The recent recommendations from the joint ESHRE/ASRM expert panel regarding the definition of PCOS (the so-called Rotterdam Criteria) have given rise to considerable controversy, particularly over the inclusion of anovulatory women with polycystic ovaries (PCO) but without evidence of androgen excess. We reviewed the results of a large series of anovulatory women with PCO who have been treated with low-dose FSH at our centre with the aim of investigating the impact of serum testosterone (T) on outcome. Of 222 patients treated, 83 (37%) had a baseline T concentration >2.7 nmol/l (i.e. >2S.D. above the mean of our control population). The ovulation rate was slightly lower in those with elevated T (68%) than in those with normal T (76%; P=0.014, Fishers Exact test) as was the percentage of uni-ovulatory cycles (82 vs 86%; P=0.039). The proportion of cycles abandoned because of ovarian hyperstimulation was higher in the high T group (7% vs 1.5%; P<0.0001). The overall pregnancy rate was similar in the 2 groups (High T: 49% normal T: 48%) (as was the cumulative conception rate) and there was no difference in multiple pregnancies (low in both groups: 2% and 4% respectively) or miscarriage rate. After correction for multiple comparisons, only the difference in the percentage of abandoned cycles remained significant (P<0.001). In summary, a raised serum testosterone concentration had little impact on the response to FSH in anovulatory women, suggesting that, in terms of reproductive implications at least, normo-androgenaemic women with PCO are part of the same spectrum of disorders as those with the classic combination of anovulation and hyperandrogenism.
06 - 07 Nov 2006
Society for Endocrinology