Abstract / Summary
Abstract Background Endocrine causes of secondary hypertension carry important cardiovascular consequences but require different disease-specific diagnostic pathways. We evaluated whether a small set of baseline endocrine measurements could support triage among adults already undergoing etiologic evaluation for hypertension. Methods In this prospective single-center cohort, adults evaluated for hypertension in Shenzhen between July 2023 and January 2025 underwent baseline measurement of adrenocorticotropic hormone, plasma renin activity (PRA), angiotensin II, aldosterone, and cortisol before final endocrine adjudication. Participants with complete candidate-predictor and independent reference-standard data were included. Nested logistic models were compared using a stratified 70:30 development-validation split. Discrimination, calibration, Brier score, sensitivity, specificity, predictive values, and likelihood ratios were assessed. Results Among 5,986 participants in the screening cohort, 1,502 completed the reference-standard pathway and 412 had confirmed primary aldosteronism, pheochromocytoma/paraganglioma, or Cushing syndrome. In the validation cohort (n = 451; 124 events), the aldosterone-PRA-cortisol model achieved an area under the receiver operating characteristic curve of 0.889 (95% CI 0.852–0.922), sensitivity of 72.6%, specificity of 86.9%, positive predictive value of 67.7%, negative predictive value of 89.3%, and Brier score of 0.109. Adding angiotensin II and adrenocorticotropic hormone did not materially improve the linear model. An exploratory nonlinear five-marker model achieved an area under the curve of 0.900 with higher sensitivity but lower specificity. Conclusions A three-marker model retained most of the discrimination of the five-marker panel and may help prioritize disease-specific confirmatory evaluation in selected patients undergoing hypertension work-up. It should not replace established endocrine testing or be used to withhold guideline-recommended primary aldosteronism screening. External validation is required before clinical implementation.