|Home | About | Journals | Submit | Contact Us | Français|
Conventional abdominoperineal excision for low rectal cancer has a higher local recurrence and reduced survival compared to anterior resection. An extralevator abdominoperineal excision (ELAPE) may improve outcome through removal of increased tissue in the distal rectum. Experience with ELAPE is limited and no studies have reported on quality of life (QOL) following this procedure. We describe a minimally invasive approach to ELAPE within an enhanced recovery programme, and present short-term results and QOL analyses.
All laparoscopic ELAPEs were included in a prospective database. Demographics, intra-operative and post-operative outcomes were evaluated. Postoperative QOL was assessed using the European Organisation for Research and Treatment of Cancer (EORTC) questionnaires QLQ-C30 and QLQ-CR29.
Thirteen laparoscopic ELAPEs were performed over a two-year period. All were enrolled in an enhanced recovery programme. The median age was 76. The median tumour height was 20mm (range: 0–50mm) from the dentate line and all patients received neoadjuvant treatment. The median duration of surgery was 300 minutes (range: 120–488 minutes), the mean blood loss was 150ml and one procedure was converted to open surgery. There was no circumferential resection margin involvement or tumour perforation. The median duration of use of intravenous fluid, patient controlled analgesia and urinary catheterisation was 2, 2 and 2.5 days respectively and the median length of hospital stay was 7.5 days. Two patients developed perineal wound dehiscence. QOL analysis revealed high global health status (90.8), physical (91.3), emotional (98.3) and social functioning (100) scores, which compared favourably with EORTC reference values and published QOL scores following conventional abdominoperineal excision.
Laparoscopic ELAPE within an enhanced recovery setting is a feasible and safe approach with acceptable short-term outcomes and post-operative quality of life.
Abdominoperineal excision (APE) for low rectal cancer is associated with considerably higher rates of circumferential resection margin (CRM) involvement, intra-operative tumour perforation (IOP) and local recurrence, and leads to poorer survival when compared to anterior resection (AR).1–4 This difference may be attributable in part to the anatomical tapering of the distal mesorectum with smaller tissue volumes surrounding lower tumours and a greater risk of IOP and CRM involvement, both of which are strong predictors of local recurrence and survival.
In response to these concerns, Holm et al5 among other groups4,6,7 proposed a change in approach to APE surgery towards that described more than a century ago by Miles,8 namely a radical removal of the perineal and pelvic structures, referred to as ‘the cylinder’.9 This extralevator abdominoperineal excision (ELAPE) or cylindrical APE aims to improve oncological outcome through removal of increased tissue in the distal rectum and en bloc excision of the levator ani. This creates a more cylindrical specimen without a waist and is associated in early reports with reduced CRM involvement, IOP and local recurrence when compared with conventional APE.5,7
In theory, this more radical approach also has the potential to cause greater morbidity and reduced quality of life (QOL). However, to date no studies have considered these outcomes. A number of studies and meta-analyses have considered QOL following conventional APE,10–16 which may allow some comparison to QOL following ELAPE.
A further consideration is the approach to the abdominal component of APE. Laparoscopic colorectal resection is widely established and the benefits and safety of laparoscopic colorectal surgery have been reported extensively.17,18 Nevertheless, so far no studies with an exclusively laparoscopic approach to ELAPE have been described. Additionally, enhanced recovery after surgery (ERAS) programmes that optimise pre-, peri- and post-operative factors to minimise the stress of surgery are commonly practised but not reported in the context of ELAPE surgery.19 The present study presents our short-term outcomes and QOL analysis after laparoscopic ELAPE (lap-ELAPE) within the context of an ERAS programme.
All patients who underwent lap-ELAPE for low rectal cancer at our institution between March 2009 and March 2011 were identified from a prospectively populated database. All had been enrolled in a standard enhanced recovery programme based on the description by King et al.20 Institutional review board approval was obtained prior to commencement of this study. Data collected included neoadjuvant therapy use, operation duration, ASA (American Society of Anesthesiologists) grade, tumour height, tumour histology, CRM involvement, IOP, relevant or major morbidity, mortality, return to theatre, ERAS-related outcomes, readmission and local or distant recurrence at follow-up.
As recommended by the Association of Coloproctology of Great Britain and Ireland guidelines for the management of colorectal cancer,21 the fifth edition of the American Joint Committee on Cancer TNM (tumour, lymph nodes, metastasis) classification system was used. Neoadjuvant therapy was administered as short course pre-operative radiotherapy defined as 25Gy in 5 consecutive daily fractions or long course chemoradiotherapy with 45Gy in 25 fractions with twice daily capecitabine (900mg/m2). The plane of surgery was graded as described by Nagtegaal et al4 although the additional classification system around the levator complex was not applied until more recently. Tumour height was determined by pathological examination of the specimen and was defined as distance from the dentate line.
QOL was assessed post-operatively on a single occasion using the European Organisation for Research and Treatment of Cancer (EORTC) core questionnaire (QLQ-C30) and colorectal cancer module (QLQ-CR29).22–24 All scales were compared using Student's t-test with stage-matched and unmatched published EORTC reference values for patients with colorectal cancer produced by the EORTC with pooled results from clinical trials and epidemiological studies. QOL results were also compared to the QOL score following conventional APE in studies identified by Cornish et al in their meta-analysis.16
Lap-ELAPE was performed by a laparoscopic technique following the principles laid down by Holm et al.5 Patients were placed in the Lloyd-Davies position and a standard four-port technique was adopted. After a thorough evaluation, the omentoplasty was commenced by dividing the attachments of the greater omentum to the transverse colon in a right to left direction with care to preserve omental vasculature. When the lesser sac was entered, the right omental vessels were divided close to their origin at the right gastroepiploic artery, thereby creating an omental flap based on the left omental vessels. Once adequate length had been obtained for the omental flap to reach the pelvis, the left colonic dissection was commenced. A medial to lateral or lateral to medial dissection technique was used according to preference. The sigmoid colon was mobilised from retroperitoneal structures and the inferior mesenteric vessels were ligated. Where possible, the ascending left colic artery was preserved.
Pelvic dissection was then commenced and the rectum mobilised posteriorly to the upper border of the coccyx, which is the lower limit of the ELAPE procedure as described by Holm et al.5 The exact position of the coccyx was confirmed by digital rectal examination to palpate the coccyx from below while manipulating a laparoscopic bowel grasper on the coccyx from above (Fig 1). Where this was not possible, we performed a limited posterior perineal dissection while the patient was still supine to assist in identification of the lower limit of the laparoscopic posterior dissection. We dissected posterior to the coccyx and then used this to palpate the abdominal grasper while in this position. The perineal dissection could then be completed in the prone position as described below.
In either case, the lateral and anterior limits of the abdominal phase were then pursued with the aid of the posterior landmark, terminating the dissection below the autonomic nerves laterally and just inferior to the seminal vesicles in men or the cervix in women. The proximal colon was divided at a position to allow optimal formation of an end left-sided colostomy and the omental flap sutured to the specimen side of the colon to allow delivery of the flap into the perineum during extraction. Haemostasis was confirmed and the colostomy fashioned in the pre-marked location on the anterior abdominal wall.
The patient was subsequently positioned in the prone jack-knife position and the perineal dissection conducted as previously described by Holm et al.5 Pelvic floor reconstruction was performed using a combination of biologic mesh and omentoplasty, and supplemented by retroverting the uterus in women. Biologic grafts used in the present study were cross-linked porcine dermis (Perma-col®, Tissue Science Laboratories plc, Aldershot, UK) and porcine small intestinal submucosa (Surgisis®, Cook Surgical, Bloomington, IN, US).
Thirteen consecutive lap-ELAPE cases were performed over a two-year period. Patient demographics, pre-operative, intra-operative and post-operative data are summarised in Table 1. The median age of patients was 76 (range: 49–84) and 7 patients were female. All cases were predicted TNM stage II/III pre-operatively, with T3 or T4 disease. All patients received neoadjuvant treatment, with five patients receiving short-course preoperative radiotherapy and eight patients receiving long-course chemoradiotherapy. The median operative duration was 300 minutes. Repair of the perineal defect was performed in four patients with biologic mesh alone, in one patient with omentoplasty only and in four patients with both. One procedure, in a patient with a body mass index of 30.3kg/m2, required conversion owing to patient habitus and intolerance of a head-down position.
Pathologic assessment revealed a median tumour height of 20mm above the dentate line. All resections were R0, there were no perforations and the median CRM clearance was 6.5mm. Macroscopic evaluation revealed the plane of surgery for all specimens to be in the mesorectal plane with no significant breaches to the smooth mesorectal surface. Three patients had a complete pathological response. Post-operatively, the median duration of use of intravenous fluid therapy and patient controlled analgesia was 2 days and the urinary catheter was removed at a median of 2.5 days. Stoma competency was achieved at a median of 6 days post-operatively and the overall length of stay was a median of 7.5 days. There were no deaths.
One patient required a return to theatre for drainage of an infected collection and two patients developed perineal wound dehiscence. Of these two, one had received neoadjuvant short-course treatment and the other had long-course treatment. Neither patient had an omentoplasty or perineal mesh placement. One of them received vacuum-assisted therapy over a seven-week period. The other was readmitted, treated with simple dressings and discharged after ten days. At their follow-up appointment, both patients had fully healed wounds with no evidence of perineal herniation. Three of the six male patients in our series still required long-term catheterisation at the last follow-up visit.
The EORTC QLQ-C30 and QLQ-CR29 questionnaires were completed by all except three patients (owing to insufficient follow-up time) at a median of 194 days after surgery. The one patient who received adjuvant therapy answered the QOL questionnaire during this treatment course. The median global health status score was 90.8 (95% confidence interval: 85.7–96.0). Raw scores were transformed using a linear transformation as described by the EORTC scoring manual. For functional scales, a higher score represents a better level of functioning, while for symptom scales, a higher score represents a greater severity of symptoms.
Comparison to EORTC reference values revealed favourable global health status, function and symptom scores in our cohort (Table 2) while comparison to QOL results following conventional APE in studies identified by Cornish et al16 that used EORTC QLQ-C30 did not reveal QOL deterioration in any field (Table 3). Statistical comparison with such trials was not possible without raw scores or standard deviation yet it has previously been suggested that a difference of >10 points indicates clinical significance.25
None of the patients in this series had any significant chronic perineal pain although three patients reported having ‘a little’ pain in the perineum on the QLQ-CR29 questionnaire. Pooled scores from the EORTC QLQ-CR29 questionnaire revealed high symptom scoring for impotence in males while both men and women had moderate urinary frequency (Table 4). At a median follow up of 186 days, no local or distant recurrences were identified.
The development of total mesorectal excision (TME) has served as a paradigm for improving local recurrence and survival in upper and middle rectal cancer. However, APE is still required for 30% of rectal cancers and remains associated with higher rates of local recurrence and lower survival when compared to AR.26 This has led to the development of the more radical ELAPE and early results suggest improved oncological outcomes.5,7,27 To date, ELAPE with a laparoscopic abdominal component in all cases has not been described and the role of an ERAS programme after this technique or short-term outcomes and QOL results remain unclear.
In the present study, we describe our early experience with lap-ELAPE. Minimally invasive TME surgery has continued to evolve in recent times, with fully laparoscopic, hand-assisted or robotic series.28–30 This has prompted a change in outlook to laparoscopic surgery in the deep pelvis. Outcomes following an exclusively laparoscopic approach to ELAPE have not yet been described. West et al recently presented a multicentre experience of ELAPE and identified a small percentage of laparoscopic cases.27 They noted no difference in IOP between an open or laparoscopic approach. Nevertheless, the technique used was not described and no other results from a laparoscopic approach were presented. In addition, no data on the use of ERAS or assessment of QOL have been put forward to date.
The lack of published reports may, in part, reflect the inherent difficulties of identifying the necessary landmarks in the deep pelvis that dictate when one should stop the abdominal phase of ELAPE. Excessive dissection in the TME ‘holy plane’ during the abdominal approach would separate the levator muscle from the rectum and risk exposing T3/T4 low rectal tumours, thereby resulting in a positive CRM. At our unit, we use a combination of methods to identify the posterior attachments of the levator ani onto the coccyx at laparoscopy. We subsequently use this as the landmark to define the limits of the anterolateral dissections. Using these approaches, we found that the plane of surgery for all specimens was in the mesorectal plane with no significant breaches to the smooth mesorectal surface.
An additional three-stage classification system has been proposed by Nagtegaal et al for the grading of APE specimens at the level of the levators.4 We recently adopted this system into our routine pathological reporting and found that all lap-ELAPE specimens (n=3) examined in this method were classified as extralevator with a CRM lying outside the levator complex (Fig 2).
Patient selection is an important consideration in ELAPE surgery and few reports have described the means by which surgical teams have selected patients for this technique. Similar to the original description of ELAPE, our preference is to select patients with magnetic resonance predicted T3/T4 disease, while conventional APE or an ultra-low anterior resection are favoured for earlier stage disease. Our early findings indicate that lap-ELAPE in these advanced cases gives low rates of CRM involvement and IOP. The duration of operation for lap-ELAPE in our series is high and reflects the challenges of adopting a new technique.
A further question in ELAPE surgery is which method of perineal reconstruction to use. In our study, two patients suffered wound dehiscence and no perineal hernias were reported at the last follow-up appointment. Myocutaneous flaps are one possible solution for reduction of perineal wound related complications, yet they add significantly to the duration of the procedure and all carry the risk of longer bed rest, flap-related complications and donor site morbidity.5,31 Our study adds to the work of other groups on biologic grafts for repair of large perineal wounds32,33 and indicates that good outcomes can be obtained without the addition of a flap procedure.
One important observation in this study was the high rate of urinary retention in male cases and patient-reported impotence. Three of the six male patients required a long-term catheter, while the average impotence scores from QLQ-CR29 was 75/100. West et al reported genitourinary dysfunction of 30% after ELAPE with 46% of these cases suffering urinary dysfunction.27 Fowler et al reported bladder dysfunction of up to 30% following conventional APE.34 A number of other studies also reported a high incidence of sexual dysfunction following both laparoscopic and open APE.35,36 Formal assessment of genitourinary function pre- and post-operatively is required to investigate further the effect of lap-ELAPE in this domain.
The present study is the first to report the use of ERAS following ELAPE surgery. ERAS programmes improve health outcomes, optimise the use of limited healthcare resources, and are an efficient intervention in high volume, major surgical procedures with prolonged hospital stays.37 Our data show that an accelerated recovery can be achieved after lap-ELAPE, with short durations for intravenous fluid therapy, opiate and anti-emetic use, and time to removal of the urinary catheter.
These durations are consistent with published data relating to enhanced recovery outcomes following colorectal resection. Gatt et al reported the duration of intravenous fluid use to be approximately 33 hours and the duration of urinary catheterisation to be 48 hours38 while King et al reported a median of 3 anti-emetic injections per patient.20 The median length of stay in our series was longer than conventional enhanced recovery interventions, principally owing to the lengthy time to stoma proficiency. In a randomised study of intensive pre-operative community-based stoma education against conventional post-operative training, Chaudhri et al demonstrated significantly shorter length of stays in the intervention arm.39 Similar to our results, they noted a median length of stay of 10 days in the control arm. Improved pre-operative stoma training may help further reduce the length of stay of our patients.
QOL is increasingly recognised as an important outcome measure, particularly when comparing a novel treatment to an established technique. While it has previously been thought that the presence of a stoma must lead to a reduced QOL after APE when compared to AR, recent studies and a Cochrane review have suggested that poor functional results after sphincter preserving surgery also compromise QOL.11,40,41 We demonstrate acceptable QOL scores following lap-ELAPE using the extensively-used EORTC QOL instruments. The average global health status score noted in the present analysis was 90.8, which was significantly higher than the published EORTC reference values and higher than five studies investigating QOL following conventional APE identified in a meta-analysis by Cornish et al16 (Table 3).
The conclusions that can be drawn from such comparisons are limited, however. It is acknowledged that both the EORTC reference values and those taken from the work of Cornish et al are not matched to our cohort for neoadjuvant or adjuvant therapy use, tumour stage or questionnaire timing. Despite these limitations, the absence of any reduction in QOL following lap-ELAPE using either approach suggests that this more radical procedure is not associated with a significant reduction in QOL. Prospective and comparative assessment of QOL following lap-ELAPE, laparoscopic APE and open APE would be required to confirm this.
Despite the small sample size and non-comparative nature of the present study, this paper nevertheless adds to the evolving body of information on ELAPE and illustrates the feasibility of a minimally invasive approach to the abdominal phase of this procedure. Our results also encourage adoption of an ERAS programme in ELAPE and support the notion that post-operative QOL in these radically treated patients is acceptable. Future studies with longer follow-up periods and comparative analyses with conventional APE are necessary to allow the role of laparoscopic and open ELAPE to be defined more precisely in low rectal cancer and to help refine patient selection for neoadjuvant treatments and surgery as well as selection of the optimal method of perineal reconstruction.
AHM is supported by grant funding from Wessex Medical Research and Cancer Research UK (C28503/A10013).