In our modified ELAPE, the perineal procedure in prone jackknife position was firstly performed, and the abdominal procedure in supine position was then conducted. In the perineal operation, the coccyx was not routinely surgically removed, and the pelvic peritoneum was directly closed by laparoscopic approach without the application of biological mesh. The modified ELAPE was successfully performed in all the patients in this study, and this novel surgical technique was demonstrated to be safe, even in patients with advanced age (range 75–83 years). In this novel ELAPE, the sequence of perineal operation first and abdominal operation later avoids the squeeze of abdominal incision and colostomy stoma, and thus lowers postoperative complications associated with colostomy stoma, such as necrosis, stenosis, and parastomal hernia . Moreover, the operative field in the perineal operation would not be influenced by the errhysis following the abdominal operation as seen in Holm’s ELAPE, which is critical to avoid pelvic vascular and nerval injury.
In our opinion, prone jackknife position provides a better surgical field to the surgeon, which enables a comfortable manipulation. In addition, the improved visual field helps the surgeon to clearly define the resection range, which in consequence reduce the incidence of intestinal perforation. Meanwhile, vascular and genital nerval injury could be avoided and thus the incidence of intraoperative bleeding and sexual dysfunction could be reduced. Research results by Dalton et al.  and Shiha et al.  suggested that the prone jackknife position in perineal operation in ELAPE was more suitable for patients with tumor infiltrating the prostate and posterior vaginal wall. No positive CRM and intraoperative perforation were found in our study, and only 1 male patient developed sexual dysfunction, the incidence of which was lower than that reported previously . In our experience, although the change of the patient’s position seemed to increase surgical risk and prolong the operative time, but the application of prone jack-knife position to the perineal operation was convenient for the surgeon to operate under the direct vision and clear surgical level, in addition, it could shorten the learning curve. Generally speaking, this reduced perineal complications and shortened the overall operation time instead. The mean operative time was 213.5 min, which was less than that reported in the previous studies [10, 23, 24]. With the accumulation of technical experience and improvement of laparoscopic techniques, the overall operation time of this novel ELAPE performed by our surgical team had been reduced to currently around or less than 180 minutes.
Perineal pain was the common postoperative complication in ELAPE. It has been indicated that the postoperative perineal pain may be related to the coccygectomy, the activation of inflammatory cytokines at the mesh site, the damage to the pudendal nerve, the wider excision of the levator ani muscles and ischiorectal fossa fat, and the suturing of the mesh itself close to the pelvic wall , among which the coccygectomy may be the main relation . In ELAPE of Holm et al. , the coccyx is routinely resected to permit entry into the pelvic cavity at the point where the intra-abdominal dissection stopped, and the mesorectum needs to be turned out from the pelvis, followed by removal of the specimen from the perineal incision. Partial distal sacrum may even be resected in case the mesorectum is hypertrophy or the tumor is relatively huge. In our modified ELAPE, the specimen was removed from the location of colostomy, thus coccygectomy was not required in all cases. Whether to perform coccygectomy should be determined by the location of rectal tumor and the extent of invasion, the information of which was obtained by careful evaluation of the preoperatively MRI imaging. If the tumor locates at the anterior and lateral wall of the rectum, the coccyx could be retained. In case the tumor locates at the posterior wall of the rectum, the coccyx could still be retained in the condition of ensuring negative CRM. No patients in the present study underwent coccygectomy, and no sacrococcygeal pain occur in any patient.
The wider excision of the levator ani muscles and ischiorectal fossa fat leads to a large perineal defect at the level of the pelvic floor, which might result in increasing incidences of perineal complications, such as perineal wound infection and perineal hernia. The reconstruction of pelvic floor is critical for decreasing perineal morbidity. The currently reported methods of pelvic floor reconstruction mainly include primary closure , reconstruction with myocutaneous autologous flaps , reconstruction with biologic meshes [22, 25], and the pedicled omentoplasty . Though the studies reported acceptable or favorable results regarding the methods mentioned above, no consensus was achieved on the optimal method. A recent multi-center retrospective study indicated that the application of biological mesh could not reduce the incidence of perineal hernia, and even increase perineal morbidity . Result from a meta-analysis study found that compared to primary closure, reconstruction with biologic mesh was associated with a lower hernia rate, but it had no effect on perineal wound complications . Therefore, the benefit of the application of biologic mesh in pelvic floor reconstruction remains controversial. In our opinion, wider excision of the ischiorectal fat is not necessary if the tumor do not infiltrate the ischiorectal fossa, and in this situation, the usage of gluteal muscle flap or biological mesh implants to pelvic floor reconstruction could be avoided. In all the cases in our study, the perineum incision was easily sutured in two layers and the pelvic peritoneum was closed laparoscopically, without coccygectomy and complex pelvic reconstruction. The closure of pelvic peritoneum has been reported to prevent the small bowel from descending into the pelvic cavity, thus avoiding perineal hernia and obstruction caused by adhesion of the small bowel in the pelvis . In our study, no small-bowel obstruction or perineal wound infection was observed, and only 1 patient had perineal hernia during the long-term follow-up period. The low incidence of perineal complication in our study may not only be related to the closure of pelvic peritoneum, but be also associated with the retention of coccyx and the routine use of presacral drains.
Discrepancies in regard to the long-term outcomes of ELAPE can be found in previous studies. A retrospective study with long-term follow-up period showed that the local recurrence rate was 7% in ELAPE group, whereas long-term survival did not differ between ELAPE group and APE group . Results from a recent single-center study revealed that the local recurrence rate of ELAPE reached 6.7%, and the 3-year and 5-year OS rates were 86.4 and 58.8%, respectively . In the present study, the local recurrence rate of the modified ELAPE was 1.9% and the median observation time was 65 months. Pulmonary metastasis was the most frequently observed distant metastasis, followed by bone metastasis, and hepatic metastasis. This result was in consistent with a previous study conducted by Qiu et al.  The 5-year OS and DFS rates of this study were 76.4 and 70.9%, respectively. As is well known, ELAPE procedure is recommended for T4 tumors or advanced T3 tumors. Some patients with T1/2 tumors more than 3 cm from the anal verge in our consecutive series were performed with ELAPE procedure for reasons mentioned earlier. According to the pathologic and 5-year oncologic outcomes of these patients, we suggested that extended excision of the pelvic floor was not necessary in higher T1/2 tumors without infiltration of the pelvic floor or incontinence as indication. The main limitation of the present study is that it was a retrospective cohort study, which lacked control groups. Another limitation of this study is the small sample size with single institution.