Oocytes retrieval difficulties in women with ovarian endometriomas.

Study question: What is the impact of the presence of ovarian endometriomas on the technique of oocytes retrieval? Summary answer: The presence of ovarian endometriomas is associated with increased technical difficulties during oocytes retrieval. What is known already: Considering that ovarian reserve may be injured following surgery for ovarian endometriomas, conservative management has grown in recent years. There is now an increasing conformity that endometriomas, in particular those with a mean diameter below 4 cm, should not be systematically removed before IVF. However, conservative management is not without potential drawbacks and risks. A neglected issue here is the possibility that the presence of these cysts may interfere with oocytes retrieval. In particular, one may hypothesize that the procedure may be more difficult and dangerous. Study design, size, duration: Prospective cross-sectional study. All women undergoing oocytes retrieval between January and December 2015 in a single institution were recruited. Cases were women with an ultrasound diagnosis of one or more endometriomas. Controls were women without a diagnosis of endometriosis. Women could be included only for their first treatment cycle. Participants/materials, setting, methods: We recruited 56 women with endometriomas and 227 unaffected controls. Each clinician was requested to complete a questionnaire regarding the technical difficulties encountered during the oocytes retrieval. Moreover, patients were interviewed two weeks after the procedure to assess complications occurring after they were discharged. A binomial distribution model was used to calculate the 95% Confidence Interval (95%CI) of the most relevant proportions. Main results and the role of chance: The endometrioma had to be transfixed during oocytes retrieval in 8 cases (14%, 95%CI: 7–25%). Follicular fluid contamination with the endometrioma content occurred in 9 cases (16%, 95% CI: 8–27%). The number of retrieved oocytes was significantly inferior in cases compared to controls (5.6 ± 4.4 and 7.2 ± 5.2, respectively), (p = 003). However, the proportion of retrieved oocytes per aspired follicles did not differ. The median (interquartile range) was 61% (40–80%) and 67% (50–87%), respectively (p = 0.31). The procedure of oocytes retrieval was scored as difficult or very difficult in 5 (9%) cases and 8 (4%) controls (p = 0.14). The aspiration of follicles was not complete (some follicles were not punctured because of technical difficulties) in 8 cases (14%) and 10 controls (4%) (p = 0.01). The corresponding Odds Ratio (OR) was 3.6 (95%CI: 1.4–9.6). Finally, four women with endometriomas (7%) and four controls (2%) referred to the hospital during the two weeks period after oocytes retrieval because of pain (p = 0.05), corresponding to an OR of 4.3 (95%CI: 1.0–17.8). None of these cases was diagnosed with clinically relevant complications and all recovered with pain killers. No ovarian abscess occurred (0%, 95%CI: 0–5%). Limitations, reasons for caution: Personnel was aware of the condition of the patients and study purposes. It cannot thus be totally excluded that differences could have been inflated for some of the outcomes. Type II errors cannot be fully excluded in some outcomes due to the relatively small samples size. Wider implications of the findings: The presence of ovarian endometriomas exposes women to higher risks. However, the magnitude of the observed differences is modest and we did not observe any clinically demanding complication. Advocating systematic surgery before IVF based on this evidence is thus not justified. Trial registration number: Not applicable.