To the Editor:
We read with interest the review by Tsoumpou et al (1) that continues the debate on the surgical treatment of endometrioma and the effect of such surgery on the outcome of assisted reproductive techniques (ART).
Although they demonstrate no significant difference in the primary outcome of clinical pregnancy rate/cycle (OR 1.34; CI 0.82-2.20), what their careful analysis most clearly demonstrates is the paucity of robust data on which clinical decisions are made. Only five studies compare women with endometrioma with women with untreated endometrioma, only three presented data on the primary outcome and only one study presented data on the more appropriate outcome of livebirth rate. Furthermore, none were randomized controlled trials, cyst size is undocumented in three and the lack of documentation in any study of time period between surgical removal and treatment cannot exclude the possibility of early recurrence.
Although we concur that an adequately powered multicenter RCT would overcome some of the limitations, the current evidence does not lend itself to “at-a-glance” guidance of the management of endometrioma incorporating arbitrary cut-offs such as the >5cm for ovarian cystectomy prior to IVF suggested, not dissimilar to the “Good Practice Point” 4cm cut-off advocated by the ESHRE consensus guideline. Such arbitrary cut-offs neglect the importance of managing endometrioma within the greater context of the infertile couple where endometriosis is one of potentially many relevant factors (clinical, resource and patient driven).
The superiority of surgical excision of endometrioma over ablation in terms of symptom recurrence (OR 0.08-0.15) and recurrence rate (OR 0.41) is well documented (2). Similarly the subsequent decrease in risk of repeat surgery with excision (OR 0.21) appears to overcome any postulated increase in risk of destruction of normal ovarian tissue, supported by the increased spontaneous pregnancy rate following excision (OR 5.21).
Despite these advantages it remains clear that a finite volume of normal ovarian tissue will be destroyed at an excisonal procedure in keeping with the ‘step-reduction’ in the follicular pool theory of early ovarian aging (3). However, this potential effect on ovarian reserve but not egg quality would be unlikely to affect short-term outcome.
Applying an assumptive mathematical model (surface area of endometriotic sphere of 4πr2), Figure 1 demonstrates the increase in surface area of cyst wall removed and potential ovarian volume lost (using a maximum ovarian tissue thickness of 0.8mm (4)) as cyst diameter increases. Based on this loss of increasing follicular reserve with delay in endometrioma excision, we would support early surgery in all symptomatic women, supported by evidence that resolution of endometriomas greater than 3cm in diameter is rarely seen even after six months of medical therapy (5). The decision in asymptomatic women in whom ART is required is less clear-cut and current evidence does not support absolute cut-offs.
We therefore advocate the focus of future studies changing to address potential effects of endometriosis in the short term on markers of oocyte quality rather than quantity and the effect of surgery on long-term markers of ovarian reserve.James D.M. Nicopoullos, MBBSa,b Dimitrios Nikolaou, MDa,b
Robert Richardson, FRCOGb aAssisted Conception Unit bEndometriosis and Minimal Access Therapy Team, Department of Gynaecology Chelsea & Westminster Hospital London, United Kingdom
1. Tsoumpou I, Kyrgiou M, Gelbaya TA, Nardo LG. The effect of surgical treatment for endometrioma on in vitro fertilization outcomes: a systematic review and meta-analysis. Fertil Steril 2008; In Press DOI: 10.1016/j.fertnstert.2007.10.053.
2. Hart RJ, Hickey M, Maouris P, Buckett W. Excisional surgery versus ablative surgery for ovarian endometriomata. Cochrane Database Syst Rev 2008;2:CD004992.
3. Nikolaou D, Templeton A. Early ovarian ageing. EJOG 2003;113:126-133.
4. Muzii1 L, Bellati F, Bianchi A, Palaia I, Manci N, Zullo MA, Angioli R, Panici PB. Laparoscopic stripping of endometriomas: a randomized trial on different surgical techniques. Part II: pathological results. Hum Reprod 2005; 20:1987–92.
5. Buttram VC, Reiter RC, Ward S. Treatment of endometriosis with Danazol: report of a 6-year prospective study. Fertil Steril 1985;43:353-60.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2008.12.145
The Authors Respond:
We agree with the Dr. Nicopoullos and co-authors about the paucity of robust data for the management of endometrioma prior to IVF. This was clearly expressed in our conclusion (1). The development of guidelines is important to help women and fertility specialists in decision making. Clearly, such guidelines will be based on good practice due to absence of evidence. Anecdotally we feel that there is currently great variation in the practice of management of endometrioma prior to IVF. We have stated in the conclusion of the manuscript that all the therapeutic options, including conservative, medical or surgical treatments, as well as their advantages and disadvantages should be fully discussed with the patient. Any decision for surgery should be carefully considered and balanced against the risks, especially in women with previous adnexal surgery or women with suboptimal ovarian reserve. Our guidance on the management of endometrioma depends mainly on whether the woman had previous surgery or not. The arbitrary cut-off we proposed is based on our clinical experience.
Our review did not compare excision of endometrioma with ablation. Dr Nicopoullos and co-authors referred to a Cochrane review that showed evidence of increased spontaneous pregnancy rate following excision of endometrioma compared with ablation (not expectant management) in women who subsequently attempted to conceive (2). The review concluded that there is insufficient evidence to determine the favoured surgical approach in women who may subsequently undergo fertility treatment. We do not agree with Dr. Nicopoullos and co-authors that the increase in pregnancy rate appears to overcome any postulated risk of destruction of normal ovarian tissue. There is no evidence to support the paradigm that reduced ovarian performance is unlikely to be immediately clinically apparent and unlikely to impact on an IVF cycle immediately after surgery. IVF may not be successful in the first cycle. And indeed, it may not be successful at all for some patients. Therefore, the effect of surgery on ovarian reserve, whether it is immediate or delayed should be carefully discussed with the woman. The consent form should include a possibility of a loss of normal ovarian tissue during surgery.Ioanna Tsoumpou, MB, ChBa Maria Kyrgiou, MDb Tarek A. Gelbaya, MDc Luciano G. Nardo, MDc aDepartment of Obstetrics and Gynecology Royal Lancaster Infirmary Lancaster, United Kingdom bDepartment of Obstetrics and Gynecology Lancashire Teaching Hospitals Preston, United Kingdom cDepartment of Reproductive Medicine
St. Mary’s Hospital Central Manchester and Manchester Children’s University Hospitals Manchester, United Kingdom
1. Tsoumpou I, Kyrgiou M, Gelbaya TA, Nardo LG. The effect of surgical treatment for endometrioma on in vitro fertilization outcomes: a systematic review and meta-analysis. Fertil Steril. 2008; In Press DOI: 10.1016/j.fertnstert.2007.10.053.
2. Hart RJ, Hickey M, Maouris P, Buckett W.Excisional surgery versus ablative surgery for ovarian endometriomata. Cochrane Database Syst Rev 2008; 16 (2):CD004992.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2009.06.001