Statin Use and Risk of Atrial Fibrillation or Flutter

American Journal of Therapeutics 22, 186–194 (2015) Statin Use and Risk of Atrial Fibrillation or Flutter: A Population-based Case–Control Study Giacomo Veronese, BSc, Jonathan Montomoli, MD,* Morten Schmidt, MD, Erzsebet Horváth-Puhó, MSc, PhD, and Henrik Toft Sørensen, MD, PhD, DMSc The pleiotropic effects of statins have been suggested to prevent atrial fibrillation onset. We conducted a population-based case–control study using medical databases from Northern Denmark (population: 1.8 million) to examine the association between statin use and atrial fibrillation or flutter. We identified 51,374 patients with atrial fibrillation or flutter between 1999 and 2010 and 513,670 matched population controls. We collected data on statin prescriptions redeemed within 90 days (current users) or longer (former users) before the diagnosis date of atrial fibrillation or flutter. We stratified statin users by duration of exposure, determined by the number of days between first and last redeemed prescription before the diagnosis date (,365, 365–1094, and $1095 days). We used conditional logistic regression to compute odds ratios (ORs) and 95% confidence intervals (CIs), controlling for potential confounders. We defined people without previous statin use as never users (reference). A total of 7360 (14.3%) cases and 55,699 (10.8%) controls were current statin users. Among current users (adjusted OR: 0.96, 95% CI, 0.93–0.99), the preventive effect of statins on atrial fibrillation or flutter was related to duration of use: adjusted ORs decreased from 1.35 (95% CI, 1.28– 1.42) for users who were prescribed statins for ,365 days to 0.85 (95% CI, 0.81–0.89) for users who were prescribed statins for $1095 days compared with never users. For former users (adjusted OR: 0.94, 95% CI. 0.90–0.98), the ORs did not change with varying lengths of exposure. In conclusion, long-term statin use may reduce the risk of atrial fibrillation or flutter compared with never use. Keywords: statin, atrial fibrillation, prevention, epidemiology INTRODUCTION Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark. The Department of Clinical Epidemiology, Aarhus University Hospital, receives funding for other studies from companies in the form of research grants to (and administered by) Aarhus University. None of these studies has any relation to the present study. The study was supported by the Clinical Epidemiology Research Foundation, Denmark. These funding sources had no role in the design, conduct, analysis, or reporting of this study. The authors have no conflicts of interest to declare. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions this article on the journal’s Web site (www.americantherapeutics.com). *Address for correspondence: Department of Clinical Epidemiology, Aarhus University Hospital, Olof Palmes Alle 43-45, Aarhus N DK-8200, Denmark. E-mail: [email protected] Hydroxymethylglutaryl-coenzyme A reductase inhibitors, commonly known as statins, represent the cornerstone of blood cholesterol-lowering therapy, and thus they are unequivocally associated with a substantial reduction in cardiovascular morbidity and mortality.1,2 Atrial fibrillation is the most common arrhythmia encountered in clinical practice,3 with a general population prevalence increasing from 0.5% at age 50–59 years to above 10% at age 80–89 years.4 Atrial fibrillation is associated with increased mortality and morbidity,5–7 mainly due to hemodynamic impairments and thromboembolic events.8 Projections based on population-based studies in the United States indicate that the number of patients with atrial fibrillation may triple by 2050.9 Thus, there is a significant need for primary prevention. 1075–2765 Copyright Ó 2013 Wolters Kluwer Health, Inc. All rights reserved. www.americantherapeutics.com Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. 187 Statin Use and Atrial Fibrillation Recent evidence suggests that the pleiotropic effects of statins may become the rationale for statin use in preventing and treating atrial fibrillation.10 Results from a UK case–control study based on the General Practice Research Database suggest that statin use may have a sustained protective effect against atrial fibrillation onset.11 Other studies have investigated the effects of statin use on atrial fibrillation occurrence in subgroups of patients with congestive heart failure, hypertension, acute coronary syndrome, stroke, or paroxysmal atrial fibrillation, and after surgery, permanent pacing, cardioversion, or pulmonary vein ablation. These results were inconsistent and varied according to the patient population examined.12,13 In addition, recent meta-analyses of the preventive effects of statins on atrial fibrillation reported heterogeneous findings.14–19 We therefore conducted a large population-based case–control study to investigate whether statin use was associated with the prevention of atrial fibrillation or flutter. MATERIALS AND METHODS We conducted this population-based case–control study in northern Denmark, within a population of 1.8 million inhabitants (approximately 33% of the total Danish population).20 Denmark has a tax-financed health care system, providing unfettered access to general practitioners and hospitals and partial reimbursement for prescribed drugs, including statins. Each Danish resident is registered in the Danish Civil Registration System at birth or upon immigration and is assigned a unique personal identification number (CPR number), which permits accurate linkage among all Danish administrative and medical registries. Since 1968, the Civil Registration System, which is updated daily, has collected vital statistics on the Danish population, including changes in address, date of emigration, and date of death.21 We searched the Danish National Registry of Patients (DNRP) to identify all patients with a first-time diagnosis of atrial fibrillation or flutter from January 1, 1999, through December 31, 2010. Since 1977, the DNRP has recorded information on all nonpsychiatric hospital admissions and, since 1995, also on outpatient and emergency visits. Diagnoses are coded according to the International Classification of Diseases (ICD, 8th revision until the end of 1993 and 10th revision thereafter).22 Because atrial fibrillation and flutter share the same risk factors and to some degree pathophysiology,23,24 we combined the diagnoses for the 2 diseases into 1 entity.25 We considered the date of the first diagnosis of atrial fibrillation or flutter as the index date for cases. www.americantherapeutics.com We restricted the study to patients who had resided in Denmark for 365 days or more before the index date, to ensure at least 1 year of prescription history. For each case, we selected from the Danish Civil Registration System 10 population controls matched on age, gender, and county of residence.21 Controls were chosen using risk set sampling26 and assigned an index date identical to that of the corresponding case. Since statins are available in Denmark by prescription only, we used the prescription database to identify all prescriptions for statins filled by cases and controls before their index date. Computerized prescription data are complete for the population of Northern Denmark from 1998 onward. The prescription database contains information on the type of drug, coded according to the Anatomical Therapeutic Chemical classification (ATC), fill date, number of packages redeemed, and dosage.20 We defined current and former statin users according to the time elapsed between the last statin prescription and the index date: Current users were persons who redeemed their most recent prescription within 90 days before the index date and former users were persons whose most recent statin prescription was redeemed more than 90 days before the index date. We further categorized current users in 2 groups: new users, defined as current users who filled their first-ever statin prescription 90 days or less before the index date, and long-term users, defined as current users who filled their first statin prescription more than 90 days before the index date. The advantages of the so-called new user design have been discussed previously.27 Never users were defined as persons with no filled prescriptions for statins registered in the prescription database. Data on potential confounding factors were collected from the DNRP and the prescription database. We searched the DNRP starting with 1977 for any diagnosis before the index date of diseases associated with an increased risk of atrial fibrillation or flutter (Table 1).6,28 We searched the prescription database from 1998 onward for any redeemed prescriptions for drugs associated with atrial fibrillation or flutter (Table 1).25,29,30 We also used the DNRP to collect information on all cardioversions performed within 1 year after the index date. (see Appendix, Supplement Digital Content 1, which reports ICD and ATC codes used in the study, available at http://links.lww.com/AJT/A12). Statistical analysis We first tabulated the frequency and proportion of cases with atrial fibrillation or flutter and population controls within categories of demographic characteristics, potential confounding factors, and statin exposure (Table 1). Using conditional logistic regression, American Journal of Therapeutics (2015) 22(3) Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. 188 Veronese et al Table 1. Characteristics of patients with atrial fibrillation or flutter and controls. Northern Denmark, 1999–2010. Characteristics Gender Female Age groups (yr) 0–49 50–59 60–69 70–79 $80 Use of statins Current New Long-term Former Never Comorbidity Alcoholism* Cancer† Cardiovascular disease‡ Hospital diagnosis† Chronic kidney disease† COPD or asthma‡ Diabetes mellitus‡ IBD† Hyperthyroidism‡ Liver disease or chronic pancreatitis† Hyperlipidemia‡ Drugs Cardiovascular drugs§ ACE or A2R inhibitors Acetylsalicylic acid Beta-blockers CCBs Diuretics Nitrates Digoxin Vitamin K antagonists Other antihypertensives Current use of oral glucocorticoids¶ Non-aspirin NSAIDs¶ Cases, n (%), Total: 51,374 Population controls, n (%), Total: 513,670 23,635 (46.0) 236,350 (46.0) 2352 5031 10,313 15,507 18,171 10,288 7360 932 6428 2928 41,086 (4.6) (9.8) (20.1) (30.2) (35.4) (20.0) (14.3) (1.8) (12.5) (5.7) (80.0) 23,520 50,310 103,130 155,070 181,640 78,884 55,699 3221 52,478 23,185 434,786 (4.6) (9.8) (20.1) (30.2) (35.4) (15.4) (10.8) (0.6) (10.2) (4.5) (84.6) 1803 7332 42,306 19,937 1583 13,559 5447 465 2644 525 2885 (3.5) (14.3) (82.3) (38.8) (3.1) (26.4) (10.6) (0.9) (5.1) (1.0) (5.6) 12,240 55,814 312,198 115,317 6475 92,400 38,273 3598 16,563 3363 17,346 (2.4) (10.9) (60.8) (22.4) (1.3) (18.0) (7.5) (0.7) (3.2) (0.7) (3.4) 41,793 18,665 23,889 20,494 15,792 29,312 10,819 8347 8773 1614 3633 5727 (81.4) (36.3) (46.5) (39.9) (30.7) (57.1) (21.1) (16.2) (17.1) (3.1) (7.1) (11.1) 308,466 131,521 161,046 110,772 105,407 206,221 64,150 16,170 14,029 10,641 15,775 43,747 (60.1) (25.6) (31.4) (21.6) (20.5) (40.1) (12.5) (3.1) (2.7) (2.1) (3.1) (8.5) *Alcoholism-related disorders other than those affecting the liver or pancreas. †Any hospital discharge diagnosis recorded in the Danish National Registry of Patients since 1977. ‡Any redeemed prescription recorded in the prescription database since 1998. §Any hospital discharge diagnosis since 1977 or any redeemed prescription since 1998 of associated drugs. ¶Prescription redemption within 60 days before the index date. A2R, angiotensin-2 receptor; ACE, angiotensin-converting enzyme; CCB, calcium channel blockers; COPD, chronic obstructive pulmonary disease; IBD, inflammatory bowel disease; NSAIDs, nonsteroidal anti-inflammatory drugs. American Journal of Therapeutics (2015) 22(3) www.americantherapeutics.com Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. 189 Statin Use and Atrial Fibrillation we computed odds ratios (ORs) with 95% confidence intervals (CIs) associating current statin use, including new and long-term use, and former statin use with the occurrence of atrial fibrillation and flutter. As we used risk set sampling, these ORs provide unbiased estimates of the incidence rate ratios.26 We fitted several logistic regression models to adjust for the potential confounders listed in Table 1. We then repeated the analysis in predefined subgroups of sex, age (#49, 50–59, 60–69, 70–79, and $80 years), presence or absence of chronic kidney disease, cardiovascular disease (overall, acute myocardial infarction, stroke, and heart failure), diabetes, and hyperlipidemia. In this analysis, we disregarded the matching used to collect population controls and performed conventional logistic regression with additional adjustments for the matching factors. To investigate the association between different duration of statin use and atrial fibrillation or flutter, we stratified all current and former statin users by duration of exposure. Duration of use was determined by counting the number of days between the dates of the first and the last prescription redemption before the index date (,365, 365–1094, and $1095 days). To examine a possible dose–response relation, we stratified by intensity of statin use. We defined intensity of use as the number of pills prescribed multiplied by drug dosage (in milligrams) divided by the total duration of use (in days). Intensity was classified as low-intensity (lower quartile: #16.3 mg/d), mediumintensity (interquartile range: 16.4–41.1 mg/d), and high-intensity (upper quartile: $41.2 mg/d). To explore whether the associations depended on statin type, we repeated the analysis for the most frequently used lipophilic and hydrophilic statins in Denmark, which are simvastatin/atorvastatin and pravastatin/rosuvastatin, respectively. As a sensitivity analysis, we repeated the adjusted analyses changing the exposure window for current and former users from 90 days to 60, 120, or 180 days. Finally, as a measure of severity, we performed a subanalysis limited to atrial fibrillation or flutter cases undergoing cardioversion within 1 year after their first arrhythmic episode. We used SAS software (version 9.2; SAS Institute, Cary, NC) for all analyses. RESULTS We identified 51,374 patients with atrial fibrillation or flutter and 513,670 population controls. The median age was 75 years, and 46% were women. A total of 7360 (14.3%) cases and 55,699 (10.8%) controls were current statin users, 2928 (5.7%) cases and 23,185 (4.5%) controls were former users, and 41,086 (80.0%) cases and 434,786 (84.6%) controls were never users. Among cases, 42,306 (82.3%) had been diagnosed previously with cardiovascular disease compared with 312,198 (60.8%) controls. All other comorbidities considered in the analysis also occurred more frequently among cases than controls. Risk of atrial fibrillation or flutter The adjusted OR associating statin use with atrial fibrillation or flutter was 0.96 (95% CI, 0.93–0.99) for current users and 0.94 (95% CI, 0.90–0.98) for former users compared with never users (Table 2). Among current users, the adjusted OR was 2.14 (95% CI, 1.98–2.32) for new users and 0.88 (95% CI, 0.85–0.91) for long-term users. After stratifying the analysis by presence or absence of chronic kidney disease, cardiovascular disease (overall, acute myocardial infarction, stroke, and heart failure), diabetes, and hyperlipidemia, adjusted ORs among new users remained elevated in all subgroups (Figure 1). In contrast, among long-term users, statin use seemed to be more protective in patients with a previous diagnosis of acute myocardial infarction, stroke, heart failure, and diabetes than among patients without these comorbidities (Figure 1). Moreover, the Table 2. Association between statin use and atrial fibrillation or flutter. OR (95% CI) Number of cases, n 5 51,374 Number of controls, n 5 513,670 41,086 7360 932 6428 2928 434,786 55,699 3221 52,478 23,185 Never users Current users New users Long-term users Former users Unadjusted* 1.00 1.48 3.17 1.36 1.39 (reference) (1.44–1.52) (2.94–3.43) (1.32–1.41) (1.33–1.45) Adjusted† 1.00 0.96 2.14 0.88 0.94 (reference) (0.93–0.99) (1.98–2.32) (0.85–0.91) (0.90–0.98) *Matched on age, gender and county of residence. †Adjusted for all covariates listed in Table 1. www.americantherapeutics.com American Journal of Therapeutics (2015) 22(3) Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. 190 Veronese et al FIGURE 1. Association between use of statins and atrial fibrillation or flutter in patients with or without specific comorbidities. Estimates are computed using conventional logistic regression and adjusted for age, sex, and county of residence and for all covariates listed in Table 1. AMI, acute myocardial infarction; CHF, chronic heart failure; CKD, chronic kidney disease; TIA, transient ischemic attack. protective effect among long-term users seemed to be most pronounced in the youngest age groups, with ORs being 0.70 (95% CI, 0.51–0.95) for patients younger than 50 years and 0.88 (95% CI, 0.83–0.94) for those aged 80 years or older (data not shown). Sex did not modify the effect estimates. For current users, the effect of statins on preventing atrial fibrillation or flutter was related to duration of use: Adjusted ORs decreased from 1.35 (95% CI, 1.28– 1.42) for users who were prescribed statins for less than 365 days to 0.85 (95% CI, 0.81–0.89) for users who were prescribed statins for more than 1095 days (Table 3). For former users, ORs did not change with varying length of exposure. Stratifying on intensity of use also did not substantially change the effect estimates for new, long-term, or former users (Table 4). When we repeated the analyses for the most frequently used statins in Denmark, we found no substantial difference in the ORs (data not shown). Altering the exposure window used to define current and former users did not markedly affect our findings either (Table 5). The results of the subanalysis restricted to cases undergoing cardioversion within 1 year after the first episode of atrial fibrillation or flutter and their controls also supported our main findings. Among the 5627 cases (10.9%) treated with cardioversion within 1 year after their first atrial fibrillation or flutter diagnosis, the Table 3. Duration of statin use and risk of atrial fibrillation and flutter. OR (95% CI) Never users Current users ,365 d 365–1094 d $1095 d Former users ,365 d 365–1094 d $1095 d Number of cases, n 5 51,374 Number of controls, n 5 513,670 Unadjusted* Adjusted† 41,086 7360 2014 1918 3428 2928 1106 769 1053 434,786 55,699 10,948 16,699 28,052 23,185 8465 6404 8316 1.00 (reference) 1.00 (reference) 2.02 (1.92–2.12) 1.26 (1.20–1.33) 1.36 (1.30–1.41) 1.35 (1.28–1.42) 0.84 (0.80–0.89) 0.85 (0.81–0.89) 1.41 (1.32–1.51) 1.31 (1.22–1.42) 1.41 (1.32–1.51) 0.97 (0.91–1.04) 0.87 (0.81–0.95) 0.92 (0.85–0.99) Duration of use was defined as the number of days from the date of the first prescription to the date of the last prescription before the index date. *Matched on age, gender, and county. Unadjusted for other covariates. †Adjusted for all covariates listed in Table 1. American Journal of Therapeutics (2015) 22(3) www.americantherapeutics.com Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. 191 Statin Use and Atrial Fibrillation Table 4. Intensity of statin use and risk of atrial fibrillation or flutter. OR (95% CI) Number of cases, n 5 51,374 Number of controls, n 5 513,670 41,086 7360 1659 3605 2096 932 167 207 558 6428 1492 3398 1538 2928 835 1261 832 434,786 55,699 12,867 29,174 13,658 3221 631 825 1765 52,478 12,236 28,349 11,893 23,185 6869 10,614 5702 Never users Current users Low-intensity Medium-intensity High-intensity New users Low-intensity Medium-intensity High-intensity Long-term users Low-intensity Medium-intensity High-intensity Former users Low-intensity Medium-intensity High-intensity Unadjusted* Adjusted† 1.00 (reference) 1.00 (reference) 1.41 (1.33–1.48) 1.38 (1.33–1.43) 1.70 (1.62–1.79) 0.92 (0.87–0.98) 0.88 (0.84–0.92) 1.13 (1.07–1.19) 2.90 (2.43–3.46) 2.72 (2.32–3.19) 3.52 (3.18–3.89) 2.01 (1.67–2.40) 1.84 (1.56–2.16) 2.36 (2.12–2.62) 1.33 (1.25–1.40) 1.33 (1.28–1.39) 1.43 (1.36–1.52) 0.87 (0.81–0.92) 0.85 (0.81–0.89) 0.94 (0.88–1.00) 1.34 (1.24–1.44) 1.31 (1.23–1.39) 1.58 (1.46–1.71) 0.91 (0.84–0.98) 0.87 (0.82–0.93) 1.05 (0.97–1.14) Intensity was defined as a measure of the number of pills prescribed multiplied by the dosage of the drug and divided by the time from the first prescription to the last prescription before the index date (mg/d). *Matched on age, gender, and county. Unadjusted for other covariates. †Adjusted for all covariates listed in Table 1. adjusted OR was 0.87 (95% CI, 0.79–0.96) for current users and 0.80 (95% CI, 0.69–0.92) for former users (data not shown). Among current users, the adjusted OR was 2.05 (95% CI, 1.61–2.60) for new users and 0.79 (95% CI, 0.71–0.87) for long-term users (data not shown). DISCUSSION In this large, population-based case–control study, long-term statin users had a reduced rate of atrial fibrillation or flutter compared with never users. The association was strongest among those who had used Table 5. Sensitivity analysis of the impact of different exposure windows for current versus former users on the association between statin use and atrial fibrillation or flutter. Adjusted ORs (95% CI)* Exposure window† 60 d Never users Current users New users Long-term users Former users 1.00 0.96 2.38 0.87 0.95 (reference) (0.93–1.00) (2.17–2.62) (0.84–0.91) (0.91–0.98) 90 d 1.00 0.96 2.14 0.88 0.94 (reference) (0.93–0.99) (1.98–2.32) (0.85–0.91) (0.90–0.98) 120 d 1.00 0.97 1.96 0.88 0.93 (reference) (0.94–0.99) (1.82–2.10) (0.85–0.91) (0.88–0.98) 180 d 1.00 0.96 1.73 0.87 0.93 (reference) (0.93–0.99) (1.63–1.85) (0.84–0.89) (0.87–0.99) *Adjusted for all covariates listed in Table 1. †Time between most recent statin prescription and index date. www.americantherapeutics.com American Journal of Therapeutics (2015) 22(3) Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. 192 statins for more than 1 year, which indicates a possible long-term protective effect of statins against atrial fibrillation or flutter. Clinically important, the protective effect was prominent among patients with cardiovascular diseases and diabetes. The strengths of our study include its populationbased design and setting within a free tax-supported universal healthcare system, which greatly reduced the risk of referral and diagnostic biases. We also controlled for several important risk factors for atrial fibrillation and flutter.28,29 The positive predictive value of a diagnosis of atrial fibrillation and flutter has been estimated to be 97% in the DNRP.31 Moreover, data on the other confounding variables have high coding accuracy,32,33 and the prescription database has virtually no missing data.20 Our study also has limitations. We used redeemed prescriptions as a surrogate for actual statin use. Thus, any nonadherence would bias our estimates toward the null. However, a previous validation study showed good correlation between general practitioner–reported medication use and timing of prescription dispensation.34 In addition, changes in the exposure window used to distinguish current users from former users did not produce relevant differences in our results. Although, we could not differentiate atrial fibrillation from atrial flutter, we know from a previous study that approximately 92% of these patients had atrial fibrillation.31 Surveillance bias also may have affected our results, that is, diagnoses of atrial fibrillation or flutter may be more common among statin users than among never users because users are more likely to undergo routine physical examinations. This bias might have led to underestimation of the preventive effect of statins on atrial fibrillation or flutter occurrence among statin users compared with never users. As well, inclusion of prevalent cases might have affected our results for new users and for those who were prescribed statins for less than 365 days. The finding that risk estimates for new users decreased as the time window of exposure was extended supports this hypothesis. Another concern is that we were unable to differentiate paroxysmal, persistent, and permanent atrial fibrillation. However, the restriction to cases who underwent cardioversion within 1 year after the first atrial fibrillation or flutter diagnosis did not change the results. Our results may also have been affected by confounding from unmeasured factors, such as lifestyle, socioeconomic status, smoking, and body mass index. Nevertheless, we adjusted for lifestyle-related diseases such as alcoholism, diabetes, cancer, chronic obstructive pulmonary disease, and asthma. In addition, a recent Danish cross-sectional study found no differences in smoking and alcohol consumption between American Journal of Therapeutics (2015) 22(3) Veronese et al statin users and nonusers.35 Conversely, statin users were found to have a slightly higher body mass index than persons who did not take statins. Our study is largely consistent with a recent UK case–control analysis including 55,412 patients from the General Practice Research Database diagnosed with new-onset atrial fibrillation between 1987 and 2007.11 Among statin users, 7144 were classified as current users (prescribed statins in the 3 months before the atrial fibrillation diagnosis), 597 as recent users (prescribed statins 3–12 months before the atrial fibrillation diagnosis), and 571 as past users (last statin prescription more than 12 months before the atrial fibrillation diagnosis). Compared with never users, a 15%–20% protective effect was found among current users (OR: 0.82, 95% CI, 0.73–0.92) and past users (OR: 0.84, 95% CI, 0.75–0.94). The effect among recent users was not statistically significant (OR: 0.94, 95% CI, 0.84– 1.05). However, the study may have underestimated the protective effect among both current and recent users, due to the lack of a new user design.27 Similarly, since no information regarding a possible association between different duration of statin use and atrial fibrillation or flutter was reported, the study did not reveal the long-term protective effect of statins on atrial fibrillation development. Furthermore, the study did not stratify the analyses by sex, age, and presence or absence of major comorbidities. Likewise, a recent meta-analysis of 11 randomized controlled trials (14,090 patients, 875 events) investigating the association between statin therapy and atrial fibrillation found that when used for primary prevention, statins have a preventive effect on atrial fibrillation (OR: 0.54; 95% CI, 0.40–0.74).19 However, most of the trials included in the analysis investigated the use of statins in prevention of atrial fibrillation in patients with cardiac surgery, and the results may not apply to other patient categories. In our study, we found that new users were at increased risk of atrial fibrillation and flutter compared with never users. Although this result might be explained both by the presence of confounding by indication (eg, acute coronary syndromes)36 and inclusion of prevalent cases, we cannot exclude the existence of a fast-acting adverse effect. To our knowledge, there is only one case report study suggesting that transient atrial fibrillation can be induced by the administration of simvastatin.37 However, data from 13 randomized controlled short-term trials (mean follow-up varied from 18 to 153 days), including 4414 randomized patients mostly undergoing cardiac surgery or electrical cardioversion, support the hypothesis that statin treatment reduces the odds of an atrial fibrillation episode (OR: 0.61, 95% CI, 0.51–0.74) even in the short term.15 www.americantherapeutics.com Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Statin Use and Atrial Fibrillation Although our data cannot elucidate the mechanism underlying our findings, the statin-induced attenuation of the electrical and structural atrial remodeling that contributes to the development of atrial fibrillation may explain the preventive effect of a long-term use of statins on atrial fibrillation onset.10 In addition, recent evidence on canine pulmonary vein sleeve preparations suggests that simvastatin exerts a direct antiarrhythmic effect by suppressing triggers responsible for the genesis of atrial fibrillation.38 In conclusion, our findings suggest that long-term statin use may have a preventive effect against new-onset atrial fibrillation or flutter. REFERENCES 1. Baigent C, Keech A, Kearney PM, et al. Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet. 2005;366:1267–1278. 2. Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet. 2002;360:7–22. 3. Fuster V, Ryden LE, Cannom DS, et al. 2011 ACCF/ AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 Guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines developed in partnership with the European Society of Cardiology and in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. J Am Coll Cardiol. 2011;57:e101–e198. 4. Heeringa J, van der Kuip DA, Hofman A, et al. Prevalence, incidence and lifetime risk of atrial fibrillation: the Rotterdam study. Eur Heart J. 2006;27:949–953. 5. 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. 6. Benjamin EJ, Levy D, Vaziri SM, et al. Independent risk factors for atrial fibrillation in a population-based cohort. The Framingham Heart Study. JAMA. 1994;271: 840–844. 7. Chien KL, Su TC, Hsu HC, et al. Atrial fibrillation prevalence, incidence and risk of stroke and all-cause death among Chinese. Int J Cardiol. 2010;139:173–180. 8. Kannel WB, Benjamin EJ. Status of the epidemiology of atrial fibrillation. Med Clin North Am. 2008;92:17–40, ix. 9. Miyasaka Y, Barnes ME, Gersh BJ, et al. Secular trends in incidence of atrial fibrillation in Olmsted County, Minnesota, 1980 to 2000, and implications on the projections for future prevalence. Circulation. 2006;114:119–125. 10. Savelieva I, Camm J. Statins and polyunsaturated fatty acids for treatment of atrial fibrillation. Nat Clin Pract Cardiovasc Med. 2008;5:30–41. www.americantherapeutics.com 193 11. Hodgkinson JA, Taylor CJ, Hobbs FD. Predictors of incident atrial fibrillation and influence of medications: a retrospective case-control study. Br J Gen Pract. 2011;61:e353–e361. 12. Savelieva I, Kakouros N, Kourliouros A, et al. Upstream therapies for management of atrial fibrillation: review of clinical evidence and implications for European Society of Cardiology guidelines. Part I: primary prevention. Europace. 2011;13:308–328. 13. Savelieva I, Kakouros N, Kourliouros A, et al. Upstream therapies for management of atrial fibrillation: review of clinical evidence and implications for European Society of Cardiology guidelines. Part II: secondary prevention. Europace. 2011;13:610–625. 14. Patel AA, White CM, Shah SA, et al. The relationship between statin use and atrial fibrillation. Curr Med Res Opin. 2007;23:1177–1185. 15. Rahimi K, Emberson J, McGale P, et al. Effect of statins on atrial fibrillation: collaborative meta-analysis of published and unpublished evidence from randomised controlled trials. BMJ. 2011;342:d1250. 16. Liu T, Li L, Korantzopoulos P, et al. Statin use and development of atrial fibrillation: a systematic review and meta-analysis of randomized clinical trials and observational studies. Int J Cardiol. 2008;126:160–170. 17. Xu Q, Guan YQ, Zhang D, et al. The effects of statin on atrial fibrillation: a meta-analysis of published data from randomized controlled trials. Curr Med Res Opin. 2011; 27:1771–1779. 18. Wang Z, Zhang Y, Gao M, et al. Statin therapy for the prevention of atrial fibrillation: a meta-analysis of randomized controlled trials. Pharmacotherapy. 2011;31:1051–1062. 19. Fang WT, Li HJ, Zhang H, et al. The role of statin therapy in the prevention of atrial fibrillation: a meta-analysis of randomized controlled trials. Br J Clin Pharmacol. 2012;74: 744–756. 20. Ehrenstein V, Antonsen S, Pedersen L. Existing data sources for clinical epidemiology: Aarhus University Prescription Database. Clin Epidemiol. 2010;2:273–279. 21. Pedersen CB. The Danish Civil Registration system. Scand J Public Health. 2011;39:22–25. 22. Andersen TF, Madsen M, Jorgensen J, et al. The Danish National Hospital Register. A valuable source of data for modern health sciences. Dan Med Bull. 1999;46:263–268. 23. Waldo AL. Mechanisms of atrial flutter and atrial fibrillation: distinct entities or two sides of a coin? Cardiovasc Res. 2002;54:217–229. 24. Waldo AL, Feld GK. Inter-relationships of atrial fibrillation and atrial flutter mechanisms and clinical implications. J Am Coll Cardiol. 2008;51:779–786. 25. Schmidt M, Christiansen CF, Mehnert F, et al. Non-steroidal anti-inflammatory drug use and risk of atrial fibrillation or flutter: population based case-control study. BMJ. 2011; 343:d3450. 26. Rothman KJ, Greenland S, Lash TL. Modern Epidemiology. Philadelphia, PA: Lippincott Williams & Wilkins; 2008. 27. Ray WA. Evaluating medication effects outside of clinical trials: new-user designs. Am J Epidemiol. 2003;158: 915–920. American Journal of Therapeutics (2015) 22(3) Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. 194 28. Krahn AD, Manfreda J, Tate RB, et al. The natural history of atrial fibrillation: incidence, risk factors, and prognosis in the Manitoba Follow-Up Study. Am J Med. 1995;98: 476–484. 29. van der Hooft CS, Heeringa J, van Herpen G, et al. Druginduced atrial fibrillation. J Am Coll Cardiol. 2004;44: 2117–2124. 30. Christiansen CF, Christensen S, Mehnert F, et al. Glucocorticoid use and risk of atrial fibrillation or flutter: a population-based, case-control study. Arch Intern Med. 2009;169:1677–1683. 31. Frost L, Vestergaard P. Alcohol and risk of atrial fibrillation or flutter: a cohort study. Arch Intern Med. 2004; 164:1993–1998. 32. Mosbech J, Jorgensen J, Madsen M, et al. The national patient registry. Evaluation of data quality [Article in Danish]. Ugeskr Laeger. 1995;157:3741–3745. 33. Thygesen SK, Christiansen CF, Christensen S, et al. The predictive value of ICD-10 diagnostic coding used to American Journal of Therapeutics (2015) 22(3) Veronese et al 34. 35. 36. 37. 38. assess Charlson comorbidity index conditions in the population-based Danish National Registry of Patients. BMC Med Res Methodol. 2011;11:83. Johannesdottir SA, Maegbaek ML, Hansen JG, et al. Correspondence between general practitioner-reported medication use and timing of prescription dispensation. Clin Epidemiol. 2012;4:13–18. Thomsen RW, Nielsen RB, Norgaard M, et al. Lifestyle profile among statin users. Pharmacoepidemiol Drug Saf. 2011;20:S111–S112. Jabre P, Jouven X, Adnet F, et al. Atrial fibrillation and death after myocardial infarction: a community study. Circulation. 2011;123:2094–2100. Akahane T, Mizushige K, Nishio H, et al. Atrial fibrillation induced by simvastatin treatment in a 61-year-old man. Heart Vessels. 2003;18:157–159. Sicouri S, Gianetti B, Zygmunt AC, et al. Antiarrhythmic effects of simvastatin in canine pulmonary vein sleeve preparations. J Am Coll Cardiol. 2011;57:986–993. www.americantherapeutics.com Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.