In rectal carcinoma surgery, the minimization of local recurrence and preservation of sphincter function is a concern. Historically, RA has been the preferred method for low-lying rectal cancer near the anus. Owing to ISR, sphincter-preserving resection has made great progress. According to Rullier et al., ISR made it possible to preserve the anus in patients with low-lying type–3 tumors near the anus . It is difficult to accurately diagnose the invasion depth of tumors for low-lying tumors near the anus preoperatively; therefore we have no choice but to diagnose the tumors during surgery. If conversion to RA from ISR is required, RA with an initial perineal approach must be reasonable and may have benefits. However, there are no reports on the feasibility of this approach. In this study, we compared the surgical and oncological outcomes between the two RA approaches for primary anorectal cancer. Consequently, we found three advantages to the perineal approach, which should be the first RA technique for the selected primary anorectal carcinoma.
The first advantage was that the operation time in the perineal group was significantly shorter than that in the conventional group (313 vs. 388 minutes; p = 0.0275). Laparoscopic surgery for RA remains challenging. It is very difficult to perform TME toward the pelvic floor laparoscopically, especially in males with a narrow pelvis and in patients with a bulky tumor located in the pelvis. In RA, a perineal retrograde anorectum dissection prior to the transabdominal maneuver might make RA easier and decrease the operation time.
The second advantage was that the occurrence of PWC was significantly lower in the perineal group than in the conventional group (22.9% vs. 57.1%; p = 0.006). In colorectal surgery, surgical site infection (SSI) was reported more frequently. Ata et al.  reported that SSI in colorectal surgery developed 3.8 times more often than SSI in noncolorectal general surgery. Additionally, the incidence of SSI in rectal surgery is higher than that in colon surgery. In particular, SSI following abdominoperineal resection (APR) is common . In Japan, the SSI occurrence rate was reported to be 25–47% [8, 9]. The large amount of dead space in the pelvis following RA, closure under tension and the closure of a wound in an area that has a high bacterial count [10, 11] may be causes for the rate of SSI. In this series, PWC occurred in 31.9% of all patients who underwent RA for primary anorectal carcinoma.
Various risk factors such as smoking [12 - 14] hypoalbuminemia , ASA classification  obesity [17, 18], weight loss , the number of comorbidities [20, 21] as a patient-related factor, APR for recurrent cancer , neoadjuvant radiation therapy/chemoradiation therapy [23 - 25] as a tumor-related factor, flap reconstruction [26, 27] and extralevator APR [28, 29] as an operation-related factor are reported risk factors for PWC after APR. Generally, SSI risk factors, including a prolonged operation time, extensive bleeding, intraoperative blood transfusion and other risk factors related to surgery, are well known. Shortening the operation time by starting with a perineal approach might have reduced the incidence of PWC in this study.
Traditionally, certain treatments, including omentoplasty, perineal mesh placement, and flap reconstruction, have been performed to reduce PWC incidence following RA . More recently, negative pressure wound therapy (NPWT) for the perineal wound following RA was reported as a new wound management technique . For primary closed perineal wounds following RA, NPWT was associated with a reduced incidence of perineal SSI compared with only a gauze dressing . On the other hand, van der Valk MJM et al.  reported in a pilot study that incisional NPWT decreased the duration of wound healing but did not reduce the rate of wound complications. The significance of NPWT for PWC following RA is still unknown.
The third advantage was that positive CRM was significantly lower in the perineal group than in the conventional group. CRM is an important aspect in minimizing local recurrence after rectal cancer surgery [34, 35]. In this series, although there were no patients with positive CRM in the perineal group, four male patients in the conventional group presented positive CRM. For these four patients, the tumor was located at the anterior aspect of the lower rectum. RA with an initial perineal approach has the advantage of surgical margin safety due to direct visual observation during rectum dissection, especially when the tumor is located at the anterior aspect of the rectum. The result in this study might be due to the advantage of this approach, but there was no difference in the rate of recurrence between the two groups.
There were some limitations in this study. First, the sample number of patients who underwent RA, especially those who underwent laparoscopic surgery, was very small; there were 12 patients in the perineal group and 4 patients in the conventional group. This was a retrospective and nonrandomized study. The results may have been affected by its retrospective design for selective primary anorectal carcinoma. Additionally, although possible PWCs were diagnosed by two surgeons who routinely conducted rounds after surgery to observe the wounds, there was a risk of underreporting the incidence of PWC following RA. Second, the decision of whether a patient should undergo the initial perineal approach or initial abdominal approach was not standardized. In this study, the surgical approach for RA was chosen according to surgeon’s preference. It was not known whether the decision was made based on tumor characteristics or other factors. The surgeon alone characterized the operation as either an initial perineal or initial abdominal approach.