The Incidence of Negative Laparoscopy for Pelvic Pain Stratified by Level of Training and Location of Service Provision

The authors have nothing to report.

A totla of 1309 women underwent a laparoscopy for pelvic pain between January 2018 and July 2023. Of these, 174/1309 (13%) had a negative laparoscopy (Figure  1 ). Proportion of negative laparoscopy over 5 years in the two public and private sector sites. In reference to both Figures  1 and 2 , the Z test for comparison of two proportions between sequential year proportions reveals no statistical significance in negative laparoscopy rates between each year group of the study. No differences between the proportions per year were noted on the chi‐square test for multiple proportions ( p  = 0.21, χ 2  = 7.43). Variation in the rates of negative laparoscopy over 5 years in the two public and private sector sites. Z test for comparison of two proportions between sequential year proportions reveals no statistical significance in negative laparoscopy rates between each year group of the study. No differences between the proportions per year were noted on the chi‐square test for multiple proportions ( p  = 0.21, χ 2  = 7.43). As shown in Table  1 , a statistically significant increased proportion of privately funded patients in the positive laparoscopy group compared to the negative laparoscopy group. Demographic data of both positive and negative laparoscopies over 5 years across public and private sectors. Note: Bold values indicates ≤ 0.05. A statistically significant increased proportion of privately funded patients in the ‘fellowship trained private’ group was noted, as demonstrated in Table  2 . Demographic data of all negative laparoscopies over 5 years subcategorised into Non fellowship trained public vs. fellowship trained public vs. fellowship trained private. Note: Bold values indicates ≤ 0.05. As shown in Table  3 , there is a statistically significant difference in the rate of negative laparoscopy between the public (non‐fellowship), public (fellowship) and private (fellowship) sectors. Variables associated with negative laparoscopy—public (non‐fellowship trained gynaecologists) vs. public (AGES) vs. private. Note: Bold values indicates ≤ 0.05. Negative laparoscopy was more likely to occur in non‐fellowship trained gynaecologists compared with fellowship trained (AGES) laparoscopists (OR = 2.48; 95% CI 1.76–3.43, p  < 0.05) and the rate of negative laparoscopy in the public sector 148/867 (17%) was significantly higher than the private sector 26/442 (6%) (OR 3.3, 95% CI 2.13–5.08, p  < 0.05). Of those with a negative laparoscopy, fellowship trained surgeons (77/87) compared to non‐fellowship trained were more likely to perform laparoscopy due to patient preference (53/87) (OR = 3.49; 95% CI 1.54–7.88, p  < 0.05). Similarly, surgeons in the private sector (24/26) were more likely to perform a laparoscopy due to patient preference compared to public surgeons (106/148) (OR = 4.75; 95% CI 1.08–21.01, p  < 0.05). Patients with a negative laparoscopy were more likely to be undergoing their first laparoscopy in the public sector (112/148) compared to the private sector (12/26) (OR = 3.62; 95% CI 1.54–8.59, p  < 0.05) but there was no difference in the negative laparoscopy rate in patients having a primary laparoscopy between fellowship‐trained surgeons (61/87) and non‐fellowship‐trained surgeons (63/87) (OR = 1.11; 95% CI 0.58–2.16, p  = 0.73). In patients with a negative laparoscopy, there was a statistically significant reduction in the number of patients with previous biopsy of confirmed diagnosis of endometriosis between public (11/36) and private (11/14) sectors (OR = 0.12; 95% CI 0.03–0.52, p  < 0.05) and this difference was also demonstrated amongst non‐fellowship trained (7/24) compared to fellowship trained gynaecologists (15/26) (OR = 3.3; 95% CI 1.02–10.72, p  < 0.05). In a negative laparoscopy, private sector surgeons (26/26) were more likely to take a peritoneal biopsy than in the public sector (100/148) (OR = 25.58; 95% CI 1.52–428.59, p  < 0.05) and fellowship‐trained gynaecologists (80/87) were more likely to take a peritoneal biopsy than non‐fellowship‐trained gynaecologists (46/87) (OR = 10.18; 95% CI 4.22–24.55, p  < 0.05). Fellowship trained gynaecologists were more likely to suspect endometriosis or ‘possible endometriosis’ in a histologically negative laparoscopy (58/87) than non‐fellowship trained gynaecologists (41/87) (OR = 2.24; 95% CI 1.21–4.14, p  < 0.05). However, there was no significant difference between intraoperative suspicion of endometriosis and possible endometriosis in a negative laparoscopy between the public (81/148) and the private sector (18/26) (OR = 0.54; 95% CI 0.22–1.31, p  = 0.17). The specificity of visual impression of ‘no endometriosis’ was 46/87 (52.9%) in public (non‐fellowship), 21/61 (34.4%) in public (AGES) and 8/26 (30.7%) in the private sector. Multivariable logistic regression analysis demonstrated, adjusting for all other relevant variables (previous confirmed endometriosis, prior laparoscopy, patient vs. surgeon decision, hormonal suppression) confirmed that AGES Fellowship training was independently associated with significantly lower odds of negative laparoscopy. Compared with public non‐fellowship trained gynaecologists, surgery performed by AGES Fellowship trained gynaecologists in public units was associated with reduced odds of negative laparoscopy (OR = 0.58, 95% CI 0.40–0.83, p  = 0.003), while surgery performed in the private sector by AGES Fellowship surgeons was associated with markedly reduced negative laparoscopy rate (OR = 0.23, 95% CI 0.15–0.37, p  < 0.0001). Multivariate analysis demonstrated that the decision to biopsy (OR = 0.18, 95% CI 0.05–0.64, p ≈0.01) and prior histologically confirmed endometriosis (OR = 0.31, 95% CI 0.11–0.89, p ≈0.03) were also associated with reduced odds of negative laparoscopy. Patient vs. Surgeon Decision (OR = 0.59, 95% CI 0.25–1.36, p  = 0.21), hormonal suppression usage (OR = 0.72, 95% CI 0.33–1.57, p  = 0.41) and prior laparoscopy (OR = 0.64, 95% CI 0.29–1.41, p  = 0.27) did not impact on negative laparoscopic rates.

Using the definition of a negative laparoscopy, this study highlights the need to define what an ‘acceptable’ rate of negative laparoscopy is. The overall rate identified here is 13%, which contrasts with a 2023 systematic review assessing 200 women undergoing laparoscopy for pelvic pain identifying a negative laparoscopy rate of 30%–55% [ 8 ] and a separate retrospective cohort study of 296 women with chronic pelvic pain reporting a negative laparoscopy rate of 47.6% [ 9 ]. These variations are likely multifactorial. In this study, patients were selected from two sites, both of which had fellowship trained endo‐gynaecologists involved in surgical decision‐making. The public sector was a tertiary referral hospital for endometriosis and the patients undergoing operations in the private sector were performed by two high volume fellowship trained endo‐gynaecologists, which may have improved patient selection, counselling and intraoperative detection rates. The Australian Endometriosis Clinical Practice Guidelines [ 10 ] and the current ESHRE Guidelines suggest that laparoscopy for endometriosis should be confirmed by histology [ 7 ]. This is not being universally applied, with a total of 27% of laparoscopies in this study not having had a biopsy taken, all notably from the public sector. The non‐fellowship trained gynaecologists did not obtain a biopsy in 47% of their negative laparoscopies in comparison to the 11.4% by the fellowship trained gynaecologists. This may be due to poor intraoperative detection rates and lack of clinician confidence to perform peritoneal biopsies, which may be related to surgeon experience and training. These results raise the question whether more representative peritoneal biopsies need to be performed in the context of a diagnostic laparoscopy to minimise interobserver variability. A 2024 retrospective cohort study reviewing 56 patients undergoing laparoscopic evaluation of pelvic pain identified an 89% discordance between surgeon intraoperative visualisation and pathology [ 11 ]. This supports our overall findings that of those with a histologically negative laparoscopy, the surgeon clinically suspected endometriosis in 47% of cases in public (non‐fellowship), 66% in public (fellowship) and 69% in the private sector. A 2020 retrospective cohort study further supported this by assessing 200 women undergoing laparoscopic evaluation of suspected endometriosis, suggesting that there was a 39% rate of occult microscopic endometriosis in the context of a ‘visually’ negative laparoscopy [ 12 ]. A ‘visual’ diagnosis of endometriosis yielded a sensitivity of 90%, specificity of 40%, positive predictive value of 81% and a negative predictive value of 59% with histopathology identified as the gold standard for women undergoing laparoscopy for pelvic pain—further justifying these findings [ 13 ]. The wide variability in detection rates raises the concern whether a significant proportion of visually negative laparoscopies are in fact histologically ‘positive’ for endometriosis. It may be that higher rates of ‘negative’ laparoscopy described in the literature are not negative at all but represent endometriosis that is unrecognised by the surgeon and again identifies the flaws in a visual classification system. A molecular classification system may be a more appropriate method for identifying early endometriosis [ 14 ] but further study needs to be undertaken in regards to such a system. This includes the impact on symptoms and whether ‘visually negative’ but ‘histologically positive’ tissues are clinically relevant and that this may be an appropriate time for early intervention that changes the symptomatic course. A subgroup analysis was performed to investigate whether prior hormonal uptake and previous laparoscopic excision would increase the rates of negative laparoscopy. Whilst it remains unclear whether hormonal uptake simply suppresses or regresses endometriotic lesions our findings do not reveal any statistically significant correlation regarding the above, considering the inherent limitations of the retrospective study. Surgical impression was pertinent to analyse the proportion of discordant results and further stress the need for peritoneal sampling at the time of a laparoscopy for pelvic pain. There is a significantly higher proportion of surgeons who are performing laparoscopies directed by patients in the private sector compared to those in the public. With consideration needing to be given to the financial incentive that may be associated, for example, excisional or ablative treatment of minimal endometriosis (MBS item number 35637), this is associated with a minimum fee of AUD $463 (282 EUR) [ 15 ]. Whilst this may be a potential contributing factor, patient request is an important consideration as for a proportion of patients, they continue to seek answers for symptoms despite non‐interventional management with surgery a method to achieve that answer. A negative laparoscopy visually and histologically may help them to resolve some aspects of their chronic pelvic pain and the purely surgical/biomedical approach to future management. The strengths of this study include its large sample size and data collection conducted over multiple sites spanning across the public and private sector. This increases its external validity, considering these patients are likely to arise from a variety of socioeconomic demographics. The binary outcome of histological evidence of endometriosis is well defined even in this retrospective setting. Considering the retrospective nature, confounding bias is a limitation. The patients from the private sector represent a skewed population operated on by two fellowship‐trained endo‐gynaecological surgeons, reducing the generalisability of the findings due to a biassed sample of patient selection, surgeon counselling and operative technique. Additionally, all preoperative symptoms were not collected via a uniform questionnaire and several gynaecologists performed pre‐operative visits explaining the heterogeneity of the quality of the documented medical record and increasing the risk of observer bias. A defined standard for the rate of negative laparoscopy is yet to be specified. More evidence, teaching, and research into how to improve surgical intraoperative detection rates of endometriosis are necessary. This must be considered in the face of still having little idea of the pathogenesis of endometriosis and how lesions progress from superficial to deep/invasive [ 16 ], including disease recurrence. Few data exist regarding negative laparoscopy for pelvic pain. A marked discrepancy in negative laparoscopy rates between private and public hospitals potentially relates to patient selection, sociodemographic factors, clinician bias and experience, operative technique and surgical training. Peritoneal biopsy rates also differed substantively, raising concerns around missed superficial peritoneal disease. Peritoneal biopsy should be considered in all patients undergoing laparoscopy for pelvic pain for histological evidence and patient counselling. This study highlights the differences in negative laparoscopy rates between the public and private sector, as well as fellowship and non‐fellowship trained gynaecologists. These data have implications for surgical research and training to improve decision making for women presenting with pelvic pain and suspected endometriosis.

There is considerable variability on the published rate of ‘negative’ laparoscopy (defined as no pathology determined visually and histologically) for the investigation of pelvic pain. Persistent pelvic pain is reported by 47% of Australian women [ 1 ], with 40% of laparoscopies performed as part of the diagnosis [ 2 ]. The cost burden for all laparoscopies in Australia is approximately $16,000 for in network services per patient with more than 40,000 endometriosis‐related hospitalisations [ 3 ]. One of the most common diseases identified in ‘positive’ laparoscopy is endometriosis, with a prevalence of 11.4% by age 45 [ 4 ], 14% by age 49 [ 5 ] and a societal burden of $9.7 billion annually, with $2.5 billion attributed to direct healthcare costs [ 6 ]. Accurate determination of the true negative laparoscopy rate is essential to reduce the financial burden on the healthcare system, improve patient satisfaction and minimise surgical morbidity. Whilst laparoscopic evaluation has been previously considered a routine part of the evaluation, the updated European Society of Human Reproduction and Embryology (ESHRE) Guideline has advised that laparoscopy should only be performed if medical treatment has been deemed inappropriate by the treating team or consumer in the context of suspected endometriosis [ 7 ]. It is therefore necessary to review patient selection, recognise variations in practices, highlight intraoperative detection rates and compare adherence to clinical guidelines. This study aims to report the incidence of ‘negative laparoscopy’ for women presenting with pelvic pain and to consider underlying causes for negative laparoscopies in this context.

J.A. is a member of the Endometriosis Advisory Group to the Australian Government Department of Health and Aged Care and was the chair of the Australian Guideline for the Diagnosis and Management of Endometriosis. He was a contributor to the National Action Plan on Endometriosis for the Australian Government. He is the chair of the National Endometriosis Clinical and Scientific Trials (NECST) Network. He has received consulting fees from Hologic, Gedeon Richter and BD. He is on the advisory boards for Hologic and Gedeon Richter. J.A. has received NHMRC and MRFF competitive research grants for endometriosis studies. Cecilia Ng manages research grant funding from the Australian Government and MRFF and was a previous employee with CSL Vifor (formerly Vifor Pharma Pty. Ltd.).

An ethics approved 2024/ ETH00248 multicentre, retrospective cohort study was conducted for patients undergoing elective laparoscopy for pelvic pain between 2018 and 2023 at two tertiary hospitals in Sydney, Australia. Patients were identified using the Surginet (Cerner Corporation, 2025) electronic database of a tertiary referral training hospital with an Australasian Gynaecological Endoscopic Surgery (AGES) training unit and the specialist private gynaecological clinic of two fellowship trained endo‐gynaecologists. Clinicians at the public hospital included both non‐fellowship and fellowship trained (AGES) gynaecologists. AGES training involves a two‐year accredited programme in advanced endoscopic gynaecological surgery with the intent of standardisation of training, education, and access to resources across Australia and New Zealand to those who have completed the fellowship. Primary data sources were Surginet and Powerchart (Cerner Corporation, 2025) electronic medical records from the public hospital and private clinic. The cohort was identified using the terms ‘laparoscopy’ and ‘laparoscopic excision of endometriosis’. Medicare Benefits Schedule (MBS) codes 35631, 35632, 35637, 35638 and 35641, for laparoscopy, cystectomy, and endometriosis excision of mild, moderate and severe disease were analysed. All elective surgeries during this timeframe were identifiable within the records. Inclusion criteria included women undergoing diagnostic laparoscopy between January 2018 and July 2023 for persistent pelvic pain symptoms including dysmenorrhoea, dyschezia, dyspareunia and non‐menstrual pelvic pain. Women undergoing laparoscopy for other indications were excluded. Data were collected electronically, entered into a REDCap database, and securely stored via OneDrive at The University of New South Wales, Sydney. The primary outcome assessed was the rate of negative laparoscopy over the 5‐year study duration. Negative laparoscopy was defined as surgery where no pathology is identified, either intraoperatively by surgeon visualisation or negative histopathology. The secondary outcomes analysed were: presenting symptoms, decision for surgery, number of previous laparoscopies for pelvic pain, results from a prior laparoscopy, uptake of hormonal suppression prior to surgery, types of hormonal suppression, uptake of peritoneal biopsy at the time of laparoscopy, reported histopathology of a negative biopsy and intra‐operative impression. All primary and secondary outcomes were stratified based off the level of training of the primary clinician that is, public (non‐fellowship), public (fellowship), private (fellowship). A further subgroup analysis was performed investigating the indication for the decision making around surgery, number of previous surgeries, outcome from previous laparoscopy, uptake of peritoneal biopsy and the primary surgical impression at the time of the laparoscopy. All data were collected based on documentation by healthcare providers in the electronic medical record. Statistical analysis was performed using Statistical Package for the Social Sciences (SPSS Version 29, IBM). Simple descriptive data are used to report outcomes. Data for this study were analysed using odds ratio for proportional comparison and Chi‐square test, which were utilised to determine differences between multiple populations. A multivariate regression analysis was performed to control for potential confounders influencing negative laparoscopy rates.