Abstract / Summary
Abstract Background Primary aldosteronism (PA) is the leading cause of secondary hypertension. While unilateral adrenalectomy achieves biochemical cure, postoperative hypertension resolution varies widely, and existing tools such as the Aldosteronoma Resolution Score perform inconsistently in Asian populations. We aimed to develop, temporally validate, and update a nomogram for predicting hypertension cure after adrenalectomy. Methods We retrospectively analyzed 112 patients with unilateral PA who underwent adrenalectomy (2018–2021) as the derivation cohort and 75 consecutive patients (2023) for temporal validation. Hypertension outcomes were classified by PASO criteria. Multivariable logistic regression identified independent predictors for nomogram construction. Discrimination, calibration, and clinical utility (decision curve analysis) were assessed, followed by model updating via parsimony reduction and logistic recalibration. Results In the derivation cohort (cure rate 45.5%), age, waist circumference, highest systolic blood pressure, and cerebro-cardiovascular disease history were independent predictors of non-cure. The five-variable nomogram demonstrated excellent discrimination (C-index 0.907) but showed moderate performance (C-index 0.734) with overfitting (calibration slope 0.32) upon temporal validation. A parsimonious four-variable model (excluding hypertension duration) improved discrimination (C-index 0.812) and calibration, with further enhancement after logistic recalibration. Decision curve analysis confirmed superior net benefit across clinically relevant threshold probabilities. Conclusions Temporal external validation substantially tempered the derivation-stage performance. The recalibrated four-variable model using routinely available preoperative variables offers the best transportability and is proposed for individualized preoperative counseling in Chinese patients with unilateral PA. Clinical trail number: not applicable.