Previous studies have revealed that, for patients with histologically confirmed DIE, an associated ovarian OEM is a marker for the severity of the disease . However, the related research on the distribution characteristics of the DIE lesions in patients with OEM is still limited. The present study investigated lesion distribution characteristics in women diagnosed with histologically confirmed OEM and DIE, as this combined diagnosis is often indicative of a more multifocal and severe disease. There was a greater frequency of anatomic DIE lesion sites in patients with unilateral OEM (unilateral OEM size ≤50 mm) than in patients with bilateral OEM (unilateral OEM size ≤50 mm), and DIE lesions were often associated with intestinal and vaginal infiltration.
These data suggested that DIE lesions were widely distributed. There were 30 anatomical sites observed in the present study, which were primarily located in the posterior pelvic cavity, but also in the uterosacral ligaments, intestines, ureter, vagina, posterior fornix and rectovaginal pouch. The distribution of DIE lesions was associated with the flow pattern of peritoneal fluid and the morphology of the pelvic cavity . When compared with the uterosacral ligaments, intestines and vagina, the reduced frequency of deep bladder endometriosis observed can be explained by anatomical location as the lower limit of the vesico-uterine pouch is located well above the lower limit of the Pouch of Douglas. The anatomical differences between the left- and right-hand sides of the hemipelvis are that the sigmoid colon is located on the left-hand side of the hemipelvis, and when combined with the left adnexa it forms a barrier to prevent menstrual blood reflux, resulting in an anatomical situation that could promote adhesions and the growth of refluxed endometrial cells on the left-hand side of the pelvic wall.
According to the literature, intestinal endometriosis and urinary tract endometriosis account for ~5-12% and 0.3-12%, respectively, of all women with endometriosis [16-18]. However, in the present study, intestinal endometriosis accounted for 30.9%, and urinary tract endometriosis accounted for 15.2% of cases, which are greater frequencies than those recorded by previous research. According to the theory of Kondo et al, women presenting with OEM had a stronger association with the presence of DIE lesions and intestinal DIE. There is lack of research on OEM associated with urinary tract endometriosis. The present results indicated that, for patients diagnosed with histologically verified OEM associated with DIE, there was an increased risk of lesions involving the intestine and ureter. Thus, when the clinical examination suggests OEM with DIE, the practitioner should search for severe lesions, especially intestinal and ureteral lesions. Surgery for intestinal endometriosis and urinary endometriosis is difficult, with a number of postoperative complications[20, 21]. The literature reports distinguish between major and minor complications. Major complications include anastomotic insufficiencies, intestinal perforation, retovaginal fistulas, severe infections, and bleeding requiring transfusion, which are reported in 7.4% to 25%. Minor complications include slight-to-moderate infections, peripheral sensory disturbances, bladder voiding dysfunction, and postoperative urinary obstruction, which are reported in 0.6% to 57%. Therefore, full clinical evaluations, as aforementioned, before operating may be very important for OEM patients with DIE.
As mentioned above, intestinal endometriosis refers to the rectum and sigmoid endometriosis. However, there are a special intestinal endometriosis, appendiceal endometriosis, which is considered as an uncommon finding, in the literature its prevalence varies widely . In our study, there are 4 patients with appendiceal endometriosis，are responsible for approximately 4% of all intestinal lesions. Diego Raimondo et al. pointed out that appendiceal endometriosis was associated with adenomyosis, large right endometrioma, deep posterior pelvic endometriosis, left deep lateral pelvic endometriosis, and ileocecal involvement. The 4 patients in our study who had appendiceal endometriosis were all combined with adenomyosis, and the diameter of OEM was greater than 5cm, which was further validated the conclusions of Diego Raimondo et al.
Patients with simple OEM often have no typical clinical manifestations, and some patients only find the presence of OEM in routine physical. Previous studies have conducted multivariate regression analysis on the relationship between dysmenorrhea and OEM. The results showed that the severity of dysmenorrhea had nothing to do with the existence of OEM. Intestinal endometriosis and deep pelvic invasive endometriosis were the main related factors of dysmenorrhea. Of the 304 patients in our study, 272 (89.5%) with symptoms of dysmenorrhea, 39 (12.8%) with dyspareunia, and 17 (5.6%) with chronic pelvic pain. Dai et al reported that 61.6% of patients with OEM but non-DIE had dysmenorrhea, and 10.2% had severe dysmenorrhea. It is suggested that when patients with OEM have obvious pain symptoms, it should be considered that they may be complicated with DIE.
During the early stages of ureteral endometriosis, the clinical symptoms are not typical, and are easily overlooked during surgery. Some researchers have suggested that urinary tract endometriosis may occur more frequently than it is currently thought to[30, 31]. Raimondo D et al. observed that ureteral involvement was always associated with endometriosis in other locations in the pelvis. And the most frequent endometriosis associations with ureteral involvement are ovarian lesions. In this case, when there is an associated OEM, ultrasonic examination of the urogenital system is necessary. Ureteral endometriosis should be considered when B mode ultrasound reveals ureter stenosis and hydronephrosis. Hydroureter and hydronephrosis are the severe forms of ureteral endometriosis. In case of hydronephrosis, renal scintigraphy to evaluate renal function should be needed.
Patients with associated unilateral OEM have an increased number of DIE lesions and have a greater risk of lesions involving the intestine and vagina when compared with patients with bilateral OEM. It was hypothesized that a clinical case of bilateral OEM may be more complex than unilateral OEM, and the distribution of the DIE lesions may be more extensive and deeper in patients with larger OEM. However, some researchers maintain that in patients with OEM ≥30 mm, OEM size was the most influential contributor to the total number of follicles and oocytes retrieved. OEM results in a reduced response to ovarian stimulation, when compared with the response of the contralateral normal ovary in the same individual . The oppression of the ovary was more obvious in bilateral OEM and the response to ovarian stimulation was markedly reduced. In the clinic, patients with unilateral OEM or an OEM with a diameter of ≤50 mm should not be taken lightly, and should be evaluated for the presence of DIE lesions, especially intestinal DIE.
The aim of the present study was to potentially provide the basis for avoiding underestimation of the extent of the DIE lesions. Misunderstanding the severity of DIE lesions is the main reason why it is difficult to completely remove lesions during surgical treatments. Incomplete exeresis of DIE lesions explains the high risk of recurrence; this recurrence is in fact the continued progression of the lesions left behind during previous operations[34, 35]. Repeated surgery is positively associated with increased health care costs and morbidity . Repeated surgery with damage to ovarian reserves is particularly frequent in ovarian endometriosis[37-39]. In addition, the risk of repeated operations is associated with uncertainty regarding surgical outcomes and pain. Therefore, completely excising the DIE lesions during the initial surgery is particularly important.
For the first time, our study analyses the characteristic of DIE lesion distribution in patient with ovarian endometrioma. However, this was a retrospective single center study, results were limited by a lack of random, patient selection, and incomplete data acquisition. The data on the history of previous endometriosis treatment were incomplete, which prevented us from evaluate the effect of previous treatment on the recurrence. A large multicenter prospective trial will be necessary to further assess this lesion distribution characteristic.