14,000 new cases of rectal cancer are diagnosed each year, frail and elderly patients represent a rising proportion of these patients. Whilst the gold standard is often to remove the tumour and restore bowel continuity, surgeons will often avoid this procedure in this group of patients as they unfortunately tolerate surgical complications very poorly. Such surgical complications may present with life threatening sepsis, can prolong hospital stay, delay further cancer therapy and in the elderly or frail patient often leads to loss of independence and quality of life. In this setting, there are two alternative procedures (Hartmann's procedure OR intersphincteric APE) that may be used and these are employed in roughly equal measure in the UK (nationwide survey, Dec 2013, unpublished data). It is anecdotally felt that Hartmann's procedure (HP) has a greater risk of surgical complications (30%) and a few small retrospective studies have shown this (1-3), however there are no prospective data to support this view. Whilst some surgeons do choose intersphincteric APE (IAPE) on the basis of a lower surgical complication rate, many do not due to perceived limitations in the technique (longer operating time, risk of tumour perforation), which are unproven. We feel that a larger, prospective dataset is required to demonstrate the superiority of IAPE over HP and convince the remaining surgeons to change procedure. We have explored the possibility of a full randomised trial to answer this question, however this is not feasible due to the difficulty of randomisation of patients. Very little data are available regarding the use of IAPE in the setting of rectal cancer, however many surgeons who do employ the technique, specifically adapt their technique in this setting to reduce the chances of tumour perforation (two stage, stapling off rectum before removing anal canal separately). It is possible that those surgeons who prefer HP have not considered this, and combined with the lack of prospective data are reluctant to change technique. We are confident that if we can demonstrate a significant difference in surgical complication rate and promote a modification to the IAPE surgical technique then we can significantly reduce surgical harm to these frail patients.

Having established a network of units (UK and Europe) who are also keen to answer this question we have powered a prospective observational cohort study to demonstrate a significant reduction in surgical complication rate (30% to 15%). In conjunction with an on going Swedish randomised trial we hope that the data from this study will provide compelling evidence for UK surgeons to change practice. If the study reveals only a modest difference in complications, we will use the data gathered to design and reapply again for full funding for an RCT with the additional advantage of being able to clearly demonstrate a network of units capable of recruiting the necessary number of patients. The overarching aims of this study are therefore to:

1. Determine the difference in surgical complication rates between HP and IAPE
2. Assess the effect of IAPE technique on intra operative tumour perforation rate

The objectives of the study are to:

1. Determine surgical complication rate for each procedure (graded by Clavien-Dindo)
2. Assess impact on secondary outcomes (length of stay, readmission, reintervention, medical complications, time to chemotherapy, quality of life)
3. Determine if the complication rate in IAPE is dependant on surgical technique
4. Determine patient and clinician acceptability to randomisation if required

Data Collection

Data will be collected on specific CRF at baseline and 30 post operative days, all data will be completely anonymised and no further data will be collected past 30 days. The easy to use and well validated Clavien-Dindo scale will be use to grade surgical complications, but we will also calculate the Comprehensive Complication Index which integrates all medical and surgical complications and is felt to represent a more accurate estimation of post operative complications (4). Data will be collected at the following time points:

BASELINE INTRAOPERATIVE POST OPERATIVE (30 days) Age Surgical approach (lap, open) Pathological stage Sex Anaesthetic type TME quality BMI Operative time Length of stay Comorbidities Intra operative perforation Readmission ASA score Length of anorectal stump (HP) Surgical complications (Clavien-Dindo scale) Radiological stage Antibiotics Medical complications Distance of tumour from anal verge Method of IAPE (one vs two stage) Comprehensive Classification Index score Preoperative therapy 30 day quality of life Quality of life Patient willingness to randomise Reason for avoiding primary anastomosis Surgeon willingness to randomise Surgeon's reason for op choice

Data Analysis As this is an observation analysis, reporting only the unadjusted difference between HP and IAPE may result in a biased comparison. The primary analysis will therefore be a multivariable analysis adjusting for any confounding factors. To ensure the analysis is in keeping with the sample size calculation, evaluations of the primary outcome will be based on a 90% confidence interval.

Inclusion Criteria:

• • Patients aged over 18 years

  • Able to provide informed consent
  • Undergoing elective, locally curative surgery for rectal cancer
  • Recurrent rectal cancer not a contraindication if pre operative imaging suggests that the tumour can be removed with clear margins
  • Primary anastomosis not appropriate for reasons of frailty, poor function or risks of anastomotic leak
  • Local staging completed by MRI
  • Histological confirmation of adenocarcinoma
  • Fit for major resection

Exclusion Criteria:

• • Pregnant patients

  • Patients unable to consent
  • Local palliative resection (systemic metastatic disease not a contraindication)
  • Suspicion of tumour perforation
  • Rectal tumours requiring a formal APE due to distal tumour involvement of anorectal junction or pelvic floor

Frail patients with rectal cancer