Poor diuretic response associated with worse clinical outcomes

These novel data from the RELAX-AHF trial were presented by lead author Professor Adriaan A. Voors at the Heart Failure Congress 2014, held 17-20 May in Athens, Greece. The Congress is the main annual meeting of the Heart Failure Association of the European Society of Cardiology.

RELAX-AHF was a double blind, placebo-controlled trial of 1161 patients admitted to hospital with acute heart failure. Patients were randomized to receive 48-hour infusions of placebo or serelaxin (30µg/kg per day) within 16 hours from presentation. Primary analysis of the study showed that serelaxin reduced dyspnoea and decreased 180 day mortality.

The current substudy, presented for the first time today, had two aims. The first was to confirm preliminary findings of the PROTECT trial, which found that poor diuretic response is a serious clinical problem.2 The second was to investigate whether the beneficial effects of serelaxin discovered in the primary analysis were related to improvement of the diuretic response.

Diuretic response was defined as kilograms of weight loss per 40mg of the loop diuretic furosemide. Professor Voors said: "Congestion is the main problem in patients with acute decompensated heart failure and we give them loop diuretics to get rid of the excess fluid. Patients who respond to the loop diuretic have diuresis and lose weight. But a substantial proportion do not respond to the diuretic, do not diurese and do not lose weight."

He added: "Diuretic response could be measured by urinary output but when patients no longer need a catheter it is difficult to capture how much fluid they have lost. We therefore used weight loss as a more objective and reliable measurement of diuretic response. The more weight patients lost per mg of furosemide given, the better their diuretic response."

The study found that diuretic response was an important clinical problem, confirming preliminary results from the PROTECT trial. Patients with a poor diuretic response had less dyspnoea relief (p=0.0001) and a higher risk of cardiovascular death or rehospitalisation for heart failure or renal failure through day 60 (p<0.004). Poor diuretic response was not associated with increased 180-day cardiovascular mortality (p=0.546).

Professor Voors said: "The most important message from this substudy is that if patients are hospitalised for acute heart failure and do not respond well to their diuretic then they are in trouble. Their recovery in hospital is likely to be worse and when they are sent home there is a good chance that they will be rehospitalised again or even die."

The next question was whether serelaxin influenced patients' response to diuretics. The researchers found that serelaxin had no effect on diuretic response. Patients in the serelaxin group improved clinically and were given fewer diuretics but they lost less weight than the placebo group.

Professor Voors said: "Patients felt better with serelaxin and so physicians gave them less diuretics. But these patients did not diurese more, so the overall effect of serelaxin on diuretic response was neutral."

He added: "We will continue to look for the mechanisms by which serelaxin exerts its positive effects. The current study shows that the beneficial effects of serelaxin are not through improvement of diuretic response. We have already shown that serelaxin prevents the organ damage that occurs during an episode of acute heart failure3 and we will continue to investigate other modes of action."

Professor Voors concluded: "The major finding from this substudy of the RELAX-AHF trial is that diuretic resistance is a clinical problem that leads to worse in-hospital and early post-discharge clinical outcomes in patients with acute heart failure. More insights are needed into the problem of poor diuretic response so that better therapies can be found to improve the diuretic response and improve clinical outcomes."