Dose-Related Effect of Statins on Atrial Fibrillation After Cardiac Surgery

Transcription

1 ORIGINAL ARTICLES: ADULT CARDIAC ADULT CARDIAC SURGERY: The Annals of Thoracic Surgery CME Program is located online at To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal. ADULT CARDIAC Dose-Related Effect of Statins on Atrial Fibrillation After Cardiac Surgery Antonios Kourliouros, MRCS, Ayesha De Souza, PhD, Neil Roberts, MRCS, Anna Marciniak, MD, Athanasios Tsiouris, MD, Oswaldo Valencia, MD, John Camm, FRCP, and Marjan Jahangiri, FRCS Department of Cardiothoracic Surgery, St. George s Hospital, London; and Department of Cardiac and Vascular Sciences, St. George s University of London, London, United Kingdom Background. Atrial fibrillation (AF) is the most common 38.3% in the no-statin group (adjusted odds ratio [OR], heart rhythm abnormality after cardiac surgery. It 2.00; 95% confidence interval, 1.24 to 3.24; p 0.004). increases morbidity and prolongs hospital stay. A role forsimvastatin (40 mg) and atorvastatin (40 mg) demon- the greatest effect on postoperative AF at 15.6% statins in the prevention of AF has been suggested. Westrated hypothesized that the incidence of postoperative AF due and 21.2%, respectively, vs no statins (respective adjusted to statin therapy is dose-related. Methods. A retrospective study of 680 consecutive ORs, 3.89 [p < ] and 2.76 p [ 0.012]). Intermediatedose (20 mg) statins were also effective against AF, at patients undergoing coronary bypass graft surgery 24.4% for simvastatin (adjusted OR, 2.32; p 0.004) and and/or aortic valve replacement was done. Excluded were 26.4% for atorvastatin (adjusted OR, 1.99, p 0.047). 57 patients (8.4%) with history of AF, permanent pace-low-dosmakers, statins, simvastatin or atorvastatin (10 mg), did and those receiving antiarrhythmic medication. Preoperative statin treatment and occurrence of postoperative not influence postoperative AF. Conclusions. Statin treatment may reduce the incidence AF were examined using propensity score match- of AF after cardiac surgery. Higher-dose statins have the ing to adjust for differences in patient characteristics between the statin and no-statin groups. Results. The cohort comprised 623 patients. The statin group had a 27.1% incidence of postoperative AF vs greatest preventative effect, whereas low-dose statins do not influence postoperative AF. (Ann Thorac Surg 2008;85: ) 2008 by The Society of Thoracic Surgeons Atrial fibrillation (AF) is the most common arrhythmia in clinical practice. The incidence of AF has been reported to be 19.2/1000 patient-years in men aged older than 65 [1]. Patients undergoing cardiac operations are more likely to develop AF during their postoperative period, with an incidence ranging from 11% to more than 50% in some series [2]. Moreover, postoperative AF is associated with increased morbidity and prolonged hospitalization [3]. Although the exact cause of AF remains unknown, several theories have been suggested. Atrial fibrosis, electrophysiologic remodeling, and inflammation are some of the factors that have been proposed to initiate and possibly precipitate AF [4, 5]. Specifically to postoperative AF, the metabolic and autonomic response to trauma, transient cardiac ischemia, and the inflammatory response associated with cardiopulmonary bypass are possible contributing factors [6]. Accepted for publication Jan 11, Address correspondence to Prof Jahangiri, Department of Cardiac Surgery, St. George s Hospital, Blackshaw Rd, London, SW17 0QT, United Kingdom; marjan.jahangiri@stgeorges.nhs.uk. The possible role of statins (3-hydroxy-3-methylglutarylcoenzyme A reductase inhibitors) in the prevention of postoperative AF has recently been suggested. In a prospective nonrandomized study, preoperative statin treatment significantly reduced the incidence of AF after coronary artery bypass graft (CABG) operations [7]. However, the statin type, dose, and duration of treatment varied between patients. In a recent randomized controlled trial, patients who received 40 mg of atorvastatin 7 days before CABG had a lower incidence of AF compared with placebo (35% vs 57%, p 0.003) [8]. The aim of this study was to identify the effect of varying doses of statins in the occurrence of AF after cardiac procedures and raise the importance of highdose statin therapy as a possible preventative measure against postoperative AF. Material and Methods Patient Population and Data A retrospective analysis was done of 680 consecutive patients who underwent CABG or aortic valve replace by The Society of Thoracic Surgeons /08/$34.00 Published by Elsevier Inc doi: /j.athoracsur

2 ADULT CARDIAC 1516 KOURLIOUROS ET AL Ann Thorac Surg DOSE-RELATED STATIN EFFECTS ON POST-OP AF 2008;85: Table 1. Patient Characteristics and Perioperative Variables for No Statin Versus Statin Groups Variable a No Statins (n 81) Statins (n 542) p Value Age, years Males 56 (69) 427 (79) Medical history Hypertension 31 (38) 321 (59) Diabetes 10 (12) 124 (23) Previous CVA 9 (11) 40 (7) 0.24 Medications -blockers 29 (36) 407 (75) ACE inhibitors 35 (43) 368 (68) Calcium channel blockers 16 (20) 170 (31) Cardiac history CCS class III-IV 15 (19) 213 (40) NYHA class III-IV 15 (19) 79 (15) 0.36 Previous MI 13 (16) 216 (40) Previous PCI 1 (1) 58 (11) Ejection fraction (11) 82 (15) 0.34 Left main stem disease 8 (10) 147 (27) Three-vessel disease 23 (28) 389 (72) Operative characteristics CABG 26 (32) 459 (85) CABG AVR 9 (11) 51 (9) AVR 46 (57) 32 (6) Nonelective 17 (21) 152 (28) 0.18 CPB time (minutes) Post-op atrial fibrillation 31 (38) 147 (27) a Categoric data are numbers (%); continuous data as means standard deviation. ACE angiotensin-converting enzyme; AVR aortic valve replacement; CABG coronary artery bypass graft; CCS Canadian Cardiovascular Society; CPB cardiopulmonary bypass; CVA cerebrovascular accident; NYHA New York Heart Association; PCI percutaneous coronary intervention. ment (AVR), or both, between January and December 2005 in the Department of Cardiothoracic Surgery at St George s Hospital, London. Initially, 57 patients (8.3%) were excluded because of preoperative AF, they were receiving antiarrhythmic medications (except -blockers), or they had pacemakers. From the 623 patients included, demographic characteristics, medical history, and drug therapy were evaluated. Preoperative cardiac investigations and clinical findings were also recorded. Operative variables and postoperative events were evaluated with emphasis on the occurrence of AF and hospital length of stay. A detailed assessment of medical records, daily electrocardiograms (ECGs), and drug charts was combined with data obtained from the audit database for outcomes in cardiac surgery at St George s Hospital. All patients included in the statin group received statins for a minimum of 2 months before operation. Routine postoperative heart rhythm assessment was achieved with continuous telemonitoring for the first 72 hours after the operation and with 4-hourly clinical examinations thereafter. Twelve-lead ECGs were also obtained and assessed daily. The primary end point was the development of postoperative AF during the hospital stay. We compared the incidence of postoperative AF in patients taking statins with patients not taking statins. The incidence of postoperative AF was then assessed in patients receiving different treatment doses of atorvastatin and simvastatin compared with patients not receiving statins. Six additional patients were further excluded because they received an unusual statin dose (simvastatin, 30 mg, 1; atorvastatin, 30 mg, 2; simvastatin, 80 mg, 3). Patients taking other statin types (pravastatin, 16; rosuvastatin, 9) accounted for only 4.6% of the total statin population and these subgroups were not analyzed; however, these statin subgroups were included in the statin vs no-statin analysis. Statistical Analysis Univariate analysis was performed to examine differences in variables between patients on statins and patients not on statins. Variables with a significance level of less than 0.2 were entered into a multivariable regression model and predictors of statin group membership were identified. Logistic regression with forward elimination determined the most important denominators (CABG, AVR, and -blocker use) for receiving statin treatment or not. A propensity model was then constructed based on these variables and was used to calculate the propensity score, which reflects the probability of a patient s assignment to statin treatment. Having calculated the propensity score for each pa- Table 2. Propensity Score for Statins Coefficient SE p Value CABG blockers Previous PCI Aortic valve replacement Calcium channel blockers ACE inhibitors Diabetes mellitus Hypertension Left main stem disease Three-vessel disease CCS class III-IV Previous MI Sex Constant Logistic regression CABG blockers AVR Constant ACE angiotensin converting enzyme; CCS Canadian Cardiovascular Society; CABG coronary artery bypass graft; MI myocardial infarction; PCI percutaneous coronary intervention; SE standard error.

3 Ann Thorac Surg KOURLIOUROS ET AL 2008;85: DOSE-RELATED STATIN EFFECTS ON POST-OP AF Table 3. Patient Characteristics and Perioperative Variables in the Atrial Fibrillation and No-Atrial Fibrillation Groups Variables a tient, we then adjusted the variables of interest and compared surgical outcomes. Furthermore, a univariate analysis of the study population was done by the development (or not) of postoperative AF. Variables that were found to have a value of p 0.2 in the univariate analysis were examined by multivariate logistic regression to identify independent predictors of postoperative AF and correlate with the outcomes of the propensity score matching. Categoric and dichotomous data were examined with 2 and Fisher exact tests where appropriate. The Mann-Whitney U test was used for continuous variables with nonparametric distribution. Values of p 0.05 were considered statistically significant. Statistical analysis was performed with SPSS 14.0 software (SPSS Inc, Chicago, IL) Results No AF (n 445) AF (n 178) p Value Age (years) Males 348 (78) 135 (76) 0.52 Medical history Hypertension 245 (55) 107 (60) 0.25 Diabetes 101 (23) 33 (19) 0.25 Previous CVA 36 (8) 13 (7) 0.74 Medication -blockers 319 (72) 117 (66) 0.14 ACE inhibitors 292 (66) 111 (62) 0.44 Calcium channel blockers 133 (30) 53 (30) 0.98 Cardiac history CCS class III-IV 169 (38) 59 (33) 0.26 NYHA class III-IV 61 (14) 33 (19) 0.13 Previous MI 171 (38) 58 (33) 0.17 Previous PCI 47 (11) 12 (7) 0.14 Ejection fraction (14) 27 (15) 0.80 Left main stem disease 110 (25) 45 (25) 0.88 Three-vessel disease 295 (66) 117 (66) 0.58 Operative characteristics CABG 361 (81) 124 (70) CABG AVR 32 (7) 28 (16) AVR 52 (12) 26 (15) Nonelective 129 (29) 40 (22) CPB time, min Statin use 395 (89) 147 (83) a Categoric data are numbers (%); continuous data are means standard deviation. ACE angiotensin converting enzyme; AF atrial fibrillation; AVR aortic valve replacement; CABG coronary artery bypass graft; CCS Canadian Cardiovascular Society; CPB cardiopulmonary bypass; CVA cerebrovascular accident; MI myocardial infarction; NYHA New York Heart Association; PCI percutaneous coronary intervention. From the 623 patients studied, 485 (77.8%) had CABG procedures, 78 (12.5%) had AVR, and 60 (9.6%) CABG Table 4. Independent Risk Factors of Postoperative Atrial Fibrillation as Identified by Multivariate Logistic Regression Analysis Variable Coefficient Standard Error Exponential Coefficient 95% CI p Value Statins Age Constant CI confidence interval. and AVR. The 542 patients on statins accounted for 87% of the population, and the 81 patients not on statins accounted for 13%. Patient characteristics and procedural variables are reported in Table 1. Analysis of cardiac history, medical treatment, and operative characteristics demonstrated heterogeneity between the two groups. Patients taking statins were more likely to have systemic hypertension, diabetes mellitus, and previous myocardial infarction (MI). They were more likely to have left main stem disease, three-vessel disease, and a higher Canadian Cardiovascular Society (CCS) classification score. Patients taking statins were also more likely to receive -blockers, angiotensin-converting enzyme inhibitors, and calcium channel blockers, and 459 (85%) underwent CABG, whereas 46 patients (57%) not taking statins underwent AVR. Cardiopulmonary bypass (CPB) times were similar in both groups. Predictors for inclusion in the statin group, as identified by multivariate regression analysis of covariables, followed by logistic regression (Table 2), were used to create the propensity score model to adjust for outcomes. Atrial fibrillation occurred in 147 patients (27.1%) taking statins, which was significantly lower compared with the 31 patients (38.3%) with AF who did not take statins (OR, 1.67; 95% confidence interval [CI], 1.02 to 2.71; p 0.038). After adjustment for propensity score, the reduction in postoperative AF in the statin group was more pronounced (adjusted OR, 2.00; 95% CI, 1.24 to 3.24; p 0.004). Table 5. Incidence of Postoperative Atrial Fibrillation in Each Statin Group Compared With No-Statin Group Statin Dose, mg No. AF, No. (%) Adjusted Analysis 1517 OR 95% CI a p Value a No statins (38.3) Simvastatin (53.3) (24.4) (15.6) Atorvastatin (41.9) (26.4) (21.2) (35.7) a Adjusted for propensity score. AF atrial fibrillation; CI confidence interval; OR odds ratio. ADULT CARDIAC

4 ADULT CARDIAC 1518 KOURLIOUROS ET AL Ann Thorac Surg DOSE-RELATED STATIN EFFECTS ON POST-OP AF 2008;85: Hospital stay was also significantly shorter in patients on statins, with fewer patients requiring hospitalization for more than 8 days compared with patients not on statins (21% vs 35%; adjusted OR, 2.01; 95% CI, 1.22 to 3.31; p 0.006). Overall morbidity and length of stay in the intensive care unit were not different. Further univariate analysis of the same patient variables by the development (or not) of postoperative AF (Table 3) demonstrated no significant difference in most preoperative variables between patients who developed postoperative AF compared with those who did not; the only exceptions were age, type of operation, and statin use. More specifically, mean age was significantly higher in the AF group ( vs , p 0.001) as was the proportion of patients who received CABG with or without AVR (p 0.002). Statin use was significantly higher in the group that remained in sinus rhythm postoperatively (p 0.038). Multivariate logistic regression analysis demonstrated that increasing age and no use of statins were the only independent risk factors for postoperative AF (Table 4). This finding is in line with the outcome of the propensity score matching, which identified a significant reduction in postoperative AF in patients on statins. The incidence of postoperative AF in each group of the study population is reported in Table 5. It is apparent that any increase in statin dose decreases the incidence of AF in that subgroup, with the exception of atorvastatin at 80 mg. Furthermore, comparison of each statin subgroup with the no-statin group demonstrates a significant reduction in the incidence of AF in patients taking 40 mg of atorvastatin (adjusted OR, 2.76; 95% CI, 1.24 to 6.15; p 0.012), atorvastatin at 20 mg (adjusted OR, 1.99; 95% CI, 1.00 to 3.94; p 0.047), simvastatin at 40 mg (adjusted OR, 3.89; 95% CI, 2.03 to 7.45; p ), and simvastatin at 20 mg (adjusted OR, 2.32; 95% CI, 1.30 to 4.11; p 0.004). Doses of simvastatin and atorvastatin at 10 mg did not influence the development of postoperative AF. Comment The effect of statin treatment on AF was first described in nonsurgical patients. A prospective study of 449 patients with stable coronary artery disease during a 5-year follow-up period showed that the risk of AF was significantly reduced in patients who took statins [9]. A smaller retrospective study was the first to demonstrate a decrease in recurrence of AF in patients on statins who underwent successful cardioversion for lone AF [10]. However, a prospective randomized study of 114 patients on the effect of pravastatin at 40 mg in the recurrence of AF after electrical cardioversion did not demonstrate any significant reduction in the rate of recurrent AF [11]. The aim of our study was to assess the effect of varying doses of statins on AF after CABG, AVR, and CABG and AVR. We have shown that preoperative statin treatment is significantly and independently associated with reduced AF; however, further analysis demonstrates that this effect is primarily attributed to higher doses of statin. The low-dose statin regimens that are commonly used in clinical practice for lipid control do not seem to be effective in reducing postoperative AF. Perioperative -blocker therapy is known to exhibit a protective effect against AF [12]. In our study, preoperative -blocker treatment did not influence postoperative AF. However, -blockers commenced in the early postoperative period significantly reduced new-onset AF (p 0.001). Angiotensin enzyme-converting inhibitors have also been suggested to exhibit a possible protective effect on postoperative AF [13], but this was not observed in our study. Others have shown an association between statin therapy and postoperative AF. Marin and colleagues [7] demonstrated a significant reduction in postoperative AF (p 0.038) associated with previous statin use in a prospective nonrandomized study of 234 patients undergoing CABG; however, the statin type and dose were not uniform in all patients. A recent observational study of 362 patients showed a significant reduction in AF after CABG (p 0.03) and decreased duration of AF in patients taking statins [14]. The third study, Atorvastatin for Reduction of Myocardial Dysrhythmia After Cardiac Surgery (ARMYDA-3) [8], is the only randomized double-blind trial to demonstrate a reduction in AF and hospital stay after cardiac surgery in patients who received atorvastatin (40 mg) 7 days before operation compared with placebo. The incidence of AF was 35% in the statin group vs 57% in the placebo group (p 0.003). This rate of postoperative AF is higher than expected in the current literature, as most other studies have shown AF rates of 20% to 40% in their control groups [15 17]. The same trial correlated high postoperative C-reactive protein (CRP) levels with an increased risk of AF, a finding that was not confirmed in a recent cohort study [17]. Finally, statin use was associated with a decreased occurrence of AF after thoracic operations (pulmonary lobectomy, esophagectomy) compared with patients not using statins (11% vs 29%, p 0.025) [18]. Statins demonstrate a pleiotropic effect compared with other lipid-lowering agents [19]. The antiinflammatory properties of statins have been suggested as one of the mechanisms by which they exhibit their protective role in atherogenesis [20] and possibly in the development of AF. Two independent clinical trials that appeared almost simultaneously were the first to report an association between CRP and AF in non-postoperative patients. The first was a case-control study [21] that included 131 patients with atrial arrhythmias. The concentration of CRP was significantly elevated in patients with AF and was greatest with persistent AF. In the second study [22], 50 patients with paroxysmal AF who underwent pharmacological cardioversion were compared with age and sex-matched controls. Levels of CRP were higher in patients with AF and significantly associated with unsuccessful cardioversion to sinus rhythm. These findings have also been confirmed by others [23]. The concept of an association between AF and inflammation has also been supported by population-based studies, where CRP levels were associated with the development of AF and predicted patients at risk for future AF [24]. The antiinflammatory role of statins as a regulatory mechanism for AF has been suggested in two

5 Ann Thorac Surg KOURLIOUROS ET AL 2008;85: DOSE-RELATED STATIN EFFECTS ON POST-OP AF studies. Kumagai and colleagues [25] evaluated the effect of atorvastatin in AF in a canine sterile pericarditis model. The atorvastatin group had lower CRP levels, less pronounced fibrosis in the atrial myocardium, and a shorter duration of AF compared with the control group. This study deserves attention as the first one to emphasize the role of inflammation on atrial electrophysiologic as well as on atrial structural changes. These findings were supported by another study on mongrel dogs that underwent atrial tachypacing and simvastatin treatment [26]. Atrial tachypacing-induced AF was virtually abolished and effective refractory period shortening was significantly suppressed in the simvastatin treated dogs. Emerging data suggest that statin treatment is a safe preventative measure for postoperative AF. This study shows that high-dose statin therapy is more effective in reducing AF than low-dose statins. The mechanisms by which statins exhibit this beneficial effect are currently unknown. Further research into the structural changes of the heart and the inflammatory response associated with AF may provide more information on its pathogenesis and guide future treatment strategies. One study limitation is that postoperative assessment of heart rhythm was performed with telemonitoring for 72 hours and with daily ECGs and clinical examination thereafter. Self-terminating episodes of AF or other arrhythmias may have occurred and not been recorded during these intervals. This study is also subject to limitations inherent to any retrospective analysis. The indications for preoperative statin therapy were not uniform in our study group and not always recorded in the hospital notes. The duration of preoperative treatment was variable between patients. Despite using a well-established balancing score to adjust for patient characteristics between the statin and no-statin groups, selection bias may not have been completely eliminated. The observed reduction in postoperative AF with increasing statin dose can only suggest a dose-related effect. Despite having used a validated balancing score to adjust for possible covariates, a direct etiologic association cannot be confirmed and is beyond the scope of this study. References 1. Psaty BM, Manolio TA, Kuller LH, et al. Incidence of and risk factors for atrial fibrillation in older adults. Circulation 1997;96: Creswell LL, Schuessler RB, Rosenbloom M, Cox JL. Hazards of postoperative atrial arrhythmias. Ann Thorac Surg 1993; 56: Aranki SF, Shaw DP, Adams DH, et al. Predictors of atrial fibrillation after coronary artery surgery. Current trends and impact on hospital resources. Circulation 1996;94: Fuster V, Ryden LE, Cannom DS, et al. ACC/AHA/ESC 2006 Guidelines for the Management of Patients with Atrial Fibrillation: a report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Circulation 2006;114: Jahangiri M, Weir G, Mandal K, Savelieva I, Camm J. Current strategies in the management of atrial fibrillation. Ann Thorac Surg 2006;82: Hogue CW Jr, Creswell LL, Gutterman DD, Fleisher LA. Epidemiology, mechanisms, and risks: American College of Chest Physicians guidelines for the prevention and management of postoperative atrial fibrillation after cardiac surgery. Chest 2005;128(suppl): Marin F, Pascual DA, Roldan V, et al. Statins and postoperative risk of atrial fibrillation following coronary artery bypass grafting. Am J Cardiol 2006;97: Patti G, Chello M, Candura D, et al. Randomized Trial of Atorvastatin for Reduction of Postoperative Atrial Fibrillation in Patients Undergoing Cardiac Surgery. Results of the ARMYDA-3 (Atorvastatin for Reduction of Myocardial Dysrhythmia After cardiac surgery) Study. Circulation 2006;114: Young-Xu Y, Jabbour S, Goldberg R, et al. Usefulness of statin drugs in protecting against atrial fibrillation in patients with coronary artery disease. Am J Cardiol 2003;92: Siu CW, Lau CP, Tse HF. Prevention of atrial fibrillation recurrence by statin therapy in patients with lone atrial fibrillation after successful cardioversion. Am J Cardiol 2003; 92: Tveit A, Grundtvig M, Gundersen T, et al. Analysis of pravastatin to prevent recurrence of atrial fibrillation after electrical cardioversion. Am J Cardiol 2004;93: Bradley D, Creswell LL, Hogue CW Jr, Epstein AE, Prystowsky EN, Daoud EG. Pharmacologic prophylaxis: American College of Chest Physicians guidelines for the prevention and management of postoperative atrial fibrillation after cardiac surgery. Chest 2005;128(suppl): White CM, Kluger J, Lertsburapa K, Faheem O, Coleman C. Effect of preoperative angiotensin converting enzyme inhibitor or angiotensin receptor blocker use on the frequency of atrial fibrillation after cardiac surgery: a cohort study from the atrial fibrillation suppression trials II and III. Eur J Cardiothorac Surg, 2007;31: Ozaydin M, Dogan A, Varol E, et al. Statin use before by-pass surgery decreases the incidence and shortens the duration of postoperative atrial fibrillation. Cardiology 2007; 107: Calo L, Bianconi L, Colivicchi F, et al. N-3 Fatty acids for the prevention of atrial fibrillation after coronary artery bypass surgery: a randomized, controlled trial. J Am Coll Cardiol 2005;45: Barnes BJ, Kirkland EA, Howard PA, et al. Risk-stratified evaluation of amiodarone to prevent atrial fibrillation after cardiac surgery. Ann Thorac Surg 2006;82: Ahlsson AJ, Bodin L, Lundblad OH, Englund AG. Postoperative atrial fibrillation is not correlated to C-reactive protein. Ann Thorac Surg 2007;83: Amar D, Zhang H, Heerdt PM, Park B, Fleisher M, Thaler HT. Statin use is associated with a reduction in atrial fibrillation after noncardiac thoracic surgery independent of C-reactive protein. Chest 2005;128: Landmesser U, Bahlmann F, Mueller M, et al. Simvastatin versus ezetimibe: pleiotropic and lipid-lowering effects on endothelial function in humans. Circulation 2005;111: Shishehbor MH, Brennan ML, Aviles RJ, et al. Statins promote potent systemic antioxidant effects through specific inflammatory pathways. Circulation 2003;108: 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: Dernellis J, Panaretou M. C-reactive protein and paroxysmal atrial fibrillation: evidence of the implication of an inflammatory process in paroxysmal atrial fibrillation. Acta Cardiol 2001;56: ADULT CARDIAC

6 ADULT CARDIAC 1520 KOURLIOUROS ET AL Ann Thorac Surg DOSE-RELATED STATIN EFFECTS ON POST-OP AF 2008;85: Conway DS, Buggins P, Hughes E, Lip GY. Predictive value of indexes of inflammation and hypercoagulability on success of cardioversion of persistent atrial fibrillation. Am J Cardiol 2004;94: Aviles RJ, Martin DO, Apperson-Hansen C, Houghtaling PL, Rautaharju P, Kronmal RA, et al. Inflammation as a risk factor for atrial fibrillation. Circulation 2003;108: Kumagai K, Nakashima H, Saku K. The HMG-CoA reductase inhibitor atorvastatin prevents atrial fibrillation by inhibiting inflammation in a canine sterile pericarditis model. Cardiovasc Res 2004;62: Shiroshita-Takeshita A, Schram G, Lavoie J, Nattel S. Effect of simvastatin and antioxidant vitamins on atrial fibrillation promotion by atrial-tachycardia remodeling in dogs. Circulation 2004;110: INVITED COMMENTARY The study by Kourliouros and colleagues [1] is a thorough and clearly presented retrospective assessment describing the relationship between hydroxymethylgluteryl-coenzyme A reductase inhibitor (statin) use and postoperative atrial fibrillation. The authors found a significant reduction in postoperative atrial fibrillation in patients taking higher doses of statins at the time of their procedures. This effect was not seen in patients on lower doses of the same agents. When considering the results of this study it is important to understand the limitations of the methodology. In a retrospective analysis it is difficult to control for baseline differences between the treatment and nontreatment groups. In this study, there were important differences between the patients with and without statins. The statins were more likely to be prescribed for patients with coronary artery disease. Therefore, the underlying heart disease and surgical procedures in the statin group were significantly different than those in the non-statin patients. Furthermore, there may have been differences in unreported variables, such as left atrial size. The authors have used appropriate statistical methods to adjust for these differences, but the results still can not be accepted with the same confidence as those of a randomized trial. The study clearly demonstrates an association between high-dose statin use and a reduction in postoperative atrial fibrillation. However, the conclusion that statins prevented atrial fibrillation is more tenuous, given the methodology used. The current article should be considered within the context of previous studies examining the role of statins for prevention of atrial fibrillation. These include some observational studies and a single, small randomized study as noted by Kourliouros and colleagues [1] (in Reference 8 of their article). The randomized trial called ARMYDA-3 (Atorvastatin for Reduction of MYocardial Dysrhythmia After cardiac surgery) was well conducted, but there were some unexpected differences in the baseline patient characteristics and a higher than expected incidence of postoperative atrial fibrillation that cast some residual doubt on the conclusions. The current study makes an important contribution by lending weight to these earlier findings and adding more data regarding variations in statin dose and the incidence of postoperative atrial fibrillation. Stuart P. Thomas, MD Department of Cardiology Westmead Hospital Westmead NSW 2145, Australia stuartpt@yahoo.com Reference 1. Kourliouros A, De Souza A, Roberts N, et al. Dose-related effect of statins on atrial fibrillation after cardiac surgery. Ann Thorac Surg 2008;85: by The Society of Thoracic Surgeons /08/$34.00 Published by Elsevier Inc doi: /j.athoracsur