{"paper_id":"47d65324-df6f-47a3-bb45-bc6cef735038","body_text":"Proposed simplified protocol for initial assessment of endometriosis with transvaginal ultrasound\nAbstract\nBackground\nTransvaginal ultrasound (TVUS) is a valuable tool for the diagnosis and staging of endometriosis. The 2022 update of the European Society of Human Reproduction and Embryology (ESHRE) guidelines for diagnosis and management of endometriosis1 suggests that laparoscopy should no longer be considered the gold standard in diagnosis, favoring imaging as the first-line investigation. The 2016 consensus from the International Deep Endometriosis Analysis (IDEA) group eloquently laid out a systematic approach to performing TVUS to assess for endometriosis (eTVUS)2. In comparison to the established standard-of-care routine TVUS (rTVUS), in which only the uterus and ovaries are assessed3-5, they proposed assessment of many additional anatomical structures2. Although studies have shown an achievable learning curve for this technique within specialist centers6, it is likely that the learning curve in general clinical settings, where disease prevalence is lower, would be longer and it would therefore be harder to achieve proficiency. Furthermore, performing a comprehensive eTVUS examination adds a significant amount of scan time7. Similarly, the interpretation and reporting of eTVUS is more onerous than that of rTVUS. These factors have likely contributed to the slow uptake of eTVUS amongst the general imaging community, which has made access to eTVUS outside of specialist centers limited.\nIt is well known that endometriosis carries a significant diagnostic delay, which can impact quality of life and may contribute to disease progression8. Furthermore, as endometriosis is common, affecting approximately 190 million people worldwide, if ultrasound is to be a practical diagnostic option, it needs to be accessible in community imaging facilities and not limited to specialist centers. Wider access to TVUS diagnosis could lead to a reduction in diagnostic delay and improve the quality of life for people with endometriosis by facilitating faster access to treatment. A consensus paper from the Society of Radiologists in Ultrasound (SRU) recently proposed a protocol to augment the rTVUS examination, in order to address this important issue9. The SRU consensus provides much-needed guidance on how to document eTVUS through cine clips for offline review, which will aid radiologists considerably in assessing for endometriosis. However, its inclusion of ‘indirect’ features and ‘optional’ maneuvers potentially reduces diagnostic accuracy and certainty for both clinicians and patients. Furthermore, it lacks the systematic approach to both scanning and describing disease with consistent nomenclature that is a cornerstone of eTVUS as proposed by the IDEA consensus2.\nIn our opinion, a simplified, yet still consistently systematic protocol of the IDEA consensus could be performed within community ultrasound settings, with the aim of diagnosing endometriosis in symptomatic patients, yielding a high-level of disease detection yet reducing complexity and scan times. By utilizing a systematic approach, such a protocol could help to expand access to eTVUS for the diagnosis of endometriosis beyond specialist centers, resulting in a reduction in diagnostic delay and thereby enabling earlier validation and treatment, whilst maintaining the strong diagnostic accuracy of eTVUS that has evolved to date10-15. Additionally, from a training perspective, the step-by-step nature of the IDEA method provides a clear learning pathway, whereby learners master in turn each anatomical structure to become proficient in performing a comprehensive eTVUS examination. Although the precise learning curve of this proposed simplified method is unknown, reducing the number of steps to be learnt will undoubtably create an easier educational pathway for generalists.\nProposed method\nOur proposed method, as it compares with both rTVUS and comprehensive eTVUS in line with the IDEA consensus method2, is outlined in Table 1 and can be viewed in Videoclip S1. We propose that, in addition to the structures assessed during rTVUS (uterus and ovaries), assessment of the uterosacral ligaments (USLs), pouch of Douglas (POD) (for endometriotic nodules and adhesions causing obliteration) and upper rectum be performed as a simplified eTVUS protocol. The remaining anatomical locations and techniques proposed by the IDEA consensus (vaginal wall, rectovaginal septum, rectosigmoid and sigmoid colon, bladder, ureters, vesicouterine space, parametrium16, ovarian mobility and anterior sliding sign) do not need to be assessed routinely. The steps required to perform our proposed simplified eTVUS method are shown in Figure 1.\n| Parameter | rTVUS | eTVUS (simplified) | eTVUS (comprehensive) |\n|---|---|---|---|\n| Anatomical structure | |||\n| Uterus (including cervix and endometrium) | ✓ | ✓ | ✓ |\n| Ovaries | ✓ | ✓ | ✓ |\n| Ovarian mobility | × | × | ✓ |\n| Uterine sliding sign with/without triangle sign (for POD obliteration) | × | ✓ | ✓ |\n| Uterosacral ligaments/torus uterinus | × | ✓ | ✓ |\n| Upper rectum | × | ✓ | ✓ |\n| Lower rectum/rectosigmoid/sigmoid colon | × | × | ✓ |\n| Vaginal wall/posterior vaginal fornix | × | × | ✓ |\n| Rectovaginal septum | × | × | ✓ |\n| Parametrium | × | × | ✓ |\n| Bladder | × | × | ✓ |\n| Ureters | × | × | ✓ |\n| Vesicouterine space | × | × | ✓ |\n| Anterior sliding sign | × | × | ✓ |\n| Pouch of Douglas | Assessed for fluid only | Assessed for fluid, obliteration and nodules of endometriosis | Assessed for fluid, obliteration and nodules of endometriosis |\n| Appointment time required (min)* | 20 | 30 | 45 |\n- *Estimated, including both scanning and reporting7.\n- ×, not assessed; ✓, assessed; POD, pouch of Douglas.\nJustification\nGiven that rTVUS is a commonly performed ultrasound study, we feel that most sonographers, sonologists and radiologists have a satisfactory level of confidence in performing this examination and detecting endometriomas within the ovaries. We do not advocate the assessment of ovarian mobility for the purpose of diagnosing endometriosis, due to the low diagnostic accuracy of this technique, risking false-positive results due to technical factors or poor technique17.\nThe USLs have been shown repeatedly to be the location most frequently affected by deep endometriosis (DE)10-12, 18, 19, with USL nodules occurring in up to 66% of cases of DE, of which 83% do not have disease present at other sites18. The USLs are, therefore, irrefutably the most important structure to assess if the intention is to provide an initial sonographic diagnosis of endometriosis. Furthermore, recent studies have shown that with a methodical scanning technique via the posterior vaginal fornix20, high levels of diagnostic accuracy can be achieved for detecting DE within the USLs14, 21. As such, including an assessment of the USLs, performed systematically via the posterior vaginal fornix during TVUS, will lead to a diagnosis of endometriosis in many cases without the need for further extended scanning. Despite the high prevalence of endometriosis of the USLs, these are challenging structures to assess sonographically. However, as this simplified eTVUS examination requires assessment of fewer structures than does the comprehensive eTVUS scan, the overall learning curve is reduced, making the detection of endometriosis on ultrasound more attainable and less daunting for sonographers. However, no formal learning-curve studies specific to USL assessment for endometriosis have been published, and such an investigation would be valuable to broaden our understanding of the pathway to proficiency.\nThe uterine sliding sign2, considered to be a simple maneuver to perform, has high diagnostic value for the detection of POD obliteration, with an achievable learning curve, even amongst non-experts15. Furthermore, given the need to triage cases of endometriosis with POD obliteration to an advanced laparoscopic surgeon, we feel that inclusion of the sliding sign offers the potential to alter greatly the patient journey, with minimal effort. In a recent systematic review and meta-analysis, Alcázar et al.13 revealed that not only does a negative sliding sign have a strong positive predictive value for detecting POD obliteration, but it also acts as an indirect diagnosis of bowel DE, with a sensitivity of 81% and specificity of 95%. The value of the sliding sign as an indirect marker for DE was also highlighted by Young et al.22, who reported a 3-fold increase in the detection of DE when the sliding sign was added to rTVUS examinations. As such, it could act as a suitable surrogate marker for overall bowel DE on a simplified eTVUS, without the additional learning curve or time required to assess the entire bowel satisfactorily.\nThe uterine sliding sign is limited in those with a retroverted uterus, in whom it is not possible to assess the status of the POD between the posterior uterus and posterior vaginal fornix. It is usually in this region that USL- or POD-DE is present, leading to POD obliteration. The triangle sign is a technique that has been proposed to overcome this limitation; however, the diagnostic accuracy of this is, as yet, unknown23. Similar to the traditional sliding sign, the technique for its performance is thought to be straightforward and easy to learn and implement.\nDespite considerable understanding of POD obliteration within the eTVUS literature, the POD peritoneum as an anatomical site for DE nodules has been described poorly. However, nodules within the POD have been shown to be evident in approximately 10% of people with endometriosis visible sonographically19. By definition, the POD is the potential space, lined by peritoneum, between the posterior uterus, cervix and posterior vaginal fornix anteriorly, and the rectum posteriorly. It is bounded laterally by the USLs24. As such, when sweeping between the left and right USLs, the POD is scanned incidentally, so assessing for nodules in this location takes little extra effort. Furthermore, as literature describing the sonographic detection of superficial endometriosis within the POD is emerging25-27, sonographers becoming familiar with the anatomy of the POD will likely increase early detection of endometriosis (of all subtypes).\nDespite its significance surgically, as the occurrence of bowel DE without DE in other anatomical areas is uncommon28, we do not advocate assessment of the entirety of the bowel in this simplified protocol, as omitting this will almost certainly reduce both scan time and the learning curve. However, the upper rectum will often be seen when sweeping through the POD in the sagittal plane and is the most common location of bowel DE (mean ± SD lesion-to-anal-verge distance = 114.8 ± 36.5 mm)29. Furthermore, many rectal DE nodules are contiguous with USL- or POD-DE nodules, and/or adherent to the uterus, causing a negative sliding sign, as noted above. Therefore, we feel that DE within the upper rectum will likely be detected with this simplified eTVUS protocol, without the need to perform a full assessment of the bowel.\nThe anterior compartment, vagina, rectovaginal septum and parametrium are shown consistently within the literature to be less commonly affected locations of DE11, 19. Moreover, these locations are almost never affected in isolation, without the more commonly involved locations such as the USLs or endometriomas. Therefore, we feel that excluding these from this simplified protocol will result in a more achievable examination for the generalist practitioner, with minimal impact on the diagnosis of endometriosis.\nTo ensure diagnostic clarity for patients, it is vital that examination reports utilizing this simplified protocol clearly document exactly which anatomical structures have been assessed, whilst ensuring patients understand that a negative ultrasound examination does not exclude endometriosis.\nWhen to consider simplified eTVUS vs comprehensive eTVUS\n- surgery for endometriosis is intended; we strongly support routine preoperative mapping (with comprehensive eTVUS) to provide important information regarding possible involvement of bowel, bladder, parametrium or ureters, given the surgical complexity of these anatomical sites;\n- ultrasound is required to evaluate ‘severe’ endometriosis noted on another diagnostic test or diagnostic laparoscopy;\n- the clinical scenario warrants comprehensive eTVUS, such as when there are concerns regarding bowel endometriosis (e.g. obstruction symptoms), ureteral endometriosis (e.g. flank pain) or visible/palpable vaginal endometriosis on clinical examination;\n- an individual patient desires full mapping of endometriosis to feel more equipped in their treatment consent process.\nThe future of rTVUS and eTVUS\nThe rise of imaging diagnosis of endometriosis has been invaluable to our understanding of the prevalence, presentation and disease behavior of endometriosis. A recent study by Chaggar et al.30 found that 19% of people presenting to a gynecology clinic for any clinical reason had endometriosis visible on ultrasound. Similarly, Alson et al.31 found that 22% of people presenting with subfertility had endometriosis detected at baseline fertility TVUS. Furthermore, in those with endometriosis and subfertility, obtaining the endometriosis diagnosis prior to initiation of assisted reproduction has been shown to be associated with higher live-birth rates32. As such, we believe that the evidence currently presents a compelling argument for the standard-of-care gynecological ultrasound examination, rTVUS, to include an assessment for endometriosis. Our simplified eTVUS protocol, potentially with the addition of an antral follicle count for fertility planning, would be well placed to achieve this. Whilst additional training beyond that currently required for the performance of rTVUS would be needed, and modestly extended scanning times required, the benefit to patients in terms of improved quality of life and fertility planning would far outweigh these impacts.\nConclusions\nBased on the current literature and our extensive clinical experience, our proposed simplified eTVUS protocol for the diagnosis of endometriosis could be utilized in the community imaging setting to diagnose many cases of DE. We believe that, by following a systematic approach to scanning, as advocated by the IDEA consensus2, diagnostic accuracy, and clarity with respect to precisely which structures have been assessed, can be maintained. However, by excluding anatomical locations less likely to be affected by DE, whilst focusing on those with a higher prevalence of disease, we believe that uptake of TVUS to detect endometriosis could be expanded into more general clinical settings, helping to reduce the diagnostic delay associated with endometriosis. This would undoubtedly have a positive impact on wait times for comprehensive eTVUS examinations, which are currently a precious resource, being performed in only very few locations. When DE is detected with this method in a general imaging setting, the patient could then be referred on to a specialist center for comprehensive preoperative mapping of the disease, in line with the IDEA consensus approach2, if necessary. Furthermore, this protocol has potential value as a triage tool: (1) those with identified endometriosis can receive a diagnosis and be triaged to an appropriate managing clinician; and (2) those with POD obliteration can be referred directly to an advanced laparoscopic surgeon. It is critical to emphasize, however, that a negative scan result (i.e. no endometriosis is detected), for either this simplified eTVUS protocol or a comprehensive eTVUS examination, does not rule out the possibility of endometriosis, and further diagnostic investigations should be considered if symptomatology warrants this. As our understanding of endometriosis prevalence and impact continues to evolve, we believe there is a compelling argument for the standard-of-care gynecological ultrasound examination, rTVUS, to include an assessment for endometriosis. Accordingly, we propose that our simplified eTVUS protocol could replace rTVUS, with optimized impact on patient care and minimal impact on the gynecological scan time.\nDisclosures\nA.D. reports grants from The Robinson Research Institute and the Wilson foundation, speaker fees from Radiopaedia, Philips Healthcare and GE Healthcare, and honoraria from Vagenius Training, outside the submitted work. M.L. reports grants from Australian MRFF, AbbVie, CanSAGE, CIHR, Hamilton Health Sciences, Hyivy and Pfizer, honoraria for lectures/writing from AIUM, GE Healthcare, Bayer and AbbVie, consultancy work with Abbvie, Hologic, Chugai, Gesynta, Roche Diagnostics, Afynia and Pfizer, and affiliations with Imagendo and SUGO (Specialized Ultrasound in Gynecology & Obstetrics), outside the submitted work.\nDATA AVAILABILITY STATEMENT\nData sharing not applicable to this article as no datasets were generated or analysed during the current study.","source_license":"CC0","license_restricted":false}