Impact of renin-angiotensin-aldosterone blockade by angiotensin-converting enzyme inhibitors or AT-1 blockers on frequency of contrast medium-induced nephropathy: A post-hoc analysis from the Dialysis-versus-Diuresis (DVD) trial

Abstract

Background. After exposure to contrast medium (CM), about 10% of patients will develop contrast medium-induced nephropathy (CIN), with severe consequences for their prognosis. Although numerous studies evaluated risk factors for CIN development, it is still a matter of debate whether treatment with angiotensin-converting enzyme inhibitors (ACE-I) or AT-1 blockers increases the frequency of CIN after exposure to CM or not.Methods. We performed a prospective, single-centre study (January 2001-July 2004) to compare different treatments for CIN prevention. Creatinine levels within 72 h after CM application and in-hospital outcomes were documented. The impact of RAAS blockade on the frequency of CIN was assessed retrospectively.Results. Four hundred twelve patients were included (83.5% men, 29.1% diabetes mellitus, 74.6% hypertension). Of these, 269 patients (65.3%) were taking ACE-I (n = 236) or AT-1 blockers (n = 33). There were no significant differences in mean age (P = 0.075), creatinine levels (P = 0.113), gender (P = 0.281), diabetes mellitus (P = 0.172) or left ventricular ejection fraction (P = 0.09) between patients treated or not treated with RAAS blockade. Univariate analyses concerning development of CIN depending on treatment with RAAS blockade within 72 h found CIN to be significantly higher in patients treated with RAAS blockade (11.9 vs 4.2%, P = 0.006). Multivariate analyses (logistic regression) identified RAAS blockade to be an independent predictor of CIN (odds ratio 3.082, 95% confidence interval 1.234-7.698, P = 0.016).Conclusion. Patients treated with RAAS blockade before exposure to CM develop significantly more often CIN within 72 h. Even after adjustment for confounding comorbidities, treatment with ACE-I or AT-1 blockers turned out to be an independent risk predictor.