Incidence and Clinical Consequences of Atrial Fibrillation Within 1 Year of First-Time Isolated Coronary Bypass Surgery
Background— Atrial fibrillation (AF) is the commonest complication during cardiac surgery, however, the long-term prevalence of AF following surgery and its clinical consequences remain unclear.
Patients and Methods— To investigate this, 877 consecutive patients undergoing first time CABG were followed for 1 year. Rhythm disturbances were diagnosed from serial ECGs and documented notes. The arrhythmia was treated medically and/or by cardioversion.
Results— Out of 877, 17 patients (1.9%) died in the hospital and out of the remaining 860 patients 844 (98.1%) had a complete 1-year follow-up. Patients were divided according to their age: Group I (50 to 59 years), Group II (60 to 69 years) and Group III (70 to 79 years). The prevalence of AF in the general population was taken from the Framingham Heart Study. Patients in groups I and II had a higher incidence of AF before the operation than the general population (1.5% versus 0.4% and 3.1% versus 1.6%, respectively, P<0.05) and also higher incidence of AF at the 1-year follow-up (2% versus 0.4% and 4.6% versus 1.6% respectively, P<0.05). The incidence of AF in group III did not differ from the general population before operation, at the 6-week and 1-year follow-ups. As expected most of the patients with preoperative AF remained in AF after 1-year of CABG surgery. Importantly, the incidence of newly developed AF was higher in patients that developed infection and renal dysfunction in the postoperative period. AF did not predict embolic events at any stage of the study.
Conclusion— In conclusion, the incidence of AF for the first year following CABG is higher in patients <70 years but not in those >70 years when compared with the general population. AF was also associated to the occurrence of postoperative infection and renal dysfunction. Patients in this study were closely monitored and received timely appropriate treatment, and this may account for the absence of a relationship between AF and embolic events.
Atrial fibrillation (AF), with an incidence reported to vary from 10% to 50%1, is the commonest complication following coronary artery bypass grafting (CABG) and it is associated with a two-fold increase in cardiovascular mortality2 and morbidity.3 The lengthening of hospital stay and the consequent financial burden caused by the occurrence of perioperative AF has led to an extensive scrutiny of predisposing factors. Advanced age4, anatomical distribution of coronary artery disease5, poor left ventricular function6, prolonged aortic cross clamp time7, abrupt withdrawal of beta blockers8 and renal failure1 have been reported to influence its occurrence.
AF is known to be an important cause of stroke, heart failure and death in the general population;2 however the persistence of AF following CABG and its clinical consequences remain unclear, since most of the previous studies on the subject have focused on the perioperative period and have failed to include long-term follow-ups. Recently, Loubani et al9 studied the clinical consequences of newly developed atrial dysrhythmias at a 6-month follow-up. They found that newly developed atrial dysrhythmias persisted for 6 months in a remarkable proportion of patients (39%) and concluded that they were not associated with adverse embolic effects.
The present study was designed to investigate: (i) whether the occurrence of postoperative AF is short lived and confined to early postoperative period or persistent beyond hospital discharge and for the first year following surgery, (ii) the influence of perioperative factors on its incidence, and (iii) its clinical consequences.
Patients and Methods
A total of 877 consecutive patients undergoing first time isolated CABG in a single institution between December 1999 and December 2000 were included in the study. All the patients operated as elective, urgent or emergency CABG were included. Patients who had undergone heart surgery previously and those with associated valvular disease were excluded. If a degree of valvular dysfunction was suspected before surgery, a perioperative transesophageal echocardiogram was performed. Therefore, only patients with coronary artery disease and without significant valvular disease took part in the study. Baseline demographic and clinical data were available for all the patients. Initial data (eg, early postoperative period and 6-weeks follow-up) were collected from the case notes and the 1-year follow-up was completed by contacting the patients in person (67%) or via their general practioners (33%).
The preoperative medications, including beta blockers, angiotensin converting enzyme inhibitors and calcium antagonist were continued up to the day of surgery, except acetylsalicylic acid, that was discontinued at least one week before surgery. All the patients received standard anesthesia and surgical procedures.
Rhythm disturbances were diagnosed from ECG and documented notes. Following the diagnosis of AF, patients were cardioverted and/or received one or a combination of the following therapies according to physician’s choice: digoxin, amiodarone or beta-blockers that were maintained for at least 6 weeks. At the end of this period and coinciding with the first postoperative visit at the outpatient clinic, the medication was discontinued if the rhythm had reverted to sinus. If still in AF other medications were added or cardioversion was considered.
Postoperative stroke was defined as a clinically evident temporary or permanent new neurologic focal deficit. The neurologic deficit was confirmed by a computed tomography scan of the brain and clinically by a neurologist physician. Renal dysfunction was defined as serum creatinine elevation above 120 mg/L with or without need for hemodialysis. The elevation of creatinine values is confirmed with more than one determination before surgery. All the patients with preoperative high serum creatinine exhibited similar or even greater values in the perioperative period and during the follow-up. Diagnosis of postoperative infection was established if at least one of the following criteria were present: (i) positive blood, urine or surgical wound discharge cultures, (ii) pyrexia >38°C for at least two days beyond the first two postoperative days, associated with white cell count elevation, (iii) radiologic indication of significant lung consolidation, and (iv) clinical signs of significant surgical wound infection (erythema and discharge).
Data were expressed as means ± standard error of the mean. Statistical analysis was performed using the SPSS program in the following manner:
Kolmogorov-Smirnof for testing the distributions of the continuous variables, chi-square analysis for comparison of categorical variables, one-way analysis of variance for comparison of continuous variable in groups, logistic multiple regression analysis for the identification of variables that may influence AF (each predictive variable was also tested in univariate analysis). A value of P<0.05 was taken as significant. When small number of patients, were analyzed, univariate analysis with Chi-square analysis, and Yates’ correction and Mann-Whitney U test were performed.
The demographics, preoperative medications, surgical data, outcome anti-arrhythmic management of the studied population are presented in Tables 1 & 2⇓. The 1-year survival follow-up was 100% complete and 844 survivors (98.1%) provided complete and reliable information regarding arrhythmia and embolic events. There were no late deaths within the 1-year follow-up. Intra-operative stroke was diagnosed in 17 patients (1.9%), early post-operative stroke in 6 (0.7%) and late stroke in the period between 6 weeks and 1 year after the operation in 3 (0.35%).
The overall incidence of AF was 2.2% before the operation and it rose to 15.5%, in the early postoperative period, and declined to 4.2% at the 6-week follow-up and to 3% at the 1-year follow-up. A comparison between perioperative AF in different age groups and the corresponding values in the general population (Framingham Heart Study) are presented in Figure 1. The incidence of AF before surgery was greater only in patients <70 years old but not in patients >70 years old compared with the general population. The incidence of AF in the early postoperative period was, however, greater with increasing age. The prevalence of AF 6 weeks after surgery was greater than in the general population and again greater in older patients. Interestingly, at 1-year follow-up AF remained elevated to a similar extent in the <70 year age groups but it decreased by almost half the rate in the >70 year age group. This represents actual decrease in the AF prevalence, since there were no late deaths during the follow-up period.
Most patients with preoperative AF remained in AF for the 1 year of follow-up, while the incidence of AF in patients with sinus rhythm prior to surgery significantly declined by 6 weeks and 1 year after surgery but still it was higher in patients <70 years than in the general population of the same age (Table 3). This makes the pre-operative AF the most significant predictor of postoperative AF.
The predictors for newly developed AF and its persistence are presented in Table 4. Age, postoperative renal dysfunction and infections were associated with higher incidence of early postoperative AF and they exhibited strong interaction. Thus, patients complicated with either infections or renal failure had a higher early postoperative incidence of AF compared with patients without these complications (21%, 30% and 12%, respectively). The incidence of AF was 2-fold greater when these complications occurred in combination (75%). Interestingly, only those patients with renal dysfunction experienced a greater incidence of persistent AF at 6 weeks and 1 year after surgery (12.5% and 12.5%, respectively).
The peri-operative management of newly developed AF, that varied according to practitioner’s preference, did not influence the persistence of AF (Table 5). Demographics, surgical data and other postoperative complications did not differ between different treatment groups.
Univariate analysis showed that intra-operative stroke was related to old age (P=0.03), while delayed stroke during the one year of follow-up was exclusively related to previous intra-operative stroke (P=0.001). There was no relationship between stroke and pre-existing or concurrent AF at any stage of the follow-up (Table 6).
The present study has shown that the incidence of AF during the first year following CABG is greater than in the general population in patients <70 years, but not in those >70 years and that this is associated to the occurrence of infections and renal dysfunction. The results of this study are supported by our previous findings9 in a smaller population that was followed for a shorter period (6 months). The association between AF and advanced age in the general population is well documented,12 as it is the greatest incidence of AF in elderly patients in the early postoperative period after CABG.13 Our study has confirmed the above findings during the postoperative period, but it is the first to show that beyond the 6-week postoperative period the incidence of AF is unrelated to age. Information available in this study is not enough to explain this finding.
The demonstration that the persistence of AF is associated to the occurrence of infection and renal dysfunction suggests that inflammatory factors may play a role in the generation of this arrhythmia. This thesis is supported by a recent report showing a positive relationship between AF and C-reactive protein, a marker of inflammation.3 The incidence of AF in patients on chronic dialysis has been shown to be 23%, which is close to the 30% incidence of AF in patients developing postoperative renal impairment in the present study.10 Although it has been suggested that the high incidence of AF in patients with renal failure may be ischemic in origin due to the high rate of concomitant coronary artery disease, this may be unlikely in the present study since the revascularization procedures should have reduced the contribution of ischemia to the development of AF. Interestingly, the demonstration that C-reactive protein levels are increased in 30% to 50% of dialysis patients and that it predicts cardiovascular morbidity and mortality11 further support the view that inflammatory factors may be the common cause of persistence of AF following cardiac surgery. Clearly, more studies are required to demonstrate the potential relationship between inflammatory factors and the occurrence of AF.
It is of interest that in spite of the demonstration that AF is linked to stroke in the general population,14 a relationship between persistent AF for the first year following surgery and neurological complications could not be demonstrated in the present study. A possible explanation for this may be that the anticoagulation treatment administrated to patients in our study (aspirin, warfarin) reduced the risk of stroke.
A limitation of the present study is that it was a retrospective investigation and the diagnosis of AF was performed by ECG recordings and by documentation in the patients’ notes and therefore the occurrence of arrhythmia may have been underestimated. However, this shortcoming does not detract from the clinically important findings that the incidence of AF is elevated beyond the perioperative period and for the first year after bypass graft surgery and that this is associated to the presence of inflammation and renal dysfunction. The results imply that this group of patients may require a close surveillance of the heart rhythm and possibly the design of appropriate antiarrhythmic prophylaxis.
Benjamin EJ, Wolf PA, D’Agostino RB et al. Impact of atrial fibrillation on the risk of death: the Framingham Heart Study. Circulation. 1998; 98: 946–952.
Chung MK, Martin DO, Sprecher D, et al. C-Reactive protein elevation in patients with Atrial Arrhythmias Inflammatory mechanisms and persistence of atrial fibrillation. Circulation. 2001; 104: 2886–2891.
Kolvekar S, D’Souza A, Akhtar P, Reek C, Garratt C, Spyt T, Akhatar P. Role of atrial ischaemia in development of atrial fibrillation following coronary artery bypass surgery. Eur J Cardiothorac Surg. 1997; 11: 70–5.
Mathew JP, Parks R, Savino, JS, Friedman AS, et al. Atrial Fibrillation following coronary artery bypass graft surgery: Predictors, outcomes and resource utilization. JAMA. 276; 24: 300–306.
Ali IM, Sanalla AA, Clark V. Beta-blocker effects on postoperative atrial fibrillation. Eur J Cardiothorac Surg. 1997; 11: 1154–7.
Fabbian F, Catalano C, Lambertini D, Tarroni G, Bordin V, et al. Clinical characteristics associated to atrial fibrillation in chronic hemodialysis patients. Clinica Nephrol. 2000; 54: 234–9.
Kannel WB, Wolf PA., Benjamin EJ, Levy D. Prevalence, incidence, prognosis and predisposing conditions for atrial fibrillation: population-based estimates. Am J Cardiol. 1998; 82: 2–9.
Hogue CW, Murphy SF., Schechtman KB, Davila Roman VG. Risk factors for early or delayed stroke after cardiac surgery. Circulation. 1999; 100: 642–647.