Un(b)locking Therapeutic Options: Potential for Calcium Channel Blockers in Heart Failure with Non-Reduced Left Ventricular Ejection Fraction

This article refers to ‘Calcium channel blocker use and outcomes in patients with heart failure and mildly reduced and preserved ejection fraction’ by S. Matsumoto et al., published in this issue on pages 2202–2214. Learnings from recent clinical trials has led to an emphasis on provision of evidence-based medicines to treat patients with heart failure with mildly reduced ejection fraction (HFmrEF) and preserved ejection fraction (HFpEF).1 Several of these therapies, such as sodium–glucose cotransporter 2 inhibitors and sacubitril/valsartan, had previously been approved for the treatment of heart failure with reduced ejection fraction (HFrEF) and were subsequently demonstrated to provide benefits for those with HFmrEF and HFpEF.2 However, this is not universal, such as in the case of beta-blockade, where the evidence base for treatment of HFpEF is very poor, heterogeneous, and contains no randomized trials that show clinical benefit. Despite this, it is known that use of beta-blockers typically exceeds 70% of patients with HFpEF. This can be explained by ‘indication creep’, where the association of beta-blocker efficacy in HFrEF is so strong that clinicians are willing to expand their use to the HFpEF population. As we will see, this may be a double-edged sword. The other side of this issue relates to the risk from therapies which have classically been contraindicated in patients with HFrEF such as calcium channel blockers (CCBs).3 The quality of evidence related to harm does not rise to the same high level of evidence demanded for efficacy. In the case of CCBs, guideline panels have relied upon reports publishing secondary endpoints and sub-group analyses of neutral trials,4-6 leading to a guilt by association. While it is reasonable to avoid heart rate lowering CCBs in deference to the powerful benefit of beta-blockade in HFrEF, this does not apply to HFpEF. Further, this latter group is characterized by a high prevalence of conditions for which CCB use is indicated such hypertension (present in >80% of cases) and stroke.7 In this issue of the Journal, Matsumoto et al.8 present a timely analysis from four randomized clinical trials evaluating the association of CCBs on outcomes in patients with either HFmrEF or HFpEF. The authors were able to extract patient-level data from over 16 000 patients and employed appropriate and rigorous means to mitigate bias related to imbalances between the two groups. The most important findings are that utilization of CCBs was associated with no difference in the combined outcome of cardiovascular death or heart failure hospitalization. There were no differences in other outcomes with CCB use apart from a lower risk of death due to pump failure and a higher risk of non-fatal stroke. This association was persistent across important sub-groups, seemed to be consistent across the spectrum of ejection fraction and type of CCB (rate-lowering versus non-rate-lowering). This is a particularly important finding for patients with HFpEF, who frequently have poorly controlled hypertension as a cause or contributor to their underlying disease.9 This suggests that physicians could consider utilizing dihydropyridine CCBs for managing these comorbidities in patients with HFmrEF and HFpEF, without concern for increasing the risk of heart failure-related outcomes. The limitations have been well outlined by the authors. Utilization of randomized controlled trial data is a dual-edged sword; the patients are well characterized but the applicability to general populations is limited. Secondly, CCB therapy was not randomized and while persistence was similar, several patient factors which would be expected to be associated with CCB use may explain the findings. For example, patients receiving CCBs were more likely to have a history of hypertension and stroke and tended to have higher blood pressure. Patients maintaining a higher blood pressure despite CCB therapy may have greater functional reserve,10 and therefore could be at lower risk of pump failure. This may not be fully accounted for in the statistical adjustments made by the authors, since CCB therapy would already be influencing baseline blood pressure. Additionally, other unmeasured confounders (such as frailty) may have accounted for differences between groups. As mentioned by the authors, the indication for using CCBs (hypertension or atrial fibrillation) may also help explain the observation of excess risk of stroke. Lastly, most patients were receiving dihydropyridine CCBs. Given the relatively low number of patients receiving non-dihydropyridines (728 or 12.4% of 5874 patients) it remains uncertain if these CCBs confer identical risk as their dihydropyridine counterparts. A larger, dedicated analysis would provide further insights. The study by Matsumoto et al. is timely in the context of drastic changes in our approach to managing patients with heart failure, regardless of left ventricular ejection fraction (LVEF). In patients with HFrEF, an increasing array of medicines call for simple yet thoughtful therapeutic approaches, irrespective of LVEF.11, 12 However, it is important to evaluate the potential influence of therapeutic combinations.13 Additionally, there are patient-specific features which may modulate the effects of medical therapy,14 which should be investigated to allow physicians to better personalize therapies.15 Now is also the time to reconsider traditional concepts regarding indicated and contraindicated therapies across the spectrum of LVEF. Matsumoto et al. have addressed an important knowledge gap, which may be used to inform a ‘heat map’ of medicine options for heart failure treatment where treatments are used to improve outcomes in heart failure, as well as those used to primarily treat comorbid conditions, rather than heart failure (Figure 1). We hope and expect this to be one of many future studies that will clarify the use of medications to treat comorbidities in patients with HFpEF. Conflict of interest: none declared.