A multicentre prospective randomised controlled trial of preoperative endoluminal stenting versus conventional surgery in large bowel obstruction.

Co-ordinating Centre:

Darent Valley Hospital

Darenth Wood Road
Dartford
Kent
DA2 8DA

Main Investigator:

Mr. M.C. Parker

Department of Surgery
Darent Valley Hospital
Tel: 01322 428100
Fax: 01322 428635

Aim of the Trial

To evaluate the clinical outcome and to analyse the cost-benefit of pre-operative endoluminal stenting of obstructing left sided colo-rectal lesions compared with urgent surgical decompression.

Background

The morbidity and mortality rates after emergency surgery on obstructing colo-rectal lesions are high, the latter has been quoted as up to 40% ( Cuschieri, 1997) although most papers report values of around 30 % (Anderson and Hall, 1992). This is partially attributed to proximal faecal loading causing technical difficulties leading to anastomotic dehiscence (Irnin and Goligher, 1973). Pre-operative preparation of the whole gut has made it possible for the loaded but unobstructed colon to be mechanically prepared (Hewitt et al, 1973). However this method is not applicable in obstructed cases where a defunctioning colostomy or Hartmann's procedure would be the operation of choice (Bakker et al, 1982). On-table antegrade irrigation of the loaded colon, as described by Dudley and Radcliffe in 1980, allows primary resection and anastomosis of the obstructing lesion. This procedure is more technically difficult and increases the operation time significantly in these often-frail patients.

Endoluminal stenting of obstructing colo-rectal lesions has been performed as a palliative treatment of surgically inoperable and unfit patients (Turegano-Fuentes et al, 1998). Stenting has been employed as a temporary pre-operative measure (Saida et al, 1996), as well as a permanent curative procedure for benign strictures. The papers so far report low complication and failure rates. A pilot study at Joyce Green Hospital confirmed these reports and showed a potential clinical and economic advantage with stenting / colonic resection when compared with an emergency colectomy, (Boorman et al, 1999).

Hypothesis:

Preoperative stenting of obstructing left-sided colo-rectal lesions allows proper preparation and staging of the patient. Hence, the surgical outcome could be significantly improved by converting an emergency into an elective procedure.

Inclusion criteria

All patients with obstructing left sided (up to and including the splenic flexure) colo-rectal lesions (proven radiologically) will be included in this trial and will be taken on with the intention to treat. Previous bowel conditions will not exclude patients from the trial.

Exclusion criteria

1. Patients with signs of peritonitis and/or perforation
2. Patients with "closed loop obstruction" and right iliac fossa tenderness
3. Patients with lesions less than 6 cms from the anal verge
4. Surgically unfit patients
5. Patients with advanced disseminated disease
6. Patients refusing entry into the trial
7. Patients refusing surgery
8. Demented patients (unable to understand explanation of trial)
9. Pregnant patients

Patients and methods:

Patients admitted with obstructing left-sided colo-rectal lesions would be randomised to either pre-operative endoscopic stenting of the obstruction (Group A) or surgical decompression by a technique according to the operating surgeons preference and medical condition of the patient (Group B).

We will need a minimum of 140 patients in order to obtain statistically significant results. This figure has been calculated from an expected reduction in mortality rate, from 27.5% (in patients undergoing emergency surgery) to 7.5% (in the stented / elective surgery group). This gives the trial a power of 90% with a 10% value (false positive rate). Dr.T.Crayford, from The Department of Public Health at Kings' College Hospital, London, has provided the statistical advice.

Randomisation will be performed on admission if the patient agrees to enter the trial. Patients will be fully informed of the details and aim of the trial, as well as of the particular procedure to which they have been randomised. Block randomisation will be used to ensure a relatively normal distribution of patients in each group. The study shall extend over a period of two years with a minimum follow-up period of six months.

On admission, routine laboratory investigations will be performed (a full blood count, urea and electrolyte estimation and liver function tests). A plain abdominal and erect chest X-ray will be taken to exclude established perforation. In both groups the diagnosis will be confirmed by a water-soluble contrast enema.

After resuscitation, patients in Group A will have a "Bard memotherm" colorectal stent (Angiomed GmbH & Co.Medizintechnik KG, Wachhausstrasse 6, D-76227 Karlsruhe) deployed across the obstructing tumour by a fluoroscopic +/- endoscopic technique. The patency and position of the stent will be checked radiologically by means of a contrast enema. In the event of proximal faecal loading a flatus tube will be passed (under radiological guidance) through the deployed stent to allow irrigation of the proximal colon until the bowel is empty (confirmed by a plain abdominal X-ray). These patients will be given 10mls of lactulose twice daily for one week.

Patients may be discharged post-procedure, as long as normal bowel function has returned, following which proper staging and assessment for further surgery will be performed. Improvement of any medical problems will also be undertaken at this time. Routine laboratory investigations, radiological imaging of the liver and according to the clinical condition of the patient a CT scan, will be performed, after which patients will undergo delayed definitive surgery. Prior to operation patients will have standard bowel preparation (as for non-obstructed cases) and an appropriate elective colonic resection will be performed.

In those patients proven to suffer from advanced local disease, metastatic disease or who are unfit for further major surgery, the stenting procedure will be considered to have been a palliative measure and no further surgical intervention will be necessary.

Patients from Group A, in whom stenting fails will undergo an appropriate emergency surgical decompression along similar lines to those in Group B.

Following similar resuscitation, patients in Group B will have emergency surgery to perform bowel decompression as dictated by the surgeons' preference and the clinical condition of the patient.

A standardised antibiotic prophylactic regimen using intravenous antibiotics will be administered to both groups according to the surgeons' preference.

Results:

Analysis of the short and long-term clinical outcome in both groups, as well as, a cost benefit analysis, will be performed. All complications and costs will be recorded for later assessment. The cost of the various procedures involved in the study will be calculated according to National Health Service rates as obtained from the financial departments.

Measurable parameters

All patients

1. Patient demographic data
2. Site of obstruction
3. Duration of symptoms of absolute constipation
4. Co-morbidity/ concomitant diseases affecting outcome
5. Haematological and biochemical results

1. Site, length of stricture (in cms), distance from anal verge (in cms)
2. Duration of procedure and radiation dose received
3. Number of stents per patient
4. Complications (morbidity/mortality)
5. Length of hospital stay
6. Category i.e. Palliative or preoperative (potentially curative)
7. Staging of tumor

1. Name of surgical procedure
2. Length of operation
3. Intra-operative staging
4. Curative/palliative procedure
5. Length of hospital stay
6. Intensive care admission (reason and duration)
7. Complications
8. 30 day mortality
9. pathological classification

1. survival/symptom free duration
2. stent related problems
3. recurrent obstruction and it's management
4. restenting

1. name of procedure
2. intra-operative staging
3. curative/palliative
4. length of operation
5. length of hospital stay
6. intensive care admission (reason/duration)
7. complications
8. pathological classification
9. permanent stoma rate
10. 30 day mortality
11. survival

Costs:

The cost of a single stent is £795. Other points of expenditure include secretarial support, questionnaires sent to both General Practitioners and patients, presentations and computer programs etc. Bard Ltd. has kindly agreed to provide help with the financial expenditure.

References:

1. McIntyre R., Reinbach D., Cuschieri A. Emergency abdominal surgery in the elderly. J.R.Coll.Surg.Edinb. 1997; 42: 173-8.
2. Anderson J.H., Hole D. Elective surgery versus emergency surgery for patients with colorectal cancer. B.J.S. 1992; 7: 706-9.
3. Irvin T.T., Goligher J.C. Aetiology of disruption of intestinal anastomoses. B.J.S. 1973; 60: 461-4.
4. Hewitt J., Rigby J., Reeve J., Cox A.G. Whole gut irrigation in preparation for large bowel surgery. Lancet 1973; 2: 337.
5. Bakker F.C., Hoitsma H.F.W. Hartmann procedure. B.J.S. 1982; 69: 580-2.
6. Dudley H.A.F., Radcliffe A.G., McGeehan D. Intraoperative irrigation of the colon to permit primary anastomosis. B.J.S. 1980; 67:80-1.
7. Turegano-Fuentes F. et al. Transanal self-expanding metal stents as an alternative to palliative colostomy in selected patients with malignant obstruction of the left colon. B.J.S. 1998; 85: 232-5.
8. Saida Y., Sumiyama Y., Nagao J., Takase M. Stent endoprosthesis for obstructing colorectal cancers. Dis. Colon Rectum 1996; 39:552-5.
9. Boorman P., Soonawlla Z., Sathananthan N., MacFarlane P., Parker M. Endoluminal stenting of obstructed colorectal tumours. Ann.Roy.Coll.Surg.Eng. 1999