Reproducibility of a novel computed-tomography based measurement of renal papillary density in the Framingham Heart Study

Abstract

Background: Renal papillary calcification is a compelling candidate risk factor for chronic kidney disease (CKD) and nephrolithiasis. Renal papillary density (RPD), as assessed by computed tomography (CT), is a potential marker for calcification that has not been well studied. We developed a protocol to measure RPD using CT scans and assessed its reproducibility in participants from the Framingham Heart Study. Methods: We assessed RPD of right kidneys from a single abdominal CT slice in 100 representative participants from the Framingham Heart Study (47 % female, mean age 59.9 years) using a novel protocol. We selected the kidney slice with the most open sinus space and assessed RPD using the average of three 20 mm2 ellipses from upper, middle and lower papillary regions. Two different readers performed RPD measurements and the first reader repeated all measurements to determine both intra- and inter-reader reproducibility, respectively. Results: Of 100 total individuals included in the replication dataset, six were excluded for poor scan quality. Average RPD across all individuals was 48.7 ± 4.7 (range 38.7-61.7) Hounsfield Units (HU). The intra- and inter-reader correlation coefficients were 0.86 and 0.79, respectively. Bland-Altman analysis suggested no systematic bias between the different reads. Conclusion: Measuring RPD is practical and reproducible using MDCT scans from a small sample of a community-based cohort.