PITUITARY IMAGING BY MRI AND ITS CORRELATION WITH BIOCHEMICAL PARAMETERS IN THE EVALUATION OF MEN WITH HYPOGONADOTROPIC HYPOGONADISM

Objective: A significant ambiguity still remains about which patient deserves a magnetic resonance imaging (MRI) scan of the pituitary during evaluation of hypogonadotropic hypogonadism (HH) in men. Methods: Retrospective case series of 175 men with HH referred over 6 years. Results: A total of 49.7% of men had total testosterone (TT) levels lower than the Endocrine Society threshold of 5.2 nmol/L. One-hundred forty-two patients (81.2%) had normal appearance of pituitary MRI, whereas others had different spectrum of abnormalities (empty sella [n = 16], macroadenoma [n = 8], microadenoma [n = 8], and pituitary cyst [n = 1]). In men with TT in the lowest quartile, MRI pituitary findings were not significantly different from men in the remaining quartiles (P = .50). Patients with raised prolactin had higher number of abnormal MRI findings (38.9% vs. 13.7%; P = .0014) and adenomatous lesions (macro and micro) (27.8% vs. 43%; P = .01) in comparison to men with normal prolactin. The prolactin levels (median [interquartile range]) were highest in men with macroadenomas in both groups (9,950 [915]; P = .007 and 300 [68.0] mU/L; P = .02, respectively), with concomitant lower levels of other pituitary hormones. Multivariate logistic regression showed an association of abnormal pituitary MRI with insulin-like growth factor 1 (IGF-1) standard deviation score (SDS) (odds ratio [OR], 1.78 [95% confidence interval (CI), 1.15 to 2.77]; P = .009) and prolactin (OR, 1.00 [95% CI, 1.00 to 1.03]; P = .01). Conclusion: MRI of the pituitary is not warranted in all patients with HH, as the yield of identifiable abnormalities is quite low. Anatomic lesions are likely to be present only when low levels of TT (<5.2 nmol/L) are found concomitantly with high levels of prolactin and/or low IGF-1 SDS.