Detection and prognostic impact of renal dysfunction in patients with chronic heart failure and normal serum creatinine

Abstract Background Accurate identification of renal dysfunction (RD) is crucial to risk stratification in chronic heart failure (CHF). Patients with CHF are at special risk of having RD despite normal serum creatinine (SCr), owing to a decreased Cr generation. At low levels of SCr, the equations estimating renal function are less accurate. This study was aimed to assess and compare the prognostic value of formulas estimating renal function in CHF patients with normal SCr. Methods We studied 462 patients with systolic CHF and normal SCr. Creatinine clearance was estimated by the Cockcroft–Gault (eCrCl) and glomerular filtration rate by the 4-variable MDRD equation (eGFR); eCrCl normalized for body-surface area (eCrCl BSA ) was calculated. The primary outcome was all-cause mortality at 2years. Results Seventy five patients died. At multivariate Cox regression analysis, only eCrCl BSA was significantly associated with mortality ( p =0.006); eGFR ( p =0.24), eCrCl ( p =0.09) and BUN ( p =0.14) were not statistically significant predictors. The patients in the lowest eCrCl BSA quartile had an adjusted 2.1-fold (CI: 1.06–4.1) increased risk of mortality, compared with those in the referent quartile. Two-year survival was 70.4% in the lowest eCrCl BSA quartile and 89.7% in the referent quartile. Other independent predictors of mortality were ischemic etiology (RR: 2.16 [CI: 1.3–3.5], p =0.0017), NYHA III/IV class (RR: 2.45 [CI: 1.51–3.97], p =0.0003), LVEF p =0.014), and anemia (RR: 1.86 [CI: 1.16–2.99], p =0.009). Conclusions A sizeable proportion of CHF patients have prognostically significant RD despite normal SCr. Such patients represent a high-risk subgroup and can more accurately be identified by the CG formula corrected for BSA than the MDRD.