Dear Editor,
We thank the Bano et al.1, for their interest in our article. We appreciate the comments on the lack of details of type, dosage and duration of gonadotrophin releasing hormone (GnRH) agonist treatment; location, size and depth of lesions; fertility outcomes and surgeon experience. Our study was a pragmatic analysis of a very large, multicentre surgical database [British Society for Gynaecological Endoscopy (BSGE) Endometriosis Centres Database] from 101 centres, including 9,433 patients.2 Although the data were entered from six different countries, the vast majority of the entries were from the United Kingdom.
The database did not include the type, dosage or duration of GnRH agonists. However, in practice, the majority of women would have been on long term depot preparations such as leuprorelin acetate 3.75 mg/month or 11.25 mg/3 months, goserelin acetate 3.6 mg/month or 10.8 mg/3 months, or triptorelin 11.25 mg/3 months. The decision to use GnRH agonists, including the specific agent, dosage, and duration, was according to local protocols and clinician discretion. While we agree that stratification by agent and duration could provide further insights, this analysis was not possible. Importantly, the lack of uniformity reflects actual clinical heterogeneity and enhances external validity, while the large sample size provides robustness.
The lesion characteristics were collected to include the location but not depth or size of deep endometriotic nodules. Data regarding the size and depth of deep nodules are disreputably difficult to collect, especially when different teams use different diagnostic methods. In addition, surgical estimates regarding size of these lesions are not reliable. Data collection in the BSGE database started in 2009, when the concept of nodule size was only starting to develop (and the first version of the ENZIAN classification did not exist).3, 4 Thus, analysis of outcomes according to lesion depth and size was not possible with the data available. However, we used proxy measures to reflect surgical complexity instead. This included analysis based on presence or absence of bowel resection, bladder or ureteric nodule excision and hysterectomy. Nevertheless, we agree that lesion-specific staging would improve risk stratification and could be incorporated into future registry studies. It is our understanding that a revision of the BSGE database is close to completion and set to include the latest ENZIAN classification. Fertility related outcomes were not collected in the BSGE Database as it was designed to focus on surgical outcomes and health-related quality of life.
We agree that heterogeneity in surgeon experience and perioperative management protocols are important considerations. These parameters are difficult to determine and collect objectively. However, the accredited centres were expected to reach certain requirements, including a minimum number of operations carried out by the centres or individual surgeons annually and submission of an edited video to the BSGE Scientific Advisory Group for assessment every year. The large number of participating centres inevitably introduced practice variability, which we acknowledge as both a strength and a limitation. Further work, including randomised trials and standardised data collection frameworks, is needed to refine and personalise GnRH analogues regimens and predict long-term reproductive and surgical outcomes.
