This series conducted at a Canadian tertiary care center for endometriosis confirmed that an advanced TVUS performed by an expert clinician trained in focused endometriosis scanning is likely to be more advantageous (sensitivity 77.5 %) in the diagnosis of endometriosis, especially for sites of involvement other than ovary, when compared to routine pelvic ultrasound performed by a sonographer (sensitivity 25 %).
Similar low sensitivity in detecting endometriosis by routine pelvic ultrasound has been shown by earlier studies. Friedman et al. in their review of 37 women assessed for infertility depicted successful detection of endometriosis in only 10.8 % (4/37) with routine ultrasound, concluding that it is neither sensitive nor specific in diagnosing endometriosis . It is important to note that the focus with routine ultrasound is on the disease’s presentation as a discrete pelvic mass, mostly endometriomas [6–8], and not on other forms of endometriosis, a fact that was replicated in this retrospective review. The diagnosis of endometriosis in all the ten women was based exclusively on the presence or absence of endometrioma with features other than endometriomas identified in only two women. TVUS, otherwise, is considered a useful test both to make and to exclude the diagnosis of an ovarian endometrioma, as concluded by a systematic review assessing the use of gray-scale imaging in the diagnosis of ovarian endometriomas specifically rather than the full spectrum of endometriosis lesions . A definitive advantage of routine TVUS over ETVUS remains its easy availability, cost and time efficacy as well as no requirement for additional, specialized training.
A good quality TVUS when performed by an expert has shown a high degree of accuracy in the diagnosis of endometriosis [9–11]. The sensitivity of 77.5 % achieved in the current review is comparable to 78.5 % reported by Bazot et al. in their study assessing the diagnostic accuracy of TVUS in deep endometriosis . The scan when performed by an experienced radiologist is highly sensitive in detecting endometriosis of the pelvis not only involving the ovaries but also structures such as the vagina, retrocervical space, uterosacral ligaments, bladder, and rectal wall [9, 15–20]. Two essential components of ETVUS are a detailed compartmentalized assessment of the disease to identify the extent as well as severity of involvement, and detection of different types of lesions pertaining to the deep involvement of the disease such as nodules, dense adhesions, implants, and obliteration of the Pouch of Douglas. Both of these factors are critical for symptomatic evaluation and formulating a management plan. Figure 5 describes a case where although TVUS depicts typical features of endometriosis but are not reported due to lack of familiarity of the lesions. Expert TVUS images in the same case depict similar findings missed on TVUS.
Imaging of the urinary tract to identify endometriotic disease is important . Disease may be easily overlooked with the TVUS as urinary tract imaging does not form a part of the standard protocol used for pelvic pain. ETVUS, on the other hand with its detailed compartmentalized assessment, is more likely to pick up the disease. Three women in this series, missed by TVUS, were taken up for surgical removal of the disease. ETVUS appears a useful tool in experienced hands but its performance in the absence of clinical suspicion or a negative clinical exam, requirement of specific training to understand the radiological features, and limited availability are considerable limitations.
The major implications of missed diagnosis especially with a nonexpert scan are delay in diagnosis, rationalizing patient’s symptoms as normal, delay in referral to a specialized center, inappropriate management, and incomplete surgeries amounting to patient dissatisfaction. A total of 68 surgeries were performed for the management of endometriosis in this study group of forty, averaging 1.7 per patient. Twenty-seven women (27/40, 67.5 %) in this review attended a specialized center for further management after being treated at other non-specialized centers. All of these women had undergone prior surgery, with incomplete surgical management or continued signs and symptoms. Sixteen out of these 27 (16/27, 59.2 %) women underwent a repeat definitive surgery after being evaluated by an expert scan at the specialized center. The duration between primary management at other centers and attendance at the specialized center ranged from less than 1 year to as long as 13 years. These unplanned or incomplete surgeries as well as long diagnostic delays can add to economic burden on the healthcare system emphasizing the need for an accurate and early preoperative diagnosis of the disease.
The results show that there should be no argument against a high quality ultrasound; however, it is difficult to completely eliminate the inherent biases in the comparative study as this one. The significant ones being that the TVUS was not standardized for technique, time period, ultrasound machine, or the imager’s experience. It is extremely difficult to standardize these factors as the TVUS was performed at different locations; however, to minimize this bias, we ensured that satisfactory quality of imaging with a reasonable evaluation of uterus and bilateral ovaries was available. It is also difficult to estimate the extent of influence on ETVUS when performed after the prior information of TVUS results. Considering that endometriosis has a chronic progression over years, the median time difference of 294 days between TVUS and ETVUS seems reasonable, though, it is hard to be certain if the higher detection rate with ETVUS pertains to progression of disease or missed diagnosis with TVUS. In the authors' experience, it would be highly unlikely that 30 out of 40 women had a progressive endometriosis developing within a relatively short-time period. The reason it is so important to share this knowledge is the fact that 30 out of 40 scans in patients with surgically confirmed endometriosis-related pelvic pain were reported as “normal”. It is essential to share with our colleagues the fact that a “normal” routine ultrasound may not be enough in patients who are suffering with severe pain.
In light of the above-discussed facts, it would be reasonable to state that routine pelvic ultrasound is a valuable first line of investigation for patients with pelvic pain, but it can neither reliably diagnose nor convincingly rule out the presence of endometriosis, especially the nonovarian form. The future emphasis, thus, should be toward developing TVUS imaging predictors other than endometriomas that are more reliable, easily identifiable, reproducible, and require a short learning curve with minimal training. There is a need to develop and disseminate the concept of a standardized level I screening ultrasound for endometriosis (that includes predictors other than endometriomas) to help primary healthcare providers in making a definitive diagnosis without delay as well as to assess the need for referral to a specialized centre, and a level II expert ultrasound for endometriosis, intended towards knowing the extent and severity of disease to help in surgical planning.