Atrial Fibrillation reduces the Functional REServe of the Heart by a fifth: A pilot FRESH-AF study

Abstract

The advent of catheter ablation has reignited the controversy about whether rhythm or rate control should be the preferred mode of therapy for atrial fibrillation (AF) [1]. It is well known that compared to sinus rhythm (SR), AF is associated with increased mortality and morbidity including stroke, heart failure, and impaired quality of life. Furthermore, patients may experience adverse effects from drug treatment that may negate thebeneficial effects ofmaintaining SR [2].What is still unknown is inwhatways and by howmuch overall cardiac function is impairedwhen the atrial “kick” in SR is lost after the onset of AF. To fill this gap in knowledge, it is essential to assess cardiac functionnot onlyat rest but also during maximal stress, and not merely piecemeal aspects of cardiac function. This can be achieved by conducting full cardiopulmonary exercise testing (CPX) in combination with non-invasive evaluation of central haemodynamics [3]. Thiswouldprovide a quantitative assessment of aerobic physical exercise capacity represented by peak O2 consumption (VO2max), together with peak cardiac power output (CPOmax) which measures the overall cardiac dysfunction [4], and is by far the strongest predictor of prognosis in heart failure patients [5]. As metabolic and circulatory demands increase during severe exercise, patients with AF have lower cardiovascular reserve. Previous studies have shown that VO2max improved by 5–13% after restoration of SR in patients with AF [6], buthowmuchcardiac functional gain is achieved thereby is still unknown. We therefore measured changes in CPOmax after successful DC cardioversion (DCCv) in patients with optimally rate-controlled AF. In so doing, we tested the hypothesis that the cardiac pumping capability of patients in AF is diminished by about a fifth. Twelve unselected ambulatory patients (10 males, age: 57.5 ± 13.7 (mean ± SD) years) with symptomatic persistent AF recruited from outpatient clinics underwent full CPX with measurements of non-invasive haemodynamics before and after successful DCCv. The aetiologies (including overlaps) of their AF were ischaemic heart disease (17%), hypertension (25%), LV dysfunction (42%), valve disease (25%), lone AF (25%), and obesity (BMI N 35, 25%). None of the patients had previous catheter ablation, or any contraindication to undergomaximal exercise, such as significant outflow obstruction, haemodynamically serious structural heart disease, uncontrolled ventricular tachyarrhythmia, uncontrolled hypertension, limiting symptoms of ischaemic heart disease, inability or deemed unsafe to exercise on treadmill, contraindications to electrical cardioversion or full anticoagulation, or inability to provide full informed consent. Standardexerciseparametersweremeasuredaspartof an incremental cardiopulmonary exercise test performed on a Trackmaster TMX425 treadmill (Full Vision, Kansas, USA) using the Bruce or modified Bruce protocol, as previously described [3]. Rates of oxygen consumption (VO2) and other gaseous exchanges were recorded breath-by-breath using the Medgraphics Ultima CardiO2 analytic system (Medgraphics, Minnesota, USA). Using the CO2-rebreathing technique (indirect Fick), cardiac output were measured at rest and during peak exercise. Mean arterial pressure was calculated from the equation, MAP=DBP+ 0.412 ∗ (SBP− DBP) [7]. CPOmax in watts is the product of peak cardiac output (COmax in l.min−1)multiplied by theMAP inmm Hg at peak exercisemultiplied by a conversion factor (2.22 × 10−3) [8]. A paired sample t-test was used to assess the change in cardiac reserve variables pre and post DCCv. This study was approved by the National Research Ethics Service Committee for Yorkshire & Humber in Leeds East. After successful cardioversion, subjectively all patients claimed to feel better and more energetic and, as shown in Table 1, objectively their treadmill exercise duration increased by 19.3% (P b 0.05), coupled with an improvement in peak O2 consumption by 16.5% (P b 0.01). Their resting ventricular rate decreased from 83.1 ±20.3 to 66.9 ± 20.5 min− 1 (P b 0.05), while peak exercise HR decreased from 163 ± 42 to 128 ± 18 min−1 to (P b 0.05). Secondarily, these negative chronotropic effects led to heightened stroke volume at peak exercise by 46% (P b 0.0001), thereby raising the peak cardiac output by 16.5% (P b 0.0001). Due to a concomitantly greater