Acute declines in kidney function with thiazide plus loop diuretics vs loop diuretics alone in acute decompensated heart failure
W Mccallum, H Tighiouart, M Tuttle, T Oka, T Kaur, J Casado, J Morales-Rull, F Formiga, L Manzano, J Testani, M J Sarnak, J C TrullasAbstract
Background
The association of acute declines in kidney function with mortality in patients with acute decompensated heart failure (ADHF), particularly in the setting of thiazide plus loop diuretic therapy has not been evaluated. Our objective was to examine whether acute declines in estimated glomerular filtration rate (eGFR), and the timing of these declines are associated with mortality and a composite outcome of mortality or HF hospitalization, and whether these relations differ by randomization status in the Combination of Loop with Thiazide-type Diuretics in Patients with Decompensated Heart Failure (CLOROTIC) Trial.
Methods
Patients admitted for ADHF were enrolled into the CLOROTIC trial and randomized to treatment with thiazide vs placebo (plus standard loop therapy) for 5-days. The exposure was %eGFR change at 2-days and 4-days after randomization. Multivariable Cox regression models were used to evaluate the association between % eGFR change and a primary outcome of mortality and secondary outcome of composite of mortality or HF hospitalization.
Results
Randomization to thiazide was associated with greater degree of eGFR decline at 2-days, with median %eGFR change of -9.7% (IQR -22.1, 5.4) vs -0.3% (-7.5, 10.1) in the thiazide (n=111) and placebo (n=114) arms, respectively. Changes were similar at 4-days, with median %eGFR change of -14.4% (-25.1, 7.8) and 0.0% (-10.0, 16.2) in the thiazide (n=108) and placebo (n=110) arms, respectively. Over a median follow-up of 3 months, 41 (18%) patients died and 98 (44%) either died or were hospitalized for HF. The eGFR decline at 2-days after randomization to either thiazide or placebo was not associated with risk of mortality (HR=0.96 [95% CI 0.47, 1.97] and HR=0.89 [0.37, 2.10] per 30% eGFR decline, respectively). The eGFR decline at 4-days after randomization was not associated with risk of mortality in the thiazide arm (HR=0.86 [0.48, 1.55] per 30% eGFR decline) but showed a trend toward increased risk in the placebo arm (HR=1.63 [0.82, 3.26] per 30% eGFR decline, p-interaction=0.19). Associations were similar for the composite outcome.
Conclusions
Among patients admitted for ADHF randomized to thiazide vs placebo plus loop diuretic therapy, early acute declines in eGFR particularly among those randomized to thiazide had no association with increased risk of mortality or the composite outcome.For image description, please refer to the figure legend and surrounding text.