Predictors of contrast induced nephropathy in patients with normal renal function undergoing percutaneous coronary intervention - ADAPH score

Contrast-induced nephropathy (CIN) remains a common complication, affecting 5–10% of patients undergoing percutaneous coronary intervention (PCI). While its association with pre-existing renal disease and excessive use of iodinated contrast has been extensively studied, other precipitating factors should also be identified to enable the prompt adoption of preventive strategies. This study aims to identify the main predicting factors of CIN and develop a score that could be used to define preventive strategies.

A retrospective study was designed, including all patients who underwent urgent or elective PCI between January 1, 2019, and May 31, 2024, with a creatinine clearance ≥60 ml/min (calculated with the Cockroft-Gault formula) and serum creatinine ≤1.2 mg/dL. CIN was defined as an increase of ≥0.5 mg/dL in serum creatinine or ≥0.25% increase from basal levels within the first 48 hours post-procedure. Comorbidities and concomitant medications were studied in a multivariate analysis and a score named ADAFH was developed consisting in the cumulative impact of fibrinolysis (p=0.017 - 1 point), Heart Failure (p= 0.02, 1 point), Anaemia (p= 0.025, 2 points), Diabetes (p=0.007, 2 points) and the protective Alopurinol role (p=0.04), -1 point). High-risk patients were considered with a score equal or above 3. Then we studied its impact on CIN development using binary logistic regression and compared it with validated association such as volume of iodinated contrast/creatinine clearance ratio (Vc/Clcr) > 4.

487 patients were included, 59.1% (n=289) were men, with a mean age of 66.6 ±10 years, 31.0% (n=151) diabetic, 8,2% (n=40) who underwent fibrinolysis before PCI, 14.6% (n=71) currently under Allopurinol, 23.4% (n=114) with Hb < 13 g/dl and 8.0% (n=39) with history of heart failure. Mean ClCr was 91.92 ml/min/1.73m² ± 28.26 and an average dose of iodinated contrast used of 208,61 ± 2.48 ml. 10.5% (n=51) patients developed CIN. Mean ADAFH score was 1.11 ± 2.1, with 13.3% (n=65) patients classified in the high-risk group for diabetic nephropathy according to the ADAFH score. We found that CIN was more common in patients who scored equal or above 3 in the ADAFH score (52.9% vs 10.6%: χ² = 64.39 p=0.001). In binary logistic regression, the ADAPH score seemed to be an independent predictor of CIN (OR: 9.538, 95% CI: 5.085–9.538, p = 0.001). ROC curve analysis demonstrated an acceptable capacity of predicting CIN, better than the Vc/Clcr > 4 (AUC 0.81 vs 0.66; p=0.04) (Figure 1).

This study highlights the likely impact of the ADAFH score and its’ variables on the development of CIN, further demonstrating a real risk of this complication even in patients with good renal function at admission, with an acceptable performance in predicting CIN. It also emphasizes the need to adopt preventive strategies, particularly in patients that despite its apparently normal creatinine clearance, might be in greater risk for CIN.