If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Requests for reprints should be addressed to Renato D. Lopes, MD, PhD, MHS, Duke Clinical Research Institute, Box 3850, 2400 Pratt Street, Room 0311, Terrace Level, Durham, NC 27705
Division of Invasive Clinical Electrophysiology, Department of Cardiology, Universidade Federal de São Paulo, BrazilBrazilian Clinical Research Institute, São Paulo, Brazil
Division of Invasive Clinical Electrophysiology, Department of Cardiology, Universidade Federal de São Paulo, BrazilBrazilian Clinical Research Institute, São Paulo, Brazil
We sought to assess the effect of naproxen versus placebo on prevention of atrial fibrillation after coronary artery bypass graft (CABG) surgery.
Methods
In this randomized, double-blind, placebo-controlled, single-center trial of 161 consecutive patients undergoing CABG surgery, patients received naproxen 275 mg every 12 hours or placebo at the same dosage and interval over 120 hours immediately after CABG surgery. The primary outcome was the occurrence of atrial fibrillation in the first 5 postoperative days.
Results
The incidence of postoperative atrial fibrillation was 15.2% (12/79) in the placebo versus 7.3% (6/82) in the naproxen group (P=.11). The duration of atrial fibrillation episodes was significantly lower in the naproxen (0.35 hours) versus placebo group (3.74 hours; P=.04). There was no difference in the overall days of hospitalization between placebo (17.23±7.39) and naproxen (18.33±9.59) groups (P=.44). Intensive care unit length of stay was 4.0±4.57 days in the placebo and 3.23±1.25 days in the naproxen group (P=.16). The trial was stopped by the data monitoring committee before reaching the initial target number of 200 patients because of an increase in renal failure in the naproxen group (7.3% vs 1.3%; P=.06).
Conclusions
Postoperative use of naproxen did not reduce the incidence of atrial fibrillation but decreased its duration, in a limited sample of patients after CABG surgery. There was a significant increase in acute renal failure in patients receiving naproxen 275 mg twice daily. Our study does not support the routine use of naproxen after CABG surgery for the prevention of atrial fibrillation.
Postoperative atrial fibrillation is a common complication following cardiac surgery, occurring in 25% to 40% of the patients undergoing coronary artery bypass graft (CABG) surgery
In this setting, atrial fibrillation is associated with increased hospital costs, prolonged hospitalizations, and intensive care readmission, and can cause stroke, congestive heart failure, and hemodynamic instability.
Postoperative use of naproxen does not reduce the incidence of atrial fibrillation but decreases its duration.
•
Naproxen was associated with a significant increase in the incidence of new-onset postoperative renal failure.
•
The routine use of naproxen after coronary artery bypass graft surgery for the prevention of atrial fibrillation should be avoided.
Several clinical factors such as age, sex, prior history of atrial fibrillation, hypertension, obesity, chronic obstructive pulmonary disease, left atrial size, left ventricular ejection fraction, and postoperative withdrawal of beta-blocker or an angiotensin-converting enzyme (ACE) inhibitor have been shown to be associated with an increased incidence of postoperative atrial fibrillation.
Conversely, the use of beta-blockers, ACE inhibitors, statins, amiodarone, and hydrocortisone have been associated with a reduction in the incidence of atrial fibrillation in patients undergoing CABG surgery.
Randomized trial of atorvastatin for reduction of postoperative atrial fibrillation in patients undergoing cardiac surgery: results of the ARMYDA-3 study.
Several clinical studies have investigated the relationship between inflammation and atrial fibrillation. These studies have shown that vascular markers of inflammation such as high-sensitivity C-reactive protein and interleukin-6 are higher in patients with atrial fibrillation compared with those without it, supporting the inflammatory hypothesis in the generation of postoperative atrial fibrillation.
Given the possible inflammatory mechanism of its genesis, anti-inflammatory drugs such as nonsteroidal agents may have a role in reducing the incidence of postoperative atrial fibrillation.
We performed a prospective, double-blind, randomized trial to compare the effects of naproxen (a nonsteroidal anti-inflammatory drug) with placebo on the incidence of postoperative atrial fibrillation in patients undergoing CABG surgery.
Methods
Study Design
The study protocol was approved by the Pontifícia Universidade Católica do Rio Grande do Sul ethics committee. All consecutive patients who were scheduled for CABG surgery between January 2005 and January 2009 were evaluated for inclusion in the study.
Immediately after surgery, patients were randomized and admitted to the intensive care unit. The 2 groups were block randomized with block sizes of 50. The pharmacist made the randomization list and allocated the placebo and naproxen pills without the knowledge of any other person. The blocks of 50 drug containers were delivered to the pharmacy for the intensive care unit and each container was labeled with a number. The containers were identical in size, shape, color, consistency, and packaging, and contained pills that appeared identical. The numbers of the 50 drug containers were recorded and sealed in envelopes by a nurse not involved in the protocol. Patients and investigators were blinded to group assignments. During enrollment, staff nurses not involved in the investigation opened the envelopes and administered the pills (naproxen or placebo; 275 mg every 12 hours) from the respective container for 5 days postoperatively. The anti-inflammatory agent naproxen was chosen to evaluate its efficacy in reducing postoperative atrial fibrillation, which is thought to be related to inflammation caused by cardiopulmonary bypass as well as manipulation of cardiac structures and pericardium by surgeon(s) during CABG surgery.
Study Population
Patients were eligible if they were 18 years or older, able to provide written informed consent, were in sinus rhythm before the surgical procedure, and were scheduled to undergo isolated on-pump CABG surgery.
We excluded patients who had an allergy to naproxen or other nonsteroidal anti-inflammatory drug; were enrolled in another clinical trial; were pregnant; had off-pump surgery; had a history of gastrointestinal bleeding; had chronic liver disease, renal insufficiency (serum creatinine concentration >132.6 μmol/L), or thrombocytopenia (platelets below 50×109/L); reported preoperative use of glucocorticoids; or had a previous diagnosis of atrial fibrillation. Patients with valve disease have been shown to have a higher incidence of postprocedural atrial fibrillation compared with those undergoing CABG surgery. In addition to inflammation caused by cardiopulmonary bypass, other mechanisms such as elevated left atrial pressure, larger left atrial size, and direct mechanical manipulation of the atria also contributed significantly to this higher propensity of atrial fibrillation. As such, we excluded patients undergoing valve replacement to keep the population more homogeneous.
Coronary Artery Bypass Graft Surgery
All patients underwent on-pump CABG surgery by standard surgical technique. The surgery was performed through midline sternotomy incision. Cannulation of the right atria and aorta was performed in all patients, and systemic hypothermia was maintained at 30-32°C. The cardioplegia solution consisted of cold crystalloid.
After surgery, patients were admitted to the intensive care unit, and the pills were administered by a nasogastric tube while on mechanical ventilation and an oral route after weaning off mechanical ventilation. Patients with at least 1 risk factor for gastrointestinal damage (eg, age >60 years, previous history of peptic ulcer, or anticoagulant use) received omeprazole 20 mg/day.
All patients were continuously monitored with a bedside electrocardiographic monitor and daily 12-lead electrocardiographic recordings while in the postoperative intensive care unit. After discharge from the intensive care unit, all patients received a 12-lead electrocardiogram daily and at any time they exhibited signs and symptoms defined as palpitations, thoracic pain, hypotension, or pulmonary congestion.
Study Outcomes
The primary outcome was the incidence of atrial fibrillation after CABG during the first 5 postoperative days. Atrial fibrillation was defined as an irregular rhythm with a lack of discernible P waves on the electrocardiogram. Atrial fibrillation episodes lasting longer than 3 minutes, or less if accompanied by symptoms of hemodynamic instability, were identified. The secondary outcomes were length of stay in the intensive care unit, overall length of hospital stay, number of atrial fibrillation episodes, and mean duration of atrial fibrillation episodes. We also assessed other outcomes such as myocardial infarction, pericarditis, hypokalemia, hypomagnesemia, gastrointestinal bleeding, angioedema, agranulocytosis, jaundice, renal failure, mediastinitis, wound infection, stroke, and in-hospital mortality. All of these events were identified by the investigators.
Data Monitoring Committee
The safety outcomes established by the data monitoring committee during the development of the study protocol were the occurrence of statistically significant upper gastrointestinal hemorrhage (defined as hemorrhage that emanates proximal to the ligament of Treitz); agranulocytosis (defined as neutrophil levels <100/μL); renal failure (defined as ≥50% serum creatinine elevation from baseline); mediastinitis; acute myocardial infarction (defined as troponin elevation >5×the upper limit of normal, or creatine kinase-MB associated with new pathological Q-waves, or new left bundle branch block, or imaging evidence of new loss of viable myocardium); stroke (defined as a sudden, nontraumatic, focal neurological deficit lasting more than 24 hours); and in-hospital mortality in the naproxen group. Renal failure was determined as a criteria proposed by the Second International Consensus Conference of the Acute Dialysis Quality Initiative Group and ratified by the Acute Kidney Injury Network. The committee analyzed the safety outcomes for each group of 50 patients randomized in an unblinded fashion.
Statistical Analysis
The data were analyzed on an intention-to-treat principle. The sample size was calculated on the assumption of a 10% incidence of postoperative atrial fibrillation in the naproxen group and 25% in the placebo group to detect a 15% absolute reduction in both groups with a power of 80% and a 2-sided alpha of 0.05. Using these parameters, it was necessary to evaluate 200 patients for the primary outcome. Continuous variables with normally distributed values were compared with Student's t test. Categorical variables were compared with Fisher's exact test. The continuous variables are expressed as means±SD, and the categorical values are expressed as percentages. A Kaplan-Meier curve with log-rank test was used to compare the ratio of patients without atrial fibrillation in both groups and the proportion of no renal failure during 5 postoperative days.
Results
Study Population
The flowchart of the 422 screened patients is shown in Figure 1. Of these, 261 were excluded, and 161 were randomized. The study was stopped early by the data monitoring committee due to an increase in the rate of renal failure in patients assigned to receive naproxen (7.3%) when compared with placebo (1.3%).
Figure 1Flowchart of patients in the Naproxen as Prophylaxis against Atrial Fibrillation after Coronary Artery Bypass Graft Surgery (NAFARM) trial.
The baseline characteristics of the study population are shown in Table 1. The groups were well matched, although patients randomized to naproxen tended to more frequently have a history of myocardial infarction. The use of baseline medications was similar between groups. Table 2 shows the perioperative characteristics. Patients randomized to placebo more often used inotropics when compared with patients in the naproxen arm (73.4% vs 58.5%; P=.04). Patients randomly assigned to receive naproxen had a trend for more intraaortic balloon pump use than those receiving placebo, but this difference was not significant. There were no statistically significant differences between the groups with respect to postoperative use of beta-blockers, ACE inhibitors, or anti-arrhythmic agents. Importantly, all patients in both groups received aspirin pre- and postoperatively. In this study, no cardioversion procedure for atrial fibrillation was performed within the first 5 postoperative days (Table 2).
Table 1Preoperative Characteristics
Characteristics
Placebo (n=79)
Naproxen (n=82)
P Value
Age, mean (SD), years
58.0 (8.6)
59.7 (9.8)
.23
Male
52 (65.8)
50 (61.0)
.40
LVEF, mean (SD)
50.8 (13.5)
48.0 (11.7)
.15
History of heart failure
45 (57.0)
43 (52.4)
.56
Diabetes mellitus
22 (27.8)
27 (32.9)
.48
Systemic hypertension
66 (83.5)
67 (81.7)
.75
Chronic obstructive pulmonary disease
3 (3.8)
6 (7.3)
.33
Current tobacco use
18 (22.8)
20 (24.4)
.81
Previous MI
29 (36.7)
41 (50)
.08
Previous CABG surgery
2 (2.5)
1 (1.2)
.53
Medications
Aspirin
79 (100.0)
82 (100.0)
1.00
Beta-blockers
72 (91.1)
72 (87.8)
.49
ACE inhibitors
60 (75.9)
61 (74.4)
.81
Calcium channel blocker
5 (6.3)
12 (14.6)
.08
Amiodarone
0
1 (1.2)
.32
Digitalis
2 (2.5)
3 (3.7)
.68
HMG-CoA reductase inhibitors
73 (92.4)
78 (95.1)
.47
Values presented as no. (%), unless otherwise indicated.
The primary and secondary outcome data are summarized in Table 3. There were no statistically significant differences between the groups with respect to pericarditis, hypokalemia, hypomagnesemia, upper gastrointestinal hemorrhage, angioneurotic edema, agranulocytosis, jaundice, mediastinitis, stroke, and in-hospital mortality (Table 3). Patients randomly assigned to receive naproxen had a trend towards more postoperative myocardial infarction than those receiving placebo, but this difference was not significant (15.9% vs 7.6%; P=.10).
The incidence of postoperative atrial fibrillation was 15.2% (12/79) in the placebo group versus 7.3% (6/82) in the naproxen group (P=.11) (Figure 2). The majority of the patients (94.4%) developed atrial fibrillation up to the fourth postoperative day, while most of them were still in the intensive care unit. The mean duration of atrial fibrillation episodes was significantly lower in the naproxen group than in the placebo group (P=.04). There were no differences in the overall length of hospitalization (P=.44) or length of stay in the intensive care unit (P=.16) between the placebo and naproxen groups. The incidence of renal failure was 7.3% (6/82) in the naproxen group versus 1.3% (1/79) in the placebo group (Figure 3). The incidence of atrial fibrillation in patients without renal failure was 10.4% (16/154) and 28.6% (2 of 7 patients) in patients with renal failure (P=.20). No patients needed to go back to the operating room because of bleeding complications.
Figure 2Effect of naproxen and placebo on the incidence of postoperative atrial fibrillation after coronary artery bypass graft surgery.
Our study has 2 main findings. First, we showed that naproxen, a nonsteroidal anti-inflammatory drug, is not effective at reducing the incidence of atrial fibrillation, but significantly reduced the duration of this arrhythmia when it did occur in a limited sample of patients undergoing isolated CABG surgery who were well treated with evidence-based medications. Second, the use of naproxen at a dose of 275 mg twice a day caused a significant increase in acute renal failure. These data suggest that the use of naproxen as a prophylaxis for atrial fibrillation in patients undergoing CABG surgery is associated with more harm than benefit and should not be used routinely.
A nonsteroidal anti-inflammatory drug was chosen because of the proposed role of inflammation
The 174G/C interleukin-6 polymorphism influences postoperative interleukin-6 levels and postoperative atrial fibrillation Is atrial fibrillation an inflammatory complication?.
in postoperative atrial fibrillation, and because a prospective observational study with 4657 patients showed that postoperative anti-inflammatory use was associated with a reduced incidence of atrial fibrillation.
Some studies have shown a positive relationship between inflammatory markers (serum levels of high-sensitivity C-reactive protein, interleukin-6, tumor necrosis factor) and the incidence of postoperative atrial fibrillation.
The 174G/C interleukin-6 polymorphism influences postoperative interleukin-6 levels and postoperative atrial fibrillation Is atrial fibrillation an inflammatory complication?.
Furthermore, the 174G/C interleukin-6 promoter variant appears to modulate the inflammatory response and influence the development of postoperative atrial fibrillation.
The 174G/C interleukin-6 polymorphism influences postoperative interleukin-6 levels and postoperative atrial fibrillation Is atrial fibrillation an inflammatory complication?.
The relationship between atrial inflammation and the inhomogeneity of atrial conduction has been demonstrated in an animal study, which showed that anti-inflammatory therapy significantly decreases the inhomogeneity of atrial conduction after atriotomy and shortens the duration of postoperative atrial fibrillation.
Naproxen was selected because it has been associated with fewer side effects than other nonsteroidal anti-inflammatory drugs, it is already used postoperatively following CABG surgery, and no study has shown a correlation between naproxen and adverse cardiac events.
Our study is the first double-blind, prospective, randomized clinical trial to analyze the effects of naproxen on postoperative atrial fibrillation. When compared with placebo, we found that naproxen did not significantly reduce the incidence of postoperative atrial fibrillation in patients optimally treated with evidence-based medication and undergoing isolated CABG surgery. Naproxen reduced the relative risk of postoperative atrial fibrillation by 55%, but this was not statistically significant. Because our study was prematurely stopped and we did not reach our initial sample size, it is possible that the lack of statistical significance in our primary outcome was due to lack of statistical power. The trial intended to enroll 200 patients, but it was stopped early because of the high incidence of postoperative renal failure in the naproxen group. The duration of atrial fibrillation was significantly shorter, by about 3 hours, in patients receiving naproxen. This is significant clinically as the shorter duration in the naproxen patients could be associated with a lower incidence of adverse events. The incidence of atrial fibrillation in both groups was lower than reported previously, possibly because our patients were optimally treated with beta-blockers, ACE inhibitors, and statins preoperatively and most of them remained on these drugs postoperatively. Some meta-analyses of preoperative beta-blocker therapy have shown a relative risk reduction of 65% in the incidence of atrial fibrillation compared with placebo.
Randomized trial of atorvastatin for reduction of postoperative atrial fibrillation in patients undergoing cardiac surgery: results of the ARMYDA-3 study.
An open-label randomized study of CABG surgery with postoperative ibuprofen showed a 9.8% incidence of atrial fibrillation on ibuprofen versus 28.6% on placebo.
Several differences between these trials and ours could explain the lower atrial fibrillation rates we observed. In the ibuprofen study, an episode of atrial fibrillation was considered significant if it persisted for more than 30 seconds; there is no mention of the preoperative use of statins or ACE inhibitors or maintenance on beta-blocker treatment during the postoperative period. A randomized, double-blind study reported an incidence of postoperative atrial fibrillation of 30% in a hydrocortisone group versus 48% in the placebo group, but they also included patients scheduled to undergo aortic valve replacement.
There is no report about the use of statins and ACE inhibitors preoperatively and the use of beta-blockers postoperatively, factors that influence the incidence of postoperative atrial fibrillation.
Randomized trial of atorvastatin for reduction of postoperative atrial fibrillation in patients undergoing cardiac surgery: results of the ARMYDA-3 study.
Importantly, renal failure occurred 6 times more often in the naproxen group than in the placebo arm of this study (7.3% vs 1.3%). This variable was defined as an increase ≥50% in postoperative creatinine, a stricter criterion than that used by Cheruku et al,
who reported no difference in the incidence of renal failure between ibuprofen and placebo groups. In that trial, renal failure was defined as an increase in serum creatinine of 176.8 μmol/L.
We believe that a stricter criterion was necessary because we wanted to verify if naproxen could provide additional benefits when used together with medications with proven efficacy and safety in preventing postoperative atrial fibrillation, including beta-blockers. In our study, the combination of naproxen and ACE inhibitors may have amplified the adverse effect on renal function, leading to worse acute renal outcome in the naproxen group. In addition, it is well known that renal failure following CABG surgery increases short- and long-term mortality; the incidence of postoperative complications such as respiratory infections, sepsis, and gastrointestinal bleeding; and intensive care unit and hospital lengths of stay.
Despite the higher incidence of atrial fibrillation in acute renal failure patients when compared with patients without renal dysfunction, our study was not statistically powered enough to address this issue. Over the years, the prognostic importance of mildly elevated preoperative serum creatinine levels or small increases in postoperative creatinine values became apparent, and mild elevation of the preoperative creatinine level significantly increases the probability of perioperative mortality, low cardiac output, hemodialysis, and prolonged hospital length of stay.
A recent study demonstrated that acute renal failure in patients undergoing CABG surgery was associated with worse in-hospital events and higher long-term mortality, even in patients with complete recovery of renal function.
Thus, we did not want to expose our patients to these increased risks.
Study Limitations
Our study has some limitations. First, the patients were not continuously monitored by telemetry during the hospital stay. However, all relevant clinically symptomatic episodes of atrial fibrillation were recorded, and these episodes usually required some type of treatment. Nonetheless, we might have missed some asymptomatic and transient atrial fibrillation episodes. Second, because this is a single-center study, our results might not be generalizable to other centers. However, our institution is an academic and tertiary hospital with a high volume of procedures; therefore, our findings might apply to similar centers across the world. Third, all patients included in our study underwent on-pump CABG surgery as very few patients at our institution undergo surgery off pump. Therefore, our results might not be applicable for patients undergoing off-pump CABG surgery. Fourth, unfortunately, we did not collect information on red blood cells and other blood product transfusions. Fifth, our results are best applied to patients undergoing isolated CABG surgery on pump. Finally, due to the premature study interruption, the lack of statistical power could have played a role in the absence of significance in our primary outcome; however, the increase in renal impairment (using a very strict criterion) and its well-known clinical implications in this setting are already enough to contraindicate the use of naproxen for prevention of atrial fibrillation in patients undergoing isolated CABG surgery.
Conclusions
Postoperative use of naproxen did not reduce the incidence of atrial fibrillation but decreased its duration, in a limited sample of patients after CABG surgery. In contrast, naproxen was associated with a significant increase in the incidence of new-onset postoperative renal failure. Our findings do not support the routine use of naproxen as a prophylaxis for preventing atrial fibrillation in patients undergoing isolated CABG surgery.
Appendix
Naproxen as Prophylaxis against Atrial Fibrillation after Coronary Artery Bypass Graft Surgery (NAFARM) Study Investigators
Raquel Pessoa de Carvalho, Fabrício M. Velho, Guilherme Guaragna Filho, Angela G. Bertaso, Iloni T. Rotta, from the Division of Cardiology of Pontifícia Universidade Católica do Rio Grande do Sul, Porto Alegre, Brasil.
Betânia Braga Silva from the Division of Arrhytmology, Department of Cardiology, Universidade Federal de São Paulo, São Paulo, Brasil.
Franciele B. Horbach from the Division of Internal Medicine of Hospital Geral, Universidade de Caxias do Sul, Rio Grande do Sul, Brasil.
References
Crystal E.
Connolly J.S.
Sleik K.
et al.
Interventions on prevention of postoperative atrial fibrillation in patients undergoing heart surgery: a meta-analysis.
Randomized trial of atorvastatin for reduction of postoperative atrial fibrillation in patients undergoing cardiac surgery: results of the ARMYDA-3 study.
Data analyzed during the 5 postoperative days.
00644-9&id=gr1.jpg){kind=link}
00644-9&id=gr2.jpg){kind=link}
00644-9&id=gr3.jpg){kind=link}