Diagnosing testicular function using 31P magnetic resonance spectroscopy: a current review.

Abstract

Patients with low sperm counts combined with normal concentrations of gonadotrophins, and in whom physical examination and post-ejaculatory urine analysis are normal, present a diagnostic dilemma. This situation can be caused by testicular failure or by ductal obstruction, which have very different clinical prognoses. Ductal obstruction might be correctable by microsurgical vasovaso/vasoepididymostomy, whereas this approach is of no use in primary testicular failure. A possible diagnostic step for these patients is a testicular biopsy to differentiate between hypospermatogenesis and a normal gonad. However, to date testicular biopsy is seldom performed because of its invasive character. An alternative accurate, non-invasive method to assess testicular function could be very helpful in the evaluation of idiopathic azoospermia or idiopathic oligozoospermia. During the past decade, magnetic resonance (MR) spectroscopy has been developed from a scientific tool into a non-invasive clinical diagnostic tool and has also been used to study testicular function. Recent studies have shown that 31P-MR spectroscopy, based upon differences in the ratio of peaks of phosphomonoester to beta-adenosinetriphosphate, is a non-invasive technique able to differentiate between groups of patients with testicular failure and ductal obstruction, and it correlates reasonably well with the averaged mean Johnsen score of testicular biopsy. The role for a non-invasive technique in the diagnosis of male infertility, such as 31P-MR spectroscopy, can be manifold. It serves not only as an alternative for biopsy but can also be used to assess obstruction as the cause of infertility in patients with subnormal sperm counts, and to predict the chances of pregnancy in patients planned for vasovasostomy to correct a prior vasectomy. However, the main limitation to MR spectroscopy becoming a universal clinical diagnostic technique is the limited availability of 1.5 Tesla MR scanners.