Antigen carbohydrate 125 and creatinine on admission for prediction of renal function response following loop diuretic administration in acute heart failure

Abstract

BackgroundThe use of loop diuretics in acute heart failure (AHF) is largely empirical and has been associated with renal function impairment by reducing renal perfusion but also renal improvement by decreasing renal venous congestion. Antigen carbohydrate 125 (CA125) has emerged as a proxy for fluid overload. We sought to evaluate whether the early changes in creatinine (ΔCr) induced by intravenous furosemide doses (ivFD) differ among clinical groups defined by overload status (CA125) and creatinine on admission (Cr). Methods and resultsWe included 526 consecutive patients admitted for AHF. All patients received intravenous furosemide for the first 48hours. CA125 and Cr were dichotomized at 35 U/ml and 1.4mg/dl, respectively, and grouped as follows: C1 [Cr <1.4, CA125 ≤35 (n=151)]; C2 [Cr <1.4, CA125 >35 (n=241)]; C3 [Cr ≥1.4, CA125 ≤35 (n=45)]; and C4 [Cr ≥1.4, CA125 >35 (n=89)]. Clinicians in charge of the management of patients were blind to CA125 values. ΔCr was estimated as the absolute difference in Cr between admission and 48–72 hours. Multivariable linear regression analysis was used for modeling purposes. The adjusted analysis showed a differential effect of ivFD on ΔCr. Per increase in 20mg/day of ivFD, the mean ΔCr was 0.010mg/dl (p=0.464) in C1, 0.002mg/dl (p=0.831) in C2, 0.045mg/dl (p=0.032) in C3, and −0.045mg/dl (p<0.001) in C4 (omnibus p<0.001). A similar pattern of response was observed in a validation cohort. ConclusionsIn patients with AHF, the magnitude and direction of ΔCr attributable to ivFD were differentially associated with values of CA125 and Cr on admission.