Mehran 2.0 Risk Score for Prediction of CA-AKI After PCI

Overview

Consecutive patients having percutaneous coronary intervention (PCI) over a period of 9 years at a large tertiary care center with available creatinine measurements both before and within 48 hours after the procedure were included; patients on chronic dialysis were excluded. Patients treated between 2012 and 2017 comprised the derivation cohort (n=14,616) and those treated from 2018 to 2020 formed the validation cohort (n=5,606). The primary endpoint is contrast-associated acute kidney injury (CA-AKI, defined per Acute Kidney Injury Network \[AKIN\]). In addition, independent predictors of CA-AKI will be derived from multivariate logistic regression analysis. Model 1 will include only preprocedural variables, while Model 2 will also include procedural variables. A weighted integer score based on the effect estimate of each independent variable will be used to calculate the final risk score for each patient. Impact on 1-year mortality will be also evaluated.

Consecutive patients who underwent PCI at a tertiary hospital (Mount Sinai Hospital, New York, NY, USA) between 2012 and 2020 with documented serum creatinine both pre- procedure and within 48 hours after PCI were identified from our existing prospectively populated database. Only patients who expired within 48 hours after PCI and with pre-existing end-stage renal disease requiring dialysis were excluded from the present analysis. CA-AKI was defined according to the Acute Kidney Injury Network as an increase by ≥50% or ≥0.3mg/dl within 48 hours after PCI compared to pre-PCI serum creatinine. Anemia was defined as hemoglobin plasma level ≤13 g/dl for men and ≤12 g/dl for women. Chronic kidney disease (CKD) was defined as an estimated glomerular filtration rate (eGFR) of ≤60 ml/min/1.73 m2, per the Chronic Kidney Disease-Epidemiologic (CKD-EPI) formula. A dedicated data coordinating center performed all data management. Prespecified clinical and laboratory demographic information was obtained from hospital charts that were reviewed by independent research personnel who were unaware of the objectives of the study; accumulated data were then entered prospectively in the database. Patients treated between 2012 and 2017 will comprise the derivation cohort (n=14,616) and those treated from 2018 to 2020 will form the validation cohort (n=5,606). Independent predictors of CA-AKI (defined per Acute Kidney Injury Network) will be derived from multivariate logistic regression analysis. Model 1 will include only preprocedural variables, while Model 2 will also include procedural variables. A weighted integer score based on the effect estimate of each independent variable will be used to calculate the final risk score for each patient. Impact on 1-year mortality of CA-AKI occurrence will be also evaluated with Cox regression analysis. Multivariable Cox regression analysis will be also performed to look for independent association of CA-AKI occurrence with mortality.