|
|
||||||||
Ann Thorac Surg 1995;59:1351-1355
© 1995 The Society of Thoracic Surgeons
Department of Psychiatry, University College London Medical School, London, United Kingdom
Abstract
Neuropsychologic assessment in the context of cardiac surgery is constrained by the clinical time available to see patients, and as such test selection needs to be carefully planned. Besides the time limitations, it differs from clinical neuropsychologic assessment primarily because it tends to involve at least two assessments, a comparison with performance before operation, and a limited number of tests. The analysis and design issues involved in both incidence studies and intervention studies are discussed in this article. Incidence studies customarily involve a single group assessed on at least two occasions and are designed to identify the numbers of individuals who show clear evidence of neuropsychologic changes after a cardiac operation. Intervention studies involve at least two groups where one factor (eg, surgical equipment) is varied systematically. The research on neuropsychologic deficits after cardiac operations has progressed from incidence studies, which involve a conventional definition of deficit, to intervention studies, in which specific test performance can be compared.
An increasing focus now is being placed on neuropsychologic assessment in the context of cardiac surgery. In part this development derives from the need to develop more sensitive tools of brain functioning and brain changes after cardiac operations. With the decline in mortality after cardiac procedures an interest developed in neurologic damage [13]. The reduction in the incidence of stroke made it less useful and discriminating as a measure of central nervous system functioning in cardiac surgery. More recent studies, therefore, have included not only an evaluation of the patients' neurologic status by clinical assessment but also more formal evaluations of cognition by means of neuropsychologic assessments to provide a sensitive index of brain function [46].
Cardiac Versus Clinical Neuropsychologic Assessment
Neuropsychologic assessment in the context of clinical practice frequently is pursued to examine whether a patient with a known brain lesion shows significant cognitive disturbance, to establish the pattern of cognitive abilities, or to establish whether there is evidence for brain damage. This form of assessment customarily uses tests that have standardized norms, and the assessment includes reference to these norms. In neuropsychologic assessments in cardiac surgery the task and issues are different (Table 1
). The principal difference is that the objective is to determine change in performance and, unlike most other assessments, at least two assessments are performed with relatively little time separation. As a result of this design difference, the selection of tests and analysis are different than those used in traditional neuropsychologic assessment. The tests used in cardiac surgery need, in particular, to show small learning as they are repeated. Parallel forms of the test are a great advantage as these enable repetition but reduce learning of items. The clinical environment also restricts the time available and therefore restricts the number of tests that can be performed. The selection of tests therefore needs to be considered carefully, and cognizance must be taken of the areas of the brain and cognitive abilities that are likely to be affected by the cardiac operation.
|
More recently specific batteries for use in cardiac surgery have been developed that have involved carefully selected, sensitive tests incorporating the use of computers for increased objectivity and of a duration that makes them feasible to perform within the time constraints of clinical practice [4, 9].
Incidence Studies
Rationale
To establish that a problem existed in cardiac surgery early studies involved a single group with at least two performances over time: one before operation and one after operation. These studies are essentially descriptive in nature and provide information on the behavior of individuals over time. It is important that different forms of analysis of this data are likely to lead to different conclusions.
Group Comparisons
To assess a group change over time involves a number of pitfalls [5, 6]. To examine group performance over time rather than analyzing individual data fails to take into account the possibility of learning on the neuropsychologic tests and also assumes deficits in all patients. Figure 1
illustrates this difficulty with a hypothetic example. If approximately 25% of patients show some deterioration and approximately 75% show some learning effects, then the overall trend will be toward no difference and lead to the conclusion that cardiac does not influence brain functioning (see Fig 1
). Some have used this form of analysis [10]. This type of analysis of group data fails to identify those patients with deficits. To identify those individuals who are performing at a significantly lower level than preoperatively a different form of analysis is required.
|
In our own work we have taken a deterioration in performance equal to or greater than 1 standard deviation of the preoperative performance on that test. We also only consider an individual to have a deterioration in performance if he or she shows this degree of deterioration on two or more tests [4, 9]. Another approach is to take a drop of 20% in the test performance to reflect a significant deterioration [11]. Both of these classifications are conservative in that most neuropsychologic tests when repeated on the same individual would be expected to produce some improvement in performance, and it is against this improvement that any deteriorations are to be considered.
The arbitrariness of this type of decision needs to be placed into context. Much of clinical science involves establishing a point that if exceeded is considered to reflect a pathologic state. Some examples of this approach are the definitions of hypertension, obesity, and learning disabilities. One important difference is that these measures normally do not reflect change scores and are population referenced rather than individual referenced. In other words, they do not take into account the level of a prior performance.
There are also important statistical as opposed to conventional difficulties in defining a change as reflecting significant neuropsychologic deterioration. For example, individuals with high preoperative scores are more able to show a deterioration of a fixed magnitude as they do not encounter floor effects. This problem, coupled with the associated difficulty of regression to the mean, creates some problems in examining change scores. In other areas of research it is customary to construct a regression line that compensates for the differential between poor and high achievers.
Comparison/Control Groups
One design that is potentially useful in this area is to introduce a comparison or control group. This control or comparison group has tended to be either a nonsurgical control group to examine for learning effects [11] or a surgical control group [12]. The former strategy enables the researcher to assess learning effects and the stability of their tests, but it does not control for the effects of operation, which may yield dramatic results in the short term. The difficulty with the latter approach concerns matching the control group to the surgical group of interest. In the case of coronary artery bypass grafting one of the areas of interest is the effects of extracorporeal circulation. It is extremely difficult to find a group of patients with a similar age and degree of disability who undergo an operation of a similar duration and complexity without extracorporeal circulation.
Findings in Incidence Studies
Neuropsychologic changes within the days after operation have been reported in a number of studies [4, 1218] with the incidence of neuropsychologic morbidity ranging from 12% to 79%. A number of studies have considered longer term changes beyond the first few weeks after bypass, and some of these findings, contrary to some reviews [19], do suggest a proportion of patients do have persisting neuropsychologic problems [5, 6, 11, 202323]. The number of patients deemed to have neuropsychologic deficits in these studies differs considerably [6].
Variability in Findings
One criticism held against neuropsychologic assessment revolves around the issue of variability in findings [6]. The neuropsychologic changes observed in cardiac surgery first must be seen against the variable incidence of stroke [24] and also the changes that have occurred and are occurring in cardiac surgery. These include variations on the use of hypothermia [25], reductions in bypass time, and significant changes in equipment. These all have led to a reduction in the more severe consequences of cardiac operations [26] and a reduction in neuropsychologic changes over time [9, 14, 27, 28]. In addition, any comparison of studies performed in different centers involves not only differences in surgical technique, but also population and sampling differences as well as variations in the number and type of tests used and the assessment of deficit.
Associated Factors
Studies of incidence have examined for associated patient factors such as age [4, 14, 20, 29] sex [20], co-morbidities such as diabetes [30] and cerebrovascular disease [31], operative variables such as arterial pressure [32, 33] and time on extracoproreal circulation [4, 12, 14, 15, 34], and associated mechanisms such as microemboli [3537] or cerebral blood flow [4, 12, 38, 39].
Intervention Studies
Design and Analysis
Studies attempting to investigate a proposed intervention such as a comparison between two anesthetic regimens [40], a variation in bypass equipment [37], or the introduction of a neuroprotective drug use a two or more group design where one group receives the intervention of interest while the other acts as a control group. Although it is possible to compare incidence measures between the intervention and the control group, these scores are both crude and conventional. The two-group design enables the scores on each test to be used in the analysis. The distribution of individual change scores can be examined where the preoperative score serves as a baseline and performance differences from this baseline may be plotted (see Fig 2
for an example). This form of analysis not only enables an examination to be made of individual performance but also makes possible a sensitive examination of each of the tests and their associated cognitive domains.
|
Limitations of Current Neuropsychologic Assessments
Current neuropsychologic assessments of patients undergoing cardiac operations are intended to offer a sensitive tool of brain functioning. A somewhat different question concerns the impact of cardiac surgery and potential cognitive disturbance on daily life. These questions commonly are swept up into the all-embracing but variably used term ``quality of life.'' Quality of life includes such issues as psychologic well being, social and leisure activities, and work. It frequently is considered that there should be a relationship between cognitive disturbance after a cardiac operation and some indicator of quality of life.
The difficulty of seeking a direct relationship between the measures of cognition employed in cardiac surgery and measures of quality of life will be illustrated by considering the ability to return to work after a cardiac operation. Studies that have examined return to work at various times after the operation have reported rates of employment ranging from 17% to 90% [4954]. Reasons given by individuals for not returning to work include physical difficulties, emotional problems, disability and pension benefits, age, and personality factors. It is important to recognize that cognitive changes are only one factor in a complex range of factors likely to affect return to work.
A further question concerns whether the cognitive changes observed are likely to affect work performance. The incidence measures described above use a deterioration in performance that is substantial; a reduction in the region of 20% in any measurable performance such as typing speed is clearly something that will affect work performance. The important question concerns the nature and demands of work and whether the individual is working or expected to work at full intellectual capacity. In many areas of employment individuals are not working at their full intellectual capacity and hence some decline in cognitive and work performance may not have any perceptible impact or may not be detected by colleagues or by the individual [55]. That is not to imply that the changes do not reflect significant alterations in cognition, but the translation from cognitive change to work performance is one that involves a range of social and psychologic factors. To tease out the role of cognitive disturbance among these other factors would require a study of substantial proportions.
Footnotes
Presented at the Conference on CNS Dysfunction After Cardiac Surgery: Defining the Problem, Fort Lauderdale, FL, Dec 1011, 1994.
Address reprint requests to Dr Newman, Department of Psychiatry, UCL Medical School, Middlesex Hospital, Mortimer St, London W1N 8AA, UK.
References
This article has been cited by other articles:
![]() |
F. D. Rubens, M. Boodhwani, and H. Nathan Interpreting studies of cognitive function following cardiac surgery: a guide for surgical teams Perfusion, May 1, 2007; 22(3): 185 - 192. [Abstract] [PDF] |
||||
![]() |
S. Dogan, T. Aybek, P. S. Risteski, F. Detho, A. Rapp, G. Wimmer-Greinecker, and A. Moritz Minimally Invasive Port Access Versus Conventional Mitral Valve Surgery: Prospective Randomized Study Ann. Thorac. Surg., February 1, 2005; 79(2): 492 - 498. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Motallebzadeh and M. Jahangiri Benefits of off-pump bypass on neurologic function Ann. Thorac. Surg., September 1, 2004; 78(3): 1131 - 1132. [Full Text] [PDF] |
||||
![]() |
W. Nagels, R. Demeyere, J. Van Hemelrijck, E. Vandenbussche, K. Gijbels, and E. Vandermeersch Evaluation of the Neuroprotective Effects of S(+)-Ketamine During Open-Heart Surgery Anesth. Analg., June 1, 2004; 98(6): 1595 - 1603. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Jahangiri and R. Motallebzadeh Neurocognitive deficit following mitral valve surgery Eur. J. Cardiothorac. Surg., October 1, 2003; 24(4): 665 - 665. [Full Text] [PDF] |
||||
![]() |
D. Wang, X. Wu, J. Li, F. Xiao, X. Liu, and M. Meng The Effect of Lidocaine on Early Postoperative Cognitive Dysfunction After Coronary Artery Bypass Surgery Anesth. Analg., November 1, 2002; 95(5): 1134 - 1141. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Van Dijk, E. W. L. Jansen, R. Hijman, A. P. Nierich, J. C. Diephuis, K. G. M. Moons, J. R. Lahpor, C. Borst, A. M. A. Keizer, H. M. Nathoe, et al. Cognitive Outcome After Off-Pump and On-Pump Coronary Artery Bypass Graft Surgery: A Randomized Trial JAMA, March 20, 2002; 287(11): 1405 - 1412. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. J. Fearn, R. Pole, K. Wesnes, E. B. Faragher, T. L. Hooper, and C. N. McCollum Cerebral injury during cardiopulmonary bypass: Emboli impair memory J. Thorac. Cardiovasc. Surg., June 1, 2001; 121(6): 1150 - 1160. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Millar, A. J. Asbury, and G. D. Murray Pre-existing cognitive impairment as a factor influencing outcome after cardiac surgery Br. J. Anaesth., January 1, 2001; 86(1): 63 - 67. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Grimm, M. Czerny, H. Baumer, J. Kilo, C. Madl, L. Kramer, A. Rajek, and E. Wolner Normothermic cardiopulmonary bypass is beneficial for cognitive brain function after coronary artery bypass grafting - a prospective randomized trial Eur. J. Cardiothorac. Surg., September 1, 2000; 18(3): 270 - 275. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Crawley, J. Stygall, S. Lunn, M. Harrison, M. M. Brown, and S. Newman Comparison of Microembolism Detected by Transcranial Doppler and Neuropsychological Sequelae of Carotid Surgery and Percutaneous Transluminal Angioplasty Stroke, June 1, 2000; 31(6): 1329 - 1334. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. M. Browne, P. W. Halligan, D. T. Wade, and D. P. Taggart COGNITIVE PERFORMANCE AFTER CARDIAC OPERATION: IMPLICATIONS OF REGRESSION TOWARD THE MEAN J. Thorac. Cardiovasc. Surg., March 1, 1999; 117(3): 481 - 485. [Abstract] [Full Text] [PDF] |
||||
![]() |
J.E. Arrowsmith, M.J.G. Harrison, S.P. Newman, J. Stygall, N. Timberlake, and W.B. Pugsley Neuroprotection of the Brain During Cardiopulmonary Bypass : A Randomized Trial of Remacemide During Coronary Artery Bypass in 171 Patients Stroke, November 1, 1998; 29(11): 2357 - 2362. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. C. Kneebone, M. J. Andrew, R. A. Baker, and J. L. Knight Neuropsychologic Changes After Coronary Artery Bypass Grafting: Use of Reliable Change Indices Ann. Thorac. Surg., May 1, 1998; 65(5): 1320 - 1325. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Kumar, K. Dhital, M. Hossein-Nia, S. Patel, D. Holt, and T. Treasure S-100 PROTEIN RELEASE IN A RANGE OF CARDIOTHORACIC SURGICAL PROCEDURES J. Thorac. Cardiovasc. Surg., May 1, 1997; 113(5): 953 - 954. [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |