Association of serum testosterone levels and testicular volume in adult patients

A retrospective observational study was undertaken to gain new insight into the relationship between total testicular volume and levels of serum testosterone, luteinising hormone, follicle-stimulating hormone, prolactin and clinical variables. A total of 312 men with sexual dysfunction or infertility were divided into groups A and B (156 each) on the basis of basal plasma testosterone 5nmol/L of 12nmol/L respectively. Group A was subclassified in A(1) (primary hypogonadism) and A(2) (secondary hypogonadism). There were significant differences in total testicular volume between group A (15.33 +/- 11.94ml) and group B (36.74 +/- 6.9; p<.001) and also between subgroup A(1) (11.07 +/- 8.49ml) and subgroup A(2) (23.62 +/- 13.04ml; p<.001). Only 13.5% of patients in group B had a total testicular volume <30ml. Differences in all studied parameters were found between group A and group B. There were no variations when comparing age, body mass index and testosterone in groups A(1) and A(2). The use of total testicular volume and body mass index together for predicting testosterone levels yields a sensitivity and specificity of 85.3% and 86.5% respectively. Logistic regression analysis, univariate and multivariate models, using the measurement of total testicular volume resulted in a high capacity to predict testosterone levels.