#1829 Intravenous contrast exposure in stable CKD 4 and 5 patients and adverse renal outcomes: a propensity-score matched study

Abstract

Abstract Background and Aims Previous studies of intravenous (IV) contrast administration in patients with CKD have focused on the association between contrast exposure and AKI. Here we explore the association of IV contrast administration for CT imaging with a broader range of adverse renal outcomes in patients with advanced CKD (stages 4 and 5). Method Data covering nephrology care in the West of Scotland were obtained from the Strathclyde Electronic Renal Patient Record. We included adult patients with an incident CT scan and an eGFR < 30 mL/min/1.73 m2 between April 2008 and May 2020. Patients with AKI were excluded. Structured query language (SQL) was used to analyse CT scan reports and identify whether IV contrast had been administered or not. Patient demographics (age, sex, Scottish index of multiple deprivation), co-morbidities, laboratory tests (including eGFR slope for 6 months preceding the study calculated by linear regression, eGFR at time of CT, C-reactive protein, urine protein:creatinine ratio), anthropometric measurements (blood pressure, body mass index), and medications were collected. The primary outcome was a composite renal outcome (AKI, sustained eGFR drop, renal replacement therapy [RRT], and renal death). Secondary outcomes included individual components of the composite outcome and all-cause death. AKI was defined as a 25% creatinine rise from index creatinine between days 7 and 14 post CT. Sustained eGFR drop was defined as two consecutive eGFR < 15 mL/min/1.73 m2 more than 4 weeks apart. Renal death was defined as death from acute kidney injury, chronic kidney disease or end stage renal disease without dialysis. A propensity-score matched dataset was created using a 1:2 ratio with no replacement. Multivariable logistic, proportional hazards and Fine-Gray (competing risks) regressions were used to identify independent correlates of the primary and secondary outcomes. Missing data were multiply imputed using the “mice: Multivariate Imputation by Chained Equations” package on R studio. Results 9226 patients with an eGFR < 30 mL/min/1.73 m2 underwent an incident CT scan during the study period. Among 6371 patients meeting the inclusion criteria (3067 (48.1%) men; mean [SD] age 72.8 [13.8] years; mean [SD] eGFR 20.4 [6.9] ml/min/1.73 m2), 875 had received IV contrast. A propensity-score matched dataset was generated consisting of 875 patients who had a contrast CT and 1750 patients with non-contrast CT. On multivariable analysis, contrast administration was associated with a higher risk of AKI (HR 1.49, 95% CI 1.06-2.10, p = 0.02) but not with the composite renal outcome (HR 0.89, 95% CI 0.66-1.20, p = 0.45). Contrast administration did not change the risk of a sustained eGFR drop (HR 0.88, 95% CI 0.54-1.43, p = 0.60), RRT (HR 0.85, 95% CI 0.61-1.20, p = 0.36), or all-cause death (HR 0.95, 95% CI 0.76-1.19, p = 0.65). We did not identify an effect of contrast administration on the time to composite renal outcome, RRT, or all-cause death by Cox regression (HR 0.92, 95% CI 0.70-1.22, p = 0.56; HR 0.88, 95% CI 0.64-1.23, p = 0.46; and HR 1.07, 95% CI 0.94-1.20, p = 0.30, respectively). On Fine-Gray regression, we did not identify an impact of contrast administration on the risk of RRT (HR 0.84, 95% CI 0.63-1.12, p = 0.24) or death (HR 1.07, 95% CI 0.95-1.20, p = 0.27). In subgroup analysis in patients with an index eGFR < 20 mL/min/1.73 m2, contrast administration was not associated with the composite renal outcome (HR 0.79, 95% CI 0.51-1.22, p = 0.28), RRT (HR 0.85, 95% CI 0.52-1.39, p = 0.51), AKI (HR 1.28, 95% CI 0.66-2.45, p = 0.46), or death (HR 0.94, 95% CI 0.64-1.39, p = 0.76). Conclusion In a stable population with advanced CKD, we identified an association between IV contrast administration and the risk of AKI, but this did not translate in adverse long-term renal outcomes. These findings suggest that increases in serum creatinine post-IV contrast in patients with stable CKD are not clinically significant.