Defining poor ovarian response during IVF cycles, in women aged <40 years, and its relationship with treatment outcome

BACKGROUND: Poor ovarian response limits IVF success but assessing interventions is difficult because of the wide variation in definition. This study attempts to derive objective definitions of poor response. METHODS: A retrospective study of a consecutive series of 1190 patients aged <40 years undergoing their first IVF/ICSI cycle was undertaken. Factors adversely affecting implantation, including advanced female age, were excluded. Clinical outcome in cycles reaching oocyte retrieval (n = 1036) were evaluated with respect to gonadotrophin dose used and oocyte number. Cancelled cycles (n = 154) were analysed in relation to the stimulation dose at cancellation and outcome of their subsequent cycle. RESULTS: Cycle cancellation for patients on >/=300 IU FSH/day compared to those on a lower dose was associated with a significantly worse outcome in the subsequent cycle. If <3000 IU FSH/cycle were administered, clinical pregnancy rates remained favourable if <4 eggs were recovered (29 versus 33% for >/=5 eggs). By contrast, if >/=3000 IU FSH was required, the pregnancy rate was 25% if >/=5 eggs were recovered but declined to 7% if <4 were obtained. CONCLUSIONS: Definitions of poor response should include the degree of ovarian stimulation used. A low oocyte number is only detrimental if the cumulative dose is >3000 IU FSH. Cancellation at >/=300 IU FSH/day is associated with a significantly worse prognosis and could define poor response.