Evaluation of two classical contrast-induced nephropathy definitions for predicting long-term mortality in patients undergoing elective percutaneous coronary intervention

Abstract

Abstract Background Contrast-induced nephropathy (CIN) is a frequent complication after percutaneous coronary intervention (PCI), and is associated with poor outcome. However, the optimal definition of CIN has been debated because of its different incidence and influence on prognosis. At present, there are limited data regarding th impact of different CIN definitions on long-term mortality in patients undergoing elective PCI. Purpose To explore the influence of two classical CIN definitions on long-term mortality and identify which definition was more suitable for predicting long-term mortality in patients undergoing elective PCI.. Methods We prospectively observed 5600 consenting patients undergoing PCI from January 2012 to December 2018. Two classical CIN definitions include those defined by ESUR[Contrast-media-induced nephrotoxicity (CMN)] and AKIN[contrast induced acute kidney injury (CI-AKI)]. CMN was defined as an increase in serum creatinine (SCr) ≥25% or 0.5 mg/dLabove thebaseline level within 3 days,while CI-AKI wasdefined as an increase in SCr ≥50% or 0.3 mg/dL within 48hs after contrast medium exposure.The association of CIN with long-term mortality was investigated by Cox regression analysis.Interaction analyses were performed for long-term mortality across subgroups. Results The incidence of CIN according to ESUR (CMN) and AKIN (CI-AKI) definition were18.3% (n=1023) and 6.1% (n=342), respectively. During a median follow-up of 2 years, after adjusting other potential risk factors, multivariable cox regression analysis revealed CIN was a risk factor for long-term mortality [hazard ratio (HR): 2.021, 95% confidence interval (CI): 1.389–2.938, P<0.0001] according to AKIN definition, but not for ESUR definition (HR: 1.344, 95% CI: 0.982–1.838, P>0.05). Further interaction analysis showed that there was a significant interaction between age >75ys and CMN for long-term mortality (P=0.042) while no such association was observed between age >75ys and CI-AKI (P=0.806). Conclusions CIN defined by AKIN may be more suitable for predicting long-term mortality in patients undergoing elective PCI. However, in elderly patients, CIN defined by ESUR could also be used for predicting long-term mortality. Association Between CIN and mortality Funding Acknowledgement Type of funding source: None