A diagnostic strategy for primary aldosteronism diagnosis based on urinary aldosterone assay
In brief
Chemiluminescent urine aldosterone tests match mass-spectrometry accuracy (~0.81 AUC) for primary aldosteronism detection
In 705 hypertensive inpatients, 24-hour urinary aldosterone measured by two chemiluminescent immunoassays correlated closely with LC-MS/MS and achieved similar overall diagnostic performance (AUC ≈0.81). Stratifying results by direct renin concentration raised accuracy to about 0.88 in low-renin patients, but the proposed cut-offs need external validation before routine use.
- Journal
- Clinical chemistry and laboratory medicine (Q1)
- Published
- 1 September 2026
- Study design
- Prospective / inception cohort
- Evidence level
- Level 2, Moderate (CEBM 2b)
- Authors
- Zhangwei Gao, Xiaojing Gao, Hongying Cong, Wei Zhang, Xiaocui Shi, Yutao Liu, et al.
- PMID
- 42670711
- DOI
- 10.1515/cclm-2026-0865
Why clinicians should know about it
- Picked for Biochemistry (medical) (paper of the day, 1 September 2026): Urinary aldosterone assay, endocrine assay validation
Abstract
OBJECTIVES: To compare 24-h urinary excretion of aldosterone (24-h UEA) measured by competitive chemiluminescent immunoassay (cCLIA) and sandwich chemiluminescent immunoassay (sCLIA) with liquid chromatography-tandem mass spectrometry (LC-MS/MS), and to evaluate a direct renin concentration (DRC)-stratified interpretive strategy for primary aldosteronism (PA). METHODS: We prospectively enrolled 705 hypertensive inpatients, including 105 with PA and 600 with essential hypertension. 24-h UEA was measured by cCLIA, sCLIA, and LC-MS/MS, and DRC was measured by CLIA. Method agreement was assessed using Passing-Bablok regression and Bland-Altman analysis. Diagnostic performance was evaluated using receiver operating characteristic (ROC), with internal validation by 1,000 outcome-stratified bootstrap resamples. RESULTS: Both CLIAs correlated strongly with LC-MS/MS (Spearman r=0.946 for cCLIA and 0.950 for sCLIA). Overall areas under the curve (AUCs) were 0.809, 0.814, and 0.812 for cCLIA, sCLIA, and LC-MS/MS, respectively, with no statistically significant differences among the methods. A DRC cutoff of 8.1 mU/L was used to define low-DRC (≤8.1 mU/L) and high-DRC (>8.1 mU/L) subgroups. In the low-DRC subgroup, AUCs were 0.880, 0.879, and 0.881; in the high-DRC subgroup, they were 0.871, 0.878, and 0.859, respectively. Optimism-corrected AUCs remained close to the original estimates. Method-specific high-sensitivity and high-specificity thresholds were used to construct an exploratory DRC-stratified interpretive strategy. CONCLUSIONS: CLIA-based 24-h UEA measurements showed strong correlation and small mean bias with LC-MS/MS and similar diagnostic discrimination in this cohort. DRC stratification may support method-specific interpretation of 24-h UEA, but the proposed thresholds require prospective external validation before clinical implementation.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.