Operative management of deeply infiltrating endometriosis: results on pelvic pain symptoms according to a surgical classification.

Study Objective: To assess the results of complete surgical excision for patients with painful functional symptoms in a context of histologically proven deeply infiltrating endometriosis (DIE).

Design: Retrospective analysis (Canadian Task Force classification II-2).

Setting: University-affiliated hospital.

Patients: One hundred thirty-two patients with pelvic pain symptoms and histologically proved DIE. The DIE lesions were classified according to surgical classification: uterosacral ligaments (USL), vagina, bladder, or intestine.

Intervention: Complete surgical excision of DIE lesions.

Measurements And Main Results: A retrospective analysis was made of medical, operative, and pathologic reports as well as of questionnaires mailed to patients. Efficiency of surgical excision was assessed according to two methods: objective evaluation (numerical rating scale) and subjective evaluation (patients were asked to classify the improvement after surgery with one of the following: excellent, satisfactory, slight, or no improvement). For each symptom, the mean scores according to the numerical rating scale were significantly lower postoperatively. The difference between the preoperative and postoperative scores was 5.2 points +/- 3.6 for dysmenorrhea, 4.6 points +/- 3.1 for deep dyspareunia, 4.4 points +/- 3.7 for painful defecation during menstruation, 4.9 +/- 3.2 for lower urinary tract symptoms during menses, and 4.6 points +/- 3.4 for noncyclic chronic pelvic pain. Comparable results were observed for patients in each group according to the surgical classification of their DIE lesions: USL (n = 78 patients); vagina (n = 25 patients); bladder (n = 13 patients); and intestine (n = 16 patients). Subjective evaluation showed that the improvement was considered to be excellent in 40.2% of women (53 patients), satisfactory in 42.4% (56 patients), slight in 14.4% (19 patients), and nonexistent in 3.0% (4 patients). The patients' characteristics (i.e., age, gravidity, parity, body mass index, preoperative medical treatment, follow-up after surgery, number and location of DIE lesions, revised American Fertility Society stage, associated endometrioma) did not differ significantly according to whether the improvement was considered to be excellent (Group A: 53 patients) or not (Group B: 79 patients). Among the infertile patients (n = 78; 59.1%), there was no difference in pain improvement if the patient was pregnant or not in the 42 women who achieved pregnancy after the surgery.

Conclusion: Complete surgical excision of DIE lesions results in a statistically significant reduction in painful functional symptoms. These results are observed whatever the main location of DIE lesions. The patients' preoperative characteristics have no significant influence on the result.