Estimating the Lifetime Benefits of Treatments for Heart Failure

João Pedro Ferreira, MD, PHD; Kieran F. Docherty, MBCHB; Susan Stienen, MD, PHD; Pardeep S. Jhund, MBBCH, PHD; Brian L. Claggett, PHD; Scott D. Solomon, MD; Mark C. Petrie, MBCHB; John Gregson, PHD; Stuart J. Pocock, PHD; Faiez Zannad, MD, PHD; John J.V. McMurray, MD

Disclosures

JACC Heart Fail. 2020;8(12):984-995. 

In This Article

Results

Trials Analyzed

The overall results for each of the 4 trials analyzed are shown in Table 1 and Figure 1.

Figure 1.

RMST Using Age Instead of Time for the Trials Included
The restricted mean survival time (RMST) is expressed in potential years gained without an event from 60 to 80 years of age. The number of patients at risk represents the number of patients who were enrolled in the trial(s) in whom the event of interest had not occurred at a given age. A longer potential life extension (area under the Kaplan-Meier curve) was observed for patients at lower risk. Because they were less likely to have events or dying during follow-up, their potential life extension was improved with treatment. (A) PARADIGM-HF (Angiotensin–Neprilysin Inhibition versus Enalapril in Heart Failure): potential years gained without event were observed across all the age groups in patients treated with sacubitril/valsartan (vs. enalapril), especially for the composite outcome of cardiovascular death (CVD) or heart failure hospitalization (HFH). (B) RALES (The Effect of Spironolactone on Morbidity and Mortality in Patients with Severe Heart Failure): high-risk population, the potential years gained without event in favor of spironolactone (vs. placebo) were observed especially at younger ages within the trial. (C) EMPHASIS-HF (Eplerenone in Patients with Systolic Heart Failure and Mild Symptoms): a lower risk population than that in the RALES trial with a treatment effect of eplerenone (vs. placebo) on life extension that was observed across all the ages (60 to 80 years) within the trial. (D) DIG (The Effect of Digoxin on Mortality and Morbidity in Patients with Heart Failure): digoxin extended the time free of HFH but did not prolong life. ACM = all-cause mortality.

PARAGIGM-HF

Overall, in the PARADIGM-HF trial, 25.8% (13.2 per 100 patient-years) of the patients in the enalapril group versus 20.0% (10.6 per 100 patient-years) of the patients in the sacubitril and/or valsartan group experienced the primary composite outcome (HF hospitalization or cardiovascular death) at 3 years of follow-up, which gave an HR of 0.80 (95% CI: 0.73 to 0.88), and a NNT of 25 (18 to 43). The RMST days gained over this follow-up was +37 (23 to 52) days, and the potential extension of life without an event was estimated at +1.5 (0.7 to 2.4) years. The findings for all-cause death and an increase in overall survival are shown in Table 1 and Figure 1.

RALES

In the RALES study, 60.9% (45.5 per 100 patient-years) of the patients in the placebo group versus 46.4% (29.3 per 100 patient-years) of the patients in the spironolactone group experienced the primary outcome over 3 years of follow-up, with a corresponding HR of 0.67 (95% CI: 0.59 to 0.77) and NNT of 9 (7 to 14). The RMST days gained during follow-up were +119 (78 to 160) days and an extension of event-free survival of +1.1 (−0.1 to +2.3) years. The findings for all-cause death and increase in overall survival are shown in Table 1 and Figure 1.

EMPHASIS-HF

In the EMPHASIS-HF trial, 25.4% (16.7 per 100 patient-years) of the patients in the placebo group versus 17.3% (10.5 per 100 patient-years) of the patients in the eplerenone group experienced the primary outcome at 3 years of follow-up, which gave an HR of 0.64 (95% CI: 0.54 to 0.75) and a NNT of 14 (10 to 22). The RMST days gained were +78 (51 to 106) days and an extension of event-free survival of +2.9 (1.2 to 4.5) years. The findings for all-cause death and increase in overall survival are shown in Table 1 and Figure 1.

DIG

In the DIG trial, 44.5% (22.0 per 100 patient-years) of the patients in the placebo group versus 39.3% (18.0 per 100 patient-years) of the patients in the digoxin group experienced the primary outcome over 3 years of follow-up; the HR was 0.82 (95% CI: 0.76 to 0.88) in favor of digoxin, and the NNT was 20 (15 to 32). Patients who took digoxin gained +63 (44 to 82) days of follow-up and +0.8 (0.2 to 1.5) years of event-free survival. Digoxin did not reduce all-cause death because there was no increase in event-free survival (Table 1).

Subgroups in PARADIGM-HF

The results for each of the 3 subgroups analyzed are shown in Table 2, Figure 2, and Supplemental Table 1.

Figure 2.

PARADIGM-HF: RMST Using Age Instead of Time in Subgroups Reflecting Patient Risk and Age
The number of patients at risk represents the number of patients who were enrolled in the trial(s) in whom the event of interest had not occurred at a given age. A longer potential life extension (area under the Kaplan-Meier curve) was observed for patients with lower N-terminal pro–B-type natriuretic peptide (NT-proBNP) levels, who were in New York Heart Association (NYHA) functional class I and/or II, and were younger than 65 years, which suggested that the long-term benefits of sacubitril/valsartan were particularly important in less symptomatic and younger patients (i.e., patients with lower risk). Abbreviations as in Figure 1.

Lower Baseline NT-proBNP Level Versus Higher Baseline NT-proBNP Level. Low NT-proBNP was defined as <1,000 pg/ml and high NT-proBNP as ≥1,000 pg/ml (Table 2). The analysis was also performed using the trial median NT-pro BNP of 1, 615 pg/ml (Supplemental Table 1).

Among patients with a NT-pro BNP <1,000 pg/ml, the levels were 680 pg/ml (quartile 1 to quartile 3 [Q1 to Q3]: 527 to 827 pg/ml), and the mean ± SD age was 61.8 ± 11.2 years. The primary outcome was experienced in 16.8% (7.6 per 100 patient-years) of those in the enalapril group and 12.6% (5.6 per 100 patient-years) in the sacubitril/valsartan group over 3 years of follow-up, which gave a HR of 0.73 (95% CI: 0.59 to 0.91), a NNT of 27 (16 to 83), and +26 (6 to 47) days of RSMT gained during follow-up. The estimated extension of event-free survival was +2.3 (0.6 to 4.0) years.

Among patients with a NT-proBNP ≥1,000 pg/ml, the levels were 2,330 pg/ml (Q1 to Q3: 1,512 to 4,361 pg/ml), and mean age was 64.6 ± 11.4 years. Among these patients, the event rates were higher (30.7%; 16.0 per 100 patient-years in the enalapril group vs. 25.7%; 12.9 per 100 patient-years in the sacubitril/valsartan group), and the relative risk reduction was smaller (HR: 0.81; 95% CI: 0.73 to 0.89), but there was a lower NNT of 24 (16 to 44). RMST days gained during the follow-up were similar +42 (24 to 61) days, and the estimated extension of event-free survival was smaller, at +1.4 (0.4 to 2.4) years.

Among patients with NT-proBNP below the median, the levels were 888 pg/ml (Q1 to Q3: 642 to 1,201 pg/ml), and the mean age was 62.8 ± 11.1 years. These patients experienced the primary outcome in 18.2% (8.8 per 100 patient-years) of those in the enalapril group and in 13.9% (6.5 per 100 patient-years) in the sacubitril/valsartan group over 3 years of follow-up, which gave an HR of 0.74 (95% CI: 0.63 to 0.86), a NNT of 24 (16 to 48), and +34 (17 to 51) days of RSMT gained during follow-up. The estimated extension of event-free survival was +2.4 (1.1 to 3.8) years.

Among patients with a NT-pro BNP above the median, the levels (Q1, Q3) were 3,231 (2,186, 5,593) pg/ml; and mean (SD) age was 64.8 (11.6) years. Among these individuals, the event rates were higher (33.4%, 18.4 per 100 patient-years in the enalapril group vs. 28.7%, 15.2 per 100 patient-years in the sacubitril/valsartan group). The relative risk reduction was smaller (HR: 0.83; 95% CI: 0.75 to 0.93), but there was a similar NNT of 25 (16 to 61). RMST days gained during the follow-up were similar at +41 (18 to 63) days, and the estimated extension of event-free survival was smaller 0.9 (−0.1 to 2.0) years.

Patients in NYHA Functional Class I/II Versus NYHA Functional Class III/IV. Among individuals in NYHA functional class I and/or II (mean age: 63.1 ± 11.5 years), 30.2% (12.2 per 100 patient-years) of those in the enalapril group and 23.4% (9.2 per 100 patient-years) of patients in the sacubitril/valsartan group experienced the primary outcome over 3 years of follow-up, which gave an HR of 0.75 (95% CI: 0.68 to 0.84), a NNT of 21 (15 to 34), and +42 (26 to 58) days of RSMT gained during follow-up. The estimated extension of event-free survival was +1.9 (0.8 to 2.9) years.

In patients in NYHA functional class III and/or IV (mean age: 66.0 ± 10.6 years), the event rates were higher (36.9%, 16.2 per 100 patient-years in the enalapril group vs 36.7%, 15.5 per 100 patient-years in the sacubitril/valsartan group), and the relative risk reduction was smaller (HR: 0.93; 95% CI: 0.80 to 1.09). The RMST days gained during the follow-up were 19 (−12 to 49) days, and the estimated extension of event-free survival was 0.6 (−1.0 to 2.2) years, both of which were nonsignificant and of smaller magnitude than in the NYHA functional class I and/or II group.

Younger Patients Versus Older Patients

Among individuals aged younger than 65 years (mean age: 54.9 ± 8.0 years), 25.1% (13.0 per 100 patient-years) of those in the enalapril group and 20.0% (10.0 per 100 patient-years) of patients in the sacubitril/valsartan group experienced the primary outcome over 3 years of follow-up, which gave an HR of 0.77 (95% CI: 0.68 to 0.88), a NNT of 23 (15 to 44) and +40 (21 to 61 days) of RSMT gained during follow-up. The estimated extension of event-free survival was +1.7 (0.6 to 2.8) years.

In patients aged 65 years or older (mean age: 73.0 ± 5.7 years), the event rates were higher (26.6%; 13.5 per 100 patient-years in the enalapril group vs. 22.7%; 11.2 per 100 patient-years in the sacubitril/valsartan group), the relative risk reduction was smaller (HR: 0.83; 95% CI: 0.73 to 0.94), and the NNT was larger at 29 (17 to 83). RMST days gained during the follow-up was similar +33 (13 to 54) days, and the estimated extension of event-free survival was smaller +0.9 (0.2 to 1.6) years.

The summary of the main characteristics of the RMST compared with absolute and relative risk metrics is provided in the Central Illustration.

Central Illustration.

Comparison of Age-Based Lifetime Estimates Using the Restricted Mean Survival Time With Conventional Measures of Risk Reduction
The restricted mean survival time (RMST) using age instead of time allows the estimation of long-term, event-free survival, which is a clinically meaningful metric for both the clinicians and patients. The projections of long-term survival may be particularly useful for explaining the potential long-term benefits of treatments to less symptomatic/lower risk patients. In contrast, the absolute risk reduction may be less pronounced in lower risk patients, which may discourage them from taking additional therapies that could substantially increase the long-term event-free time. DIG = The Effect of Digoxin on Mortality and Morbidity in Patients with Heart Failure; EMPHASIS-HF = Eplerenone in Patients with Systolic Heart Failure and Mild Symptoms; Eple. = eplerenone; HF = heart failure; HR = hazard ratio; PARADIGM-HF = Angiotensin–Neprilysin Inhibition versus Enalapril in Heart Failure; RALES = The Effect of Spironolactone on Morbidity and Mortality in Patients with Severe Heart Failure Sac./Val. = sacubitril/valsartan; Spiro. = spironolactone.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....