乌鸦传媒

The patients who participated in the RELAX-AHF trial of serelaxin for acute heart failure were not representative of the larger heart failure population, researchers found.

Only 20.7% of patients from a real-world U.S. registry and 16.2% of those from a similar international registry , according to , of the Duke Clinical Research Institute, and colleagues.

Notably, only 60% of the U.S. patients and 51% of the international patients had a systolic blood pressure greater than 125 mm Hg, one of the major factors used to select patients for the trial, the researchers reported online in Circulation: Cardiovascular Quality and Outcomes.

And the patients in the registries who would have qualified for showed numerous differences in clinical characteristics and outcomes -- including a lower rate of in-hospital mortality -- compared with those who would not have qualified.

"Although the RELAX-AHF trial demonstrated the efficacy of a novel therapy, serelaxin, in acute heart failure, future studies will be necessary to evaluate the extent of benefit for the remainder of the acute heart failure population," Hernandez and colleagues wrote.

"The results of RELAX-AHF will undoubtedly require replication before broader use," they added.

These issues may be considered when later this month.

"If serelaxin is approved for use in acute heart failure," the researchers noted, "it will be important to recognize the population studied (in addition to those who derive the greatest benefit) and consider how to appropriately use the treatment in that population."

The main findings of the trial -- presented at the 2012 meeting of the American Heart Association -- showed that patients admitted with acute heart failure had significant improvements on one of the two primary measures of dyspnea, as well as reductions in cardiovascular and all-cause death, when they took serelaxin versus placebo.

But like other randomized controlled trials, RELAX-AHF had a highly selected patient population, raising questions about the extent to which the results could be applied to the wider heart failure population. The generalizability issue has particular importance in acute heart failure "given the extent of heterogeneity based on the marked variation in presenting symptoms, clinical characteristics, disease severity, and outcomes," according to the researchers.

To assess how representative the trial population was, Hernandez and colleagues examined data from 196,770 acute heart failure admissions included in the Acute Decompensated Heart Failure National Registry (ADHERE) for the U.S. and for Latin America and the Asia-Pacific region; 95% of the admissions were in the U.S.

For the current analysis, patients were considered to be like those in RELAX-AHF if they met all of the following major criteria:

• A discharge diagnosis of heart failure
• Systolic blood pressure greater than 125 mm Hg
• Dyspnea at rest or with mild exertion
• Intravenous diuretic use
• Glomerular filtration rate of 30 to 75 mL/min/1.73 m²
• Hemoglobin greater than 8 g/dL
• No use of IV inotropes or vasopressors

Only about one in every five patients in the registries would have qualified for RELAX-AHF using those criteria. And those who would have been candidates for the trial differed significantly from the rest of the heart failure population.

Those meeting all of the criteria were older, had higher systolic blood pressure and ejection fraction, and were more likely to be female and to have a prior history of hypertension and better renal function. Anemia, chronic renal insufficiency, chronic hemodialysis, and use of implantable cardioverter-defibrillators were all less frequent in the RELAX-AHF-like group.

That group also was given fewer interventions -- including cardiac catheterization and mechanical ventilation -- in the hospital and was more likely to be asymptomatic and to receive prescriptions for ACE inhibitors/angiotensin receptor blockers at discharge.

In addition, the RELAX-AHF-like patients fared better in the hospital with a lower rate of mortality both among U.S. patients (1.6% versus 4.4%) and international patients (1% versus 5.7%). The hazard ratio for the combined cohorts was 0.59 (95% CI 0.53-0.66).

The authors acknowledged some limitations of the study, including the ability to use only the major entry criteria for the trial in the analysis, the uncertain generalizability of the findings to European cohorts or to centers not included in the registries, the difference in time periods of enrollment in the registries and RELAX-AHF, and the possibility of residual confounding.

From the American Heart Association:

Comment