Key Points
Question Does dapagliflozin reduce the risk of total episodes of worsening heart failure (HF; defined as hospitalization for HF or urgent HF visit requiring intravenous HF therapies) and cardiovascular death in patients with mildly reduced or preserved ejection fraction heart failure?
Findings In this prespecified analysis of the DELIVER trial including 6263 patients, dapagliflozin reduced the risk of total HF events and cardiovascular death by 23%, and this was consistent across a range of subgroups, including across the spectrum of ejection fraction.
Meaning In this study, dapagliflozin demonstrated no reduction in efficacy in reducing second or subsequent HF events.
Abstract
Importance In the Dapagliflozin Evaluation to Improve the Lives of Patients With Preserved Ejection Fraction Heart Failure (DELIVER) trial, dapagliflozin reduced the risk of time to first worsening heart failure (HF) event or cardiovascular death in patients with HF with mildly reduced or preserved ejection fraction (EF).
Objective To evaluate the effect of dapagliflozin on total (ie, first and recurrent) HF events and cardiovascular death in this population.
Design, Setting, and Participants In this prespecified analysis of the DELIVER trial, the proportional rates approach of Lin, Wei, Yang, and Ying (LWYY) and a joint frailty model were used to examine the effect of dapagliflozin on total HF events and cardiovascular death. Several subgroups were examined to test for heterogeneity in the effect of dapagliflozin, including left ventricular EF. Participants were enrolled from August 2018 to December 2020, and data were analyzed from August to October 2022.
Interventions Dapagliflozin, 10 mg, once daily or matching placebo.
Main Outcomes and Measures The outcome was total episodes of worsening HF (hospitalization for HF or urgent HF visit requiring intravenous HF therapies) and cardiovascular death.
Results Of 6263 included patients, 2747 (43.9%) were women, and the mean (SD) age was 71.7 (9.6) years. There were 1057 HF events and cardiovascular deaths in the placebo group compared with 815 in the dapagliflozin group. Patients with more HF events had features of more severe HF, such as higher N-terminal pro–B-type natriuretic peptide level, worse kidney function, more prior HF hospitalizations, and longer duration of HF, although EF was similar to those with no HF events. In the LWYY model, the rate ratio for total HF events and cardiovascular death for dapagliflozin compared with placebo was 0.77 (95% CI, 0.67-0.89; P < .001) compared with a hazard ratio of 0.82 (95% CI, 0.73-0.92; P < .001) in a traditional time to first event analysis. In the joint frailty model, the rate ratio was 0.72 (95% CI, 0.65-0.81; P < .001) for total HF events and 0.87 (95% CI, 0.72-1.05; P = .14) for cardiovascular death. The results were similar for total HF hospitalizations (without urgent HF visits) and cardiovascular death and in all subgroups, including those defined by EF.
Conclusions and Relevance In the DELIVER trial, dapagliflozin reduced the rate of total HF events (first and subsequent HF hospitalizations and urgent HF visits) and cardiovascular death regardless of patient characteristics, including EF.
Patients with heart failure (HF) are frequently hospitalized for decompensation of HF. While the risk of death declines as ejection fraction (EF) increases, the risk of hospitalization for HF remains relatively static across the spectrum of EF.1 Therefore, repeated hospitalizations account for a greater proportion of the burden of disease in patients with HF with mildly reduced EF (HFmrEF) or HF with preserved EF (HFpEF) compared with HF with reduced EF (HFrEF). These repeated hospitalizations are the major driver of the burden of HF on patients and health care systems. In HFmrEF and HFpEF, as with HFrEF, these repeated hospitalizations are also associated with a higher subsequent risk of death.2 The gradient of risk is linear; as the number of repeated hospitalizations increases, the subsequent risk of both cardiovascular and all-cause mortality also increases.3
Recognizing the importance of repeated hospitalizations in patients with HF, analysis of repeated or total hospitalizations for HF was the primary outcome in a trial of sacubitril/valsartan in patients with HFmrEF or HFpEF.4 More recently, there has also been a recognition that urgent visits for treatment for HF are associated with worse outcomes, and these events have been incorporated into time to first event composites along with HF hospitalizations.5–7 Trials of the sodium-glucose cotransporter 2 (SGLT2) inhibitor dapagliflozin were designed with a primary outcome of time to first worsening HF event (first hospitalization for HF or urgent HF visit) or cardiovascular death, in recognition of the prognostic impact of both of these nonfatal events.8,9 However, to our knowledge, the trials of SGLT2 inhibitors in HF still only examine the total number of hospitalizations for HF as a secondary outcome7,10–12 and not the effect of treatment on the total burden of this condition reflected by the full spectrum of HF events from urgent visits through to cardiovascular death. In this prespecified analysis, we describe in detail the efficacy of dapagliflozin on total HF events, ie, first and subsequent HF hospitalizations or urgent visits for HF and cardiovascular deaths, in the population with HFmrEF or HFpEF enrolled in the Dapagliflozin Evaluation to Improve the Lives of Patients With Preserved Ejection Fraction Heart Failure (DELIVER) trial.
In patients with HFmrEF or HFpEF, dapagliflozin reduced the risk of total HF events, ie, repeated in addition to first events. This benefit was observed in all the prespecified DELIVER subgroups and across the spectrum of EF. The characteristics associated with multiple HF events in this population with HFmrEF or HFpEF were similar to those in patients with HFrEF experiencing multiple hospitalizations.18
The reduction in burden of total HF events with dapagliflozin was evident regardless of the method used to analyze the total events and whether we examined total HF events including urgent HF visits or HF hospitalizations without urgent visits. The point estimates were more favorable to that obtained in the time to first event analysis, which was the primary outcome of the DELIVER trial,10 ie, dapagliflozin demonstrated no reduction in efficacy in reducing second or subsequent events. The estimates were also consistent with other trials of SLGT2 inhibitors in patients with HFmrEF HFpEF. In the Empagliflozin Outcome Trial in Patients With Chronic Heart Failure With Preserved Ejection Fraction (EMPEROR-Preserved) trial,12 the SGLT2 inhibitor empagliflozin reduced the rate of total HF hospitalizations by 27% (rate ratio, 0.73; 95% CI, 0.61-0.88; P < .001) using the joint frailty approach (compared with 0.71 [95% CI, 0.63-0.80; P < .001] using the same outcome and model in the DELIVER trial).10 In a separate analysis of the DELIVER trial, we also examined if this reduction in total HF events was also observed for all-cause hospitalizations and found that there was a similar but smaller relative risk reduction of 11%.19 The observation that the benefits of dapagliflozin were consistent across the range of EF is important, as an earlier pooled analysis of total HF hospitalizations in the EMPEROR-Reduced and EMPEROR-Preserved trials reported that the effect of empagliflozin on total HF hospitalizations appeared to diminish at higher EFs.20 We did not see any evidence of an attenuation of the benefit of dapagliflozin on total HF events at higher EFs in this analysis or for total HF hospitalizations in our pooled analysis.21 We also found that the elevated risk for subsequent death in a patient whose first event was an urgent HF visit, compared with a patient who did not have any worsening HF event, was elevated, in keeping with reports from HFrEF trials.5,22 Furthermore, we found that although urgent HF visits were associated with a higher risk of death, the excess risk was not as high as in patients in whom the first event experienced during follow-up was an HF hospitalization, in keeping with the findings of a similar analysis in the Prospective Comparison of ARNI With ARB Global Outcomes in HF With Preserved Ejection Fraction (PARAGON-HF) trial.6
As most prior trials enrolling patients with an EF greater than 40% have been neutral, there are few data with which to compare the relative efficacy of SGLT2 inhibitors in reducing total HF events. A post hoc analysis of the Candesartan in Heart Failure Assessment of Reduction in Mortality and Morbidity–Preserved (CHARM-Preserved) trial, which enrolled patients with an EF greater than 40%, suggested that candesartan reduced total HF hospitalizations and cardiovascular death using a negative binomial model.23 In that analysis, candesartan reduced total HF hospitalizations by 25%, ie, the rate ratio for total HF hospitalizations and cardiovascular death was 0.75 (95% CI, 0.62-0.91; P = .003). This post hoc analysis appeared to provide enough power to detect a treatment effect that was not evident in a time to first event analysis. In theory, total events should require a smaller sample size to demonstrate a treatment effect, as not only are subsequent nonfatal HF events counted but cardiovascular deaths that occur after these events are also counted (whereas both are ignored after a first HF event in a traditional time to first event analysis). In our trial, these repeated events contributed a further 193 deaths and 557 HF events that would otherwise have been ignored. Consequently, power calculations in the setting of total events are more complex, and factors such as heterogeneity of patient risk have to be incorporated, which is not currently part of routine sample size estimation strategies.24 However, use of total HF events as a primary outcome may result in a smaller sample size than is needed for a time to first event primary end point (or provide more power for secondary total events end points in trials powered for a time to first event primary outcome).
One attraction of recurrent or total events analysis is that they describe the full burden of disease and are potentially more meaningful to patients, representing the full disease experience. Therefore, describing reductions in total events may be helpful to explain treatment effects to patients. Explaining treatment efficacy is difficult in a clinical setting, and it is well known that relative risks are poorly understood by patients and by some clinicians. Other methods of expressing treatment benefits, such as the number needed to treat, are equally, if not more difficult, in the setting of total events.25 We reported that the AUC ratio for dapagliflozin vs placebo was 0.72, not dissimilar to the estimates from the conventional model-based approaches that we had prespecified. Although the relative risk reduction can be described as a ratio, perhaps more usefully the absolute risk reduction of 2.2 months over a 3-year period is easily explained to clinicians and patients, ie, a gain of 2.2 months of event-free survival. This absolute risk reduction with the accompanying time scale is a directly interpretably into clinically relevant terms for the patient. Simulation studies of power calculations using this approach suggest that sample sizes based on time to first events may be 20% larger than samples based on an AUC.17 The results that we observed using the AUC method (which is an extension of the restricted mean survival time used in multiple disease areas26–29) were consistent with the more traditional model-based approaches to analyzing total events. The technique may be useful in situations where model assumptions are not met, as the approach does not require a statistical model to be constructed and could be used in addition to other metrics, such as days alive and out of hospital.
In summary, among patients with HFmrEF or HFpEF, dapagliflozin reduced the risk of total (first and recurrent) HF events or cardiovascular deaths compared with placebo. HF events are common and preventable, and the efficacy of dapagliflozin in reducing the number of these events is consistent across a broad range of subgroups and across the spectrum of EF.
ARYAN'S BLOG 