Heart Rate in Patients with Atrial Fibrillation and Heart Failure with Preserved Ejection Fraction: A Prognosticator Like in Sinus Rhythm?

This article refers to ‘Association of heart rate with mortality in sinus rhythm and atrial fibrillation in heart failure with preserved ejection fraction’ by U. Sartipy et al., published in this issue on pages 471–479. Atrial fibrillation (AF) and heart failure with preserved ejection fraction (HFpEF) are two extremely common diseases, each leading to adverse prognosis.1, 2 The number of patients suffering from both AF and HFpEF is growing rapidly, due to ageing populations, increasing incidence of AF in older patients,3 and better treatment of preceding cardiovascular diseases. Shared risk factors, reciprocal causation and poor prognosis with concomitant disease make the combination of AF and HFpEF a major clinical burden.4 So far, no treatment strategies are available, and progress in finding novel treatments is limited.5 In this issue of the Journal, Sartipy et al.6 describe the association between AF and all-cause mortality among patients with HFpEF, and the role of heart rate in that relationship. The findings were based on the Swedish Heart Failure Registry, and involves a sample of 9090 patients with clinician-diagnosed HFpEF. As expected, a higher heart rate, measured once at inclusion, was associated with higher risk of mortality in sinus rhythm patients, an association well-established for sinus rhythm patients with heart failure, and especially in heart failure with reduced ejection fraction (HFrEF). Patients with HFpEF and AF had higher risk of death than those in sinus rhythm, again confirming prior reports. Within the first few years of follow-up, higher heart rate was associated with worse outcomes in AF patients. This could be explained by the observational biases of treatment in sicker patients, and patients not being free of negative chronotropic medication when heart rate was measured (74% treated with beta-blockers, and in AF patients this rose to 83% with 29% using digoxin but with dosages unknown). In longer follow-up, outcomes were similar for patients with HFpEF and AF regardless of baseline heart rate, nicely summarized by Figure 3 in the paper. As demonstrated also for HFrEF,7 this goes against the conventional idea that HFpEF patients have reduced ventricular compliance and impaired diastolic filling, and as a consequence are more dependent on atrial contribution and diastolic filling which should benefit from lowering heart rate.8-10 However, the situation during AF may be different. Already in 1990 Rawles11 demonstrated in AF patients that a reduction of a rapid ventricular rate in AF does result in a longer diastolic filling period and a higher left ventricular stroke volume, but this effect is offset by reduced contractility and fewer beats per minute. It should be noted that accurate determination of stroke volume in AF is challenging, less reproducible and dependent on the degree of irregularity. So where does this leave us from a clinical standpoint? It is important to note that in patients with HFpEF there is no proven prognostic benefit from beta-blockers.12 Even in HFrEF, the beneficial effect of beta-blockers is no longer present with AF.13, 14 In AF patients it has been shown that reducing heart rate to strict rate control targets increases the risk of bradycardia and need for pacemakers, without improving prognosis.15 In the RACE II trial of permanent AF, with 49% HFpEF and HFrEF patients, it was demonstrated that a lenient rate controlling strategy was non-inferior to strict rate control for a composite clinical outcome.16 Current guidelines suggest that heart rate control in AF should be pursued to alleviate symptoms, with initial lenient control (<110 b.p.m.) acceptable regardless of heart failure status.17 The current study seems to support that approach in HFpEF patients. A number of limitations should be mentioned. With violation of proportional hazards in HFpEF with AF, we should be cautious not to over-interpret the findings of the study for these patients. Stratification by time period is a statistical ‘fix’ for this issue, but the resultant findings provide less power and limited generalizability to clinical practice. Further, only 5% of the cohort had complete case data for all covariates, hence the need for multiple imputation. Last, the methods of diagnosing AF and measuring heart rate are important to consider. Heart rate was measured in different and non-standardized ways by regular hospital staff. The method (e.g. palpation or electrocardiogram) and rhythm (AF or sinus rhythm during measurement) were not recorded. Pulse deficit often occurs as a consequence of high and irregular ventricular rate during AF, and hence radial pulse taking will underestimate heart rate. A single non-standardized measurement could also potentially be recorded at rest in some patients and after exertion in others. This has particular consequence for AF patients, since heart rate increases the most when starting exercise (whereas it is modest in sinus rhythm). The findings of this study underscore the need for more research in this field to better understand the combination of AF and HFpEF, and define treatment targets to reduce symptoms and improve prognosis. Whether heart rate during AF in HFpEF patients should be a therapeutic target is still an unanswered question. Drs. Rienstra and Van Veldhuisen are principal investigators of the VIP-HF study and acknowledge financial support for this study by an unrestricted grant of St. Jude Medical/Abbott to the University Medical Center Groningen. Dr. Kotecha is Chief Investigator for the RATE-AF trial (NCT02391337) and the Beta-blockers in Heart Failure Collaborative Group (NCT00832442); he acknowledges grants from the National Institute of Health Research (NIHR; CDF-2015-08-074); British Heart Foundation (PG/17/55/33087), EU Innovative Medicines Initiative Collaboration Grant (116074 [email protected]), Menarini and IRCCS San Raffaele. The opinions expressed are those of the authors and do not represent the NIHR or the UK Department of Health. Conflict of interest: none declared.