Body surface area-based kidney length percentiles misdiagnose small kidneys in children with overweight/obesity

Background We evaluated the diagnostic performance of height-, age- and body surface area (BSA)-based kidney length (KL) percentiles in the identification of at least one small kidney (KL < 3(rd)) and in the prediction of reduced estimated glomerular filtration rate (eGFR) and/or elevated blood pressure (BP) in children with and without overweight (OW)/obesity(OB).Methods In this cross-sectional study, 744 apparently healthy children (mean age 8.3 years) were recruited in a primary care setting. Clinical data were collected, and serum creatinine and KL were measured. Height-, age- and BSA-based percentiles of KL were calculated and the association of at least one small kidney per subject with reduced eGFR and/or elevated BP was explored by logistic regression.Results Two hundred fifty-seven out of seven hundred forty-four (34.5%) subjects were OW/OB and 127 (17.1%) had reduced eGFR or elevated BP. In separate analyses in children with OW/OB, the KL percentiles calculated on the basis of BSA were lower compared with height- and age-based KL percentiles. Consequently, the prevalence of a small kidney was significantly higher when evaluating percentiles of KL based on BSA compared with other percentiles. In logistic regression analysis, a small kidney was significantly associated with reduced eGFR and/or elevated BP only when using height-based KL percentiles. The KL percentiles according to BSA for the ideal weight (iBSA) showed similar performance compared with height-based percentiles. No differences in the diagnostic performance of different percentiles were found in children with normal weight.Conclusions BSA-based percentiles underestimate KL in children with OW/OB. In these subjects, the use of height-based or iBSA-based percentiles should be preferred.