Alan Trounson,     DEPARTMENT OF OBSTETRICS AND GYNAECOLOGY, MONASH UNIVERSITY, QUEEN VICTORIA MEDICAL CENTRE, MELBOURNE, AUSTRALIA

Publisher Summary

Successful in vitro fertilization (IVF) may be approached by either of two methods that differ substantially in concept. One method of obtaining normal oocytes for IVF is to rely on the precise determination of periovular changes to predict the exact time of ovulation in the natural or spontaneous ovulatory cycle. This approach necessitates complete reliance on a method for determining the onset of the preovulatory luteinizing hormone surge and requires personnel and facilities to be available 24 hours a day to recover oocytes at the appropriate time. The difficulties of this approach reside in the variable and often unpredictable rise in LH, the absolute necessity of recovering the single mature oocyte in each ovulatory cycle, and the inefficient use of available facilities and personnel. The alternative approach is to control the ovulatory response by the judicious use of clomiphene citrate or exogenous gonadotropins to initiate follicle growth and the administration of human chorionic gonadotropin or gonadotropin releasing hormone to control the time of the final stages of oocyte maturation. This method has many advantages when compared to IVF in the natural cycle.

I The Natural Ovulatory Cycle and IVF

The variability of the time of ovulation for any woman during the menstrual cycle has been often described and may be clearly demonstrated by the data reported by Mcintosh et al. (1980). In their study, the time of the LH surge was recorded for patients attending an artificial insemination by donor semen (AID) service. Most of the husbands of these patients were azoospermic, so that the majority of the women could be considered to have normal fertility and ovulated regularly. They were able to compute the 95% confidence limits for the day of ovulation from the mean menstrual cycle length and sample standard deviation. The validity of their predictions were confirmed by comparing the predicted time of ovulation with subsequent determinations of the day of the peripheral plasma LH surge. When examining these data it is apparent that the variation in the day of ovulation is considerable and if it is necessary to predict ovulation within 2 to 6 hours for oocyte recovery, precise tests and considerable patience are necessary.

During the follicular phase of the ovulatory cycle a number of antral follicles of between 0.8 and 1.5 cm in diameter can be detected in the growth phase by ultrasound but it is unusual to observe more than one follicle larger than 1.7 cm within 3 days of ovulation. It is not known when the final commitment is made as to which follicle is destined for the final growth phase and ovulation in the human. Even though the growth of the ovulating follicle appears to be continuous, it is not known if the increase in follicular diameter is constant or characterized by periods of expansion. Disturbance to constant follicular growth and expansion would result in difficulties for the use of ultrasound, as the primary monitor for assessing the approach of ovulation. Furthermore, the final size of the follicle at the time of ovulation may vary between patients and between ovulatory cycles. This would appear to be the case in studies with our own patients. The system for monitoring approaching ovulation should not rely on a single parameter, but rather a combination of a number of indicators.