This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: 112/21/3247    most recent CIRCULATIONAHA.105.553743v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zacharias, A. Articles by Habib, R. H. Search for Related Content PubMed PubMed Citation Articles by Zacharias, A. Articles by Habib, R. H. Pubmed/NCBI databases Medline Plus Health Information Heart Surgery Nutrition Obesity Related Collections CV surgery: coronary artery disease CV surgery: valvular disease Arrhythmias, clinical electrophysiology, drugs

(Circulation. 2005;112:3247-3255.)
© 2005 American Heart Association, Inc.

Cardiovascular Surgery

Obesity and Risk of New-Onset Atrial Fibrillation After Cardiac Surgery

Anoar Zacharias, MD; Thomas A. Schwann, MD; Christopher J. Riordan, MD; Samuel J. Durham, MD; Aamir S. Shah, MD; Robert H. Habib, PhD

From the Division of Cardiovascular Surgery, St. Vincent Mercy Medical Center (A.Z., T.A.S., C.J.R., S.J.D., A.S.S., R.H.H.), Toledo, Ohio; Division of Cardiovascular Surgery, St. Luke’s Hospital (A.Z., T.A.S., C.J.R., S.J.D., A.S.S., R.H.H.), Maumee, Ohio; and Departments of Surgery (A.Z., T.A.S., C.J.R., S.J.D., A.S.S.) and Medicine (R.H.H.), Medical University of Ohio, Toledo, Ohio.

Correspondence to Robert H. Habib, PhD, Director, Cardiopulmonary Research, St. Vincent Mercy Medical Center, 2213 Cherry St, ACC Bldg, Suite 309, Toledo, OH 43608. E-mail Robert_Habib{at}mhsnr.org

Received April 4, 2005; revision received August 30, 2005; accepted September 1, 2005.

	Abstract

Top Abstract Introduction Methods Results Discussion Conclusions References

 
Background— New-onset postoperative atrial fibrillation (AF) is a common complication of cardiac surgery that has substantial effects on outcomes. In the general (nonsurgical) adult population, AF has been linked to increasing obesity, which correlates with left atrial enlargement. It is not known whether postoperative AF is similarly linked to obesity.

Methods and Results— This was a retrospective analysis of the incidence of AF in terms of body mass index (BMI). A total of 8051 consecutive cardiac surgery patients (1994 to 2004; mean age 64 [SD 11] years; 5372 men [67%]) who were free of any history of preoperative AF or flutter were included in the analysis. This series included 3164 obese patients (39%; median age 62 years) and 4887 nonobese patients (61%; median age 66 years), who were further divided on the basis of BMI (kg/m²) into 6 groups: BMI <22 kg/m², 22BMI25 kg/m² (normal), 25<BMI30 kg/m² (overweight), 30<BMI 35 kg/m² (obese I), 35<BMI40 kg/m² (obese II), and BMI >40 kg/m² (obese III). Unadjusted AF incidence was similar in obese and nonobese patients (n=742 [23.5%] versus n=1068 [21.9%], respectively; P=0.099). Covariate-adjusted ORs for AF were systematically greater for larger patients than for patients in the normal group (adjusted OR [95% CI]=1.18 [1.00 to 1.40], 1.36 [1.14 to 1.63], 1.69 [1.35 to 2.11], and 2.39 [1.81 to 3.17] for overweight, obese I, obese II, and obese III, respectively). Other AF predictors included age (adjusted OR=1.52 [95% CI 1.46 to 1.58] per 10 years), mitral valve surgery (adjusted OR=2.42 [95% CI 1.92 to 3.06]), aortic valve surgery (adjusted OR=1.79 [95% CI 1.45 to 2.22]), chronic obstructive pulmonary disease (adjusted OR=1.28 [95% CI 1.12 to 1.46]), male gender (adjusted OR=1.24 [95% CI 1.10 to 1.40]), preoperative ß-blocker use (adjusted OR=1.17 [95% CI 1.05 to 1.32]), vascular disease (adjusted OR=1.18 [95% CI 1.05 to 1.32]), white race (adjusted OR=1.33 [95% CI 1.07 to 1.66]), history of arrhythmia other than AF/flutter (adjusted OR=0.80 [95% CI 0.68 to 0.96]), ejection fraction <40% (adjusted OR=1.16 [95% CI 1.03 to 1.31]), left main disease (adjusted OR=1.15 [95% CI 1.00 to 1.32]), and off-pump surgery (adjusted OR=0.61 [95% CI 0.44 to 0.83]). The obesity-AF association was confirmed in 4 1-to-1 propensity-matched obese versus nonobese comparisons and in 2 separate derivation/validation subcohort analyses.

Conclusions— Obesity is an important determinant of new-onset AF after cardiac surgery. Future postoperative AF risk models should incorporate BMI or obesity levels. Studies examining the efficacy of AF-minimizing prophylactic interventions in high-BMI patients, particularly in the elderly, may be warranted.

Key Words: arrhythmia • cardiopulmonary bypass • complications • multivariate analysis • propensity matching

	Introduction

Top Abstract Introduction Methods Results Discussion Conclusions References

 
Atrial fibrillation (AF) is the most common form of cardiac arrhythmia¹ and is associated with significant morbidity and mortality.^2,3 From the Framingham Heart Study, Wang and colleagues⁴ recently provided the strongest evidence to date showing that obesity is an important predictor of development of AF in adults. Then, via echocardiographic data, they demonstrated that this body size and AF association is explained by left atrial enlargement in larger patients. The implications of these findings are amplified by the fact that obesity has reached epidemic proportions both in the United States and worldwide, which in turn raises the potential of it contributing to an AF epidemic.⁵

New-onset AF is the most common complication after cardiac surgery (20% to 40%) and is linked to other serious perioperative complications (eg, stroke), prolonged hospital stays, increased frequency of readmissions, and greater operative mortality.^6–13 Moreover, 1 group has demonstrated a significantly worse 0- to 4-year survival in CABG patients who develop postoperative AF.¹⁴ Obesity is increasingly more prevalent among cardiac surgery patients, with 40% of CABG patients having a body mass index (BMI) >30 kg/m².¹⁵ Whether and to what extent this prevalence of obesity explains new-onset AF in this surgical population is not fully known. To date, no studies exploring predictors of AF after cardiac surgery have linked increased body size or obesity to this complication.^6–13

We reasoned that the association of left atrial enlargement with increasing adiposity in adult cardiac surgery patients is most likely similar to that in the general adult population.^4,16 In fact, the increased potential of other cardiovascular risk factors (eg, hypertension and left ventricular dysfunction) in this surgical cohort are likely to exacerbate this effect. The primary objective of the present study was to elucidate the role of body size (with primary emphasis on BMI) in the development of postoperative AF after cardiac surgery.

	Methods

Top Abstract Introduction Methods Results Discussion Conclusions References

 
Study Sample, Clinical Data, and End Point
We examined a retrospective series of 8727 consecutive cardiac surgery patients (mean age 64±11 years; 5372 men [67%]) at 2 community hospitals (Saint Vincent Mercy Medical Center, Toledo, Ohio, 1994 to 2004, and Saint Luke’s Hospital, Maumee, Ohio, 2001 to 2004). Given their high propensity for postoperative AF, patients with a history of preoperative AF or flutter (676 [7.7%] of 8727) were excluded from analysis. The remaining 8051 patients constituted the overall study population, which included 6749 patients (84%) who underwent isolated CABG only. A total of 416 patients underwent surgery without (off-pump: 5.2%) cardiopulmonary bypass (CPB). Normothermic CPB was used in 98% of on-pump patients. This investigation was approved by the institutional review boards at both hospitals. Because of the retrospective nature of the study, requirement for informed consent was waived.

Clinical data for all patients were collected prospectively and entered into identical cardiac surgery databases at both institutions as described previously.¹⁵ Collected variables included demographics, risk factors and comorbidities, preoperative medications, operative data, and postoperative complications and outcomes. All entries were based on definitions of the Society of Thoracic Surgeons and were regularly reported to the national cardiac surgery database. The single end point in this investigation was new-onset AF (occurring anytime perioperatively through hospital discharge), which was collected as a postoperative complication. AF was diagnosed on the basis of an ECG, telemetry (100% of patients), or physician findings and was confirmed from hospital or physician chart notes and discharge summaries. Database quality checks for missing values and incorrect entries were performed routinely. Additionally, complication rates and resource utilization (transfusion, cardiovascular intensive care unit [CVICU]/hospital stays) were always cross-checked for the separate CVICU and cardiac surgery databases, as well as against patients’ discharge summaries and hospital medical chart attestation (or coding summary).

Postoperative Medications
Aspirin and ß-blockers use was consistent throughout this 1994 to 2004 series. Aspirin (325 mg via nasogastric tube) was always given to patients on arrival at the CVICU and was decreased to 162 mg PO daily on postoperative day 1. In all patients, ß-blockers were routinely given by mouth until the morning of surgery and reinstituted soon after surgery, being given intravenously in the CVICU if the patient was unable to swallow pills. Accordingly, time of ß-blocker withdrawal (if any) was minimized in the present series of patients. Amiodarone (intravenous or by mouth) was always used after onset of AF, including intraoperatively. Since July 1999, postoperative prophylactic amiodarone (daily 200 mg PO started on postoperative day 1) has been used routinely in all patients. An ACE inhibitor was used routinely in patients, including those who were taking one preoperatively, who were in congestive heart failure or who had low ejection fraction (<35%), if hemodynamically tolerated (systolic blood pressure >100 mm Hg).

Data Analysis and Statistical Methods
Patients were divided into those who developed the complication (AF) versus those who did not (No-AF). BMI (in kg/m²) and body surface area (BSA; in m²) were used to characterize the role of patient size and adiposity and were considered both as continuous and categorical variables (6 categories each). BMI was derived as the weight (in kilograms) divided by the square of height (in meters). BMI categories were largely based on the World Health Organization/National Institutes of Health classification scheme¹⁷ as follows: BMI <22 kg/m² (underweight), 22BMI25 kg/m² (normal), 25<BMI30 kg/m² (overweight), 30<BMI35 kg/m² (obese I), 35<BMI40 kg/m² (obese II), and BMI >40 kg/m² (obese III). BSA was calculated with the Mosteller formula as BSA (m²)=[(height (cm)xweight (kg)]/3600)1/2 and categorized as follows: BSA <1.5 m² (very small), 1.5BSA1.75 m² (small), 1.75<BSA2.0 m² (intermediate), 2.0<BSA2.25 m² (large I), 2.25<BSA2.5 m² (large II), and BSA >2.5 m² (large III).

Continuous data were expressed as mean (SD). Baseline variables for AF versus No-AF groups were compared by use of the Wilcoxon rank sum test, t test, or the ² test as appropriate. A 2-sided P<0.05 was considered statistically significant. Body size–AF relations were derived with and without adjustment for clinical and operative confounders via multivariable logistic regression techniques for (1) all cardiac surgery and (2) isolated CABG patients. Separate multivariable model analyses were performed with either BMI and BSA to define body size. Here, "normal" BMI and "intermediate" BSA were used as reference categories, respectively, when size was described as a categorical variable. For CPB, short-duration CPB (60 minutes) was used as the reference category, with "off-pump" included as a subgroup. Covariates selected for adjustment were based on prior reports^6–13 and included all the variables listed in Table 1 in addition to a month-of-surgery variable, which varied between 1 (January 1994) and 132 (December 2004). Regression model selection was done with backward elimination (Wald statistic, confirmed by forward and stepwise selection). A P<0.05 significance level was used for model inclusion and P>0.1 for exclusion (SPSS version 10.0, SPSS Inc).

View this table: [in this window] [in a new window]   TABLE 1. Comparison of Selected Patient Characteristics for AF vs No-AF Cohorts

The obesity-AF association derived in the primary multivariable analysis was further confirmed via additional complementary analyses. First, we compared AF incidence in 1-to-1 or greedy (ie, no shared or repeated matches) propensity-matched obese (BMI >30 kg/m²) versus nonobese (BMI 30 kg/m²) isolated CABG patient groups. Here, a total of 4 matched patient group pairs were obtained via a nonparsimonious obese model^18,19 that incorporated all variables in Table 1 except for body size. These were as follows: obese (all) versus nonobese, obese I versus nonobese, obese II versus nonobese, and obese III versus nonobese. Propensity score matching was strict, with a maximum of ±1% score difference to ensure that all covariates (demographics, risk factors, and operative data) for the obese versus nonobese comparisons groups, other than size, were statistically similar. Here, a significantly greater difference in AF incidence after matching would indicate a link between obesity (or size) and AF while controlling for confounders. Second, we used 2 additional analyses in which the 8051 patients were divided into derivation (n=4026) and validation (n=4025) subcohorts of equal size (1) via random selection without replacement of the derivation group and (2) based on time of surgery, with recent patients (September 1999 to December 2004) used as the derivation group and remote patients used as the validation cohort (January 1994 to September 1999).

	Results

Top Abstract Introduction Methods Results Discussion Conclusions References

 
Incidence of New-Onset Postoperative AF
The 8051 cardiac surgery study population included 1810 patients with (22.5%) and 6241 patients without (87.5%) documented postoperative new-onset AF. This incidence was similar after isolated CABG (AF 1496 of 6749; 22.2%). Baseline characteristics of the AF and No-AF cohorts are compared in Table 1. Gender was not a predictor of AF before risk adjustment but became significant for men when adjusted for age (P=0.041; Figure 1, top). AF patients were older and included more whites. AF was systematically and substantially more frequent with increasing age in both men and women (Figure 1) increasing from 6.2% in patients <40 years old to 52% for patients >85 years old (P<0.001).

View larger version (27K): [in this window] [in a new window]   Figure 1. Top, Incidence of AF (A-Fib) after cardiac surgery increased with increasing patient age in male and female patients; youngest and oldest groups included all patients aged 40 years and >85 years, respectively. Intermediate groups were derived for 5-year age increments. Bottom, Incidence of AF in age groups subdivided into short (CPB 100 minutes), long (CPB >100 minutes), and no CPB (off-pump), indicating increased AF with time on CPB. Inset, Unadjusted incidence of AF in unadjusted CPB duration categories (0=off-pump).

AF patients were characterized by a higher incidence of hypertension, chronic obstructive pulmonary disease, peripheral and/or cerebrovascular disease (vascular disease), congestive heart failure, triple-vessel disease, and left main disease; lower ejection fractions; less preoperative use of digitalis, diuretics, or steroids; and low cardiac output (use of intra-aortic balloon pump). Current smokers, patients with angina (any kind), and preoperative ß-blocker use were less frequent among AF patients. Operatively, valve surgery and increasing time on CPB (Figure 1, bottom; P<0.001) were associated with increased frequency of AF, whereas off-pump (CPB=0) surgery was associated with less AF both before and after age adjustment.

Role of Body Size in New-Onset AF
Of 8051 overall study patients, 3164 were obese (39.3%; median age 62 years; BMI >30 kg/m²) and 4887 were nonobese (61%; median age 66 years). The overall unadjusted AF rates in obese (742 of 3164; 23.5%) versus nonobese (1068 of 4887; 21.9%) patients were similar (P=0.099); however, the mean age of cardiac surgery patients decreased systematically for increasing BMI categories (both P<0.001; Figure 2). When BMI was considered as a continuous variable (Table 2), unadjusted BMI was significantly associated with AF (OR 1.01, 95% CI 1.00 to 1.02; P=0.046), equivalent to a 1% increase in AF incidence per 1-kg/m² increase in BMI. This effect was increased after adjustment for age (OR=1.03; P<0.001) and other covariates (OR=1.04; P<0.001). These findings were largely mirrored when BMI was included as a categorical variable (Table 2; Figure 2). Age-adjusted AF rates were increased from lowest to highest BMI groups (P<0.001): 20.0%, 19.8%, 21.8%, 23.3%, 26.59%, and 33.2%, respectively. The corresponding analyses when BSA was used for body size instead of BMI are provided in the online-only Data Supplement (Table I, Figure I).

View larger version (37K): [in this window] [in a new window]   Figure 2. Left, Mean age at time of surgery in BMI-based patient size subgroups was systematically lower for larger patients. Right, Unadjusted and age-adjusted incidence of new-onset AF for size cohorts. Pts indicates patients; Prob., probability; UW, underweight; N, normal weight; and OW, overweight.

View this table: [in this window] [in a new window]   TABLE 2. Continuous and Categorical Body Size Indices (BMI and BSA) as Predictors of New-Onset AF With (Multivariable Models) and Without (Univariate Model) Adjustment

Multivariable Analyses
With adjustment for covariates, underweight (OR 0.97, 95% CI 0.73 to 1.27) and normal (OR 1.0 [reference]) BMI groups were equally likely to develop AF. The adjusted ORs for developing AF were systematically greater for the larger patients (1.18 [95% CI 1.00 to 1.40], 1.36 [95% CI 1.14 to 1.63], 1.69 [95% CI 1.35 to 2.11], and 2.39 [95% CI 1.81 to 3.17] for overweight, obese I, obese II, and obese III, respectively). In case of BSA (Table I in the Data Supplement), the risk-adjusted OR for AF was increased systematically from smallest (OR [very small]=0.67 [95% CI 0.43 to 1.06]) to largest (OR [large III]=2.60 [95% CI 1.87 to 3.63]).

Results of the multivariable logistic regression models for AF, with patient size represented by 6 BMI categories, for all cardiac surgery and isolated CABG are shown in Table 3. The model discriminated between AF and No-AF fairly well (area under the receiver operator characteristic curve 0.68 and 0.67 [isolated CABG]), and the model calibration was good (Hosmer-Lemeshow [goodness-of-fit test]: P=0.36 and 0.58 [isolated CABG]). Other than BMI category, predictors of AF included age (adjusted OR 1.52 [95% CI 1.46 to 1.58] per 10 years), mitral valve surgery (adjusted OR 2.42 [95% CI 1.92 to 3.06]) and aortic valve surgery (adjusted OR 1.79 [95% CI 1.45 to 2.22]), chronic obstructive pulmonary disease (adjusted OR 1.28 [95% CI 1.12 to 1.46]), male gender (adjusted OR 1.24 [95% CI 1.10 to 1.40]), preoperative ß-blockers (adjusted OR 1.17 [95% CI 1.05 to 1.32]), vascular disease (adjusted OR 1.18 [95% CI 1.05 to 1.32]), white race (adjusted OR 1.33 [95% CI 1.07 to 1.66]), history of arrhythmia other than AF/flutter (adjusted OR 0.80 [95% CI 0.68 to 0.96]), ejection fraction <40% (adjusted OR 1.16 [95% CI 1.03 to 1.31]), left main disease (adjusted OR 1.15 [95% CI 1.00 to 1.32]), and off-pump surgery (adjusted OR 0.61 [95% CI 0.44 to 0.83]). Obviously, for isolated CABG patients only, the model did not include mitral and aortic valve surgery as predictor variables. Also, left main disease, triple-vessel disease, and vascular disease were no longer significant predictors. Preoperative congestive heart failure was a predictor of AF in the CABG-only subpopulation.

View this table: [in this window] [in a new window]   TABLE 3. Multivariable Predictors of New-Onset AF After Any Cardiac Surgery or Isolated CABG, With BMI (6 Categories) Used as the Only Patient Size Parameter

Validation of Findings
Propensity-Matched Comparisons
Of the 6749 isolated CABG patients, 2712 (40.2%) were obese (BMI >30 kg/m²). The derived propensity score model for obese versus nonobese patients was of good calibration (Hosmer-Lemeshow P=0.59) and good discrimination (area under the receiver operator characteristic curve 0.74). Obese versus nonobese propensity scores were substantially different (0.457±0.149 versus 0.358±0.134; P<0.001) and increased systematically and significantly with increasing BMI category (online Data Supplement; Figure II, top; P<0.001, all pairwise and overall [trend]). Greedy 1-to-1 matchings were obtained for 2360 (87%) of 2712 when all obese patients were matched as 1 group. Nearly all patients were matched when obese subgroups were matched separately: 1730 obese III (99.5%), 659 obese II (100%), and 314 obese I (100%). In all cases, with the exception of size variables, all demographics, risk factors, preoperative medications, and operative variables (grafting and CPB) were similar for the matched comparison groups (see Table 1 for list). Overall, the incidence of new-onset AF was significantly greater in obese than in nonobese patients (P<0.001), mostly owing to the increased rate of AF in moderately and severely obese patients (obese II and III; Data Supplement Figure II, bottom).

Derivation Versus Validation Subcohort Analyses
Figure 3 shows the derived adjusted OR for developing new-onset AF for each of the BMI categories (normal=reference category) compared with those derived in the overall population using independent derivation and validation subpopulations of equal size. The derived BMI-AF associations were similar for the 2 subcohorts and were within the confidence bounds derived from the overall cardiac surgery series. This finding was true when the patient subgroups were based on random selection (Figure 3, top) or based on date of surgery (Figure 3, bottom). Importantly, in the latter case, the derivation (recent) group included patients for whom postoperative amiodarone prophylaxis began on postoperative day 1, whereas this was not true for the overwhelming majority (3861 [96%] of 4025) of the validation (remote) cohort.

View larger version (24K): [in this window] [in a new window]   Figure 3. Comparison of adjusted ORs for AF (A-Fib) for the different BMI categories (normal=reference category) in independent derivation and validation subcohorts of equal size. Top, Patients in the derivation cohort (n=4026) were randomly selected, with the remaining patients used as the validation cohort (n=4025). Bottom, Patients in the derivation cohort (n=4026) were recent patients (September 1999 to December 2004), whereas earlier or remote patients were used as the validation cohort (n=4025; January 1994 to September 1999). Pts indicates patients; UW, underweight; N, normal weight; and OW, overweight.

	Discussion

Top Abstract Introduction Methods Results Discussion Conclusions References

 
New-onset AF remains the most frequent postoperative complication of cardiac surgery,⁶ and contributes substantially to increased morbidity, resource utilization, and even early and possibly late deaths.^7–12 For these reasons, there is considerable interest in developing risk models for new-onset AF to potentially guide and identify patients at high risk for prophylactic intervention, and multiple such models have been published to date.^12,13

Prior studies have demonstrated that advanced age, history of AF, male gender, digoxin use, hypertension, longer CPB cross-clamp time, chronic obstructive pulmonary disease, valve surgery, postoperative withdrawal of ß-blockers, and postoperative low cardiac output increase the risk of developing new-onset AF after cardiac surgery.^6–13 Many of these factors were confirmed in our present analysis of a large cardiac surgery population (Table 3). Others have also investigated the possible role of ECG characteristics, and they have linked prolonged P-wave duration (indicative of intra-atrial conduction delay) and increased PR interval to postoperative AF.¹³ No previous studies have previously associated obesity or large body habitus with increased AF after cardiac surgery.

The primary finding in the present study is that we demonstrated, for the first time, the existence of a strong independent association between increasing body size (especially obesity) and new-onset postoperative AF in cardiac surgery patients. These results were confirmed via multiple complementary analyses, which included (1) comparisons of AF rates in obese versus nonobese groups (and subgroups) that were closely matched for all other covariates and AF predictors via propensity modeling techniques and (2) 2 comparisons of covariate-adjusted AF-BMI category relations (Figure 3) in equally sized independent derivation/validation cohorts based on random selection and chronological order (time of surgery).

This multivariable modeling also showed, that after risk adjustment, off-pump CABG surgery was associated with a significantly lower incidence of new-onset postoperative AF (Table 3). The following CPB-related mechanisms may have contributed to these findings: (1) Myocardial ischemia during CPB is global compared with localized or regional ischemia during off-pump surgery; (2) myocardial injury (elevated creatine kinase-MB) is more common in CPB versus off-pump patients; and (3) increased inflammatory injury to all organs with CPB is mediated by complement activation, cytokine release, endothelial activation, and neutrophil induction and adhesion. The latter can lead to abnormal gas exchange and electrolyte imbalance, both of which can contribute to atrial irritability.

Potential Mechanisms of the AF-Obesity Association
Left atrial enlargement, a recognized precursor of AF, is strongly correlated with increasing BMI or adiposity.^16,20,21 Other factors characterizing obesity are ventricular remodeling,²² elevated plasma volume,²³ autonomic tone,²⁴ ventricular diastolic dysfunction,²⁵ and enhanced neurohormonal activation.²⁶ These too can contribute to left atrial dilatation and its electrical dysfunction. Also, increased oxidative stress²⁷ and/or lipoapoptosis,²⁸ which are seen with increasing adiposity, may lead to myocardial structural changes (including atrial changes) such that the likelihood of AF is increased.

Our finding of an association between increasing body size and new-onset AF after cardiac surgery is in concert with the recent study by Wang and colleagues⁴ in the general adult population (Framingham Heart Study). There, they described an association between greater incidence of newly diagnosed AF over time with increasing body size. Importantly, they showed via echocardiography data that left atrial dilatation is a critical intermediate phenotype and a physiologically plausible and mechanistic explanation for the obesity-AF association. This is further supported by another population-based study by Pritchett et al,¹⁶ who found a similar association between large atrial size and greater AF/flutter incidence. There, they found a 13.6% AF/flutter rate in the upper quartile of left atrial size (quartile IV) compared with lower rates in quartiles I, II, and III (1.1%, 1.7%, and 2.3%, respectively). We do not routinely collect data on left atrial size in cardiac surgery patients and are unable to provide the corresponding data in this retrospective surgical population. However, we suggest that the distribution of left atrial size as a function of BMI is likely similar to that of the general adult population.^4,16

Diabetes mellitus has been associated with increased AF in the general adult population, most notably by Kannel and colleagues²⁹ from the Framingham Heart Study. Incidence of diabetes is also systematically greater with increasing obesity.³⁰ Yet, the present data, similar to what has been reported in several recent large analyses,^7–13 did not indicate the presence of an independent association between new-onset postoperative AF and diabetes. Although a diabetes effect in this form of AF may be implicit in the AF-obesity association, this possibility is downplayed by the fact that as in most of the other cardiac surgery series,^7–13 the incidence of diabetes in the AF versus No-AF cohorts was essentially identical (Table 1). This may indicate a distinctly different role for diabetes in the development of new-onset postoperative AF versus AF in the general adult population. Elucidation of this role is outside the scope of our present report.

Study Limitations
This study is characterized by several important strengths that support the primary finding of increasing new-onset AF with increasing patient adiposity after cardiac surgery. These included the large study population (n=8051), the number of AF cases analyzed (n=1810), the multiple converging complementary statistical analyses, and finally, supporting data from the nonsurgical general population by other authors.^4,16 Yet, several study limitations are worth noting. Most obvious is the aforementioned lack of direct left atrial data to support the left atrial dilatation mechanistic explanation that we propose. Although likely true, a BMI–left atrial size relation in cardiac surgery patients specifically is more relevant. Second, we cannot exclude the possibility that episodes of AF were missed in some patients because of their minimal and or short-lived nature and that these patients were grouped as No-AF. The prevalence of AF we report in the present series of patients lies within the range of the national data reported by the Society of Thoracic Surgeons. However, it appears unlikely that such misclassifications would occur disproportionately among the different size groups, and therefore, they should have minimal effects on the results reported here. In addition, such random misclassification might have led to a conservative bias. Third, the present study is a retrospective, nonrandomized analysis from 1 surgical team serving 2 hospitals; however, all postoperative complications were prospectively and rigorously tracked and reported to the Society of Thoracic Surgeons national cardiac surgery database. Large observational studies have been shown to closely predict the results of corresponding randomized trials.^31,32 To minimize the potential inaccuracies of nonrandomized retrospective data, we implemented several analytical methods, including propensity modeling and matching,^18,19 in addition to 2 separate derivation and validation cohort analysis comparisons. We contend that the fact that all the methods we used converged to produce the same result significantly advances our primary findings.

	Conclusions

Top Abstract Introduction Methods Results Discussion Conclusions References

 
Obesity has become increasingly prevalent in the general adult population,³⁰ in particular among cardiac surgery patients, with nearly twice the incidence of the general adult population.¹⁵ The above-presented data indicate that increasing levels of obesity represent systematically greater risk for the development of new-onset AF after cardiac surgery, the most common form of cardiac dysrhythmia, responsible for substantial utilization of resources and increased morbidity and mortality. Accordingly, we propose that future risk-modeling efforts for this form of AF should incorporate (or at least examine) body size variables such as BMI as potential determinants. Although we did not report left atrial size data in the present series of patients, we suggest that left atrial dilatation is an important intermediate phenotype of new-onset postoperative AF, as has been reported for the general adult population.⁴ Future prospective studies are needed to confirm this contention.

	Acknowledgments

 
This study was supported by institutional (Saint Vincent Mercy Medical Center, Toledo, Ohio) and departmental (Cardiothoracic Surgery) funds.

	Footnotes

 
The online-only Data Supplement can be found at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.105.553743/DC1.

	References

Top Abstract Introduction Methods Results Discussion Conclusions References

 

1. Go AS, Hylek EM, Phillips KA, Chang Y, Henault LE, Selby JV, Singer DE. Prevalence of diagnosed atrial fibrillation in adults: national implications for rhythm management and stroke prevention: the Anticoagulation and Risk Factors in Atrial Fibrillation (ATRIA) Study. JAMA. 2001; 285: 2370–2375.[Abstract/Free Full Text]
   
2. Benjamin EJ, Wolf PA, D’Agostino RB, Silbershatz H, Kannel WB, Levy D. Impact of atrial fibrillation on the risk of death: the Framingham Heart Study. Circulation. 1998; 98: 946–952.[Abstract/Free Full Text]
   
3. Wyse DG, Waldo AL, DiMarco JP, Domanski MJ, Rosenberg Y, Schron EB, Kellen JC, Greene HL, Mickel MC, Dalquist JE, Corley SD, for the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) Investigators. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med. 2002; 347: 1825–1833.[Abstract/Free Full Text]
   
4. Wang TJ, Parise H, Levy D, D’Agostino RB Sr, Wolf PA, Vasan RS, Benjamin EJ. Obesity and the risk of new-onset atrial fibrillation. JAMA. 2004; 292: 2471–2477.[Abstract/Free Full Text]
   
5. Coromilas J. Obesity and atrial fibrillation: is one epidemic feeding the other? JAMA. 2004; 292: 2519–2520.[Free Full Text]
   
6. Lauer MS, Eagle KA, Buckley MJ, DeSanctis RW. Atrial fibrillation following coronary artery bypass surgery. Prog Cardiovasc Dis. 1989; 31: 367–378.[CrossRef][Medline] [Order article via Infotrieve]
   
7. Hravnak M, Hoffman LA, Saul MI, Zullo TG, Whitman GR, Griffith BP. Predictors and impact of atrial fibrillation after isolated coronary artery bypass grafting. Crit Care Med. 2002; 30: 330–337.[CrossRef][Medline] [Order article via Infotrieve]
   
8. Zaman AG, Archbold RA, Helft G, Paul EA, Curzen NP, Mills PG. Atrial fibrillation after coronary artery bypass surgery: a model for preoperative risk stratification. Circulation. 2000; 101: 1403–1408.[Abstract/Free Full Text]
   
9. Almassi GH, Schowalter T, Nicolosi AC, Aggarwal A, Moritz TE, Henderson WG, Tarazi R, Shroyer AL, Sethi GK, Grover FL, Hammermeister KE. Atrial fibrillation after cardiac surgery: a major morbid event? Ann Surg. 1997; 226: 501–511.[CrossRef][Medline] [Order article via Infotrieve]
   
10. Aranki SF, Shaw DP, Adams DH, Rizzo RJ, Couper GS, VanderVliet M, Collins JJ Jr, Cohn LH, Burstin HR. Predictors of atrial fibrillation after coronary artery surgery: current trends and impact on hospital resources. Circulation. 1996; 94: 390–397.[Abstract/Free Full Text]
    
11. Mathew JP, Parks R, Savino JS, Friedman AS, Koch C, Mangano DT, Browner WS. Atrial fibrillation in coronary artery bypass grafting surgery: predictors, outcomes, and resource utilization. JAMA. 1996; 276: 300–306.[Abstract]
    
12. Mathew JP, Fontes ML, Tudor IC, Ramsay J, Duke P, Mazer CD, Barash PG, Hsu PH, Mangano DT, for the Investigators of the Ischemia Research and Education Foundation, Multicenter Study of Perioperative Ischemia Research Group. A multicenter risk index for atrial fibrillation after cardiac surgery. JAMA. 2004; 291: 1720–1729.[Abstract/Free Full Text]
    
13. Amar D, Shi W, Hogue CW Jr, Zhang H, Passman RS, Thomas B, Bach PB, Damiano R, Thaler HT. Clinical prediction rule for atrial fibrillation after coronary artery bypass grafting. J Am Coll Cardiol. 2004; 44: 1248–1253.[Abstract/Free Full Text]
    
14. Villareal RP, Hariharan R, Liu BC, Kar B, Lee VV, Elayda M, Lopez JA, Rasekh A, Wilson JM, Massumi A. Postoperative atrial fibrillation and mortality after coronary artery bypass surgery. J Am Coll Cardiol. 2004; 43: 742–748.[Abstract/Free Full Text]
    
15. Habib RH, Zacharias A, Schwann TA, Riordan CJ, Durham SJ, Shah A. Effects of obesity and small body size on operative and long term outcomes of coronary artery bypass surgery: a propensity-matched analysis. Ann Thorac Surg. 2005; 79: 1976–1986.[Abstract/Free Full Text]
    
16. Pritchett AM, Jacobsen SJ, Mahoney DW, Rodeheffer RJ, Bailey KR, Redfield MM. Left atrial volume as an index of left atrial size: a population-based study. J Am Coll Cardiol. 2003; 41: 1036–1043.[Abstract/Free Full Text]
    
17. Executive summary of the clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults. Arch Intern Med. 1998; 158: 1855–1867.[Free Full Text]
    
18. Rubin DB. Estimating causal effects from large data sets using propensity scores. Ann Intern Med. 1997; 127: 757–763.[Abstract/Free Full Text]
    
19. Blackstone EH. Comparing apples and oranges. J Thorac Cardiovasc Surg. 2002; 123: 8–15.[Free Full Text]
    
20. Vaziri SM, Larson MG, Lauer MS, Benjamin EJ, Levy D. Influence of blood pressure on left atrial size: the Framingham Heart Study. Hypertension. 1995; 25: 1155–1160.[Abstract/Free Full Text]
    
21. Gerdts E, Oikarinen L, Palmieri V, Otterstad JE, Wachtell K, Boman K, Dahlof B, Devereux RB, for the Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) Study. Correlates of left atrial size in hypertensive patients with left ventricular hypertrophy: the Losartan Intervention for Endpoint Reduction in Hypertension (LIFE) Study. Hypertension. 2002; 39: 739–743.[Abstract/Free Full Text]
    
22. Lauer MS, Anderson KM, Kannel WB, Levy D. The impact of obesity on left ventricular mass and geometry: the Framingham Heart Study. JAMA. 1991; 266: 231–236.[Abstract]
    
23. Messerli FH, Ventura HO, Reisin E, Dreslinski GR, Dunn FG, MacPhee AA, Frohlich ED. Borderline hypertension and obesity: two prehypertensive states with elevated cardiac output. Circulation. 1982; 66: 55–60.[Abstract/Free Full Text]
    
24. Pelat M, Verwaerde P, Merial C, Galitzky J, Berlan M, Montastruc JL, Senard JM. Impaired atrial M(2)-cholinoceptor function in obesity-related hypertension. Hypertension. 1999; 34: 1066–1072.[Abstract/Free Full Text]
    
25. Iacobellis G, Ribaudo MC, Leto G, Zappaterreno A, Vecci E, Di Mario U, Leonetti F. Influence of excess fat on cardiac morphology and function: study in uncomplicated obesity. Obes Res. 2002; 10: 767–773.[Medline] [Order article via Infotrieve]
    
26. Engeli S, Sharma AM. The renin-angiotensin system and natriuretic peptides in obesity-associated hypertension. J Mol Med. 2001; 79: 21–29.[CrossRef][Medline] [Order article via Infotrieve]
    
27. Vincent HK, Powers SK, Stewart DJ, Shanely RA, Demirel H, Naito H. Obesity is associated with increased myocardial oxidative stress. Int J Obes Relat Metab Disord. 1999; 23: 67–74.[CrossRef][Medline] [Order article via Infotrieve]
    
28. Zhou YT, Grayburn P, Karim A, Shimabukuro M, Higa M, Baetens D, Orci L, Unger RH. Lipotoxic heart disease in obese rats: implications for human obesity. Proc Natl Acad Sci U S A. 2000; 97: 1784–1789.[Abstract/Free Full Text]
    
29. Kannel WB, Wolf PA, Benjamin EJ, Levy D. Prevalence, incidence, prognosis, and predisposing conditions for atrial fibrillation: population-based estimates. Am J Cardiol. 1998; 16: 82:2N–9N.
    
30. Mokdad AH, Ford ES, Bowman BA, Dietz WH, Vinicor F, Bales VS, Marks JS. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA. 2003; 289: 76–79.[Abstract/Free Full Text]
    
31. Concato J, Shah N, Horwitz RI. Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med. 2000; 342: 1887–1892.[Abstract/Free Full Text]
    
32. Ioannidis JP, Haidich AB, Pappa M, Pantazis N, Kokori SI, Tektonidou MG, Contopoulos-Ioannidis DG, Lau J. Comparison of evidence of treatment effects in randomized and nonrandomized studies. JAMA. 2001; 286: 821–830.[Abstract/Free Full Text]
    

This article has been cited by other articles:

				N. Echahidi, P. Pibarot, G. O'Hara, and P. Mathieu Mechanisms, Prevention, and Treatment of Atrial Fibrillation After Cardiac Surgery J. Am. Coll. Cardiol., February 26, 2008; 51(8): 793 - 801. [Abstract] [Full Text] [PDF]

				M. A. Albert, N. Halevy, and E. M. Antman Preoperative Evaluation for Cardiac Surgery Card. Surg. Adult, January 1, 2008; 3(2008): 261 - 280. [Full Text]

				D. P. Mason, D. H. Marsh, J. M. Alster, S. C. Murthy, A. M. McNeill, M. M. Budev, A. C. Mehta, G. B. Pettersson, and E. H. Blackstone Atrial Fibrillation After Lung Transplantation: Timing, Risk Factors, and Treatment Ann. Thorac. Surg., December 1, 2007; 84(6): 1878 - 1884. [Abstract] [Full Text] [PDF]

				N. Echahidi, D. Mohty, P. Pibarot, J.-P. Despres, G. O'Hara, J. Champagne, F. Philippon, P. Daleau, P. Voisine, and P. Mathieu Obesity and Metabolic Syndrome Are Independent Risk Factors for Atrial Fibrillation After Coronary Artery Bypass Graft Surgery Circulation, September 11, 2007; 116(11_suppl): I-213 - I-219. [Abstract] [Full Text] [PDF]

				N. Echahidi, P. Pibarot, J.-P. Despres, J.-M. Daigle, D. Mohty, P. Voisine, R. Baillot, and P. Mathieu Metabolic Syndrome Increases Operative Mortality in Patients Undergoing Coronary Artery Bypass Grafting Surgery J. Am. Coll. Cardiol., August 28, 2007; 50(9): 843 - 851. [Abstract] [Full Text] [PDF]

				Writing Committee Members, V. Fuster, L. E. Ryden, D. S. Cannom, H. J. Crijns, A. B. Curtis, K. A. Ellenbogen, J. L. Halperin, J.-Y. Le Heuzey, G. N. Kay, et al. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: full text: A report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation) Developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society Europace, September 1, 2006; 8(9): 651 - 745. [Full Text] [PDF]

				V. Fuster, L. E. Ryden, D. S. Cannom, H. J. Crijns, A. B. Curtis, K. A. Ellenbogen, J. L. Halperin, J.-Y. Le Heuzey, G. N. Kay, J. E. Lowe, et al. ACC/AHA/ESC 2006 Guidelines for the Management of Patients With Atrial Fibrillation: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation) Developed in Collaboration With the European Heart Rhythm Association and the Heart Rhythm Society J. Am. Coll. Cardiol., August 15, 2006; 48(4): e149 - e246. [Full Text] [PDF]

				W. J. Manning and E. V. Gelfand Left Atrial Size and Postoperative Atrial Fibrillation: The Volume of Evidence Suggests it Is Time to Break an Old Habit J. Am. Coll. Cardiol., August 15, 2006; 48(4): 787 - 789. [Full Text] [PDF]

				V. Fuster, L. E. Ryden, D. S. Cannom, H. J. Crijns, A. B. Curtis, K. A. Ellenbogen, J. L. Halperin, J.-Y. Le Heuzey, G. N. Kay, J. E. Lowe, et al. ACC/AHA/ESC 2006 Guidelines for the Management of Patients With Atrial Fibrillation: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): Developed in Collaboration With the European Heart Rhythm Association and the Heart Rhythm Society Circulation, August 15, 2006; 114(7): e257 - e354. [Full Text] [PDF]

				M. A. Arias Prevention of atrial fibrillation following cardiac surgery. JAMA, May 17, 2006; 295(19): 2247 - 2247. [Full Text] [PDF]

				M. A. Arias, J. Sanchez-Gila, and D. Amar Obesity As a Risk Factor for Developing Postoperative Atrial Fibrillation Chest, March 1, 2006; 129(3): 828 - 829. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: 112/21/3247    most recent CIRCULATIONAHA.105.553743v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar