This Article Abstract 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 Aranki, S. F. Articles by Burstin, H. R. Search for Related Content PubMed PubMed Citation Articles by Aranki, S. F. Articles by Burstin, H. R. Pubmed/NCBI databases Medline Plus Health Information Coronary Artery Bypass Surgery

(Circulation. 1996;94:390-397.)
© 1996 American Heart Association, Inc.

Articles

Predictors of Atrial Fibrillation After Coronary Artery Surgery

Current Trends and Impact on Hospital Resources

Sary F. Aranki, MD; David P. Shaw, MD; David H. Adams, MD; Robert J. Rizzo, MD; Gregory S. Couper, MD; Martha VanderVliet, RN; John J. Collins, Jr, MD; Lawrence H. Cohn, MD; Helen R. Burstin, MD, MPH

the Division of Cardiac Surgery (S.F.A., D.P.S., D.H.A., R.J.R., G.S.C., J.J.C., L.H.C.), Department of Surgery, Brigham and Women's Hospital; and the Division of General Medicine (H.R.B., M.V.V.), Department of Medicine, Harvard Medical School, Boston, Mass.

Correspondence to Sary F. Aranki, MD, Division of Cardiac Surgery, Brigham and Women's Hospital, 75 Francis St, Boston, MA 02115.

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Background Atrial fibrillation (AF) after coronary artery bypass surgery (CABG) is the most common sustained arrhythmia. Its pathophysiology is unclear, and its prevention and management remain suboptimal. The aim of this prospective study was to determine the current incidence of AF, identify its clinical predictors, and examine its impact on resource utilization.

Methods and Results Over a 12-month period ending July 31, 1994, a CABG procedure was performed on 570 consecutive patients (age range, 32 to 87 years; median age, 67 years; 232 [41%] were 70 years; 175 [31%] were women; 173 [30%] were diabetics; 364 [65%] required nonelective surgery; 86 [15%] had had a prior CABG; and 86 [15%] had had prior percutaneous transluminal coronary angioplasty). AF occurred in 189 patients (33%). The median age for patients with AF was 71 years compared with 66 for patients without (P=.0001). Multivariate logistic regression analysis (odds ratio, ±95% CI, P value) was used to identify the following independent predictors of postoperative AF: increasing age (age 70 to 80 years [OR=2; CI, 1.3 to 3; P=.002], age >80 years [OR=3; CI, 1.6 to 5.8; P=.0007]), male gender (OR=1.7; CI, 1.1 to 2.7; P=.01), hypertension (OR=1.6; CI, 1.0 to 2.3; P=.03), need for an intraoperative intra-aortic balloon pump (OR=3.5; CI, 1.2 to 10.9; P=.03), postoperative pneumonia (OR=3.9; CI, 1.3 to 11.5; P=.01), ventilation for >24 hours (OR=2; CI, 1.3 to 3.2; P=.003), and return to the intensive care unit (OR=3.2; CI, 1.1 to 8.8; P=.03). The mean length of hospital stay after surgery was 15.3±28.6 days for patients with AF compared with 9.3±19.6 days for patients without AF (P=.001). The adjusted length of hospital stay attributable to AF was 4.9 days, corresponding to $10 055 in hospital charges.

Conclusions AF remains the most common complication after CABG and consequently is a drain on hospital resources. Concerted efforts to reduce the incidence of AF and the associated increased length of stay would result in substantial cost savings and decrease patient morbidity.

Key Words: atrial flutter • atrial fibrillation • coronary disease • bypass surgery

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Atrial fibrillation was recognized as a major cause of morbidity soon after the introduction of CABG for the treatment of atherosclerotic CAD.¹ Its rate of occurrence varies widely depending on the definition used, criteria for diagnosis, and mode of postoperative monitoring (intermittent or continuous) and on the changing profile of patients undergoing coronary surgery^{2 3} ; age is consistently identified as a major independent predictor for the development of postoperative AF.^{3 4 5 6 7 8} The evolution of postoperative continuous on-line ECG monitoring and Holter monitoring and the increasing age of the patient population are the major factors that have contributed to the consistently higher incidence of AF in recent years.^{2 3}

The pathophysiological mechanisms responsible for the high incidence of AF after cardiac surgery in general and after CABG surgery in particular remain unclear. What is certain is that its incidence far exceeds its reported prevalence in the general population and in patients with atherosclerotic CAD.³ Similarly, it is significantly higher than the reported incidence of AF after major noncardiac surgery regardless of CAD status.⁹ It seems, therefore, that there are inherent, yet undetermined mechanisms that predispose a large proportion of CABG patients to develop AF. Some of these responsible mechanisms include ß-blocker withdrawal, the use of cardiopulmonary bypass, inadequate atrial protection, and overmanipulation of the right atrium. However, the maintenance of SR in the majority of CABG patients who are subjected to the same conditions would be difficult to explain. A more plausible explanation would probably be the preexistence of electrophysiological abnormalities that are amplified during surgery and, when subjected to an adverse trigger(s) postoperatively, could be manifested as AF.¹⁰ Unfortunately, preoperative and intraoperative electrophysiological identification of these patients is highly complex, time consuming, and expensive. Consequently, identification of clinical predictors for the development of AF remains the most practical approach. Targeting patients at risk of postoperative AF with intensive prophylactic measures may drastically reduce the length of hospital stay and the associated high costs.

The objective of this prospective study was to determine the current incidence of AF and identify its clinical predictors. Hospital resource utilization and consequent financial impacts were also examined.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
As part of a multi-institutional project of the AMCC, Quality Measurement and Management Initiative, the Coronary Revascularization Project was designed to measure the quality of care among patients who undergo either CABG or PTCA. As part of this effort to collect and analyze process and outcomes data at each of the 12 AMCC institutions, the Brigham and Women's Hospital began collection of prospective data for all patients who underwent CABG or PTCA as of September 15, 1993. In this study, we report clinical data for 570 consecutive patients who underwent CABG at the Brigham and Women's Hospital between September 15, 1993, and July 31, 1994. Patients who underwent valve replacements were excluded. Discharge abstract data confirmed capture of 97% of all patients who underwent CABG at Brigham and Women's Hospital during this period.

Data Collection
As part of a standard protocol, medical record review for completion of a clinical data form was performed. Physicians prospectively collected clinical data on cardiac history, including angina, prior myocardial infarction, and prior cardiac surgeries, and assessment of standard comorbidities. Intraoperative and postoperative data, including complications and adverse events, were assessed through a thorough medical record review with the use of standardized AMCC definitions.

The major outcome measure for this study was the development of postoperative AF. The relation of postoperative AF to preoperative, intraoperative, and postoperative parameters was assessed. Secondary outcomes for this study included LOS and hospital charges. LOS was assessed from admission to discharge, as well as from date of CABG to date of discharge. The latter measure of surgical LOS was examined to more directly assess the increased LOS attributable to postoperative complications, such as AF. We also examined hospital charges for the increased LOS. Although charges could not be directly attributed to the effect of postoperative AF, we used multivariate modeling to examine AF as well as other potential correlates of increased LOS.

AF rates were also risk stratified with the use of the Cleveland Clinic Foundation CABG clinical severity scoring system.¹¹ The prospective data collection allowed us to capture all preoperative factors in the clinical severity score, with the exception of operative aortic valve stenosis.

Operative Techniques
Cardiopulmonary bypass with moderate hemodilution, (hematocrit, 20% to 25%), moderate systemic hypothermia (28°C to 30°C), flow rates of 1.5 to 2 L·min^-1·m^-2, and a mean pressure of 50 to 70 mm Hg was used. Intermittent cold hyperkalemic cardioplegia was used in all patients; this included crystalloid or blood cardioplegia and delivery of cardioplegia through the antegrade or the antegrade/retrograde routes. An initial volume of 500 mL (30 mEq potassium chloride and 5 mEq sodium bicarbonate in 1 L of dextrose 2.5% and 1/2 normal saline) of cardioplegia was given at a rate of 100 mL/min. A 250-mL supplement of a lower concentration (10 mEq of potassium chloride) was given every 20 minutes. Blood cardioplegia had the same concentration (4:1 blood/5% dextrose) and was given at the same volume and rate as the crystalloid solution. The distal and proximal anastomoses were constructed during a single period of total aortic occlusion,¹² or the proximal anastomoses were constructed after removal of the total occluding clamp. A left ventricular vent was not routinely used in all patients.

Postoperative Protocols
Patients were weaned off of the ventilator as soon as they met the following criteria: hemodynamic stability, no major bleeding, normothermia, and consciousness with adequate pain control.¹³ Potassium and magnesium supplements were given as necessary to maintain electrolyte balance within the normal range. All patients were started on aspirin within the first 24 hours after surgery. AF prophylaxis was used in all patients and was also started within the first 24 hours after surgery; it was continued for 6 weeks. Digoxin was given to patients with an ejection fraction of <30% and for those in whom the use of ß-blockers was contraindicated because of low blood pressure or for another noncardiac cause. ß-Blockers (usually metoprolol 25 mg BID to 50 mg TID) were used in the remainder of the patients. All patients were routinely monitored through telemetry with continuous display of the ECGs on multiple oscilloscopes simultaneously in the intensive or continuous care unit. The monitoring continued until the day of discharge. Twelve-lead ECGs were done routinely for the first 3 postoperative days and as necessary afterward to confirm and document any ischemic or rhythm incidents.

Definitions
All definitions for this study are based on the AMCC study protocol; these included definitions from the Northern New England Cardiovascular Disease Study Groups and The Duke Databank for Cardiovascular Diseases. Significant postoperative atrial arrhythmias were defined as arrhythmias that required either medications or pacing. At our institutions, these significant postoperative atrial arrhythmias were almost all secondary to AF. Mortality was defined as in-hospital mortality only. An MI was defined as chest pain, nausea, diaphoresis, or hypotension associated with the development of new Q waves on the ECG.

Data Analysis
The association of all preoperative, intraoperative, and postoperative factors with the occurrence of postoperative AF was evaluated with the use of ² and Student's t tests. Univariate factors significant to P<.10 were entered into logistic regression models. We used logistic regression to assess the independent correlates of AF. All models were examined in both forward and backward logistic regression with comparable results. For this analysis, we present the OR and 95% CI for each significant correlate from backward elimination models. In a subanalysis, we also used logistic regression to determine the preoperative correlates of AF. Linear regression models for LOS were used to estimate the number of hospital days attributable to AF, controlling for all clinical factors significantly associated with the development of AF in the logistic regression model. Hospital charges were based on the number of inpatient days attributable to AF in multivariate analysis.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Patient age ranged from 32 to 87 years, with a median age of 67 years (232 [41%] were 70 years, 175 [31%] were women, 173 [30%] were diabetics, 343 [60%] had had a prior MI, 112 [20%] had a history of congestive heart failure, 364 [65%] required nonelective surgery, 61 [11%] had had a preoperative IABP, 86 [15%] had had prior CABG, and 86 [15%] had had prior PTCA). The age distribution of the patient population is shown in Fig 1. The increasing age of our current patient population is reflected by the fact that 178 (31%) of the patients were between 70 and 80 years of age, and 54 (9%) were >80 years.

View larger version (13K): [in this window] [in a new window]   Figure 1. Age distribution of the overall patient population (n=570).

AF occurred in 189 patients (33%), and SR was maintained in 381 (67%). The time of occurrence of AF was clustered around the first 4 postoperative days, with 70% of the patients developing AF during those days and only 6% developing AF after the sixth postoperative day (Fig 2). The median age of patients with AF was 71 compared with 66 years for SR patients (P=.0001). The age distribution for AF patients is shown in Fig 3.

View larger version (15K): [in this window] [in a new window]   Figure 2. Time of occurrence of postoperative AF.

View larger version (12K): [in this window] [in a new window]   Figure 3. Age distribution of AF patients (n=189).

Baseline Patient Characteristics
The preoperative baseline patient characteristics are shown in Table 1 for each group (SR or AF), with the appropriate univariate P values. Patients in the AF group were significantly older and were more likely to be males, to be hypertensive, to have had a prior MI, and to have coexisting congestive heart failure, peripheral vascular disease, and chronic lung disease. The use of preoperative ß-blockers and calcium channel blockers did not differ significantly between the two groups. These preoperative characteristics showed that the patients in the AF group were older and had more severe cardiac disease and comorbid conditions. Coronary artery dominance derived from preoperative coronary angiograms was similar for AF and SR patients.

View this table: [in this window] [in a new window]   Table 1. Preoperative Patient Characteristics

With the use of the Cleveland Clinic Risk Stratification System,¹¹ patients with AF had a mean clinical severity score of 7.5±3.9 compared with 6.7±4.1 for the SR group (P=.03). The distribution of the clinical severity score for the entire patient cohort is shown in Fig 4. The distribution of AF according to the preoperative clinical severity score is shown in Fig 5. Patients with moderate (score, 6 to 8) or severe (score, 9+) scores accounted for three fourths of all AF patients.

View larger version (14K): [in this window] [in a new window]   Figure 4. Preoperative risk stratification of the overall patient population (n=570). Based on the Cleveland Clinic clinical severity scores.11

View larger version (15K): [in this window] [in a new window]   Figure 5. Distribution of AF patients (n=189) stratified according to the Cleveland Clinic Preoperative Clinical Severity Scores.11

Operative Variables
The operative variables are summarized in Table 2. The number of bypass grafts, use of the internal mammary artery, need for coronary endarterectomy, ischemia time, cardiopulmonary bypass time, and use of hemofiltration were similar for the SR and AF groups. The only significant difference was in the need for an operative IABP (P<.001).

View this table: [in this window] [in a new window]   Table 2. Operative Variables

Postoperative Outcomes
There were 13 perioperative deaths, for an overall OM of 2.3% (SR group, 1.8%; AF group, 3.9%; P=.15). The OM correlated well with the Cleveland Clinic clinical severity score categories. The mean risk score was 10.5±3.9 for the patients with perioperative death versus 6.9±4 for the early survivals (P=.0047). The various perioperative complications are shown in Table 3.

View this table: [in this window] [in a new window]   Table 3. Postoperative Complications

Multivariate Analysis
Multivariate logistic regression analysis identified age, male gender, history of hypertension, need for an intraoperative IABP, postoperative pneumonia, need for prolonged ventilation (>24 hours), and return to the intensive care unit to be independent correlates for AF. The OR, the 95% CI, and probability value for each multivariate predictor are shown in Table 4.

View this table: [in this window] [in a new window]   Table 4. Multivariate Predictors of AF

In a subanalysis of potential correlates of AF available before surgery, male gender (OR=1.6; 95% CI, 1.1 to 2.4; P=.02), age 70 to 80 years (OR=2.2; 95% CI, 1.5 to 3.2; P=.0001), age >80 years (OR=3.2; 95% CI, 1.8 to 5.8; P=.0001), and history of prior MI (OR=1.6; 95% CI, 1.1 to 2.3; P=.01) were independently associated with AF.

Resource Utilization
The overall mean hospital LOS (including catheterization) was 12.9±17.5 days and a median of 10 days. The overall mean LOS after surgery was 11.3±23.2 days and a median of 7 days. The corresponding values for patients who had SR or AF are shown in Table 5. The proportion of patients in each group discharged on different postoperative days is given in Fig 6. Only 17% of the patients were discharged after the 10th postoperative day in the SR group compared with 47% for the AF group.

View this table: [in this window] [in a new window]   Table 5. Hospital LOS

View larger version (35K): [in this window] [in a new window]   Figure 6. LOS for patients in SR and for AF patients.

In a linear regression model for LOS, we examined potential correlates of increased LOS. In addition to AF, other univariate correlates of increased LOS included in these models included gender, age, and postoperative complications (postoperative balloon pump, pneumonia, return to operating room, prolonged ventilation, postoperative sternal wound infection). After adjusting for age and other potential correlates of LOS, we found that AF remained an independent correlate of LOS. From these linear regression models, we examined the adjusted mean LOS for AF. After adjusting for the factors listed above, we found that AF was independently associated with an increased LOS of 4.9 days. The additional hospital charges associated with the increased LOS were calculated, after excluding those hospital charges directly attributable to the surgical procedure. For patients who underwent cardiac catheterization and CABG (diagnosis-related group 106), we found that the increased LOS attributable to AF resulted in extra hospital charges of $11 500 per patient. For patients who underwent CABG only (diagnosis-related group 107), the corresponding extra charges were $10 055 per patient.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
Major advances in cardiac surgery in general, and coronary surgery in particular, were not paralleled by a major decline in the incidence of postoperative AF. On the contrary, such advances may have paradoxically contributed to an increase in the incidence of AF that has been seen in recent years.³ Improvements in surgical techniques, myocardial protection, cardiopulmonary bypass, and care of the critically ill patient in the operating room and the intensive care unit have resulted in a major rearrangement of the risk profile of CABG patients. Coupled with the rapid evolution of the technology and practice of percutaneous revascularization procedures, current surgical patients are significantly older and sicker than their counterparts were 10 or 15 years ago. Such patients are at a higher risk for an increased morbidity and mortality.¹⁴ Advances in continuous monitoring technology have led to more frequent diagnosis of AF, which remains the most common cause of morbidity after coronary surgery.² Although the increased hospital LOS attributable to AF is modest compared with other more serious complications such as acute respiratory distress syndrome, septicemia, or pneumonia,¹⁵ the cumulative impact of AF that occurs with greater incidence on hospital resource utilization is far more significant.

The results of this study confirm the high incidence of postoperative AF (33%) in patients undergoing CABG and identify its independent clinical predictors. In addition, risk stratification of CABG patients appears to be closely related to the development of AF, with a significantly higher incidence in patients with moderate or severe clinical scores. The adjusted hospital LOS attributed directly to AF and derived from a linear regression model was 4.9 days for every patient who developed AF.

Incidence of AF
The prevalence of AF in the general population, in patients with CAD, and after major nonthoracic procedures should serve as a base reference to appreciate the magnitude of AF after CABG. The Framingham Study¹⁶ reported a 1.7% overall incidence of AF that incrementally increased with age. The incidence of AF in patients <70 years old is estimated to be 0.4%, whereas for patients >70 years old, the incidence is 2% to 4%.¹⁷ In a recent report from the Cardiovascular Health Study Collaborative Research Group,¹⁸ AF was diagnosed in 4.8% of women and 6.2% of men. The incidence was 9.1%, 4.6%, and 1.6% for patients with evidence of clinical, subclinical, or no cardiovascular disease, respectively. In the presence of atherosclerotic CAD, the incidence of AF was 3.6%. The incidence of AF after major nonthoracic procedures is reported to be 5%.^{9 10 19} The reported incidence of AF after CABG varies widely, with a reported incidence of 5% to 40%²⁰ ; an incidence of well over 70% has been reported.^{21 22} A more accurate estimate can be found in the meta-analysis of 24 controlled, randomized trials (total number of patients, 2458) by Andrews et al² at the Brigham and Women's Hospital, who reported a pooled proportion of supraventricular arrhythmias (fibrillation or flutter) to be 26.7% with a 95% CI of 24.7 to 29.1. They attributed the variation in the reported incidence of these arrhythmias to the intensity and duration of postoperative monitoring. The incidence of arrhythmias in the trials that used a period of Holter monitoring was 41.3% compared with 19.9% for the trials that did not use any period of Holter monitoring. A similar trend was also reported by Michelson et al.²³

Another important factor contributing to the increase in the incidence of AF in recent years is the changing profile of patients undergoing CABG. Increasing age has been a consistent independent predictor for AF after CABG.^{2 3 4} At the Brigham and Women's Hospital, the mean age of our CABG patients increased by 12 years between 1970 and 1992.²⁴ The incidence of AF has increased during this period from 5%²⁵ to 33% in the present study. A similar trend was reported by Creswell et al³ ; the incidence of atrial arrhythmias increased from 25% to 36% in only 6 years, during which the mean age of the patients increased by 2.2 years.

Clinical Predictors of AF
Increasing age and systemic hypertension are probably the two most commonly identified risk factors for the development of AF in the general population.^{18 26} In addition, increasing age has been consistently shown to be an independent predictor for AF after CABG.^{3 4 5 6 7 8} In contrast, no other study has identified the independent role of hypertension as a predictor of AF in the postoperative setting. Although older patients are more likely to be hypertensive, we found both advanced age and hypertension to be independent predictors of postoperative AF. Age-related structural changes, such as increased fibrosis and atrial dilatation,²⁷ and age-related comorbidities seem to be responsible for the increased incidence of AF with in-creasing age.³ Similarly, hypertension-related structural changes may have a significant role in the genesis of associated arrhythmias. A number of changes in the hypertrophied heart, such as fibrosis, may act as a substrate for reentry arrhythmias.^{28 29} This is supported by results of examination of endomyocardial biopsy samples from hypertensive patients, showing increased myocardial fibrosis in patients with documented arrhythmias.³⁰ Other biopsy studies showed associated electrophysiological abnormalities in the hypertrophied heart, causing abnormal depolarization, conduction velocities, and impulse propagation that predispose to reentrant arrhythmia.³¹ In addition, increased myocardial fibrosis may lead to a common abnormality related to age and hypertension, which is the decline in left ventricular diastolic function and compliance, and such abnormalities may be compounded by the presence of CAD^{4 6} and, therefore, increased incidence of arrhythmias.³

Another independent predictor for the development of postoperative AF in this study was male gender. A similar finding was reported by Fuller et al,⁴ who had 190 female patients of a total of 1666 patients (11%). They postulated that perhaps a hormone-related protective mechanism may account for the lower incidence of AF in female patients or that other gender-related factors were not accounted for in the multivariate analysis. They advised caution in interpreting these results because of the small number of female patients and because of the weak association (P=.02). In the present study, there were 175 female patients, accounting for 31% of the total, with a strong association between male gender and AF (P=.01). A similar trend has been reported in the elderly, where AF was significantly associated with the male gender in the general population as a univariate correlate.¹⁸ It appears, therefore, that an unknown protective mechanism may be associated with the reduced incidence of postoperative AF in female patients.

The need for an intraoperative IABP was also an independent predictor for the development of postoperative AF. An intraoperative IABP is usually necessary because of severe myocardial dysfunction secondary to myocardial necrosis or a stunned myocardium, resulting in heart failure. Poor left ventricular function and congestive heart failure are associated with a greater risk for the development of AF.^{16 17} The other predictors for the development of postoperative AF were pneumonia, the need for extended ventilation (>24 hours), and return to the intensive care unit for any reason. These factors are all associated with increased vulnerability to AF due to hypoxia, hypovolemia, sepsis, and electrolyte imbalances.^{7 17 26 32 33 34 35} Although the incidence of AF was significantly associated with postoperative complications (Table 3), it is unlikely that AF was the responsible etiology but rather was the result of the above systemic disturbances associated with these complications.

The clinical predictors of postoperative AF in this study may be consistent with the pathophysiological mechanisms responsible for the pathogenesis of AF as proposed by Cox¹⁰ and based on an experimental study by his group.³⁶ An underlying electrophysiological basis appears to be responsible for the increased vulnerability of certain patients for the development of postoperative AF. Such vulnerability is based on the concept of "dispersion of refractoriness," which describes the nonuniformity of an infinite number of local atrial refractory periods. As a result, one or more regions of the atria would have relatively short refractory periods in close proximity to much longer ones, rather than the gradual transition that occurs in nonvulnerable people. It is further stipulated¹⁰ that such vulnerability to AF requires an appropriate trigger or triggers for the actual development of AF. The fact that inadequate atrial cooling and protection^{32 37} are common with modern cardioplegia techniques and extended periods of ischemia led Cox¹⁰ to postulate that atrial ischemia was probably the trigger responsible for the development of AF in vulnerable patients. This was supported by the clinical studies that showed extended cross-clamp times to be an independent predictor for AF.^{3 38 39} However, in the present study and others,^{4 7} the cross-clamp time was not an independent predictor. In addition, the incidence of AF in noncardioplegic operations with short ischemia time^{40 41} was as high and similar to that in cardioplegic procedures with an extended ischemia time. Also, the very low incidence of AF after pediatric cardiac surgery and cardiac transplantation does not support the hypothesis that inadequate atrial protection or atrial ischemia is the sole trigger for postoperative AF in vulnerable patients. It is more likely that multifactorial triggering mechanisms are responsible. Preexisting structural changes, such as those related to age and long-standing hypertension; the effects of cardiopulmonary bypass and cardioplegia; and the presence of postoperative electrolyte imbalance, hypoxia, hypovolemia, and sepsis may all be important triggers when they occur in different combinations. This predisposition for AF appears to be an attractive and plausible explanation for the vulnerability of certain patients to AF after CABG. This is supported by the presence of electrophysiological predictors of AF. Preoperative predictors can reveal intra-atrial conduction defects through measurement of prolonged P-wave duration,^{42 43} and intraoperative predictors can reveal atrial vulnerability through splitting of the atrial electrogram during atrial premature stimulation⁴⁴ or through pace ineducability of AF before cardiopulmonary bypass.⁴⁵ However, these techniques are cumbersome and time consuming and are not practical when large numbers of patients are involved.

The multivariate analyses used in the present study were used to identify clinical predictors for postoperative AF that are consistent with those associated with AF in the general population. It is likely that these different variables in the presence of CAD will predispose certain patients to an electrophysiological abnormality¹⁰ that is amplified during surgery with consequent AF in the presence of several postoperative triggers.

Resource Utilization
The number of hospital days attributable to postoperative AF was an additional 4.9 days per patient. This corresponds to an additional $10 055 to $11 500 of hospital charges per patient and $2 million of additional hospital charges for this cohort of 570 patients as a result of postoperative AF. It is estimated that annually, 300 000 patients have CABG in the United States.⁴⁶ Although our hospital charges might not be generalizable, if one third to one fourth of these patients develop AF, the financial implications would be enormous.

AF certainly is not the most expensive complication that can occur after CABG. However, because it is the most frequent complication, the cumulative cost of AF will exceed that of any other complication. In a study by Taylor et al⁴⁷ that examined the economic consequences of postoperative complications associated with CABG, AF was one of the least expensive complications, but it was the most common, occurring in 20% of the patients. Respiratory failure and sternal wound infection were the most expensive complications, but they occurred in only 3% and 0.4% of patients, respectively. The increased cost associated with AF compared with patients with no complications was $5000 per patient. In a similar study by Mauldin et al,¹⁵ a major arrhythmia occurred in 25% of the patients and was one of the least expensive complications. Septicemia and adult respiratory distress syndrome were the most expensive complications, but they occurred in only 0.5% and 0.4% of patients, respectively. The increase in cost associated with a major arrhythmia compared with patients without any complications was $6000 per patient.

It seems that AF is a costly complication after CABG, and any reduction in its rate of occurrence will result in enormous savings. Identifying patients at risk through the use of clinical predictors and, eventually, through more practical and less expensive electrophysiological predictors and targeting these patients with more intensive prophylactic measures may result in a reduction in the incidence of postoperative AF. An alternative, and probably a more practical, approach would be an early discharge of AF patients soon after initiation of treatment if the patient has an adequate rate of control and no contraindications for outpatient anticoagulation. Most patients would cardiovert to SN within 6 weeks,⁴⁸ and electrical cardioversion could be done on an outpatient basis for the few patients who do not cardiovert to SN on pharmacological treatment. Such an approach would result in a considerable reduction in hospital LOS and significant cost reduction.

Summary and Conclusions
AF is the most common complication after CABG. Its clinical predictors are consistent with those associated with AF in the general population. Age continues to be a major and consistent predictor of AF as has been shown by the majority of similar studies. AF is a major drain on hospital resources with a significant contribution to the escalating health costs associated with CABG. Identifying and targeting patients at risk with aggressive prophylactic measures may lead to reduced patient morbidity and could lead to major cost savings.

	Selected Abbreviations and Acronyms

 

AF = atrial fibrillation AMCC = Academic Medical Center Consortium CABG = coronary artery bypass graft surgery CAD = coronary artery disease IABP = intra-aortic balloon pump LOS = length of stay MI = myocardial infarction OM = operative mortality OR = odds ratio PTCA = percutaneous transluminal coronary angioplasty SR = sinus rhythm

	Acknowledgments

 
The authors thank Kathleen LaMae, Nicole Beckel, and Timothy Sullivan for their valuable assistance with preparation of the manuscript. We also acknowledge Dr Anthony Komaroff, Dr Thomas Lee, and Dr Elliott Antman for comments on earlier drafts of the manuscript.

	Footnotes

 
Presented at the 67th Scientific Sessions of the American Heart Association, Dallas, Tex, November 14-17, 1994.

Received July 31, 1995; revision received January 23, 1996; accepted January 29, 1996.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Favaloro RG, Effler DB, Groves LK, Sheldon WC, Riahi M. Direct myocardial revascularization with saphenous vein autograft. Dis Chest. 1969;56:279-283.[Abstract]
   
2. Andrews TC, Reimold SC, Berlin JA, Antman EM. Prevention of supraventricular arrhythmias after coronary artery bypass surgery. Circulation. 1991;84(suppl III):III-236-III-244.
   
3. Creswell LL, Schuessler RB, Rosenbloom M, Cox JL. Hazards of postoperative atrial arrhythmias. Ann Thorac Surg. 1993;56:539-549.[Abstract]
   
4. Fuller JA, Adams GG, DiComp MS, Buxton B. Atrial fibrillation after coronary artery bypass grafting. J Thorac Cardiovasc Surg. 1989;97:821-825.[Abstract]
   
5. Leitch JW, Thomson D, Baird DK, Harris PJ. The importance of age as a predictor of atrial fibrillation and flutter after coronary artery bypass grafting. J Thorac Cardiovasc Surg. 1990;100:338-342.[Abstract]
   
6. Crosby LH, Pifalo WB, Woll KR, Burkholder JA. Risk factors for atrial fibrillation after coronary artery bypass grafting. Am J Cardiol. 1990;66:1520-1522.[Medline] [Order article via Infotrieve]
   
7. Hashimoto K, Ilstrup DM, Schaff HV. Influence of clinical and hemodynamic variables on risk of supraventricular tachycardia after coronary artery bypass. J Thorac Cardiovasc Surg. 1991;101:56-65.[Abstract]
   
8. Frost L, Molgaard H, Christiansen EH, Hjortholm K, Paulsen PK, Thomsen PE. Atrial fibrillation and flutter after coronary artery bypass surgery: epidemiology, risk factors and preventive trials. Int J Cardiol. 1992;36:253-261.[Medline] [Order article via Infotrieve]
   
9. Goldman L. Supraventricular tachyarrhythmia in hospitalized adults after surgery: clinical correlates in patients over 40 years of age after major noncardiac surgery. Chest. 1978;73:450-454.[Abstract/Free Full Text]
   
10. Cox JL. A perspective of postoperative atrial fibrillation in cardiac operations. Ann Thorac Surg. 1993;56:405-409.[Medline] [Order article via Infotrieve]
    
11. Higgins TL, Estafanous FG, Loop FD, Beck GJ, Blum JM, Paranadi L. Stratification of morbidity and mortality outcome by preoperative risk factors in coronary artery bypass patients: a clinical severity score. JAMA. 1992;267:2344-2348.[Abstract]
    
12. Aranki SF, Rizzo RJ, Adams DH, Couper GS, Kinchla NM, Gildea JS, Cohn LH. Single-clamp technique: an important adjunct to myocardial and cerebral protection in coronary operations. Ann Thorac Surg. 1994;58:296-303.[Abstract]
    
13. Antman EM. Medical management of the patient undergoing cardiac surgery. In: Braunwald E, ed. Heart Disease: A Textbook of Cardiovascular Medicine. 4th ed. Philadelphia, Pa: WB Saunders; 1992:1670-1693.
    
14. Jones EL, Weintraub WS, Craver JM, Guyton RA, Cohen CL. Coronary bypass surgery: is the operation different today? J Thorac Cardiovasc Surg. 1991;101:108-115.[Abstract]
    
15. Mauldin PD, Weintraub WS, Becker ER. Predicting hospital costs for first time coronary artery bypass grafting from preoperative and postoperative variables. Am J Cardiol. 1994;74:772-775.[Medline] [Order article via Infotrieve]
    
16. Kannell WB, Abbott RD, Savage DD, McNamara PM. Epidemiologic features of chronic atrial fibrillation: the Framingham Study. N Engl J Med. 1982;306:1018-1022.[Abstract]
    
17. Alpert JS, Petersen P, Godtfredsen J. Atrial fibrillation: natural history, complications, and management. Annu Rev Med. 1988;39:41-52.[Medline] [Order article via Infotrieve]
    
18. Furberg CD, Psaty BM, Manolio TA, Gardin JM, Smith VE, Rautaharju PM. Prevalence of atrial fibrillation in elderly subjects (the Cardiovascular Health Study). Am J Cardiol. 1994;74:236-241.[Medline] [Order article via Infotrieve]
    
19. Johnston KW. Multicenter prospective study of nonruptured abdominal aortic aneurysm, part II: variables predicting morbidity and mortality. J Vasc Surg. 1989;9:437-447.[Medline] [Order article via Infotrieve]
    
20. Lauer MS, Eagle KA, Buckley MJ, DeSanctis RW. Atrial fibrillation following coronary artery bypass surgery. Prog Cardiovasc Dis. 1989;5:367-378.
    
21. White HD, Antman EM, Glynn MA, Collins JJ, Cohn LH, Shemin RJ, Friedman PL. Efficacy and safety of timolol for prevention of supraventricular tachyarrythmias after coronary artery bypass surgery. Circulation. 1984;70:479-484.[Abstract/Free Full Text]
    
22. Matangi MF, Strickland J, Garbe GJ, Habib N, Basu AK, Burgess JJ, Maitland A, Busse EFG. Atenolol for the prevention of arrhythmias following coronary artery bypass grafting. Can J Cardiol. 1989;5:229-234.[Medline] [Order article via Infotrieve]
    
23. Michelson EL, Morganroth J, MacVaugh H. Postoperative arrhythmias after coronary artery and cardiac valvular surgery detected by long-term electrocardiographic monitoring. Am Heart J.. 1979;97:442-448.[Medline] [Order article via Infotrieve]
    
24. Aranki SF, Cohn LH. Coronary artery bypass grafting in the elderly. J Myocard Ischemia. 1994;6:15-20.
    
25. Satinsky JD, Collins JJ Jr, Dalen JE. Conduction defects after cardiac surgery. Circulation. 1974;49(suppl II):II-170. Abstract.
    
26. Repique LJ, Shah SN, Marais GE. Atrial fibrillation 1992: management strategies in flux. Chest. 1992;101:1095-1103.[Free Full Text]
    
27. Davies MJ, Pomerance A. Pathology of atrial fibrillation in man. Br Heart J. 1972;34:520-525.[Free Full Text]
    
28. Toyoshima H, Park YD, Ishikawa Y, Nagata S, Hirata Y, Sakakibara H, Shimon W, Nakayam R. Effect of ventricular hypertrophy on conduction velocity of activation front in the ventricular myocardium. Am J Cardiol. 1982;49:1938-1945.[Medline] [Order article via Infotrieve]
    
29. McLenachan JM, Dargie HJ. Ventricular arrhythmias in hypertensive left ventricular hypertrophy. Am J Hypertens. 1990;3:735-740.[Medline] [Order article via Infotrieve]
    
30. Peters NS, Green CR, Poole-Wilson PA, Severs NJ. Reduced content of connexin-43 gap junctions in ventricular myocardium from hypertrophy and ischemic human hearts. Circulation. 1993;88:864-875.[Abstract/Free Full Text]
    
31. Topol EJ, Traill TA, Fortuin NJ. Hypertensive hypertrophic cardiomyopathy of the elderly. N Engl J Med. 1985;312:277-283.[Abstract]
    
32. Tchervenkov CI, Wynands JE, Symas JF, Malcolm ID, Dobell AR, Morin JE. Electrical behavior of the heart following high-potassium cardioplegia. Ann Thorac Surg.. 1983;36:314-319.[Abstract]
    
33. Petersen P, Godtfredsen J. Atrial fibrillation: a review of course and prognosis. Acta Med Scand. 1984;216:5-9.[Medline] [Order article via Infotrieve]
    
34. Kleiger RE, Senior RM. Longterm electrocardiographic monitoring of ambulatory patients with chronic airway obstruction. Chest. 1974;65:483-487.[Abstract/Free Full Text]
    
35. Incalzi RA, Pistelli R, Fuso L, Cocchi A, Bonetti MG, Giordano A. Cardiac arrhythmias and left ventricular function in respiratory failure from chronic obstructive pulmonary disease. Chest. 1990;97:1092-1097.[Abstract/Free Full Text]
    
36. Sato S, Yamaguchi S, Schuessler RB, Boineau JP, Matsunaga Y, Cox JL. The effect of augmented atrial hypothermia on atrial refractory period, conduction, and atrial flutter/fibrillation in the canine heart. J Thorac Cardiovasc Surg. 1992;104:297-306.[Abstract]
    
37. Smith PK, Buhrman WC, Levett JM, Ferguson TB Jr, Holman WL, Cox JL. Supraventricular conduction abnormalities following cardiac operations: a complication of inadequate atrial preservation. J Thorac Cardiovasc Surg. 1983;85:105-115.[Medline] [Order article via Infotrieve]
    
38. Ormerund OJM, McGregor CGA, Stone DL, Wisbey C, Petch MC. Arrhythmias after coronary bypass surgery. Br Heart J. 1984;51:618-621.[Abstract/Free Full Text]
    
39. Caretta Q, Mercanti CA, DeNardo D, Chiarotti F. Ventricular conduction defects and atrial fibrillation after coronary artery bypass grafting: multivariate analysis of preoperative, intra-operative and postoperative variables. Eur Heart J. 1991;12:1107-1111.[Abstract/Free Full Text]
    
40. Akins CW. Noncardioplegic myocardial preservation for coronary revascularization. J Thorac Cardiovasc Surg. 1984;88:174-181.[Abstract]
    
41. Butler J, Chong JL, Rocker GM, Pillai R, Westaby S. Atrial fibrillation after coronary artery bypass grafting: a comparison of cardioplegia versus intermittent aortic cross-clamping. Eur J Cardiothorac Surg. 1993;7:23-25.[Abstract]
    
42. Steinberg JS, Zelenkofski S, Wong SC, Gelernt M, Sciacca R, Menchavez E. Value of the P-wave signal-averaged ECG for predicting atrial fibrillation after cardiac surgery. Circulation. 1993;88:2618-2622.[Abstract/Free Full Text]
    
43. Guidera SA, Steinberg JS. The signal-averaged P wave duration: a rapid and noninvasive marker of risk of atrial fibrillation. Am J Cardiol. 1993;21:1645-1651.
    
44. Capucci A, Frabetti L, Turinetto B, Pierangeli A, Magnani B. Fibrillazione atriale nei post operati de by-pass aortocoronarica. G Ital Cardiol. 1987;17:575-582.[Medline] [Order article via Infotrieve]
    
45. Lowe JE, Hendry PJ, Hendrickson SC, Wells R. Intraoperative identification of cardiac patients at risk to develop postoperative atrial fibrillation. Ann Surg. 1991;213:338-392.
    
46. American Heart Association. 1991 Heart and Stroke Facts. Dallas, Tex: American Heart Association; 1992:2-19.11.
    
47. Taylor GJ, Mikell FL, Moses W, Dove JT, Katholi RE, Malik SA, Markwell SJ, Korsmeyer C, Schneider JA, Wellons HA. Determinants of hospital charges for coronary artery bypass surgery: the economic consequences of postoperative complications. Am J Cardiol. 1990;65:309-313.[Medline] [Order article via Infotrieve]
    
48. Landymore RW, Howell F. Recurrent atrial arrhythmias following treatment for postoperative atrial fibrillation after coronary bypass operations. Eur J Cardiothorac Surg. 1991;5:436-439.[Abstract]
    

This article has been cited by other articles:

				O. Adam, H.-R. Neuberger, M. Bohm, and U. Laufs Prevention of Atrial Fibrillation With 3-Hydroxy-3-Methylglutaryl Coenzyme A Reductase Inhibitors Circulation, September 16, 2008; 118(12): 1285 - 1293. [Full Text] [PDF]

				R. Cavolli, K. Kaya, A. Aslan, O. Emiroglu, S. Erturk, O. Korkmaz, M. Oguz, R. Tasoz, and U. Ozyurda Does Sodium Nitroprusside Decrease the Incidence of Atrial Fibrillation After Myocardial Revascularization?: A Pilot Study Circulation, July 29, 2008; 118(5): 476 - 481. [Abstract] [Full Text] [PDF]

				D. Annane, V. Sebille, D. Duboc, J.-Y. Le Heuzey, N. Sadoul, E. Bouvier, and E. Bellissant Incidence and Prognosis of Sustained Arrhythmias in Critically Ill Patients Am. J. Respir. Crit. Care Med., July 1, 2008; 178(1): 20 - 25. [Abstract] [Full Text] [PDF]

				D. E. Singer, G. W. Albers, J. E. Dalen, M. C. Fang, A. S. Go, J. L. Halperin, G. Y. H. Lip, and W. J. Manning Antithrombotic Therapy in Atrial Fibrillation: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition) Chest, June 1, 2008; 133(6_suppl): 546S - 592S. [Abstract] [Full Text] [PDF]

				A. Kourliouros, A. De Souza, N. Roberts, A. Marciniak, A. Tsiouris, O. Valencia, J. Camm, and M. Jahangiri Dose-related effect of statins on atrial fibrillation after cardiac surgery. Ann. Thorac. Surg., May 1, 2008; 85(5): 1515 - 1520. [Abstract] [Full Text] [PDF]

				M. A.J.M. Huybregts, R. de Vroege, and W. van Oeveren A New System for Right Atrial Cooling Ann. Thorac. Surg., April 1, 2008; 85(4): 1421 - 1424. [Abstract] [Full Text] [PDF]

				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]

				Y. M. Kim, H. Kattach, C. Ratnatunga, R. Pillai, K. M. Channon, and B. Casadei Association of atrial nicotinamide adenine dinucleotide phosphate oxidase activity with the development of atrial fibrillation after cardiac surgery. J. Am. Coll. Cardiol., January 1, 2008; 51(1): 68 - 74. [Abstract] [Full Text] [PDF]

				L. R. Zebis, T. D. Christensen, I. S. Kristiansen, and V. E. Hjortdal Amiodarone Cost Effectiveness in Preventing Atrial Fibrillation After Coronary Artery Bypass Graft Surgery Ann. Thorac. Surg., January 1, 2008; 85(1): 28 - 32. [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]

				K. Iino, N. Yui, T. Ooya, R. Kawabata, S. Tomita, and G. Watanabe Successful low-energy cardioversion using a novel biodegradable gel pad: Feasibility of treating postoperative atrial fibrillation in animals. J. Thorac. Cardiovasc. Surg., December 1, 2007; 134(6): 1519 - 1525. [Abstract] [Full Text] [PDF]

F. Roshanali, M. H. Mandegar, M. A. Yousefnia, H. Rayatzadeh, F. Alaeddini, and F. Amouzadeh Prediction of Atrial Fibrillation via Atrial Electromechanical Interval After Coronary Artery Bypass Grafting Circulation, October 30, 2007; 116(18): 2012 - 2017. [Abstract] [Full Text]