Phenome-Wide Association Study of Kidney Volume in a General Population Biobank Stratified by Diabetes Status

Background: Total kidney volume (TKV) is an emerging quantitative trait linked to kidney function, yet its broader clinical associations in the general population remain understudied. Given the potential for metabolic diseases such as diabetes to modify kidney structure and its downstream effects, we conducted a phenome-wide association study (PheWAS) of TKV stratified by diabetes status. Methods: We leveraged linked EHR and imaging data from BioVU, a large academic biobank, to conduct PheWAS analyses of TKV among 8,355 individuals with segmented kidney volumes derived from CT imaging. TKV was inverse-normalized and modeled as a continuous predictor. Associations with 1,400+ phecodes were tested using logistic regression adjusted for age and sex, and in secondary models, additionally adjusted for BMI. Analyses were conducted in the overall population and stratified by sex and diabetes status. Results: In the full cohort, larger TKV was associated with multiple phenotypes including obesity, hypertension, and glomerular diseases. Stratified analyses revealed distinct association profiles by diabetes status. Among individuals with diabetes, larger TKV was significantly associated with retinopathy, neuropathy, and macrovascular complications. In contrast, among non-diabetic individuals, associations were more limited to obesity-related traits and cystic kidney disease. Adjustment for BMI attenuated some but not all associations, suggesting that TKV captures phenotypic variation not fully explained by body size. Conclusion: TKV is associated with a diverse range of clinical phenotypes in the general population, with distinct profiles emerging by diabetes status. These findings highlight the utility of imaging-derived kidney traits in capturing differential risk and comorbidity patterns, and support further investigation of TKV as a potential biomarker in diabetes-related kidney and systemic disease.