Between January, 2013 and December, 2014, a cohort of 51 patients (22 men and 29 women), with a median age of 52 years (range, 36–75 years) who had undergone LLAR at the Department of General Surgery, Sixth People's Hospital Affiliated to Shanghai Jiao Tong University (Shanghai, China) were included in our study. All the patients were clinicopathologically diagnosed with adenocarcinoma of the rectum (within 5–12 cm above the anal verge). Tumor location was assessed by preoperative examination, including digital rectal examination and colonoscopy. In addition, the postoperative anastomotic height (measured from the anastomosis to the anal verge) was assessed by colonoscopy and the patients were allocated into two groups: The high anastomosis (HA) group, comprising those with anastomosis >5 cm above the anal verge, and the low anastomosis (LA) group, comprising those who had an anastomosis within 5 cm of the anal verge. There was no significant difference in general information between the two groups (all P>0.05) as shown in Table I. The exclusion criteria were applied as follows: i) Abdominal computed tomography and magnetic resonance imaging showing distal metastases or local invasion into the levator ani muscle and the external sphincter; ii) laparoscopic surgery converted to open surgery due to unexpected factors; and iii) patients with accompanying symptoms including intestinal obstruction or perforation, dementia and cognitive dysfunction. Furthermore, another cohort of 50 patients (32 men and 18 women), with a median age of 51 years (range, 32–72 years) who had undergone other abdominal surgeries (not involving the pelvis) were also included in our study as the control group to test the consistency of the baseline and reduce bias. All the patients received the same chemotherapy (12 courses of FOLFOX4) postoperatively and patients who received radiotherapy were not included in our study. The study was approved by the Ethics Committee of The Sixth People's Hospital Affiliated to Shanghai Jiao Tong University. Written informed consent was obtained from all the patients prior to enrolment.

All the surgical procedures were conducted by the same group of surgeons experienced in laparoscopic colorectal surgery; bowel preparation and perioperative intravenous antibiotic prophylaxis were routinely performed. The laparoscopic process for tumor resection has been previously described (20). In brief, after establishing a pneumoperitoneum, a sharp dissection involving a total mesorectal excision was performed by an ultrasound knife. The inferior mesenteric vessels were ligated following identification of the left ureter, and the distal rectum was successively separated. The proximal end of the bowel was delivered through a small incision. Finally, the bowel was resected with a distal mesorectal margin of ≥2 cm whenever possible, and the double-stapling technique was used to perform an anastomosis as described by Kosmidis et al (21). The presence of an anastomotic leak was checked by transrectal insufflation of air. For postoperative oncological follow-up, the patients underwent laboratory tests every 3 months and radiological examination every 6 months to monitor local recurrence and/or distal metastasis. Endoanal ultrasonography was also employed to investigate the integrity and thickness of the internal anal sphincter (IAS) and the external anal sphincter (EAS) postoperatively.

For anorectal manometry, an 8-channel water-perfused catheter with an external diameter of 5.5 mm and a computer system (all from Medical Measurement Systems Corporation, Enschede, Netherlands) were employed. Each patient assumed the left lateral decubitus position and the stationary technique was used for catheter insertion. Anorectal manometric parameters, including mean/maximal anal resting pressure (mean/max ARP), maximal squeeze pressure (MSP), initial/strong sensory volume (ISV/SSV), maximal tolerable volume (MTV) and rectoanal inhibitory reflexes (RAIR) were recorded. In a proportion of patients with low anastomosis, postoperative manometry performed within a short period may result in rupture of the anastomosis. Therefore, manometry was performed at 1 week preoperatively and at 3, 6 and 9 months postoperatively.

The anal function was assessed based on Wexner incontinence grading scale, while QoL was assessed by the QLQ-C30 and QLQ-CR38 questionnaires developed by the European Organisation for Research and Treatment of Cancer. The Chinese versions of QLQ-C30 and QLQ-CR38, the clinical validity and reliability of which have been previously confirmed, were used for the questionnaire survey (10,22). For this questionnaire survey, all the patients were interviewed with standardized questionnaires at 1 week preoperatively and at 3, 6 and 9 months postoperatively.

The results are presented as mean ± standard deviation (SD). The Student's t-test and the χ² test were performed by SPSS 17.0 statistical software (SPSS, Inc, Chicago, IL, USA). A P-value of <0.05 was considered to indicate statistically significant differences.

All the patients were followed up for at least 1 year and there were no reported deaths. Furthermore, radiological examination indicated no local recurrence or distant metastasis. Four patients had anastomotic leakage and received conservative treatment. One patient of the LA group complained of severe stool incontinence and his manometric parameters were by 50% lower compared with those of other patients at 9 months postoperatively. Therefore, this case was not included in our data.

First, as shown in Table II, there was no significant difference in the preoperative pressure between the control and trial groups (all P>0.05). Generally, all the manometric parameters changed to different degrees following LLAR (Fig. 1A-C). For the HA group, the mean/maximum ARP and MSP were decreased at 3 months postoperatively (all P<0.05), but had returned to preoperative levels at 6 months postoperatively (all P>0.05). For the LA group, the observations were similar at 3 months postoperatively (all P<0.05), but the parameters gradually returned to normal at 9 months postoperatively (P>0.05), except for MSP, which had returned to normal level at 6 months (P>0.05). With regard to intergroup comparison, the MSP of the HA group was significantly higher compared with that of the LA group at 3 months (P<0.05). In addition, the mean/maximal ARP of the HA group was significantly higher compared with that of the LA group at 6 months postoperatively (all P<0.05).

The ISV, SSV and MTV were significantly decreased in both groups at 3 months postoperatively (all P<0.05), but had gradually improved at 6 and 9 months postoperatively (Fig. 1D-F). However, all the volume parameters remained significantly lower compared with the preoperative level at 9 months postoperatively (all P<0.05). Furthermore, the MTV of the HA group was significantly higher compared with that of the LA group at 9 months postoperatively (P<0.05).

The changes in RAIR following LLAR are shown in Fig. 2A. Generally, RAIR was absent in 2 of the 50 patients at 1 week preoperatively. For the HA group, RAIR was absent in 12 of the 31 patients at 3 months postoperatively, whereas RAIR was absent in none of the patients at 6 and 9 months postoperatively. For the LA group, RAIR was absent in 10 of the 19 patients at 3 months postoperatively. However, RAIR was induced in 16 of the 19 patients at 6 and 9 months postoperatively, with 3 patients remaining negative for RAIR (including 2 patients who were negative for RAIR preoperatively). In addition, the RAIR rate of the HA group was significantly higher compared with that of the LA group between 3 and 9 months postoperatively (all P<0.05). Endoanal ultrasonography demonstrated rupture of IAS in 6 patients (1 in the HA and 5 in the LA group), with full integrity of EAS in all the patients. The changes in the thickness of IAS and EAS are shown in Fig. 2B and C. There was no difference in the thickness of IAS/EAS between the control and trial groups (all P>0.05, Table II). For the HA group, although a decreasing tendency was observed in the thickness of IAS and EAS at 3 months, the results were not statistically significant (all P>0.05). Furthermore, at 6 and 9 months, the thickness of both the IAS and EAS had increased to the preoperative level (all P>0.05). For the LA group, similar changes were only found in EAS; the thickness of the IAS was significantly reduced at 3 months postoperatively (P<0.05), with an insignificant increase at 6 and 9 months postoperatively (all P>0.05).

The changes in gas, liquid and solid incontinence rates are presented in Fig. 2D-F. The patients of the control and trial groups reported no incontinence prior to surgery. However, both subgroups experienced incontinence to different degrees at 3 months after LLAR (as shown in Fig. 2D-F). The liquid incontinence rate of the LA group was significantly higher compared with that of the HA group postoperatively (all P<0.05), but gradually improved. The changes of the Wexner scores are demonstrated in Fig. 2G. For the HA group, the Wexner score exhibited an increase at 3 months postoperatively, but was decreased at 6 and 9 months, which was also observed in the LA group. Moreover, the Wexner scores of the HA group were significantly lower compared with those of the LA group at 3 and 9 months postoperatively (all P<0.05).

The functioning scales of QoL-C30 are shown in Fig. 3. Compared with the preoperative level, the function scores of both groups decreased at 3 months postoperatively, but gradually increased by 9 months postoperatively, except for cognitive function (Fig. 3E). For the symptom scales of QoL-C30, as shown in Fig. 4, the reverse trends were observed, except for dyspnea (Fig. 4D) and constipation (Fig. 4G). On intergroup comparison, the HA group had a significantly better role, emotional and social function (Fig. 3C, D and F), and fewer diarrhea problems (Fig. 4H) at 6 months postoperatively compared with the LA group (all P<0.05).

With regard to QoL-CR38 scores, as shown in Fig. 5, certain scales had changes similar to QoL-C30, such as sexual function (Fig. 5A), body image (Fig. 5B) and micturition problems (Fig. 5D). Chemotherapy-related side effects and gastrointestinal tract symptoms, which were increased prior to 6 months, had decreased to preoperative levels at 9 months postoperatively (Fig. 5E and F). Although the future perspective score was decreased at 3 months and slowly increased by 9 months postoperatively, it did not return to the preoperative level (Fig. 5C, P<0.05). Moreover, there was a decreasing trend in the scores of defecation problems (Fig. 5G) and weight loss (Fig. 5H) postoperatively. On intergroup comparison, the HA group appeared to have a better sexual function (Fig. 5A) and fewer micturition problems (Fig. 5D) at 3 and 6 months postoperatively, with a better body image (Fig. 5B) and fewer chemotherapy-related side effects (Fig. 5E) and defecation problems (Fig. 5G) at 6 months compared with the LA group (all P<0.05).