MedicineLakex

 

Oncological outcome of malignant colonic obstruction in the dutch stentin 2 trial

Oncological outcome of malignant colonic obstruction
in the Dutch Stent-In 2 trial
D. A. M. Sloothaak1, M. W. van den Berg2, M. G. W. Dijkgraaf3, P. Fockens2, P. J. Tanis1,
J. E. van Hooft2 and W. A. Bemelman1 on behalf of the collaborative Dutch Stent-In study group
1Department of Surgery, 2Department of Gastroenterology and Hepatology, and 3Clinical Research Unit, Academic Medical Centre, Amsterdam,The NetherlandsCorrespondence to: Professor W. A. Bemelman, Department of Surgery, Academic Medical Centre, PO Box 22660, 1100 DD Amsterdam, The Netherlands(e-mail: [email protected]) Background: The Stent-In 2 trial randomized patients with malignant colonic obstruction to emergency
surgery or stent placement as a bridge to elective surgery. The aim of this study was to compare the
oncological outcomes.
Methods: Disease recurrence, and disease-free, disease-specific and overall survival were evaluated,
including a subgroup analysis of patients with a stent- or guidewire-related perforation.
Results: Of 98 patients included in the original Stent-In 2 trial, patients with benign (16) or incurable
(23) disease were excluded from this study, along with a patient who had withdrawn from the trial. Of
the remaining 58 patients, 32 were randomized to emergency surgery (31 resection, 1 stoma only) and
26 to stenting. Unsuccessful stenting required emergency surgery in six patients owing to wire or stent
perforation. Locoregional or distant disease recurrence developed in nine of 32 patients in the emergency
surgery group and 13 of 26 in the stent group. Disease-free survival was worse in the subgroup with stent-
or guidewire-related perforation. Five of six patients in this subgroup developed a recurrence, compared
with nine of 32 in the emergency surgery group and eight of 20 who had unperforated stenting.
Conclusion: Stent placement for malignant colonic obstruction was associated with a risk of recurrence
in this trial, but the numbers are small. There is not enough evidence to refute the approach strongly.
Registration number: ISRCTN46462267 (http://www.controlled-trials.com).
Presented to the spring meeting of the Association of Surgeons from the Netherlands, Veldhoven, The Netherlands,May 2013, and to United European Gastroenterology Week, Berlin, Germany, October 2013 Paper accepted 8 August 2014
Published online 9 October 2014 in Wiley Online Library (www.bjs.co.uk). DOI: 10.1002/bjs.9645
(6 of 47), and occult perforations were revealed in a further10 per cent of the resected specimens (3 of 31)6.
Stenting malignant colonic obstruction as a bridge to The possibility of tumour cell dissemination result- elective surgery has become a widely used treatment ing from stent insertion and/or tumour perforations is option. Initial cohort studies1–4 suggested that the use worrying, and potential oncological consequences have of a stent was associated with lower mortality, morbidity been debated extensively7. Moreover, interest in potential and colostomy rates than emergency surgical interven- tumour reactivity to stent placement is increasing8. In the tion. Randomized trials thereafter showed conflicting present update of the Stent-In 2 trial, the impact of colonic results, and meta-analyses5 did not confirm a reduction in stent placement as a bridge to elective surgery was assessed mortality or secondary stoma rate.
for oncological outcome in patients presenting with malig- The Dutch Stent-In 2 randomized clinical trial nant colorectal obstruction.
(ISRCTN46462267) compared stenting as a bridge tosurgery with emergency surgery. One of the poten- tial complications of endoscopic stent placement inthis trial was tumour perforation. The clinical stent- Patients were identified from the database of the Stent-In or procedure-related perforation rate was 13 per cent 2 trial9. The trial was stopped prematurely in March 2010 2014 BJS Society Ltd BJS 2014; 101: 1751–1757
Published by John Wiley & Sons Ltd D. A. M. Sloothaak, M. W. van den Berg, M. G. W. Dijkgraaf, P. Fockens, P. J. Tanis, J. E. van Hooft and W. A. Bemelman
in accordance with advice from the Data Safety Monitor- ing Board. Interim analysis of the first 60 included patients, Data were analysed based on the on-treatment principle10.
and thereafter the first 90 patients, revealed an increased Continuous data are presented as median (i.q.r.) and were risk of 30-day morbidity (relative risk 1⋅60 (95 per cent c.i.
compared using the Mann–Whitney U test. For dichoto- 0⋅85 to 3⋅01) and 1⋅62 (0⋅94 to 2⋅78) respectively) for the mous outcomes, the stent and emergency surgery groups stent group compared with the emergency surgery group.
were compared by means of χ2 or Fisher's exact test. The In addition, interim analysis recorded no difference in the Kaplan–Meier method was used for survival analysis, with primary outcome of the study, which was global health sta-tus. Details of the study design and short-term outcomes comparison between stent and emergency surgery groups have been published previously6. Patients randomized to using the log rank test. In addition, a subanalysis was per- emergency surgery underwent resection of the primary formed of all patients who had a stent-related perfora- tumour, with either primary anastomosis, temporary stoma tion within the stent group. Inclusion in this subgroup was or definitive stoma, at the discretion of the surgeon. Those based on the presence of tumour perforation on patholog- randomized to stent placement received either an enteral ical examination; this included both clinical and subclini- Wallstent™ (diameter 22 mm; Boston Scientific, Marlbor- cal perforations. Estimated survival rates are reported with ough, Massachusetts, USA) or a WallFlex™ colonic stent(diameter 25 mm; Boston Scientific).
Table 1 Baseline characteristics of eligible patients
Patients with non-malignant obstruction, those in whom malignancy was not confirmed histopathologically and patients who were eventually treated with palliative intent were excluded from the present study. The decision regarding intentionally palliative or curative treatment was made during multidisciplinary team meetings. Patients with resectable liver metastasis were treated with curative Sex ratio (M : F) ASA fitness grade Follow-up and data collection
In the Dutch Stent-In 2 trial, patients were initially followed for at least 6 months after randomization.
Prospectively collected patient demographics, treatment characteristics and pathology reports were complemented retrospectively with data on adjuvant treatment, recurrence (locoregional recurrence or distant metastasis) and survival.
Urgency of resection Information was obtained from hospital medical records and general practitioners. The total follow-up was calcu- lated from the date of randomization in the Stent-In 2 trial.
Endpoints of the study were overall and loco- regional disease recurrence, disease-free survival (DFS), disease-specific survival (DSS) and overall survival after 4 years. Diagnosis of disease recurrence was based on radi- ological imaging or histopathological investigation. Loco- Lymph node count* regional recurrence was defined as intestinal, regional Positive lymph node count* Radical resection lymph node or peritoneal recurrence. DFS was defined as Adjuvant chemotherapy the time between resection of the primary tumour and the Follow-up (months)* diagnosis of disease recurrence or death from any cause.
*Values are median (i.q.r.). †Insufficient deployment of inserted stent (1), DSS was defined as the time to cancer-specific death, and inability to insert guidewire (3), suspicion of guidewire-induced overall survival was defined as the time to death from any perforation (2). ASA, American Society of Anesthesiologists. ‡Fisher's exact test, except §Mann–Whitney U test and ¶χ2 test.
2014 BJS Society Ltd BJS 2014; 101: 1751–1757
Published by John Wiley & Sons Ltd Oncological outcome after stenting for malignant colonic obstruction
Emergency surgery Time after surgery (months) Emergency surgery a Disease-free survival
Disease-specif 0·2 Time after surgery (months) Emergency surgery b Disease-specific survival
Time after surgery (months) Emergency surgery c Overall survival
Fig. 1 Kaplan – Meier survival curves for patients randomized to emergency surgery or colonic stenting: a disease-free survival,
b disease-specific survival and c overall survival. a P = 0⋅149, b P = 0⋅061 and c P = 0⋅468 (log rank test)
2014 BJS Society Ltd BJS 2014; 101: 1751–1757
Published by John Wiley & Sons Ltd D. A. M. Sloothaak, M. W. van den Berg, M. G. W. Dijkgraaf, P. Fockens, P. J. Tanis, J. E. van Hooft and W. A. Bemelman
95 per cent c.i.11. All tests were two-sided and P < 0⋅050 Table 2 Comparison of patients with and without stent
indicated statistical significance. Statistical analyses were performed with SPSS® for Windows® version 19.0 (IBM, Armonk, New York, USA).
Sex ratio (M : F) A total of 98 patients were included in the original Stent-In ASA fitness grade 2 trial between 2007 and 20096. Thirty-nine patients were excluded from the present analysis because of benign dis- ease (16) or palliative treatment (7 in whom the primary tumour was never resected, 9 with unresectable liver metas- tases, 4 with incurable peritoneal metastases, and 3 with distant metastases at multiple sites). One additional patient had withdrawn from the trial and was not included in the analysis. Characteristics of the 58 remaining patients are summarized in Table 1.
Six patients randomized to receive a stent as a bridge to surgery underwent emergency operation. In one of these patients the stent did not deploy well enough to result Lymph node count* in clinical decompression. The guidewire could not be Positive lymph node count* Adjuvant chemotherapy positioned in the lumen of the tumour, resulting in a false Follow-up (months)* route in three patients and guidewire-induced perforation *Values are median (i.q.r.). ASA, American Society of Anesthesiologists.
†Fisher's exact test, except ‡Mann–Whitney U test.
Stent-related perforations occurred in six of 26 patients in the stent group. Three stent-related perforations presentedclinically (including 2 after complicated stent placement stent-related perforation (2 locoregional metastasis, 6 procedures) and a further three became apparent during The 4-year DFS rate was 0 (0 to 0) per cent in patients with a stent-related perforation, and was worse than Recurrence and survival
the rate of 45 (22 to 68) per cent in patients withoutstent-related perforation (Fig. 2a). DSS rates were 60 (17 Median follow-up was 45 (35–60) and 41 (19–55) months to 100) and 69 (46 to 92) per cent respectively (Fig. 2b).
in the emergency surgery and stent groups respectively.
The overall survival rate was 50 (10 to 90) per cent for Three of the 58 patients were lost to follow-up after 6–11 patients with a stent-related perforation and 62 (39 to 84) per cent among those without perforation (Fig. 2c).
A recurrence developed in nine of 32 patients after emer- gency surgery (2 locoregional recurrence, 7 distant metas-tasis) and in 13 of 26 patients after stent placement (5 locoregional metastasis, 8 distant metastasis).
The estimated 4-year DFS was 49 (95 per cent c.i. 32 The results of this study are in line with data underlining to 67) and 30 (10 to 51) per cent in the emergency surgery concerns about potential negative oncological outcomes and stent groups respectively (Fig. 1a). Four-year DSS rates of colonic stent placement8,12–14. Examination of resected were 87 (73 to 100) and 66 (37 to 95) per cent respectively stented specimens showed that tumour and peritumoral (Fig. 1b), and overall survival rates were 67 (50 to 84) and ulceration occurred more frequently as a result of stents15.
58 (38 to 78) per cent (Fig. 1c).
In addition, tumour manipulation during guidewire inser- For the subgroup analysis of stent-related perforations, tion, dilatation and stent deployment can cause disrup- endpoints were calculated separately for patients with (6) tion with potential spread of cancer cells (or their shed and without (20) stent-related perforation (Table 2).
particles)7. Only two of the six patients with stent-related Five of six patients developed a recurrence after a perforation in the present study received adjuvant systemic stent-related perforation (3 locoregional recurrence, 2 dis- chemotherapy. Although it is unknown whether adjuvant tant metastasis) compared with eight of 20 patients without systemic chemotherapy is able to prevent outgrowth of 2014 BJS Society Ltd BJS 2014; 101: 1751–1757
Published by John Wiley & Sons Ltd Oncological outcome after stenting for malignant colonic obstruction
Emergency surgeryStent, no perforation Stent, perforation Time after surgery (months) Emergency surgery Stent, no perforation Stent, perforation a Disease-free survival
Time after surgery (months) Emergency surgery Stent, no perforation Stent, perforation b Disease-specific survival
Time after surgery (months) Emergency surgery Stent, no perforation Stent, perforation c Overall survival
Fig. 2 Kaplan – Meier survival curves for patients in the emergency surgery group and those who had colonic stenting with or without
perforation: a disease-free survival, b disease-specific survival and c overall survival. a P = 0⋅007, b P = 0⋅099 and c P = 0⋅478 (log rank
test)
2014 BJS Society Ltd BJS 2014; 101: 1751–1757
Published by John Wiley & Sons Ltd D. A. M. Sloothaak, M. W. van den Berg, M. G. W. Dijkgraaf, P. Fockens, P. J. Tanis, J. E. van Hooft and W. A. Bemelman
direct tumour spread after perforation, this may have influ- seems to become less important than the increased risk of enced the overall recurrence rate16.
postoperative death in these frail patients.
Literature on the oncological consequences of colonic stent placement in the curative setting is scarce. Recently, cohort studies have compared the oncological outcome ofstenting as a bridge to surgery with emergency surgery.
Members of the Dutch Stent-In study group are col- The group from St Vincent's University Hospital in laborators in this study: J. E. van Hooft, P. Fockens, Dublin12 showed similar overall and cancer-specific sur- W. A. Bemelman, M. G. Dijkgraaf, M. A. Sprangers, C.
vival in 49 patients. This has been echoed by others17, J. Buskens (Academic Medical Centre, Amsterdam); J. M.
but there have also been reports of a negative impact on Jansen, M. F. Gerhards (Onze Lieve Vrouwe Gasthuis, cancer outcomes. One retrospective study8 used a propen- Amsterdam); R. Timmer, B. van Ramshorst (St Antonius sity score to adjust for allocation bias in 48 patients who Hospital, Nieuwegein); B. Oldenburg, R. van Hilligersberg received a bridging stent and 39 who underwent emergency (University Medical Centre, Utrecht); C. M. Bakker, M.
surgery. The 5-year overall survival rate was significantly Sosef (Atrium Hospital, Heerlen); P. Witteman, P. Kruyt lower in the stent group than in the emergency surgery (Gelderse Vallei Hospital, Ede); W. R. ten Hove, L. N.
group (25 versus 62 per cent respectively) and the 5-year Tseng (Groen Hart Hospital, Gouda); K. van der Linde, S.
cancer specific mortality rate was significantly higher in A. Koopal (Medical Centre Leeuwarden, Leeuwarden); A.
the stent group (48 versus 21 per cent). Finally, in another W. Marinelli, L. Perk (Medical Centre Haaglanden, Den study14 of patients aged 75 years or younger, 38 of whom Haag); M. F. Lutke Holzik (Medisch Spectrum Twente, were stented and 24 had emergency surgery, the local Enschede); M. J. Grubben, J. Heisterkamp (Sint Elisabeth recurrence-free survival rate was significantly worse after Hospital, Tilburg); A. C. Depla, E. Derksen (Slotervaart stenting (8 versus 32 per cent; P = 0⋅038).
Hospital, Amsterdam); A. H. Naber, A. A. van Geloven The prevalence of stent-related complications is likely (TerGooi Hospitals, Hilversum); R. Breumelhof, P. H.
to be underestimated, because subclinical perforations Davids (Diakonessenhuis, Utrecht); H. Akol, E. van der may be identified only at the time of pathological Zaag (Gelre Hospitals, Apeldoorn); E. Schenk, G. A. Patijn examination18–20. The prevalence of clinical stent-related (Isala Hospitals, Zwolle); R. A. Veenendaal, R. A. Tollenaar perforation is 6⋅9 per cent based on meta-analysis5. The (Leiden University Medical Centre, Leiden); A. van Berkel risk of perforation may be related to local anatomy, tumour (Rode Kruis Hospital, Beverwijk); L. P. Gilissen, G. A.
factors (length and constitution), stent design and expe- Nieuwenhuijzen (Catharina Hospital, Eindhoven); L. A.
rience of the clinician placing the stent1,18,21–23. Because van der Waaij, P. C. Baas (Martini Hospital, Groningen); it is still not clear how stent-related perforations can be H. Cense (Rode Kruis Hospital, Beverwijk); P. Scholten, B.
prevented, this risk should be given serious consideration van Wagensveld (St Lucas Andreas Hospital, Amsterdam); in clinical decision-making. Even if stent placement seems J. J. Koornstra, K. Havenga (University Medical Centre to be uncomplicated clinically, tumour biology may be Groningen, Groningen); M. van Milligen de Wit, A. M.
altered owing to the mechanical stress, and tumour spillage Rijken (Amphia Hospital, Breda); M. Cazemier, O. R.
may occur because of tumour disruption. A Markov chain Guicherit (Bronovo Hospital, Den Haag); M. H. Houben, Monte Carlo decision analysis model determined differ- W. H. Steup (Haga Hospitals, Den Haag).
ences in effectiveness and costs between the two strategiesfor different patient scenarios based on perioperative risk factors24. Probabilities of outcome variables were based onsystematic review of 54 studies including 1198 patients.
The authors declare no conflict of interest.
Based on this analysis, the authors concluded that the ben-efits of colonic stenting are modest in low-risk patients, and may not outweigh the risks. Therefore, emergencysurgery is probably the preferred treatment strategy for 1 Saida Y, Sumiyama Y, Nagao J, Uramatsu M. Long-term prognosis of preoperative ‘bridge to surgery' expandable patients without significantly increased operative risk.
metallic stent insertion for obstructive colorectal cancer: On the other hand, stenting is probably the preferred comparison with emergency operation. Dis Colon Rectum strategy in patients at increased operative risk (age above 2003; 46(Suppl): S44–S49.
70 years, American Society of Anesthesiologists fitness 2 Martinez-Santos C, Lobato RF, Fradejas JM, Pinto I, grade III or more). The potentially impaired oncological Ortega-Deballón P, Moreno-Azcoita M. Self-expandable outcome related to the risk of stent-related perforation stent before elective surgery vs. emergency surgery for the 2014 BJS Society Ltd BJS 2014; 101: 1751–1757
Published by John Wiley & Sons Ltd Oncological outcome after stenting for malignant colonic obstruction
treatment of malignant colorectal obstructions: comparison Oncologic outcomes of self-expanding metallic stent of primary anastomosis and morbidity rates. Dis Colon insertion as a bridge to surgery in the management of Rectum 2002; 45: 401–406.
left-sided colon cancer obstruction: comparison with 3 Ng KC, Law WL, Lee YM, Choi HK, Seto CL, Ho JW.
nonobstructing elective surgery. World J Surg 2009; 33:
Self-expanding metallic stent as a bridge to surgery versus emergency resection for obstructing left-sided colorectal 14 Gorissen KJ, Tuynman JB, Fryer E, Wang L, Uberoi R, cancer: a case-matched study. J Gastrointest Surg 2006; 10:
Jones OM et al. Local recurrence after stenting for obstructing left-sided colonic cancer. Br J Surg 2013; 100:
4 Park JS, Kim BG, Chang IT, Choi YS, Kwak BK, Shim HJ et al. Placement of stents in proximal colonic obstructions 15 Sabbagh C, Chatelain D, Trouillet N, Mauvais F, using a percutaneous retroperitoneal colostomy. Surg Bendjaballah S, Browet F et al. Does use of a metallic colon Laparosc Endosc Percutan Tech 2009; 19: e202–e205.
stent as a bridge to surgery modify the pathology data in 5 Tan CJ, Dasari BV, Gardiner K. Systematic review and patients with colonic obstruction? A case-matched study.
meta-analysis of randomized clinical trials of self-expanding Surg Endosc 2013; 27: 3622–3631.
metallic stents as a bridge to surgery versus emergency 16 Benson AB III, Schrag D, Somerfield MR, Cohen AM, surgery for malignant left-sided large bowel obstruction. Br Figueredo AT, Flynn PJ et al. American Society of Clinical J Surg 2012; 99: 469–476.
Oncology recommendations on adjuvant chemotherapy for 6 van Hooft JE, Bemelman WA, Oldenburg B, Marinelli AW, stage II colon cancer. J Clin Oncol 2004; 22:
Lukte Holzik MF, Grubben MJ et al. Colonic stenting versus emergency surgery for acute left-sided malignant colonic 17 Kim HJ, Huh JW, Kang WS, Kim CH, Lim SW, Joo YE obstruction: a multicentre randomised trial. Lancet Oncol et al. Oncologic safety of stent as bridge to surgery compared 2011; 12: 344–352.
to emergency radical surgery for left-sided colorectal cancer 7 Maruthachalam K, Lash GE, Shenton BK, Horgan AF.
obstruction. Surg Endosc 2013; 27: 3121–3128.
Tumour cell dissemination following endoscopic stent 18 Baron TH. Colonic stenting: technique, technology, and insertion. Br J Surg 2007; 94: 1151–1154.
outcomes for malignant and benign disease. Gastrointest 8 Sabbagh C, Browet F, Diouf M, Cosse C, Brehant O, Bartoli Endosc Clin N Am 2005; 15: 757–771.
E et al. Is stenting as ‘a bridge to surgery' an oncologically 19 Small AJ, Coelho-Prabhu N, Baron TH. Endoscopic safe strategy for the management of acute, left-sided, placement of self-expandable metal stents for malignant malignant, colonic obstruction? A comparative study with a colonic obstruction: long-term outcomes and complication propensity score analysis. Ann Surg 2013; 258: 107–115.
factors. Gastrointest Endosc 2010; 71: 560–572.
9 van Hooft JE, Bemelman WA, Breumelhof R, Siersema PD, 20 Watt AM, Faragher IG, Griffin TT, Rieger NA, Maddern Kruyt PM, van der Linde K et al. Colonic stenting as bridge GJ. Self-expanding metallic stents for relieving malignant to surgery versus emergency surgery for management of colorectal obstruction: a systematic review. Ann Surg 2007; acute left-sided malignant colonic obstruction: a multicenter randomized trial (Stent-in 2 study). BMC Surg 2007; 7: 12.
21 Binkert CA, Ledermann H, Jost R, Saurenmann P, 10 Chêne G, Morlat P, Leport C, Hafner R, Dequae L, Decurtins M, Zollikofer CL. Acute colonic obstruction: Charreau I et al. Intention-to-treat vs. on-treatment analyses clinical aspects and cost-effectiveness of preoperative and of clinical trial data: experience from a study of palliative treatment with self-expanding metallic stents – a pyrimethamine in the primary prophylaxis of toxoplasmosis preliminary report. Radiology 1998; 206: 199–204.
in HIV-infected patients. ANRS 005/ACTG 154 Trial 22 Khot UP, Lang AW, Murali K, Parker MC. Systematic Group. Control Clin Trials 1998; 19: 233–248.
review of the efficacy and safety of colorectal stents. Br J 11 Twisk JWR. Inleiding in de Toegepaste Biostatistiek (2nd edn).
Surg 2002; 89: 1096–1102.
Elsevier: Amsterdam, 2010.
23 van Halsema EE, van Hooft JE, Small AJ, Baron TH, 12 Kavanagh DO, Nolan B, Judge C, Hyland JM, Mulcahy García-Cano J, Cheon JH et al. Perforation in colorectal HE, O'Connell PR et al. A comparative study of short- and stenting: a meta-analysis and a search for risk factors.
medium-term outcomes comparing emergent surgery and Gastrointest Endosc 2014; 79: 970–982.e7.
stenting as a bridge to surgery in patients with acute 24 Govindarajan A, Naimark D, Coburn NG, Smith AJ, Law malignant colonic obstruction. Dis Colon Rectum 2013; 56:
CH. Use of colonic stents in emergent malignant left colonic obstruction: a Markov chain Monte Carlo decision 13 Kim JS, Hur H, Min BS, Sohn SK, Cho CH, Kim NK.
analysis. Dis Colon Rectum 2007; 50: 1811–1824.
2014 BJS Society Ltd BJS 2014; 101: 1751–1757
Published by John Wiley & Sons Ltd

Source: http://www.colonobstructie.nl/Richtlijnen_en_literatuur_files/Oncological%20outcome%20of%20the%20Dutch%20Stent-In%202%20trial.pdf