Endocrine Abstracts (2019) 63 P1126 | DOI: 10.1530/endoabs.63.P1126

Hormonal predictors of pathologic findings at magnetic resonance imaging in secondary hypogonadal men

Sarah Cipriani1, Giovanni Corona2, Mario Maggi1 & Giulia Rastrelli1


1University of Florence, Florence, Italy; 2Maggiore Hospital, Bologna, Italy.


Introduction: Secondary hypogonadism (sHG) is the most common form of hypogonadism. sHG can arise from any dysregulation of hypothalamic-pituitary (HP) axis due to functional or organic disorders. HP organic disorders are often diagnosed using pituitary magnetic resonance imaging (MRI). However, MRI is an expensive and not widespread exam and it cannot be routinely offered to all sHG patient. The Endocrine Society guidelines suggest MRI scan in sHG men with total testosterone (TT) below 5.2 nmol/l. However, this suggestion is not supported by robust experimental evidence.To evaluate whether hormonal parameters can predict a pathologic finding at MRI in sHG men.

Methods: A consecutive series of 126 men (exploratory sample) attending the Outpatient Clinic of the University of Florence for sexual dysfunction and diagnosed as sHG according to the European Male Ageing Study criteria [LH<9.4 U/l and TT <10.5 nmol/l] performed a pituitary MRI. A cohort of 50 men attending the Outpatient Clinic of the Endocrinology section of the Maggiore Hospital of Bologna for sexual dysfunction and diagnosed as sHG was used as a validation sample.

Results: Among men of the exploratory sample, 46 reported pathologic findings at pituitary MRI (15 microadenomas, six macroadenomas, two pituitary hypoplasia, two Rathke’s cleft cysts, four pituitary stalk diseases, one radiology signs of iron overload and 16 empty sella). These men did not differ from those with normal MRI, except for LH, FSH and TT. In the exploratory sample, the ROC curve analysis for the accuracy of TT in predicting a pathologic MRI was 0.62[0.52;0.73], P=0.021) and the best threshold was identified by the Youden index at 6.0 nmol/l (sensitivity=48.9% and specificity 83.1%). The threshold of 5.2 nmol/l proposed by the Endocrine Society had a sensitivity of 36.2% and a specificity of 94.4%. Concerning LH and FSH, their accuracy was 0.65[0.55;0.75], P=0.007 and 0.68[0.58;0.77], P=0.002, respectively) showing the best threshold value at 1.9 U/l for LH (sensitivity=61.4% and specificity 62.0%) and 4.2 U/l for FSH (sensitivity=90.2% and specificity 43.8%). When applying these thresholds in the validation sample, TT<6.0 nmol/l and LH<1.9 U/l adequately recognized pathologic MRI, whereas TT<5.2 nmol/l and FSH<4.2 U/l did not.

Conclusions: In sHG subjects consulting for sexual dysfunction, hormonal parameters can help in recognizing men who deserve performing pituitary MRI. In particular, TT<6.0 nmol/l and LH<1.9 U/l adequately recognize patients with hypothalamic-pituitary organic disorders.

Article tools

My recent searches

No recent searches.