Answers are grounded in this paper and cite their sources.
INTRODUCTION: Adrenal venous sampling (AVS) remains the gold standard for primary aldosteronism (PA) subtyping but is invasive and technically demanding. We investigated whether peripheral steroid profiles could serve as a noninvasive tool for PA subtyping, characterizing the presence of coexisting mild autonomous cortisol secretion (MACS), and predicting biochemical outcomes after adrenalectomy. METHODS: In total, 216 patients with PA who underwent AVS between 2018 and 2022 were enrolled. Serum steroid profiling quantified 19 adrenal steroids and 17 metabolic ratios using liquid chromatography-mass spectrometry. Feature selection was performed using random forest-based recursive feature elimination. Logistic regression models were developed to predict lateralized PA and postoperative biochemical outcomes. Model performance was evaluated using receiver operating characteristic analysis. RESULTS: Lateralized PA showed significantly higher aldosterone and 18-hydroxycortisol (18-OHF) (P<0.001) but lower 11-hydroxyandrostenedione (11-OHA4) (P=0.001) than bilateral PA. A three-steroid model (18-OHF, aldosterone, 11-OHA4) showed discrimination comparable to a clinical model based on serum potassium, plasma aldosterone concentration, and plasma renin activity (AUC 0.821 vs. 0.790; P=0.351). A combined model integrating the clinical parameters with 18-OHF and 11-OHA4 achieved the highest performance (AUC 0.852), significantly outperforming the clinical model (P=0.002). Patients with lateralized PA and MACS had higher 18-OHF levels (age/sex-adjusted P=0.038). Baseline steroid markers were not significantly associated with post-surgical biochemical success (P>0.05). CONCLUSION: Serum steroid profiling may facilitate the noninvasive identification of lateralized PA beyond clinical parameters alone. Coexisting MACS was associated with androgen suppression and altered steroid patterns, while the prognostic role of steroid profiles for treatment outcomes requires further investigation.
No takes yet. Share an insight, caveat, or question.
Park et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: