4721 RCSI Colorectal Guide covsurvival in patients with colorectal cancer. In 1995 there was some 17,000 deaths due to colorectal cancer in the U.K. The overall survival quoted in

Colorectal CancerManagementClinical Guidelines

Prepared by The Clinical Guidelines CommitteeRoyal College of Surgeons in Ireland

November 2002

Royal College of Surgeons in Ireland123 St. Stephen’s Green, Dublin 2, Ireland

Tel: 353-1 402 2100. Fax: 353-1 402 2460. Web: www.rcsi.ie

FOREWARDThe Clinical Guidelines Committee is very pleased tobe in a position to publish guidelines in themanagement of colorectal cancer. This form ofmalignant disease is common in Ireland and is of greatimportance to all general surgeons in the country. It isstill usual for the disease to present at an advancedstage, either as an emergency with intestinalobstruction or with metastatic disease. Advances andimprovements in treatment are recognised and havemade a significant impact to prognosis, particularly inpatients with rectal cancer. All surgeons treatingpatients with colorectal cancer are aware of the rapidrate of change in the management of patients withrectal cancer. Newer methods of staging have led tochanges in the sequence of treatment in many instancesso that primary radiotherapy followed by meticulousradical surgery is now appropriate for many patients.The role, the scheduling and the composition ofadjuvant chemotherapy all remain under continualevaluation. The value of adjuvant chemotherapy innode-negative colon cancer remains uncertain.Continuing challenges for the surgeon include themanagement of patients with unresectable andmetastatic disease as well as those with recurrence afterresection. The increasing value of multidisciplinarymeetings for patients with cancer will require thesurgeon to retain a central role in the management ofall patients with malignant disease.

It should be emphasised that guidelines are not strictprotocols and that they always leave scope for theindividual judgement of the experienced surgicalclinician. It is hoped that these guidelines will bereviewed and renewed periodically and that they willbe found to be useful to the surgical community inplanning investigation and treatment of their patients.

Although drawing extensively from existing publishedguidelines, the development of this booklet has requireda great deal of effort and consultation in its preparation.Mr. Peter Gillen has consulted extensively withcolleagues and with relevant societies, has co-ordinatedall the available documentation and has assembled it ina coherent fashion. The Clinical Guidelines Committeeis most grateful to him for his efforts.

We are also most grateful to the Irish Society ofGastro-enterology, the Association of Coloproctologyof Great Britain and Ireland, the ScottishIntercollegiate Guidelines Network, the Irish Society ofColoproctology and many individual surgeons fromthese organisations whose advice and suggestions havebeen incorporated into this booklet.

The assistance of Ms. Paula Wilson and heradministrative colleagues and that of Ms. BeatriceDoran, Librarian, is acknowledged gratefully and thesupport and encouragement of the President andCouncil has also been genuinely appreciated by theClinical Guidelines Committee.

Niall O’HigginsChairman, Clinical Guidelines Committee

November, 2002.

Colorectal Cancer Management Clinical Guidelines 1

Foreward

Colorectal Cancer Management Clinical Guidelines2

Colorectal Cancer Management Clinical Guidelines 3IntroductionContents

Foreward 1

Introduction 4

Aims/Purpose 4

Validity 4

Summary of Guidelines 5

Investigations 10

Methods of Investigation 12

Pre-Operative Assessment 14

Preparation for Surgery 14

Surgical Technique 17

Emergency Surgery 20

Adjuvant Chemotherapy 21

Treatment of Advanced Disease 24

Outcome 28

Follow-Up 29

Nursing Care 31

Histopathology Reporting 32

References 39

Introduction


Colorectal cancer constitutes the single largestdiagnostic group of noncutaneous cancers in Ireland.The number of colorectal cancers exceeds the totalnumber of breast and lung cancers and deaths fromcolorectal cancer were second only in number to thosefrom lung cancer. Colorectal cancer represents about9% of all noncutaneous cancers diagnosed and 13% ofcancer related deaths are due to colorectal cancer(National Cancer Registry in Ireland, 1997).

The incidence of colorectal cancer is relatively high inIreland which has the 5th highest incidence amongstwomen and the third highest incidence amongst men inEuropean countries. In Ireland the cumulativeprobability of survival following a diagnosis ofcolorectal cancer is 71% at one year, 59% at two yearsand 50% at three years (National Cancer Registry).Colorectal cancer is frequently advanced at the time offirst presentation and metastatic disease is present inup to 20% of patients at time of diagnosis. Heightenedawareness of colorectal cancer among the generalpublic and health care professionals may lead to earlierpresentation, diagnosis and treatment with improvedlong term survival.

PURPOSEThe purpose of these guidelines is to assist cliniciansand other health care professionals in decision-makingand practice by removing uncertainty in areas where itis possible to do so. The guidelines are to serve as abenchmark of good clinical care and to be prescriptiveof unacceptable clinical standards. They are nothowever intended to create a rigid framework insituations where there is reasonable difference ofopinion. Thus clinical freedom, within limits definedby good practice, is preserved.

It is hoped that the establishment of nationalguidelines for the treatment of colorectal cancer willlead to introduction of local protocols, better clinicalcare leading to improved survival of patients withcolorectal cancer.

AIMS■ To improve awareness of colorectal cancer

in the community.

■ To facilitate early diagnosis and treatment.

■ To improve referral patterns of patients withproven colorectal cancer.

■ To improve quality of life for patients withcolorectal cancer.

■ To improve disease-free intervals and overallsurvival in patients with colorectal cancer.

In 1995 there was some 17,000 deaths due to colorectalcancer in the U.K. The overall survival quoted in theliterature is currently running at five year survivals ofless than 40%. The high incidence of the disease andthe fact that improvement in mortality has been modesthighlights the need for better prevention, diagnosis andtreatment. Advanced disease at first presentation is stillcommon and has been quoted at over 20%. It is hopedthat this percentage will reduce with heightenedawareness of the disease among the general public.

VALIDITYDefinition of types of evidence taken from the USAgency for Health Care Policy & Research and are setout in the following tables.

Level Type of Evidence

Ia. Evidence obtained from meta-analysis ofrandomised controlled trials.

Ib. Evidence obtained from at least onerandomised controlled trial.

IIa. Evidence obtained from at least one well-designed controlled study without randomisation.

IIb. Evidence obtained from at least one other type of well-designed quasi-experimental study.

III. Evidence obtained from well-designednon-experimental descriptive studies, suchas comparative studies, correlation studiesand case studies.

IV. Evidence obtained from expert committeereports or opinions and/or clinicalexperience of respected authorities.

Colorectal Cancer Management Clinical Guidelines 5Introduction

SUMMARY OF GUIDELINES

Investigation

i. It is recommended that patients with higher-risksymptoms should be fast- tracked either in specialclinics or with urgent appointments to routineclinics. Patients referred through such clinicsshould be investigated with either flexible or rigidsigmoidoscopy plus a high quality double contrastbarium enema or colonoscopy, when appropriate. Grade B

ii. Pre-operative histology should be obtained fromall rectal tumours. Grade C

iii. Doctors carrying out colonoscopy should audittheir results and expect to achieve a high rateof complete colonoscopy with a lowperforation rate. Grade B

iv. It is acceptable for non-consultant staff to performdouble contrast barium enemas, provided theyhave completed a recognised training programme,the examinations are performed to strict protocolsand supervised by a consultant radiologist. Grade C

v. All patients, particularly those with rectal cancer,should have pre-operative staging to determine thelocal extent of the disease and the presence of lungand liver metastases. Endoanal rectal ultrasoundscan should be performed to identify T1 rectalcancers and CT or preferably MRI scan performed

to assess involvement of adjacent organs in moreadvanced tumours. Grade C

vi. Surveillance and genetic testing should be offeredto all families with Familial AdenomatonsPolyposis (FAP) or families that either meet withAmsterdam criteria for Hereditary Non-PolyposisColorectal Cancer (HNPCC) or have a confirmedmismatch repair gene mutation. Grade A

vii. First-degree relatives of patients who developcolorectal cancer before the age of 45 years andmembers of families in which multiple cancershave occurred should be seen by a specialist,preferably with experience in genetic counselling,who can evaluate their risk of developing thedisease and advise on appropriate investigationsand surveillance. Grade B

Access to Treatment

i. Patients should expect to receive initial treatmentwithin 4 weeks between making a diagnosis ofcolorectal cancer and start of therapy. Grade B

ii. Colorectal cancer should be treated by surgeonswith appropriate training and experience and whowork as part of a multidisciplinary team.

iii. All patients with colorectal cancer should have thebenefit of a suitably informed surgical opinion andtheir management should be considered by themultidisciplinary team.

A Requires – at least one randomised controlled trial as part of the body of literatureof overall good quality and consistency addressing specific recommendation.

B Requires – availability of well conducted clinical studies but no randomised clinicaltrials on the topic of recommendation.

C Requires – evidence obtained from expert committee reports or opinions and/orclinical experiences of respected authorities.Indicates absence of directly applicable clinical studies of good quality.

GRADE RECOMMENDATION

(Evidence Levels Ia, Ib)

(Evidence Levels IIa, IIb, III)

(Evidence Level IV)

Where numerals or capital letters appear in the text of these guidelines, they refer to levels of evidence and grades ofrecommendations as outlined above.


Preparation for Surgery

i. All patients undergoing surgery for colorectalcancer should give informed consent. Informedconsent implies being given information about thelikely benefits and risks of the proposed treatmentand details of any alternatives. Informed consentshould be obtained by the operating surgeon or byanother senior doctor. Grade C

ii. The patient who may require a stoma should beseen by a stoma therapist prior to surgery and thereferral should be made at the earliest opportunityto allow adequate time for preparation. Grade C

iii. Blood should not be withheld if there is a clinicalindication for blood transfusion. Preparation forblood transfusion should be made for all patientsundergoing surgery for colorectal cancer exceptwhere an individual patient refuses.

iv. Mechanical bowel preparation prior to surgery isrecommended. Grade C

v. Subcutaneous heparin and/or intermittent calfcompression should be employed asthromboembolism prophylaxis in surgery forcolorectal cancer unless there is a specificcontraindication. Grade A

vi. Patients undergoing surgery for colorectal cancershould have antibiotic prophylaxis. Precise regimensvary but a single dose of appropriate intravenousantibiotic appears to be effective.Grade A

Elective Surgical Treatment

i. It is recommended that the term curative resectionshould be based on histological confirmation ofcomplete excision or residual tumour. Surgeonsshould expect to achieve an overall curativeresection rate of 60%, but it is appreciated thatthis will depend at least in part on the stage atwhich patients present. Grade B

ii. Any cancer in which the distal margin is seen at15cm or less from the anal verge using a rigidsigmoidoscope should be classified as rectal. Grade C

iii. It is recommended that total mesorectal excisionshould be performed for cancer in the lower two-thirds of the rectum, either as part of a lowanterior resection or an abdomino-perinealresection (APER). In tumours of the upper rectumthe mesorectum should be divided no less than5cm below the lower margin of the tumour. Careshould be taken to preserve the pelvic autonomicnerves and plexuses, and perforation of thetumour during operation should be avoided.Grade B

iv. Although no definite recommendations can bemade regarding anastomotic technique, theinterrupted serosubmucosal method has thelowest reported leak rate and stapling facilitatesultra-low pelvic anastomoses. After low anteriorresection and total mesorectal excision thejudicious use of a temporary defunctioning stomais recommended, and the formation of a colonicpouch may be considered.

v. Cytocidal washout of the rectal stump should beundertaken prior to anastomosis. Grade C

vi. The proportion of rectal cancers treated byabdomino-perineal excision of the rectum (APER)should be less than 40%, and, if distal clearanceof 1cm can be achieved, a low rectal cancer maybe suitable for anterior resection. If a surgeon hasany doubt regarding the choice between these twooperations, an experienced second opinion shouldbe sought. Grade B

vii. Local excision for cure in rectal cancer should berestricted to pT1 cancers with well or moderatedifferentiation less than 3cm in diameter. It mustbe accepted that subsequent histopathologicalexamination of cancers thought to be suitable forlocal excision will identify a small proportionwhich require more radical surgery. Grade B

viii. Laparoscopic surgery for colorectal cancer shouldonly be performed by experienced laparoscopicsurgeons who have been properly trained incolorectal surgery and who are entering theirpatients into multi-centre trials. Grade B

Introduction


Record Keeping

i. It is recommended that detailed clinical notes andoperative findings and procedures are kept for allcolorectal cancer patients.

ii. Sound record-keeping facilitates accurate audit ina colorectal unit.

iii. Accurate record keeping promotes good qualityclinical research

Emergency Treatment

i. Emergency surgery should be carried out duringdaytime hours as far as possible, by experiencedsurgeons and anaesthetists. Grade C

ii. For patients presenting with obstruction, stepsshould be taken to exclude pseudo-obstructionbefore operation. Grade B

iii. Stoma formation should be carried out in thepatient’s interests only, and not as a result of lackof experienced surgical staff. Grade B

Adjuvant Therapy

i. Patients with Dukes’ C colon cancer should beconsidered for adjuvant chemotherapy. Grade A

ii. Patients with Dukes’ B colon cancer should beconsidered for entry into randomised trials ofadjuvant chemotherapy.

iii. Patients with high risk Dukes’ B colon cancershould be individually counselled about their levelof risk and possible benefits of chemotherapy.

iv. There is no evidence to support the use ofadjuvant chemotherapy in Dukes’ A cancers ofcolon or rectum.

v. No definite recommendation can be maderegarding adjuvant chemotherapy for patients with Dukes’ C rectal cancer. Patients may be either offered chemotherapy or be considered for clinical trials, in addition to appropriateadjuvant radiotherapy. Grade B

vi. Systemic chemotherapy should only beadministered by clinical staff with appropriatetraining and experience, according to IrishOncology guidelines. Grade C

vii. Patients with a T3 rectal cancer should beconsidered for entry into clinical trials of pre-operative radiotherapy. Grade C

viii. Patients with rectal cancer in whom the tumour istethered or in whom local imaging indicates a highrisk of incomplete resection should be selected forlong course pre-operative radiotherapy to obtaintumour downstaging.Grade B

ix. In patients with rectal cancer pre-operativeradiotherapy using short course (25 Gray (Gy) in 5fractions in one week) or longer course (40-45 Gy in20-25 fractions over 4-5 weeks) are both acceptable. Grade A

x. In patients with rectal cancer who have not hadpre-operative radiotherapy, post-operativeradiotherapy and chemotherapy should be selectedwhere there are well-established predictors ofhigh-risk (e.g. evidence of tumour at thecircumferential resection margins). Grade A

xi. In all patients with rectal cancer post-operativeradiotherapy doses should be 40-50 Gy in 20-25fractions or a suitable biological equivalent using aplanned volume. Grade B

xii. A planned radiotherapy volume using three orfour fields is recommended for rectal cancers asthis results in less morbidity and mortality. Grade B

xiii. Patients with potentially operable rectal cancershould always be considered for entry into trials of adjuvant radiotherapy. Grade B

Introduction


Treatment of Advanced Disease

i. For patients with inoperable rectal carcinomawithout evidence of metastatic disease, primaryradiotherapy alone or in combination withchemotherapy should be considered. Grade B

ii. Patients with metastatic disease who are fit foractive therapy should be accurately staged withCT scans of abdomen and thorax.

iii. Patients with evidence of unresectable metastaticdisease should be considered for palliativechemotherapy as soon as the diagnosis ofmetastatic disease is made, but this may not beappropriate for all patients.Grade A

iv. Chemotherapy for metastatic colorectal cancershould only be given after discussion at aMultidisciplinary Team meeting. Grade C

v. Entry into clinical trials evaluating the benefits ofnovel therapeutic regimens in colorectal cancershould be encouraged. All trials involving surgeryeither at diagnostic or therapeutic stage shouldhave a surgeon as a principal investigator.

vi. Palliative treatment should be 5-fluorouracil (5FU)given by infusion combined with the use ofirinotecan in the first line or on 5FU failure if thepatient remains fit for chemotherapy. Grade A

vii. Hepatic arterial infusional chemotherapy remainsof unproven benefit. Grade A

viii. Patients with metastatic disease limited to the liverwhich is potentially resectable should beconsidered for partial hepatectomy by anexperienced liver surgeon. Grade B

ix. Surgeons and medical and radiation oncologistswho deal with colorectal cancer should make it apriority to build close links with palliative carespecialists and units. Grade B

x. All clinicians who deal with colorectal cancer shouldhave good communication skills and experience inthe control of pain and other cancer symptoms. Grade C

xi. Patients with colorectal cancer should be offered the opportunity to ask questions and tohave important information repeated. Provision of information should be an essential part of every consultation. Grade C

OutcomesMeasurement of outcomes is an essential part ofcolorectal cancer care. In order to undertakemeasurement of outcomes manpower resources andinformation technology facilities are required. These facilities are currently lacking in many hospitals in Ireland.

Hospitals treating colorectal cancer should carefullyaudit the outcome of treatment and achieve:

i. An operative mortality of less than 25% foremergency surgery and less than 7% for electivesurgery with colorectal cancer. Grade B

ii. Wound infection rates after elective surgery for colorectal cancer in the region of 10% or even lower. Grade A

iii. A clinical anastomotic leak rate of around 8% foranterior resections and around 4% for other typesof resection. However ultra-low pelvicanastomoses will have higher leak rates (around15%) and therefore the judicious use of adefunctioning stoma is recommended. Grade B

iv. Local recurrence rates after curative resection forrectal cancers in the region of 10% or even lowerwithin 2 years of operation. Grade B

Intensive care and high dependency care are essentialparts of peri-operative colorectal cancer care andshould be available in hospitals undertaking colorectalcancer surgery.

Introduction


Follow-Up

i. There is no evidence that intensive follow up forthe detection of recurrent disease improvessurvival. However, liver imaging may be offered toasymptomatic patients during the first two post-operative years for the purpose of detectingoperable liver metastases. Grade B

ii. There is no evidence that colonoscopic follow-upimproves survival, but it has been shown to yieldadenomatous polyps and metachronous cancers. Ifsuch a policy is pursued, it is recommended that a“clean” colon should be examined by colonoscopyat 3-5 year intervals. Grade B

iii. Audit and follow-up are necessary to determinepost-operative mortality, anastomotic leak rates,colostomy rates and 5-year survival. This shouldbe regarded as routine. Grade C

iv. All patients with a stoma should have ready accessto specialist nursing staff. Grade C

Pathology

i. All resected polyps and cancers should besubmitted for histopathological examination.Grade B

ii. Pathology reports (See Appendix 2) should containinformation on all of the data items contained inthe Joint National Guidelines Minimum Data Setfor Colorectal Cancer Histopathology Reports.Grade C

iii. Pathology laboratories should store stainedhistology slides for a minimum of 20 years, andtissue blocks from specimens indefinitely, in order to facilitate future case review, clinical audit, and research.Grade B

iv. Pathological examination of colorectal cancerspecimens should be carried out in laboratorieswhich attain high technical standards and whichparticipate in external quality assessment schemes and regular audit of technical procedures and diagnosis. Grade B

Introduction

Detailed Guidelines


INVESTIGATIONS

Screening in high risk groupsA small percentage (about 10%) of patients have agenetic predisposition for colorectal cancer.Approximately one tenth of these (1%) are ‘inheritedcancers’, as occurs in familial adenomatous polyposis(FAP). The remainder are ‘familial cancers’ asexemplified by hereditary non polyposis colorectalcancer (HNPCC). HNPCC is an autosomal dominantcondition characterised by:

i. at least three relatives with a confirmed colorectalcancer (one of whom should be a first degreerelative of the other two);

ii. at least two consecutive generations should beaffected; and

iii. colorectal cancer should be diagnosed under theage of 50 years of age in one affected member.

Various clinical subtypes exist, with or withoutextracolonic tumours (e.g. endometrium, skin).

Until genetic screening is established, colonoscopy willform the mainstay of screening of patients and theirrelatives and should probably be carried out every 2-3years. Screening should commence in the 3rd decadefor those at risk of HNPCC and in the 2nd decade forthose at risk of FAP.

In the majority of patients presenting with colorectalcancer who have a family history of the disease, and inmany families presenting to genetic clinics, the degreeof risk to relatives and the nature of the underlyinggenetic defect is unclear. There is little evidence as tothe value of screening relatives of affected individualsin these circumstances.

Family screening should be offered to first degreerelatives when FAP or HNPCC is diagnosed. Totalcolonoscopy, or double contrast barium enema andsigmoidoscopy, should be offered and started in thesecond decade when FAP is diagnosed, and in the thirddecade when HNPCC is diagnosed.

Colonoscopy should be performed every 2-3 yearstaking into account the individual’s general physicalcondition and compliance. For those at risk of FAPmore frequent (every 1-3 years) sigmoidoscopicscreening should also be considered. Grade C, level IV

Patients with FAP should be offered total colectomywith ileorectal anastomosis or proctocolectomy withrestorative ileoanal pouch reconstruction once colonicpolyps have developed.

Subsequent management should include lifelongsurveillance of any residual rectal stump and regularupper gastrointestinal tract endoscopy to detectadenomata and early malignant change. Grade B, level III

Primary CareIt is estimated that a general practitioner will see 3 new gastro-intestinal cancers per 10,000consultations. There are usually significant delays (19to 29 weeks) from the onset of symptoms to surgicaltreatment. Almost one-third of patients with colorectalcancer undergoing surgery are admitted as emergenciesand approximately 45% are over 75 years of age.

Patients with significant and persistent (4 weeks) newsymptoms (e.g. change in bowel habit, rectal bleeding,passage of mucus, abdominal pain), particularly inmiddle-aged and older patients, and/or those with anysignificant clinical abnormality (e.g. abdominal mass,tenderness, rectal lesion) need an urgent referral forassessment. Similarly, patients with clinical features of anemia and haematological evidence of irondeficiency anemia, in the absence of any overt cause,should be referred to exclude occult colonic cancer. For the majority of patients referral should be directlyto a specialist with an interest in colorectal cancer.

All patients attending a general practitioner with new, significant and persistent (4 weeks or more)colorectal symptoms should have a careful history(including family history) taken and undergo a physical, including rectal, examination.

If any significant clinical abnormality is found, urgentreferral should be made to a specialist with an interestin colorectal cancer for investigation within 2-4 weeks.

If the patient is under 45 years age but in a high risk group, whether or not there is any significant clinical abnormality, urgent (2-4 weeks) referral should be made.

If the patient is over 45 years of age with colorectalsymptoms that are not explained by the generalpractitioner assessment, urgent referral should be made.

If the patient is less than 45 years and does not belongto a high risk group, there is no relevant past family ormedical history and no clinical abnormality, reviewwithin a period of four weeks should be carried out asappropriate, depending on the severity of thesymptoms, their persistence or recurrence. If symptomspersist at review, referral should be made. Grade C, level IV

Detailed Guidelines


High risk groups and low risk groups are defined as follows.

High risk criteriaOnly patients with new and persistent symptoms listedbelow should be referred to the fast-track system.These criteria should include 80-90% of all colorectalcancers presenting to outpatient clinics.

Low Risk CriteriaThe risk of cancer is never zero even in patients withno symptoms as shown by screening studies. Somecancers will be diagnosed incidentally in patients beinginvestigated for symptoms of benign disease. Otherswill have symptomatic cancers, which develop inpatients already symptomatic from functional boweldisease or haemorrhoids. This means that all low-riskpatients with persistent symptoms not responding totreatment or recurring after stopping treatment, shouldbe referred to routine clinics.

Criteria indicating patients at very low risk of cancer are:

■ Rectal bleeding with anal symptoms

■ Rectal bleeding with an obvious external visiblecause such as prolapsed piles

It is recommended that in order to reduce patientanxiety surgeons should expect to achieve averagewaiting times of four weeks or less between makinga diagnosis of colorectal cancer and treatment. Thisrequires adequate access to appropriate facilities atthe referral centre.

It is axiomatic that doctors who decide ontherapeutic strategy and implement that strategyshould have sufficient experience in themanagement of colorectal cancer.

Thus, it is not yet possible to specify a minimumnumber of operations which should be performedeach year. It is recommended that colorectal cancershould be treated by surgeons with appropriatetraining and experience.

Rectal bleeding with a change in bowel habit to increased frequencyof defaecation and/or looser stools and persistent for at least 6 weeks All ages

Rectal bleeding persistently without anal symptoms Over 60 years

Change in bowel habit to increased frequency of defaecation and/or looser stools persistent for at least 6 weeks Over 60 years

Patients with an easily palpable right iliac fossa mass All ages

Patients with an easily palpable rectal mass All ages

Patients with an unexplained iron deficiency anaemia:below 11 grams in men All agesbelow 10 grams in women Post-menopausal

Symptoms Age Threshold

Methods of Investigation


ASSESSMENT OF THE COLONA complete examination of the large bowel should beundertaken by either:

a. total colonoscopyor

b. adequate endoscopic visualisation of the rectumplus a double contrast barium enema.

It must be accepted however, that both investigationsmay vary in quality, and the choice betweencolonoscopy and barium enema for total colonicexamination will depend on local availability andexpertise. Histology should be considered mandatory ina rectal cancer which might result either in permanentstoma formation or an ultra-low anterior resection, orwhen pre-operative radiotherapy is being considered.

Regardless of whether colonoscopy or barium enema isemployed, certain minimum levels of quality should beachieved with both of these investigations.

Colonoscopy should usually be done as a day-caseprocedure after full bowel preparation, and theendoscopist should be prepared to biopsy or removeappropriate lesions (i.e. areas of inflammation, sessiletumours and polyps). The patient should be warned ofpossible discomfort and the risks of perforation andbleeding. If sedation is used, care should be taken toavoid complications and guidelines have been issued.

It is important that the endoscopist can recognise whena total colonoscopy has been achieved, and this canonly be guaranteed when the terminal ileum has beenunequivocally identified (Cotton & Williams 1990. IV).This is generally not practicable. A printed picture ofthe ileo-caecal valve may be a reasonable compromise.

Barium enemas should all now be double contrastexaminations (Laufer 1979. III), and should be carriedout by, or under the supervision of, a radiologistexperienced in the performance and interpretation ofthis technique. Every attempt should be made toexamine the whole of the large bowel and particularattention should be paid to the sigmoid colon andcaecum, as failure to display this area properly canlead to lesions being missed (Lauer et al 1965. III). Inaddition, inexperience combined with failure to distendthe caecum can produce misleading appearances whichcan be misinterpreted as malignancy.

Non-committal reporting of barium enemas by theresponsible radiologist can be a problem for theclinician, and may be due to lack of experience in theinterpretation of films, a technically inadequateexamination or medico-legal concerns. For a bariumenema to be of use in reaching a clinical decision, a firmopinion as to the most likely process giving rise to theradiological appearances should be given on the report.

Despite good radiological technique, it may beimpossible to be sure of excluding neoplasia with anycertainty, particularly where there is severe diverticulardisease of the sigmoid colon. In such cases,supplementary endoscopy either in the form of flexiblesigmoidoscopy or colonoscopy is mandatory.

In summary, it is recommended that patients withsuspicious symptoms or a proven colorectal cancer beinvestigated with either endoscopic visualisation of thewhole rectum plus a high quality double contrastbarium enema or total colonoscopy. Supplementaryflexible endoscopy should be carried out where it isimpossible with any certainty to exclude neoplasia onbarium enema. Grade B

Histology should be obtained from all rectal tumours.Grade B

Doctors carrying out colonoscopy should audit theirresults, and expect to achieve a high total colonoscopyrate with a low perforation rate. Grade B

Assessment of the Rectum: Local extensionand peri-rectal lymph nodesWhere possible magnetic resorance imaging (MRI)rather than Computed Temography (CT) should beundertaken when rectal cancers are being consideredfor resection as it is important to determine whetheradjacent organs are involved (Padhani 1999. III, Kumar & Scholefield 2000. III, Heriot et al 1999. III).The degree of local extension may determine whether a curative excision can be achieved, and whether pre-operative adjuvant radiotherapy should be given(Brown et al 1999. IIb) and (Beets-Tan et al 2001. IIb).In patients with rectal cancer where local excision isbeing considered (T1 lesions), staging by endorectalultrasound scanning to determine the depth ofpenetration is recommended.

Methods of Investigation


MRI can identify small peri-rectal lymph nodes byvirtue of their morphology (Brown et al 1999a. IIb).The use of size criteria alone for defining lymph nodeinvolvement is unreliable and must be used with greatcaution. Lymph nodes greater than 1cm in diameterare more likely to be involved (Padhani 1999. III,Kumar & Scholefield 2000. III, Heriot et al 1999. III),The majority of involved lymph nodes in colorectalcancer specimens measure less than 5mm (Dworak1991. III) and there is currently no method fordifferentiating these from reactive nodes.

The value of MRI lies not so much in early T1-T2staging (where rectal endosonography is currently moreaccurate) but in assessing the tumour extent(particularly in the lateral and anterior planes).Involvement of the mesorectum is easily demonstratedon MRI. A histological clearance of less than 1mm maybe safely predicted when this appears to be less than5mm on MRI (Beets Tan, 2001. III). The high soft tissuedifferentiation, multiplanar imaging and improvedresolution of current scanners dictate that MRI is fastbecoming the investigation of choice for rectal cancerregardless of position (Brown 1999. IIb).

Transrectal endosonography is recommended whereclinical examination suggests that local excision maybe feasible.

Pre-operative Assessment


Pre-operative assessment is carried out to confirm thediagnosis. All rectal cancers should be histologicallyproven. It is important to determine extent and spreadand distant spread (in particular hepatic metastases byultrasound, computed tomography (CT) or magneticresonance imaging (MRI) and establish the presence orabsence of synchronous lesions (4-5%), usingcolonoscopy and/or double contrast barium enema(DCBE) and sigmoidoscopy. Pre-operative staging(extent of local spread) of distal rectal cancers may becarried out using CT, MRI or endoluminal ultrasound.Results of these staging procedures should bedocumented clearly in the notes.

Where possible histological confirmation of diagnosisshould be obtained prior to surgery, particularly inrectal cancer.

In rectal cancer, distance from the anal verge, extent ofcircumferential involvement and degree of fixity shouldbe noted.

The large bowel should be assessed either bycolonoscopy or a combination of rigid/flexiblesigmoidoscopy and DCBE.

A chest radiograph and liver imaging should beperformed to determine tumour dissemination. Grade C, level IV

COMMUNICATION WITH PATIENTSMost patients with cancer wish to have informationabout their condition and proposed management, andto be involved in decisions about their treatmentoptions. This should be readily available in clear andreadable publications as well as through direct patientcontact. There should not be undue delay in impartingthe results of investigations. Breaking ‘bad news’ isinitially likely to be done by the surgeon. It isimportant to establish what patients know, fear andwant to know. Discussion of surgical treatment mayneed to address stoma formation and possibledifficulties with sexual, urinary and rectal function.

Patients should be given the information they wishover a period of time.

Coming to terms with the diagnosis and treatmenttakes time and the information which patients wish orneed may vary during this period.

Prior to carrying out investigations to establish thediagnosis of colorectal cancer patients should beprovided with relevant information about theprocedures to be used.

Consideration should be given to increasing patientacceptability of the different investigative procedures.

Patients should have the opportunity to discusstreatment options and be appraised of the likelyconsequences of different procedures (e.g. stomaformation). They should be involved in decision-making to the extent they wish. Grade B, levels IIb & III

PREPARATION FOR SURGERYThe risks of surgery and possible morbidity (e.g.impotence in males), likelihood and effects of stomaformation need to be discussed with the patient.Bowel preparation and prophylaxis for deep veinthrombosis (DVT) and sepsis are all initiated in theperioperative period.

All elective patients should have mechanical bowelpreparation and antibiotic prophylaxis. Grade A, level IB

All elective patients should have DVT prophylaxis

Patients requiring formation of a stoma should be seen bya stoma therapist prior to surgery and those undergoingpelvic dissection should be informed about possiblebladder and sexual dysfunction following surgery. Grade C level IV

In general terms, surgery for colorectal cancer shouldbe avoided if the hazards are deemed to outweigh thepotential benefits, i.e. in the patient who is medicallyunfit for surgery or who has advanced disease which isnot amenable to surgical therapy. As the decision notto operate depends on highly individual factors it isimpossible to provide such guidelines, but in makingsuch a decision it is important to involve the patientand/or close relatives so that the underlying reasoningis clear and acceptable to all concerned.

Given that surgery is to proceed, there are certainfundamental aspects of preparation which deserveconsideration, and these are listed below:

a. Informed consent.

b. Anorectal manometry.

c. Preparation for stoma formation.

d. Cross-matching of blood.

e. Bowel preparation.

f. Thrombo-embolism prophylaxis.

g. Antibiotic infection prophylaxis.

Colorectal Cancer Management Clinical Guidelines 15IntroductionPre-operative Assessment

a. Informed consentAll patients undergoing surgery for colorectalcancer should give informed consent unless theyare unable to do so in which case it may benecessary to obtain consent from a relative. Theconsent should be obtained by a doctor who fullyunderstands the nature of the operation, and isable to answer any pertinent questions the patientor relatives may have. The risks of death andmorbidity must be carefully explained; inparticular, the likelihood of requiring a stoma anddeveloping urinary problems and impotence afterrectal surgery should be discussed. In addition,patients should be warned of the increased stoolfrequency often experienced after righthemicolectomy or subtotal colectomy, and the riskof a poor functional result after a low anteriorresection should be explained. Grade C

b. Anorectal manometryIt is desirable that patients proposed for lowanterior resection, particularly females, undergoevaluation of anal sphincter function bymanometry pre-operatively.Grade C

c. Preparation for stoma formationIf a patient may require a stoma the nature andconsequences of this should be carefully explained. It is also important that the site of the stoma bemarked prior to surgery to ensure optimum fitting of the appliance (Devlin 1982. IV). The patientshould be seen by a stoma therapist prior tosurgery (Saunders 1976. IV) and the referralshould be made at the earliest possibleopportunity to allow adequate time forpreparation. It is recognised that this may not bepossible in some emergency situations and in thiscase the stoma site should be marked by anexperienced surgeon.

d. Cross-matching of bloodThere is evidence that blood transfusion mayincrease the likelihood of recurrence of colorectalcancer, and immunological mechanisms have beeninvoked (Burrows & Tartter 1982. III). This hasnot been unequivocally proven, however (Bentzenet al 1990. III) and a trial comparing the use ofautologous and allogeneic blood in patientsundergoing resection of colorectal cancer showedno difference in prognosis (Busch et al 1993. Ib). Itis recommended that blood should not be withheldif there is a clinical indication to give it, andpreparation for blood transfusion should be madein all patients undergoing surgery for colorectalcancer except where an individual patient refuses.For an uncomplicated right hemicolectomy, bloodgrouping and saving of serum may be sufficient butformal cross-matching is recommended for moreextensive operations, especially rectal resections(Harrison et al 1992. Ib. A)Grade A

e. Bowel preparationMechanical bowel preparation is generallyregarded as mandatory before elective colorectalsurgery, and all surgeons in Trent/Wales (Ib) usedsome form of such preparation. However, not allsurgeons hold this view (Irving et al 1987. III) andthere has been a recent randomised study whichshowed no benefit from sodium picosulphatepreparation when compared with no preparationin left colonic or rectal resection (Burke et al1994. IIb). No definite recommendations can begiven therefore, but the consensus view is still infavour of mechanical bowel preparation. Grade C


f. Thrombo-embolism prophylaxisPatients undergoing surgery for colorectal cancerare at risk of deep vein thrombosis (DVT) andpulmonary embolism (PE) (Salzman and Davies1980. III). The most widely studied prophylacticmeasure against these complications is the use ofsubcutaneous heparin, and although there havebeen no studies confined to patients with colorectalcancer, a meta-analysis of appropriate trials hasindicated that rates of DVT, PE and death from PEcan all be significantly reduced in general surgicalpatients (Collins et al 1988. Ia). Low molecularweight heparin (LMWH) has attracted recentattention and although a large randomised trial inpatients undergoing abdominal surgery has shownit to be of similar efficacy to standard heparin,bleeding-related complications were less common(Kakkar et al 1993. Ib). Other measures which canbe taken are intravenous dextran, the use ofintermittent pneumatic compression devices andthe use of graduated stockings. Dextran does notappear to be as effective as heparin (Salzman &Davies 1980. III), but there has been one trialindicating that intermittent compression isequivalent to heparin in reducing the incidence ofDVT (Persson et al 1991. Ib). Graduated stockingsalone are less effective than other measures(Persson et al 1991. Ib). It is recommended thateither of these strategies should be employed insurgery for colorectal cancer unless there is aspecific contraindication.

g. Antibiotic prophylaxisThere is now very good evidence that prophylacticadministration of antibiotics can decreasemorbidity, shorten hospital stay and reduceinfection-related costs after general surgicaloperations (Page et al 1993. Ib). Variousantibiotics and combinations of antibiotics havebeen shown to be effective, and in Trent/Wales themost favoured regime was a combination ofcephalosporin and metronidazole. Gentamicinwith metronidazole and augmentin alone was alsoused. If intravenous cephalosporin andmetronidazole are used, there is evidence from arandomised, controlled trial that a single dosegiven immediately before operation is asefficacious as a three-dose regimen in preventingwound infection (Rowe Jones et al 1990. Ib). Grade A

It is therefore recommended that all patients undergoingsurgery for colorectal cancer have antibioticprophylaxis. It is impossible to be dogmatic as regardsthe precise regimen, but a single dose of appropriateintravenous antibiotics appears to be effective. Grade A

Pre-operative Assessment

Surgical Technique


SURGICAL PRINCIPLESThe surgery of colorectal cancer is based ononcological and anatomical principles.

■ Colorectal cancer spreads in a predictable mannerto the regional draining lymph nodes.Locoregional control is best achieved by surgeryand is essential for satisfactory outcome.

■ Survival after ‘curative resection’ is related to thestage of the disease – depth of tumour penetrationand/or the presence of lymph node metastases(biological markers of disseminated disease).

The fixity of the primary tumour will determineresectability and the extent of spread the ultimatesurvival. Tethering or local infiltration is notnecessarily a contraindication for radical surgery.

SURGICAL TECHNIQUE

a. ResectionThere is little controversy regarding the resection ofcolonic tumours. There has been a tendency to moveaway from segmental resections for tumours of thetransverse colon and splenic flexure in favour ofextended right hemicolectomy and although there hasbeen no randomised trials, this is widely accepted asbeing safer. The no-touch isolation technique in whichthe vascular supply to a tumour is divided before thetumour is handled has been tested in a randomisedcontrolled trial and shown to confer no significantadvantage (Wiggers et al 1988. Ib).

In rectal cancer, however, resection technique is ofgreat importance.

In 1999 representatives of the American Society ofColon and Rectal Surgeons and the Association ofColoproctology met with Australian Societies to definethe rectum and the procedures used to treat cancer ofthe rectum. As the treatment of rectal cancer differsfrom the treatment of colonic cancer in someimportant respects, particularly in the areas of surgeryand radiotherapy, it is important to have a clearanatomical definition of the rectum. Strictly, the rectumis that part of the large bowel distal to the sigmoidcolon and its upper limit is indicated by the end of thesigmoid mescolon. Standard anatomical texts put thisat the level of the 3rd sacral vertebra (Williams &Warwick 1980. IV). This is not particularly helpfulpre-operatively, however, and it has been agreed by theExpert Advisory Committee of the Association ofColoproctology of Great Britain and Ireland, that any

tumour for which the distal margin is seen at 15cm orless from the anal verge using a rigid sigmoidoscopeshould be classified as rectal. Grade C

Although most local recurrences after resection ofcolonic cancer occur together with disseminated disease(Abulafi & Williams 1994. III), local recurrences afterrectal excision are often isolated, and reported rates ofrecurrence after curative rectal resection varies from2.6% (Karanjia et al 1990. III) to 32% (Hurst et al1982. III) Individual surgeons vary greatly in thisrespect, with two studies showing a variation of 0 to21% in recurrence rate among participating surgeons(Phillips et al 1984. IIb, McArdle & Hole 1990. IIb).

The reason for this variation is not entirely clear,although there is good evidence that surgical techniqueis a critical factor. Complete excision of the mesorectumis associated with a low rate of local recurrence (Healdet al 1982, Heald & Ryall 1986. III), and a pathologystudy has shown that distal mesorectal spread oftenextends further than intramural spread, withsecondaries being found up to 3cm distal to the primarycancer (Scott et al 1995. IIb). Evidence from Europe hasshown that education of established surgeons can leadto improvements in technique which result in areduction in local recurrence and a reduction in theabdomino-perineal resection rate (Martling et al, 2000.IIa). It is recommended that lymph node clearanceshould extend for 5cm beyond the distal margin of arectal cancer, and in tumours of the middle and lowerthirds of the rectum the only practical way of achievingthis is by total mesorectal excision. When this is done,care must be taken to preserve the autonomic nervesand plexuses on which sexual potency and bladderfunction depend.

There was a concern that a tendency to avoidabdomino-perineal excision (APER) in favour ofanterior resection might account for high localrecurrence rates (Phillips et al 1984. IIb, Neville et al1987. III), but several series show no differencebetween the operations (Dixon et al 1991. III, Morsonet al 1963. III, Patel et al 1977. III, Williams et al1984. III, Holm et al 1995. IIa). Randomisedcontrolled studies comparing APER and anteriorresection are not available, but where differences in localrecurrence rates for the two operations exist it has beensuggested that they may be explained by the plane ofdissection being nearer the rectum in anterior resection – a problem which can be avoided by total mesorectalexcision (Heald 1988. III).


Perforation of the tumour during resection is also animportant factor, as it is associated with local recurrence(Phillips et al 1984. IIb, Patel et al 1977. III, Zirngibil etal 1990. III). This phenomenon appears to beindependent of tumour stage or fixity (Wiggers et al1988. IIb).

In summary, it is recommended that total mesorectalexcision be performed for tumours in the lower two-thirds of the rectum, either as part of a low anteriorresection or an APER. In tumours of the upper rectumthe mesorectum should be divided no less than 5cmbelow the lower margin of the tumour. Care should betaken to preserve the pelvic autonomic nerves andplexuses, and perforation of the tumour duringoperation should be avoided.Grade B

b. AnastomosisAnastomotic dehiscence is a major source of operativemorbidity and mortality after resection for colorectalcancer. Its rate is known to vary greatly from onesurgeon to another and it is known to be morecommon after anterior resection of the rectum thanafter colonic resection (Fielding et al 1980. IIb,McArdle & Hole 1991. IIb).

Although the best published results involved the use of a single layer interrupted serosubmucosal technique(Matheson et al 1985. III, Carty et al 1991. III), this mayhave been due to the skill of the surgeon and/or caseselection rather than the technique itself. Staplingmethods have been compared with manual suturingtechniques in several randomised trials (Beart et al 1981.Ib, Brennan et al 1982. Ib, Everett et al 1986. Ib,McGinn et al 1985. Ib, West of Scotland and HighlandAnastomosis Group 1991. Ib), and overall there is noobservable difference in leak rates with colorectal surgery.

Stapling has, however, made the performance of theultra-low anastomosis after anterior resection muchmore feasible. As it is known that distal intramuralspread rarely extends more than 1cm beyond thepalpable edge of the tumour (Williams et al 1983. IIb),the ability to obtain distal clearance of 1cm of moreshould therefore allow an anterior resection which isoncologically sound so long as it is combined with totalmesorectal excision (vide supra).

Unfortunately, such anastomoses are associated with a high leakage rate, even when the same surgeon hasvery acceptable leakage rates from other types ofresection (Karanjia et al 1994. III). However, there isevidence that the defunctioning stoma can amelioratethe consequences of leakage, decreasing the risk ofdeath and need for a permanent stoma (Karanjia et al 1994. III).

Several trials have compared a defunctioning ileostomywith defunctioning colostomy with mixed outcomes.There are advantages and disadvantages for each typeof stoma. The balance of evidence slightly favours adefunctioning ileostomy over transverse colostomy(Williams et al 1996. Ib, Gooszen et al 2000. Ib).

Other problems associated with the low anastomosisare functional; many patients have urgency andincreased frequency action (Williams & Johnston1983. Ib) after low anterior resection, and this has been attributed to loss of the reservoir function of therectum. Formation of a colonic J-pouch may overcomethis difficulty, and several studies now attest to theefficacy of this procedure (Seow-Choen & Goh 1995.Ib, Mortensen et al 1995. IIb).

Finally, as large numbers of viable tumour cells can be demonstrated in the lumen of the colon at the timeof operation (Umpleby et al 1984. IIb) the use ofcytocidal washout prior to anastomosis is generallyaccepted as a sensible precaution to reduce the risk of anastomotic recurrence.

Although no definite recommendations can be maderegarding anastomotic technique, the interruptedserosubmucosal method has the lowest reported leakrate and stapling facilitates ultra-low pelvicanastomoses. After anterior resection and totalmesorectal excision, the judicious use of a temporary de-functioning stoma is recommended and the formationof a colonic pouch should be considered. Grade B

Cytocidal washout of the rectal stump prior toanastomosis should be used.

Surgical Technique


RATES OF PERMANENT STOMAFORMATIONThe lowest rate of permanent stoma formation forrectal cancer in the literature is 9% in a unit routinelyemploying a stapled anastomotic technique for lowanterior resection (Karanjia et al 1994. III). Otherspecialist units have reported rates of 10% (Williams et al 1985. III) and 19% (Matheson et al 1985. III).However, in a report from the West Midlands CancerRegistry the proportion of rectal tumours treated byabdomino-perineal excision (APER) between 1957 and 1981 was 68%(Allum et al 1994. IIb). In theLarge Bowel Cancer Project, 56% of patientsundergoing curative resection of the rectum had anAPER (Phillips et al 1984. IIb) in the Trent/Wales auditthe figure was 37% (IIb). It is not clear why thesedifferences exist. There seems to be a general reductionin the proportion of rectal cancer treated by APERwith the passage of time, but there is still markedindividual variation. Case-mix and an increasinglyelderly population may explain some of this variation.As staged above, distal intramural spread rarelyextends more than 1cm beyond the palpable edge ofthe tumour (Williams et al 1983. IIb), and it is possible that failure to recognise this finding results in aninappropriate number of APERs being performed by non-specialist surgeons.

In low rectal cancers, a surgeon may be unsure of the feasibility of anterior resection. In such a case, it is strongly recommended that a second opinion froman experienced rectal surgeon is obtained (IV).

It is difficult to determine what the ideal ratio ofanterior resection to APER should be, but it isrecommended that the overall proportion of rectalcancers treated by APER should be less than 40%.Depending on stage of disease at presentation and ifdistal clearance of 1cm can be achieved, a low rectalcancer may be suitable for anterior resection. If asurgeon has any doubt regarding the choice betweenthese two operations, an experienced second opinionshould be sought. Grade B

LOCAL EXCISIONOccasionally small pT1 rectal cancers are technicallysuitable for a local excision transanally and somepolyps excised by snare diathermy will contain invasivecarcinoma. Careful studies have shown that cancersfulfiling the histological criteria can be regarded ascured by a local excision (Whiteway et al 1985. IIb)whereas pT2 tumours are associated with a higher risk of lymph node involvement and of local recurrencewithout further treatment (Graham et al 1990. IIb).Local excision of rectal adenomas using transanalendoscopic microsurgery has become popular over the last five years. Published data suggest that this is atleast as good as traditional transanal resection and mayoffer advantages for polyps in the middle third of therectum (Steele et al 1996. IIb).

In summary local excision for cure in rectal cancershould be restricted to pT1 cancers with well-ormoderate-differentiation and less than 3cm in diameter.It must be accepted that subsequent histopathologicalexamination of cancers thought to be suitable for localexcision will identify a small proportion which requiremore radical surgery. Grade B

The anastomotic leak rate should be less than 8% foranterior and less than 4% for other types of resection.

Operative mortality for colorectal surgery should beless than 5% for elective resections and below 25% foremergency procedures. B level IIb & III

Surgical Technique

Emergency Surgery


In population-based studies, approximately 30% of colorectal cancers present as emergencies – most with obstruction – a few with perforation or bleeding.Patients presenting as emergencies tend to be older,have concomitant disease, a more extensive malignantprocess, spend longer in hospital and are more likely to have a permanent stoma. Post-operative morbidity(19%) and mortality (8%) is higher and survival poorer (30% at five years), compared with patientspresenting electively.

A clinical diagnosis of obstruction should be confirmedby a plain abdominal radiograph and a water solublecontrast enema or endoscopy to exclude pseudo-obstruction (Koruth et al 1985. IIb).

In the absence of perforation or life-threateningbleeding, operation for large bowel obstruction can be regarded as an urgent rather than emergencyprocedure, and every effort should be made to operateduring the day with experienced surgeons andanaesthetists. An exception to this may be the situationwhere the ileo-caecal valve is competent, and thecaecum is in danger of perforation.

The patient with obstruction should be carefullyprepared for surgery, with adequate fluid resuscitationmonitored by blood pressure and urine outputmeasurements at the very least. Antibiotics and DVTprophylaxis should be administered. Centresundertaking this type of surgery should have anintensive care unit or high-dependency unit, and theseshould be used for post-operative, and, occasionally,pre-operative care when appropriate.

The type of surgery which should be undertaken forlarge bowel obstruction is to some extent controversial,but broad guidelines can be given. For right-sidedlesions, primary resection and ileocolic anastomosis is usually feasible (Deans et al 1994. III). For left-sidedlesions, the use of simple defunctioning colostomy is notgenerally favoured except in extreme circumstanceswhere the patient is not considered fit for a moreextensive procedure. Rather, immediate resection of theobstructing cancer should be carried out, either as aHartmann’s procedure with end colostomy, or, whenconditions are favourable, as a primary resection withanastomosis (Deans et al 1994. III). If the latter option ischosen, this can be done either as a segmental resectionwith on-table colonic lavage (Koruth et al 1985. IIb), oras a subtotal colectomy with ileorectal anastomosis(Dorudi et al 1990. III). A recent randomised trial hasindicated that these two procedures are roughlyequivalent, although long term bowel habit is betterwith the former (SCOTIA 1994. Ib).

In summary, emergency surgery should be carried outduring daytime hours as far as possible, by experiencedsurgeons and anaesthetists. Grade C

In patients presenting with obstruction, measuresshould be taken to exclude pseudo-obstruction beforeoperation and stoma formation should be carried outin the patient’s interest only – not as a result of lack ofexperienced surgical staff. The overall mortality foremergency/urgent surgery should be less than 25%.


COLON CANCERThere is evidence that in patients who have had acurative resection for Dukes’ C carcinoma of the colon,adjuvant systemic chemotherapy has a significantimpact on population survival (Moertel et al 1995. Ib,IMPACT, 1995. Ib, Zaniboni et al 1998. Ib). Tworecent meta-analyses have been performed. In thecolorectal cancer collaborative group meta-analysis12,000 patients in 33 randomised controlled trials wereincluded. This concluded that prolonged use of 5FUbased regimens for greater than three months canimprove survival in colorectal cancer. The size of thebenefit for Dukes’ C colon cancer was estimated atabout 6% (range 2-10%) (NHS Executive 1997. IV). A different meta-analysis of 39 trials indicates the sizeof the benefit is about 5% absolute improvement in 5-year survival in those receiving chemotherapy (Dubeet al 1997. Ia). The side-effects of the treatment havebeen shown to cause only minor psychological distress(Norum 1997. IIb). However, in individual patients thesurvival benefit is small and there will be some patientsfor whom systemic chemotherapy is inappropriate inview of age or comorbidity.

Patients with Dukes’ C colon cancer should beconsidered for adjuvant chemotherapy. Grade A

In node-negative colon cancer (Dukes’ B), there remainsuncertainty of the value of adjuvant chemotherapy. A pooled analysis of 1116 patients with B2 (AstlerColler) colon cancer randomised to chemotherapyversus observation showed no significant improvementin overall survival (HR 0.86, CI 0.72-1.07), 5-yearsurvivals were 80% for control and 82% forchemotherapy patients, (IMPACT B2 1999. Ib). Incontrast, a grouped analysis of the National SurgicalAdjuvant Bowel Project (NSABP) trials C-01-04, whichincluded 1565 Dukes’ B patients, concluded that a 30%proportional reduction in mortality resulted from theuse of chemotherapy (Mamounas et al 1999. Ia)consistent with a 5% absolute reduction in death at 5-years. This data together suggests a small (perhaps 2-5%) absolute increase in survival for Dukes’ B cancerpatients for adjuvant chemotherapy. Given the toxicityof treatment, this is inadequate justification for routineuse of chemotherapy in these patients. The size ofbenefit and its balance with toxicity needs furtherclarification in clinical trials.

Some poor risk features can be identified in Dukes’ B cancers (serosal involvement, perforated tumours,extramural vascular invasion, poorly-differentiatedhistology, and in rectal cancer, involvement of the

circumferential resection margin). These patients haveas poor a prognosis as node-positive patients.Individual patients should be assessed for their specificrisk on this basis and counselled regarding the relativelack of evidence to support adjuvant chemotherapy.However, it may be appropriate to offer chemotherapyto such patients on the basis of the proportionatereduction in risk observed in the clinical trials.

Patients with Dukes’ B colon cancer should beconsidered for entry into randomised trials ofadjuvant chemotherapy

Patients with high risk Dukes’ B colon cancer shouldbe individually counselled about their level of risk andpossible benefits of chemotherapy.

There is no evidence to support the use of adjuvant chemotherapy in Dukes’ A cancers of thecolon or rectum.

RECTAL CANCERThe impact of adjuvant chemotherapy alone has beendifficult to identify because many trials have assessedthe combination of chemotherapy and radiotherapytogether. Combined chemotherapy and radiotherapyhas been associated with a survival benefit in patientswith Dukes’ B. Combined chemotherapy andradiotherapy has been associated with a survivalbenefit in patients with Dukes’ B and C rectal cancer,when compared with radiotherapy alone (Krook et al1991. Ib). The meta-analysis of adjuvant chemotherapyof colorectal cancer (Dube et al 1997. Ia) showed agreater benefit for rectal cancer than for colon cancer(OR for mortality 0.64 95% C1 0.48 – 0.85) andestimated the size of benefit to be a 9% increase insurvival for rectal cancer patients. NSABP R-20evaluated chemotherapy with or without post-operative radiotherapy and showed no advantagefor survival in adding radiotherapy to adjuvantchemotherapy (Wolmark et al 2000. Ib). Conversely,early data from the Dutch study of adjuvant therapy in rectal cancer shows no advantage for adjuvantchemotherapy and they conclude it is acceptable tocontinue randomising into the ongoing trial (Kapiteijnet al, 2001. Ib). However, the indirect evidence suggeststhat there is a benefit for post-operative adjuvantchemotherapy in rectal cancer.

Adjuvant Chemotherapy


No definite recommendation can be made regardingadjuvant chemotherapy for patients with Dukes’ Crectal cancer. Patients may be offered eitherchemotherapy or be considered for clinical trials, in addition to appropriate adjuvant radiotherapy. Grade B

ADJUVANT CHEMOTHERAPYREGIMENSCertain statements can now be made regarding theschedule and duration of adjuvant chemotherapy if it isused. 5FU remains the basis for all adjuvant regimens.There is no advantage to high dose versus low dosefolinic acid (QASAR 2000. Ib, Haller et al 1998. Ib).Levamisole does not add any further benefit in place ofor in addition to folinic acid (QASAR 2000. Ib, Halleret al 1998. Ib, Wolmark et al 1998. Ib). Using astandard 5FU and folinic regimen, there is noadvantage in a longer duration of therapy over 6months (O’Connell et al 1998. Ib, Haller et al 1998.Ib), North Central Cancer Treatment Group). Thetoxicity of a weekly bolus 5FU and folinic acid regimenis less than the conventional 5-day fractionatedregimen but in a non-randomised comparison in theQASAR trial no reduction in effectiveness was shown(QASAR, 2000. Ib).

There is no consensus beyond the above comments asto the optimum chemotherapy regimen. All 5FU basedregimens are associated with a risk of toxicity, notablydiarrhoea, stomatitis and leucopenia. A higher incidenceof severe infections has been observed in older patients,but in spite of increased toxicity patients over 70 areequally likely to complete treatment as younger patients(Moore & Haller 1999. IIb).

Systemic chemotherapy should be administered onlyby clinical staff with appropriate training andexperience, according to Joint Council for ClinicalOncology guidelines. Grade C

The role of portal vein infusion (PVI) of chemotherapyremains uncertain. The meta-analysis of 4000 patients in 10 studies showed a possible survival advantage ofabout 5% (Liver Infusion Meta-analysis Group 1997.Ia). Since then the EORTC study has been negative(Rougier et al 1998. Ib). The UKCCX AXIS trial, whichwas powered to look for an overall survival benefit of5%, showed no overall advantage for PVI 5FU in theinitial analysis (James 1999. Ia). However, the estimated 5-year survival benefit in a curatively resected colon

cancer group was 5% compared with 0% for curativelyresected rectal cancer patients. Similar trends were seenin the meta-analysis, although at present it is notpossible to say conclusively that any benefit to PVI 5FUis likely to be greater than 5%. However, molecularanalysis of the curatively resected colon cancers in AXIS(a group chosen before the main AXIS results wereavailable) showed significantly larger treatment effects in those with certain molecular markers on genetictesting, compared to those without these markers(Barratt et al 1999. IIb).

Therefore following full evaluation with large scale trialsaround the world the level of benefit from a one weekinfusion of 5FU does not indicate it should be routinelyapplied. However, its role is not yet finally resolved.

ADJUVANT RADIOTHERAPYThere is no established role for radiotherapy in themanagement of colonic carcinoma. The following appliesto carcinoma of the rectum. A systemic overview hasbeen undertaken by the colorectal cancer collaborativegroup and was published in the NHS executive researchevidence of colorectal cancer in 1997. This includes 13pre-operative radiotherapy studies and 8 post-operativeradiotherapy studies.

PRE-OPERATIVE RADIOTHERAPYThere is consistent evidence that local recurrence of rectal cancer can be reduced by about 50% byadequate dose pre-operative radiotherapy given overone week for five weeks. The proportional reduction is not influenced by stage of tumour or fixity. In earliertrials, parallel opposed treatment fields were used, andan increased post-operative mortality fromcardiovascular causes was observed. This effect wasnot present in other trials when improvedradiotherapeutic technique employing three or fourfields was used.

However, in all these trials local recurrence rates on the surgery alone arm were in the region of 20 – 30%.With improvements in surgical techniques, includingtotal mesorectal excisions (TME), local recurrence ratesare falling to less the 10% in many specialistspractices. Preliminary results from the Dutch rectalcancer trial show that local recurrence rates werereduced from 8.2% with TME surgery alone to 2.4%with the addition of a one-week course of pre-operative radiotherapy (Kapiteijn et al 2001. Ia).



This was at a cost of increased perianal infection ratesof 26% versus 18% in the abdomino-perineal group (n = 485, p = 0.05) and increased sexual difficulties(impotence and dyspareunia) in the irradiated group.No survival benefit is apparent at this time. Thereremains uncertainty as to whether this level ofreduction in local recurrence rates in the absence of asurvival advantage and in view of the side-effects issufficient to use pre-operative radiotherapy in all casesof mobile rectal cancer.

POST-OPERATIVE RADIOTHERAPYMeta-analysis of the post-operative radiotherapy trialsalso shows an effect on local recurrence, but theevidence is less consistent and the size of benefit issmaller (33% reduction SD 11%, p = 0.002) than forpre-operative radiotherapy. No effect on survival hasbeen confirmed. The combination of chemotherapy andradiotherapy has been shown to improve survival(Krook et al 1991. III) and has been the standardAmerican pattern of care for Dukes’ B and C rectalcancer since the 1990 National Institutes of Healthconsensus statement (IV). This has evolved into acombination of 5FU by infusion during radiotherapywith improvement in relapse and overall survival(O’Connell et al 1994. Ib). The question remains as towhether the improved survival was due to the synergyof 5FU and radiotherapy or simply an effect ofimproved chemotherapy. The NASBP R-02 trial showsthat the addition of post-operative radiotherapy tochemotherapy only reduced local recurrence (13% to8%, p0.02) but had no effect on relapse free survival oroverall survival. However, the more effective 5FU + LVchemotherapy did improve survival (Wolmark et al2000. Ib). Thus the evidence for the benefit of adjuvantchemotherapy in rectal cancer is increasing, while theevidence for post-operative radiotherapy seems toindicate that it only has an effect on local recurrence.

The toxicity of post-operative radiotherapy issignificant. There is an increased likelihood of smallbowel being in the radiation field. Grade 3-4 toxicitiesare seen in over 70% of patients with leucopenia (23 – 33%) and diarrhoea (37 – 49%) being mostfrequently reported (Tepper et al 1997. Ib).

PRE-OR POST-OPERATIVERADIOTHERAPYThe weight of evidence is that pre-operativeradiotherapy in rectal cancer has a greater effect onlocal recurrence than post-operative radiotherapy. Pre-operative radiotherapy also has a small effect onoverall survival which is not the case for post-operativeradiotherapy as distinct from chemotherapy. Theadditional argument for using pre-operativeradiotherapy is that it results in a separation in timebetween pre-operative radiotherapy and the post-operative adjuvant chemotherapy and thereby areduction in toxicity.

Recommendations on Adjuvant Therapy

Patients with mobile rectal cancer should beconsidered for entry into clinical trials of pre-operative radiotherapy. Grade C

Patients in whom the tumour is tethered or in whomlocal imaging indicates a high risk of incompleteresection should be selected for long course pre-operative radiotherapy to obtain tumour downstaging. Grade B

Pre-operative radiotherapy using short courses (25Gyin 5 fractions in one week) or longer course (40-45 GYin 20-25 fractions over 4-5 weeks) are both acceptable. Grade A

In patients who have not had pre-operativeradiotherapy, post-operative radiotherapy andchemotherapy should be considered for well-established predictors or in the presence of high riskfor tumour recurrence (e.g. evidence of tumour at thecircumferential resection margins or radial margins of less than 1mm). Grade A

Post-operative radiotherapy doses should be 40-50 Gyin 20-25 fractions or a suitable biological equivalentusing a planned volume. Grade B

A planned radiotherapy volume using three or fourfields is recommended as this results in less morbidityand mortality. Grade B

Patients with potentially operable rectal cancer should always be considered for entry into trials ofadjuvant radiotherapy. Grade B




a. Locoregional recurrenceThe three year survival of patients withlocoregional recurrence of colorectal cancer is inthe region of 10% (Nicholls 1986. III). There issome evidence that resection of locally recurrentdisease may improve survival (Schiessel et al 1986.III, Pollard et al 1989. III), but this has not beenproven in a randomised trial. Pre-operativechemoradiation prior to surgical resection ofrecurrent disease has increased resectability ratesto 60% (Rodel et al 2000. III) but remainsunproven in Phase 3 trials. A systematic review ofthe literature has been performed in an attempt todefine the optimal radiotherapy dose forsymptomatic treatment of recurrent disease (Wonget al 1998. IIa). No randomised trials wereidentified. Symptom relief occurred in around 70-80% of patients after radiotherapy but themedian duration of relief was only 3 months withbetween 25 and 50% being symptom-free at 6months in different series.

b. Inoperable primary diseaseInoperable primary disease is commonest in therectum, and is associated with a very poor prognosis(Baigrie & Berry 1994. III). There is no evidencethat palliative resection improves survival (Baigrie &Berry 1994. III). In a group of patients, clinicallyunresectable tumours may be rendered operable byradiotherapy (Brierley et al 1995. IIa, MRC RectalCancer Working Party. Ib. 1996, Marsh et al 1994.Ib), and such patients have a much better prognosisthan those whose tumours remain in-situ (Pahlman& Glimelius 1992. III). Currently, trials areunderway investigating the role of combinedchemoradiation schedules to downstage suchpatients prior to attempted resection.

For patients unfit for such an approach, palliationis the objective of therapy. In the patient with aninoperable obstructing rectal cancer, adefunctioning colostomy, preferably in the form of a Hartmanns’ procedure, may provide usefulpalliation. Transanal tumour ablation using laser,electrocoagulation or resectional techniques mayprovide better palliation, and should be consideredin these cases (Baigrie & Berry 1994. III).Radiotherapy is useful for relieving pelvic pain(Sischy et al 1982) but side-effects can occur(Puthwala et al 1982. III) and duration of relief isrelatively short-lived. Chemotherapy alone may be useful in the presence of metastatic disease butlocal treatments for the primary tumour may stillneed to be considered.

For fit patients with inoperable rectal carcinomawithout evidence of metastatic disease, primaryradiotherapy alone or in combination withchemotherapy should be considered Grade B

c. Metastatic diseaseThe liver and the lung are the commonest sites for metastatic colorectal cancer. In most instances,systemic treatment is the only therapeutic option,although in a small number of cases surgicalexcision of metastases or in-situ destructivetherapy may be feasible.

Patients with a small number of metastases in theliver or lung may benefit from appropriateresection, and with careful patient selection,hepatectomy for colorectal metastases can beassociated with a 5-year survival of around 33%(Scheele et al 1990. III). The hypothesis that thisapproach prolongs life has not been tested by arandomised trial, but a retrospective review of2040 patients with a metachronous isolatedhepatic metastasis compared the outcome of thosewho did not undergo resection with those whodid. After resection of hepatic metastases meansurvival was 31 months (projected 5-year survival25%) compared with those (887) who did nothave resection whose mean survival was 11months (projected 5-year survival 2%; p<0.01)(Wade et al 1996. III). Non-randomised evidenceexists to support the use of pre-operativechemotherapy prior to resection in those withpotentially operable liver metastases (Giachetti etal 1999. III). It is important to note that goodresults from this type of intervention depend on alow operative mortality, and should only beattempted where this can be achieved.

Colorectal Cancer Management Clinical Guidelines 25Treatment of Advanced Disease

In-situ destructive therapies (interstital laserablation, cryotherapy, radiofrequency ablation)have been in use for over the last decade forcolorectal liver metastases. In Phase 2 trials inexcess of 50% of patients are alive at two years,which compares favourably with systemicchemotherapy (Seifert & Morris 1998. III, Curleyet al 1999. III, Rossi et al 1996. IIb). However,patient selection for fitness to travel for invasivetherapy and metastases limited to the liverinfluence the overall validity of this percentage.There is no Phase 3 evidence of benefit and thesetherapies require further evaluation.

A meta-analysis of 5 trials of palliativechemotherapy (3 systemic, 2 regionalchemotherapy, Scheithauer et al 1993. Ib,NGTATG 1992. Ib, Beretta et al 1994. III, Allen-Mersh et al 1994. IIb) has demonstrated animproved survival with chemotherapy comparedwith best supportive care (p=0.002). The evidenceindicates that early chemotherapy (12 weekstherapy) prior to clinical deterioration foradvanced disease improves survival by 3 to 6months without any adverse impact on quality oflife. A rest from treatment with close observationuntil disease progression is not detrimental tosurvival and contributes to improved quality oflife (Maughan 2001).

Selection of patients for chemotherapy requires theopinion of a medical oncologist experienced incolorectal cancer chemotherapy. Poor performancestatus, low serum albumin, high alkalinephosphatase and liver involvement wereindependent predictors of progression, and lowserum albumin, high y-glutamyl transferase andhigh CEA predicted poor survival (Fontzilas et al1996. III). Performance status is a particularlypotent indicator. In a meta-analysis of patientstreated in trials of 5FU-based chemotherapy,median survivals were 4, 10 and 14 months forpatients with ECOG performance status scores of2, 1 and 0 respectively (Thirion et al 1999. Ia) (SeeAppendix 3). However, whilst these variables mayhelp identify patients with an overall poorprognosis, they do not necessarily predict who willor will not benefit from chemotherapy.

The mainstay of treatment is still 5-fluorouracil. Thefollowing conclusions can be drawn regarding 5FUtherapy in metastatic colorectal cancer:

1. Infusional regimens (at least 24h duration) of 5FUdouble response rates (23% v 13%P<0.003)compared with bolus regimens with a smallimprovement in median survival (Meta-analysisGroup in Cancer (MGC), 1998. Ib) and reducedtoxicity (MGC, 1998. IIa).

2. Hepatic arterial infusion (HAI): HAI with 5FU orFUDR increases response rate for isolated hepaticmetastatic disease from 14% to 41% (p<0.0001)compared with IV regimens (MGC 1996. IIa). Nosurvival advantage is proven at present.

3. Biomodulation is the use of a second agent tomodulate the cellular response to 5FU therapy.Biomodulation of 5FU with folinic acid doublesresponse rates and has become a standardcomponent of therapy (23% v 11% p< 0.0001).(ACCMP 1992. IIa). Low-dose folinic acid hasbeen shown to be as effective as high-dose whenused in combination with bolus regiments of 5FU.

4. Combined therapy with established drugs: Thecombination of protracted venous infusion of 5FUwith mitomycin C (I RCT) has shown improvedresponse rate, progression-free survival andsurvival compared with 5FU alone (Ross et al1997. Ib). Biomodulation of 5FU withmethotrexate doubles response rate (19% v10%p<0.0001) with a small improvement insurvival. (Median survival 10.7 v 9.1 months;p=0.02) NHS Exec 1997. IV). It has not beenshown to be superior to folinic acid modulation.Neither cisplatin or interferon have been shown tobe beneficial when combined with 5FU (NHS Exec1997. IV).

Patients with metastatic disease who are fit for activetherapy should be accurately staged with CT scans ofabdomen and thorax.


New Chemotherapy Agents

Several new chemotherapeutic agents have beenevaluated in the last five years.

1. Thymydilate synthase (TS) targeted chemotherapy

A number of new agents have been evaluatedwhich also target TS. These include 5FU prodrugs(e.g. capecitabine) and agents using the reducedfolate pathway (raltitrexed). No evidence ofincreased efficacy over optimal usage of 5FU hasbeen demonstrated for any of the agents to date.Equivalent survival benefit with increased ease ofadministration has been documented forcapecitabine and 5FU with respect to bolus 5FUand low dose folinic acid (Pazdur et al 1999. Ib,Carmichael et al 1999. Ib, Twelves et al 1999. Ib,Cox et al 1999. Ib). For raltitrexed, equivalentsurvival and response rates have beendemonstrated (Cunningham et al 1996. Ib,Cocconi et al 1998. Ib, Pazdur & Vincent 1997.Ib, Maughan et al 1999. Ib) but with increasedtoxicity when compared with infusional 5FUschedules (Maughan et al 1999. Ib). Treatment-related deaths range from 3-6% in the trials andparticular attention to renal function, patientselection and supervision is required for safe usageof this agent.

2. Irinotecan

Irinotecan is a topoisomerase-1 inhibitor. In second-line therapy, two trials show improvedsurvival for chemotherapy with single agentirinotecan in 5FU refractory patients versus BSC(Cunningham et al 1998. Ib) and versus alternative5FU regimens (Rougier et al 1998. Ib) withmaintained quality of life. In first-line therapy, twotrials show improved response rate, progression-freeand overall survival using irinotecan in combinationwith 5FU for metastatic disease (Douillard et al2001. Ib, Saltz et al 2001. Ib). Side-effects ofirinotecan include diarrhoea, myelotoxicity, alopeciaand anticholingergic syndrome.

3. Oxaliplatin

Oxaliplatin is a novel platinum agent which actsby DNA cross-linking. For second line therapy,Phase 2 trials show activity in 5FU-refractorycolorectal cancer by the addition of oxaliplatin toa 5FU regimen (Raymond et al 1998. Ib). In first-line therapy, two trials show improvedresponse rate and progression-free survival usingoxaliplatin in combination with 5FU in first-linetreatment versus 5FU alone, but no overall

survival benefit has been documented, although ahigh proportion of patients crossed over to thecombination therapy on progression (de Gramontet al 2000. Ib, Giacchetti et al 1997. Ib). Side-effects of oxaliplatin include some enhanced5FU toxicity (diarrhoea, mucositis, hand-footsyndrome, myelotoxicity, nausea, vomiting andreversible cold-induced sensory neuropathy).

RecommendationsPatients with evidence of unresectable metastaticdisease should be referred to a medical oncologist forconsideration of palliative chemotherapy as soon as thediagnosis of metastatic disease is made.

Grade A

Chemotherapy for metastatic colorectal cancer shouldonly be given after discussion at the MultiDisciplinary Team Meeting and under the direction ofrecognised oncologists within facilities conforming toJCCO guidelines.

Grade C

Entry into clinical trials evaluating the benefits of novel chemotherapy regimens in colorectal should be encouraged.

Grade C

On current evidence, standard therapy should includean infusional 5FU regimen combined with the use ofirinotecan in first line or on 5FU failure if the patientremains fit for chemotherapy (Performance Score 0-1).

Grade A

Hepatic arterial infusional chemotherapy remains ofunproven benefit.

Grade A

Patients with metastatic disease limited to the liverwhich is potentially resectable should be considered forpartial hepatectomy by an experienced liver surgeon.

Grade B

D Palliative CareThe diagnosis and treatment of cancer can have adevastating impact on the quality of patients’ lives and that of their families and carers. Cancer patientsface uncertainty and may have to undergo unpleasantand sometimes debilitating treatments. Patients,families and carers need access to support from thetime that cancer is first suspected through to deathand into bereavement.

Good communications between health professionalsand patients is essential for delivery of high qualitycare. It is also central in empowering patientsinvolvement in decision-making. All cancer


Colorectal Cancer Management Clinical Guidelines 27Treatment of Advanced Disease

patients, but particularly those with advanced orincurable disease, need to receive high qualityinformation, symptom control, psychological, socialand spiritual support.

In the past patients tended to be referred for palliativecare only when they were in the terminal phase of theirillness. But increasingly palliative care is being seen asan integral part of care, often being delivered alongsidecurative treatment. Careful and expert symptomcontrol is an important aspects of quality of care.

All patients should have access to specialist palliativecare advice and services appropriate to their needs.Services should be provided in the community and inhospitals as well as in specialist palliative care units. Theoverall management plan agreed with the patient andfamily should include the understanding of the extent towhich the patient wishes to be informed and involved indecision-making, how far active treatment should bepursued and where the patient would prefer to die.

RecommendationsSurgeons and other oncologists who deal withcolorectal cancer should make it a priority to buildclose links with palliative care specialists and units.

Grade B

All clinicians who deal with colorectal cancer shouldbe trained in communication skills and in the controlof pain and other cancer symptoms.

Grade C

It is important that patients with colorectal cancer areoffered the opportunity to ask questions and to haveimportant information repeated. Information-givingshould be seen as an essential part of every consultation.

Grade C

Outcome


a. Operative mortalityOperative mortality for operations for colorectalcancer varies according to whether the operationcategory is elective or emergency and whether it iscurative or palliative. For curative resections,considerable variation exists among surgeons, with 30day mortality ranging from 0 to 20% (McArdle &Hole 1990. IIb).

It is therefore recommended that surgeons shouldexpect to achieve an operative mortality of less than20% for emergency surgery and 5% for electivesurgery for colorectal cancer. Grade B

b. Wound InfectionWith modern antibiotic prophylaxis the rates of woundinfection (as evidenced by pus) should be less than10% (Page et al 1992. Ib, Rowe-Jones et al 1990. Ib). It should be noted, however, that a rate of 2% forelective colorectal surgery has been reported (Mathesonet al 1985. III).

It is therefore recommended that wound infectionrates after surgery for colorectal cancer should be lessthan 10%.Grade A

c. Recurrence ratesAs indicated in the sections on surgical technique andadjuvant therapy, local recurrence after resection ofrectal cancer may be influenced by surgical techniqueand the use of radiotherapy. Current evidence suggeststhat, with the use of optimal surgical techniques andpre-operative radiotherapy for tethered or fixedtumour, local recurrence rates of 10% or less shouldbe achievable.

It is therefore recommended that surgeons should audittheir results and aim to achieve local recurrence ratesafter curative resection of 10% or less.Grade A

d. Survival ratesThe overall 5-year survival rate for colorectal cancer inthe UK is currently in the region of 38% (CRC 1993 2).Data from the Birmingham Cancer Registry between1977 and 1981 indicates that after curative resection, 5-year age-adjusted survival rates for colon cancer are85%, 67% and 37% for Dukes’ stage A, B and Crespectively. For rectal cancer, the equivalent figures are 80%, 55% and 32% (Slaney et al 1991. IIb).

It is recommended that surgeons should audit thesurvival rates of patients and examine carefully theirpractice with a view to meeting or improving ontargets set by national statistics. Grade B


Reasons for Follow-upThe value of routine follow-up investigations forcolorectal cancer patients after treatment iscontroversial. Trials show that intensive routineinvestigations are not associated with survivalimprovement and no established strategy of routinefollow-up is agreed. Reasons for follow-up afterapparently curative operation for colorectal cancer canbe summarised as follows:

a. Detection of recurrent disease at an early or pre-symptomatic stage when further attempts at cure might be possible.

b. Surveillance in order to detect metachronoustumours.

c. Provision for psychological support for the cancer patient.

d. Facilitation of audit and outcome.

All patients should have access to follow-upfacilities, and the evidence relating to methods,frequency and efficacy of clinical follow-up isexamined in the next section.

Nature and Frequency of Follow-upFollow-up policy after curative resection for colorectalcancers is controversial, and there is very littleconsensus among surgeons. In the Trent/Wales audit(IIb), although all surgeons reviewed their patientsafter operation, the frequency of follow-up varied froma single appointment to lifelong surveillance. Variouscombinations of 12 different imaging, biochemical andendoscopic investigations were used to detectrecurrence, but only 22 surgeons employed comparableregimens. A similar variation in practice is reported bymembers of the American Society of Colon and RectalSurgeons (1994). To examine the evidence relating tothe efficacy of follow-up, it has been subdivided intofour categories based on the reasons for follow-up.

a. Detection of recurrent diseaseThe case for frequent follow-up during the firsttwo years after surgery is based on the observationthat 8% of recurrences are detected during thisperiod (Umpleby et al 1984. III). Despite frequentclinical review during this time, many patientsbecome symptomatic from recurrence betweenhospital appointments (Cochrane et al 1980. III,Hulton and Hargreaves 1989. III) and even theaddition of an intensive investigation programmefails to detect approximately 50% of

asymptomatic recurrences (Tornqvist et al 1982.IIb, Bohm et al 1993. IIb). Symptomaticrecurrence is rarely amenable to curative surgery(Camunas et al 1991. III, Safi et al 1993. III Wyatt& Aitkin 1994. III), but the diagnosis ofasymptomatic recurrence, particularly in the liver,is more likely to result in attempts at curative re-operation (Ovaska et al 1990. IV).

Bruinvels and colleagues (1994. III) set out toperform a meta-analysis of published studies todetermine whether intensive follow-up isassociated with increased 5-year survival rates.They were unable to identify a single randomisedtrial with patients allocated to follow-up or nofollow-up groups. They therefore investigated non-randomised studies in which controls were eitherhistorical or self-selected (defaulted from follow-up). Of the seven such studies available in theliterature the authors were unable to draw definiteconclusions. Their suggestions included regularfollow-up and monthly carcino-embryonic antigen(CEA) measurements for the first two or threeyears, combined with aggressive hepatic surgery asindicated. This conclusion, however, was notbased on the results of their meta-analysis.

There is little doubt that an elevation of CEA afterapparently curative resection of colorectal cancer isfrequently associated with recurrent disease.Although this precedes clinical evidence ofrecurrence, there is still no evidence that the leadtime provided by CEA monitoring confers anysurvival benefit. In a 400 patient multicare study75 patients underwent second-look surgery,prompted either by a rise in CEA in 43 (57%) orclinical signs in 32 (43%). Resection for cure andsurvival was the same for patients in each group(Martin et al 1985. IIa). A further largeprospective, but not randomised, study found thatthe proportion of patients alive and clinicallydisease-free at least one year after CEA promptedre-operation was similar to that of patients whodid not have CEA monitoring (Moertel et al 1993). In summary, although there is no evidence thatintensive follow-up for the detection of recurrentdisease improves survival, it is reasonable to offerliver imaging to asymptomatic fit patients duringthe first two post-operative years for the purposeof detecting operable liver metastases.

Follow-up


b. Surveillance for metachronous cancersPatients with colorectal cancer are at increased riskof developing adenomas and a secondary primary(metachronous) cancer in the remaining largebowel (Heald & Lockhart Mummery 1972. IIb,Tornqvist et al 1981. IIb), and surveillancecolonoscopy after the initial resection results in asubstantial yield of such tumours (Juhl et al 1990.III). On this basis it may be recommended thatpatients should undergo colonoscopic follow-up,and if the colon is free of tumours then furthercolonoscopy should be repeated at three to fiveyear intervals (Brady et al 1990. III, Winawar et al1993. Ib), at least until the age of 70 years(Kronborg et al 1983. IV, Barlow & Thompson1994. IV). If tumours are found, however, theexamination should be repeated sooner. It must bestressed, however, that there is no evidence thatcolonoscopic follow-up has a significant impact onsurvival following surgery for colorectal cancer.

Colonoscopy cannot be recommended for thedetection of anastomotic recurrence which islargely a feature of low colorectal anastomoses(Juhl 1990. III). Recurrence at the site should bepalpable by digital rectal examination, or seen atrigid sigmoidoscopy. It has been recommendedthat a rectal anastomosis is examined every threemonths for two years, every six months for twoyears and then annually (Rosen et al 1992. IV).There is no evidence, however, that this practiceimproves survival. The results or re-operation inthis group are disappointing, although anoccasional patient enjoys long term survival (Juhl et al 1990. III).

c. Provision of Psychological SupportThe social and psychological morbidity associatedwith anorectal excision (Devlin et al 1973. III) canbe minimised by a combination of attention tosurgical technique, provision of communityservices and support from a stoma specialist.However, surgery for colorectal cancer gives riseto considerable morbidity from impaired bowel,psychological and sexual function (Sprangers et al1993. III). Little is written about the effects ofroutine follow-up on quality of life after surgeryfor colorectal cancer. A study of patients withvarious cancers, including colorectal, found thatthe majority were in favour of regular follow-upand thought that the advantages outweighed thedisadvantages (Kiebert et al 1993. IIb). Patientswith breast cancer prefer follow-up and hospital

visits do not increase the stress and anxiety (GIVO1994. IIb, Morris et al 1992). There are nocomparable studies in patients who haveundergone curative resection for colorectal cancer.

d. Facilitation of AuditIt is only by audit that surgeons can evaluate theirresults at professional standards. Without thisinformation the stimulus to investigate andperhaps change personal practice is lost. Ifguidelines are to be of value, surgeons must audittheir results and for this some form of follow-up isessential. Currently adequate audit facilities arenot available in most Irish hospitals due to lack ofcentral funding.Grade B

There is no evidence that colonoscopic follow-upimproves survival but it has been shown to produce ahigh yield of treatable tumours. If such a policy ispursued, it is recommended that a “clean” colon shouldbe examined by colonoscopy at 3-5 year intervals.

In the absence of randomised trials, the mostpersuasive arguments for routine follow-up are patientsupport and audit. Audit should be structured withparticular reference to outcome measures, and shouldbe regarded as a routine part of the consultants work.Grade C

All patients with a stoma should have ready access tospecialist nursing staff. Grade C

Follow-up


NURSING CARECancer care is a specialty in which specially trainednurses play an important role. Nurses involved in thecare of patients with colorectal cancer need to be ableto provide information and support for both physicaland psychological needs

Locally agreed standards of care should beimplemented specific to each treatment modality andcover all aspects of the cancer nursing practice. Theservices of a colorectal clinical nurse specialist shouldbe viewed as a desirable component of the multi-disciplinary approach. Grade C Level IV

Nursing assessment is of value in the planning oftreatment of patients with colorectal cancer. Thisassessment should be carried out by a specialist nursewho should address the needs not only of patients butalso of their families, as well as other personal issues.Access to information and advice concerning colorectalcancer is an essential part in ensuring that patientscope adequately with their diagnosis. Despiteclinicians’ involvement with their patients, it is oftenthe nurse who plays a key role insuring that vitalinformation is retained as well as re-explained at alater occasion.

The appropriate information should be made availableto the patients and their relatives in order to promotemaximum understanding and assist copingmechanisms. Patients should be made aware of theavailability of written materials and telephone lineservices at all stages of disease management. Grade ALevel 1b

It is of considerable help to patients if specially trainednurses can recognise anxiety and depression inpatients with colorectal cancer and assist in referringto appropriate personnel, to reduce patients’emotional distress and improve their chances ofcoping with their disease.

All patients should have access to appropriately trainednursing staff. Patients shown to be suffering frompsychological or psychiatric morbidity should bereferred for specialist care. Grade C Level IV

All patients who require a stoma following colorectalsurgery should be seen by a stoma therapist prior tosurgery where possible, and for on-going advice andsupport. Grade C Level IV

It is well-documented that patients with stomas havehigher levels of psychological distress, restriction ofsocial activity and impaired sexual function. Access to a stoma therapist increases patient satisfaction andability to both cope with the disease and with thestoma. After discharge from hospital, patients shouldhave access either directly or by telephone link to aspecialist nurse.

Such access allays fears and helps assist recovery.Continuity of care in the community is essential and stoma therapists have a pivotal role to play byhome visits.

Nursing Care

Appendix 1


HISTOPATHOLOGY REPORTING

Accessi. Indications

Accurate, detailed and consistent pathology reportingis important for estimating prognosis and planningfurther treatment. When applied to groups of patientsit is also an index of any shift towards earlier diagnosiswhich may result from screening programmes.Unfortunately, the quality of pathology reporting hasbeen found to be highly variable (Bull et al 1997. III),and this has important implications for theinterpretation of differences in outcomes in differentareas of the country. The use of structured proformashas been demonstrated to improve the informationalcontent of pathology reports (Cross et al 1998. IIb).

The structure of a pathology report depends on whetherthe tissue submitted is a locally resected carcinoma or afull resection specimen. Such reporting should beavailable for all patients, and it is the surgeon’sresponsibility to ensure that all resection specimens,including polyps, are sent for histological examination.

Processi. Local resections

Local resection includes polyps excised endoscopicallywhich are found to be malignant on subsequenthistological examination and sessile tumours whichare electively treated by formal surgical transanalexcision. In each case it is essential that thepathologist assess all excision lines. For polypectomyspecimens this requires careful examination of thestalk at the base of the polyp, usually requiringmultiple sections. For formal excisions it is importantto assess the whole of the deep resection plane, andfor the pathologist to be able to do this adequately thesurgeon should pin the specimen out on a cork matbefore fixation, so that multiple properly orientatedblocks can be taken for histological examination.

When invasive malignancy is identified in apolypectomy or formal excision specimen, more radicalsurgery is indicated if:-

- there is doubt about completeness of excision of thecarcinoma

- there is invasion of muscularis propria

- the invasive tumour is poorly differentiated (criteriaof Morson 1985. IIb)

The pathology report of locally resected carcinomamust therefore make specific mention of each of these parameters.

There is considerable evidence to suggest thatlymphatic or vascular invasion in the submucosa(including the polyp stalk) is also an indication forfurther surgery (Coverlizza et al 1989. IIb), but this hasnot been confirmed in other studies (Geraghty et al1991. IIb).

There is also some controversy over the managementof locally excised pT1 tumours in which the carcinomainvades the full thickness of the submucosa (so-calledKikuchi type Sm3 tumours) as some authorities alsoregard this as an indication for further surgery(Kikuchi et al 1995. IIb).

ii. Full Resection Specimens

It is important to know if a tumour has beencompletely excised and how advanced it is, as both ofthese parameters may affect further treatment. In orderto provide this information there must be properfixation of the specimen and careful pathologicaldissection prior to selecting tissue blocks for histology.In order to assess the circumferential resection marginin rectal cancers, it is strongly recommended that thismargin is painted with ink before sectioning and slicingat 3 – 4mm transversely through the whole of thetumour and the entire mesorectum distally andproximally. Dukes’ staging requires the separateidentification of the “apical” lymph node, i.e. the nodeclosest to the main vascular ligature.

Pathology reports should contain information on all ofthe data items contained in the Joint NationalGuidelines Minimum Data Set for Colorectal CancerHistopathology Reports as set out below.

Joint National Guidelines Minimum Dataset for Colorectal Cancer Histopathology ReportingThese proposals for reporting of colorectal cancershould be implemented for the following reasons:

1. Patients who have lymph node involvement(Dukes’ stages C1 and C2, TMN stages pN1,pN2) are likely to receive adjuvant chemotherapywhich is of possible benefit but is somewhat toxicand expensive (Moertel et al 1995. Ib, IMPACT1995. Ib).

2. Patients with rectal adenocarcinoma andcircumferential margin involvement are at highrisk of local recurrence (Quirke et al 1986. IIb,

Colorectal Cancer Management Clinical Guidelines 33Appendix 1

Adam et al 1994. IIb, Ng et al 1993. IIb) and mayreceive post-operative radiotherapy with orwithout chemotherapy which is toxic and costlybut may decrease the likelihood (MRC RectalCancer Working Party 1996. Ib, Thomas &Linbald 1988. Ib) of death from this unpleasantand nearly uniformly fatal complication. Thefrequency of circumferential margin involvementfound may reflect the quality of the surgicaloperation (Quirke 1997. IIb).

3. To confirm that radical surgery was necessary andto place the patient in the appropriate stage sothat the individual can be given a prognosis. Inaddition surgeons can accurately audit theiroutcomes while avoiding case-mix selection bias.

4. To identify whether the anal sphincter has beenlost. The frequency of abdomino-perinealresections is an indicator of the quality of surgery.

5. To allow the equitable comparison of surgeons incolorectal cancer audits, (McArdle & Hole 1991.IIb, Hermanek et al 1995. IIb) to identify goodsurgical practice (Quirke 1997. IIb) and to allowcomparison of patients in clinical trials.

The form reproduced on the next page has beendevised to include the minimum amount of datarequired for a careful assessment of a colorectalcancer specimen. It is evidence-based and has beenwidely discussed. It has been approved by the RoyalCollege of Pathologists and Surgeons (England), theAssociations of Coloproctology and ClinicalPathologists, the United Kingdom Co-ordinatingCommittee for Cancer Research Colorectal CancerSubcommittee, the Scottish IntercollegiateGuidelines Network, the Welsh CROPS Project andthe UK Association of Cancer Registries.


JOINT NATIONAL GUIDELINES MINIMUM DATA SET FOR COLORECTAL CANCERHISTOPATHOLOGY REPORT

Surname ________________________________________________________________ Forenames __________________________________________________________ Sex____________________________________________________________________________

Hospital__________________________________________________________________ Hospital Number ____________________________________________________________________________________________________________________________________

Date of receipt____________________________________________________ Date of reporting ____________________________________________ Report Number ________________________________________________

Pathologist____________________________________________________________ Surgeon __________________________________________________________________________________________________________________________________________________________

Gross Description Metastatic Spread

Site of tumour Number of lymph nodes examinedMaximum tumour diameter Number of positive lymph nodesDistance tumour to nearer margin (cut end) (pN1 1-3 nodes, pN2 4+ nodes involved)Presence of tumour perforation (pT4) Yes No

For rectal tumours Apical node positive (Dukes’ C2) Yes NoTumour is Above At Below Extramural vascular invasion Yes Nothe peritoneal reflectionDistance from the dentate line ............................... (mms)

Histology type Background abnormalities

Adenocarcinoma ____________________ Yes No Adenoma(s) Yes No(to include mucinous and signet ring adenocarcinomas) Synchronous carcinoma(s) Yes NoIf No, specify ________________________________________________________________________________________________ (Complete a separate form for each cancer)

Differentiation by predominant area Ulcerative colitis Yes NoWell/moderate Poor Crohn’s disease Yes No

Familial adenomatous polyposis Yes No

Other Comments..............................................................

Local invasion Pathological stagingSubmucosa (pT1) Complete resection at all margins Yes NoMuscularis propria (pT2)Beyond muscularis propria (pT3) TNM T N MTumour cells have breached the peritoneal surface

or invaded adjacent organs(pT4)

Margins Dukes’Tumour involvement N/A Yes No Dukes’ A (growth limited to wall, nodes negative)Doughnut Dukes’ B (growth beyond muscularis propria,Margin (cut end) nodes negative)Circumferential Dukes’ C1 (nodes positive and apical node negative)margin involvement Dukes’ C (apical node positive)

Histological measurement from tumour to Histologically confirmed circumferential margin .......................................... (mms) liver metastases Yes No

Signature ____________________________________________________________________________________________________________________________________________________________________________________________ Date____________________________________

Appendix 1

Colorectal Cancer Management Clinical Guidelines 35Appendix 1

NOTES FOR PATHOLOGISTS FOR RECORDINGData items for all primary colorectal cancers should berecorded, with all measurements in mm.

GROSS DESCRIPTION

Site of TumourThis will usually be stated on the request form.However, if examination of the specimen suggests thatthe stated site is incorrect this should be queried withthe surgeon and corrected if necessary.

Maximum Tumour DiameterMeasured from the luminal aspect of the bowel. The thickness of the tumour is ignored for thismeasurement.

Distance of Tumour to Nearest MarginMeasured from the nearest cut end of the specimen,not the circumferential margin. It is only necessary to examine the margins histologically if tumourextends macroscopically to within 30mm of one ofthese. For tumours further than this it can be assumedthat the cut ends are not involved. Exceptions to thisrecommendation are adenocarcinomas that are foundon subsequent histology to have an exceptionallyinfiltrative growth pattern, show extensive vascular or lymphatic permeation, or are pure signet ringcarcinomas, small cell carcinomas or undifferentiated carcinomas.

Presence of Tumour PerforationIf the tumour has perforated into the peritoneal cavitythis should be recorded. Such cases are alwaysregarded as pT4 in the TNM staging system. Ifperforation does not involve the tumour the “No” box should be marked.

FOR RECTAL TUMOURS

Relationship to the Peritoneal ReflectionThe crucial landmark for recording the site of rectaltumours is the peritoneal reflection. This is identifiedfrom the exterior surface of the anterior aspect of thespecimen.

Rectal tumours are classified according to whether they are:-

a. entirely above the level of the peritoneal reflection anteriorly

b. astride (or at) the level of the peritoneal reflection anteriorly

c. entirely below the level of the peritoneal reflection anteriorly

Tumours below the peritoneal reflection have thehighest rate of local recurrence.

Distance from Dentate LineThis can only be measured for low rectal tumours inabdomino-perineal excision of rectum (APER)specimens. This measurement is important as itidentifies patients who have lost their internal sphincter.

HISTOLOGY

TypeVirtually all colorectal cancers are adenocarcinomas.Other rare forms worthy of special mention are:

■ Adenosquamous carcinomas

■ True squamous carcinomas (not includingupwardly spreading tumours)

■ Adenocarcinoid (composite carcinoma/carcinoid)tumours

■ Small cell carcinomas

■ Totally undifferentiated carcinomas

Mucinous carcinomas and signet ring carcinomas arerecorded as adenocarcinomas.

Differentiation by Predominant AreaPoorly-differentiated carcinomas should be separatedfrom other types, but only if this forms thepredominant area of the tumour. Small foci of apparentpoor differentiation are not uncommon at theadvancing edge of tumours, but these are insufficient to classify the tumour as poorly-differentiated.

The criteria for poorly-differentiated tumours are eitherirregularly folded, distorted and often small tubules or the absence of any tubular formation.


Local InvasionThe maximum degree of local invasion into or throughthe bowel wall is recorded, so only one of the fourboxes should be marked.

Sufficient blocks of the tumour should be taken toassess local invasion. It is recommended that the wholetumour and attached mesentery (or mesorectum) areserially sliced at 3-4mm intervals with a sharp knife inorder to identify the areas of deepest invasionmacroscopically, which should be blocked forhistological confirmation.

Involvement of the serosal (peritoneal) surface isdefined as the presence of tumour cells on theperitoneal surface. Thus tumour cell penetration of theserosa needs to be seen by penetration or ulceration.This does not constitute circumferential margininvolvement since there is no involvement of aretroperitoneal margin.

Margins

Tumour Involvement

DoughnutsIt is not necessary to examine doughnuts histologicallyif the main tumour is less than 30mm from the cutend of the main specimen or in other rare casesdescribed above but this is a decision to be made inaccordance with local practice.

When doughnuts from stapling devices are examinedhistologically the presence or absence of tumour isrecorded. If doughnuts are not sectioned because it isdeemed unnecessary, or if no doughnuts are submittedfor examination by the surgeon, this item should berecorded as “not applicable”.

Margin (cut end)When cut ends are examined histologically (see criteriaabove) the presence or absence of tumour should berecorded. If margins are not examined histologicallythey should be recorded as “not applicable”.

Circumferential Margin (rectal cancers only)Accurate assessment of the circumferential (radial)margins of rectal tumours is most important because itinfluences post-operative therapy.

The circumferential margin is reported only for rectalcancers; for tumours at other sites the “not applicable”box is marked. It represents involvement of the surgicalmargins of the connective tissues around the rectum inan area where there is no peritoneal covering. Henceinvolvement of this margin is different from, and quiteunrelated to serosal involvement.

Anteriorly the rectum is covered by peritoneum andonly the area below the peritoneal reflection is at risk ofcircumferential margin involvement. Posteriorly thisarea and the area above it, a triangular-shaped barearea running up to the start of the sigmoid mescolon,are at risk from not only direct tumour spread but alsometastatic deposits in lymph nodes that lie against thecircumferential margin.

It is recommended that the whole of this margin (i.e.the mesorectum) is painted with a marker such assilver nitrate or India ink before dissecting thespecimen. The tumour is then best sliced serially at 3-4mm intervals to select blocks from areas that areclosest macroscopically to the circumferential margin.Slices should then be made of the area above andbelow the tumour to look for metastatic deposits. Iflymph nodes lie against the circumferential marginthen this margin should be included in the block.

The minimum distance between the tumour and thecircumferential margin in millimetres is also recordedfrom the histological slides. If this is less than 1mmthen the circumferential margin is regarded as involvedin the assessment on completeness of resection later inthe proforma. Such involvement may be through directcontinuity with the main tumour, by tumour in veins,lymphatics or lymph nodes, or by tumour depositsdiscontinuous from the main growth.

Metastatic Spread

Number of lymph nodes examinedAll lymph nodes found in the specimen should besampled and counted, regardless of their site or size.

Number of positive lymph nodesThis must be equal to or less than the number oflymph nodes sampled.

Extramural tumour deposits measuring more than3mm are counted as involved lymph nodes even if noresidual lymph node structure can be identified.Smaller deposits are regarded as apparentdiscontinuous extensions of the main tumour.

In the TNM staging system, pN1 corresponds toinvolvement of 1-3 nodes and pN2 to involvement of 4 or more nodes (A previously used pN3 category wasdropped in the 1997 TNM revision).

Appendix 1


Apical node positiveFor Dukes’ staging the pathologist will only need toidentify separately the apical lymph node closest to themain vascular tie. This is not defined by any measure ofdistance, but is simply the first node identified by slicingthe mesentery serially and distally from the vascular tie.

Extramural vascular invasionThis is recorded when tumour is present within anextramural endothelium-lined space that is eithersurrounded by a rim of muscle or contains red blood cells.

Background abnormalitiesThe presence or absence of the following in thebackground bowel is recorded

■ Adenoma(s)

■ Synchronous carcinoma(s) (each of which willrequire a separate proforma)

■ Ulcerative colitis

■ Crohn’s disease

■ Familial adenomatous polyposis

PATHOLOGICAL STAGINGIt is recommended that Dukes’ and TNM staging isused. The proforma is designed for both systems.

Complete resection at all marginsThis includes the doughnuts, the ends of the specimenand, for rectal tumours, the mesorectal circumferentialresection plane.

Where doughnuts and the ends of the specimen are notexamined histologically because the tumour is morethan 30mm away these are assumed to be tumour-free.Circumferential margins of rectal tumours are regardedas involved if tumour extends histologically to within1mm of this margin.

Peritoneal (serosal) involvement alone is not a reasonto categorise the tumour as incompletely excised.

TNMHere the T stage and the N stage are derived from theextent of local spread and lymph node metastases, thecriteria for each stage being defined on the form. Theappropriate figure is entered in each box. The prefixP is used to indicate pathological staging. If thepatient has had pre-operative chemotherapy orradiotherapy then the prefix YP should be used toindicate that the stage found may not be thepresenting stage of the tumour.

The following should be noted:

i. In determining the pT stage, tumours that haveperforated into the peritoneal cavity are regardedas pT4, irrespective of other factors.

ii. Direct intramural spread of caecal carcinomas intothe terminal ileum does not affect the pT stage.However direct extramural spread (across theserosa) of a colorectal carcinoma into another partof the large or small intestine corresponds to pT4.

iii. Extramural deposits of tumour that are notobviously within lymph nodes are regarded asdiscontinuous extensions of the main tumour if theymeasure less than 3mm in diameter, but as lymphnodes if they measure more than 3mm in diameter.

iv. The difference between stages pN1 and pN2 is thenumber of lymph nodes involved (pN1 = 1-3nodes, pN2 = 4+ nodes), irrespective of their sitein the resection specimen.

v. Pre-operative radiotherapy (including short course)diminishes lymph node yield and downstagestumours. Identification of such tumours isessential when comparing outcomes. Thepathological staging of these tumours can beidentified by insertion of a y prefix (for example a“pT3” tumour becomes a “ypT3” tumour) toindicate that this tumour has received per-operative irradiation (TNM classification ofMalignant Tumours 5th edition. III).

vi. Pathological M staging can only be based ondistant metastases that are submitted for histologyby the surgeon and will therefore tend tounderestimate the true M stage. Pathologists willtherefore only be able to use M1 (distantmetastases present) or MX (distant metastasesunknown). Note that metastatic deposits in lymphnodes distant from those surrounding the maintumour or its main artery in the specimen, whichwill usually be submitted separately by thesurgeon (e.g. in para-aortic nodes or nodessurrounding the external iliac or common iliacarteries), are counted as distant metastases andhence pM1.

Dukes’Here one of the four boxes is marked, correspondingto the Dukes’ stage. Criteria used for Dukes’ stagingare given on the form. Note that Dukes’ so-called stageD is not used.

Appendix 1


Histologically confirmed liver metastasesHere one of the two boxes is marked. If no liver biopsyis submitted with the resection specimen the “No” boxshould be marked.

RecommendationsAll resected colorectal tumours should be submitted forhistopathological examination, which should reachacceptable quality standards as outlined above.

Grade B

Pathology reports should contain information on allthe data items contained in the Joint NationalGuidelines Minimum Data Set for Colorectal CancerHistopathology Reports.

Grade C

Pathology laboratories should store stained histologyslides for a minimum of 10 years and tissue blocksfrom specimens indefinitely in order to facilitate futurecase review, clinical audit and research.

Grade B

Pathological examination of colorectal cancerspecimens should be carried out in laboratories whichperform to high technical standards and whichparticipate in external quality assessment schemes andregular audit of technical procedures and diagnosis.

Grade B

Appendix 1


1. Abulafi AM, Williams NS. Local recurrence of colorectalcancer: the problem, mechanisms, management and adjuvanttherapy. Br J Surg 1994; 81: 7-19.

2. ACCMP (Advanced Colorectal Cancer Meta-analysis Project).Modulation of 5-fluorouracil in patients with advancedcolorectal cancer: evidence in terms of response rates. J ClinOncol 1991; 10: 896-903.

3. Adam IJ, Mohamdee MO, Martin IG, et al. Role ofcircumferential margin involvement in the local recurrence ofrectal cancer. Lancet 1994; 344: 707-11.

4. Allen-Mersh TG, Earlam S, Fordy C, et al. Quality of life andsurvival with continuous hepatic-artery floxuridine infusion forcolorectal liver metastases. Lancet 1994; 344: 1255-60.

5. Allum WH, Slaney G, McConkey CC, Powell J. Cancer of thecolon and rectum in the West Midlands, 1957-1981. Br J Surg1994; 81: 1060-63.

6. Baigrie RJ, Berry AR. Management of advanced rectal cancer.Br J Surg 1994; 81: 343-52.

7. Barlow AP, Thompson MH. Colonoscopic follow-up afterresection for colorectal cancer : a selective policy. Brit J Surg1993; 80: 781-84.

8. Barratt PL, Seymour MT, Stenning S. Molecular markers topredict survival and benefit from adjuvant intraportal 5FU incolon cancer. Abs. 1030 Proc ASCO 1999.

9. Beart RW, Kelly KA. Randomised prospective evaluation of theEEA stapler for colorectal anastomoses. Am J Surg 1981; 141:143-47.

10. Beets-Tan RGH, Beets GL, Vliegen RFA et al. Accuracy ofmagnetic resonance imaging in prediction of tumour-freeresection margin in rectal cancer. Lancet 2001; 357: 497-504.

11. Bentzen SM, Balsev I, Pedersen M et al. Blood transfusion andprognosis in Dukes’ B and C colorectal cancer. Eur J Cancer1990; 26: 457-63.

12. Beretta G, Bollina R, Martignoni M et al. Fluorouracil +folates as standard treatment for advanced/metastaticgastrointestinal carcinomas. Ann Oncol 1994; 5: 239.

13. Bohm B, Schwenk W, Hucke HP et al. Does methodic long-term follow-up affect survival after curative resection ofcolorectal carcinoma? Dis Colon Rectum 1993; 36 (3):280-6.

14. Brady PG, Straker RJ, Goldschmid S. Surveillance ColonoscopyAfter Resection for Colon Carcinoma. Southern Medical J1990; 83: 765-68.

15. Brennan SS, Pickford IR, Evans M, Pollok AV. Staples orsutures for colonic anastomoses – a controlled clinical trial. BrJ Surg 1982; 69: 722-724.

16. Brierley JD, Cummings BJ, Wong CS et al. Adenocarcinoma ofthe rectum treated by external beam radiation therapy. Int JRadiat Oncol Biol Phys 1995; 31: 255-59.

17. Brown G, Richards CJ, Newcombe RG, et al. RectalCarcinoma: Thin-Section MR Imaging for Staging in 28patients. Radiology 1999; 211: 215-22.

18. Brown G, Richards CJ, Bourne MW et al. Rectal carcinoma:thin section MR imaging for staging in 28 patients. Radiology1999; 357: 497-504.

19. Brown G, Richards CJ, Bourne MW et al. Discriminantanalysis of thin slice MR imaging characteristics in detectinglocal lymph node involvement in patients with rectalcarcinoma. Radiology 1999a; 213: 299.

20. Bruinvels DJ, Stiggelbout AM, Kievit J. Follow-Up of Patientswith Colorectal Cancer. A Meta-Analysis. Ann Surg 1994; 291:174-82.

21. Bull AD, Biffin AHB, Mella J et al. Colorectal cancer pathologyreporting: a regional audit. J Clin Pathol 1997; 50: 138-42.

22. Burke P, Mealy K, Gillen P, et al Requirement for bowelpreparation in colorectal surgery. Br J Surgery 1994; 81: 907-10.

23. Burrows L, Tartter P. Effect of blood transfusions on colonicmalignancy recurrence rate. Lancet 1982; ii: 662.

24. Busch ORC, Hop WCJ, Hoynck van Pependrecht MAW, Bloodtransfusions and prognosis in colorectal cancer. N Engl J Med 1993; 328: 1372-76.

25. Camunas J, Enriquez JM, Devesa JM, Morales V, Value offollow-up in the management of recurrent colorectal cancer.Eur J Surg Oncol 1991; 17: 530-35.

26. Carmichael J, Popiela T, Radstone D et al. Randomisedcomparative study of ORZEL plus leucovorin versus parenteral5FU plus LV in patients with metastatic colorectal cancer. ProcAm Soc Clin Oncol 1999; 18:1015.

27. Carty NJ, Keating J, Campbell J, Karanjia N, Heald RJ.Prospective audit of an extramucosal technique for intestinalanastomosis. Br J Surg 1991; 78: 1439-41.

28. Cocconi G, Cunningham D, van Cutzem E et al. Openrandomised multicentre trial of raltitrexed versus fluorouracilplus high-dose leucovorin in patients with advanced colorectalcancer. J Clin Oncol 1998; 16: 2943-52.

29. Cochrane JP, Williams JT, Faber RG, et al. Value of outpatient follow-up after curative surgery for carcinoma of the largebowel. Br Med J 1980;280 (6214):593-5.

30. Collins R, Scrimgeour A, Yusuf S, Peto R. Reduction in fatalpulmonary embolism and venous thrombosis by perioperativeadministration of subcutaneous heparin. N Engl J Med 1988;318: 1162 –73.

31. Cotton PB, Williams CB. Practical Gastrointestinal Endoscopy3rd Ed Blackwell Scientific Publications 1990.

References


32. Coverlizza S, Risio M, Ferrari A, et al. Coverlizza S, Risio M,Ferrari A, Genologio – Preiser CM, Rossini FP ColorectalAdenomas containing invasive carcinoma Pathologicassessment of lymph node metatasic potential. Cancer 1989;64(9):1937-47.

33. Cox JV, Pazdur R, Thibault A et al. A phase III trial of Xeloda(Capecitabine) in previously untreated advanced/metastaticcolorectal cancer. Proc Am Soc Clin Oncol 1999; 18: 1016.

34. CRC (Cancer Research Campaign). Cancer ResearchCampaign. Cancer Stats: Large Bowel-UK. November 1999.

35. Cross SS, Feeley KM, Angel CA. The effect of fourinterventions on the informational content of histopathologyreports of resected colorectal carcinomas. J Clin Pathol 1998;51: 481-2.

36. Cunningham D, Zalcberg JR, Rath U, Olver I, Van Cutsem E,Svensson C, Seitz FJ, Harper P, Kerr D, Perez-Manga G & the‘Tomudex ‘ Colorectal Cancer Study Group. Final results of arandomised trial comparing “Tomudex”® (raltitrexed) with 5-fluorouracil plus leucovorin in advanced colorectal cancer.Annals Oncol 1996; 7: 961-5.

37. Cunningham D, Pyrhonen S, James RD et al. Randomised trialof irinotecan plus supportive care versus supportive care aloneafter fluorouracil failure for patients with metastatic colorectalcancer. Lancet. 1998; 352(9138): 1413-1418.

38. Curley SA, Izzo F, Delrio P et al. Radiofrequency ablation ofunresectable primary and secondary hepatic malignancies. AnnSurg 1999; 230: 1-8.

39. Deans GT, Krukowski ZH, Irwin ST. Malignant obstruction ofthe left colon. Br J Surg 1994; 81: (9)1270-6.

40. de Gramont A, Figer A, Seymour M et al. Leucovorin andfluorouracil with or without exaliplatin as first-line treatmentin advanced colorectal cancer. J Clin Oncol. 2000; 18(16):2938-2.

41. Devlin HB. Stoma Therapy Review. Coloproctology 1982.

42. Dixon AR, Maxwell WA, Thornton Holmes J. Carcinoma ofthe rectum: a 10-year experience. Br J Surg 1991; 78: 308-11.

43. Douillard JY, Cunningham D, Roth AD et al. Irinotecancombined with fluorouracil compared with fluorouracil aloneas first-line treatment for metastatic colorectal cancer: amulticentre randomised trial. Lancet 2000; 355:1041-47.

44. Dorudi S, Wilson NM, Heddle RM. Primary restorativecolectomy in malignant left-sided large bowel obstruction. AnnRoy Coll Surg Eng 1990; 72: 393-5.

45. Dube S, Heyen F, Jenicek M. Adjuvant chemotherapy incolorectal carcinoma: results of a meta-analysis. Diseases of theColon & Rectum. 1997; 40: 35-41.

46. Dworak O. Morphology of lymph nodes in the resected rectumof patients with rectal carcinoma. Path Res Pract 1991; 187:1020-24.

47. Everett WG, Friend PJ, Forty J. Comparison of stapling andhand suture for left-sided large bowel anastomosis. Br J Surg1986; 73: 345-8.

48. Fielding LP, Stewart-Brown S, Blesovsky L, Kearney G.Anastomotic integrity after operations for large bowel cancer: amulti-centre study. Br Med J 1980; 281: 411- 4.

49. Fontzilas G, Gossios K, Zisiadis A et al. Prognostic variable inpatients with advanced colorectal cancer treated withfluorouracil and leucovorin-based chemotherapy. Medical andPaediatric Oncol 1996; 26: 305-17.

50. Geraghty JM, Williams CB, Talbot IC. Malignant colorectalpolyps: venous invasion and successful treatment by endoscopicpolypectomy. Gut 1991;32(7):774-8.

51. Giacchetti S, Itzhaki M, Gruia G, Adam R, Zidani R,Kunstlinger F, Brienza S, Alafaci E, Bertheault-Cvitkovic F,Jasmin C, Reynes M, Bismuth H, Misset JL, Levi F. Long-termsurvival of patients with unresectable colorectal cancer livermetastases following infusional chemotherapy with 5-fluorouracil, leucovorin, oxaliplatin and surgery. Annals Oncol.1999: 10(6): 663-9.

52. GIVIO Investigators: Impact of follow-up testing on survivaland health-related quality of life in breast cancer patients: amulticentre randomized controlled trial. JAMA 1994; 271:1587-92.

53. Gooszen AW, Geelkerken RH, Hermans J, Lagaay MB,Gooszen HG. Quality of life with a temporary stoma. Dis.Colon Rectum. 2000; 43: 650-5.

54. Graham RA, Garnsey L, Jessup JM. Local excision of rectalcarcinoma. Am J Surg 1990; 160: 306-12.

55. Haller DJ, Catalano PJ, MacDonald RJ et al. Fluorouracil (FU)leucovorin (LV) and levamisole (LEV) adjuvant therapy forcolon cancer: Five year final report of Int-0089. Proc Am SocOncol 17: 256, 1998 (abstr).

56. Harrison S, Steele RJ, Johnston AK, et al. Predepositautologous blood transfusion in patients with colorectal cancer:a feasibility study. Br J Surg 1992;79(4):355-7.

57. Heald RJ. The “holy plane” of rectal surgery. J R Soc Med1988; 81: 503-8.

58. Heald RJ, Husband EM, Ryall RDH. The mesorectum in rectalcancer surgery – the clue to pelvic recurrence? Br J Surg 1982;69: 613-16.

59. Heald RJ, Lockhart-Mummery HE. The lesion of the secondcancer of the large bowel. Br J Surg 1972; 59: 16-9.

60. Heald RJ, Ryall RDH. Recurrence and survival after totalmesorectal excision for rectal cancer. Lancet 1986; I: 1479-82.

61. Heriot AG, Grundy A, Kumar D. Pre-operative staging ofrectal carcinoma. Br J Surg 1999; 86:17-28.

References

Colorectal Cancer Management Clinical Guidelines 41References

62. Hermanek P, Wiebelt H, Staimmer D, Riedi S and the GermanStudy Group for Colorectal Cancer. Prognostic factors of rectalcancer – experience of the German multicentre study SGCRC.Tumori 1995; 81: 60-4.

63. Holm T, Rutqvist LE, Johansson H, Cedermark B.Abdominoperineal resection and anterior resection in thetreatment of rectal cancer: results in relation to adjuvant pre-operative radiotherapy. Br J Surg 1995; 82: 1213-16.

64. Hulton NR, Hargreaves AW. Is long-term follow-up of allcolorectal cancer necessary? J R Coll Surg Edinb1989;34(1):21-4.

65. Hurst PA, Proust WG, Kelly JM, Bannister JJ, Walker RT.Local recurrence after low anterior resection using the staplegun. Br J Surg 1982; 69: 275-6.

66. IMPACT (International Multicentre Pooled Analysis of Coloncancer Trials) investigators. Efficacy of adjuvant fluorouraciland folinic acid in colon cancer. Lancet 1995; 345: 939-44.

67. IMPACT B2 (International Multicentre Pooled Analysis of B2Colon cancer Trials) investigators. Efficacy of adjuvantfluorouracil and folinic acid in B2 colon cancer. J Clin Oncol17: 1356-63, 1999.

68. Irving AD, Scrimgeour D. Mechanical bowel preparation forcolon resection and anastomosis. Br J Surg 1987; 74: 580-1.

69. James RD et al AXIS Proc Am Soc Clin Oncol. 1999.

70. Juhl G, Larson GM, Mullins R, Bond S, Polk HC. Six-YearResults of annual colonoscopy after resection of colorectalcancer. World J Surg 1990; 14: 255-61.

71. Kakkar VV, Cohen AT, Edmonson RA et al. Low molecularweight versus standard heparin for prevention of venousthromboembolism after major abdominal surgery. Lancet 1993;341: 259-65.

72. Kapiteijn E, Marijnen CA, Nagtegaal ID et al. Pre-operativeradiotherapy combined with total mesorectal excision forresectable rectal cancer. N Engl J Med 2001; 345: 638-46.

73. Karanjia ND, Corder AP, Bearn P, Heald RJ. Leakage fromstapled low anastomosis after total mesorectal excision forcarcinoma of the rectum. Br J Surg 1994; 81: 1224-6.

74. Karanjia ND, Schache DJ, North WRS, Heald RJ. “Closeshave” in anterior resection. Br J Surg 1990; 77: 510-2.

75. Kiebert GM, Welvaart K, Kievit J. Psychological effects ofroutine follow-up on cancer patients after surgery. Eur J Surg1993; 159: 601-7.

76. Kikuchi R, Takano M, Takagi K et al. Management of earlyinvasive colorectal cancer. Risk of recurrence and clinicalguidelines. Dis Colon Rectum 1995; 38: 1286-95.

77. Kronborg O, Hage E, Deichgraeber E. The remaining colonafter radical surgery for colorectal cancer. The first three yearsof a prospective study. Dis Colon Rectum 1983; 26: 172-6.

78. Koruth NM, Koruth A, Matheson NA. The place of contrastenema in the management of large bowel obstruction. J RoyColl Surg Edin 1985; 30: 258-60.

79. Koruth NM, Krukowski ZH, Youngson GG, Hendry WS,Logie JRC, Jones PF, Munro A. Intraoperative colonicirrigation in the management of left-sided large bowelemergencies. Br J Surg 1985; 72: 708-11.

80. Krook JE, Moertel CG, Gunderson LL et al. Effective surgicaladjuvant therapy for high-risk rectal carcinoma. N Engl J Med1991; 324: 709-15.

81. Kumar A, Scholefield JH. Endosonography of the Anal Canaland Rectum. World J Surg 2000; 24:208-15.

82. Lauer JD, Carlson HC, Wollaeger EE. Accuracy ofroentgenologic examination in detecting carcinoma of thecolon. Dis Colon Rectum 1965; 8: 190.

83. Laufer I. Tumours of the colon. Double ContrastGastrointestinal Radiology with Endoscopic Correlation.Laufer I (ed) Philadelphia: Saunders 1979.

84. Liver Infusion Meta-analysis Group. Portal vein Chemotherapyfor colorectal cancer: a meta-analysis of 4000 patients in 10studies. J Natl Cancer Inst 1997; 89: 497-505.

85. Mamounas E, Wieand S, Wolmark N et al. Comparativeefficacy of adjuvant chemotherapy in patients with Dukes’ Bversus Dukes’ C colon cancer: results from four nationalsurgical Adjuvant Breast and Bowel Project adjuvant studies(C-01, C-02, C-03 and C-04). J Clin Oncol 1999; 17: 1349-55.

86. Marsh PJ, James RD, Schofield PF. On behalf of the NorthwestRectal Cancer Group. Adjuvant pre-operative radiotherapy forlocally advanced rectal carcinoma: results of a prospectiverandomised trial. Dis Colon Rectum 1994; 37: 1205-14.

87. Martin EW Jr, Minton JP, Carey LC. CEA-directed second-looksurgery in the asymptomatic patient after primary resection ofcolorectal carcinoma. Ann Surg 1985;202(3):310-7.

88. Martling Al, Holm T, Rutqvist LE, Moran BJ, Heald RJ,Cedemark B. Effect of a surgical training programme onoutcome of rectal cancer in the county of Stockholm. Lancet2000; 356:93-6.

89. Matheson NA, McIntosh CA, Krukowski ZH. Continuingexperience with single layer appositional anastomosis in thelarge bowel. Br J Surg 1985; 70: S104-6.

90. Maughan TS, James RD, Kerr D et al. Preliminary results of amulticentre randomised trial comparing 3 chemotherapyregimens in metastatic colorectal cancer. Proc Am Soc ClinOncol 1999; 18: 1007.

91. Maughan TS, James RD, Kerr D et al. Continuous versusintermittent chemotherapy for advanced colorectal cancer:preliminary results of MRC CR06b randomized trial. Proc AmSoc Clin Oncol 2001; 20: 125a.

Colorectal Cancer Management Clinical Guidelines

92. Medical Research Council Rectal Cancer Working Party.Randomised trial of surgery alone versus surgery followed byradiotherapy for mobile cancer of the rectum. Lancet 1996;348: 1610-14.

93. Meta-analysis Group in Cancer. Efficacy of intravenouscontinuous infusion of fluorouracil compared with bolusadministration in advanced colorectal cancer. Journal ofClinical Oncology. 1998; 16(1):301-308.

94. Moertel CG, Fleming TR, Macdonald JS et al. Fluorourcil andlevamisole as effective adjuvant therapy after resection of stageIII colon carcinoma: a final report. Ann Intern Med 1995; 122:321-326.

95. Moertel CG, Fleming TR, Macdonald JS, Haller DG, LaurieJA, Tangen C. An evaluation of the carcinoembryonic antigen(CEA) test for monitoring patients with resected colon cancer.JAMA 1993; 270: 943-947.

96. Moore HCF & Haller DG. Adjuvant therapy of colon cancer.Semin Oncol 1999; 26: 545-555.

97. Morris S, Corder AP, Taylor I. What are the benefits of routinebreast cancer follow-up? Postgrad Med J 1992; 68: 904-7.

98. Morson BC, Vaughan EG, Bussey HJR. Pelvic recurrence afterexcision of rectum for carcinoma. Br Med J 1963; ii: 13-18.

99. Mortensen NJM, Ramirez JM, Takeuchi N, SmilginHumphreys MM. Colonic J pouch-anal anastomosis afterrectal excision for carcinoma: functional outcome. Br J Surg1995; 82: 611-613.

100. McArdle CS, Hole D. Impact of variability among surgeons onpostoperative morbidity, mortality and ultimate survival. BrMed J 1991; 302: 1501-1505.

101. McArdle CS, Hole D, Hansell D, Blumgart LH, Wood CB.Prospective study of colorectal cancer in the west of Scotland:ten-year follow-up. Br J Surg 1990; 77: 280-282.

102. McGinn FP, Gartell PC, Clifford PC, Brunton FJ. Staples orsutures for low colorectal anastomoses: a prospectiverandomised trial. Br J Surg 1985; 72: 603-605.

103. Neville R, Fielding LP, Amendola C. Local tumour recurrenceafter curative resection of rectal cancer. A ten-hospital review.Dis Colon Rectum 1987; 30: 12-17.

104. NGTATG (Nordic Gastrointestinal Tumour AdjuvantherapyGroup). Expectancy or primary chemotherapy in patients withadvanced asymptomatic colorectal cancer: a randomized trial. JClin Oncol 1992; 10: 904-911.

105. NHS Executive. Guidance on Commissioning Cancer Services.Improving outcomes on Colorectal Cancer, The Manual. 1997,Dept of Health, London (97CV0119).

106. NHS Executive. Guidance on Commissioning Cancer Services.Improving outcomes on Colorectal Cancer, The ResearchEvidence. 1997, Dept of Health, London (97CV0120)

107. NHS Executive. Recurrent and advanced disease. In: Improvingoutcomes in colorectal cancer: The research evidence. 1997, CCO1200, p98-125.

108. NHS Executive. NHS Cancer Plan, A plan for investment , Aplan for reform, London. September 2000.

109. Nicholls J. Large bowel cancer. In: Slevin ML, Staquet MJ, eds.Randomised Trials in Cancer: A Critical Review by Sites. NewYork: Raven Press, 1986: 241-271.

110. Norum J. Adjuvant chemotherapy in Dukes’ B and C colorectalcancer has only a minor influence on psychological distress.Supportive Care in Cancer. 1997; 5 318-21.

111. O’Connell MJ, Martenson JA, Weiand HS et al. Improvingadjuvant therapy for rectal cancer by combining protractedinfusional fluorouracil with radiation therapy after curativesurgery. N Engl J Med 1994; 331: 502-7.

112. O’Connell MJ, Laurie JA, Kahn M et al. Prospectivelyrandomised trial of adjuvant chemotherapy in patients withhigh risk colon cancer. J Clin Oncol. 1998; 16: 295-300.

113. Ovaska J, Jarvinen H, Kujari H, Pertfila I, Mecklin JP. Follow-up of patients operated on for colorectal carcinoma. Am J Surg 1990; 159: 593-6.

114. Padhani AR. Advances in imaging of colorectal cancer. Clinical Reviews in Oncology/Hematology 1999; 30:189-99.

115. Page CP, Bohnen JMA, Fletcher JR, McManus AT, SolomkinJS, Wittmann DH. Antimicrobial prophylaxis for surgicalwounds. Guidelines for clinical care. Arch Surg 1993; 128: 79-88.

116. Pahlman L, Glimelius B. Pre-operative and post-operativeradiotherapy and rectal cancer. World J Surg 1992; 16: 858-65.

117. Patel SC, Tovee EB, Langer B. Twenty-five years experiencewith radical surgical treatment of carcinoma of theextraperitoneal rectum. Surgery 1977; 82: 460-5.

118. Pazdur R, Vincent M for the Tomudex Colorectal Study group:Raltitrexed versus 5-fluorouracil and leucovorin in patientswith advanced colorectal cancer: results of a randomizedmulticentre North American trial. Proc ASCO 1997; 1(16): 801(abstract).

119. Pazdur R, Douillard J-Y, Skilings JR et al. Multicentre phase IIIstudy of 5FU or UFT in combination with leucovorin inpatients with metastatic colorectal cancer. Proc Am Soc ClinOncol 1999; 18: 1009.

120. Persson AV, Davis RJ, Villavicencio JL. Deep VenousThrombosis and Pulmonary Embolism. Surg Clin N Amer1991; 71: 1195-9.

121. Phillips RKS, Hittinger R, Blesovsky L, Fry JS, Fielding LP.Local recurrence following “curative” surgery for large bowelcancer: I. The overall picture. Br J Surg 1984; 71: 12-6.

122. Phillips RKS, Hittinger R, Blesovsky L, Fry JS, Fielding LP.Local recurrence following “curative” surgery for large bowelcancer: II. The rectum and rectosigmoid. Br J Surg 1984; 71: 17-20.

42References

123. Pollard SG, Macfarlane R, Everett WG. Surgery for recurrentcolorectal carcinoma – is it worthwhile? Ann R Coll Surg Engl1989; 71: 293-8.

124. Puthwala AA, Syed AMN, Gates TC, McNamara C. Definitivetreatment of extensive anorectal carcinoma by external andinterstitial irradiation. Cancer 1982; 50: 1746-50.

125. QASAR colorectal cancer study group. Adjuvant chemotherapywith 5-fluorouracil, 1-folinic acid and levamisole for patientswith colorectal cancer: comparison of weekly versus monthlyschedules. Ann Oncol 2000; 11: 947-55.

126. Quirke P, Durdey P, Dixon MF, Williams NS. Local recurrenceof rectal adenocarcinoma due to inadequate surgical resection:histopathological study of lateral tumour spread and surgicalexcision. Lancet 1986; ii: 996-9.

127. Quirke P. Limitations of existing systems of staging for rectalcancer: The forgotten margin. Rectal Cancer Surgery. Ed.Soreide O & Norstein J. Springer-Verlag 1997, 63-81.

128. Raymond E, Chaney SG, Taamma A, Cvitkovic E. Oxaliplatin:a review of preclinical and clinical studies. Ann Oncol. 1998; 91053-1971.

129. Rodel C, Grabenhauer GG, Matzel KE et al. Extensive surgeryafter high-dose pre-operative chemoradiotherapy for locallyadvanced recurrent rectal cancer. Dis Colon Rectum 2000; 43, 312-9.

130. Rosen M, Chan L, Beart RW et al. Follow up of colorectalcancer: a meta-analysis. Dis Colon Rectum 1998; 41:1116-26.

131. Ross P, Norman A, Cunningham D et al. A prospectiverandomised trial of protracted venous infusion 5-fluorouracilwith or without mitomycin C in advanced colorectal cancer.Ann Oncol. 1997; 8: 995-1001.

132. Rossi S, Di Stasi M, Buscarini E et al. Percutaneous RFinsterstitial thermal ablation in the treatment of hepatic cancer.Am J Roentgenology 1996; 167 (3): 759-68.

133. Rougier P, Van Cutsem E, Bajetta E et al. Randomised trial ofirinotecan versus fluorouracil by continuous infusion afterfluorouracil failure in patients with metastatic colorectal.Lancet. 1998; 352(9138): 1407-1.

134. Rougier P, Sahmoud T, Nitti D et al. Adjuvant portal veininfusion of fluorouracil and heparin in colorectal cancer: arandomised trial. Lancet 1998; 351:1677-81.

135. Rowe-Jones DC, Peel ALG, Kingston RD, Shaw JFL, TeasdaleC, Cole DS. Single dose cefotaxime plus metronidazole versusthree dose cefotaxime plus metronidazole as prophylaxisagainst wound infection in colorectal surgery: multicentreprospective randomised study. Br Med J 1990; 300: 18-22.

136. Safi F, Link KH, Beger HG. Is follow-up of colorectal cancerpatients worthwhile? Dis Colon Rectum 1993; 36: 636-42.

137. Saltz LB, Cox JV, Blanke C, et al. Irinotecan plus fluorouraciland leucovorin for metastatic colorectal cancer. IrinotecanStudy Group. New England Journal of Medicine 2000; 343:905-914.

138. Salzman EW, Davies GC. Prophylaxis of VenousThromboembolism. Ann Surg 1980; 191: 207-218.

139. Saunders B. The nurse’s role in the care of patients with stoma.Br J Clin Pract 1976; 30: 81-2.

140. Scheele J, Stangl R, Altendorf-Hofmann A. Hepatic metastasesfrom colorectal carcinoma: impact of surgical resection on thenatural history. Br J Surg 1990; 77: 1241-46.

141. Scheithauer W, Rosen H, Kornek GV, Sebesta C, Depisch D.Randomised comparison of combination chemotherapy plussupportive care alone in patients with metastatic colorectalcancer. Br Med J 1993; 306: 752-5.

142. Schiessel R, Wunderlich M, Herbst F. Local recurrence ofcolorectal cancer: effect of early detection and aggressivesurgery. Br J Surg 1986; 73: 342-4.

143. SCOTIA Study Group. Sub-total or segmental resection formalignant left-sided colonic obstruction: a randomised clinicaltrial. Gut 1994; 35 (suppl 2): S16.

144. Scott N, Jackson P, Al-Jaberi T, Dixon MF, Quirke P, Finan PJ.Total mesorectal excision and local recurrence: a study oftumour spread in the mesorectum distal to rectal cancer. Br JSurg 1995; 82: 1031-3.

145. Seifert JK, Morris DL. Prognostic factors after cryotherapy forhepatic metastases from colorectal cancer. Ann Surg 1998; 228:201-8.

146. Seow-Choen F, Goh HS. Prospective randomised trialcomparing J colonic pouch-anal anastomosis and straightcoloanal reconstruction. Br J Surg 1995; 82: 608-10.

147. Sischy B. The place of radiotherapy in the management ofrectal adenocarcinoma. Cancer; 50: (11 Suppl): 2631-7 1982 Dec.

148. Sprangers MAG, Te Velde A, Aaronson NK, Taal B. Quality oflife following surgery for colorectal cancer: A literature review. Psycho-oncology 1993; 2: 247-59.

149. Steele RJC, Hirshman MJ, Mortensen NJMcC, ArmitageNCM, Scholefield JH. Trans Anal Endoscopic Micro-Surgery –initial experience from three centres in the UK. Br J Surg 1996;83: 207-10.

150. Slaney G et al (eds) Cancer of the large bowel. Clinical CancerMonograph Vol 4. MacMillan Press 1991.

151 Tepper JE, O’Connell MJ, Petroni GR et al. Adjuvant post-operative fluorouracil modulated chemotherapy combined withpelvic radiation therapy for rectal cancer: initial results of INT0114. J Clin Oncol. 197; 15: 2030-9.

Colorectal Cancer Management Clinical Guidelines 43References

152. Thirion P, Wolmark N, Haddad E et al. Survival Impact ofChemotherapy in Patients with Colorectal MetastasesConfirned to the Liver: A re-analysis of 1,458 Non-OperablePatients Randomized in 22 Trials and 4 Meta Analyses. ProcAm Soc Clin Oncol 1999; 10: 1317-20.

153. Thomas PR, Linbald AS. Adjuvant postoperative radiotherapyand chemotherapy in rectal carcinoma: A review of theGastrointestinal Tumor Study Group Experience. RadiotherOncol 1988; 13: 245-52.

154. Tornqvist A, Eklund G, Leandoer L. Early diagnosis ofmetachronous colorectal carcinoma. Aust N Z J Surg. 1981;51: 442-5.

155. Twelves C, Harper P, Van Cutsem E et al. A phase III trial ofXeloda (Capecitabine) in previously untreatedadvanced/metastatic colorectal cancer. Proc Am Soc Clin Oncol1999; 18: 1010.

156. Umpleby HC, Bristol JB, Rainey JB, Williamson RCN. Survivalof 727 patients with single carcinomas of the large bowel. DisColon Rectum 1984; 27: 803-10.

157. Umpleby HC, Fermor B, Symes MO, Williamson RCN.Viability of exfoliated colorectal carcinoma cells. Br J Surg1984; 71: 659-663.

158. Wade TP, Virgo KS, Li MJ et al. Outcomes after detection ofmetastatic carcinoma of the colon and rectum in a nationalhospital system. J Am Coll Surgeons, 1996; 182: 353-61.

159. West of Scotland and Highland Anastomosis Study Group.Stapling or suturing in gastrointestinal surgery: a prospectiverandomised study. Br J Surg 1991; 78: 337-341.

160. Whiteway J, Nicholls RJ, Morson BC. The role of surgicallocal excision in the treatment of rectal cancer. Br J Surg 1985;72: 694-7.

161. Wiggers T, Arends JW, Volovics A. Regression analysis ofprognostic factors in colorectal cancer after curative resections/Dis Colon Rectum 1988; 31: 33-41.

162. Wiggers T, Jeekel J, Arends JW et al. No-touch isolationtechnique in colon cancer: a controlled prospective trial. Br JSurg 1988; 75: 409-15.

163. Williams NS, Johnston D. Survival and recurrence aftersphincter saving resection and abdominoperineal resection forcarcinoma of the middle third of the rectum. Br J Surg 1984;71: 278-82.

164. Williams PL, Warwick R. Gray’s Anatomy. ChurchillLivingstone 1980. p1356.

165. Williams NS, Johnston D. The quality of life after rectalexcision for low rectal cancer. Br J Surg 1983; 70: 460-2.

166. Williams NS, Dixon MF, Johnston D. Reappraisal of the 5 cmrule of distal excision for carcinoma of the rectum: a study ofdistal intramural spread and of patient survival. Br J Surg1983; 70: 278-282.

167. Williams NS, Nasmyth DG, Jones D, Smith AD. Defunctioningstomas: a prospective controlled trial comparing loop ileostomywith transverse loop colostomy. Br J Surg 1996; 73: 566-70.

168. Williams NS, Durdey P, Johnston D. The outcome followingsphincter-saving resection and abdomino-perineal resection forlow rectal cancer. Br J Surg 1985; 72: 595-8.

169. Winawar SJ, Zauber AG, O’Brien MJ et al. Randomisedcomparison of surveillance intervals after colonoscopic removalof newly diagnosed adenomatous polyps. N Engl J Med 1993;328: 901-6.

170. Wolmark N, Bryant J, Smith R et al. Adjuvant 5-fluorouraciland leucovorin with or without interferon alfa-2a in coloncarcinoma: National Surgical Adjuvant Breast and BowelProject protocol C-05. J Nat Cancer Inst 1998; 90(23): 1810-1816.

171. Wolmark N, Wieand HS, Hyams DM et al. Randomized trialof postoperative adjuvant chemotherapy with or withoutradiotherapy for carcinoma of the rectum: National SurgicalAdjuvant Breast and Bowel Project Protocol R-02. J NatCancer Inst. 2000; 92: 388-96.

172. Wong R, Thomas G, Cummings B, Froud P, Shelley W, WithersR, Williams J. In search of a dose-response relationship withradiotherapy in the management of recurrent rectal carcinomain the pelvis: a systematic review. Int J Radiat Oncol Biol Phys1998; 40): 437-46.

173. Wyatt JP, Aitken RJ. Evaluation of hospital and generalpractice follow-up after surgery for colorectal cancer. Br J Surg1994; 81(1):145.

174. Zaniboni A, Labianca R, Marsoni S et al. GIVIO-SITAC 01: Arandomized trial of adjuvant 5-fluorouracil and folinic acidadministered to patients with colon carcinoma – long termresults and evaluation of the indicators of health-related qualityof life. Cancer. 1998; 82: 2135-44.

175. Zirngibil H, Husemann B, Hermanek P. Intraoperative spillageof tumours cells in surgery for rectal cancer. Dis Colon Rectum 1990; 33: 610-14.

Colorectal Cancer Management Clinical Guidelines44References

Colorectal CancerManagementClinical Guidelines

Prepared by The Clinical Guidelines CommitteeRoyal College of Surgeons in Ireland

November 2002

Royal College of Surgeons in Ireland123 St. Stephen’s Green, Dublin 2, Ireland

Tel: 353-1 402 2100. Fax: 353-1 402 2460. Web: www.rcsi.ie

Documents

4721 RCSI Colorectal Guide covsurvival in patients with colorectal cancer. In 1995 there was some 17,000 deaths due to colorectal cancer in the U.K. The overall survival quoted in