The concordance between serum anti-Mullerian hormone and testosterone concentrations depends on duration of hCG stimulation in boys undergoing investigation of gonadal function

Background In boys undergoing investigation of gonadal function, the relationship between a single measurement of serum anti-Mullerian hormone (AMH) and hCG stimulated serum testosterone is unclear. Aim The aim of the study was to assess concordance between serum AMH and testosterone concentrations following hCG stimulation of two different durations. Methods Samples from 284 children (M : F, 154 : 130) with a median age of 8 years (10th, 90th centiles, 0 25, 14) were used to establish an AMH reference range. Clinical data were reviewed in boys undergoing investigation of gonadal function and who had an AMH measurement and a hCG stimulated (3-day or 3-week) (n - 26) testosterone. Of these 26 boys, 11 had combined genital anomalies, whereas the rest had conditions such as isolated hypospadias, undescended testes or microphallus. Normal testosterone response to hCG stimulation was defined as a level greater than 3 5 nmol at day 4 and 9 5 nmol/l at day 22. Results In the reference group, the 5th centile AMH for boys below 1 year was 215 pmol/l and between 1 and 8 years 180 pmol/l. The 95th centile for girls for these respective age groups was 30 pmol/l and 25 pmol/l. In those cases where serum testosterone concentrations were available at day 1, day 4 and day 22 of the 3 week-hCG test, five cases had a normal serum testosterone at day 4 and three cases only showed such a response by day 22. In those where serum AMH was less than 180 pmol/l, a poor testosterone response of less than 3 5 nmol was observed in approximately seven of eight (88%) cases with a 3-day hCG stimulation test or the 3-week test. An AMH of greater than 180 pmol/l was associated with a normal testosterone response at day 4 in 10 out of 15 (67%) cases and at day 22 in eight of 11 (73%) cases. However, a low serum testosterone concentration of less than 3 5 nmol after the 3-day hCG test was only associated with a likelihood of a low AMH in three of eight (37%) cases. With the 3-week hCG test, a low day 22 testosterone of 9 5 mmol/l or less was associated with a low AMH of 180 pmol/l or less in four of seven (57%) cases. Conclusion In boys undergoing investigation of gonadal function, the concordance between AMH and testosterone is better at day 22 than day 4. A normal AMH may provide useful information on overall testicular function but does not exclude the need for an hCG stimulation test.