The impact of dilated left atrium on rhythm control in patients with newly diagnosed persistent atrial fibrillation: the Belgrade atrial fibrillation project

For patients with persistent atrial fibrillation (AF) who remain highly symptomatic even with satisfactory ventricular rate control, rhythm control [cardioversion and maintenance of sinus rhythm (SR)] may be a management option (1). There is a well-established relationship between increased left atrial (LA) size and the risk of incident AF or recurrent AF following cardioversion, but data on the association of increased LA size with immediate outcome of AF cardioversion or long-term rhythm control are conflicting (2-14). We examined the impact of increasing LA anteroposterior diameter (LA-APD) on cardioversion and rhythm control outcomes in 335 consecutive patients with newly diagnosed non-valvular symptomatic persistent AF scheduled for elective external direct-current (DC) cardioversion (Table 1). We tested the hypothesis that a greater extent of LA-APD dilatation confers a greater risk of cardioversion or rhythm control failure. After a 4- to 6-week pretreatment with oral amiodarone 400 mg daily (because of a relatively long AF duration prior to cardioversion) (11) and oral anticoagulation with target International normalised ratio 2.0–3.0, electrocardioversion was attempted using ≤ 2 successive external monophasic DC-shocks of 300–360J, under light general anaesthesia (etomidate 10–20 mg intravenously). Overall, cardioversion was successful (SR ≥ 24 h, documented by 12-lead electrocardiogram) in 256 patients (76.4%). In 60/335 patients (17.9%), cardioversion occurred during the amiodarone pretreatment phase, while 275 patients underwent DC-cardioversion, which was successful in 196/275 patients (71.3%). In patients with preserved left ventricular ejection fraction (LVEF ≥ 50%) amiodarone was subsequently substituted by no treatment [20 patients (7.8%)], beta blockers [53 (20.7%)], propafenone [90 (35.2%)] or sotalol [12 (4.7%)]. In patients with prior AF> 12 months (regardless of LVEF) and patients with LVEF < 50% (regardless of prior AF duration) amiodarone was continued (81 [31.6%]). If needed, electrocardioversion was repeated, but not within 6 months following prior cardioversion and further anti-arrhythmic treatment was applied according to clinical judgment. Rhythm control failure was defined as progression to permanent AF despite active treatment (repeated electrocardioversions and long-term anti-arrhythmic drug treatment). LA-APD was normal (≤ 40 mm) in 137 patients (40.9%), 41–45 mm in 130 (38.8%), 46–50 mm in 51 (15.2%) and 50–60 mm in 17 (5.1%). In multivariate analysis, adjusted for baseline clinical and echocardiographic characteristics, each 5-mm increase in LA-APD was associated with an increasing risk of cardioversion failure (OR 2.8, 95%CI: 1.3–5.8, OR 3.3, 95%CI: 1.4–7.8 and OR 6.0, 95%CI 1.6–21.9, respectively, all p < 0.01). During follow up (mean 4.5 ± 3.1 years, range 1–8), AF progressed to permanent form in 118/256 patients (46.1%), despite active treatment. Overall, 197/335 patients (58.8%) had permanent AF at the end of follow up, either because of unsuccessful cardioversion at baseline (79 patients, 23.6%) or because of AF progression during follow up (118, 35.2%). Only 40 patients (11.9%) were continuously in SR (for mean 2.9 ± 2.2 years, range 1–7) and 98 (29.3%) needed repeated cardioversions (1–8, mean 2.0 ± 1.5 cardioversions per patient) to maintain SR. In multivariate analysis, adjusted for baseline clinical (including AF duration) and echocardiographic characteristics and anti-arrhythmic therapy, each 5-mm increase in LA-APD was associated with an increasing risk of rhythm control failure [HR 1.8, 95%CI:1.2–2.7 and HR 2.9, 95%CI:1.8–4.9 (for LA-APD 46–60 mm), all p < 0.01]. An analysis without including patients who were cardioverted pharmacologically, showed broadly similar results where increased LA diameter was significantly related to the risk of cardioversion failure (OR 1.9, 95%CI: 1.4–2.5; p < 0.001). A 5-mm incremental increase in LA-APD was also associated with an increasing risk of cardioversion failure (OR 3.0, 95%CI: 1.5–6.0; p = 0.001; OR 3.2, 95%CI: 1.4–7.3; p = 0.007; and OR 8.5, 95%CI: 2.8–25.9; p < 0.001, respectively). In the present study, LA size was estimated using a ‘surrogate measure’ (LA-APD), which is commonly used in clinical practice as an estimate of LA size, although it may not accurately represent the true LA dimensions (11). However, we have demonstrated that the risk of cardioversion or rhythm control failure in patients with persistent AF is significantly influenced by the degree of LA-APD dilatation and gradually increases with each 5 mm of LA-APD dilatation. None.