Low agreement between kidney volume and kidney length z-scores

Abstract

Pediatric nephrologists use kidney length and kidney volume z-scores to longitudinally assess normal nephron endowment. However, most radiologists only report kidney length. Agreement between kidney length and kidney volume z-scores in children has been understudied. This study aims to assess agreement between kidney length and kidney volume z-scores in children. This novel cross-sectional cohort study prospectively followed prematurely born babies from a large specialized prematurity follow-up center. A healthy control group matched the cases by age and sex and was recruited from schools. Children were assessed for kidney length and kidney volumes at age 5 by three independent ultrasonographers. All measurements were performed in triplicate. Detailed anthropometry, blood pressure, and kidney function were also obtained. Age-independent z-scores were calculated for all parameters according to Scholbach and Weitzel and compared using descriptive statistics. We studied 89 premature patients (median 32 weeks gestational age) and 33 healthy controls (median 38 weeks gestational age). There were 732 determinations of kidney length, width, and thickness. The mean z-score of the right kidney length was 0.65 ± 0.08 (SEM) compared with 0.88 ± 0.08 of the left kidney length (p = 0.0003, two-sided paired t test). The squared correlation coefficient for kidney volume to kidney length was 0.32 (p < 0.0001). Bland and Altman analysis revealed considerable bias with − 1.36 ± 0.76 standard deviations and 95% limits of agreement from − 2.83 to − 0.16. Reporting only kidney length results in significant overestimation of age-independent z-scores. Based on our findings, consideration to measuring all kidney dimensions may be more appropriate.