In vitro fertilisation for unexplained subfertility.

BACKGROUND: In vitro fertilisation (IVF) is now a widely accepted treatment for unexplained infertility (RCOG 1998). However, with estimated live-birth rates per cycle varying between 13% and 28%, its effectiveness has not been rigorously evaluated in comparison with other treatments. With increasing awareness of the role of expectant management and less invasive procedures such as intrauterine insemination, concerns about multiple complications and costs associated with IVF, it is extremely important to evaluate the effectiveness of IVF against other treatment options in couples with unexplained infertility.

OBJECTIVES: The aim of this review is to determine, in the context of unexplained infertility, whether IVF improves the probability of live-birth compared with (1) expectant management, (2) clomiphene citrate (CC), (3) intrauterine insemination (IUI) alone, (4) IUI with controlled ovarian stimulation, and (5) gamete intrafallopian transfer (GIFT).

SEARCH STRATEGY: We searched the Cochrane Menstrual Disorders and Subfertility Group Trials Register (searched 23 March 2004), the Cochrane Central Register of Controlled Trials (Cochrane Library Issue 3, 2004), MEDLINE (1970 to August 2004), EMBASE (1985 to August 2004) and reference lists of articles. We also handsearched relevant conference proceedings and contacted researchers in the field.

SELECTION CRITERIA: Only randomised controlled trials were included. Live-birth rate per woman was the primary outcome of interest.

DATA COLLECTION AND ANALYSIS: Two reviewers independently assessed eligibility and quality of trials.

MAIN RESULTS: Ten randomised controlled trials were identified. In two we could not extract data separately for unexplained infertility cases, two were non-randomised, one did not report valid rates (included in the review but not in the meta-analysis); leaving four trials for analysis. One trial compared two different interventions (IVF versus IUI with or without ovarian stimulation) and one study compared three interventions (IVF versus IUI with ovarian stimulation and GIFT). The numbers of trials assessing the effectiveness of IVF with the other treatments were as follows: IVF versus expectant management (two), IVF versus IUI (one), IVF versus IUI with ovarian stimulation (two) and IVF versus GIFT (three). Live-birth rate per woman was reported in three studies and three studies determined clinical pregnancy rate per woman. Multiple pregnancy rate was reported in three trials. Two studies reported ovarian hyperstimulation syndrome (OHSS) as an outcome measure. There were no comparative data for clomiphene citrate and no comparative data on live-birth rates for GIFT. There was no evidence of a difference in live-birth rates between IVF and IUI either without (OR 1.96; 95% CI 0.88 to 4.4) or with (OR 1.15; 95% CI 0.55 to 2.4) ovarian stimulation. There were significantly higher clinical pregnancy rates with IVF in comparison to expectant management (OR 3.24; 95% CI 1.07 to 9.80). There was no significant difference between IVF and GIFT for the one RCT that reported live-birth rates (OR 2.57; 95% CI 0.93 to 7.08). However, there was a significant difference in the clinical pregnancy rates between IVF and GIFT, with pregnancy rates greater for IVF (OR 2.14; 95% CI 1.08 to 4.2). There was no evidence of a difference in the multiple pregnancy rates between IVF and IUI with ovarian stimulation (OR 0.63; 95% CI 0.27 to 1.5), however, IVF had a higher rate than GIFT (OR 6.3; 95% CI 1.7 to 23). Clinical heterogeneity was present among the studies included. However, there was no evidence of statistical heterogeneity, which allowed the studies to be combined for statistical analysis.

AUTHORS' CONCLUSIONS: Any effect of IVF relative to expectant management, clomiphene citrate, IUI with or without ovarian stimulation and GIFT in terms of live-birth rates for couples with unexplained subfertility remains unknown. The studies included are limited by their small sample size so that even large differences might be hidden. Live-birth rates are seldom reported. Periods of follow up are inadequate and unequal. Adverse effects such as multiple pregnancies and ovarian hyperstimulation syndrome have also not been reported in most studies. Larger trials with adequate power are warranted to establish the effectiveness of IVF in these women. Future trials should not only report rates per woman/couple but also include adverse effects and costs of the treatments as outcomes. Factors that have a major effect on these outcomes such as fertility treatment, female partner's age, duration of infertility and previous pregnancy history should also be considered.