Ultrasound strain zero-crossing elasticity measurement in assessment of renal allograft cortical hardness: a preliminary observation.
Academic Article
Overview
abstract
To determine whether ultrasound strain zero-crossing elasticity measurement can be used to discriminate moderate cortical fibrosis or inflammation in renal allografts, we prospectively assessed cortical hardness with quasi-static ultrasound elastography in 38 renal transplant patients who underwent kidney biopsy from January 2013 to June 2013. With the Banff score criteria for renal cortical fibrosis as gold standard, 38 subjects were divided into two groups: group 1 (n = 18) with ≤25% cortical fibrosis and group 2 (n = 20) with >26% cortical fibrosis. We then divided this population again into group 3 (n = 20) with ≤ 25% inflammation and group 4 (n = 18) with >26% inflammation based on the Banff score for renal parenchyma inflammation. To estimate renal cortical hardness in both population divisions, we propose an ultrasound strain relative zero-crossing elasticity measurement (ZC) method. In this technique, the relative return to baseline, that is zero strain, of strain in the renal cortex is compared with that of strain in reference soft tissue (between the abdominal wall and pelvic muscles). Using the ZC point on the reference strain decompression slope as standard, we determined when cortical strain crossed zero during decompression. ZC was negative when cortical strain did not return or returned after the reference, whereas ZC was positive when cortical strain returned ahead of the reference. Fisher's exact test was used to examine the significance of differences in ZC between groups 1 and 2 and between groups 3 and 4. The accuracy of ZC in determining moderate cortical fibrosis and moderate inflammation was examined by receiver operating characteristic analysis. The intra-class correlation coefficient and analysis of variance were used to test inter-rater reliability and reproducibility. ZC had good inter-observer agreement (ICC = 0.912) and reproducibility (p = 0.979). ZCs were negative in 18 of 18 cases in group 1 and positive in 19 of 20 cases in group 2 (p ≪ 0.001), and were positive in 18 of 20 cases in group 3 and negative in 17 of 18 cases in group 4 (p ≪ 0.001). The area under the receiver operating characteristic curve was 0.992 ± 0.010 for fibrosis and 0.988 ± 0.021 for inflammation. ZC had 100% sensitivity and 95% specificity when zero strain was used as the cutoff value to determine moderate cortical fibrosis and 94% sensitivity and 90% specificity for inflammation. ZC is a new strain marker that could be straightforward to interpret and perform, making it a potentially practical approach for monitoring progression of cortical fibrosis or inflammation in renal allografts.
