Clinical presentation of perineal endometriosis and prognostic nomogram after surgical resection BMC Women’s Health

PEM is a rare kind of endometriosis, and accounts for 0.29% (130/44828, Fig. 1) of all kinds of endometriosis according to our data. Many theories are advocated to explain the occurrence of ectopic endometrium in the perineal area. Iatrogenic implantation of endometrial tissue into the open wound at perineum has been proposed as the predominant theory in relation to the genesis of PEM [11]. A recent systemic review incorporating 90 studies found that 95.3% of 283 patients with vulvo-perineal endometriosis had gone through perineal trauma before the onset of symptoms [5]. The results of our survey further corroborate the hypothesis of iatrogenic seeding through wound contact with viable endometrial cells. Nevertheless, spontaneous PEM without perineal injury has been documented as well. They mostly occur in the bartholin gland and labia while rarely involve the perianal muscle [5]which might be explained by coelomic metaplasia, lymphatic and/or hematogenous dissemination, and growth differentiation of bone marrow-derived stem cells (BMDSCs) [12, 13]. BMDSCs may be a source of extra-pelvic endometriosis because of their ability to differentiate directly into endometriotic cells at ectopic sites after migration through peripheral circulation [12]. Also, growing evidence has found that abnormal epigenetic expression [12]microbiome and metabolic changes [14, 15]unbalanced immune microenvironment [16] may involve the development of endometriosis by modulating the proliferation, invasiveness and adhesion of ectopic endometrial cells. However, whether they influence the pathogenesis of PEM remains to be explored.

Despite early management is recommended to prevent the infiltrative growth into adjacent structures, accurate diagnosis of PEM at the initial stage is challenging in clinical practice. Zhu et al. suggested three typical characteristics for clinical diagnosis with a high predictive value, including anamnesis of perineal tears and/or episiotomy, tender nodules or masses at perineal scar, and cyclic nature of painful complaint associated with menses [6]. However, the highly variable presentations make PEM easily to be confused with anal abscess, bartholinitis, lipoma, lymphoma, haemangioma and vulvar carcinoma, particularly those women who were transferred to general surgeons at first [5, 17,18,19,20]. Hence, clinical practitioners should realize that any cyclical symptom reported by women of reproductive age can be suspected as an indicator of endometriotic lesion.

Consistent with previous findings, serum CA-125 does not assist in PEM diagnosis in our survey since the majority of participants showed normal range concentrations [21]. Ultrasonography has been proposed as a non-invasive, reproducible and cost-effective imaging modality for visualizing the lesion diameters, localization, proximity to neighboring structures and extent of muscular invasion if present [22]. The sonographic features are usually characterized by hypoechoic solid or cystic nodules containing hyperechoic spots or bright strands with ill-defined borders at the site of episiotomy scar [3]. Peripheral vascularization may be revealed through Doppler evaluation [3, 6]. Endoanal ultrasonography is a reliable technique to distinguish the perianal abnormalities and to delineate the structural integrity of the anal sphincter with high accuracy, through which sphincter or rectal involvement can be easily determined [23]. As a complementary method to endo-sonography, trans-perineal ultrasonography has been deemed as a more widely available option to measure the size of perianal lesion and to assess its anatomic relationship to vital structures [4]. Pelvic MRI has superb contrast resolution for soft tissues and is used to identify very small lesions or deep endometriotic tissue with extensive infiltration. PEM lesions characteristically appear as hyperintense heterogeneous spots on T1-weighted images and hypointense nodules on T2-weighted images due to the periodic hemorrhage inside the ectopic foci [3]. Infiltrative PEM lesions penetrating the mucous membrane of the anal or rectal wall could lead to severe intestinal symptoms including painful defecation, rectal bleeding, and chronic diarrhea. In that case, endoscopic examination such as colonoscopy or proctoscopy should be performed for further investigation.

Despite the very low incidence of extra-pelvic endometriosis, treatment for this ectopic pathology is one of the most challenging issues in clinical practice. Generally, the therapeutic strategy varies with the disease location and symptomatic severity. Medical management of hormone deprivation could greatly relieve the pain, whereas surgical removal of the endometrial implants has been viewed as a more radical approach to restore the anatomy and organ function with a pronounced long-term benefit. In regard to the age, infiltrating depth, tumor size and the adjacent structures, different surgical techniques sparing the critical nerve, vessels and important organs should be applied for a minimally invasive treatment with the best effect [24, 25].

As the primary therapeutic modality for symptomatic patients with PEM, surgical treatment is recommended. Local complete excision is associated with lower recurrence and can avoid the possibility of malignant degeneration even though it may partially compromise sphincter muscles. Hormone suppressive therapy with GnRH-a before the surgery is effective in reducing the size of lesions, particularly for lesions adherent to or extending into perianal muscle [26]. There were nearly one third of PEM patients affected with ASI in our series, which was similar to the findings of Liu et al. [21]. Rectal examination should be performed routinely, especially in the case of perianal involvement, to assess the extent of ASI. Accordingly, we detected a positive relationship between ASI and recurrence due to increased difficulty in the process of radical removal. Since the majority of cases with ASI experienced no recurrence, radical removal of involved sphincter muscle with sphincteroplasty is preferred as a safe and curative procedure minimizing the risk of recurrence [4, 21].

Conceivably, the recurrent endometriotic lesions may arise from minimal residual lesions (MRLs) or from dae novo lesions [27]. Nevertheless, mounting evidence suggests that the former is more likely. Complete excision of the lesions together with surrounding healthy tissue to make sure that no residual disease left behind was advocated by most practitioners [5]. In the present study, the high incidence of mPM and its potential effect of facilitating post-operative recurrence implied that, merely under visual inspection, MRLs could hardly be avoided in the case of narrow excision with 0.2–0.5 cm of surgical margins. In fact, some researchers have proposed that complete resection required at least 0.5–1.0 cm free edges from the PEM nodules, and even wide excision with 1.0–2.0 cm of peripheral tissue has been also recommended [5]. A recent systemic review of vulvo-perineal endometriosis pointed out that wide complete excision produced a lower overall recurrence rate than a mix of all kinds of excision [5].

Minimal residual of PEM and its potential effect on recurrence was discovered in our study for the first time. Based on a much larger sample size with a longer follow-up period, the rate of local recurrence in our study was significantly higher than previously reported. Surprisingly, late recurrence beyond two years accounts for 43.8%, indicating a continuous surveillance of PEM patients after surgical treatment is recommended. In the circumstances of bowel endometriosis, the impact of mPM on clinical outcome remains inconclusive. Nirgianakis et al. proposed that positive margins in segmental bowel resection might predict higher recurrence (HR = 6.5, 95% CI 1.8–23.5, P= 0.005) [28]. Despite that Roman et al. found mPM had no impact on postoperative pain and digestive function [29]. Further investigation of prospective study is required for a better understanding of clinical significance of mPM.

In our study, hormonal medication did not seem to interfere with the risk of recurrence in a statistically significant way, which deviates from the rudimentary results that GnRH-a correlated with reduced recurrence in previous observations. However, after the diagnosis of recurrence, the pain symptoms of most recurrent cases were well relieved after timely hormone intervention. Likewise, it has been reported that PEM lesions could spontaneously regress after pregnancy, suggesting the hormone-responsive feature of the disease [30]. In agreement with our findings, Seong et al. found hormone therapy was associated with longer recurrence-free interval from the time of surgery to the onset of recurrence after primary surgery for ovarian endometrioma [31]. Hormonal therapy maintains the minimal disease state by slowing down the regrowth rather than eliminating residuals, as revealed by Sharpe et al. in a rat model that the implant lesion was significantly inhibited by GnRH-a while regrowth sustained spontaneously after the cessation of hormone suppressive treatment []. Taken together, these results suggest postoperative hormonal suppression has beneficial effects on extending disease-free interval but does not completely prevent recurrence of PEM.

To our best knowledge, the present study is unique in that it includes the largest number of consecutive patients with PEM to date and conducts a profound investigation into clinical characteristics, treatment modalities and postsurgical outcomes of this rare entity. Moreover, the postoperative recurrence nomogram of PEM was established accordingly. However, there are still some limitations in the present study. Firstly, the follow-up was completed in a manner of outpatient visit and telephone enquiry. There may be recall bias about the exact time of recurrence. Secondly, it’s limited that we took symptomatic recurrence as primary outcomes on this survey because treatment was mainly dominated by symptomatic recurrence. There’s only 37.5% (5/16) of patients with recrudescence of symptoms had histopathological confirmation which can produce more convincing evidence. Thirdly, the 20% of non-respondent rate may cause some uncertainty on the interpretation of results. We lost touch with them because of invalid contact information so that their current situation cannot be obtained. Finally, due to the small number of cases, the effectiveness of this nomogram was achieved through internal verification. Therefore, it is still necessary to use external data for further verification in the future.

Leave a Comment