Evidence-based reproductive surgery: endometriosis

In order to answer the question as to whether surgical removal of implants should be performed at all, one should first address the question as to whether endometriosis per se affects fertility. For this, one could study spontaneous pregnancy rates in untreated control subjects in randomized controlled trials (RCTs). Six trials have been published with respect to medical treatment; two reported surgical treatment. The overall pregnancy rate in the (untreated) controls of all eight RCTs together was 28% [95% confidence interval (CI), 24-33%]. This rate does not differ significantly from the one reported by Taylor and Collins in their review of 20 studies of 2026 couples with essentially unexplained subfertility of 33% (95% CI, 31-35%). Two RCTs studied the effect on pregnancy rates of surgical resection or ablation of the endometriosis lesions. Neither study allowed for estimating the effect of ablation of lesions on pregnancy chances since, in both studies, apart from ablation of the lesions, lysis of adhesions was also performed. The larger of the two studies did show an overall benefit from surgical removal of the endometriotic implants; the smaller one did not. We conclude that surgery for minimal or mild endometriosis might modestly enhance fecundity in women with otherwise unexplained subfertility, but it cannot be excluded that this improvement is due to removal of adhesions rather than implants. (C) 2004 Published by Elsevier B.V.