al. [47] also found cognitive decline in 20% of patients 6 months after CABG. In another study, the incidence of cognitive decline at 6 months was found to be much higher. It was reported that 40% of patients remained significantly impaired [48]. When reviewing the literature it soon becomes apparent that it is exceedingly difficult to make a general statement regarding the incidence or duration of cognitive deficits following CABG. This lack of consensus is the result of great variation between studies particularly in the definition of decline, the number and type of tests included in test batteries, the timing of testing and patient characteristics [9,49].

Problems with the definition of decline

The concept of decline is of central importance in research of this nature, yet no single, standard criterion has been consistently adopted by researchers. This is particularly problematic as different criteria identify cognitive decline in different individuals and therefore alter any estimates of the prevalence of this decline. In addition, these changing criteria make it difficult to evaluate the severity of post-CABG cognitive decline. These inconsistencies also complicate any attempt to investigate the clinical significance of such deficits. Furthermore, the reason for the selection of particular criteria is often idiosyncratic and arbitrary [49–51].

A large number of researchers have adopted the ‘one standard deviation’ definition of decline. This criterion identifies cognitive decline as any instance when a patient's postoperative performance on a neuropsychological test declines by more than one standard deviation from the baseline distribution of scores on that test. There are, however, several limitations associated with this method. First, there is the problem of a ‘floor effect’ where it simply may not be possible for some patients with very low baseline scores to decline by one standard deviation. In a study by Mahanna et al. [49] it was found that 35% of patients were misclassified because of this effect. Secondly, patients with high scores require a proportionately larger drop from baseline before they are considered to have declined [51]. Another limitation associated with this method is that cross-study comparisons are difficult as estimates of standard deviations vary considerably across groups [49].

To add further confusion, studies have also varied in the number of tests that a patient's score must have declined on before a subject can be said to have declined. This has varied from a one or more standard deviation decrease on only one test, while others require decline on two or more tests and others state that decline must be observed on three or more tests (Table 1). This design is also complicated because studies differ considerably in the number of neuropsychological tests used. A criterion of decline on two tests becomes easier to satisfy as the number of tests increases [52]. For a battery of 20 neuropsychological tests a definition that insists upon a decrease in two tests will be less stringent than the same definition of decline applied to a battery of five tests.

Recently an alternative definition of decline using a 20% change criterion has been adopted by a number of researchers (e.g. [5,33,53]). Using this method a decrease in scores by 20% from baseline on 20% of tests is said to constitute significant decline. After comparing the different indices of decline, Mahanna et al. [49] considered this to be the most sensitive method for identifying patients with signs of cognitive decline, as this approach overcomes the problems associated with floor effects. Thus, this criterion, unlike the standard deviation criterion, can identify cognitive decline in patients with low baseline scores. Furthermore, results obtained using this criterion can be generalised to other studies. Low scores, however, remain problematic, as a minor change in performance will have a greater impact upon the score of a patient with a low score, compared with the same degree of change for someone with a high score.

Testing intervals

Researchers have conducted follow-up neuropsychological assessments across a range of different testing intervals. It is likely that this has also contributed to the great variability in reports of the incidence of postoperative cognitive decline. Furthermore, there are a number of limitations associated with each testing interval that must be considered. Preoperative testing is frequently conducted the night before patients undergo surgery. At this interval limited patient availability is usually the greatest difficulty encountered, as patients must undergo a variety of preoperative examinations and interviews with medical staff. Patients may also have elevated anxiety levels.

Short-term postoperative assessment is the most controversial testing interval, because it is difficult, if not impossible to separate the potential effects of microembolic damage from the effects of other postoperative factors also shown to influence performance on neuropsychological tests. During this interval patients usually receive sizeable doses of narcotics and benzodiazepines for anaesthesia during surgery and later for analgesia and sedation postoperatively. Patients may also show significant postoperative pain or fatigue at this interval [20]. Long-term follow-up assessments have been made as early as 1 month [9], and as late as 5 years [21]. The greatest difficulty when attempting to determine the most appropriate time for this assessment is that it is unclear when stabilisation from possible cerebral insult has occurred, however, this issue of when it is best to conduct long-term follow-up assessments is critically important. During the acute and post-acute stages following cerebral insult, the brain may be in a period of active recovery [23,43]. Consequently any neuropsychological assessment made during this interval is not expected to reflect the long-term neuropsychological status of patients [12]. Furthermore, the process of recovery may be influenced by variability in the size and location of the lesion, as well as by the individual's health and other sociocultural factors [43].

Physical recovery from CABG is generally considered to be complete within 3 months, and in their ‘Statement of Consensus’, Murkin et al. [23] recommend that this is also the optimum time for the reassessment of cognitive function. In accordance with this recommendation, it follows then that the reassessment of cognitive function at intervals beyond 3 months may be problematic. For example, if an assessment was extended beyond a year after the operation it is difficult to be sure that any signs of subtle cognitive decline are not the result of other factors unrelated to the operation. Of particular concern, are neurodegenerative diseases (such Alzheimer's disease and Parkinson's disease) the prevalance of which increases in older individuals and which may be exacerbated by surgery requiring a general anaesthetic [54]. Some researchers have, however, reported new instances of cognitive decline at longterm follow-up assessments and suggested that this may be evidence of a delayed response to injury [9,47,48]. Yet, given that cognitive decline following CABG is often only subtle and that there is no consistent, domain-specific and therefore readily identifiable pattern of cognitive decline, it is highly possible that these findings may reflect changes not the result of the CABG procedure.

Patient characteristics

Increasing age has been identified as an incremental risk factor for cognitive decline following CABG, especially in patients over 70 years of age [2,7,17,20,33,53,55]. It would seem likely then that some of the variability between studies may be due to age differences between samples, particularly as some studies have excluded patients older than 70 years (e.g. [15,19,56]). Populations of CABG patients also appear to be changing, which makes comparisons between studies difficult. Since 1974 the average age of patients undergoing CABG has increased from 52 years to 65 years of age in 1996 [8,51]. This trend is a reflection of advances in cardiac surgery and patient management that have now made it safer for older individuals to undergo CABG. Indeed, the situation is paradoxical, older patients benefit most from surgery as they are more likely to have advanced coronary artery disease, yet as a population they are at the greatest risk of postoperative morbidity and mortality. One reason that has been advanced for this is that older patients are also more likely to have advanced atherosclerosis of the aorta. In addition, older patients are also more likely to present with a greater number of comorbid risk factors, such as diabetes and hypertension [50].

As a consequence of the advances in surgical techniques, patients undergoing CABG today are generally older and sicker than they were several years ago. Similarly, surgical techniques and instruments have also been extensively modified (for example arterial line filters and membrane oxygenators have been introduced) [15,34]. These factors reduce the validity of any comparisons made between studies of cognitive function following CABG from different years.

Generally, it appears that age-related rates of decline differ with different cognitive abilities [12]. In a repeated measures study design, patients act as their own controls and, provided follow-up intervals are reasonably short, age should not influence performance. It is only when follow-up intervals are more extended that age may become a potentially confounding factor.

Education level has also been examined and found to be inversely related to the risk of cognitive decline [57]. However, education level is seldom systematically controlled or even reported in CABG studies. The influence of mood, specifically depression and anxiety on patients performances on neuropsychological tests has also been considered. An early report suggested depression, as measured by the Centre for Epidemiologic Studies-Depression Self Rating Scale, was an incremental risk factor significantly associated with the development of cognitive deficits following surgery. Anxiety and psychosocial functioning were not found to be related to the outcome [57]. However, cognitive function was only assessed using the Mini-Mental State Examination and it is uncertain whether the deficits reported were truly the result of cerebral insult or lack of patient motivation. In a more recent study, no correlation was found between depression and cognitive decline. Patients depressed before surgery were also more likely to be depressed after surgery. It was concluded that the reduction in test performance observed after CABG was not a consequence of depression [58]. Using Hammon depression scale ratings, Townes et al. [20] also concluded depression did not account for impairments in performance, and nor could improvements at follow-up testing be attributed to a decrease in anxiety.

Conclusion: The assessment of cognitive decline following coronary artery bypass grafting

This review indicates that questions about the incidence and severity of post-CABG cognitive decline are still unresolved. While some studies have reported high rates of poor performance on neuropsychological tests postoperatively, these results may have been confounded by variability in postoperative testing intervals, the definition of decline and the neuropsychological test batteries used. In addition, improvements in surgical techniques and changes in patient characteristics may have altered the extent to which cognitive abnormalities occur following CABG, consequently these changes may limit the comparisons which can be made between a study published, for example, in 1988 and one published in 1998. Moreover, the nature of CNS damage after CABG has not been established, although microembolic damage seems to be the most likely factor.

At a more practical level, the development of a neuropsychological test battery to assess cognitive function following CABG involves a series of compromises. Firstly, limited patient availability the night before surgery, as well as pain and fatigue in the short-term follow-up period, restricts the length of the battery. Yet conversely, a test battery must be sufficiently detailed to detect those patients exhibiting signs of cognitive decline. Second, it is also preferable to include tests to assess performance across a wide range of cognitive domains as diffuse damage is expected following CABG and because individuals may be affected differently [9,10,59]. Due to the need for brevity, however, it is not possible to conduct exhaustive interviews with patients to establish with greater specificity the cognitive functions or brain systems involved. Finally, the tests also must be able to be administered at the bedside to enable testing in the short-term follow-up interval. Therefore, the difficulty which has confronted researchers has been to develop a test battery able to assess a number of cognitive domains, while also managing to balance issues of assessment depth and administration time [10,13,50,51].

Some of these issues have already been encountered by researchers examining cognitive decline in patients with neurodegenerative diseases (e.g. [60,61,62,63]). This research has provided a number of useful tools and recommendations which can be applied to studies investigating cognitive decline following CABG. For example, in seeking to develop a sensitive, valid and reliable index to distinguish the cognitive change associated with Alzheimer's disease (a cortical pathology) from that associated with normal ageing, the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) developed a short battery of neuropsychological tests. These tests have minimal practice effects, are appropriate for use with an aged sample and also have low floor and high ceiling effects. They also take only a short time to administer and do not require any special apparatus [60,64].

Another similar battery has been developed by the Multicentre AIDS Cohert Study (MACS) group, although the aim of this battery was to detect the subtle cognitive changes which are associated with the sub-cortical pathology of HIV dementia (HIV-D). The tests in this battery are minimally affected by factors related to fatigue, depression and anxiety that commonly occur in advanced stages of HIV infection [61,65].

Some of the tests included in the CERAD and MACS batteries have already been used in studies investigating cognitive decline in CABG patients. Accordingly, the neuropsychological tests included in the CERAD and MACS batteries include all those suggested recently by Murkin et al. [23] for use in CABG.

Finally, changes in mood do occur following CABG, and these influence performance on neuropsychological tests [20,57]. However, the assessment of mood often requires that patients complete a lengthy battery of detailed and time-consuming questionnaires. Therefore, the assessment of mood may also need to be restructured so that measures are less detailed, but more rapid. This has been achieved in the past with the use of visual analogue mood scales [66].

References

1. Klonoff

Clark

Kavanagh-Gray

Mizgala

Munro

. Two-year follow up study of coronary artery bypass surgery. Psychologic status, employment status and quality of life. Journal of Thoracic and Cardiovascular Surgery 1998; 97: 78–85.

2. Vanninen

Aikia

Kononen

. Subclinical cerebral complications after coronary artery bypass grafting: prospective analysis with magnetic resonance imaging, qualitative electroencephalography and neuropsychological assessment. Archives of Neurology 1998; 55: 618–627.

3. Newman

Reves

. Toward a new frontier in cardiac surgery. Annals of Thoracic Surgery 1997; 63: 322–323.

4. Roach

Kanchuger

Mangano

. Adverse cerebral outcomes after coronary artery bypass surgery. New England Journal of Medicine 1996; 335: 1857–1863.

5. Stump

Rogers

Hammon

Newman

. Cerebral emboli and cognitive outcome after cardiac surgery. Journal of Cardiothoracic and Vascular Anaesthesia 1996; 10: 113–119.

6. Barbut

Hinton

Szatrowski

. Cerebral emboli detected during bypass surgery are associated with clamp removal. Stroke 1994; 25: 2398–2402.

7. McLean

Wong

Naylor

. Cardiopulmonary bypass, temperature and central nervous system dysfunction. Circulation 1994; 90: 250–255.

8. Gardner

Horneffer

Manolio

. Stroke following coronary artery bypass grafting: a ten year study. Annals of Thoracic Surgery 1985; 40: 576–581.

9. McKhann

Goldsborough

Borowicz

. Cognitive outcome after coronary artery bypass: a one year prospective study. Annals of Thoracic Surgery 1997; 63: 510–515.

10.

10. Stump

. Selection and clinical significance of neuropsychologic tests. Annals of Thoracic Surgery 1995; 59: 1340–1344.

11.

11. Stump

Rogers

Hammon

. Neurobehavioural tests are monitoring tools used to improve cardiac surgery outcome. Annals of Thoracic Surgery 1996; 61: 1295–1296.

12.

12. Lezak

. Neuropsychological Assessment3rd edn. Oxford University Press, New York 1995.

13.

13. Newman

. Analysis and interpretation of neuropsychologic tests in cardiac surgery. Annals of Thoracic Surgery 1995; 59: 1351–1355.

14.

14. Heyer

Delphin

Adams

. Cerebral dysfunction after cardiac operations in elderly patients. Annals of Thoracic Surgery 1995; 60: 1716–1722.

15.

15. Aris

Solanes

Camara

Junque

Escartin

Caralps

. Arterial line filtration during cardiopulmonary bypass. Journal of Thoracic and Cardiovascular Surgery 1986; 91: 526–533.

16.

16. Clark

Brillman

Davis

Lovell

Price

TRP

Magovern

. Microemboli during coronary artery bypass grafting. Journal of Thoracic and Cardiovascular Surgery 1995; 109: 249–258.

17.

17. Newman

Croughwell

Blumenthal

. Effect of aging on cerebral autoregulation during cardiopulmonary bypass – association with postoperative cognitive dysfunction. Circulation 1994; 90: 243–249.

18.

18. Pugsley

Klinger

Paschalis

Treasure

Harrison

Newman

. The impact of microemboli during cardiopulmonary bypass on neuropsychological functioning. Stroke 1994; 25: 1393–1409.

19.

19. Shaw

Bates

Cartlidge

NEF

. Early intellectual dysfunction following coronary bypass surgery. Quarterly Journal of Medicine, New Series 1986; 58: 59–68.

20.

20. Townes

Bashein

Hornbein

. Neurobehavioural outcomes in cardiac operations – a prospective controlled study. Journal of Thoracic and Cardiovascular Surgery 1989; 98: 774–782.

21.

21. Sotamiemi

Mononen

Hokkanen

. Long-term cerebral outcome after open heart surgery. A five year neuropsychological follow-up study. Stroke 1986; 17: 410–416.

22.

22. Hlatky

Bacon

Boothroyd

. Cognitive function 5 years after randomisation to coronary agnioplasty or coronary artery bypass graft surgery. Circulation 1997; 96(Suppl. II)II-11–II-15.

23.

23. Murkin

Newman

Stump

Blumenthal

. Statement of consensus on assessment of neurobehavioural outcomes after cardiac surgery. Annals of Thoracic Surgery 1995; 59: 1289–1295.

24.

24. Jacobs

Neveling

Horst

. Alterations of neuropsychological function and cerebral glucose metabolism after cardiac surgery are not related only to intraoperative microembolic events. Stroke 1998; 29: 660–667.

25.

25. Gilman

. Neurological complications of open heart surgery. Annals of Neurology 1990; 28: 475–476.

26.

26. Shaw

Bates

Cartlidge

NEF

Heaviside

Julian

Shaw

. Early neurological complications of coronary artery bypass surgery. British Medical Journal 1985; 291: 1384–1387.

27.

27. Blauth

Arnold

Schulenberg

McCartney

Taylor

. Cerebral microembolism during cardiopulmonary bypass. Retinal microvascular studies in vivo with fluorescein angiography. Journal of Thoracic and Cardiovascular Surgery 1998; 95: 668–676.

28.

28. Miller

Keane

. Encyclopedia and dictionary of medicine, nursing and allied health3rd edn. W.B. Saunders, Philadelphia 1983.

29.

29. Mills

. Risk factors for cerebral injury and cardiac surgery. Annals of Thoracic Surgery 1995; 59: 1296–1299.

30.

30. Padayachee

Parsons

Theobold

Linely

Gosling

Deverall

. The detection of microemboli in the middle cerebral artery during cardiopulmonary bypass: a transcranial doppler ultrasound investigation using membrane and bubble oxygenators. Annals of Thoracic Surgery 1987; 44: 298–302.

31.

31. Belden

Caplan

Bojar

Payne

Blackman

. Multiple brain emboli from an ulcerated thrombogenic aorta and its treatment with aortectomy and graft replacement. Neurology 1996; 46: A257.

32.

32. Royse

Royse

Blake

Grigg

. Assessment of thoracic aorta atheroma by echocardiography: a new classification and estimation of risk of dislodging atheroma during three surgical techniques. Annals of Thoracic and Cardiovascular Surgery 1998; 4: 72–77.

33.

33. Hammon

Stump

Kon

. Risk factors and solutions for the development of neurobehavioural changes after coronary artery bypass grafting. Annals of Thoracic Surgery 1997; 63: 1608–1612.

34.

34. Pugsley

. The use of doppler ultrasound in the assessment of microemboli during cardiac surgery. Perfusion 1989; 4: 115–122.

35.

35. Sylivris

Levi

Matalanis

Buxton

Mitchell

Tonkin

. The pattern and significance of cerebral microemboli during coronary artery bypass surgery. Paper presented at the Proceedings of the Australian Cardiology Society, MelbourneAustralia, April, 11–131996.

36.

36. van der Linden

Casimir-Ahn

. When do cerebral emboli appear during open heart operations? A transcranial doppler study. Annals of Thoracic Surgery 1991; 51: 237–241.

37.

37. Stump

Tegeler

Rogers

. Neuropsychological deficits are associated with the number of emboli detected during cardiac surgery. Stroke 1993; 24: 509.

38.

38. Moody

Bell

Challa

Johnston

Prough

. Brain microemboli during cardiac surgery or aortography. Annals of Neurology 1990; 28: 477–486.

39.

39. Moody

Brown

Challa

Stump

Reboussin

Legult

. Brian microemboli associated with cardiopulmonary bypass: a histologic and magnetic resonance imaging study. Annals of Thoracic Surgery 1995; 59: 1304–1307.

40.

40. Toner

Peden

Hamid

Newman

Taylor

Smith

. Magnetic resonance imaging and P300 (event-related auditory evoked potentials) in the assessment of postoperative cerebral injury following coronary artery bypass graft surgery. Perfusion 1995; 8: 321–329.

41.

41. Toner

Peden

Hamid

Newman

Taylor

Smith

. Magnetic resonance imagining and neuropsychological changes after coronary artery bypass graft surgery: preliminary findings. Journal of Neurosurgery and Anaesthesiology 1994; 6: 163–169.

42.

42. Schmidt

Fazekas

Offenbacher

. Brain magnetic resonance imaging in coronary artery bypass grafts: a pre-and postoperative assessment. Neurology 1993; 43: 775–778.

43.

43. Kertesz

. Recovery and treatment, Recovery and treatment. Clinical neuropsychololgy3rd edn., Heilman

Valenstein

. Oxford University Press, New York 1993; 647–674.

44.

44. Savageau

Stanton

Jenkins

Klein

. Neuropsychological dysfunction following elective cardiac operation – early assessment. Journal of Thoracic and Cardiovascular Surgery 1982; 84: 585–594.

45.

45. Fish

Hellms

Sarnquist

. A prospective, randomised study of the effects of prostacyclin on neuropsychologic dysfunction after coronary artery operation. Journal of Thoracic and Cardiovascular Surgery 1987; 93: 609–615.

46.

46. Smith

PLC

Treasure

Newman

. Cerebral consequences of cardiopulmonary bypass. Lancet 1986; 355: 823–825.

47.

47. Savageau

Stanton

Jenkins

Franter

. Neuropsychological dysfunction following elective cardiac operation – a six month reassessment. Journal of Thoracic and Cardiovascular Surgery 1982; 84: 595–600.

48.

48. Shaw

Bates

Cartlidge

NEF

. Long-term intellectual dysfunction following coronary artery bypass graft surgery: a six month follow-up study. Quarterly Journal of Medicine, New Series 1987; 62: 259–268.

49.

49. Mahanna

Blumenthal

White

. Defining neuropsychological dysfunction after coronary artery bypass grafting. Annals of Thoracic Surgery 1996; 61: 1342–1347.

50.

50. Blumenthal

Mahanna

Madden

White

Croughwell

Newman

. Methodological issues in the assessment of neuropsychologic function after cardiac surgery. Annals of Thoracic Surgery 1995; 59: 1345–1350.

51.

51. Borowicz

Goldsborough

Selnes

McKhann

. Neuropsychologic change after cardiac surgery: a critical review. Journal of Cardiothoracic and Vascular Anaesthesia 1996; 10: 105–112.

52.

52. Ingraham

Aitken

. An empirical approach to determining criteria for abnormality in test batteries with multiple measures. Neuropsychology 1996; 10: 120–124.

53.

53. Heyer

Adams

Delphin

. Cerebral dysfunction after coronary artery bypass grafting done with mild or moderate hypothermia. Journal of Thoracic and Cardiovascular Surgery 1997; 114: 270–277.

54.

54. Whitehouse

Lerner

Hedera

. Dementia. Clinical neuropsychololgy3rd edn., Heilman

Valenstein

. Oxford University Press, New York 1993; 409–460.

55.

55. Stump

Newman

Coker

Wallenhaupt

Roy

. The effect of age on neurologic outcome after cardiac surgery. Anaesthesia and Analgesia 1992; 74: SI–S386.

56.

56. Shaw

Bates

Cartlidge

NEF

. Neurologic and neuropsychological morbidity following major surgery: a comparison of coronary artery bypass and peripheral vascular surgery. Stroke 1987; 18: 700–707.

57.

57. Folks

Freeman

Sokol

Govier

Reves

Baker

. Cognitive dysfunction after coronary artery bypass surgery: a case controlled study. Southern Journal of Medicine 1988; 81: 202–206.

58.

58. McKhann

Borowicz

Goldsborough

Enger

Selnes

. Depression and cognitive decline after coronary artery bypass grafting. Lancet 1997; 349: 1282–1284.

59.

59. Harrison

MJG

. Neurologic complications of coronary artery bypass grafting: diffuse or focal ischemia?. Annals of Thoracic Surgery 1995; 59: 1356–1358.

60.

60. Koss

Edland

Fillenbaum

. Clinical and neuropsychological differences between patients with earlier and later onset of Alzheimer's disease: a CERAD analysis part XII. Neurology 1996; 46: 136–141.

61.

61. Selnes

Galai

Bacellar

. Cognitive performance after progression to AIDS: a longitudinal study from the Multicentre AIDS Cohort Study. Neurology 1995; 45: 267–275.

62.

62. Steenhuis

Ostbye

. Neuropsychological test performance of specific diagnostic groups in the Canadian study of health and aging (CSHA). Journal of Clinical and Experimental Neuropsychology 1995; 17: 773–785.

63.

63. Van Gorp

Lamb

Schmitt

. Methodologic issues in neuropsychological research with HIV-spectrum disease. Archives of Clinical Neuropsychology 1993; 8: 17–33.

64.

64. Morris

Heyman

Mohs

. The consortium to establish a registry for Alzheimer's disease (CERAD). Part 1. Clinical and neuropsychological assessment of Alzheimer's disease. Neurology 1989; 39: 1159–1165.

65.

65. Concha

Selnes

McArthur

. Normative data for a brief neuropsychologic test battery in a cohort of injecting drug users. International Journal of the Addictions 1995; 30: 823–841.

66.

66. Maruff

Wood

Currie

McArthur-Jackson

Malone

Benson

. Computer-administered visual analogue mood scales: rapid and valid assessment of mood in HIV positive individuals. Psychological Reports 1994; 74: 39–42.