Contrast-enhanced CT and Acute Kidney Injury: Risk Stratification by Diabetic Status and Kidney Function

Background Diabetes mellitus may be associated with an increased likelihood of CT contrast material-induced acute kidney injury (CI-AKI), but this has not been studied in a large sample with and without kidney dysfunction. Purpose To investigate whether diabetic status and estimated glomerular filtration rate (eGFR) are associated with the likelihood of acute kidney injury (AKI) following CT contrast material administration. Materials and Methods This retrospective multicenter study included patients from two academic medical centers and three regional hospitals who underwent contrast-enhanced CT (CECT) or noncontrast CT between January 2012 and December 2019. Patients were stratified according to eGFR and diabetic status, and subgroup-specific propensity score analyses were performed. The association between contrast material exposure and CI-AKI was estimated with use of overlap propensity score-weighted generalized regression models. Results Among the 75 328 patients (mean age, 66 years ± 17 [SD]; 44 389 men; 41 277 CECT scans; 34 051 noncontrast CT scans), CI-AKI was more likely in patients with an eGFR of 30-44 mL/min/1.73 m2 (odds ratio [OR], 1.34; P < .001) or less than 30 mL/min/1.73 m2 (OR, 1.78; P < .001). Subgroup analyses revealed higher odds of CI-AKI among patients with an eGFR less than 30 mL/min/1.73 m2, with or without diabetes (OR, 2.12 and 1.62; P = .001 and .003, respectively), when they underwent CECT compared with noncontrast CT. Among patients with an eGFR of 30-44 mL/min/1.73 m2, the odds of CI-AKI were higher only in those with diabetes (OR, 1.83; P = .003). Patients with an eGFR less than 30 mL/min/1.73 m2 and diabetes had higher odds of 30-day dialysis (OR, 1.92; P = .005). Conclusion Compared with noncontrast CT, CECT was associated with higher odds of AKI in patients with an eGFR of less than 30 mL/min/1.73 m2 and in patients with diabetes with an eGFR of 30-44 mL/min/1.73 m2; higher odds of 30-day dialysis were observed only in patients with diabetes with an eGFR less than 30 mL/min/1.73 m2. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Davenport in this issue.