| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2006;82:388-390
© 2006 The Society of Thoracic Surgeons
a Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
b Department of Biostatistics, The Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
c Zanvyl Krieger Mind/Brain Institute, Baltimore, Maryland
* Address correspondence to Dr Selnes, Division of Cognitive Neuroscience, Reed Hall East - 2, The Johns Hopkins Hospital, 1620 McElderry St, Baltimore, MD 21205-1910 (Email: oselnes{at}jhmi.edu).
It has been widely assumed that coronary artery bypass grafting (CABG) is associated with cognitive decline. Nevertheless, attempts to quantify these changes with neuropsychological tests have led to highly variable results. For example, the reported incidence of cognitive decline among patients evaluated shortly before hospital discharge has ranged from 14% to a high of 48% [1]. Similar wide variations are reported at later follow-up time points. The source of this variability has been attributed to a variety of factors, including between-study differences in subject selection criteria, postsurgical follow-up times, and the number and sensitivity of neuropsychological tests used to measure cognitive change.
A major limitation of these studies, including our own [2], is that most did not compare the incidence of postoperative cognitive decline among CABG patients to that observed in a control group (ie, either healthy persons or those with similar degrees of cardiovascular and cerebrovascular disease). In the absence of a control group, the criterion for cognitive decline must be based on an arbitrary measure of change within the study population. In early studies, a decline of 1 standard deviation decrease from baseline in 20% of tests was commonly used [3]. More recently a 20% decline in 20% of neuropsychology tests has been used [4]. Mahanna and colleagues [3] applied five such arbitrary criteria for decline to the same data set, and found that the incidence of cognitive decline at 6 weeks after surgery ranged from a low of 1% to a high of 34% depending on which criterion was used [3]. They concluded that the "large variation in the reported incidence of cognitive decline after CABG can be attributed to the different criteria used to define impairment."
Although the findings of Mahanna and colleagues [3] were published more than a decade ago, some contemporary studies of cognitive outcomes after CABG have continued to rely on fixed, arbitrary definitions of cognitive decline, such as 20% decline in 20% of tests [5]. However, it is less appreciated that any estimates of the incidence of postoperative cognitive decline after CABG are essentially uninterpretable unless reference is made to a control group. A recent example is the reanalysis of data from the Octopus Study group, which originally used the 20% decline on 20% of the tests criterion in the analysis of cognitive change after conventional and off-pump CABG [4]. Because this was a randomized trial, nonsurgical controls were not included. In their original analysis, 31% of the CABG patients were classified as having decline at 3 months. In a follow-up to this study, they recruited healthy controls not undergoing surgery and applied the same criteria for decline, and it was found that an unexpected 28% of these normal controls also met this criterion for decline [6]. Using a more conservative definition to define decline, they then compared their original CABG group with this control group and found at 3 months that 7.7 % of the CABG and 4.6% of the controls were classified as having decline. These authors conclude that their previous use of the 20% decline of 20% of test criteria had thus greatly overestimated the incidence of cognitive decline after CABG.
We expand on the findings just mentioned by illustrating the effects of applying the arbitrary criterion of 20% decline on one or more tests to previously published data from our prospective study of cognitive outcomes comparing CABG and nonsurgical controls with coronary artery disease and heart-healthy controls (without risk factors for coronary artery disease) [7]. The results clearly demonstrate that there is considerable variability (both improvement and decline) in the follow-up test performance even for the control subjects without surgery. Therefore, in the absence of a control group, this normal variability associated with follow-up cognitive testing might have been incorrectly attributed to surgery-related cognitive decline.
Figure 1 displays box plots of the within-subject changes in z-scores from 3 to 12 months by study group: CABG, nonsurgical control, and heart-healthy control for the cognitive domains of memory and a global domain score. Whereas there are both decliners and improvers in all study groups, previous detailed analyses of these data have shown there is little or no evidence of a disproportionate decline in cognitive performance in the CABG group relative to the controls [7].
|
|
Because both the CABG and nonsurgical controls have known risk factors for cerebrovascular disease, it is possible that the patients whose scores declined with time had progression of their underlying vascular disease. To control for this, we also included a control group that we designated as heart-healthy; these subjects had no known risk factors for vascular disease. The distribution of within subject change scores from 3 to 12 months in the heart healthy controls is remarkably similar to that of the two other groups, supporting the interpretation that the underlying mechanism is test-retest variability rather than disease-related decline.
With the same longitudinal data set that is used in this report, more appropriate and powerful analyses to compare the trends in cognitive function between surgical and nonsurgical intervention groups have been discussed by Barry and colleagues [8]. Rather than creating a binary response of decline versus no decline for each subject, more of the available information was used by analyzing the numerical trends in the test scores themselves. Random effects regression models estimated the difference in the temporal trend between study groups, taking account of practice effect and heterogeneity among subjects in the levels and trends of their scores. In this way, all of the available information was used without discarding parts of it, as would have been the case if the individual trends had been dichotomized into decline or no decline.
There are continuing reasons to monitor the association of CABG with possible postoperative cognitive change. First, the technology associated with CABG is constantly changing, and the efficacy of these changes in terms of cognitive outcomes should be determined. Second, there is increasing use of other interventions for coronary artery disease, such as "off-pump" CABG and coronary artery stenting procedures. Studies to compare the efficacy as well as comparison of the rates of adverse cognitive outcomes of these alternate approaches to coronary artery disease are needed. Finally, these cardiac interventions provide an opportunity for evaluation of possible neuroprotective agents. If changes in cognition are to be part of this outcome research, the expression of both decline and improvement in cognition should be included, as well as comparison with appropriate control groups. Without these comparisons, estimates of cognitive decline are greatly overestimated, and virtually uninterpretable.
 |
Acknowledgments |
|---|
 Top Acknowledgments References |
|---|
|
References |
|---|
|
TopAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. van Dijk, K. G.M. Moons, H. M. Nathoe, E. H.L. van Aarnhem, C. Borst, A. M.A. Keizer, C. J. Kalkman, R. Hijman, and Octopus Study Group Cognitive Outcomes Five Years After Not Undergoing Coronary Artery Bypass Graft Surgery Ann. Thorac. Surg., January 1, 2008; 85(1): 60 - 64. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. W. Hogue, O. A. Selnes, and G. McKhann Should All Patients Undergoing Cardiac Surgery Have Preoperative Psychometric Testing: A Brain Stress Test? Anesth. Analg., May 1, 2007; 104(5): 1012 - 1014. [Full Text] [PDF]
|
|
|
|
 |
|
 D. van Dijk, M. Spoor, R. Hijman, H. M. Nathoe, C. Borst, E. W. L. Jansen, D. E. Grobbee, P. P. T. de Jaegere, C. J. Kalkman, and for the Octopus Study Group Cognitive and Cardiac Outcomes 5 Years After Off-Pump vs On-Pump Coronary Artery Bypass Graft Surgery JAMA, February 21, 2007; 297(7): 701 - 708. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
