3153Rate versus rhythm control and mortality in atrial fibrillation patients: a Danish nationwide cohort study

Abstract

Abstract Background Treatment of AF patients with rate or rhythm drug therapy have shown no difference in mortality in clinical trials. However, the generalizability of these trials to real-world populations can be questioned. Purpose We aimed to investigate the all-cause and cardiovascular (CV) mortality risk in a nationwide AF cohort by treatment strategy (rate vs. rhythm) and by individual drug classes. Methods We queried the Danish nationwide registries from 2000 to 2015 to identify patients with AF. A rate control strategy included the use of one or more of the following medications: beta-blocker, digoxin, and a class-4 calcium channel blocker (CCB). A rhythm control strategy included the use of an anti-arrhythmic drug (amiodarone and class-1C). Primary outcome was all-cause mortality. Secondary outcome was CV mortality. Adjusted incidence rate ratios (IRR) were computed using Poisson regression with time-dependent covariates allowing patients to switch treatment during follow-up. Results Of 140,697 AF patients, 131,793 were on rate control therapy and n=8,904 were on rhythm control therapy. At baseline, patients on rhythm control therapy were younger (71 yrs [IQR: 62–78] vs 74 [65–82], p<0.001) more likely male (63.5% vs 51.7% p<0.001), had more prevalent heart failure (31.1% vs 19.4%, p<0.001) and ischemic heart disease (40.1% vs. 23.3%, p<0.001), and had more prior CV-related procedures; PCI (7.4% vs. 4.0% p<0.001) and CABG (15.0% vs. 2.3%, p<0.001). During a median follow up of 4.0 (IQR: 1.7–7.3) years, there were 64,653 (46.0%) deaths from any-cause, of which 27,025 (19.2%) were CVD deaths. After appropriate adjustments and compared to rate control therapy, we found a lower IRR of mortality and CV mortality in those treated with rhythm control therapy (IRR: 0.93 [95% CI: 0.90–0.97] and IRR 0.84 [95% CI: 0.79–0.90]). Compared with beta-blockers, digoxin was associated with increased risk of all-cause and CV mortality (IRR: 1.26 [95% CI: 1.24–1.29] and IRR: 1.32 [95% CI: 1.28–1.36]), so was amiodarone: IRR for all-cause mortality: 1.16 [95% CI: 1.11–1.21] and IRR for CV mortality: 1.12 [95% CI: 1.05–1.19]. Class-1C was associated with lower all-cause (0.43 [95% CI: 0.37–0.49]) and CV mortality (0.35 [95% CI: 0.28–0.44]). Figure 1. Models were adjusted for age, sex, ischemic heart disease, stroke, chronic obstructive pulmonary disease, chronic kidney disease, valvular atrial fibrillation, bleeding, diabetes, ablation, pacemaker, implantable cardioverter defibrillator, hypertension, heart failure, use of loop diuretics, calendar year, and time on treatment. Abbreviations; CCB; calcium channel blocker, PY; person years. Conclusions In a real-world AF cohort, we found that compared with rate control therapy, rhythm control therapy was associated with a lower risk of all-cause and CV mortality. The reduced mortality risk with rhythm therapy could reflect an appropriate patient selection. Acknowledgement/Funding The Danish Heart Foundation