We conducted a prospective, double-blind, randomized, multicenter study among 241 patients who underwent their first on-pump coronary artery bypass graft (CABG) surgery, aortic valve replacement (AVR), or combined CABG and AVR surgery to determine whether the use of hydrocortisone prevents atrial fibrillation after cardiac surgery. Results showed that the incidence of postoperative atrial fibrillation was significantly lower in the hydrocortisone group compared with the placebo group.
Approximately 20% to 45% of patients undergoing coronary artery bypass graft (CABG) surgery experience atrial fibrillation, and those undergoing aortic valve replacement (AVR) or combined CABG and AVR surgery have an even greater incidence of atrial fibrillation. Atrial fibrillation results in higher stroke risk, longer hospital stays, and higher costs.1-3
The mechanisms responsible for postoperative atrial fibrillation are not well understood. One proposed mechanism is that extracorporeal circulation is associated with a systemic inflammatory response.4,5 Patients with atrial fibrillation have been shown to have increased levels of inflammatory markers, white blood cells, and C-reactive protein complex.6,7 Administration of corticosteroids has been shown to decrease recurrent atrial fibrillation episodes in nonoperative patients.8 We conducted a prospective, double-blind, randomized, multicenter study among patients undergoing their first on-pump CABG surgery, AVR, or combined CABG and AVR surgery to determine whether the use of hydrocortisone decreases the incidence of atrial fibrillation after cardiac surgery.
Subjects and methods
Details of the study have been previously published.9 A total of 241 consecutive subjects scheduled to undergo their first on-pump CABG surgery, AVR, or combined CABG and AVR surgery were enrolled in the study. Subjects with thrombophlebitis, previous peptic ulcer, a serum creatinine level over 200 µg/mL, herpes simplex keratitis, psychotic mental disorder, Cushing's syndrome, active tuberculosis, systemic mycotic infection, uncontrolled diabetes mellitus, or a previous occurrence of atrial fibrillation or atrial flutter were excluded from the study.
Subjects underwent cardiac surgery using conventional cardiopulmonary bypass. They remained in the intensive care unit postoperatively until they were stable and weaned off the ventilator. After they were transferred to the surgical ward, they received continuous electrocardiographic (ECG) monitoring for at least 84 hours. We documented all atrial fibrillation episodes that lasted more than 5 minutes.
Subjects were randomly assigned to receive either 100 mg of hydrocortisone or placebo. The first dose was given in the evening of the operative day. Then 1 dose was given every 8 hours over the next 3 days. All subjects also received 50 to 150 mg of oral metroprolol (Lopressor; Toprol XL) daily, depending on their heart rate. The occurrence of the first atrial fibrillation episode or completion of the 84-hour study protocol was the primary end point of the study. We also conducted a meta-analysis of all published reports of prospective randomized trials evaluating the use of corticosteroids for the prevention of atrial fibrillation following cardiac surgery.
The mean age of the subjects was 65.3 years (±8.9) years, with no difference between the 2 groups. The majority of subjects were men (76.8%). Subjects were similar with regard to left ventricular ejection fraction, presence of diabetes, chronic obstruction pulmonary disease, unstable angina pectoris, hypertension, 3-vessel disease, preoperative beta blocker use, Canadian Cardiovascular Society Class score, history of stroke or transient ischemic attack, and history of claudication. Subjects in the hydrocortisone group were more likely to be men and were less likely to have hypertension or unstable angina.
Most of the operations in both groups were isolated CABG surgery, and there were no differences between the groups in the number of distal anastomoses. There also were no differences between groups in pump time, gross clamp time, or first postoperative creatine kinase-MB level.
The incidence of postoperative atrial fibrillation was significantly lower in the hydrocortisone group (30.0%) than in the placebo group (47.9%; P = .004). The C-reactive protein levels on the first, second, and third postoperative days were also significantly lower in the hydrocortisone group than in the placebo group. There were no differences between the 2 groups with respect to postoperative infection, mediastinitis, stroke, myocardial infarction, conduction disturbances, or resternotomy caused by bleeding. We found no psychotic disorders in either study group. In a meta-analysis that included 2 other studies along with our study, with a total of 621 patients, corticosteroid therapy was shown to decrease the incidence of atrial fibrillation after cardiac surgery.
The most important finding of our study is that intravenous hydrocortisone decreased the incidence of atrial fibrillation substantially (37%) after cardiac surgery. In addition, hydrocortisone therapy turned out to be feasible, and there were no serious complications associated with intravenous administration. There were also no differences between the study groups with regard to any known preoperative or perioperative risk factor for postoperative atrial fibrillation.
Results of early studies on the effect of glucocorticoids on the incidence of postoperative atrial fibrillation are not entirely comparable with ours.10-12 A beneficial effect of glucocorticoid therapy on the incidence of postoperative atrial fibrillation was reported by Prasongsukarn and colleagues.10 They administered 1 g of methylprednisolone prior to surgery and 4 mg of dexamethasone or placebo every 6 hours for 1 day after surgery (43 subjects in both groups). The incidence of atrial fibrillation following surgery was 21% in the steroid group compared with 51% in the placebo group (P = .003). Subjects undergoing CABG, AVR, and combined CABG and AVR surgery were included in our study, whereas only CABG surgery subjects were included in the study by Prasongsukarn and colleagues.
Another study by Rubens and colleagues investigated the effects of intravenous administration of 1 g of methylprednisolone on inflammation reaction after CABG surgery.11 They reported that the incidence of atrial fibrillation after CABG surgery was lower in subjects treated with methylprednisolone than in control subjects. It is somewhat difficult to draw any conclusions based on the study, however, because atrial fibrillation was not a primary end point, and there was no definition of an atrial fibrillation episode in the report.
A third study by Halvorsen and colleagues showed the opposite results.12 They administered 4 mg dexamethasone after the induction of anesthesia, and a second dose was given on the first postoperative morning. They found no difference in the incidence of atrial fibrillation between the study and placebo groups. The major difference between our study and that of Halvorsen and colleagues is that we used intravenous hydrocortisone for 3 days, whereas they used only 2 doses of dexamethasone. Systemic inflammatory response to cardiopulmonary bypass and to surgical trauma has been suggested as an etiologic factor for postoperative atrial fibrillation. Studies have shown that the concentration of inflammatory cytokines,13 the number of white blood cells,7 and the concentration of complement C-reactive protein complex,6 which are all markers of increased inflammatory reaction, are greater in patients with atrial fibrillation after surgery than in patients who remain in sinus rhythm. Corticosteroids reduce excessive inflammatory reactions and have anti-inflammatory effects.14 It is thus likely that the anti-inflammatory effect of corticosteroids is responsible for the reduced incidence of atrial fibrillation. This is supported by the finding in our study that the C-reactive protein level was markedly lower in the hydrocortisone group than in the placebo group following surgery. This finding concurs with a study by Dernellis and Panaretou, who reported that glucocorticoid therapy decreased C-reactive protein levels and recurrent atrial fibrillation risk in nonoperative patients.8
Possible adverse effects of corticosteroid treatment include gastrointestinal bleeding and increased risk of wound infections.14 The subjects in our study did not experience any adverse effects related to hydrocortisone therapy. No major adverse effects occurred among subjects in the study by Prasongsukarn and colleagues, although subjects taking steroids had more minor complications.10 There were no differences between the 2 groups in our study. The difference in results between our study and that of Prasongsukarn and colleagues may be explained by the use of different corticosteroid medications (methylprednisolone and dexamethasone vs hydrocortisone).
Previous studies have shown that intravenous metoprolol,2 amiodarone (Cordarone, Pacerone), biatrial pacing, and magnesium reduce the incidence of atrial fibrillation after cardiac surgery.15 Additional studies are needed to determine whether the use of hydrocortisone in combination with these preventive measures will further decrease the risk of postoperative atrial fibrillation after cardiac surgery. Studies are also required to evaluate whether hydrocortisone might be beneficial in the prevention of atrial fibrillation after other types of cardiac surgery.
We conducted a study to determine whether intravenous corticosteroid administration after cardiac surgery would prevent atrial fibrillation. Results showed that hydrocortisone was effective in the prevention of atrial fibrillation following cardiac surgery without adverse effects.