2
Leading article Stenting and colorectal cancer J. Hill Department of Surgery, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK (e-mail: [email protected]) Published online in Wiley InterScience (www.bjs.co.uk). DOI: 10.1002/bjs.6386 Colorectal cancer (CRC) is the second commonest cancer in the United Kingdom. It affects 35 000 individuals each year in England and Wales alone. Although the age adjusted mortality from CRC remains close to 50 per cent, recent developments in diagnosis and adjuvant therapy have improved survival rates. These quantifiable survival benefits have all resulted from randomized controlled trials 1,2,3 . Up to 30 per cent of CRCs present as an emergency with large bowel obstruction. The mortality rate fol- lowing emergency surgery for obstructing lesions is high, between 15 and 30 per cent; the mortality rate following elective surgery for CRC is 5 per cent 4 . Furthermore, emergency surgery is associated with a prolonged hospital stay, a greater requirement for critical care and a high frequency of stoma formation, all of which com- promise the patient’s quality of life and recovery. For those who survive an emergency operation, early sur- vival (according to stage) is similar to that associated with the elective setting. Patients treated in the emer- gency setting, however, may have their subsequent care compromised as they are often unable to tolerate adjuvant chemotherapy. These fea- tures suggest that it is the process of care, rather than the disease itself, that influences survival adversely in the emergency setting. Published reports suggest that stenting of obstructing lesions may allow emergency surgery to be avoided. This should permit a full assessment and preparation of the patient for an ‘elective’ operation, thereby reducing operative morbid- ity and improving quality of life and survival 5 . Self-expanding metal stents were initially deployed in patients with incurable disease. More recently they have become popular as a ‘bridge to surgery’ in those with obstruct- ing CRC. Stenosing lesions in the rectum and rectosigmoid region can be treated by a radiologist work- ing alone, but the more proximal the lesion, the greater the need for a combined endoscopic/fluoroscopic approach. Following the introduc- tion of a guide wire beyond the obstruction, the stent is inserted either through the endoscope or, after removal of the endoscope, over the guide wire. Stent delivery systems require an endoscope working chan- nel of 3·7 mm. For left-sided lesions, the therapeutic gastroscope is most often used; for right-sided lesions a therapeutic colonoscope is required. A number of non-randomized studies have compared different stent designs but no single design has yet been shown to be superior. Nearly a hundred articles have been published on stent insertion for malig- nant colorectal obstruction. Sebastian and colleagues in 2004 performed a pooled analysis of 1198 patients from 54 heterogeneous cohort studies 6 . They reported median technical and clinical success rates of 94 (i.q.r 90–100) and 91 (i.q.r 84–94) per cent, respectively. The clinical suc- cess when used as a bridge to surgery was 71·7 per cent. Early complica- tions relating to stent placement included perforation (3·7 per cent) and stent migration (11·8 per cent). Stent-related mortality was 0·58 per cent. No meaningful data exist on the long-term effect of stenting on CRC survival. This is important because recent evidence suggests that the endoscopic insertion of colonic stents results in increased levels of CK20 mRNA in the peripheral circu- lation and may result in tumour cell dissemination 7 . It is also unfortunate that these small, non-randomized, series are subject to significant selec- tion bias between the study popula- tions. This calls their conclusions into question and should restrict dissem- ination of their findings into clinical practice. Despite these undoubted limita- tions in evidential quality, stent inser- tion is becoming more widely prac- tised, both in the palliative and the emergency setting. To date, no ran- domized controlled clinical trial com- paring emergency colonic stenting with emergency surgery for CRC has been undertaken. Importantly, the recent Dutch Stent-in-1 Study was closed prematurely 8 . This was a multi-centre, prospective controlled trial designed to assess the poten- tial benefit of endoluminal stenting compared to surgery in patients with incurable CRC. The trial was stopped because of the high rate of stent- related complications. By the end of the trial, six of nine stented patients had a perforation, resulting in opera- tive intervention or death. Two of the perforations occurred 12 days after stent insertion, and the other four at a later stage following chemother- apy. The closure of this study high- lights the considerable uncertainty about the role of endoluminal stent- ing. In addition, in 2007 the National Copyright 2008 British Journal of Surgery Society Ltd British Journal of Surgery 2008; 95: 1195–1196 Published by John Wiley & Sons Ltd

Stenting and colorectal cancer

  • Upload
    j-hill

  • View
    217

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Stenting and colorectal cancer

Leading article

Stenting and colorectal cancerJ. HillDepartment of Surgery, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK(e-mail: [email protected])

Published online in Wiley InterScience (www.bjs.co.uk). DOI: 10.1002/bjs.6386

Colorectal cancer (CRC) is the secondcommonest cancer in the UnitedKingdom. It affects 35 000 individualseach year in England and Walesalone. Although the age adjustedmortality from CRC remains closeto 50 per cent, recent developmentsin diagnosis and adjuvant therapyhave improved survival rates. Thesequantifiable survival benefits have allresulted from randomized controlledtrials1,2,3.

Up to 30 per cent of CRCs presentas an emergency with large bowelobstruction. The mortality rate fol-lowing emergency surgery forobstructing lesions is high, between15 and 30 per cent; the mortality ratefollowing elective surgery for CRC is5 per cent4. Furthermore, emergencysurgery is associated with a prolongedhospital stay, a greater requirementfor critical care and a high frequencyof stoma formation, all of which com-promise the patient’s quality of lifeand recovery. For those who survivean emergency operation, early sur-vival (according to stage) is similarto that associated with the electivesetting. Patients treated in the emer-gency setting, however, may havetheir subsequent care compromisedas they are often unable to tolerateadjuvant chemotherapy. These fea-tures suggest that it is the processof care, rather than the disease itself,that influences survival adversely inthe emergency setting.

Published reports suggest thatstenting of obstructing lesions mayallow emergency surgery to beavoided. This should permit a fullassessment and preparation of thepatient for an ‘elective’ operation,

thereby reducing operative morbid-ity and improving quality of life andsurvival5. Self-expanding metal stentswere initially deployed in patientswith incurable disease. More recentlythey have become popular as a ‘bridgeto surgery’ in those with obstruct-ing CRC. Stenosing lesions in therectum and rectosigmoid region canbe treated by a radiologist work-ing alone, but the more proximalthe lesion, the greater the need fora combined endoscopic/fluoroscopicapproach. Following the introduc-tion of a guide wire beyond theobstruction, the stent is insertedeither through the endoscope or, afterremoval of the endoscope, over theguide wire. Stent delivery systemsrequire an endoscope working chan-nel of 3·7 mm. For left-sided lesions,the therapeutic gastroscope is mostoften used; for right-sided lesions atherapeutic colonoscope is required.A number of non-randomized studieshave compared different stent designsbut no single design has yet beenshown to be superior.

Nearly a hundred articles have beenpublished on stent insertion for malig-nant colorectal obstruction. Sebastianand colleagues in 2004 performed apooled analysis of 1198 patients from54 heterogeneous cohort studies6.They reported median technical andclinical success rates of 94 (i.q.r90–100) and 91 (i.q.r 84–94) percent, respectively. The clinical suc-cess when used as a bridge to surgerywas 71·7 per cent. Early complica-tions relating to stent placementincluded perforation (3·7 per cent)and stent migration (11·8 per cent).Stent-related mortality was 0·58

per cent. No meaningful data existon the long-term effect of stentingon CRC survival. This is importantbecause recent evidence suggests thatthe endoscopic insertion of colonicstents results in increased levels ofCK20 mRNA in the peripheral circu-lation and may result in tumour celldissemination7. It is also unfortunatethat these small, non-randomized,series are subject to significant selec-tion bias between the study popula-tions. This calls their conclusions intoquestion and should restrict dissem-ination of their findings into clinicalpractice.

Despite these undoubted limita-tions in evidential quality, stent inser-tion is becoming more widely prac-tised, both in the palliative and theemergency setting. To date, no ran-domized controlled clinical trial com-paring emergency colonic stentingwith emergency surgery for CRChas been undertaken. Importantly,the recent Dutch Stent-in-1 Studywas closed prematurely8. This was amulti-centre, prospective controlledtrial designed to assess the poten-tial benefit of endoluminal stentingcompared to surgery in patients withincurable CRC. The trial was stoppedbecause of the high rate of stent-related complications. By the end ofthe trial, six of nine stented patientshad a perforation, resulting in opera-tive intervention or death. Two of theperforations occurred 12 days afterstent insertion, and the other fourat a later stage following chemother-apy. The closure of this study high-lights the considerable uncertaintyabout the role of endoluminal stent-ing. In addition, in 2007 the National

Copyright 2008 British Journal of Surgery Society Ltd British Journal of Surgery 2008; 95: 1195–1196Published by John Wiley & Sons Ltd

Page 2: Stenting and colorectal cancer

1196 J. Hill

Bowel Cancer Audit Project reportedan 11 per cent mortality after colonicstenting for obstructing cancers; theauthors commented that this washigher than in previously publishedreports and needed further study4.

Stent insertion needs to be prop-erly evaluated in a randomized con-trolled trial. Cancer Research UK hasrecently funded a multi-centre ran-domized trial of stenting in obstruct-ing left-sided CRC (CReST). Thistrial aims to randomize 400 patientsto answer the following questions.First, is there a worthwhile net benefit(length of hospital stay, 30-day mor-tality and presence and duration of astoma) from endoluminal stenting forpatients presenting with an obstruct-ing colonic cancer? Second, if sucha benefit exists, is this identifiable inpatients undergoing attempted cura-tive treatment, palliative treatment, orboth?

For those who are already con-vinced of the benefits of stent deploy-ment for obstructing CRC, the storyof stenting for carotid artery steno-sis is noteworthy. Initial enthusi-asm for carotid stenting led toits use in many centres. Yet asBeebe and Kritpracha9 pointed out‘Because no prospective randomizedtrial has yet been published comparingcarotid endarterectomy with carotidangioplasty and stenting, we mustuse data from statewide, population-based, and single-center reports tocompare the procedures’; such data

led to their concluding that ‘currentevidence does not indicate the useof carotid stenting as a routine alter-native to carotid endarterectomy’. Ameta-analysis has shown that carotidendarterectomy can be performedwith more safety than the endovas-cular procedure, and that endarterec-tomy remains the ‘gold standard’treatment for carotid stenosis requir-ing intervention10.

It may well be that stent inser-tion does, indeed, offer significantadvantages for patients with obstruct-ing CRC, but this is not yet proven.Demonstrating benefit in the settingof a randomized clinical trial shouldenable any advantages to be prop-erly evaluated. Benefit can then reachthe greatest number of patients in themost appropriate way.

References

1 Scholefield JH, Moss S, Sufi F,Mangham CM, Hardcastle JD. Effectof faecal occult blood screening onmortality from colorectal cancer:results from a randomised controlledtrial. Gut 2002; 50: 840–844.

2 Quasar Collaborative Group.Adjuvant chemotherapy versusobservation in patients with colorectalcancer: a randomised study. Lancet2007; 370: 2020–2029.

3 Kapiteijn E, Marijnen CA,Nagtegaal ID, Putter H, Steup WH,Wiggers T et al. Preoperativeradiotherapy combined with totalmesorectal excision for resectable

rectal cancer. N Engl J Med 2001;345: 638–646.

4 Smith JJ, Cornish J, Tekkis P,Thompson MR. The National BowelCancer Audit Project 2007, QualityImprovement and Open Reporting.Association of Coloproctology ofGreat Britain and Ireland, 2007.

5 Khot UP, Lang AW, Murali K,Parker MC. Systematic review of theefficacy and safety of colorectal stents.Br J Surg 2002; 89: 1096–102.

6 Sebastian S, Johnston S,Geoghegan T, Torreggiani W,Buckley M. Pooled analysis of theefficacy and safety of self-expandingmetal stenting in malignant colorectalobstruction. Am J Gastroenterol 2004;99: 2051–2057.

7 Maruthachalam K, Lash GE,Shenton BK, Horgan AF. Tumourcell dissemination followingendoscopic stent insertion. Br J Surg2007; 94: 1151–1154.

8 van Hooft JE, Fockens P,Marinelli AW, Timmer R, vanBerkel AM, Bossuyt PM et al. Earlyclosure of a multicenter randomizedclinical trial of endoscopic stentingversus surgery for stage IV left-sidedcolorectal cancer. Endoscopy 2008; 40:184–191.

9 Beebe HG, Kritpracha B. Carotidendarterectomy versus carotidangioplasty: comparison of currentresults. Semin Vasc Surg 2000; 13:109–116.

10 Luebke T, Aleksic M, Brunkwall J.Meta-analysis of randomized trialscomparing carotid endarterectomyand endovascular treatment. Eur JVasc Endovasc Surg 2007; 34: 470–479.

Copyright 2008 British Journal of Surgery Society Ltd www.bjs.co.uk British Journal of Surgery 2008; 95: 1195–1196Published by John Wiley & Sons Ltd