Development and Validation of Prediction Models for Subtype Diagnosis of Patients With Primary Aldosteronism

Abstract

Context Primary aldosteronism (PA) comprises unilateral (lateralized, LPA) and bilateral disease (BPA). The identification of LPA is important to recommend potentially curative adrenalectomy. Adrenal venous sampling (AVS) is considered the gold standard for PA subtyping, but the procedure is available in few referral centers. Objective To develop prediction models for subtype diagnosis of PA using patient clinical and biochemical characteristics. Design, Patients and Setting Patients referred to a tertiary hypertension unit. Diagnostic algorithms were built and tested in a training (N=150) and in an internal validation cohort (N=65), respectively. The models were validated in an external independent cohort (N=118). Main outcome measure Regression analyses and supervised machine learning algorithms were used to develop and validate two diagnostic models and a 20-point score to classify patients with PA according to subtype diagnosis. Results Six parameters were associated with a diagnosis of LPA (aldosterone at screening and after confirmatory testing, lowest potassium value, presence/absence of nodules, nodule diameter, and computed tomography results) and were included in the diagnostic models. Machine learning algorithms displayed high accuracy at training and internal validation (79.1% to 93%), whereas a 20-point score reached an AUC of 0.896, and a sensitivity/specificity of 91.7/79.3%. An integrated flow-chart correctly addressed 96.3% of patients to surgery and would have avoided AVS in 43.7% of patients. The external validation on an independent cohort confirmed a similar diagnostic performance. Conclusions Diagnostic modelling techniques can be used for subtype diagnosis and guide surgical decision in patients with PA in centers where AVS is unavailable.

Machine learning based models for prediction of subtype diagnosis of primary aldosteronism using blood test

“…The score-based algorithms reported previously revealed the importance of PAC and serum K levels 26 – 31 , 33 . However, for the first time, we demonstrated that serum Na levels were also useful for the subtype diagnosis of PA.…”

“…However, all were based on confirmatory test data, computed tomography (CT) findings, or both [26][27][28][29][30][31][32][33] . The previously reported machine learning model was developed with linear discriminant analysis and RF using six variables selected by multivariate analysis, including confirmatory test results and CT findings, and was assumed to be primarily targeted at specialized facilities where AVS is unavailable 33 . Although the algorithms reported previously can be useful in specialized centers, they are not commonly be used for predicting subtype diagnosis of patients with suspected PA in general practice.…”

Recent Development toward the Next Clinical Practice of Primary Aldosteronism: A Literature Review

“…Additionally, age at diagnosis and baseline renin levels were included in previously reported prediction scores [ 119 , 142 ], whereas the sample size of those studies was not enough for validation. Recently, Burrello et al conducted a relatively large study to develop and validate prediction models for PA subtyping, using machine learning algorithms [ 143 ]. In the study, the researchers included 215 and 118 PA patients, respectively, for the development cohort and the external cohort.…”

“…In the study, the researchers included 215 and 118 PA patients, respectively, for the development cohort and the external cohort. Based on the discriminative performance on univariate and multivariate logistic regression analysis, six parameters (baseline PAC, PAC post-confirmatory test, lowest serum potassium, the presence of adrenal nodules and its maximum diameter, and CT findings of adrenal glands) were selected for subsequent analysis, including a linear discriminant analysis model and random forest model [ 143 ]. Finally, a 20-point score developed for subtype differentiation, named as SPACE score, demonstrated a high discriminative performance with a maximum accuracy of 89.3% at the score cut-off of 12 in internal validation, whereas the diagnostic accuracy decreased to 78.8% in external validation [ 143 ].…”

Best Achievements in Pituitary and Adrenal Diseases in 2020

“…However, adrenal vein sampling is an invasive test, and the rate of successful cannulation was around 70% even with an experienced interventional radiology team. A clinical prediction score was developed for subtyping primary aldosteronism without adrenal vein sampling [15]. Six parameters (higher aldosterone at screening and after confirmatory testing, lower potassium level, presence of nodules, higher nodule diameter, and absence of contralateral adrenal nodule) were indicative of unilateral pri-mary aldosteronism.…”