Assessment of renal function with computed tomographic densitometry measurements

Abstract

Rationale and objectives: We assessed the relationship between multiple renal computed tomographic (CT) densitometry parameters and renal function.

Methods: Three hundred seventy-three patients underwent standardized helical CT of the abdomen. The ratio of mean attenuation in the renal cortex to mean attenuation in the aorta (KAR), the products of mean renal cortical attenuation with CT-estimated renal volume (KVP), and the patient's weight (KWP) were derived from scans obtained 90 sec (n = 373) and 30 sec (n = 108) after initiation of intravenous contrast material administration. These densitometry parameters were compared with renal function measured by serum level of creatinine and creatinine clearance (CrCl).

Results: Among the 373 patients in the study, we found statistically significant differences (p < .01) between the patients with normal renal function (CrCl > or = 60 ml/min, n = 300) and the patients with abnormal renal function (CrCl < 60 ml/min, n = 73) for the KAR, KVP, and KWP. The KAR was the parameter best correlated with CrCl and was an independent predictor of renal function from the patient's age, weight, and renal volume. Fifty-three patients with a KAR less than 1 had significantly worse renal function (CrCl = 60 +/- 21 ml/min) than the patients with a KAR greater than or equal to 1 (CrCl = 95 +/- 31 ml/min). Only 4% of patients with normal renal function had a KAR less than 1. A threshold value of KAR less than 1 had a sensitivity of 55%, a specificity of 96%, a positive predictive value of 75%, and a negative predictive value of 90% for predicting renal dysfunction. CT parameters obtained at 30 sec were less useful than parameters at 90 sec.

Conclusion: In patients undergoing clinically requested CT scanning, renal densitometry analysis can be used to depict patients with normal renal function; however, it has a high false-negative rate in depicting patients with diminished renal function.