Prone Versus Supine Position in Abdominoperineal Resection: Outcomes in the Laparoscopic Era

Introduction

Rectal cancer represents 30% of colorectal cancers.^1,2 Surgery remains the main treatment for these tumors, but approaches have changed notably over the last 20 years due to improvements in oncological therapies and surgical techniques.

Over time, abdominoperineal resection (APR) ³ has gradually been replaced by sphincter-saving procedures. However, APR continues to be the curative approach for selected cases of low, locally advanced rectal cancers and those with complex sphincter involvement.

A modification of the Miles procedure was described and extralevator APR (ELAPR) was introduced.^4,5

The aim of this procedure is to facilitate cylindrical resection of the mesorectum and sphincter complex and reduce local recurrence.

For perineal dissection in ELAPR, patient placement in the prone position may provide benefits such as improved visualization of perineal planes and a more comfortable position for the surgeon and assistants.

APR is currently routinely performed laparoscopically and the tumor is extracted through the perineal incision. It has been postulated that placing the patient in the prone position could improve the quality of the specimen and perineal margins. ⁶

The objective of the present study was to compare the impact of patient positioning in terms of operative, postoperative, and oncological outcomes in the perineal phase for patients undergoing laparoscopic APR for rectal tumors.

Materials and Methods

In this retrospective study, we recruited all consecutive patients who underwent laparoscopic APR with curative intent for low rectal tumors at Hospital de la Santa Creu I Sant Pau from 2005 to 2018.

Exclusion criteria were open surgery as first intent, inflammatory bowel diseases or other benign pathologies, multiple metastasis and/or peritoneal carcinosis, palliative surgery, and lost-to-follow-up cases.

The study protocol followed the ethical guidelines of the 1975 Declaration of Helsinki (as revised in Brazil 2013). The study was approved by the local ethics committee who deemed that formal approval was not required owing to the retrospective, observational, and anonymous nature of this study. The results are reported according to the principles of Strengthening the Reporting of Observational Studies in Epidemiology (STROBE). ⁷ We reviewed data for all patients from medical records available in our electronic health record (EHR).

Collected preoperatory features were sex, age, body–mass index, comorbidities, carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA 19-9) blood levels at diagnosis time, tumour, node, metastasis staging (TNM) ⁸ from pelvic magnetic resonance imaging (MRI) and abdominal computed tomography (CT) scan, measurement of tumor distance from anal verge detected by MRI and colonoscopy, and neoadjuvant therapies. Intraoperative features were operative time (OT), technical complications (vascular or uro-gynecologic lesions, adhesions, and tumor perforation), patient's position during perineal time (prone position versus supine position), use of perineal prothesis, and eventual conversion.

The procedures were performed by four colorectal surgeons and cases were distributed equally among surgeons within the study period. The supine approach was performed at the beginning of the study period, but was gradually replaced with the prone approach in the second half of the study. The surgical technique was performed according to the principles of total mesorectal excision (TME) ⁹ and respecting oncological steps. The abdominal time of laparoscopic APR remained consistent throughout the study period. We used one umbilical trocar for the camera, two trocars on the right side for the surgeon, and one trocar on the left side for the assistant. Surgery was divided into two parts, vascular dissection with high ligation of the inferior mesenteric artery and dissection of the mesorectum. Before beginning the perineal phase, all surgical wounds were sutured and a terminal colostomy was created in the left iliac fossa.

In the supine group, we performed lower laparoscopic dissection of the mesorectum. The mesorectum was mobilized from the levator muscles, not from the coccyx. The perineal phase was then started from below, with excision of the rectum, including the lower portions of the levator muscles.

In the prone position, the mesorectum was mobilized up to the level of the levators during the laparoscopic part of the ELAPR. The patient then had to be turned to start the perineal dissection. The anus was closed with a purse-string suture, followed by a circumferential incision from the anal verge. Dissection of the sphincter complex was carried out following a plane approximately in line with the coccyx posteriorly and with the prostate or vagina anteriorly. The extralevator muscles were completely removed with the mesorectum. This approach normally created a larger perineal defect that sometimes needed closure with a mesh.

Complications were classified using the Clavien-Dindo scale. ¹⁰ Perineal wound infection was defined as infection of the perineal incision occurring up to 30 days after the APR surgery and was an endpoint in comparing the two groups.

A standard evaluation of the surgical specimen was carried out by a trained pathologist. The quality of the TME was assessed by a description of the integrity of the mesorectum. The circumferential resection margin (CRM) was identified as negative if the tumor was more than 1 mm from the resected margin and positive if it was within 1 mm. ¹¹ Vascular and perineal invasion was also included in the pathology report. Adjuvant therapies were collected.

Following European Society for Medical Oncology (ESMO) guidelines, ¹² all patients with rectal cancer (RC) were followed up by colorectal surgeons and medical oncologists for 5 years. Patients were checked at three-month intervals during the first 2 years. Routine postoperative controls included physical examination, blood CEA and CA 19.9 levels, colonoscopy, and abdominopelvic CT or MRI scan.

Local recurrence was defined as the presence of a lesion with positive histology in the pelvis indicated through diagnostic imaging evidence and raised oncological marker levels.

Overall survival (OS) was defined as the time from the date of the APR to the date of death for cancer progression. The Kaplan–Meier method was used to calculate survival, considering only patients without synchronous metastasis at diagnosis. Disease-free survival (DFS) was defined as the time from the date of surgery to the date of diagnosis of recurrence (either local, distant, or both).

Statistical analysis was performed using SPSS 25 (SPSS, Inc., Chicago, IL, USA). Categorical variables were compared using the chi-square test. Continuous variables were compared using t-tests and are presented as mean ± standard deviation. Oncological and survival outcomes were analyzed using Kaplan–Meier curves. Survival curves were analyzed according to whether the histopathology was adenocarcinoma or squamous cell carcinoma.

Results

The study included 123 patients who underwent laparoscopic APR from 2005 to 2018. There were 67 (54%) men and 56 (46%) women. Mean age was 72 (41–93) years.

Sixty-five patients underwent perineal dissection in the prone position and 58 in the supine position (53% versus 47%). There were no statistically significant differences between the two groups in terms of demographics, comorbidities, clinical staging, or neoadjuvant therapy (Table 1). Thirty-eight patients from the prone cohort and 30 patients from the supine cohort underwent neoadjuvant treatment (58% versus 51%, P = .48). Radiotherapy (RT) short course was used more in the prone group (17 patients versus 8 patients).

The mean follow-up, expressed in months, was 47.8 ± 30.9 (13–158) for the prone group and 67.4 ± 45.7 (28–169) for the supine group (P = .026).

Table 2 shows the intraoperative data. The mean duration of surgery was statistically longer in the prone group than in the supine group (237 ± 52.3 versus 210 ± 56.6, P = .007).

The intraoperative complication rate was significantly higher in the supine group than in the prone group (21% versus 5%, respectively, P = .042). We also included vascular lesions, uro-gynecologic injury, adhesions, and tumor perforation (Table 2).

Iatrogenic perforation of the tumor during surgery was higher in the supine group than in the prone group, but the difference did not reach statistical significance (3% versus 9%, P = .208). The incidence of conversion to open surgery was higher in the supine group than in the prone group (12% versus 0%, P = .004). A perineal mesh was placed in 15 patients to close the wound defect.

Table 3 shows hospitalization data. There were no differences between the two groups in terms of postoperative complications (prone 47% versus supine 41%, P = .48)

Perineal wound infection did not differ statistically between the two groups (prone 20% versus supine 22%, P = .93).

Table 4 shows the anatomopathological features according to histotype (adenocarcinoma versus other malignancies). Patients with positive histology for adenocarcinoma were collected in Adenocarcinoma Group (ADK-GR). The ADK-GR included 108 patients, 56 in prone (51%) and 52 in supine (49%) positions. Pathologic T and N stage distribution in ADK-GR was not statistically different between supine and prone positions (P = .485 and P = .098, respectively). The incidence of incomplete mesorectum did not show statistical significance between the two positions, but was higher in the supine group (prone 14% versus supine 25%, P = .175). The rates of CRM involvement were higher in the supine group, but did not reach statistical significance (P = .374).

Adjuvant therapy was given more often in the supine group than in the prone group (64% versus 45%, P = .041).

Long-term outcomes are summarized only for the adenocarcinoma group (Table 5).

A perineal hernia was diagnosed in four patients, two in each group (4% prone versus 9% supine, P = .955). No prophylactic mesh was placed in any of the four cases.

Mortality was significantly higher in the supine group (24% versus 4% in the prone group, P = .034). The Kaplan–Meier curves for OS are shown in Figure 1.

OPEN IN VIEWER

FIG. 1.

Overall survival. P = .034.

Recurrence was reported in 30 patients, 20 in the supine group and 10 in the prone group (P = .177). Local recurrence (LR) and distal recurrence (DR) were higher in the supine group, but the difference was not statistically significant (LR 10% versus 4% and DR 31% versus 17%, P = .177, respectively). The Kaplan–Meier curve for DFS is shown in Figure 2.

OPEN IN VIEWER

FIG. 2.

Disease-free survival. P = .177.

Discussion

In the present study, we compared the short-term and long -term outcomes of supine and prone positioning during laparoscopic APR for rectal tumors. We summarized all the papers about this topic in Table 6. We found that the OT was significantly shorter in the supine group. This result is in contrast with two meta-analyses (Table 7),^13,14 where regarding the perioperative endpoints, the OT was shorter in the prone group. Based on our experience, a longer OT in the prone position was an expected result, possibly explained by the time required to turn the patient after the abdominal phase of laparoscopic APR. Moreover, the literature does not appear to consider that turning the patient to prone position for the perineal phase allows the surgeon to start a new cleaner surgery that would be more complicated in the supine position due to Trendelenburg angulation in the laparoscopic part of APR. The prone approach provides excellent exposure of the posterior perineum and anterior face of the rectum, improving surgeon comfort and enabling better lighting of the surgical field.

Another potentially critical aspect of APR surgery is the postoperative course with respect to the perineal wound. In our study, and similar to findings in the literature, perineal wound infection and complication rates between the two groups did not differ statistically.^13–15

An interesting aspect of our work is the trend toward better prognostic oncological factors in the prone group. We considered the quality of the mesorectum as an outcome and correlated this with intraoperative tumor perforation and CRM involvement. In a recent meta-analysis, McKechnie et al. suggested that prone position may lead to a lower perforation rate (P = .003) and lower CRM positivity. ¹⁴

We found that a small number of patients presented with perineal hernia (3%). This outcome did not differ statistically between the two groups. Han et al. ⁵ confirmed that the position of the patient did not influence the risk of perineal hernia even though the larger perineal defect in ELAPR requires repair with a mesh. In 15 patients, we decided to position a prophylactic perineal mesh and none presented hernia, thus supporting the systematic use of this technical option in selected cases.

Follow-up time between the two groups was statistically different because in more recent times, we preferentially chose the prone approach.

The prone position has been adopted in several centers in an attempt to improve oncological results. In one retrospective study, ¹⁶ De Campos et al. reported that surgical positioning during the perineal phase of APR did not affect oncological outcomes: they found that OS and DFS Kaplan–Meier results were similar. However, in our study, the OS curve was significantly better in prone position, suggesting that this approach could play a role in improving oncological results.

Most retrospective studies reported that long-term oncological results did not differ significantly between the two positions.^17–20 However, in the study by Tayyab et al., we observed a trend toward higher local and distal recurrence in the supine position. ²¹

Neoadjuvant therapy is thought to be important for reduction of local recurrence.^22,23 In this study, more cases in the prone cohort underwent neoadjuvant treatment in comparison with the supine cohort, without a significant difference between the two groups. According to guidelines, recently, the RT short course was definitely selected more and therefore involved the prone group more than the supine group. However, this did not change the rates of CRM positivity and oncological quality of the specimen.

Our research has several limitations. The first is that it was a retrospective study of data taken from our prospective database. The second limitation is the length of the follow-up, which was shorter in the prone cohort than in the supine group. Indeed, the prone approach is a relatively recent technique, but the risk of recurrence in rectal cancer is highest in the first 2 years of follow-up.

Conclusions

Our study supports the interest in improving and increasing prone positioning for laparoscopic APR for rectal cancer because results appear to be better than those for the classic approach. Morbidity is similar to that for the supine position and long-term outcomes are better. Statistical significance is lacking, however, in some endpoints. Prospective randomized studies are needed to compare the oncological results of supine and prone positioning for patients undergoing laparoscopic APR for rectal cancers.