| ||||||||||||||
|
|
|||||||||||||



Departments of *Anesthesiology,
Psychiatry and Behavioral Sciences,
Medicine (Neurology), and
Surgery, Duke University Medical Center, Durham, North Carolina
Address correspondence and reprint requests to Mark F. Newman, MD, Box 3094, Duke University Medical Center, Durham, NC 27710. Address e-mail to newma005{at}mc.duke.edu
| Abstract |
|---|
|
|
|---|
IMPLICATIONS: Neurocognitive dysfunction is common after coronary artery bypass graft surgery. The relationship between atrial fibrillation and neurocognitive dysfunction has not been examined. Our study shows that postoperative atrial fibrillation is associated with neurocognitive decline.
| Introduction |
|---|
|
|
|---|
Atrial fibrillation (AFIB) is a commonly seen condition after cardiac surgery, with an incidence of 20% to 40% (810). The incidence of AFIB is increased in patients of advanced age (8,10), male sex (10), chronic obstructive pulmonary disease (11,12), and certain electrocardiographic conditions (13). AFIB has been shown to increase intensive care unit and hospital length of stay and to increase the intensity of nursing care (810,14,15). Particularly germane to our investigation is the documented association between AFIB and an increased incidence of postoperative neurologic abnormalities (e.g., stroke and transient ischemic attack) (8,10,16). The purpose of our prospective observational study was to determine whether postoperative AFIB is associated with an increased risk of cognitive decline after cardiac surgery.
| Methods |
|---|
|
|
|---|
Patients were individually examined with a battery of well-validated and established cognitive tests by experienced psychometricians blinded to the patients cardiac rhythm status. Testing took place the day before CABG surgery (baseline) and 6 wk postoperatively. The battery of 6 tests resulted in 18 raw scores for each testing period and included the short story module of the Randt Memory Test, in which examinees are asked to recall details of a short story immediately after having it read to them and again after a 30-min delay. Scores are assessed on "verbatim" and "gist" accuracy for both testing periods, resulting in four variable scores (17). The Digit Span subtest of the Wechsler Adult Intelligence ScaleRevised examination requires the repeating of increasing series of numbers forward and backward. Scores are based on the number of series achieved forward and backward (18). The Wechsler Memory Scale Figural Memory test requires patients to reproduce three designs from memory after 10 s of exposure and after 30 min. Scoring is based on the accuracy of the reproductions and results in two variable scores (immediate and delayed) (19). The Digit Symbol subtest of the Wechsler Adult Intelligence ScaleRevised asks patients to correctly associate symbols with their respective digits on the basis of a code during a 90-s time period. Scores are based on the number of correct matches and result in one variable score (18). The Trail Making Test (Part B) is a procedure that tests the ability of patients to connect a series of numbers and letters (e.g., 1-A-2-B). Scores are based on the time necessary to complete the test and result in a single variable score (20). The Rey Auditory-Verbal Learning Test asks patients to recall a 15-word list after each of 5 presentations, 1 presentation of a second 15-word list, a sixth recall trial of the original word list, and a 30-min delayed recall. Scores are based on the number of verbatim correct answers for each trial, resulting in eight raw scores (21).
All patients were operated on by the same team of surgeons at the Duke Heart Center. Anesthesia was established with midazolam, fentanyl, and pancuronium and maintained with isoflurane (3). The perfusion apparatus consisted of the Cobe CML Membrane OxygenatorTM (COBE Chem Labs, Lakewood, CO), the Sarns 7000 max roller pump (3M, Inc., Ann Arbor, MI), and the Pall SP3840TM arterial line filter (Pall Biomedical Products Co., Glen Cove, NY). Patients were cooled to 32°C, and perfusion was maintained at pump flow rates of 2.02.4 L · min-1 · m-2 throughout cardiopulmonary bypass. The pump was primed with crystalloid, and serial hematocrits were maintained
0.18 with packed red blood cell supplementation as necessary. Mean arterial blood pressure was maintained at 5090 mm Hg. Dual-stage atrial cannulation was used. Venting was through the aortic root per institutional practice. Cardioplegia technique was per surgeon discretion, although all patients were arrested during aortic cross-clamping via antegrade cardioplegia. Arterial blood gas analyses were monitored every 1530 min to maintain arterial carbon dioxide partial pressures at 3540 mm Hg and oxygen partial pressures at 150250 mm Hg by
stat. Patient temperatures were core temperatures and were measured in the bladder and the nasopharynx. Patients were rewarmed slowly with cardiopulmonary by-pass inflow temperatures not exceeding 2°C3°C gradients compared with nasopharyngeal temperature measurements.
Patients were monitored with continuous telemetry both in the intensive care unit and in the step-down unit for 96 h postoperatively. Patients were considered to have postoperative AFIB if telemetry displayed a sustained rhythm which was confirmed by a cardiologist-interpreted 12-lead electrocardiogram and which did not convert spontaneously.
The process of assessing neurocognitive decline used a factor analysis (22). This method uses the correlation among the set of 18 raw scores to construct a smaller set of four independent factor scores, each representing a separate domain of cognitive function. The scores for the factors are derived from factor loadings and weights of each test on each domain, determined by the factor analysis. The factor scores represent performance in distinct and relatively independent areas of cognitive function and are standardized to the same scale (z scores). Each subjects factor scores for both baseline and 6-wk cognitive function were calculated by using the weights from the baseline factor analysis. In this manner, four factors were identified at baseline and remained consistent for the 6-wk test period (2).
Factor analysis on the 18 baseline neuropsychological test scores found that the four factors account for approximately 71% of the variance present in the test battery at baseline. The four factors represent the cognitive domains of 1) verbal learning, short-term and delayed; 2) discourse memory and oral language comprehension, short-term and delayed; 3) visuospatial orientation, psychomotor processing speed, and figural memory, short-term and delayed; and 4) attention and concentration.
The mean of each subjects four factor scores formed their composite cognitive index. The composite cognitive change score was then calculated by subtracting the baseline index from the 6-wk follow-up index in each patient. Our sample size was powered to detect a difference in cognitive function as assessed by a continuous method (composite cognitive change score), assuming an incidence of AFIB of approximately 25% in the population. In addition, an overall dichotomous "cognitive deficit" outcome event was defined as a decline in performance of 1 SD or more in any of the independent domains. This information is presented only descriptively, to preserve the
level of 0.05; no additional statistical tests were performed on this information.
The two groups (AFIB and no AFIB) were compared for baseline and demographic differences with Fishers exact test for categorical variables and with Wil-coxons ranked sum test for numeric variables. Significance was assessed at a two-tailed
of 0.05.
The effect of postoperative AFIB on 6-wk change in composite cognitive function was investigated with linear regression models. First, the unadjusted association between AFIB and outcome was found by simple linear correlation. Multivariable regression modeling was then used to test the effect and adjust for covariable effects of baseline cognitive function, years of education, age, diabetes, left ventricular ejection fraction, preoperative AFIB, and previous myocardial infarction. Nonsignificant covariables were dropped stepwise from the model until only significant effects remained.
With the baseline SD of the individual factor scores at 1.0, to detect a difference between the AFIB groups in a before-to-after change of approximately 0.2 points, or approximately 20% of an SD, estimates had to be based on a two-sample Students t-test with unequal sample sizes, thus allowing for approximately 25% of the total samples having AFIB and assuming an SD of a change of 0.5 on the basis of pilot data. Calculations showed that a test with a 0.050 two-sided significance level would have 84% power to detect this difference when the sample sizes in the two groups were 226 (no AFIB) and 75 (with AFIB), which required a total sample size of at least 300. To allow for a generous loss to follow-up of 25%, 400 subjects were needed.
| Results |
|---|
|
|
|---|
Sixty-nine (22%) of the 308 patients experienced postoperative AFIB during their hospital stay. Demographic differences between those patients with and without AFIB are listed in Table 1. Patients with AFIB were older (P = 0.001) and more likely to have had preoperative AFIB (P = 0.009) compared with patients without postoperative AFIB.
|
|
|
|
| Discussion |
|---|
|
|
|---|
Neurocognitive dysfunction is common after CABG surgery (27,23). Although many studies have investigated the causes of neurologic and neurocognitive dysfunction after cardiac surgery, the relationship between postoperative AFIB and neurocognitive dysfunction has until now not been examined. The paroxysmal nature of this rhythm disturbance, combined with an increased potential for thrombus formation, embolization (24), and decreases in cardiac output (8), is thought to heighten the risk of neurologic injury. Evidence for postoperative AFIB as a neurocognitive risk factor comes from a number of other studies demonstrating an association between AFIB and postoperative stroke (8,10,16).
Although frank stroke is a devastating and debilitating complication of cardiac surgery, physicians and patients have discovered that deficits in memory, concentration, or attention not only are disconcerting and demoralizing, but also can result in long-term neurologic dysfunction and reduced quality of life for the patient (2,25). Our investigation is the first study demonstrating an association between neurocognitive outcome and postoperative AFIB. These results are important because they demonstrate that postoperative factors, in addition to intraoperative and patient factors, alter the pattern of postoperative cognition. Obviously, the etiology of neurocognitive dysfunction is multifactorial.
To compare cognitive function between patients with and without postoperative AFIB, we assessed cognitive function by using a previously described continuous method of cognitive change (2). This method allows a comparison of cognitive improvement and decline. The use of factor analysis as part of this continuous scale decreases the chance of Type I errors resulting from multiple test redundancy. The use of a factor-loading system separates components of each individual test into the previously described domains, with all domains remaining consistent from baseline to 6-wk follow-up testing. Thus, a direct comparison was made between each subjects baseline and follow-up domain scores.
Limitations of our study include the increased incidence of preoperative AFIB within the postoperative AFIB group, possibly contributing to a reduction in cognitive scores at baseline. If AFIB is a cause of neurocognitive dysfunction, then those patients with preoperative AFIB may already have some dysfunction that may not improve. In fact, the composite cognitive score was lower at baseline in patients with AFIB, although not significantly so, whereas the 6-wk follow-up scores are significantly divergent. This baseline effect is negated by our use of the cognitive change score as opposed to the follow-up score alone. Additionally, patients with less baseline cognitive function are statistically less likely to demonstrate a decline. Therefore, these group differences would only have worked against finding support for our hypothesis. Finally, multivariable analysis revealed an effect of postoperative AFIB on neurocognitive function independent of age, baseline cognitive function, and preoperative AFIB.
Second, our study was powered only to detect a difference in cognitive function as assessed by a continuous composite index in two groups of unequal size (see Methods). Although we have presented the incidence of cognitive dysfunction defined dichotomously (1 SD decline in any domain) and although the observed incidence of cognitive dysfunction was more frequent in the group developing postoperative AFIB, no statistical comparisons were made to preserve the
level at 0.05.
Another limitation was in the measurement of postoperative AFIB. Monitoring for postoperative AFIB was limited to 96 hours after surgery. This opens the possibility that AFIB may have occurred after that period or even after the patient had been discharged from the hospital. However, if patients developed AFIB that was confirmed and treated after leaving the telemetry unit, we included them in the postoperative AFIB group.
A final limitation to our study is the loss of patients to follow-up. Although the effect of this limitation is unclear, the similar incidence of AFIB between the patients returning and those lost to follow-up provides some evidence that our results would probably hold in the full sample.
In conclusion, our study suggests that postoperative AFIB is associated with reduced cognitive function when compared with normal cardiac rhythm after CABG surgery. Although the exact mechanism by which this occurs is uncertain, preventing AFIB may improve postoperative neurocognitive function.
Appendix 1: Neurologic Outcome Research Group of the Duke Heart Center
Director: Joseph P. Mathew, MD. Co-Director: James A. Blumenthal, PhD. Anesthesiology: Hilary P. Grocott, MD, Steven E. Hill, MD, Mark F. Newman, MD, Joseph P. Mathew, MD, J. G. Reves, MD, Debra A. Schwinn, MD, Mark Stafford-Smith, MD, David Warner, MD, Malissa Harris, RN, BSN, Jerry L. Kirchner, BS, Brenda Mickley, BS, Mandy Barnes, RN, BSN, Elizabeth Carver, RN, BSN, Bonita L. Funk, RN, E. D. Derilus, BS, Jason Hawkins, RN, BSN, Terri Moore, BA, Chonna Campbell, BS, Amanda Cheek, AS, Tanya Kagarise, BS, Tori Latiker, BS, Erich Lauff, BA, Melanie Tirronen, BS, Regina DeLacy, BA, William Hansley, BS, Yvonne M. Connelly, MA, MPH, Barbara Phillips-Bute, PhD, and William D. White, MPH. Behavioral Medicine: Michael A. Babyak, PhD, and James A. Blumenthal, PhD. Cardiology: Daniel B. Mark, MD, MPH, and Michael H. Sketch Jr., MD. Neurology: Carmelo Graffagnino, MD, Daniel T. Laskowitz, MD, John R. Lynch, MD, Ann M. Saunders, PhD, Warren J. Strittmatter, MD, and Kathleen A. Welsh-Bonner, PhD. Pathology: Ellen Bennett, PhD. Perfusion Services: Greg Smigla, BS, CCP, and Ian Shearer, BS, CCP. Surgery: Robert W. Anderson, MD, Thomas A. DAmico, MD, R. Duane Davis, MD, Donald D. Glower, MD, R. David Harpole, MD, James Jaggers, MD, Robert H. Jones, MD, Kevin Landolfo, 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.
Appendix 2: Cardiothoracic Anesthesia Research Endeavors (CARE) Investigators of the Duke Heart Center
Director: M. Newman, MD. Anesthesiology: F. Clements, MD, N. de Bruijn, MD, K. Grichnik, MD, H. Grocott, MD, S. Hill, MD, J. Mathew, MD, J. Reves, MD, D. Schwinn, MD, M. Stafford Smith, MD, I. Weslby, MD, and J. Booth, MD.
| Acknowledgments |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. W. Hogue, R. Fucetola, T. Hershey, K. Freedland, V. G. Davila-Roman, A. M. Goate, and R. E. Thompson Risk factors for neurocognitive dysfunction after cardiac surgery in postmenopausal women. Ann. Thorac. Surg., August 1, 2008; 86(2): 511 - 516. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. J. Durham and J. P. Gold Late Complications of Cardiac Surgery Card. Surg. Adult, January 1, 2008; 3(2008): 535 - 548. [Full Text] |
||||
![]() |
S. Al-Ruzzeh and D. O'Regan Assessment of Neurocognitive Outcome After Cardiac Surgery Ann. Thorac. Surg., July 1, 2007; 84(1): 358 - 358. [Full Text] [PDF] |
||||
![]() |
Y. Kadoi Pharmacological Neuroprotection During Cardiac Surgery Asian Cardiovasc Thorac Ann, April 1, 2007; 15(2): 167 - 177. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Park, A. Hildreth, R. Thomson, and J. O'Connell Non-valvular atrial fibrillation and cognitive decline: a longitudinal cohort study Age Ageing, March 1, 2007; 36(2): 157 - 163. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Guler, C. Ozkara, H. Dulger, V. Kutay, M. Sahin, E. Erbilen, and H. A. Gumrukcuoglu Do Cardiac Neuropeptides Play a Role in the Occurrence of Atrial Fibrillation After Coronary Bypass Surgery? Ann. Thorac. Surg., February 1, 2007; 83(2): 532 - 537. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Boodhwani, F. D. Rubens, D. Wozny, R. Rodriguez, A. Alsefaou, P. J. Hendry, and H. J. Nathan Predictors of Early Neurocognitive Deficits in Low-Risk Patients Undergoing On-Pump Coronary Artery Bypass Surgery Circulation, July 4, 2006; 114(1_suppl): I-461 - I-466. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Al-Ruzzeh, S. George, M. Bustami, J. Wray, C. Ilsley, T. Athanasiou, and M. Amrani Effect of off-pump coronary artery bypass surgery on clinical, angiographic, neurocognitive, and quality of life outcomes: randomised controlled trial BMJ, June 10, 2006; 332(7554): 1365. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Whitaker, J. Stygall, M. Harrison, and S. Newman Relationship between white cell count, neuropsychologic outcome, and microemboli in 161 patients undergoing coronary artery bypass surgery J. Thorac. Cardiovasc. Surg., June 1, 2006; 131(6): 1358 - 1363. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Sendelbach, R. Lindquist, S. Watanuki, and K. Savik Correlates of Neurocognitive Function of Patients After Off-Pump Coronary Artery Bypass Surgery Am. J. Crit. Care., May 1, 2006; 15(3): 290 - 298. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. C Whitaker, J. Stygall, C. Hope-Wynne, R. K Walesby, M. J. Harrison, and S. P Newman A prospective clinical study of cerebral microemboli and neuropsychological outcome comparing vent-line and auto-venting arterial line filters: both filters are equally safe Perfusion, March 1, 2006; 21(2): 83 - 86. [Abstract] [PDF] |
||||
![]() |
M. K. Chung, L. Shemanski, D. G. Sherman, H. L. Greene, D. B. Hogan, J. C. Kellen, S. G. Kim, L. W. Martin, Y. Rosenberg, D. G. Wyse, et al. Functional Status in Rate- Versus Rhythm-Control Strategies for Atrial Fibrillation: Results of the Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) Functional Status Substudy J. Am. Coll. Cardiol., November 15, 2005; 46(10): 1891 - 1899. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. L. Fontes, J. P. Mathew, H. M. Rinder, D. Zelterman, B. R. Smith, C. S. Rinder, and the Multicenter Study of Perioperative Ischemia (M Atrial Fibrillation After Cardiac Surgery/Cardiopulmonary Bypass Is Associated with Monocyte Activation Anesth. Analg., July 1, 2005; 101(1): 17 - 23. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Kailasam, C. A. Palin, and C. W. Hogue Jr Atrial Fibrillation After Cardiac Surgery: An Evidence-Based Approach to Prevention Seminars in Cardiothoracic and Vascular Anesthesia, March 1, 2005; 9(1): 77 - 85. [Abstract] [PDF] |
||||
![]() |
C. A. Palin, R. Kailasam, and C. W. Hogue Jr Atrial Fibrillation After Cardiac Surgery: Pathophysiology and Treatment Seminars in Cardiothoracic and Vascular Anesthesia, September 1, 2004; 8(3): 175 - 183. [Abstract] [PDF] |
||||
![]() |
D. Zimpfer, M. Czerny, F. Vogt, P. Schuch, L. Kramer, E. Wolner, and M. Grimm Neurocognitive deficit following coronary artery bypass grafting: a prospective study of surgical patients and nonsurgical controls Ann. Thorac. Surg., August 1, 2004; 78(2): 513 - 518. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Mathew, M. L. Fontes, I. C. Tudor, J. Ramsay, P. Duke, C. D. Mazer, P. G. Barash, P. H. Hsu, and D. T. Mangano A Multicenter Risk Index for Atrial Fibrillation After Cardiac Surgery JAMA, April 14, 2004; 291(14): 1720 - 1729. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. N. Patel, B. L. Hamman, A. N. Patel, R. F. Hebeler, R. E. Wood, C. A. Cockerham, B. A. Willey, and H. C. Urschel Jr Epicardial atrial defibrillation: successful treatment of postoperative atrial fibrillation Ann. Thorac. Surg., March 1, 2004; 77(3): 831 - 837. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. N. Wijeysundera, W. S. Beattie, V. Rao, J. Ivanov, and K. Karkouti Calcium antagonists are associated with reduced mortality after cardiac surgery: a propensity analysis J. Thorac. Cardiovasc. Surg., March 1, 2004; 127(3): 755 - 762. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Djaiani, B. Phillips-Bute, M. Podgoreanu, R. H. Messier, J. P. Mathew, F. Clements, and M. F. Newman The Association of Patent Foramen Ovale and Atrial Fibrillation After Coronary Artery Bypass Graft Surgery Anesth. Analg., March 1, 2004; 98(3): 585 - 589. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. N. Wijeysundera, W. S. Beattie, V. Rao, and J. Karski Calcium antagonists reduce cardiovascular complications after cardiac surgery: A meta-analysis J. Am. Coll. Cardiol., May 7, 2003; 41(9): 1496 - 1505. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. Albahrani, M. Swaminathan, B. Phillips-Bute, P. K. Smith, M. F. Newman, J. P. Mathew, and M. Stafford-Smith Postcardiac Surgery Complications: Association of Acute Renal Dysfunction and Atrial Fibrillation Anesth. Analg., March 1, 2003; 96(3): 637 - 643. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Zimpfer, M. Czerny, J. Kilo, M.-T. Kasimir, C. Madl, L. Kramer, G. M. Wieselthaler, E. Wolner, and M. Grimm Cognitive deficit after aortic valve replacement Ann. Thorac. Surg., August 1, 2002; 74(2): 407 - 412. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|