Construction of a predictive scoring system as a guide to screening and confirmation of the diagnosis of primary aldosteronism

• Published in: Clinical endocrinology (Oxford) Vol. 92; no. 3; pp. 196 - 205
• Main Authors: Kietsiriroje, Noppadol, Wonghirundecha, Rawipas, Suntornlohanakul, Onnicha, Murray, Robert D.
• Format: Journal Article
• Language: English
• Published: England Wiley Subscription Services, Inc 01.03.2020
• Subjects: Adolescent; Adult; Aged; Aldosterone; Aldosterone - blood; Antihypertensives; confirmatory test; Cross-Sectional Studies; Diabetes mellitus; Diagnosis; Diagnostic Techniques, Endocrine; Endocrine disorders; endocrine hypertension; Female; Humans; hyperaldosteronism; Hyperaldosteronism - blood; Hyperaldosteronism - diagnosis; Hypertension - blood; Hypertension - diagnosis; Hypertension - etiology; Male; Mass Screening - methods; Medical diagnosis; Middle Aged; Patients; predictive score; Predictive Value of Tests; primary aldosteronism; Prognosis; Renin; Renin - blood; Research Design; Retrospective Studies; screening test; Sensitivity and Specificity; Thailand; Validation Studies as Topic; Young Adult; Thailand; endocrine hypertension; screening test; primary aldosteronism; hyperaldosteronism; diagnosis; predictive score; confirmatory test
• ISSN: 0300-0664; 1365-2265
• DOI: 10.1111/cen.14142

Background Primary aldosteronism (PA) is the most frequent cause of secondary hypertension. In Southern Thailand, the aldosterone‐renin ratio (ARR) is only available within a small number of tertiary centres, necessitating need for a simple clinical assessment to determine the requirement for ARR. Objective This study aimed to identify predictive factors for the diagnosis of PA and generate a predictive scoring system (PSS) for use in screening and diagnosis of PA. Patients and Methods A total of 420 patients aged >15 years with paired plasma aldosterone concentration and plasma renin activity values allowing calculation of ARR were identified from the electronic hospital database between 2011 and 2016. Results The overall prevalence of PA was 16.7% (range; adrenal incidentaloma 5.6% to hypokalaemia 30%). Predictive factors for diagnosis of PA were as follows: age <60 years, BMI < 25 kg/m2, presence of diabetes, ≥3 antihypertensive agents, serum sodium ≥ 141 mmol/L and serum potassium < 3.5 mmol/L. A predictive scoring system (PSS) (range −2 to 13) was generated by the coefficients of the variables with ROC curve AUC 0.87 [95% CI: 0.83‐0.91]. Using the PSS, a total score <4 provided a robust negative predictive value (sensitivity, 0.97; specificity, 0.48; NPV, 0.99; PPV, 0.27) for PA. In patients at high risk of PA (PAC > 15 ng/dL and PRA < 1.0 ng/mL/hr), a PSS score > 9 had specificity and PPV of 100%, essentially confirming PA in these individuals. Conclusion The proposed PSS for PA will enable more focused and cost‐effective use of ARR screening and confirmatory testing. In our cohort, 40% and 42% of patients would not require ARR screening or confirmatory tests, respectively.