Bryan Naelitz1, Anup Shah2, Amy S Nowacki3, Darren J Bryk2, Nicholas Farber2, Neel Parekh2, Daniel Shoskes2, Betul Hatipoglu4, Sarah C Vij2. 1. Cleveland Clinic Lerner College of Medicine, Education Institute, Cleveland Clinic, Cleveland, Ohio. 2. Department of Urology, Glickman Urological and Kidney Institute, Cleveland Clinic, Cleveland, Ohio. 3. Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, Ohio. 4. Department of Endocrinology, Diabetes, and Metabolism, Cleveland Clinic, Cleveland, Ohio.
Abstract
PURPOSE: We aimed to identify predictor variables associated with pituitary abnormalities in hypogonadal men with mild hyperprolactinemia. We also sought to develop a decision-making aid to select patients for evaluation with pituitary magnetic resonance imaging. MATERIALS AND METHODS: We retrospectively examined men with mild hyperprolactinemia (15.1-50.0 ng/ml) who presented with symptoms of hypogonadism and underwent pituitary magnetic resonance imaging. Demographics, laboratory values and clinical data were obtained from the electronic medical record. Selected predictor variables were included in multivariable logistic regression and partitioning models. Cost avoidance analysis was performed on models achieving sensitivities ≥90%. RESULTS: A total of 141 men were included in the study, of whom 40 (28%) displayed abnormalities on pituitary magnetic resonance imaging. Patients with pituitary abnormalities exhibited higher prolactin (p=0.01), lower testosterone (p=0.0001) and lower luteinizing hormone (p=0.03) levels than those with normal anatomy, as well as higher prolactin-to-testosterone ratios (p <0.0001) and lower luteinizing hormone-to-follicle-stimulating hormone ratios (p=0.0001). These serological variables were identified as the best performing predictor variables. The partition incorporating a prolactin-to-testosterone ratio cutoff of 0.10 and prolactin cutoff of 25 ng/ml achieved 90% sensitivity and 48% specificity, and reduced diagnostic expenses by 28%. CONCLUSIONS: Hypogonadal men presenting with mild hyperprolactinemia and pituitary abnormalities declare themselves via endocrine studies routinely ordered to evaluate these conditions. The prolactin-to-testosterone ratio is the best independent predictor of finding a pituitary abnormality on magnetic resonance imaging, although sensitivity improves by referencing additional serological parameters. Significant cost avoidance may result from screening this population prior to ordering pituitary magnetic resonance imaging.
PURPOSE: We aimed to identify predictor variables associated with pituitary abnormalities in hypogonadal men with mild hyperprolactinemia. We also sought to develop a decision-making aid to select patients for evaluation with pituitary magnetic resonance imaging. MATERIALS AND METHODS: We retrospectively examined men with mild hyperprolactinemia (15.1-50.0 ng/ml) who presented with symptoms of hypogonadism and underwent pituitary magnetic resonance imaging. Demographics, laboratory values and clinical data were obtained from the electronic medical record. Selected predictor variables were included in multivariable logistic regression and partitioning models. Cost avoidance analysis was performed on models achieving sensitivities ≥90%. RESULTS: A total of 141 men were included in the study, of whom 40 (28%) displayed abnormalities on pituitary magnetic resonance imaging. Patients with pituitary abnormalities exhibited higher prolactin (p=0.01), lower testosterone (p=0.0001) and lower luteinizing hormone (p=0.03) levels than those with normal anatomy, as well as higher prolactin-to-testosterone ratios (p <0.0001) and lower luteinizing hormone-to-follicle-stimulating hormone ratios (p=0.0001). These serological variables were identified as the best performing predictor variables. The partition incorporating a prolactin-to-testosterone ratio cutoff of 0.10 and prolactin cutoff of 25 ng/ml achieved 90% sensitivity and 48% specificity, and reduced diagnostic expenses by 28%. CONCLUSIONS: Hypogonadal men presenting with mild hyperprolactinemia and pituitary abnormalities declare themselves via endocrine studies routinely ordered to evaluate these conditions. The prolactin-to-testosterone ratio is the best independent predictor of finding a pituitary abnormality on magnetic resonance imaging, although sensitivity improves by referencing additional serological parameters. Significant cost avoidance may result from screening this population prior to ordering pituitary magnetic resonance imaging.
Authors: S Cipriani; T Todisco; N Ghiandai; L Vignozzi; G Corona; M Maggi; G Rastrelli Journal: J Endocrinol Invest Date: 2021-05-10 Impact factor: 4.256