- Email: [email protected]
Postoperative Atrial Fibrillation After Isolated Aortic Valve Replacement: A Cause for Concern?
Akshat Saxena, BMedSc, MBBS, William Y. Shi, BMedSc, MBBS, Shaneel Bappayya, BMedSc, Diem T. Dinh, BS, PhD, Julian A. Smith, FRACS, MS, Christopher M. Reid, MS, PhD, Gilbert C. Shardey, MBBS, FRACS, and Andrew E. Newcomb, MBBS, FRACS Department of Cardiothoracic Surgery, St. Vincent’s Hospital Melbourne, Fitzroy; Department of Epidemiology and Preventative Medicine, Monash University, Prahran; Department of Surgery (MMC), Monash University and Department of Cardiothoracic Surgery, Monash Medical Centre, Clayton; Cabrini Medical Centre, Malvern; and University of Melbourne Department of Surgery, St. Vincent’s Hospital Melbourne, Fitzroy, Victoria, Australia
Background. Several studies have shown that postoperative atrial fibrillation (POAF) is associated with poorer short-term and long-term outcomes after general cardiac operations. There is, however, a paucity of data on the impact of POAF on outcomes after isolated aortic valve replacement (AVR). Methods. Data for all patients undergoing isolated first-time AVR between June 2001 and December 2009 was obtained from the Australasian Society of Cardiac and Thoracic Surgeons (ASCTS) National Cardiac Surgery Database Program and a retrospective analysis was conducted. Preoperative characteristics, early postoperative outcome, and late survival were compared between patients in whom POAF developed and those in whom it did not. Propensity score matching was performed to correct for differences between the 2 groups. Results. Excluding patients with preoperative arrhythmia, isolated first-time AVR was performed in 2,065 patients. POAF developed in 725 (35.1%) of them. Patients with POAF were significantly older (mean age, 72
versus 65 years; p < 0.001) and presented more often with comorbidities, including hypertension, respiratory disease, and hypercholesterolemia (all p < 0.05). From the initial study population, 592 propensity-matched patient pairs were derived; the overall matching rate was 81.7%. In the matched groups, 30-day mortality was not significantly different between the POAF and non-POAF groups (1.5% versus 1%; p ⴝ 0.48). Patients with POAF were, however, at an independently increased risk of perioperative complications, including new renal failure, gastrointestinal complications, and 30-day readmission (p < 0.05). Seven-year mortality was not significantly different between POAF and non-POAF groups (78% versus 83%; p ⴝ 0.63). Conclusions. POAF is a risk factor for short-term morbidity but is not associated with a higher rate of early or late mortality after isolated AVR.
A
trial fibrillation (AF) is the most prevalent complication after cardiac operations, with a reported incidence of 17% to 35% after coronary artery bypass grafting (CABG) and more than 40% after valvular surgery [1– 4]. The health and economic repercussions of this complication are substantial. Postoperative AF (POAF) has been associated with a number of adverse outcomes, most importantly a higher risk of postoperative stroke [5], prolonged in-hospital and intensive care length of stay [6], and greater resource use [(6, 7]. Although predictors and outcomes in patients with POAF have been extensively studied, recent evidence has been presented against previous reports that POAF is transient and of insignificant clinical consequence [1, 8]. Indeed, POAF has been associated with long-term mortality in patients undergoing CABG. A 5-year survival study
undertaken by Mahoney et al [9] showed increased mortality in patients with POAF undergoing CABG. Mariscalco et al [10] also demonstrated that POAF significantly reduced long-term survival in patients undergoing CABG (hazard ratio [HR], 2.56; 95% confidence interval [CI], 1.50 – 4.37). However, although POAF occurs more frequently after valvular operations [2, 5], evidence relating to the effect of POAF after valvular operations on late mortality remains scarce and ambiguous. In the present study, we investigate the effect of POAF on late mortality in a cohort of 2,335 patients receiving primary aortic valve replacement (AVR) at hospitals in Australia participating in the Australasian Society of Cardiac and Thoracic Surgeons (ASCTS) Cardiac Surgery Database between June 1, 2001 and December 31, 2009.
Accepted for publication Aug 14, 2012.
Patients and Methods
Address correspondence to Dr Newcomb, Department of Cardiothoracic Surgery, St. Vincent’s Hospital Melbourne, Fitzroy, Victoria, Australia; e-mail: [email protected].
The inclusion criterion for the study was patients undergoing isolated first-time AVR between June 1, 2001 and
© 2013 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2013;95:133– 40) © 2013 by The Society of Thoracic Surgeons
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2012.08.077
ADULT CARDIAC
Postoperative Atrial Fibrillation After Isolated Aortic Valve Replacement: A Cause for Concern?
134 ADULT CARDIAC
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
December 31, 2009 at hospitals in Australia participating in the ASCTS Cardiac Surgery Database. Patients having concomitant CABG or other concurrent cardiac surgical procedures were excluded from this study. Patients undergoing reoperations were also excluded. Moreover, only patients with documented preoperative sinus rhythm without a history of AF were included. All 6 Victorian public hospitals that perform adult cardiac operations—The Royal Melbourne Hospital, The Alfred Hospital, Monash Medical Centre, The Geelong Hospital, Austin Hospital, and St. Vincent’s Hospital Melbourne—were involved in the prospective data collection during the entire period. Additionally, 14 cardiac surgical units from South Australia, New South Wales, and Queensland entered the database project in the last 30 months of the study period and contributed 41.4% of the total patient numbers. The ASCTS database contained detailed information on patient demographics, preoperative risk factors, operative details, postoperative hospital course, and morbidity and mortality outcomes. These data were collected prospectively using a standardized dataset and definitions. Data collection and audit methods have been previously described [11]. In the state of Victoria, the collection and reporting of public hospital cardiac surgical procedure data is compulsory and mandated by the state government; hence it is all-inclusive. Data validation has been a major focus since the establishment of the ASCTS database. The data are subjected to both local validation and an external data quality audit program, which is performed on site to evaluate the completeness (defined as ⬍ 1% missing data for any variable) and accuracy (97.4%) of the data held in the combined database. Audit outcomes are used to assist in further development of appropriate standards. The ethics committee of each participating hospital had previously approved the use of deidentified patient data contained within the database for research and waived the need for individual patient consent. POAF was defined as evidence of new AF that required treatment and was discovered by electrocardiography or continuous monitoring during the postoperative period. Although treatment of AF may vary slightly between hospitals, it is general practice in the participating institutions to restore sinus rhythm in most patients within 24 hours after the onset of POAF with the use of electrolyte replacement, antiarrhythmic drugs (AADs), or electrical cardioversion. Patients who are discharged home in AF are maintained on warfarin (in the absence of any contraindication) and usually referred for cardioversion after 3 to 6 weeks. Patients discharged home on AADs are followed up in clinic in 6 weeks. In the absence of evidence of AF recurrence, the AADs are discontinued. The decision to stop warfarin in this instance was left to the discretion of the treating physician. For the purpose of this study, patients were separated into 2 groups based on whether POAF developed (POAF group) or did not develop (no POAF group). Preoperative characteristics, early outcomes, and long-term survival were compared between the 2 groups. Late mortality was
Ann Thorac Surg 2013;95:133– 40
defined as death from any cause that occurred at any time after hospital discharge. Eleven early postoperative outcomes were analyzed: (1) 30-day mortality, defined as death within 30 days of operation; (2) permanent stroke, defined as a new central neurologic deficit persisting for longer than 72 hours; (3) postoperative acute myocardial infarction, defined as at least 2 of the following: enzyme level elevation, new cardiac wall motion abnormalities, or new Q waves on serial electrocardiograms; (4) new renal failure, defined as at least 2 of the following: serum creatinine level increased to more than 200 mol/L, a doubling or greater increase in creatinine levels versus preoperative values, or a new requirement for dialysis or hemofiltration; (5) prolonged ventilation (⬎ 24 hours); (6) readmission within 30 days of operation, defined as readmission as an in-patient within 30 days from the date of operation for any reason; (7) return to the operating room for any cause; (8) return to the operating room for bleeding; (9) gastrointestinal complications, defined as postoperative occurrence of any gastrointestinal complications; (10) red blood cell (RBC) transfusion, defined as transfusion of RBCs from the commencement of the operation to discharge; and (11) length of hospital stay in days. We also examined a composite endpoint of “any mortality/ morbidity,” which encompasses the events just listed (excluding RBC transfusion and length of hospital stay).
Statistical Analysis Categorical variables were compared using Fisher’s exact and 2 tests. Continuous variables were expressed as mean ⫾ standard deviation and compared—in the unmatched sample— using the unpaired t test. The KaplanMeier method was used to analyze survival. Propensity-score matching was performed to account for differences in baseline preoperative characteristics between patients in whom POAF developed and those in whom it did not. A propensity score was generated for each patient in the standard fashion by performing a logistic regression, with POAF as the dependent variable. Baseline clinical variables, which were expected to influence outcomes after isolated AVR, were included. These are summarized in Table 1. The C-statistic was calculated for the propensity model. Once generated, patients were matched 1:1 on their propensity score without replacement, using the “greedy” matching method with a fixed caliper width of 0.020. After matching, standardized differences were used to assess the degree of baseline variable balance in the manner described by Austin [12, 13]. A high degree of balance is reflected by a standardized difference of 10% or less. Standardized differences were also calculated and presented for the entire unmatched population to aid reader identification of imbalanced baseline variables. In the matched sample, paired t tests were used for continuous data, whereas McNemar’s 2 test—which compared the discordance of 2 dichotomous variables—was used to compare categorical postoperative outcomes. For late survival, the technique proposed by Klein and Moeschberger was used [13].
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
135
Table 1. Preoperative and Intraoperative Characteristics of Unmatched Patients With and Without Postoperative Atrial Fibrillation Variables Total number of patients (%) Preoperative Variables Age (y, mean ⫾ SD) ⬎ 70 (%) ⱕ 70 (%) Male sex (%) Diabetes (%) Hypercholesterolemia (%) Hypertension (%) Obesity (%) Creatinine ⬎ 200 (%) Preoperative dialysis (%) Estimated glomerular filtration rate (%) Cerebrovascular disease (%) Peripheral vascular disease (%) Chronic pulmonary disease (%) New York Heart Association classification I/II III/IV Missing Previous myocardial infarction (%) Active endocarditis (%) Urgency Elective (%) Urgent (%) Emergency (%) Mechanical prosthesis (%) Congestive heart failure background (%) Congestive heart failure at admission (%) Cardiogenic shock (%) Previous percutaneous coronary intervention (%) Left ventricular function Normal (ejection fraction ⬎ 0.60) (%) Mild (ejection fraction ⬎ 0.45) (%) Moderate (ejection fraction 0.30–0.45) (%) Severe (ejection fraction ⬍ 0.30) (%) Missing Aortic stenosis (%) Aortic regurgitation (%) None/Mild (%) Moderate/severe (%) Missing (%) Mitral stenosis (%) Mitral regurgitation (%) None/Mild (%) Moderate/severe (%) Intraoperative Variables Aortic cross-clamp time (%) (min) (mean ⫾ SD) ⬎ 90 (%) ⱕ 90 (%) Missing
All Patients
No POAF
POAF
p Value
Standardized Difference
2,065
1,340
725
68 ⫾ 13 990 (47.9) 1,075 (52.1) 1,145 (55.4) 399 (19.3) 970 (47) 1,319 (63.9) 758 (36.7) 42 (2) 30 (1.5) 78 ⫾ 36 184 (8.9) 89 (4.3) 284 (13.8)
65 ⫾ 14 556 (41.5) 784 (58.5) 776 (57.9) 256 (19.1) 597 (44.6) 811 (60.5) 484 (36.1) 32 (2.4) 21 (1.6) 82 ⫾ 37 116 (8.7) 54 (4) 154 (11.5)
72 ⫾ 11 434 (59.9) 291 (40.1) 369 (50.9) 143 (19.7) 373 (51.4) 508 (70.1) 274 (37.8) 10 (1.4) 9 (1.2) 71 ⫾ 32 68 (9.4) 35 (4.8) 130 (17.9)
⬍ 0.001 ... ⬍ 0.001 0.003 0.73 0.003 ⬍ 0.001 0.47 0.14 0.7 ⬍ 0.001 0.57 0.43 ⬍ 0.001
50.3 37.4 ⫺37.4 ⫺14.1 1.6 13.8 20.2 3.5 ⫺7.4 ⫺2.8 ⫺32.0 2.5 3.9 18.3
1,162 (56.3) 850 (41.2) 53 (2.6) 86 (4.2) 55 (2.7)
770 (57.5) 528 (39.4) 42 (3.1) 53 (4) 43 (3.2)
392 (54.1) 322 (44.4) 11 (1.5) 33 (4.6) 12 (1.7)
0.013 0.56 0.044
⫺6.8 10.2 ⫺10.7 3.0 ⫺10.1
1,771 (85.8) 271 (13.1) 23 (1.1) 649 (31.4) 695 (33.7) 231 (11.2) 11 (0.5) 70 (3.4)
1146 (85.5) 175 (13.1) 19 (1.4) 499 (37.2) 437 (32.6) 158 (11.8) 7 (0.5) 39 (2.9)
625 (86.2) 96 (13.2) 4 (0.6) 150 (20.7) 258 (35.6) 73 (10.1) 4 (0.6) 31 (4.3)
... ... 0.2 ⬍ 0.001 0.17 0.24 ⬎0.99 0.13
2.0 0.5 ⫺8.8 ⫺37.1 6.3 ⫺5.5 0.4 7.3
1,351 (65.4) 437 (21.2) 142 (6.9) 72 (3.5) 63 (3.1) 1,802 (87.3)
880 (65.7) 282 (21) 88 (6.6) 47 (3.5) 43 (3.2) 1,146 (85.5)
471 (65) 155 (21.4) 54 (7.4) 25 (3.4) 20 (2.8) 656 (90.5)
... ... ... 0.92 0.92 0.001
⫺1.5 0.8 3.5 ⫺0.3 ⫺2.6 15.3
1,443 (69.9) 591 (28.6) 31 (1.5) 35 (1.7)
923 (68.9) 394 (29.4) 23 (1.7) 28 (2.1)
520 (71.7) 197 (27.2) 8 (1.1) 7 (1)
... ... 0.28 0.07
6.2 ⫺5.0 ⫺5.2 ⫺9.2
2,011 (97.4) 54 (2.6)
1,303 (97.2) 37 (2.8)
708 (97.7) 17 (2.3)
... 0.67
2.6 ⫺2.6
74 ⫾ 23 396 (19.2) 1,666 (80.7) 3 (0)
73 ⫾ 24 240 (17.9) 1,097 (81.9) 3 (0.2)
75 ⫾ 22 156 (21.5) 569 (78.5) 0 (0)
0.14 ... ... 0.065
6.9 9.1 ⫺8.5 0.0 (Continued)
ADULT CARDIAC
Ann Thorac Surg 2013;95:133– 40
136 ADULT CARDIAC
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
Ann Thorac Surg 2013;95:133– 40
Table 1. Continued Variables
All Patients
No POAF
POAF
p Value
Standardized Difference
Cardiopulmonary bypass time (%) (min) (mean ⫾ SD) ⬎ 120 (%) ⱕ 120 (%) Missing (%)
98 ⫾ 34 350 (16.9) 1,713 (83) 2 (0)
97 ⫾ 36 215 (16) 1,123 (83.8) 2 (0.1)
100 ⫾ 32 135 (18.6) 590 (81.4) 0 (0)
0.11 ... ... 0.2
7.5 6.8 ⫺6.4 ⫺5.5
POAF ⫽ postoperative atrial fibrillation;
SD ⫽ standard deviation.
Statistical analysis was performed using SPSS for Windows, version 17.0 (IBM Corp, Armonk NY).
Results Patient Demographics and Preoperative Variables Isolated AVR was undertaken in 2,790 patients; of these patients 2,335 did not present with any arrhythmia and are the principal subjects of the current study. Preoperative and demographic characteristics of patients with and without POAF are provided in Table 1. Patients with POAF were older and more likely to present with chronic obstructive pulmonary disease, hypercholesterolemia, hypertension, and cerebrovascular disease. They were also more likely to present with marked or severe symptoms from their heart disease (New York Heart Association [NYHA] class III or IV).
Patient Characteristics Isolated first-time AVR was undertaken in 2,065 patients at 20 Australian institutions. Of these, POAF developed in 725 (35.1%) patients. Preoperative and intraoperative characteristics of patients with and without POAF are provided in Table 1. Inspection of the standardized difference showed substantial clinical differences between the 2 unmatched groups. Of 28 baseline and intraoperative variables, 10 were poorly balanced with a standardized difference greater than 10%.
Outcomes Overall 30-day mortality for the entire cohort of 2,065 patients was 1.2%. The unadjusted 30-day mortality was 0.8% in patients without POAF and 1.9% in patients in whom POAF developed. This difference was significant on univariate analysis (p ⫽ 0.035). A comparison of other early outcomes between the unmatched groups is provided in Table 2. The mean follow-up for this study was 44 ⫾ 30 months (range, 0 –106 months). Long-term survival at 1, 3, 5, and 7 years postoperatively was significantly lower in patients in whom POAF developed compared with those in whom it did not (95% versus 97%, 91% versus 93%, 86% versus 88%, 76% versus 84%, respectively; log-rank p ⫽ 0.02) (Fig 1).
Propensity-Matched Patients A propensity-score model was constructed by adjusting for the preoperative and intraoperative variables outlined in Table 1. The model performs well with a C-sta-
tistic of 0.67. Overall, 592 patients with POAF were matched 1:1 to nonpatients with POAF, thus representing an 81.7% matching rate. All of the 28 baseline preoperative and intraoperative variables were well balanced, with a standardized difference of 10% or less (Table 3). Among the 592 propensity-matched patient pairs, patients with POAF were more likely to experience new renal failure, gastrointestinal complications, readmission within 30 days of operation, and any mortality/morbidity. Propensity-matched patients with POAF also had significantly longer hospital stays. There was, however, no difference in the incidence of 30-day mortality between the 2 groups. The adverse event profile of the 2 groups is summarized in Table 4. There was no significant difference in long-term survival at 1, 3, 5, and 7 years between the propensity-score– matched POAF group and the nonpatient group with POAF (96% versus 96%, 92% versus 92%, 88% versus 87%, 78% versus 83%; log-rank p ⫽ 0.83, KleinMoeschberger p ⫽ 0.63) (Fig 2).
Comment Evidence assessing the effect of POAF on late mortality in patients receiving AVR is scarce and contradictory. Mariscalco et al [14] examined the association between POAF and late mortality in a cohort of 995 patients undergoing valvular operations (including aortic, mitral, and dual valve replacement) at the Heart Centre of the Umeå University Hospital in Sweden and showed that POAF does not independently affect long-term survival (HR, 1.21; 95% CI, 1.08 –1.37). These results were disputed in a study by Filardo et al [15] in which long-term survival was assessed in 1,039 patients receiving either AVR or concomitant AVR and CABG at the Baylor University Medical Center in Dallas, Texas. After adjustment for risk factors for adverse outcomes after cardiac operations identified by the Society of Thoracic Surgeons using a propensity-score method, a significant association was found between POAF and late mortality (HR, 1.48; 95% CI, 1.12–1.96). Although existing data remain ambiguous, our study provides strong evidence for a nonsignificant negative effect of POAF in a primary AVR cohort. Our data indicate that although POAF was associated with late mortality on univariate analysis (p ⫽ 0.035), among propensity-matched patients, there was no significant difference in long-term survival (p ⫽ 0.63). This suggests that a poorer perioperative and intraoperative profile in patients in whom POAF developed is likely to account for
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
137
Table 2. Preoperative and Intraoperative Characteristics of Matched Patients With and Without Postoperative Atrial Fibrillation Variables Total number of patients (%) Preoperative Variables Age (Mean ⫾ SD) ⬎ 70 (%) ⱕ 70 (%) Male sex (%) Diabetes (%) Hypercholesterolemia (%) Hypertension (%) Obesity (%) Creatinine ⬎ 200 (%) Preoperative dialysis (%) Estimated glomerular filtration rate (%) Cerebrovascular disease (%) Peripheral vascular disease (%) Chronic pulmonary pulmonary disease (%) New York Heart Association class I/II (%) III/IV (%) Previous myocardial infarction (%) Active endocarditis (%) Urgency Elective (%) Urgent (%) Emergency (%) Mechanical prosthesis (%) Congestive heart failure background (%) Congestive heart failure at admission (%) NYHA III or IV (%) Cardiogenic shock (%) Previous percutaneous coronary intervention (%) Left ventricular function Normal (ejection fraction ⬎ 0.60) (%) Mild (ejection fraction ⬎ 0.45) (%) Moderate (ejection fraction 0.30–0.45) (%) Severe (ejection fraction ⬍ 0.30) (%) Aortic stenosis (%) Aortic regurgitation (%) None/mild (%) Moderate/severe (%) Mitral stenosis (%) Mitral regurgitation (%) None/mild (%) Moderate/severe (%) Intraoperative Variables Aortic cross-clamp time (%) (min) (mean ⫾ SD) ⬎ 90 (%) ⱕ 90 (%) Cardiopulmonary bypass time (%) (min) (mean ⫾ SD) ⬎ 120 (%) ⱕ 120 (%) NYHA ⫽ New York Heart Association;
All Patients
No POAF
POAF
P Value
Standardized Difference
1,184
592
592
70.2 ⫾ 10.6 486 (41) 202 (17.1) 621 (52.4) 240 (20.3) 597 (50.4) 813 (68.7) 452 (38.2) 17 (1.4) 13 (1.1) 74.9 ⫾ 32.3 109 (9.2) 52 (4.4) 156 (13.2)
70.2 ⫾ 10.6 240 (40.5) 104 (17.6) 303 (51.2) 122 (20.6) 303 (51.2) 410 (69.3) 229 (38.7) 7 (1.2) 4 (0.7) 75.4 ⫾ 32.6 54 (9.1) 27 (4.6) 76 (12.8)
70.2 ⫾ 10.7 246 (41.6) 98 (16.6) 318 (53.7) 118 (19.9) 294 (49.7) 403 (68.1) 223 (37.7) 10 (1.7) 9 (1.5) 74.5 ⫾ 32.0 55 (9.3) 25 (4.2) 80 (13.5)
0.85 ... 0.99 0.42 0.83 0.64 0.71 0.77 0.63 0.26 ... ⬎ 0.99 0.78 0.8
0.0 2.1 ⫺2.7 5.1 ⫺1.7 ⫺3.0 ⫺2.5 ⫺2.1 4.3 8.1 2.8 0.6 ⫺1.6 2.0
671 (32.5) 513 (24.8) 54 (4.6) 23 (1.9)
327 (24.4) 265 (19.8) 29 (4.9) 12 (2)
344 (47.4) 248 (34.2) 25 (4.2) 11 (1.9)
0.94 0.68 ⬎ 0.99
5.8 ⫺5.8 ⫺3.2 ⫺1.2
1,028 (86.8) 147 (12.4) 9 (0.8) 293 (24.7) 416 (35.1) 129 (10.9) 513 (43.3) 4 (0.3) 37 (3.1)
513 (86.7) 74 (12.5) 5 (0.8) 150 (25.3) 208 (35.1) 67 (11.3) 265 (44.8) 2 (0.3) 19 (3.2)
515 (87) 73 (12.3) 4 (0.7) 143 (24.2) 208 (35.1) 62 (10.5) 248 (41.9) 2 (0.3) 18 (3)
... ... 0.94 0.69 ⬎ 0.99 0.71 0.35 ⬎ 0.99 ⬎ 0.99
806 (68.1) 250 (21.1) 83 (7) 45 (3.8) 1,059 (89.4)
402 (67.9) 126 (21.3) 38 (6.4) 26 (4.4) 527 (89)
404 (68.2) 124 (20.9) 45 (7.6) 19 (3.2) 532 (89.9)
... ... ... 0.64 0.71
874 (73.8) 310 (26.2) 12 (1)
433 (73.1) 159 (26.9) 5 (0.8)
441 (74.5) 151 (25.5) 7 (1.2)
... 0.64 0.77
1,156 (97.6) 28 (2.4)
578 (97.6) 14 (2.4)
578 (97.6) 14 (2.4)
... ⬎ 0.99
2.3 ⫺2.7 3.1 0.0 0.0 0.0
74.5 ⫾ 21.6 226 (19.1) 958 (80.9) 98.5 ⫾ 32.5 202 (17.1) 982 (82.9)
74.8 ⫾ 22.3 114 (19.3) 478 (80.7) 98.5 ⫾ 35.9 106 (17.9) 486 (82.1)
74.1 ⫾ 20.9 112 (18.9) 480 (81.1) 98.4 ⫾ 28.7 96 (16.2) 496 (83.8)
0.59 ... 0.94 0.99 ... 0.49
3.2 ⫺0.8 0.6 0.3 ⫺4.0 3.0
POAF ⫽ postoperative atrial fibrillation;
SD ⫽ standard deviation.
1.0 ⫺0.5 ⫺1.9 ⫺2.7 0.0 ⫺2.7 ⫺5.8 0.0 ⫺0.9 0.0 0.6 ⫺0.7 4.2 ⫺5.6 1.6
ADULT CARDIAC
Ann Thorac Surg 2013;95:133– 40
138 ADULT CARDIAC
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
Ann Thorac Surg 2013;95:133– 40
Fig 1. Overall survival of unmatched patients with and without postoperative atrial fibrillation (POAF).
at least some of the survival differences observed. Indeed, patients in the POAF group were older (71.55 ⫾ 10.97 years versus 65.56 ⫾ 14.03 years; p ⬍ 0.001). Advanced age has been previously reported to be an independent predictor of POAF after CABG; it is also a well-known marker for poorer perioperative and longterm outcomes [2, 3, 6]. Patients in the POAF group also presented with more comorbidities—in particular, chronic obstructive pulmonary disease, hypercholesterolemia, hypertension, cerebrovascular disease, and heart failure (NYHA class III or IV). Several mechanisms have been postulated to explain the association of POAF with late mortality. Shinbane et al [16] proposed that AF causes rapid ventricular responses, leading to ventricular dilatation and reduced cardiac output. Furthermore, reduced ventricular filling and circulatory stasis in the left atrium as a result of AF may promote embolus formation and lead to an increased risk of stroke [17]. Petersen et al [18] demon-
strated that POAF may predispose to cerebrovascular complications by impairing cerebral blood flow. However, the exact mechanism by which POAF contributes to poorer long-term survival has not been determined. Although biological explanations for the potentially deleterious impact of POAF on late mortality have been proposed, several studies have demonstrated that it is possible to achieve good outcomes in patients with POAF undergoing AVR through vigilant observation. Mariscalco et al [14] hypothesized that the demonstrated nonsignificant association between POAF and late mortality in patients undergoing AVR was a result of good postoperative follow-up with regular clinic visits and echocardiographic examinations. The authors also postulated a potentially protective effect of anticoagulation therapy from embolic events in patients receiving mechanical valves [14]. Further, we hypothesize that our statistically nonsignificant results may inadvertently be due to current postoperative anticoagulation and antiplatelet regimens. Indeed, most guidelines recommend warfarin for 3 months after tissue AVR, with antiplatelet therapy considered an acceptable alternative [19]. Of particular relevance, the European Association for Cardio-Thoracic Surgery guidelines recommend anticoagulation in patients receiving tissue AVR in whom POAF develops [19]. In contrast, the guidelines state that treatment with antiplatelet therapy is sufficient in patients who underwent AVR but did not experience POAF. The therapeutic value of warfarin in reducing the incidence of postoperative stroke may have contributed to good longterm outcomes of patients in whom POAF developed after AVR. This important clinical question warrants further investigation. Although POAF was not shown to be an independent risk factor for poor long-term survival, it is associated with poorer perioperative outcomes. Propensitymatched patients with POAF were almost twice as likely to experience new renal failure and 6 times as likely to experience gastrointestinal complications. This corresponds to findings from a previous study in a CABG cohort [20]. Importantly, patients with POAF were signif-
Table 3. A Comparison of Early Outcomes of Unmatched Patients With and Without Postoperative Atrial Fibrillation Postoperative Morbidity 30-day mortality (%) Postoperative myocardial infarction (%) Stroke (%) New renal failure (%) Prolonged ventilation (⬎ 24 h) (%) Return to operating room (%) Return to operating room for bleeding (%) Gastrointestinal complications (%) Readmission within 30 d of operation (%) Any Mortality/Morbidity (%) Red blood cell transfusion (%) Length of hospital stay (d) (mean ⫾ SD) POAF ⫽ postoperative atrial fibrillation.
All Patients
No POAF
POAF
p Value
25 (1.2) 5 (0.2) 19 (0.9) 97 (4.7) 135 (6.5) 146 (7.1) 78 (3.8) 26 (1.3) 222 (10.8) 508 (24.6) 785 (38) 10 ⫾ 9
11 (0.8) 3 (0.2) 11 (0.8) 43 (3.2) 70 (5.2) 80 (6) 41 (3.1) 9 (0.7) 130 (9.7) 289 (21.6) 462 (34.5) 9 ⫾ 10
14 (1.9) 2 (0.3) 8 (1.1) 54 (7.4) 65 (9) 66 (9.1) 37 (5.1) 17 (2.3) 92 (12.7) 219 (30.2) 323 (44.6) 11 ⫾ 9
0.035 ⬎ 0.99 0.63 ⬍ 0.001 0.001 0.009 0.022 0.003 0.037 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
139
Table 4. A Comparison of Early Outcomes of Matched Patients With And Without Postoperative Atrial Fibrillation Postoperative Morbidity 30-day mortality (%) Postoperative myocardial infarction (%) Stroke (%) New renal failure (%) Prolonged ventilation (⬎ 24 h) (%) Return to operating room (%) Return to operating room for bleeding (%) Gastrointestinal complications (%) Readmission within 30 d of operation (%) Any mortality/morbidity (%) Red blood cell transfusion (%) Length of hospital stay (d) (mean ⫾ SD) POAF ⫽ postoperative atrial fibrillation;
All Patients
No POAF
POAF
p Value
18 (1.3) 3 (0.2) 11 (0.8) 63 (4.4) 80 (5.6) 85 (4.1) 47 (2.3) 20 (1) 125 (8.7) 296 (20.7) 470 (32.9) 9.7 ⫾ 7.5
7 (1) 1 (0.1) 7 (1) 22 (3.1) 34 (4.8) 36 (2.7) 19 (1.4) 4 (0.3) 49 (6.9) 126 (17.6) 218 (30.5) 8.9 ⫾ 6.7
11 (1.5) 2 (0.3) 4 (0.6) 41 (5.7) 46 (6.4) 49 (6.8) 28 (3.9) 16 (2.2) 76 (10.6) 170 (23.8) 252 (35.2) 10.6 ⫾ 8.1
0.48 ⬎ 0.99 0.55 0.023 0.22 0.18 0.23 0.012 0.014 0.004 0.064 ⬍ 0.001
SD ⫽ standard deviation.
icantly more likely to have overall morbidity/mortality (23.8% versus 17.6%; p ⫽ 0.004). There was a nonsignificant trend toward an increased need for RBC transfusion (35.2% versus 30.5%; p ⫽ 0.064). Whether POAF is cause or consequence to such outcomes can only be speculated upon, given the dynamic complexity of the postoperative setting. However, it is likely that POAF is intricately involved in the development of such adverse outcomes. As such, our data support the need to explore prophylactic strategies aimed at the identification, management, and prevention of POAF. This may then translate into improved perioperative outcomes for patients. The current study has several limitations. First, our study did not include postoperative patient data such as anticoagulation treatments or follow-up data on rates of recurrent AF and cancer. We could not account for prophylactic antiarrhythmic strategies such as the use of AADs. Furthermore, by virtue of the multicenter approach of our study, the effect of surgeon and cardiologist variability, if any, on the intraoperative and postopera-
tive course of patients was not predictable. In addition, it must be emphasized that this observational study establishes correlations only, and as such any direct mechanistic value of our findings should be interpreted with care. The present study shows that POAF was not associated with early or late mortality in a primary AVR cohort. We have also shown that POAF is more common in patients with a higher risk profile and is independently associated with adverse outcomes postoperatively. In conclusion, our study identifies POAF as a key player in the immediate postoperative setting and suggests that a review of current prophylactic and management strategies is necessary with regard to POAF in patients undergoing AVR. The following investigators, data managers, and institutions participated in the ASCTS database: Alfred Hospital: A. Pick, J. Duncan; Austin Hospital: S. Seevanayagam, M. Shaw; Cabrini Health: G. Shardey; Geelong Hospital: M. Morteza, C. Bright; Flinders Medical Centre: J. Knight, R. Baker, J. Helm; Jessie McPherson Private Hospital: J. Smith, H. Baxter; John Hunter Hospital: A. James, S. Scaybrook; Lake Macquarie Hospital: B. Dennett, M. Jacobi; Liverpool Hospital: B. French, N. Hewitt; Mater Health Service Hospital: A.M. Diqer, J. Archer; Monash Medical Centre: J. Smith , H. Baxter; Prince of Wales Hospital: H. Wolfenden, D. Weerasinge; Royal Melbourne Hospital: P. Skillington, S. Law; Royal Prince Alfred Hospital: M. Wilson, L. Turner; St. George Hospital: G. Fermanis, C. Redmond; St. Vincent’s Hospital, VIC: M. Yii, A. Newcomb, J. Mack, K. Duve; St Vincent’s Hospital, NSW: P. Spratt, T, Hunter; The Canberra Hospital: P. Bissaker, K. Butler; Townsville Hospital: R. Tam, A. Farley; Westmead Hospital: R. Costa, M. Halaka. The Australasian Society of Cardiac and Thoracic Surgeons (ASCTS) National Cardiac Surgery Database Program is funded by the Department of Human Services, Victoria, and the Health Administration Corporation (GMCT) and the Clinical Excellence Commission (CEC), NSW.
References Fig 2. Overall survival of matched patients with and without postoperative atrial fibrillation (POAF).
1. Maisel WH, Rawn JD, Stevenson WG. Atrial fibrillation after cardiac surgery. Ann Intern Med 2001;135:1061–73. 2. Almassi GH, Schowalter T, Nicolosi AC, Aggarwal A, Moritz TE, Henderson WG, et al. Atrial fibrillation after cardiac
ADULT CARDIAC
Ann Thorac Surg 2013;95:133– 40
140 ADULT CARDIAC
3. 4. 5. 6.
7. 8. 9.
10. 11.
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
surgery: a major morbid event? Ann Surg 1997;226:501–11; discussion 11–3. Hogue CW, Jr, Hyder ML. Atrial fibrillation after cardiac operation: risks, mechanisms, and treatment. Ann Thorac Surg 2000;69:300 – 6. Aranki SF, Shaw DP, Adams DH, et al. Predictors of atrial fibrillation after coronary artery surgery. Current trends and impact on hospital resources. Circulation 1996;94:390 –7. Creswell LL, Schuessler RB, Rosenbloom M, Cox JL. Hazards of postoperative atrial arrhythmias. Ann Thorac Surg 1993; 56:539 – 49. Mathew JP, Parks R, Savino JS, et al. Atrial fibrillation following coronary artery bypass graft surgery: predictors, outcomes, and resource utilization. MultiCenter Study of Perioperative Ischemia Research Group. JAMA 1996;276: 300 – 6. Hravnak M, Hoffman LA, Saul MI, Zullo TG, Whitman GR. Resource utilization related to atrial fibrillation after coronary artery bypass grafting. Am J Crit Care 2002;11:228 –38. Ahlsson A, Bodin L, Fengsrud E, Englund A. Patients with postoperative atrial fibrillation have a doubled cardiovascular mortality. Scand Cardiovasc J 2009;43:330 – 6. Mahoney EM, Thompson TD, Veledar E, Williams J, Weintraub WS. Cost-effectiveness of targeting patients undergoing cardiac surgery for therapy with intravenous amiodarone to prevent atrial fibrillation. J Am Coll Cardiol 2002;40: 737– 45. Mariscalco G, Klersy C, Zanobini M, et al. Atrial fibrillation after isolated coronary surgery affects late survival. Circulation 2008;118:1612– 8. Dinh DT, Lee GA, Billah B, Smith JA, Shardey GC, Reid CM. Trends in coronary artery bypass graft surgery in Victoria, 2001-2006: findings from the Australasian Society of Cardiac
Ann Thorac Surg 2013;95:133– 40
12.
13.
14. 15.
16.
17. 18. 19. 20.
and Thoracic Surgeons database project. Med J Aust 2008;188:214 –7. Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009;28:3083– 107. Austin PC. Propensity-score matching in the cardiovascular surgery literature from 2004 to 2006: a systematic review and suggestions for improvement. J Thorac Cardiovasc Surg 2007;134:1128 –35. Mariscalco G, Engstrom KG. Postoperative atrial fibrillation is associated with late mortality after coronary surgery, but not after valvular surgery. Ann Thorac Surg 2009;88:1871– 6. Filardo G, Hamilton C, Hamman B, Hebeler RF, Jr, Adams J, Grayburn P. New-onset postoperative atrial fibrillation and long-term survival after aortic valve replacement surgery. Ann Thorac Surg 2010;90:474 –9. Shinbane JS, Wood MA, Jensen DN, Ellenbogen KA, Fitzpatrick AP, Scheinman MM. Tachycardia-induced cardiomyopathy: a review of animal models and clinical studies. J Am Coll Cardiol 1997;29:709 –15. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke 1991;22:983– 8. Petersen P, Kastrup J, Videbaek R, Boysen G. Cerebral blood flow before and after cardioversion of atrial fibrillation. J Cereb Blood Flow Metab 1989;9:422–5. Dunning J, Versteegh M, Fabbri A, et al. Guideline on antiplatelet and anticoagulation management in cardiac surgery. Eur J Cardiothorac Surg 2008;34:73–92. Kalavrouziotis D, Buth KJ, Ali IS. The impact of new-onset atrial fibrillation on in-hospital mortality following cardiac surgery. Chest 2007;131:833–9.
Notice From the American Board of Thoracic Surgery The 2012 Part I (written) examination was held on Monday, November 19, 2012. To be admissible to the Part II (oral) examination, a candidate must have successfully completed the Part I (written) examination. A candidate applying for admission to the certifying examination must fulfill all the requirements of the Board in force at
© 2013 by The Society of Thoracic Surgeons Published by Elsevier Inc
the time the application is received. Please address all communications to the American Board of Thoracic Surgery, 633 N St. Clair St, Suite 2320, Chicago, IL 60611; telephone: (312) 202-5900; fax: (312) 202-5960; email: [email protected].
Ann Thorac Surg 2013;95:140
•
0003-4975/$36.00
Background. Several studies have shown that postoperative atrial fibrillation (POAF) is associated with poorer short-term and long-term outcomes after general cardiac operations. There is, however, a paucity of data on the impact of POAF on outcomes after isolated aortic valve replacement (AVR). Methods. Data for all patients undergoing isolated first-time AVR between June 2001 and December 2009 was obtained from the Australasian Society of Cardiac and Thoracic Surgeons (ASCTS) National Cardiac Surgery Database Program and a retrospective analysis was conducted. Preoperative characteristics, early postoperative outcome, and late survival were compared between patients in whom POAF developed and those in whom it did not. Propensity score matching was performed to correct for differences between the 2 groups. Results. Excluding patients with preoperative arrhythmia, isolated first-time AVR was performed in 2,065 patients. POAF developed in 725 (35.1%) of them. Patients with POAF were significantly older (mean age, 72
versus 65 years; p < 0.001) and presented more often with comorbidities, including hypertension, respiratory disease, and hypercholesterolemia (all p < 0.05). From the initial study population, 592 propensity-matched patient pairs were derived; the overall matching rate was 81.7%. In the matched groups, 30-day mortality was not significantly different between the POAF and non-POAF groups (1.5% versus 1%; p ⴝ 0.48). Patients with POAF were, however, at an independently increased risk of perioperative complications, including new renal failure, gastrointestinal complications, and 30-day readmission (p < 0.05). Seven-year mortality was not significantly different between POAF and non-POAF groups (78% versus 83%; p ⴝ 0.63). Conclusions. POAF is a risk factor for short-term morbidity but is not associated with a higher rate of early or late mortality after isolated AVR.
A
trial fibrillation (AF) is the most prevalent complication after cardiac operations, with a reported incidence of 17% to 35% after coronary artery bypass grafting (CABG) and more than 40% after valvular surgery [1– 4]. The health and economic repercussions of this complication are substantial. Postoperative AF (POAF) has been associated with a number of adverse outcomes, most importantly a higher risk of postoperative stroke [5], prolonged in-hospital and intensive care length of stay [6], and greater resource use [(6, 7]. Although predictors and outcomes in patients with POAF have been extensively studied, recent evidence has been presented against previous reports that POAF is transient and of insignificant clinical consequence [1, 8]. Indeed, POAF has been associated with long-term mortality in patients undergoing CABG. A 5-year survival study
undertaken by Mahoney et al [9] showed increased mortality in patients with POAF undergoing CABG. Mariscalco et al [10] also demonstrated that POAF significantly reduced long-term survival in patients undergoing CABG (hazard ratio [HR], 2.56; 95% confidence interval [CI], 1.50 – 4.37). However, although POAF occurs more frequently after valvular operations [2, 5], evidence relating to the effect of POAF after valvular operations on late mortality remains scarce and ambiguous. In the present study, we investigate the effect of POAF on late mortality in a cohort of 2,335 patients receiving primary aortic valve replacement (AVR) at hospitals in Australia participating in the Australasian Society of Cardiac and Thoracic Surgeons (ASCTS) Cardiac Surgery Database between June 1, 2001 and December 31, 2009.
Accepted for publication Aug 14, 2012.
Patients and Methods
Address correspondence to Dr Newcomb, Department of Cardiothoracic Surgery, St. Vincent’s Hospital Melbourne, Fitzroy, Victoria, Australia; e-mail: [email protected].
The inclusion criterion for the study was patients undergoing isolated first-time AVR between June 1, 2001 and
© 2013 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2013;95:133– 40) © 2013 by The Society of Thoracic Surgeons
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2012.08.077
ADULT CARDIAC
Postoperative Atrial Fibrillation After Isolated Aortic Valve Replacement: A Cause for Concern?
134 ADULT CARDIAC
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
December 31, 2009 at hospitals in Australia participating in the ASCTS Cardiac Surgery Database. Patients having concomitant CABG or other concurrent cardiac surgical procedures were excluded from this study. Patients undergoing reoperations were also excluded. Moreover, only patients with documented preoperative sinus rhythm without a history of AF were included. All 6 Victorian public hospitals that perform adult cardiac operations—The Royal Melbourne Hospital, The Alfred Hospital, Monash Medical Centre, The Geelong Hospital, Austin Hospital, and St. Vincent’s Hospital Melbourne—were involved in the prospective data collection during the entire period. Additionally, 14 cardiac surgical units from South Australia, New South Wales, and Queensland entered the database project in the last 30 months of the study period and contributed 41.4% of the total patient numbers. The ASCTS database contained detailed information on patient demographics, preoperative risk factors, operative details, postoperative hospital course, and morbidity and mortality outcomes. These data were collected prospectively using a standardized dataset and definitions. Data collection and audit methods have been previously described [11]. In the state of Victoria, the collection and reporting of public hospital cardiac surgical procedure data is compulsory and mandated by the state government; hence it is all-inclusive. Data validation has been a major focus since the establishment of the ASCTS database. The data are subjected to both local validation and an external data quality audit program, which is performed on site to evaluate the completeness (defined as ⬍ 1% missing data for any variable) and accuracy (97.4%) of the data held in the combined database. Audit outcomes are used to assist in further development of appropriate standards. The ethics committee of each participating hospital had previously approved the use of deidentified patient data contained within the database for research and waived the need for individual patient consent. POAF was defined as evidence of new AF that required treatment and was discovered by electrocardiography or continuous monitoring during the postoperative period. Although treatment of AF may vary slightly between hospitals, it is general practice in the participating institutions to restore sinus rhythm in most patients within 24 hours after the onset of POAF with the use of electrolyte replacement, antiarrhythmic drugs (AADs), or electrical cardioversion. Patients who are discharged home in AF are maintained on warfarin (in the absence of any contraindication) and usually referred for cardioversion after 3 to 6 weeks. Patients discharged home on AADs are followed up in clinic in 6 weeks. In the absence of evidence of AF recurrence, the AADs are discontinued. The decision to stop warfarin in this instance was left to the discretion of the treating physician. For the purpose of this study, patients were separated into 2 groups based on whether POAF developed (POAF group) or did not develop (no POAF group). Preoperative characteristics, early outcomes, and long-term survival were compared between the 2 groups. Late mortality was
Ann Thorac Surg 2013;95:133– 40
defined as death from any cause that occurred at any time after hospital discharge. Eleven early postoperative outcomes were analyzed: (1) 30-day mortality, defined as death within 30 days of operation; (2) permanent stroke, defined as a new central neurologic deficit persisting for longer than 72 hours; (3) postoperative acute myocardial infarction, defined as at least 2 of the following: enzyme level elevation, new cardiac wall motion abnormalities, or new Q waves on serial electrocardiograms; (4) new renal failure, defined as at least 2 of the following: serum creatinine level increased to more than 200 mol/L, a doubling or greater increase in creatinine levels versus preoperative values, or a new requirement for dialysis or hemofiltration; (5) prolonged ventilation (⬎ 24 hours); (6) readmission within 30 days of operation, defined as readmission as an in-patient within 30 days from the date of operation for any reason; (7) return to the operating room for any cause; (8) return to the operating room for bleeding; (9) gastrointestinal complications, defined as postoperative occurrence of any gastrointestinal complications; (10) red blood cell (RBC) transfusion, defined as transfusion of RBCs from the commencement of the operation to discharge; and (11) length of hospital stay in days. We also examined a composite endpoint of “any mortality/ morbidity,” which encompasses the events just listed (excluding RBC transfusion and length of hospital stay).
Statistical Analysis Categorical variables were compared using Fisher’s exact and 2 tests. Continuous variables were expressed as mean ⫾ standard deviation and compared—in the unmatched sample— using the unpaired t test. The KaplanMeier method was used to analyze survival. Propensity-score matching was performed to account for differences in baseline preoperative characteristics between patients in whom POAF developed and those in whom it did not. A propensity score was generated for each patient in the standard fashion by performing a logistic regression, with POAF as the dependent variable. Baseline clinical variables, which were expected to influence outcomes after isolated AVR, were included. These are summarized in Table 1. The C-statistic was calculated for the propensity model. Once generated, patients were matched 1:1 on their propensity score without replacement, using the “greedy” matching method with a fixed caliper width of 0.020. After matching, standardized differences were used to assess the degree of baseline variable balance in the manner described by Austin [12, 13]. A high degree of balance is reflected by a standardized difference of 10% or less. Standardized differences were also calculated and presented for the entire unmatched population to aid reader identification of imbalanced baseline variables. In the matched sample, paired t tests were used for continuous data, whereas McNemar’s 2 test—which compared the discordance of 2 dichotomous variables—was used to compare categorical postoperative outcomes. For late survival, the technique proposed by Klein and Moeschberger was used [13].
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
135
Table 1. Preoperative and Intraoperative Characteristics of Unmatched Patients With and Without Postoperative Atrial Fibrillation Variables Total number of patients (%) Preoperative Variables Age (y, mean ⫾ SD) ⬎ 70 (%) ⱕ 70 (%) Male sex (%) Diabetes (%) Hypercholesterolemia (%) Hypertension (%) Obesity (%) Creatinine ⬎ 200 (%) Preoperative dialysis (%) Estimated glomerular filtration rate (%) Cerebrovascular disease (%) Peripheral vascular disease (%) Chronic pulmonary disease (%) New York Heart Association classification I/II III/IV Missing Previous myocardial infarction (%) Active endocarditis (%) Urgency Elective (%) Urgent (%) Emergency (%) Mechanical prosthesis (%) Congestive heart failure background (%) Congestive heart failure at admission (%) Cardiogenic shock (%) Previous percutaneous coronary intervention (%) Left ventricular function Normal (ejection fraction ⬎ 0.60) (%) Mild (ejection fraction ⬎ 0.45) (%) Moderate (ejection fraction 0.30–0.45) (%) Severe (ejection fraction ⬍ 0.30) (%) Missing Aortic stenosis (%) Aortic regurgitation (%) None/Mild (%) Moderate/severe (%) Missing (%) Mitral stenosis (%) Mitral regurgitation (%) None/Mild (%) Moderate/severe (%) Intraoperative Variables Aortic cross-clamp time (%) (min) (mean ⫾ SD) ⬎ 90 (%) ⱕ 90 (%) Missing
All Patients
No POAF
POAF
p Value
Standardized Difference
2,065
1,340
725
68 ⫾ 13 990 (47.9) 1,075 (52.1) 1,145 (55.4) 399 (19.3) 970 (47) 1,319 (63.9) 758 (36.7) 42 (2) 30 (1.5) 78 ⫾ 36 184 (8.9) 89 (4.3) 284 (13.8)
65 ⫾ 14 556 (41.5) 784 (58.5) 776 (57.9) 256 (19.1) 597 (44.6) 811 (60.5) 484 (36.1) 32 (2.4) 21 (1.6) 82 ⫾ 37 116 (8.7) 54 (4) 154 (11.5)
72 ⫾ 11 434 (59.9) 291 (40.1) 369 (50.9) 143 (19.7) 373 (51.4) 508 (70.1) 274 (37.8) 10 (1.4) 9 (1.2) 71 ⫾ 32 68 (9.4) 35 (4.8) 130 (17.9)
⬍ 0.001 ... ⬍ 0.001 0.003 0.73 0.003 ⬍ 0.001 0.47 0.14 0.7 ⬍ 0.001 0.57 0.43 ⬍ 0.001
50.3 37.4 ⫺37.4 ⫺14.1 1.6 13.8 20.2 3.5 ⫺7.4 ⫺2.8 ⫺32.0 2.5 3.9 18.3
1,162 (56.3) 850 (41.2) 53 (2.6) 86 (4.2) 55 (2.7)
770 (57.5) 528 (39.4) 42 (3.1) 53 (4) 43 (3.2)
392 (54.1) 322 (44.4) 11 (1.5) 33 (4.6) 12 (1.7)
0.013 0.56 0.044
⫺6.8 10.2 ⫺10.7 3.0 ⫺10.1
1,771 (85.8) 271 (13.1) 23 (1.1) 649 (31.4) 695 (33.7) 231 (11.2) 11 (0.5) 70 (3.4)
1146 (85.5) 175 (13.1) 19 (1.4) 499 (37.2) 437 (32.6) 158 (11.8) 7 (0.5) 39 (2.9)
625 (86.2) 96 (13.2) 4 (0.6) 150 (20.7) 258 (35.6) 73 (10.1) 4 (0.6) 31 (4.3)
... ... 0.2 ⬍ 0.001 0.17 0.24 ⬎0.99 0.13
2.0 0.5 ⫺8.8 ⫺37.1 6.3 ⫺5.5 0.4 7.3
1,351 (65.4) 437 (21.2) 142 (6.9) 72 (3.5) 63 (3.1) 1,802 (87.3)
880 (65.7) 282 (21) 88 (6.6) 47 (3.5) 43 (3.2) 1,146 (85.5)
471 (65) 155 (21.4) 54 (7.4) 25 (3.4) 20 (2.8) 656 (90.5)
... ... ... 0.92 0.92 0.001
⫺1.5 0.8 3.5 ⫺0.3 ⫺2.6 15.3
1,443 (69.9) 591 (28.6) 31 (1.5) 35 (1.7)
923 (68.9) 394 (29.4) 23 (1.7) 28 (2.1)
520 (71.7) 197 (27.2) 8 (1.1) 7 (1)
... ... 0.28 0.07
6.2 ⫺5.0 ⫺5.2 ⫺9.2
2,011 (97.4) 54 (2.6)
1,303 (97.2) 37 (2.8)
708 (97.7) 17 (2.3)
... 0.67
2.6 ⫺2.6
74 ⫾ 23 396 (19.2) 1,666 (80.7) 3 (0)
73 ⫾ 24 240 (17.9) 1,097 (81.9) 3 (0.2)
75 ⫾ 22 156 (21.5) 569 (78.5) 0 (0)
0.14 ... ... 0.065
6.9 9.1 ⫺8.5 0.0 (Continued)
ADULT CARDIAC
Ann Thorac Surg 2013;95:133– 40
136 ADULT CARDIAC
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
Ann Thorac Surg 2013;95:133– 40
Table 1. Continued Variables
All Patients
No POAF
POAF
p Value
Standardized Difference
Cardiopulmonary bypass time (%) (min) (mean ⫾ SD) ⬎ 120 (%) ⱕ 120 (%) Missing (%)
98 ⫾ 34 350 (16.9) 1,713 (83) 2 (0)
97 ⫾ 36 215 (16) 1,123 (83.8) 2 (0.1)
100 ⫾ 32 135 (18.6) 590 (81.4) 0 (0)
0.11 ... ... 0.2
7.5 6.8 ⫺6.4 ⫺5.5
POAF ⫽ postoperative atrial fibrillation;
SD ⫽ standard deviation.
Statistical analysis was performed using SPSS for Windows, version 17.0 (IBM Corp, Armonk NY).
Results Patient Demographics and Preoperative Variables Isolated AVR was undertaken in 2,790 patients; of these patients 2,335 did not present with any arrhythmia and are the principal subjects of the current study. Preoperative and demographic characteristics of patients with and without POAF are provided in Table 1. Patients with POAF were older and more likely to present with chronic obstructive pulmonary disease, hypercholesterolemia, hypertension, and cerebrovascular disease. They were also more likely to present with marked or severe symptoms from their heart disease (New York Heart Association [NYHA] class III or IV).
Patient Characteristics Isolated first-time AVR was undertaken in 2,065 patients at 20 Australian institutions. Of these, POAF developed in 725 (35.1%) patients. Preoperative and intraoperative characteristics of patients with and without POAF are provided in Table 1. Inspection of the standardized difference showed substantial clinical differences between the 2 unmatched groups. Of 28 baseline and intraoperative variables, 10 were poorly balanced with a standardized difference greater than 10%.
Outcomes Overall 30-day mortality for the entire cohort of 2,065 patients was 1.2%. The unadjusted 30-day mortality was 0.8% in patients without POAF and 1.9% in patients in whom POAF developed. This difference was significant on univariate analysis (p ⫽ 0.035). A comparison of other early outcomes between the unmatched groups is provided in Table 2. The mean follow-up for this study was 44 ⫾ 30 months (range, 0 –106 months). Long-term survival at 1, 3, 5, and 7 years postoperatively was significantly lower in patients in whom POAF developed compared with those in whom it did not (95% versus 97%, 91% versus 93%, 86% versus 88%, 76% versus 84%, respectively; log-rank p ⫽ 0.02) (Fig 1).
Propensity-Matched Patients A propensity-score model was constructed by adjusting for the preoperative and intraoperative variables outlined in Table 1. The model performs well with a C-sta-
tistic of 0.67. Overall, 592 patients with POAF were matched 1:1 to nonpatients with POAF, thus representing an 81.7% matching rate. All of the 28 baseline preoperative and intraoperative variables were well balanced, with a standardized difference of 10% or less (Table 3). Among the 592 propensity-matched patient pairs, patients with POAF were more likely to experience new renal failure, gastrointestinal complications, readmission within 30 days of operation, and any mortality/morbidity. Propensity-matched patients with POAF also had significantly longer hospital stays. There was, however, no difference in the incidence of 30-day mortality between the 2 groups. The adverse event profile of the 2 groups is summarized in Table 4. There was no significant difference in long-term survival at 1, 3, 5, and 7 years between the propensity-score– matched POAF group and the nonpatient group with POAF (96% versus 96%, 92% versus 92%, 88% versus 87%, 78% versus 83%; log-rank p ⫽ 0.83, KleinMoeschberger p ⫽ 0.63) (Fig 2).
Comment Evidence assessing the effect of POAF on late mortality in patients receiving AVR is scarce and contradictory. Mariscalco et al [14] examined the association between POAF and late mortality in a cohort of 995 patients undergoing valvular operations (including aortic, mitral, and dual valve replacement) at the Heart Centre of the Umeå University Hospital in Sweden and showed that POAF does not independently affect long-term survival (HR, 1.21; 95% CI, 1.08 –1.37). These results were disputed in a study by Filardo et al [15] in which long-term survival was assessed in 1,039 patients receiving either AVR or concomitant AVR and CABG at the Baylor University Medical Center in Dallas, Texas. After adjustment for risk factors for adverse outcomes after cardiac operations identified by the Society of Thoracic Surgeons using a propensity-score method, a significant association was found between POAF and late mortality (HR, 1.48; 95% CI, 1.12–1.96). Although existing data remain ambiguous, our study provides strong evidence for a nonsignificant negative effect of POAF in a primary AVR cohort. Our data indicate that although POAF was associated with late mortality on univariate analysis (p ⫽ 0.035), among propensity-matched patients, there was no significant difference in long-term survival (p ⫽ 0.63). This suggests that a poorer perioperative and intraoperative profile in patients in whom POAF developed is likely to account for
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
137
Table 2. Preoperative and Intraoperative Characteristics of Matched Patients With and Without Postoperative Atrial Fibrillation Variables Total number of patients (%) Preoperative Variables Age (Mean ⫾ SD) ⬎ 70 (%) ⱕ 70 (%) Male sex (%) Diabetes (%) Hypercholesterolemia (%) Hypertension (%) Obesity (%) Creatinine ⬎ 200 (%) Preoperative dialysis (%) Estimated glomerular filtration rate (%) Cerebrovascular disease (%) Peripheral vascular disease (%) Chronic pulmonary pulmonary disease (%) New York Heart Association class I/II (%) III/IV (%) Previous myocardial infarction (%) Active endocarditis (%) Urgency Elective (%) Urgent (%) Emergency (%) Mechanical prosthesis (%) Congestive heart failure background (%) Congestive heart failure at admission (%) NYHA III or IV (%) Cardiogenic shock (%) Previous percutaneous coronary intervention (%) Left ventricular function Normal (ejection fraction ⬎ 0.60) (%) Mild (ejection fraction ⬎ 0.45) (%) Moderate (ejection fraction 0.30–0.45) (%) Severe (ejection fraction ⬍ 0.30) (%) Aortic stenosis (%) Aortic regurgitation (%) None/mild (%) Moderate/severe (%) Mitral stenosis (%) Mitral regurgitation (%) None/mild (%) Moderate/severe (%) Intraoperative Variables Aortic cross-clamp time (%) (min) (mean ⫾ SD) ⬎ 90 (%) ⱕ 90 (%) Cardiopulmonary bypass time (%) (min) (mean ⫾ SD) ⬎ 120 (%) ⱕ 120 (%) NYHA ⫽ New York Heart Association;
All Patients
No POAF
POAF
P Value
Standardized Difference
1,184
592
592
70.2 ⫾ 10.6 486 (41) 202 (17.1) 621 (52.4) 240 (20.3) 597 (50.4) 813 (68.7) 452 (38.2) 17 (1.4) 13 (1.1) 74.9 ⫾ 32.3 109 (9.2) 52 (4.4) 156 (13.2)
70.2 ⫾ 10.6 240 (40.5) 104 (17.6) 303 (51.2) 122 (20.6) 303 (51.2) 410 (69.3) 229 (38.7) 7 (1.2) 4 (0.7) 75.4 ⫾ 32.6 54 (9.1) 27 (4.6) 76 (12.8)
70.2 ⫾ 10.7 246 (41.6) 98 (16.6) 318 (53.7) 118 (19.9) 294 (49.7) 403 (68.1) 223 (37.7) 10 (1.7) 9 (1.5) 74.5 ⫾ 32.0 55 (9.3) 25 (4.2) 80 (13.5)
0.85 ... 0.99 0.42 0.83 0.64 0.71 0.77 0.63 0.26 ... ⬎ 0.99 0.78 0.8
0.0 2.1 ⫺2.7 5.1 ⫺1.7 ⫺3.0 ⫺2.5 ⫺2.1 4.3 8.1 2.8 0.6 ⫺1.6 2.0
671 (32.5) 513 (24.8) 54 (4.6) 23 (1.9)
327 (24.4) 265 (19.8) 29 (4.9) 12 (2)
344 (47.4) 248 (34.2) 25 (4.2) 11 (1.9)
0.94 0.68 ⬎ 0.99
5.8 ⫺5.8 ⫺3.2 ⫺1.2
1,028 (86.8) 147 (12.4) 9 (0.8) 293 (24.7) 416 (35.1) 129 (10.9) 513 (43.3) 4 (0.3) 37 (3.1)
513 (86.7) 74 (12.5) 5 (0.8) 150 (25.3) 208 (35.1) 67 (11.3) 265 (44.8) 2 (0.3) 19 (3.2)
515 (87) 73 (12.3) 4 (0.7) 143 (24.2) 208 (35.1) 62 (10.5) 248 (41.9) 2 (0.3) 18 (3)
... ... 0.94 0.69 ⬎ 0.99 0.71 0.35 ⬎ 0.99 ⬎ 0.99
806 (68.1) 250 (21.1) 83 (7) 45 (3.8) 1,059 (89.4)
402 (67.9) 126 (21.3) 38 (6.4) 26 (4.4) 527 (89)
404 (68.2) 124 (20.9) 45 (7.6) 19 (3.2) 532 (89.9)
... ... ... 0.64 0.71
874 (73.8) 310 (26.2) 12 (1)
433 (73.1) 159 (26.9) 5 (0.8)
441 (74.5) 151 (25.5) 7 (1.2)
... 0.64 0.77
1,156 (97.6) 28 (2.4)
578 (97.6) 14 (2.4)
578 (97.6) 14 (2.4)
... ⬎ 0.99
2.3 ⫺2.7 3.1 0.0 0.0 0.0
74.5 ⫾ 21.6 226 (19.1) 958 (80.9) 98.5 ⫾ 32.5 202 (17.1) 982 (82.9)
74.8 ⫾ 22.3 114 (19.3) 478 (80.7) 98.5 ⫾ 35.9 106 (17.9) 486 (82.1)
74.1 ⫾ 20.9 112 (18.9) 480 (81.1) 98.4 ⫾ 28.7 96 (16.2) 496 (83.8)
0.59 ... 0.94 0.99 ... 0.49
3.2 ⫺0.8 0.6 0.3 ⫺4.0 3.0
POAF ⫽ postoperative atrial fibrillation;
SD ⫽ standard deviation.
1.0 ⫺0.5 ⫺1.9 ⫺2.7 0.0 ⫺2.7 ⫺5.8 0.0 ⫺0.9 0.0 0.6 ⫺0.7 4.2 ⫺5.6 1.6
ADULT CARDIAC
Ann Thorac Surg 2013;95:133– 40
138 ADULT CARDIAC
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
Ann Thorac Surg 2013;95:133– 40
Fig 1. Overall survival of unmatched patients with and without postoperative atrial fibrillation (POAF).
at least some of the survival differences observed. Indeed, patients in the POAF group were older (71.55 ⫾ 10.97 years versus 65.56 ⫾ 14.03 years; p ⬍ 0.001). Advanced age has been previously reported to be an independent predictor of POAF after CABG; it is also a well-known marker for poorer perioperative and longterm outcomes [2, 3, 6]. Patients in the POAF group also presented with more comorbidities—in particular, chronic obstructive pulmonary disease, hypercholesterolemia, hypertension, cerebrovascular disease, and heart failure (NYHA class III or IV). Several mechanisms have been postulated to explain the association of POAF with late mortality. Shinbane et al [16] proposed that AF causes rapid ventricular responses, leading to ventricular dilatation and reduced cardiac output. Furthermore, reduced ventricular filling and circulatory stasis in the left atrium as a result of AF may promote embolus formation and lead to an increased risk of stroke [17]. Petersen et al [18] demon-
strated that POAF may predispose to cerebrovascular complications by impairing cerebral blood flow. However, the exact mechanism by which POAF contributes to poorer long-term survival has not been determined. Although biological explanations for the potentially deleterious impact of POAF on late mortality have been proposed, several studies have demonstrated that it is possible to achieve good outcomes in patients with POAF undergoing AVR through vigilant observation. Mariscalco et al [14] hypothesized that the demonstrated nonsignificant association between POAF and late mortality in patients undergoing AVR was a result of good postoperative follow-up with regular clinic visits and echocardiographic examinations. The authors also postulated a potentially protective effect of anticoagulation therapy from embolic events in patients receiving mechanical valves [14]. Further, we hypothesize that our statistically nonsignificant results may inadvertently be due to current postoperative anticoagulation and antiplatelet regimens. Indeed, most guidelines recommend warfarin for 3 months after tissue AVR, with antiplatelet therapy considered an acceptable alternative [19]. Of particular relevance, the European Association for Cardio-Thoracic Surgery guidelines recommend anticoagulation in patients receiving tissue AVR in whom POAF develops [19]. In contrast, the guidelines state that treatment with antiplatelet therapy is sufficient in patients who underwent AVR but did not experience POAF. The therapeutic value of warfarin in reducing the incidence of postoperative stroke may have contributed to good longterm outcomes of patients in whom POAF developed after AVR. This important clinical question warrants further investigation. Although POAF was not shown to be an independent risk factor for poor long-term survival, it is associated with poorer perioperative outcomes. Propensitymatched patients with POAF were almost twice as likely to experience new renal failure and 6 times as likely to experience gastrointestinal complications. This corresponds to findings from a previous study in a CABG cohort [20]. Importantly, patients with POAF were signif-
Table 3. A Comparison of Early Outcomes of Unmatched Patients With and Without Postoperative Atrial Fibrillation Postoperative Morbidity 30-day mortality (%) Postoperative myocardial infarction (%) Stroke (%) New renal failure (%) Prolonged ventilation (⬎ 24 h) (%) Return to operating room (%) Return to operating room for bleeding (%) Gastrointestinal complications (%) Readmission within 30 d of operation (%) Any Mortality/Morbidity (%) Red blood cell transfusion (%) Length of hospital stay (d) (mean ⫾ SD) POAF ⫽ postoperative atrial fibrillation.
All Patients
No POAF
POAF
p Value
25 (1.2) 5 (0.2) 19 (0.9) 97 (4.7) 135 (6.5) 146 (7.1) 78 (3.8) 26 (1.3) 222 (10.8) 508 (24.6) 785 (38) 10 ⫾ 9
11 (0.8) 3 (0.2) 11 (0.8) 43 (3.2) 70 (5.2) 80 (6) 41 (3.1) 9 (0.7) 130 (9.7) 289 (21.6) 462 (34.5) 9 ⫾ 10
14 (1.9) 2 (0.3) 8 (1.1) 54 (7.4) 65 (9) 66 (9.1) 37 (5.1) 17 (2.3) 92 (12.7) 219 (30.2) 323 (44.6) 11 ⫾ 9
0.035 ⬎ 0.99 0.63 ⬍ 0.001 0.001 0.009 0.022 0.003 0.037 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
139
Table 4. A Comparison of Early Outcomes of Matched Patients With And Without Postoperative Atrial Fibrillation Postoperative Morbidity 30-day mortality (%) Postoperative myocardial infarction (%) Stroke (%) New renal failure (%) Prolonged ventilation (⬎ 24 h) (%) Return to operating room (%) Return to operating room for bleeding (%) Gastrointestinal complications (%) Readmission within 30 d of operation (%) Any mortality/morbidity (%) Red blood cell transfusion (%) Length of hospital stay (d) (mean ⫾ SD) POAF ⫽ postoperative atrial fibrillation;
All Patients
No POAF
POAF
p Value
18 (1.3) 3 (0.2) 11 (0.8) 63 (4.4) 80 (5.6) 85 (4.1) 47 (2.3) 20 (1) 125 (8.7) 296 (20.7) 470 (32.9) 9.7 ⫾ 7.5
7 (1) 1 (0.1) 7 (1) 22 (3.1) 34 (4.8) 36 (2.7) 19 (1.4) 4 (0.3) 49 (6.9) 126 (17.6) 218 (30.5) 8.9 ⫾ 6.7
11 (1.5) 2 (0.3) 4 (0.6) 41 (5.7) 46 (6.4) 49 (6.8) 28 (3.9) 16 (2.2) 76 (10.6) 170 (23.8) 252 (35.2) 10.6 ⫾ 8.1
0.48 ⬎ 0.99 0.55 0.023 0.22 0.18 0.23 0.012 0.014 0.004 0.064 ⬍ 0.001
SD ⫽ standard deviation.
icantly more likely to have overall morbidity/mortality (23.8% versus 17.6%; p ⫽ 0.004). There was a nonsignificant trend toward an increased need for RBC transfusion (35.2% versus 30.5%; p ⫽ 0.064). Whether POAF is cause or consequence to such outcomes can only be speculated upon, given the dynamic complexity of the postoperative setting. However, it is likely that POAF is intricately involved in the development of such adverse outcomes. As such, our data support the need to explore prophylactic strategies aimed at the identification, management, and prevention of POAF. This may then translate into improved perioperative outcomes for patients. The current study has several limitations. First, our study did not include postoperative patient data such as anticoagulation treatments or follow-up data on rates of recurrent AF and cancer. We could not account for prophylactic antiarrhythmic strategies such as the use of AADs. Furthermore, by virtue of the multicenter approach of our study, the effect of surgeon and cardiologist variability, if any, on the intraoperative and postopera-
tive course of patients was not predictable. In addition, it must be emphasized that this observational study establishes correlations only, and as such any direct mechanistic value of our findings should be interpreted with care. The present study shows that POAF was not associated with early or late mortality in a primary AVR cohort. We have also shown that POAF is more common in patients with a higher risk profile and is independently associated with adverse outcomes postoperatively. In conclusion, our study identifies POAF as a key player in the immediate postoperative setting and suggests that a review of current prophylactic and management strategies is necessary with regard to POAF in patients undergoing AVR. The following investigators, data managers, and institutions participated in the ASCTS database: Alfred Hospital: A. Pick, J. Duncan; Austin Hospital: S. Seevanayagam, M. Shaw; Cabrini Health: G. Shardey; Geelong Hospital: M. Morteza, C. Bright; Flinders Medical Centre: J. Knight, R. Baker, J. Helm; Jessie McPherson Private Hospital: J. Smith, H. Baxter; John Hunter Hospital: A. James, S. Scaybrook; Lake Macquarie Hospital: B. Dennett, M. Jacobi; Liverpool Hospital: B. French, N. Hewitt; Mater Health Service Hospital: A.M. Diqer, J. Archer; Monash Medical Centre: J. Smith , H. Baxter; Prince of Wales Hospital: H. Wolfenden, D. Weerasinge; Royal Melbourne Hospital: P. Skillington, S. Law; Royal Prince Alfred Hospital: M. Wilson, L. Turner; St. George Hospital: G. Fermanis, C. Redmond; St. Vincent’s Hospital, VIC: M. Yii, A. Newcomb, J. Mack, K. Duve; St Vincent’s Hospital, NSW: P. Spratt, T, Hunter; The Canberra Hospital: P. Bissaker, K. Butler; Townsville Hospital: R. Tam, A. Farley; Westmead Hospital: R. Costa, M. Halaka. The Australasian Society of Cardiac and Thoracic Surgeons (ASCTS) National Cardiac Surgery Database Program is funded by the Department of Human Services, Victoria, and the Health Administration Corporation (GMCT) and the Clinical Excellence Commission (CEC), NSW.
References Fig 2. Overall survival of matched patients with and without postoperative atrial fibrillation (POAF).
1. Maisel WH, Rawn JD, Stevenson WG. Atrial fibrillation after cardiac surgery. Ann Intern Med 2001;135:1061–73. 2. Almassi GH, Schowalter T, Nicolosi AC, Aggarwal A, Moritz TE, Henderson WG, et al. Atrial fibrillation after cardiac
ADULT CARDIAC
Ann Thorac Surg 2013;95:133– 40
140 ADULT CARDIAC
3. 4. 5. 6.
7. 8. 9.
10. 11.
SAXENA ET AL ATRIAL FIBRILLATION AFTER AVR
surgery: a major morbid event? Ann Surg 1997;226:501–11; discussion 11–3. Hogue CW, Jr, Hyder ML. Atrial fibrillation after cardiac operation: risks, mechanisms, and treatment. Ann Thorac Surg 2000;69:300 – 6. Aranki SF, Shaw DP, Adams DH, et al. Predictors of atrial fibrillation after coronary artery surgery. Current trends and impact on hospital resources. Circulation 1996;94:390 –7. Creswell LL, Schuessler RB, Rosenbloom M, Cox JL. Hazards of postoperative atrial arrhythmias. Ann Thorac Surg 1993; 56:539 – 49. Mathew JP, Parks R, Savino JS, et al. Atrial fibrillation following coronary artery bypass graft surgery: predictors, outcomes, and resource utilization. MultiCenter Study of Perioperative Ischemia Research Group. JAMA 1996;276: 300 – 6. Hravnak M, Hoffman LA, Saul MI, Zullo TG, Whitman GR. Resource utilization related to atrial fibrillation after coronary artery bypass grafting. Am J Crit Care 2002;11:228 –38. Ahlsson A, Bodin L, Fengsrud E, Englund A. Patients with postoperative atrial fibrillation have a doubled cardiovascular mortality. Scand Cardiovasc J 2009;43:330 – 6. Mahoney EM, Thompson TD, Veledar E, Williams J, Weintraub WS. Cost-effectiveness of targeting patients undergoing cardiac surgery for therapy with intravenous amiodarone to prevent atrial fibrillation. J Am Coll Cardiol 2002;40: 737– 45. Mariscalco G, Klersy C, Zanobini M, et al. Atrial fibrillation after isolated coronary surgery affects late survival. Circulation 2008;118:1612– 8. Dinh DT, Lee GA, Billah B, Smith JA, Shardey GC, Reid CM. Trends in coronary artery bypass graft surgery in Victoria, 2001-2006: findings from the Australasian Society of Cardiac
Ann Thorac Surg 2013;95:133– 40
12.
13.
14. 15.
16.
17. 18. 19. 20.
and Thoracic Surgeons database project. Med J Aust 2008;188:214 –7. Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009;28:3083– 107. Austin PC. Propensity-score matching in the cardiovascular surgery literature from 2004 to 2006: a systematic review and suggestions for improvement. J Thorac Cardiovasc Surg 2007;134:1128 –35. Mariscalco G, Engstrom KG. Postoperative atrial fibrillation is associated with late mortality after coronary surgery, but not after valvular surgery. Ann Thorac Surg 2009;88:1871– 6. Filardo G, Hamilton C, Hamman B, Hebeler RF, Jr, Adams J, Grayburn P. New-onset postoperative atrial fibrillation and long-term survival after aortic valve replacement surgery. Ann Thorac Surg 2010;90:474 –9. Shinbane JS, Wood MA, Jensen DN, Ellenbogen KA, Fitzpatrick AP, Scheinman MM. Tachycardia-induced cardiomyopathy: a review of animal models and clinical studies. J Am Coll Cardiol 1997;29:709 –15. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke 1991;22:983– 8. Petersen P, Kastrup J, Videbaek R, Boysen G. Cerebral blood flow before and after cardioversion of atrial fibrillation. J Cereb Blood Flow Metab 1989;9:422–5. Dunning J, Versteegh M, Fabbri A, et al. Guideline on antiplatelet and anticoagulation management in cardiac surgery. Eur J Cardiothorac Surg 2008;34:73–92. Kalavrouziotis D, Buth KJ, Ali IS. The impact of new-onset atrial fibrillation on in-hospital mortality following cardiac surgery. Chest 2007;131:833–9.
Notice From the American Board of Thoracic Surgery The 2012 Part I (written) examination was held on Monday, November 19, 2012. To be admissible to the Part II (oral) examination, a candidate must have successfully completed the Part I (written) examination. A candidate applying for admission to the certifying examination must fulfill all the requirements of the Board in force at
© 2013 by The Society of Thoracic Surgeons Published by Elsevier Inc
the time the application is received. Please address all communications to the American Board of Thoracic Surgery, 633 N St. Clair St, Suite 2320, Chicago, IL 60611; telephone: (312) 202-5900; fax: (312) 202-5960; email: [email protected].
Ann Thorac Surg 2013;95:140
•
0003-4975/$36.00