During the study period a total of sixty-five patients underwent nerve-sparing surgery for parametrial deep infiltrating endometriosis. Fifty-nine patients met the inclusion criteria and were included and fifty-one of these agreed to fill out questionnaires for postoperative outcomes evaluation.
Median age was 36 ± 6.2 years. The majority of these patients were married or a common-law wife (72.5%) and employed (88.2%) at the time of surgery. Demographic characteristics of patients are provided in Table 1.
Eight patients had had previous surgery for endometriosis (15.6%) and twenty-seven were nulliparous (53%).
Diagnosis of parametrial endometriosis was confirmed by histological examination in all patients included. Laparoscopic approach was used in all cases, no conversion to open surgery occurred. Laparoscopic segmental bowel resection was performed in fifteen patients, while discoid resection was performed in only one. Three patients needed protective ileostomy for ultra-low rectal anastomosis. Operative details and procedures are summarized in Table 3. The mean operating time was 239 +/- 111 minutes, the mean estimated blood loss during surgery was 165 +|- 59.9 mL (range 50-1020 mL) and the mean hospital stay was 6 days (range 3-12 days). The average length of the removed intestinal segments was 12.5 cm (+/- 5,7).
In thirty-seven patients (62,7%) parametrial endometriosis was unilateral, in 22% of cases both posterior and lateral parametrium were involved and in 15.3% of cases posterior and lateral parametria were involved bilaterally.
Right posterior parametrium was involved in 38% of cases (19/50), while the left one was involved in 40% of cases (20/50). Right lateral parametrium was involved by endometriosis in 59.1% of cases (13/22), the left one was involved in 31.8% of cases (7/22), and lateral parametria were involved bilaterally in 9.1% of cases (2/22). The average size of parametrial nodules was 2 +/- 0.5 cm. No cases of intrinsic ureteral endometriosis were found.
Endometriosis of the anterior compartment was detected in 41.2% of patients with three cases of DIE of the bladder (5.9%). Others DIE localizations are shown in Table 2. Posterior compartment was most frequently involved by DIE than the anterior with the presence of nodules of the rectosigmoid colon in 69.5 % of the cases. Ovarian endometriosis was detected in 74.5% of patients. Pre-operative pain symptoms, evaluated with VAS score, are reported on Table 5. Dysmenorrhea was the most frequent symptom (84.3%), followed by dyspareunia (74.5%), ovulatory pain (70,6%), dyschezia (51%), chronic pelvic pain (35.3%) and stranguria (27.5%).
Intraoperative complications occurred in five patients. Four patients had monolateral hypogastric nerve involvement and in two of the four patients neuroablation was necessary due to the need for radical resection of nodules. In one patient sigmoid lesion occurred during the shaving procedure and required the intervention of the general surgeon for the suture. A total of three postoperative complications were registered: one hemoperitoneum (class IIIb) that required a second surgery, one rectovaginal fistula requiring three operations for complete resolution (class IIIb) and we had one case of stenosis of colorectal anastomosis requiring endoscopic balloon dilation (class IIIa).(36)
Twenty-seven patients (52.9%) were undertaking estroprogestins at the post-operative follow up visit (3 months).
Pain symptoms (expressed in VAS score) were significantly decreased after the operation (p value < 0,05) as shown in figure 1. Post-operative use of analgesic drugs also decreased from 37.5% (19/51) to 5.9% (3/51).
The EHP-30 score variation is shown in Table 4. The comparison of pre-operative and post-operative scores showed that surgery improved significantly QoL in many of the domains analyzed such as pain, control and powerlessness, emotional well-being, social support, self-image, satisfaction of treatment, sexual life. It was not possible to evaluate the “infertility” and “relationship with children” modules because respectively 21 (41.2%) and 31 (60.8%) patients did not answer the questions respectively. Limited to the small sample analyzed, however, even in these two modules a statistically significant improvement was observed.
No differences were found in terms of urinary function between pre and post-operative questionnaires (ICIQ-FLUTS). The correlation between resected parametrium and urinary symptoms was examined: of the seven patients undergoing lateral parametrium resection, four had no symptoms (57.1%), three had mild difficulties in bladder filling (33.3%) and one complained of mild incontinence; of the thirty-six patients undergoing posterior parametrium resection, twentyfour (66.7%) had no symptoms, while twelve (33.3%) complained of mild incontinence. Six of the eight patients undergoing resection of both parametria developed difficulties in bladder filling, one was asymptomatic and one developed mild incontinence.
Bowel function has improved in patients' subjective perception, although the results are controversial: while the surgery was resolutive for diarrhea and alternating alvus, there was no improvement in constipation. In fact the NBD score showed that intestinal dysfunction in the whole group was related to constipation. Most of the patients showed mild dysfunction (76.4%), while only three patients had severe dysfunction (5.9%).
NBD scores of patients undergoing recto-sigmoid shaving was compared with that of patients undergoing bowel resection: data are shown in figure 2.