Integration of clinical parameters and CT-based radiomics improves machine learning assisted subtyping of primary hyperaldosteronism.

Objectives: The aim of this study was to investigate an integrated diagnostics approach for prediction of the source of aldosterone overproduction in primary hyperaldosteronism (PA). Methods: 269 patients from the prospective German Conn Registry with PA were included in this study. After segmentation of adrenal glands in native CT images, radiomic features were calculated. The study population consisted of a training (n = 215) and a validation (n = 54) cohort. The k = 25 best radiomic features, selected using maximum-relevance minimum-redundancy (MRMR) feature selection, were used to train a baseline random forest model to predict the result of AVS from imaging alone. In a second step, clinical parameters were integrated. Model performance was assessed via area under the receiver operating characteristic curve (ROC AUC). Permutation feature importance was used to assess the predictive value of selected features. Results: Radiomics features alone allowed only for moderate discrimination of the location of aldosterone overproduction with a ROC AUC of 0.57 for unilateral left (UL), 0.61 for unilateral right (UR), and 0.50 for bilateral (BI) aldosterone overproduction (total 0.56, 95% CI: 0.45-0.65). Integration of clinical parameters into the model substantially improved ROC AUC values (0.61 UL, 0.68 UR, and 0.73 for BI, total 0.67, 95% CI: 0.57-0.77). According to permutation feature importance, lowest potassium value at baseline and saline infusion test (SIT) were the two most important features. Conclusion: Integration of clinical parameters into a radiomics machine learning model improves prediction of the source of aldosterone overproduction and subtyping in patients with PA.

Characteristics of preoperative steroid profiles and glucose metabolism in patients with primary aldosteronism developing adrenal insufficiency after adrenalectomy.

Treatment of choice in patients with unilateral aldosterone producing adenoma (APA) is adrenalectomy. Following surgery, most patients retain normal adrenal function, while some develop adrenal insufficiency (AI). To facilitate early detection and treatment of AI, we aimed to identify variables measured pre-operatively that are associated with post-operative AI. Variables obtained from 66 patients before and after surgery included anthropometrical data, clinical chemistry, endocrine work-up. LC-MS/MS steroid hormone profiles from tests before surgery (ACTH-stimulation, saline infusion, dexamethasone suppression) were obtained. Based on 78 variables, machine-learning methods were used in model fitting for classification and regression to predict ACTH-stimulated cortisol after surgery. Among the 78 variables, insulin concentration during pre-operative oral glucose tolerance test (OGTT) correlated positively, and dexamethasone suppressed glucocorticoids correlated negatively with ACTH-stimulated cortisol after surgery. Inclusion of LC-MS/MS measurements allowed construction of better models associated with the occurrence of AI in the training data, but did not allow reliable prediction in cross-validation. Our results suggest that glucocorticoid co-secretion (low insulin during pre-operative OGTT and insufficient suppression of glucocorticoids following dexamethasone) are correlated with the development of post-operative AI. Addition of steroid profiles improved the accuracy of prediction, but cross validation revealed lack of reliability in the prediction of AI.