Decision tree prediction of saline infusion test outcomes in suspected primary aldosteronism: a multicenter common data model study
Authors' conclusion
A screening-stage decision tree provides clinically interpretable probabilities for SIT positivity, supporting risk-stratified confirmatory testing and selective omission in high-probability patients.
Study context
- Population
- Patients who underwent SIT for suspected primary aldosteronism (PA), three tertiary hospitals (2002–2020); mean age 51.2 years
- Participants
- 307 patients3 tertiary hospitals
- Exposure
- Tree-derived criteria vs conventional rigid screening criteria based on plasma aldosterone concentration (PAC) and aldosterone-to-renin ratio (ARR)
- Outcome
- SIT positivity (post-infusion PAC ≥10 ng/dL)
Methodological detailsmulticenter, Common Data Model, conditional inference classification tree
- Multicenter
- Common Data Model
- Conditional inference classification tree
Main finding
As reported in the abstract
Prediction of SIT positivity
| Group | Accuracy | Sensitivity | Specificity |
|---|---|---|---|
| Tree-derived criteria | 0.85 | 0.83 | 0.86 |
Tree-derived criteria
Accuracy0.85
Sensitivity0.83
Specificity0.86
- High-probability phenotype (PAC >25 ng/dL and PRA ≤0.55 ng/mL/h): SIT positivity rate
- 96.4%
- PAC 15.7-25.0 ng/dL with potassium ≤3.9 mmol/L: SIT positivity
- 68.8%
Observational data: associations, not effects.
Further findings
PAC, SIT-positive vs SIT-negative
P < 0.001
Serum potassium, SIT-positive vs SIT-negative
P < 0.001
SIT positivity
- SIT-positive
- 153 (49.8%)
Original publication
Decision tree prediction of saline infusion test outcomes in suspected primary aldosteronism: a multicenter common data model study. J Clin Endocrinol Metab. Published 10 August 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →