| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Department of Anesthesiology (B.P.B., M.F.N., J.P.M., H.P.G.), Department of Medicine, Division of Cardiology (D.B.M.), Department of Medicine, Division of Neurology (D.T.L.), Department of Surgery (R.H.J.), and Department of Psychiatry and Behavioral Sciences (B.P.B., J.A.B.), Duke University Medical Center, Durham, North Carolina.
Address correspondence and reprint requests to Barbara Phillips Bute, PhD, Department of Anesthesiology, Box 3094, DUMC, Duke University Medical Center, Durham, NC 27710. E-mail: phill016{at}mc.duke.edu
| ABSTRACT |
|---|
|
|
|---|
Methods: With Institutional Review Board (IRB) approval and written informed consent, 732 patients at Duke University Hospital undergoing CABG were enrolled. Five hundred fifty-one (75%) completed baseline, 6-week, and 1-year neurocognitive tests and psychometric measures designed to assess QOL. Neurocognitive status was assessed by a composite cognitive index score representing the mean of the scores in four cognitive domains. Change in QOL was assessed by subtracting baseline from 1-year scores for each of 10 QOL measures. The association between QOL and cognitive dysfunction was investigated using multivariable linear regression analysis.
Results: Cognitive decline limited improvement in QOL, with substantial correlation between change in cognition and change in QOL. One-year QOL measures are associated with both 6-week and 1-year change in cognition (Instrumental Activities of Daily Living, p < .0001; Duke Activity Status Index, p < .02; Cognitive Difficulties, p < .0001; Symptom Limitations, p = .0001; Center for Epidemiologic Study Depression, p < .0001; General Health Perception, p = .0001).
Conclusions: Postoperative cognitive decline may diminish improvements in QOL. Strategies to reduce cognitive decline may allow patients to achieve the maximum improvement in QOL afforded by CABG, as even short-term cognitive dysfunction has implications for QOL 1 year later.
Key Words: neurocognitive function CABG surgery quality of life
Abbreviations: CABG = coronary artery bypass graft; QOL = quality of life; IADL = instrumental activities of daily living; DASI = Duke Activity Status Index; STAI = Spielberger State Anxiety Inventory; WAIS-R = Wechsler Adult Intelligence Scale-Revised; CPB = cardiopulmonary bypass; CI = Cognitive Index; SF-36 = Medical Outcomes Study 36-Item Short Form Health Survey; CES-D = Center for Epidemiological Studies Depression Scale.
| INTRODUCTION |
|---|
|
|
|---|
One potential complication of CABG experienced by many individuals is postoperative impairment in cognitive function (68). Even in cases where cognitive functioning does not decline to levels recognized as clinical impairment, a persistent, measurable reduction in important domains of function may occur. Cognitive decline which can be measured within days of cardiac surgery has been shown to be associated with long-term cognitive dysfunction (7). However, there are few data that describe the relationship of postoperative cognitive decline with long-term patient outcomes. Because many patients show improved QOL after CABG (9), the general relevance of acute cognitive decline to patients lives has not been obvious.
The effect of cognition on QOL has been identified in several disease processes and health fields (1012). CABG surgery represents a unique opportunity to further define the role of cognitive change in altering QOL in an environment of substantial physical limitations and symptomatology. A significant cross-sectional association between cognition and QOL 5 years after cardiac surgery has been demonstrated (13); however, the lack of longitudinal QOL data made it impossible to assess whether a change in cognitive function resulted in a corresponding change in QOL, or if there was simply a cross-sectional association between the two. As Lewis et al. (14) specifically note, the extent to which postoperative cognitive deficit affects activities of daily living is not known. Therefore, the aim of this study is investigate whether there is an association between changes in neurocognitive function and changes in QOL 1 year after CABG surgery.
| METHODS |
|---|
|
|
|---|
Measurement of Neurocognitive Function
A well-validated battery of five neurocognitive tests was administered before surgery (baseline), 6 weeks after surgery, and 1 year after surgery. Tests included the short-story module of the Randt Memory Test (15); the Digit Span subtest of the Wechsler Adult Intelligence ScaleRevised (WAIS-R) test (16); the Modified Visual Reproduction Test from the Wechsler Memory Scale; the Digit Symbol subtest of the WAIS-R (16); and the Trail Making test (Part B) (17).
Measurement of QOL
QOL instruments were administered individually by trained psychometricians blinded to the patients neurocognitive test results. Most of the questionnaires were self-administered; however, the measures were read to the patient if requested. The following QOL measures were used:
Patient Management
Anesthetic management was composed of midazolam, fentanyl, isoflurane, and pancuronium, as has been previously described (27). All patients underwent standard nonpulsatile hypothermic (30°C to 34°C) cardiopulmonary bypass (CPB) using a membrane oxygenator, roller pump, and including an arterial line filter. Nonpulsatile perfusion of 2 to 2.4 l per minute per square meter was maintained throughout CPB. The pump was primed with crystalloid solution to achieve a hematocrit of
18% during extracorporeal circulation, with packed red blood cells added when necessary to achieve the desired hematocrit. All patients underwent CPB through an ascending aortic cannula. Throughout CPB, arterial carbon dioxide tension was maintained at 35 to 40 mm Hg with a partial pressure of oxygen at 150 to 250 mm Hg.
Statistical Analysis
Measurement of neurocognitive decline was derived from a previously described factor analysis (13) performed on baseline scores. Factor scores for follow-up time periods are calculated using weights derived from this baseline analysis, so factor definitions remain consistent across all time periods. The four factors used in the final analysis represent the cognitive domains of 1) verbal memory and language comprehension, 2) visuoconstruction, 3) abstract recall, and 4) complex attention/executive function.
Change in cognitive function is described in two different ways. A continuous change score for each factor was calculated by subtracting baseline from follow-up scores. To quantify overall cognitive function, a composite cognitive index (CI) was calculated as the mean of the four domain scores to yield a single, continuous summary measure (
CI) reflecting improvement, as well as decline. This continuous CI is our primary outcome measure. As a secondary, descriptive measure, a dichotomous outcome (neurocognitive deficit) was defined as a SD decline in one or more domains.
Because no validated methodology for combining the domains of QOL was available, we investigated each QOL instrument and CI change separately using multivariate linear regression. Change in QOL (
QOL) was assessed by subtracting baseline from 1-year scores for each QOL measure. Each
QOL measure was tested in a separate multivariable linear regression model to account for the possible competing effects of age, gender, race, baseline cognitive function, baseline QOL, and Charlson comorbidity. Covariables were selected a priori and included in all models regardless of significance level.
As a secondary analysis, 6-week cognitive change scores were used as predictors of 1-year
QOL in otherwise identical models as described above. The purpose of this analysis was to determine if impaired 1-year QOL could be anticipated by immediate, short-term change in cognition.
To determine if there is a single cognitive domain that is primarily responsible for driving the associations between overall cognitive function and QOL, associations between each QOL measure and the individual cognitive domains are explored.
| RESULTS |
|---|
|
|
|---|
|
Neurocognitive Function
Neurocognitive deficit, defined as a SD decline in one or more domains, occurred in 41% of patients at the 6-week follow-up, which is consistent with our previous publications (7). At 1 year, 36.8% of patients demonstrated a neurocognitive deficit. The percentage of patients who met the definition for cognitive deficit is shown in Table 2. The continuous measure of cognitive change, the CI, showed a small positive change overall, both at 6-week and 1-year follow-up, with substantial individual variability. The overall CI and individual domain scores are shown in Table 3.
|
|
QOL
QOL measured at baseline was consistent with previous studies of cardiac surgery (Table 4). The majority of QOL measures remained unchanged at 6 weeks; measurements of physical activity (DASI) and activities of daily living (IADL) showed significant decline, thought to be related to the ongoing physical recuperation. QOL measures at 1 year show improvement in most measures (Table 4).
|
Association of Cognition With (
) QOL
Multivariable Models
Changes in cognition and in QOL over the course of 1 year showed substantial association. Multivariable analysis controlling for baseline CI, baseline QOL, gender, race, age, and Charlson comorbidity showed a significant relationship between 1-year
CI and
QOL in IADL (p < .0001), the DASI (p = .03), symptom limitations (p = .0002), Cognitive Difficulties score (p < .0001), CES-Depression (p < .0001), and general health perception (p = .0002). Details of these models are presented in Table 5A.
|
Secondary analysis showed that 6-week
CI was predictive of most of the same
QOL measures, after adjusting for the same covariates. Specifically, 6-week
CI was associated with 1-year activities of daily living (p = .003), the DASI (p = .02), symptom limitations (p = .02), Cognitive Difficulties (p = .003), and depression (p = .003). General health perception at 1 year was marginally associated with 6-week
CI (p = .09). Details of these models are presented in Table 5B.
Change in cognition was not predictive of 1-year anxiety, perceived social support, or social activities (although this demonstrated a trend at p = .08). Work activities were also not significantly associated with cognitive function; this may be due to limited power, as only 36% of our sample was working full time before surgery.
Domain Analysis
Simple correlations between each QOL score and individual cognitive domain scores show few significant relationships (Table 6), even in instances where there is a correlation between QOL and the overall CI. The unadjusted associations with the overall CI show a similar pattern of findings to multivariable modeling.
|
| DISCUSSION |
|---|
|
|
|---|
Although the literature reports that physical functioning is typically improved by cardiac surgery (34,35), many patients experience some degree of postoperative cognitive decline (36,37), and these declines can persist long term (7,38). Although the longitudinal effects of cardiac surgery on cognitive functioning have been investigated, few studies addressed both neurocognitive functioning and QOL (7,39). Many studies that have documented postsurgical functional status and QOL did not include assessment of cognitive functioning (5).
The effect of cognition on QOL has been demonstrated in numerous medical contexts (1012). Likewise, the longitudinal effects of cardiac surgery on cognitive functioning have been investigated; however, few of these studies addressed both neurocognitive functioning and QOL (7,12,39,40). Similarly, many studies that have documented postsurgical functional status, survival, and QOL did not include assessment of cognitive functioning (5,4143). One randomized CABG trial (9) that included both cognitive and QOL evaluations demonstrated a potential cognitive benefit of off-pump CABG as compared with standard CABG, but there was no difference in QOL between treatment groups, a finding that sheds doubt on the importance of postoperative neurocognitive status on QOL. However, this study did not look directly at the association between cognition and QOL.
Our study finds that cognitive decline is significantly associated with less ability to engage in activities of daily living, less functional capacity, more depression, more self-reported mental difficulties, greater symptom limitations, and a less positive general health impression. Even after controlling for factors that are known to affect QOL, including baseline functioning and overall health, we find a strong effect of cognitive decline on important areas of both functional and emotional QOL. The association of cognitive function with activities of daily living and physical functioning scores is particularly striking, as one might expect that other physical limitations from pain or other disease states could overwhelm any measurable effects of cognition. Because our data analysis method is linear and is performed on continuous measures of cognitive and QOL, it is of note that not only do patients with decreased cognitive function experience declines in QOL but also patients who experience increased cognitive function experience improved QOL.
Our examination of cognitive domains leads us to the conclusion that no single cognitive domain is responsible for the observed associations between cognitive decline and impaired QOL. Instead, the effect seems to be cumulative over the four domains, revealing stronger correlations with the composite CI score than with any single domain.
Although previous studies have not found strong associations between subjective reports of memory loss and performance on objective neuropsychological assessments (44), our data show strong association in multivariable models between the Cognitive Difficulties Scale and both 6-week and 1-year CI scores. Interestingly, there is no significant correlation between the Cognitive Difficulties Scale and any single cognitive domain score. This observation again suggests that there is a cumulative effect across the four domains, which together affect a patients own perception of problems in long- and short-term memory, concentration, attention, and psychomotor coordination. Unlike Selnes and McKhann (31), who has found that long-term changes occur primarily in the domain of motor or psychomotor speed, we see the largest incidence of decline in the domain of verbal memory.
Our comparison of 1-year assessment of neurocognitive function and QOL suggests, in contrast to conclusions drawn by other investigators (3133), that these changes are not transitory or fleeting. Cognitive function and physical aspects of QOL have had time to achieve maximal improvement at 1 year, and the strong correlation of change in cognitive function with change in QOL supports the importance of addressing cognitive function when evaluating recovery from cardiac surgery. Of perhaps equal importance, the associations with 6-week neurocognitive function testing suggest that the QOL deficits experienced at 1 year may be anticipated by more immediate testing.
Limitations of our study include the inability to follow up or complete data for 25% of our patients. Individuals with inadequate baseline or 1-year cognitive function or QOL assessments represent a substantial part of this loss. Patients completing 1-year follow-up tended to have more education than noncompleters and were more likely to be female.
Although it is not possible to determine a strict causal relationship, the hypothesis that cognitive decline is a driving force in impaired QOL is reinforced by our findings that both short- and long-term change in cognitive function are predictive of long-term change in QOL. This suggests that the shorter term end point could be used as a surrogate end point when assessing efficacy of neuroprotective strategies.
Our study demonstrates that cognitive decline occurring after surgery limits the anticipated improvement in QOL accomplished by surgical coronary revascularization. Despite the fact that we see a slight increase in the overall CI (which is the expected response in unaffected subjects and is explained by the learning effect), the number of patients affected by postsurgical cognitive deficit is not small: 36% of patients in this study experienced a cognitive deficit 1 year after surgery. Although many patients both expect and experience improved QOL following CABG surgery, not everybody experiences these benefits.
Because cardiac procedures today are successfully performed on increasingly older patients (34,35,45), it is essential to anticipate which patients are likely to receive the same enhancements in QOL as in quantity of life. Strategies to improve cognition or reduce cognitive decline could potentially enable patients to achieve the maximum level of improvement in QOL that they expect from cardiac surgery. This study adds to our understanding of how our objective measures of patients cognitive performance are reflected in their lives and daily functioning.
Supported in part by grants from National Institutes of Health Grants R01-HL54316, R01-AG09663, and R03-AG14194, Clinical Research Centers Program, National Institutes of Health Grant M01-RR-30, American Heart Association Grant-In-Aid 95010970, and Pharmaceutical Roundtable Grant #9970128N.
| Appendix 1 |
|---|
|
|
|---|
Director
Joseph P. Mathew, MD; Codirector: James A. Blumenthal, PhD.
Anesthesiology
Hilary P. Grocott, MD; Madan Kwatra, PhD; Burkhard Mackensen, MD; Joseph P. Mathew, MD; Mark F. Newman, MD; Debra A. Schwinn, MD; Mark Stafford-Smith, MD; Madhav Swaminathan, MD; David Warner, MD; Bonita L. Funk, RN; Narai Balajonda, MD; Chonna Campbell; Maria Celerian, MD, BS; Glenn Davis, BS; Eugenie Eaborn, RN; Roger L. Hall, AAS; Marcie Hanish, RN; Michael Hill, BS; Jerry L. Kirchner, BS; Satarah Latiker, BS; Erich Lauff, BA; Richard Morris, PhD; Charles R. Peters, MA; Meredith Prince; William Hansley, BS; Barbara Phillips-Bute, PhD; Andrew Slaughter, BS; Elizabeth Perez, RN; Josephine White, AAS; William D. White, MPH; and Sarah Woodring, BS
Behavioral Medicine
Michael A. Babyak, PhD, James A. Blumenthal, PhD.
Cardiology
Daniel B. Mark, MD, MPH, and Michael H. Sketch, Jr, MD.
Neurology
Ellen R. Bennett, PhD; Carmelo Graffagnino, MD; Daniel T. Laskowitz, MD; John R. Lynch, MD; Warren J. Strittmatter, MD; and Kathleen A. Welsh-Bohmer, PhD.
Perfusion Services
Greg Smigla, BS, CCP, and Ian Shearer, BS, CCP.
Surgery
Thomas A. DAmico, MD; R. Duane Davis, MD; Donald D. Glower, MD; R. David Harpole, MD; James Jaggers, MD; Robert H. Jones, MD; Andrew Lodge, MD; James E. Lowe, MD; Robert H. Messier, MD; Carmelo Milano, MD; Peter K. Smith, MD; Eric M. Toloza, MD, PhD; and Walter G. Wolfe, MD.
| NOTES |
|---|
|
|
|---|
Received for publication July 1, 2005; revision received November 23, 2005.
DOI:10.1097/01.psy.0000221272.77984.e2
| REFERENCES |
|---|
|
|
|---|
75 years of age with younger patients. Am J Cardiol 1999;83:112932.[CrossRef][Medline]This article has been cited by other articles:
![]() |
P. J. Tully, R. A. Baker, J. L. Knight, D. A. Turnbull, and H. R. Winefield Neuropsychological Function 5 Years after Cardiac Surgery and the Effect of Psychological Distress Arch Clin Neuropsychol, December 1, 2009; 24(8): 741 - 751. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y.-H. Liu, D.-X. Wang, L.-H. Li, X.-M. Wu, G.-J. Shan, Y. Su, J. Li, Q.-J. Yu, C.-X. Shi, Y.-N. Huang, et al. The Effects of Cardiopulmonary Bypass on the Number of Cerebral Microemboli and the Incidence of Cognitive Dysfunction After Coronary Artery Bypass Graft Surgery Anesth. Analg., October 1, 2009; 109(4): 1013 - 1022. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Mathew, G. B. Mackensen, B. Phillips-Bute, H. P. Grocott, D. D. Glower, D. T. Laskowitz, J. A. Blumenthal, M. F. Newman, and for the Neurologic Outcome Research Group (NORG) o Randomized, Double-Blinded, Placebo Controlled Study of Neuroprotection With Lidocaine in Cardiac Surgery Stroke, March 1, 2009; 40(3): 880 - 887. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Dunning, J. R.L. Waller, B. Smith, S. Pitts, S. W.H. Kendall, and K. Khan Coronary Artery Bypass Grafting is Associated With Excellent Long-Term Survival and Quality of Life: A Prospective Cohort Study Ann. Thorac. Surg., June 1, 2008; 85(6): 1988 - 1993. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. P. Obrenovitch Molecular Physiology of Preconditioning-Induced Brain Tolerance to Ischemia Physiol Rev, January 1, 2008; 88(1): 211 - 247. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |