Search for a command to run...
Lukey and colleagues [1] report a population-based cohort of 7102 standalone bilateral salpingectomies in British Columbia (2008–2022), with an overall ‘any complication’ rate of 2.8% and very low rates of ICU admission or return to theatre. We congratulate the authors on their important contribution that adds to the evidence base of complications from standalone salpingectomy with future implications for counselling and informed decision-making for the emerging novel option of risk-reducing salpingectomy (RRS) for ovarian cancer (OC) prevention. The primary composite outcome was usefully broad (maximising sensitivity), including post-discharge physician visits, with few coded with diagnoses ‘likely related’ to complications. In a publicly funded system with low thresholds for postoperative review, this approach may also capture reassurance visits, mild pain or precautionary assessments that many surgical audits would not count as significant complications. It is possible the complication rate may be lower than estimated in their cohort. Importantly, most of the complications were minor and the major complication rate was very low (0.7%). The prophylactic subgroup was small (n = 132), with a numerically marginally higher effect-size for complication rate (4.5% vs. 2.7%) but wide confidence intervals after adjustment. This estimate of 4.7% with RRS is also similar to complication rates reported with risk-reducing salpingo-oophorectomy (RRSO) for OC surgical prevention in some high-risk cohorts. This important paper will likely be cited in discussions about expanding RRS outside opportunistic settings. In the SALSTER randomised trial (RCT) comparing tubal-sterilisation with bilateral (opportunistic) salpingectomy, we previously reported complication rates in women undergoing ‘salpingectomy alone’ without other concurrent procedures [2]. We found a higher overall complication rate of 8.1% using active follow-up in the Swedish population. However, importantly this was non-inferior to tubal sterilisation; the severe complication rate was only 0.2%, and Clavien–Dindo Grade 3–5 complication rate was only 0.4% [2]. While overall complication rates may vary across populations, countries and healthcare settings, reassuringly the major complication rates are consistently low. Nevertheless, minor complications may also affect quality of life, patient satisfaction and acceptability. Separating ‘hard’ or more serious surgical morbidity (return to theatre, ICU admission, visceral injury, haemorrhage, readmission for a surgical complication, etc.) from minor complications and softer administrative proxies (non-specific outpatient codes and utilisation) would further help clinicians translating these findings into patient counselling and guideline deliberations. Another unaddressed issue for the adoption of premenopausal RRS is the unanswered question of potential impact on long-term ovarian function and age at menopause. Epidemiological evidence shows that a year's earlier menopause is associated with incremental increases in cardiovascular mortality (~2%/year), all-cause mortality (~2%/year), osteoporosis, neurocognitive symptoms, sexual dysfunction and reduced quality of life, representing a substantial long-term health burden. Even a modest 1–2 year earlier menopause could thus potentially increase cardiovascular/all-cause mortality, with population-level harms outweighing the cancer-preventive benefit. We previously established the risk thresholds for offering RRSO to be 4%–5% lifetime OC risk [3]. Future research needs to address the OC risk threshold for undertaking RRS along-with patient acceptability studies, and addressing impact on menopause. Some of this work is being undertaken in the recently established Lancet Commission on OC, [4] and studies like MISSION-O. All authors contributed equally. RM is supported by grant funding from Barts Chrity, Rosetrees Trust and Cancer Research UK. The authors declare no conflicts of interest. No primary data are used in this manuscript. The corresponding author can be contacted for any potential queries.