A single centre multidisciplinary team retrospective review of fifty cases of robot-assisted surgery for diaphragmatic and thoracic endometriosis

Introduction Diaphragmatic and thoracic endometriosis (DTE) is considered rare, often presenting with non-specific cyclical thoracic symptoms. Diagnosis and surgical management remain challenging due to the need for cross-specialty expertise. This study reports the first series of robot-assisted surgeries for DTE performed by a single, consistent multidisciplinary team.

This was a retrospective case series of 50 consecutive DTE surgeries performed between July 2020 and March 2023 in a specialist private hospital in the United Kingdom. Procedures involving robot-assisted laparoscopy (RAL) for pelvic/diaphragmatic disease and/or robot-assisted video-assisted thoracoscopic surgery (RAVATS) for thoracic disease were included. All cases involved a multidisciplinary collaboration between gynecological, hepato-pancreato-biliary, and thoracic robotic surgeons. Data on presentation, operative metrics, histology, and 90-day outcomes were collected and analyzed.

Forty-six women underwent 50 procedures; 29 involved RAL only, 13 involved RAL with RAVATS, and 5 were RAVATS only. Median operative time was 236 min for three-compartment cases. No cases required conversion to open surgery. Endometriosis was confirmed histologically in 64.4% of abdominal diaphragm cases and 45% of thoracic cases. The right hemidiaphragm was most commonly affected. No Clavien–Dindo ≥II complications occurred. Combined three-compartment procedures were significantly shorter overall than staged approaches (p = 0.01).

Robot-assisted surgery for DTE is safe, feasible, and may enhance disease recognition and excision, particularly when conducted by a dedicated multidisciplinary team. Simultaneous multi-compartment surgery improves operative efficiency and reduces the risk of incomplete treatment. DTE warrants evaluation in specialized centers, in a multidisciplinary fashion. Abbreviations | | | |---|---| | | | | | | | | | | | | | | | Key Message Robot-assisted multidisciplinary surgery enables safe, efficient, and comprehensive treatment of diaphragmatic and thoracic endometriosis (DTE). Evaluation of DTE in specialized centers, in a multidisciplinary fashion, allows for patient-centered operative planning and could avoid missed disease or incomplete management. 1 INTRODUCTION Diaphragmatic and thoracic endometriosis (DTE) has long been considered to be a rare phenomenon, with individual case reports still frequently published today.1 While extra-pelvic endometriosis was first classified by Markman in 1989,2 it has recently been described in more detail.3 There exist multiple small case series, often from thoracic surgeons4; a multicenter French review of 50 patients,5 and one larger Italian series of 215 cases over a 15-year period.6 None of these series consist of patients operated on by a single multidisciplinary team. The reason for this perception of rarity is likely the result of women suffering with pelvic pain and endometriosis not offering thoracic symptoms to their gynecologists, and instead seeking help from other specialities whom may be unaware of the implication of cyclical symptoms. Catamenial pneumothorax, hemothorax and hemoptysis are less common, but not rare, signs of possible DTE, and these more unusual cases tend primarily to reach the doors of thoracic surgeons who are often not readily equipped with the skills to recognize endometriosis in its many guises. As a result, a pleurodesis or abrasion for pneumothorax is performed,7 or diaphragm suturing or mesh insertion, making future access to residual or recurrent endometriosis lesions very difficult. Consequently, there is a need for skilled multidisciplinary teams to deal with this issue effectively at the first surgical opportunity. The aim of this paper was to describe our experience in trying to achieve this, especially as there are no significant robot-assisted series reported in the literature. 2 MATERIAL AND METHODS 2.1 Study setting A retrospective review of the first 50 consecutive procedures for DTE between July 2020 and March 2023 conducted at a private hospital catering for complex surgeries in London, United Kingdom. We established a multidisciplinary team consisting of three expert surgeons; one gynecological endometriosis specialist one hepato-pancreato-biliary surgeon, and one thoracic surgeon. Surgery was always performed with two or three of the surgeons present, always including the endometriosis specialist. Surgery for DTE, in the context of this study, was any surgery that involved either one or more of the following: mobilization of the liver +/− removal of endometriosis from the underside of the diaphragm through robot-assisted laparoscopy (RAL), and/or removal of endometriosis from the pleura, superior diaphragm, or wedge resection of lung through robot-assisted videoscopic assisted thoracic surgery (RAVATS). The surgeries were performed with the Da Vinci Xi Surgical System (Intuitive, California, USA). 2.2 Multidisciplinary decision making All patients underwent structured clinical assessment in a specialist endometriosis clinic, including a transvaginal ultrasound. In women with symptoms suggestive of diaphragmatic or thoracic involvement, targeted trans-abdominal ultrasound of the diaphragm was performed. Magnetic Resonance Imaging was not routinely used, as superficial diaphragmatic disease is frequently not detectable and definitive assessment relies on surgical inspection. Cases were discussed at a dedicated multidisciplinary team (MDT) meeting involving gynecological, hepato-pancreato-biliary, and thoracic surgeons. Operative planning was based on symptom profile, prior surgical history, and available imaging. In some cases, external imaging suggested possible deep diaphragmatic disease; however, the extent of invasion was always confirmed intra-operatively. The decision to include RAVATS was made pre-operatively at MDT and incorporated into patient counseling. Thoracic exploration was offered selectively, taking into account symptom burden, anticipated pelvic complexity, operative duration, and patient preference. In complex pelvic cases, thoracic exploration was staged to reduce operative burden. The operative strategy was to excise all macroscopically visible diaphragmatic disease where safely feasible. The patient assessment and decision-making is further presented in Figure 1. This study was conducted according to STROBE guidelines.8 2.3 Surgical technique 2.3.1 RAL A 5-port technique was used, with 4 × 8 mm robotic ports and 1 × 8 mm assistant port. Our strategy was to excise all lesions found where possible in accordance with our pelvic practice and based upon our view that excision is likely more effective than ablation.9, 10 On the underside of the diaphragm, it was possible to remove all areas of suspected superficial endometriosis by wide local peritoneal excision without perforation into the thoracic cavity, apart from over the bare area where ablation can be used or full-thickness diaphragm excision employed for deeper lesions. With deeper nodules, excision was performed to the depth that resulted in full disease clearance, sometimes resulting in partial diaphragm muscle excision, and sometimes necessitating full depth diaphragmatic excision into the thoracic cavity. 2.3.2 RAVATS About 4 × 8 mm robotic ports were used. Areas over sensitive structures like IVC and aorta, or the bare area of the diaphragm were generally ablated; however, deeper lesions on the bare area were managed by excision of that area of the diaphragm with primary closure using V-Loc and sometimes Ethibond. Chest drains were used only in women who had undergone wedge resection of the lung parenchyma, or if there was thought to be a higher risk of post-operative intra-cavity bleeding. 2.3.3 Combined RAL and VATS In general, we have tried to perform the RAL ‘under-diaphragm’ abdominal surgery first before continuing onto the ‘over-diaphragm’ RAVATS procedure. This was due to pelvic surgery had been carried out initially or in cases where a portion of the diaphragm was removed; it has proved easier to perform suturing from the superior side. 2.3.4 Post-operative care Patients returned to the main surgical ward following the procedure. A routine chest X-ray was performed the morning after surgery to check for residual pneumo/hemothorax in all women undergoing RAVATS, whether they had a chest drain or not. Chest drains were removed after a normal chest X-ray finding on D1 post op. We aimed for discharge once the women were able to mobilize independently and pain was controlled with oral analgesics only (1–2 nights in hospital). 2.4 Histological assessment Histopathological specimens were assessed by specialist gynecological pathologists using standard hematoxylin and eosin staining. In cases where endometriosis was suspected but not confirmed on routine staining, CD10 immunohistochemistry was performed. A diagnosis of endometriosis required identification of endometrial glands and stroma on hematoxylin and eosin and/or supportive immunohistochemical findings. Specimens classified as ‘chronic inflammation without typical features diagnostic of endometriosis’ lacked definitive endometrial glands and stroma on both routine and immunohistochemical assessment. 2.5 Outcomes Baseline characteristics including patient age and presenting symptoms were collected during initial assessment. Operative details, including total operative time, robotic console time, conversion to laparotomy/thoracotomy, estimated blood loss, and blood transfusion were retrieved from the operation notes and discharge summaries. Histopathological details, including location of endometriosis, were retrieved from histopathology reports. Histopathology results were stratified into three categories: ‘confirmed endometriosis’, ‘no endometriosis’, and ‘chronic inflammation with no typical features diagnostic of endometriosis’. Data from the post-operative period included: total length of hospital stay, length of stay in high-dependency unit (HDU), 90-day morbidity (classified according to the Clavien-Dindo scale11), readmissions, re-operations, and 90-day mortality were collected. Where available, 3-month follow-up data on symptom improvement (pain), based on clinical assessment using Visual Analogue Scale (compared with baseline). 2.6 Statistical analysis Continuous data were summarized as median (interquartile range (IQR)), with non-parametric data identified following visual evaluation of the distribution and Shapiro–Wilk assessment for normality. Following initial data analysis, subgroups based on histological subtype and operation type were established and subsequently were compared using the Mann–Whitney U test and X2 test, as appropriate. A pre-specified exploratory subgroup analysis of stage vs. single-sitting procedure was also conducted in a similar fashion. The threshold of statistical significance was set at p < 0.05. Data were analyzed and presented using R 4.0.3 (R Foundation for Statistical Computing, Vienna, Austria). 3 RESULTS 3.1 Baseline characteristics Fifty robot-assisted DE/TE procedures on 46 individual women were undertaken. The median patient age was 32 years (IQR 26–36). Four women (8.0%) had more than one operation. Of the 50 cases, 29 (58.0%) included pelvic endometriosis surgery with exploration of abdominal underside of the diaphragm (RAL), 13 (26.0%) pelvic endometriosis surgery with exploration of both abdominal underside of the diaphragm and the thoracic side of the diaphragm (RAL + RAVATS), 5 (10.0%) RAVATS only procedures, 2 (4.0%) both abdominal underside of the diaphragm and the thoracic side of the diaphragm without pelvic endometriosis surgery, and a single case only the exploration of the abdominal underside of the diaphragm. Of the 20 cases that had a RAVATS procedure either standalone or in conjunction with another cavity exploration, 16 (80.0%) were right-sided, 3 (15.0%) left-sided, and 1 bilateral VATS (5.0%). 3.2 Presentation Presenting symptoms varied amongst women. The majority of women had a combination of 1 or more pelvic symptoms in addition to their thoracic symptoms (n = 41/46, 89.1%). All but one woman (n = 45/46, 97.8%) reported pelvic pain. All women had at least one symptom of DTE and the majority reported more than one symptom suggestive of diaphragmatic or thoracic involvement (n = 41/46, 89.1%). These are detailed in Table 1. | Symptom | Frequency, n (%) | |---|---| | Shoulder tip pain | 29 (63.0) | | Chest wall/Costal margin pain | 18 (39.1) | | Scapula pain | 11 (23.9) | | Shortness of breath | 10 (21.7) | | Right upper quadrant abdominal pain | 3 (6.5) | | Hiccups | 3 (6.5) | | Jaw pain | 2 (4.3) | | Ear pain | 2 (4.2) | | Neck pain | 2 (4.3) | | Hemoptysis | 1 (2.2) | | Dysphagia | 1 (2.2) | 3.3 Operative characteristics No operations required conversion to open. For RAVATS only cases (n = 5), median operative time (from entry into anesthetic room to leaving the operating theater) was 128 min (IQR 122–239 min), with a median console time of 70 min (IQR 64.5–117.5 min). For cases involving two compartments (n = 41) (either abdominal + thoracic or pelvic + abdominal), the median operative time was 164 min (IQR 134.75–193.5 min) with median robotic console time of 118 min (IQR 83.5–137 min). For cases involving all 3 compartments (n = 13), median operative time was 236 min (IQR 213.75–254.25 min) with median robotic console time of 171 min (IQR 155–191.5 min). Median estimated blood loss was 10 mL (IQR 6.25–27.5 mL). No women required peri-operative transfusion related to intra-operative blood loss, and none had a clinically significant estimated blood loss >500 mL. 3.4 Location of endometriosis and histology 3.4.1 Abdominal under-diaphragm Out of 45 women who had an abdominal diaphragm exploration, 29 (64.4%) returned as positive for endometriosis, with an additional 6 cases returning as ‘chronic inflammation with no typical features diagnostic of endometriosis (13.3%). In 10 women (22.2%) there was no endometriosis found macroscopically on diaphragm exploration, and no samples were sent for histology. The histogram of abdominal endometriosis locations can be found in Figure 2. The majority of the histologically positive (either positive or chronic inflammation) with no typical features diagnostic of endometriosis were found on the right hemidiaphragm (n = 28, 80.0%). 3.4.2 Thoracic Out of the 20 cases operated on the thoracic diaphragm, 9 (45.0%) were reported as positive, 8 were reported as chronic inflammation with no typical features diagnostic of endometriosis (40.0%), while 3 (15.0%) were reported as negative for endometriosis (despite macroscopic changes visualized, excised and sent for histopathological analysis). One of the negative cases had findings consistent with asbestosis which was expected from pre-surgery scans. Four cases had lung parenchymal wedge resections, 2 of which were reported as positive, 1 as chronic inflammation with no typical features diagnostic of endometriosis and 1 as negative. The histogram of thoracic endometriosis locations can be found in Figure 3. The majority of histologically confirmed positive cases were on the right side of the right thoracic cavity; either on the pleura (n = 14, 82.4%), or on the overside of the diaphragm (n = 6, 35.3%). 3.5 Coexistence of abdominal under-diaphragm endometriosis and thoracic endometriosis The macroscopic coexistence of abdominal and thoracic endometriosis is shown in Table 2. Half of the positive under diaphragm cases had an RAVATS also at which all were macroscopically positive. Out of the 10 cases that did not yield macroscopically visible endometriosis on the underside of the diaphragm, 3 also had a concurrent RAVATS exploration; one was reported as positive for endometriosis and 2 as inflammation consistent with endometriosis. | Locations and macroscopic appearance | Frequency, n (%) | |---|---| | Macroscopically +ve underside diaphragm | 35 (77.8) | | With macroscopically +ve thorax | 17 (48.6) | | With macroscopically −ve thorax | 0 (0.0) | | With thorax surgically unexplored | 18 (51.4) | | Macroscopically −ve underside diaphragm | 10 (22.2) | | With macroscopically +ve thorax | 3 (30.0) | | With macroscopically −ve thorax | 0 (0.0) | | With thorax surgically unexplored | 7 (70.0) | 3.6 Combined three-compartment surgery vs. multi-stage approach About 13 women underwent combined pelvic, abdominal and thoracic endometriosis removal, while there were 4 women that had a pelvic and abdominal procedure first (RAL), followed by an isolated RAVATS on the later date. Three out of four of the completion RAVATS procedures returned positive histological confirmation of thoracic endometriosis. The pelvic/abdominal RAL and RAVATS procedures were separately combined in terms of operative time, estimated blood loss, and length of stay, and compared with a ‘one-off’ triple compartment approach. The median operative time was significantly longer (350.5 min vs. 236 min, p = 0.01) in the two-operation subgroup. This was also true for median estimated blood loss (30.0 mL vs. 5.0 mL, p = 0.043). There was no significant difference in total length of post-operative stay (2 days vs. 2.5 days, p = 0.055). 3.7 Post-operative data Two post-operative complications were reported, both Clavien-Dindo I. These included post-operative pain (managed with PO analgesia) and a small pneumothorax, managed conservatively. There were no Clavien-Dindo scale type II or III complications. The median length of stay was 1 day (IQR 1–2), with the longest post-operative stay being 6 days (due to increased pain management requirements). Ten patients were sent to High Dependency Unit for the first night for precautionary reasons only at the beginning of the program, which was later abolished. No unexpected High Dependency/Intensive Therapy Unit transfers were required. Among the 19 patients who had a chest drain, the median duration was for 1 day (IQR 1–1). 3.8 Association of pelvic disease type with diaphragmatic and thoracic disease Of the 42 cases that had a pelvic component to their surgery, 17 had endometriosis classified as deep (40.4%). Out of those, 3 cases had macroscopically negative diaphragmatic underside (17.6%), while remaining 14 patients had either chronic inflammation with no typical features diagnostic of endometriosis (n = 3, 17.6%) or positive histology for endometriosis (n = 11, 64.7%). Five of those cases had thoracic exploration with all having either chronic inflammation with no typical features diagnostic of endometriosis (n = 2, 40.0%) or positive histology for endometriosis (n = 3, 60.0%). Of the 25 cases classified as superficial endometriosis, 9 cases (36.0%) had a macroscopically negative underside of diaphragmatic, with 1 additional case that was microscopically negative (4.0%). The remaining 15 patients had either chronic inflammation with no typical features diagnostic of endometriosis (n = 3, 12%) or positive histology for endometriosis (n = 12, 48%). Eight of those cases had concomitant thoracic exploration with 1 having negative histology (12.5%), 2 having positive histology (25.0%), and remaining 5 cases (62.5%) having chronic inflammation with no typical features diagnostic of endometriosis. When compared, there was a trend toward a higher likelihood of positive diaphragmatic histology in subgroup with deep pelvis endometriosis (OR 3.11, 95% CI 0.70–13.69, p = 0.133), although this was non-significant. 3.9 Diaphragmatic disease classification Of the 14 patients who had deep pelvic endometriosis and had concurrent diaphragmatic disease excision, 5 had deep diaphragmatic disease (35.7%) and 9 had superficial diaphragmatic disease (64.3%). Of the 15 patients who had superficial pelvic endometriosis and had concurrent diaphragmatic disease excision, 1 had deep diaphragmatic disease and 14 had superficial diaphragmatic disease (92.3%). When compared, there was a trend towards a higher likelihood of deep diaphragmatic disease in the subgroup with deep pelvic endometriosis (OR 7.78, 95% CI 0.78–77.93, p = 0.08), although this was not statistically significant. All superficial diaphragmatic endometriosis (n = 23) were resected or ablated, without any breach of the diaphragm. For deep diaphragmatic endometriosis (n = 6), 4 cases required full-thickness resection, which was primarily closed with prolene sutures/V-lock/Ethibond, while two cases required non-full thickness resections. All diaphragmatic defects resulting from abdominal resection were closed from the abdominal approach. None of the cases required mesh for diaphragmatic repair. For thoracic endometriosis, 10/20 (50.0%) thoracic cases involved diaphragmatic resection, with the remaining thoracic endometriosis foci being non-diaphragmatic (e.g., pleura, lung, pericardium). Six of the cases required full-thickness resection and V-loc closure, while 4 cases required only superficial ablation of lesions. 3.10 Long-term follow-up Quantitative long-term data were available for 11 patients. The median pain score was 8.5 before surgery (IQR 7.25–9.0), and 5 (IQR 4.0–7.75) after surgery, at follow-up visit. 4 DISCUSSION This manuscript reports the outcomes of the largest series of robot-assisted endometriosis surgery involving three compartments and specialities in a private self-referral center for DTE. A combination of pelvic and under-diaphragm surgery, especially as it is by the same incisions robotically, was deemed acceptable in our practice if deep disease exists in the pelvis. We found a multiple-surgeon combined approach to be safe and had no significant major complications. An experienced endometriosis specialist is essential in disease recognition in extra-pelvic areas and that surgeries performed without this are more likely to result in missed disease.12 These teams should be regular teams that work together frequently rather than ad hoc solutions.13 We believe women with DTE should be managed in specialist centers where these teams exist. Extending from this conclusion, robot surgery has many advantages. Robotic platforms offer enhanced visualization, instrument articulation, and ergonomic advantages that may facilitate access to the subdiaphragmatic space and thoracic cavity. These features may be particularly relevant when multi-compartment surgery is undertaken in a single sitting.14-17 Robotic surgery has long been demonstrated to give advantages in vision (3D/HD), dexterity, and surgeon ergonomics.18 For simultaneous pelvic and under-diaphragm surgery, the incisions are the same and only the robot needs to be rotated 180 degrees to switch compartment. Mobilization of the liver robotically is much easier ergonomically and results, in our opinion, in a much better view of the posterior diaphragm and much greater control when excising lesions. Additional incisions are then required for the thorax, but the switching from the abdomen is not complicated, and the benefits in the thorax are analogous to those in the other compartments. All women had symptoms of possible DTE prior to surgery, and surgery was aimed directly at investigating and treating these symptoms. In other studies, VATS cases were often derived from urgent investigations for catamenial pneumothorax.5, 6 At this point, the surgery is likely to be inadequate, firstly as there is unlikely to be an endometriosis specialist present to comment upon possible lesions in real-time, and secondly because the thoracotomy, pleural abrasion or pleurodesis procedure used to treat the pneumothorax may well have left missed, untreated disease behind that and is no longer easily accessible. A proportion of women underwent liver mobilization without identification of macroscopically visible diaphragmatic disease. This reflects the inherent diagnostic challenge in DTE, where symptom burden does not reliably correlate with visible lesions and where current imaging modalities have limited sensitivity, particularly for superficial disease. In our practice, macroscopically normal-appearing diaphragm was not routinely biopsied, as random sampling carries a theoretical risk of unnecessary diaphragmatic weakening or perforation without clear clinical benefit. The observed rate of histological ‘chronic inflammation without typical features diagnostic of endometriosis’ further highlights the biological and pathological complexity of extra-pelvic disease. It is possible that some inflammatory lesions represent secondary or partially resolved endometriotic foci; however, definitive conclusions cannot be drawn. These findings underscore the limitations of pre-operative selection tools and the need for improved diagnostic strategies to better stratify patients and minimize negative explorations. As in most reports, we also found a significant bias to right-sided symptoms (80%) and subsequently a positive finding of right-sided disease (80%). On the underside of the diaphragm, the most common area was the posterior diaphragm under the liver and also on Gerota's fascia overlying the right kidney. Therefore, thorough inspection including liver mobilization may be required before excluding diaphragmatic involvement. Also, in the right thoracic cavity, we noted that the pleura over right infero-lateral cavity, along with the right superior diaphragmatic surface, were the most common sites of disease and anatomically directly opposite the under-diaphragm disease. The right-sided bias in the upper abdomen has been used as evidence in favor of support for Sampson's theory of retrograde menstruation5, 6 whereby flow of peritoneal fluid is thought to be first to the upper right quadrant and then obstructed from moving left by the falciform ligament. It has then been postulated that diaphragmatic damage, caused by the inflammatory and fibrotic nature of endometriosis, creates fenestrations in the diaphragm allowing direct spread into the thoracic cavity, mainly on the right. Based on the fact that women in our series were all operated on for suspected DTE based upon their symptoms, we found under diaphragm disease in 4 out of 5 cases (77.8%) and in more than 4 out of 5 cases (85%) where RAVATS was performed. Where we suspected macroscopic parenchymal lung involvement, we appeared correct in 3 out of 4 wedge resections. Where the underside was positive and RAVATS had been performed (half of all positive underside cases), all RAVATS appeared positive. Even when the underside was negative, 1 in 3 RAVATS was still positive. These findings suggest that thoracic involvement may occasionally be present despite absent macroscopic under-diaphragmatic disease. In our study, we had 3 cases that were negative for endometriosis or fenestrations on full liver mobilization under the diaphragm, but simultaneously were positive in the right thorax (2 cases of chronic inflammation with no typical features diagnostic of endometriosis and 1 case definitely positive for endometriosis). This suggests that these cases are not always, or even ever, caused by trans-peritoneal spread and that lymphatic or blood borne dissemination, or metaplasia may be involved. A further intriguing finding came from our series, where we have noted, on several occasions, the presence of Walthard Nests on the underside of the diaphragm. We published this data separately in conjunction with another center in the USA who noted similar findings.19 Walthard Nests are incidental findings of urothelial-lined cells in the female reproductive tract, particularly on fallopian tubes. The exact origin is debated but theories include that these are Mullerian remnants. The presence of Walthard Nests on the diaphragm may suggest etiologies other than retrograde menstruation for the unexpected appearance of endometriosis in this area of the body. Moreover, patients with cyclical symptoms suggestive of DTE have had macroscopic tissue suggestive of possible endometriosis removed, but not proven histologically to be endometriosis either on hematoxylin and eosin stain or CD10 immunohistochemistry. The presence of Walthard nests and cases of chronic inflammation without classic histological features highlight ongoing uncertainty regarding the pathogenesis of extra-pelvic endometriosis and warrant further pathological study. In the largest existing series, nodular (deep) lesions were present in 27% of cases, with full-thickness diaphragmatic resection required in 3.1% and partial muscular nodulectomy in 10.2%.4 VATS was performed in approximately 12% of patients, and mesh reinforcement was used in 46% of thoracic procedures. In contrast, although the proportion of deep lesions in our cohort was comparable, full-thickness resections were infrequent and all defects were closed primarily without mesh, with no diaphragmatic hernias observed during available follow-up. These differences likely reflect variations in case selection, surgical philosophy, and the integration of a planned multidisciplinary robotic approach with pre-operative thoracic involvement rather than a primarily laparoscopic strategy. Ceccaroni et al.,6 also reported deep disease as being a significant, and almost exclusive, risk factor for DTE (90% of patients being rASRM Stage 4 in the pelvis). However, while there was a non-significant trend toward an association in our series (p = 0.133), 60% of pelvic superficial endometriosis cases who had the underside of the diaphragm explored, had histology either positive (n = 12, 48.0%) or suspected for endometriosis (n = 3, 12.0%). What is more, 82.5% of pelvic superficial endometriosis who had thoracic exploration had either positive histology (n = 2, 25.0%) or suspected histology outcome (n = 5, 62.5%). This study has several limitations. First, it is a retrospective analysis from a single specialist private center and therefore subject to selection bias towards a symptom-selected and potentially more complex patient population. The findings cannot be used to infer population prevalence or generalizability to non-specialist settings. Second, there was no comparator group undergoing conventional laparoscopy or thoracoscopic surgery; as such, the study describes feasibility and safety within a robotic program but does not establish comparative effectiveness. Third, the comparison between simultaneous and staged procedures was exploratory, and is likely influenced by differences in disease extent, operative complexity, patient preference, and program evolution over time. These results should therefore be interpreted cautiously and as hypothesis-generating only. Fourth, systematic long-term follow-up and validated patient-reported outcome measures were not available, limiting assessment of symptom resolution, recurrence, and durability of repair. Finally, histopathological confirmation rates reflect the recognized diagnostic challenges of extra-pelvic endometriosis, and the absence of routine biopsy of macroscopically normal diaphragm may have influenced reported confirmation rates. Longer follow-up and prospective multicenter studies incorporating standardized outcome measures are warranted. 5 CONCLUSION DTE is possibly associated with deep pelvic disease, though it is not uncommon with superficial pelvic disease either. DTE can occur in the thoracic cavity without being evident on the underside of the diaphragm. Appropriate questioning of the presence of cyclical symptoms in both thoracic and gynecological assessments should raise the possibility of DTE. Treatment of DTE using robotic platform is safe and provides an opportunity for comprehensive assessment of all anatomical regions. Yet, questions about diagnostics and symptomatology remain, thus specialists should avoid operating in isolation to reduce the risk of missing disease and potentially reducing the efficacy of further specialized treatment as a result. AUTHOR CONTRIBUTIONS Peter Barton-Smith: conceptualization; methodology; investigation; data curation; supervision; project administration; writing—original draft. Michal Kawka: formal analysis; validation; data curation; visualization; writing—original draft; writing—review and editing. Tom Routledge: investigation; methodology; resources; writing—review and editing. Rupali Arora: investigation; data curation; writing—review and editing. Long R. Jiao: supervision; methodology; resources; writing—review and editing. FUNDING INFORMATION The authors have nothing to report. CONFLICT OF INTEREST STATEMENT Tom Routledge is a proctor and medical officer for Intuitive Surgical and has received direct payments. No other authors declare conflicts of interest. ETHICS STATEMENT This retrospective review did not require formal ethical approval because it did not meet the definition of research requiring NHS Research Ethics Committee review, as confirmed using the Health Research Authority decision tool. No patient-identifiable data were used, and all data were processed in accordance with UK GDPR and the Data Protection Act 2018. DATA AVAILABILITY STATEMENT The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.