Atrial fibrillation (AF) is common after coronary artery bypass grafting (CABG) and increases the morbidity and cost. Amiodarone reduces AF after CABG. Ranolazine, an antianginal agent, also prolongs atrial refractoriness and inhibits after depolarizations and triggered activity; effects that could decrease AF after CABG. The present study compared amiodarone versus ranolazine for the prevention of AF after CABG. A retrospective cohort study of patients undergoing CABG at Aspirus Hospital from June 2008 to April 2010. The patients received either amiodarone (400 mg preoperatively followed by 200 mg twice daily for 10 to 14 days) or ranolazine (1,500 mg preoperatively followed by 1,000 mg twice daily for 10 to 14 days). The primary end point was any identified AF after CABG. A total of 393 consecutive patients undergoing CABG (mean age 65 ± 10 years, 72% men) received either amiodarone (n = 211 [53.7%]) or ranolazine (n = 182 [46.3%]). AF occurred in 26.5% of the amiodarone-treated patients compared to 17.5% of the ranolazine-treated patient (p = 0.035). The univariate predictors of AF included amiodarone use, age, chronic lung disease, and congestive heart failure. The multivariate predictors of AF included amiodarone use (odds ratio 1.7, 95% confidence interval 1.01 to 2.91, p = 0.045 vs ranolazine), age (odds ratio 2.2 per 10 years, 95% confidence interval 1.63 to 2.95, p <0.001), and chronic lung disease (odds ratio 1.86, 95% confidence interval 1.00 to 3.43, p = 0.049). No difference was found in the risk of adverse events between the 2 therapies. In conclusion, ranolazine was independently associated with a significant reduction of AF compared to amiodarone after CABG, with no difference in the incidence of adverse events. Randomized studies should be conducted to confirm these results.
Atrial fibrillation (AF) is common after coronary artery bypass grafting (CABG) and increases the morbidly and cost. Amiodarone is frequently used to decrease AF after CABG. Ranolazine is an antianginal agent that inhibits the abnormal late sodium channel current in atrial and ventricle tissue, as well as peak sodium channel current in the atrium. By this inhibition, it affects intracellular calcium handling producing an energy-sparing effect. This attenuation of Na + /Ca 2+ overload has made ranolazine a potent inhibitor of after depolarizations and triggered activity produced by a number of mechanisms in the experimental setting, including those that might be operative in AF. Additionally, ranolazine induces significant postrepolarization refractoriness in atrial tissue, making it more difficult for AF to sustain itself. Ranolazine was associated with a reduction in the number of new episodes of AF in patients with an acute coronary syndrome, in maintaining sinus rhythm in a group of patient with resistance AF, and is effective for conversion of AF to sinus rhythm. Thus, ranolazine could prove useful in the treatment of AF in general and AF after CABG in particular.
Methods
A single-center nonrandomized retrospective cohort study was used to assess the relative benefit of prophylactic amiodarone versus ranolazine for the prevention of AF after CABG. All patients without a history of permanent AF undergoing CABG without concomitant valve surgery at Aspirus Wausau Hospital from June 2008 to April 2010 were included in the present study. All patients were divided into 2 groups, amiodarone or ranolazine, depending on which agent was used for AF prophylaxis. At least 2 doses of either agent were used to define the amiodarone or ranolazine assignment. For both drugs, the dosage and timing of drug initiation was at the discretion of the treating physician. However, amiodarone was usually started at a dose of 400 mg/day for 7 days before elective CABG. For urgent cases, 400 mg of oral amiodarone was begun immediately when the decision for surgery was made and continued at a dose of 200 mg twice daily through the postoperative period for 10 to 14 days. Ranolazine was generally started at 1,500 mg the day before elective surgery or, in very urgent cases, the day of surgery. It was than continued at 1,000 mg twice daily for 10 to 14 days postoperatively.
The patient data were abstracted by trained clinical abstractors who were unaware of the study hypothesis and submitted to the Society of Thoracic Surgery (STS) database. The STS database was used to obtain the baseline clinical characteristics, operative procedure, and postoperative complications of each patient as a function of the group assignment. In addition, the electronic records were reviewed for accuracy of the documented presence or absence of AF. The vast majority of episodes were correctly abstracted in the STS database; however, on 2 occasions, an episode of documented AF was identified by chart review that had not been noted in the STS database.
Each patient underwent continuous electrocardiographic monitoring throughout their hospital stay. The monitors were followed by trained technicians, and each monitor stored the data for later recall. The presence of AF was defined as evident AF, regardless of the duration, that prompted a chart notation of AF by anyone involved in the care of the patient.
The results are expressed as the mean ± SD, unless otherwise noted. Discrete variables were compared across patient groups using the chi-square, Fisher, or Spearman test, as appropriate. Continuous variables were compared using the Student t test or Mann-Whitney U tests, as appropriate.
Univariate associations between patient variables and AF were modeled using logistic regression analysis. The independent association between the drug (amiodarone vs ranolazine) and the outcome was assessed in a multiple logistic regression model. Candidate patient demographic and clinical characteristics were selected for inclusion in the multivariate model according to clinical and statistical criteria. The independent effect of the drug was adjusted for the effects of age, gender, preoperative AF, heart failure, left ventricular ejection fraction, history of myocardial infarction, previous CABG, chronic lung disease, and β blocker and angiotensin-converting enzyme inhibitor or angiotensin receptor blocker use. Interactions among the variables in the multivariate model were not estimated. A 2-tailed p value <0.05 was considered statistically significant. All analysis was performed using the Statistical Package for Social Sciences, version 16.0 (SPSS, Chicago, Illinois).
Results
A total of 393 consecutive patients were included in the present analysis. Of these patients, 211 (53.7%) were assigned to amiodarone and 182 (46.3%) to ranolazine. The baseline clinical characteristics of the patient population are listed in Table 1 . The 2 groups were comparable with respect to age, gender, history of previous AF or atrial flutter, chronic lung disease, tobacco use, diabetes, hypertension, previous myocardial infarction, or previous CABG. However, the amiodarone patients had a slightly lower ejection fraction and greater incidence of class IV heart failure. Also, a nonsignificant trend was seen for a greater incidence of previous myocardial infarction and history of any heart failure in the amiodarone group.
| Characteristic | Amiodarone | Ranolazine | p Value |
|---|---|---|---|
| Patients (n) | 211 | 182 | NA |
| Men | 77% | 70% | 0.12 |
| Age (years) | 64.9 ± 10.9 | 66.7 ± 9.3 | 0.08 |
| Ejection fraction | 54.7 ± 12.7% | 57.7 ± 9.8% | 0.01 |
| History of heart failure | 13.7% | 7.9% | 0.07 |
| Class IV heart failure | 8.5% | 2.8% | 0.02 |
| Previous myocardial infarction | 41.7% | 32.6% | 0.07 |
| Any previous atrial fibrillation/atrial flutter | 7.6% | 4.5% | 0.21 |
| Chronic lung disease | 20.4% | 19.1% | 0.73 |
| Tobacco use | 24.2% | 19.8% | 0.29 |
| Hypertension | 86.3% | 87.1% | 0.66 |
| Previous coronary artery bypass grafting | 5.7% | 5.1% | 0.78 |
| Diabetes | 36.0% | 39.3% | 0.50 |
The details of the operative procedure and postoperative care are given in Table 2 . The amiodarone group and ranolazine group underwent similar operations and received similar postoperative care. No difference was present in the postoperative length of stay between the 2 groups.
| Characteristic | Amiodarone | Ranolazine | p Value |
|---|---|---|---|
| Elective operation | 53.1% | 55.1% | 0.70 |
| Distal arterial anastomoses | 0.91 ± 0.2 | 0.90 ± 0.2 | 0.63 |
| Distal venous anastomoses | 2.4 ± 0.81 | 2.51 ± 0.80 | 0.14 |
| Intubation duration (hours) | 9.1 ± 11.1 | 8.5 ± 6.9 | 0.53 |
| Intra-aortic balloon pump used | 12.3% | 7.8% | 0.15 |
| Without use of cardiopulmonary bypass pump | 90.0% | 91.0% | 0.75 |
| Postoperative blood products | 39.3% | 40.4% | 0.75 |
| Discharged with β blocker | 94.3% | 97.8% | 0.10 |
| Discharged with angiotensin inhibitor | 41.7% | 36.5% | 0.30 |
| Discharged with aspirin | 98.1% | 91.6% | 0.24 |
| Length of stay—surgery to discharge (days) | 4.38 ± 5.5 | 4.12 ± 6.1 | 0.66 |
The postoperative complications are listed in Table 3 . AF occurred in 26.5% of the amiodarone-treated patients and 17.5% of the ranolazine-treated patients (p = 0.035). The univariate predictors of AF included amiodarone use, advanced age, chronic lung disease, and congestive heart failure. The multivariate predictors of AF include amiodarone use (odds ratio 1.7, 95% confidence interval 1.01 to 2.91; p = 0.045 vs ranolazine), age (odds ratio 2.2 per 10 years, 95% confidence interval 1.63 to 2.95, p <0.001), and chronic lung disease (odds ratio 1.86, 95% confidence interval 1.00 to 3.43, p = 0.049). No significant difference was found in the risk of adverse events between the 2 therapies, including mortality and 30-day readmissions. Two patients in each group were readmitted within 30 days with AF. Only 1 of these patients was still receiving prophylaxis (amiodarone) when the recurrence developed. One patient in the amiodarone group was admitted with third-degree heart block and required a pacemaker.
