Effect Modification by Clinical Decongestion on the Association Between Worsening Renal Function and Mortality in Patients with Acute Decompensated Heart Failure

Abstract

Introduction Worsening renal function (WRF) is common among patients with acute decompensated heart failure (ADHF) and is associated with increased risk of mortality. Recent data suggest, however, that decongestion may serve as an effect modifier on the association between WRF and mortality. Hypothesis We hypothesized that in patients with ADHF, achievement of concomitant decongestion would abolish the signal for harm associated with WRF. Methods We searched in PubMed, EMBASE and the Cochrane Library from inception to March 2019 for studies that assessed signs of decongestion via clinical and laboratory parameters in patients with WRF during ADHF admission and performed stratified analyses for mortality by the presence or absence of decongestion and WRF. Results Among 8 studies that included 4,864 patients with ADHF, 1,435 (29.5%) had WRF, with 50.5% of those with WRF experiencing decongestion. Unstratified, patients with WRF vs. no WRF had a higher risk for mortality (pooled odds ratio [OR] 1.63, 95% CI 1.41 to 1.88; I2 = 0%) (Figure 1 upper forest plot). However, when patients with WRF were stratified by the decongestion status, there was evidence for interaction. Compared to patients without WRF, patients with WRF that had features consistent with decongestion, had similar risk of mortality (pooled OR 1.09, 95% CI 0.90 to 1.34; I2 = 0%) (Figure 1 lower forest plot). Finally, the pooled mortality was lower for patients with WRF that achieved decongestion (26.6%) than for patients without WRF that did not achieve decongestion (51.8%) (Figure 1 lower panel). Conclusions Decongestion is a powerful effect modifier that nullifies harmful associations of WRF with mortality, and patients that experience decongestion with WRF fare better than those that do not achieve decongestion with stable kidney function. Future studies should not assess WRF in isolation as an endpoint without concomitant assessment of the clinical status that accompanied WRF.