Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Collateral Damage?
- Laurie R. Davis (8 June 2009)
-
Reply to Laurie R Davis regarding the NORCCAP trial results
- Geir Hoff (9 June 2009)
-
Limitations of trial
- Richard Peto (13 June 2009)
-
Survival benefit form Sigmoidoscopy screening is likely to be confined to Left sided Cancers.
- Rajeev Peravali (20 June 2009)
-
Sigmoidoscopy screening - age of target population and threshold for work-up colonoscopy
- Geir Hoff (22 June 2009)
-
Complications due to sigmoidoscopy and colonoscopy?
- Hans-Hermann Dubben (25 November 2009)
-
Re: Complications due to sigmoidoscopy and colonoscopy?
- Geir Hoff (26 November 2009)
|
|
|||
|
Laurie R. Davis, GP South Hermitage Surgery , Shrewsbury SY3 7JS
Send response to journal:
|
A curious omission in the study report is the absence of any data on surgical intervention, beyond the colonoscopy and sigmoidoscopy stage. Did the screened group ultimately end up with no more laparotomies or colostomies than the control group? How many suffered significant surgical complications? I think this is important , as the morbidity is clearly a significant issue with colorectal cancer.Are the screened group more or less likely to end up with a stoma? It might be hard to persuade people to be screened with the modest colorectal death reduction ( so far) but easier if there was a reduction in laparotomies, stoma risk or chemotherapy. Competing interests: None declared |
|||
|
|
|||
|
Geir Hoff, Head of Research Telemark Hospital, 3710 Skien, Norway
Send response to journal:
|
Dear dr. Davis, Thank you for your valuable comments. We fully appreciate your request for data on surgical interventions in the NORCCAP trial on flexible sigmoidoscopy screening. Baseline data on additional work-up, surgery and complications among screenees were published in Gondal G et al, Scand J Gastroenterol 2003 (1), referred to in the BMJ paper. We do not, however, have such information from the control group as these individuals were not contacted and available registries do not provide these data. Yours sincerely, Geir Hoff Reference: 1. G Gondal, T Grotmol, B Hofstad, M Bretthauer, T Eide, G Hoff. The Norwegian Colorectal Cancer Prevention (NORCCAP) screening study: Baseline findings and implications for clinical work-up in age groups 50-64 years. Scand J Gastroenterol 2003;38:635-42 Competing interests: None declared |
|||
|
|
|||
|
Richard Peto, Professor of Medical Statistics and Epidemiology Oxford, UK
Send response to journal:
|
In the NORCCAP trial of colorectal cancer screening in Norway (BMJ 2009; 338: b1846), the number of patients screened and the range of intestinal sites examined was too limited for even an apparent fourfold reduction in mortality from rectosigmoid cancer to be reliably detectable by a standard intention-to-treat analysis of overall colorectal cancer mortality. In these circumstances, it seems more appropriate for the medical interpretation of the available results to be based on separate consideration of rectosigmoid cancer mortality and colon cancer mortality, and for the main emphasis to be on the separate results for those who accepted the screening invitation; those who, although allocated screening, rejected the invitation; and those allocated control. Among those screened, all had rectosigmoid examination, but only 19% had, in addition, full colonoscopy; so, protection against rectosigmoid cancer would be expected to be greater than protection against colon cancer mortality. The following table of the numbers of colorectal cancer deaths supports the authors’ suggestion that further mortality follow-up may well provide further evidence of benefit. Table: Six-year numbers of deaths from colon cancer and from rectosigmoid cancer in NORCCAP, by actual and allocated treatment
Attended screening
(n=8846) Rejected invitation
(n=4807)
Allocated control
(n=41092)
Observed Expected* Observed Expected* Observed
Colon cancer
death 6 9.0 6 4.9 42
Rectosigmoid
cancer
death
3 12.3 9 6.7 57
*Expected at the rate observed in controls
The results among those who accepted the screening invitation suggest substantial protection of a single rectosigmoid examination against death over the next 6 years from rectosigmoid cancer (and, the nice apparent reduction in the incidence of rectosigmoid cancer suggests that further mortality follow-up will confirm this). Nothing in the results among those who rejected screening weighs against this conclusion. Although intention-to-treat analyses have their uses, and are in some circumstances essential, they may in other circumstances lead to false negative interpretations of important trial findings. Richard Peto
Competing interests: None declared |
|||
|
|
|||
|
Rajeev Peravali, ST3 General Surgery East of England Deanery
Send response to journal:
|
Dear Editor, I read with great interest the article by Hoff et al (BMJ 2009; 338:b1846). There is no doubt that screening for colorectal cancer will reduce mortality. The greatest disadvantage of flexible simoidoscopy as a screening tool however is that only the lower third of the colon is examined and much controversy has surrounded the question of right sided colonic neoplasia missed by sigmoidoscopy. The effectiveness of screening sigmoidoscopy depends on the association between distal and proximal adenomas and the proportion of patients with distal adenomas who have proximal pathology. This relationship is uncertain.1 The UK flexible sigmoidoscopy trial (UKFST) proposes a screening regimen where small polyps are removed during sigmoidoscopy and colonoscopy is undertaken when characteristics suggest a “high risk” of proximal leisions. 2 The criteria used were: polyp size greater than or equal to 1 cm, villous histology, severe dysplasia, and/or the presence of 3 or more small adenomas. In our study we looked at the pre and peri operative imaging at all confirmed proximal colon cancers over a seven year period to identify any distal pathology 3. More importantly we looked at what proportion these proximal cancers would have fulfilled the criteria set by he UKFST for a colonoscopy. 348 patients with proximal cancers were identified. Nearly 80% did not have any demonstrable distal neoplasia at all and only 8% would have proceeded to colonoscopy based on the UKFST protocol. This would suggest the flexible sigmoidoscopy would be ineffective at screening for proximal cancers and any survival benefit would be confined to those patients with distal cancers. The long-term results of the UKFST are awaited with interest. References: 1. Lewis JD, Ng K, Hung KE, Bilker WB, Berlin JA, Brensinger C, Rustgi AK. Detection of proximal adenomatous polyps with screening sigmoidoscopy: a systematic review and meta-analysis of screening colonoscopy. Arch Intern Med. 2003 Feb 24;163(4):413-20. 2. UK Flexible Sigmoidoscopy Screening Trial Investigators. Single flexible sigmoidoscopy screening to prevent colorectal cancer: baseline findings of a UK multicentre randomised trial. Lancet. 2002 Apr 13;359(9314):1291-300 3. Peravali R, Kandiah K, Surah A, Murria P, Taniere P, Radley S. Retrospective analysis of pre- and peri-operative imaging in confirmed proximal colonic cancers--possible implications for screening flexible sigmoidoscopy. Colorectal Dis. 2009 Feb;11(2):146-9. Competing interests: None declared |
|||
|
|
|||
|
Geir Hoff, Head of Research Telemark Hospital, 3610 Skien, Norway
Send response to journal:
|
The comments by Rajeev Peravali on flexible sigmoidoscopy (FS) screening are much welcome. The patients in their study, however, had a mean age of 75 years and there is a proximal shift in the distribution of colorectal neoplasia with increasing age.1 Apart from questioning any screening at this age, I agree that the site distribution shift of neoplasia makes flexible sigmoidoscopy screening less suitable above a certain age. Still, we are far from seeing the ideal colorectal cancer screening modality also for younger age groups. Test sensitivity is poorer for flexible sigmoidoscopy compared to colonoscopy, but may gain on programme performance by obtaining higher attendance rates. Still, we lack follow-up results from both methods to guide screening policy. Peravali refers to the UK flexible sigmoidoscopy (UKFST) trial proposing work-up colonoscopy only for flexible sigmoidoscopy screening participants found to have a high-risk adenoma and/or three or more small adenomas at screening. 2 In the NORCCAP trial we simulated what would happen if we had applied the UKFST criteria instead of “any adenoma” which triggered colonoscopy work-up in as much as 19% of NORCCAP participants. We would have reduced our colonoscopy load by impressive 73%, but would have missed 35 out of 72 cases (49%) of proximal advanced lesions (including one cancer) compared to using “any adenoma” as qualifier for colonoscopy. By not having done a full colonoscopy on our adenoma-negatives in the NORCCAP trial, it is estimated that one in 42 of these individuals would have had a proximal advanced lesion and one in 15 a proximal adenoma. 3 Comparing follow-up results of the UKFST, the similar Italian SCORE trial and NORCCAP will no doubt provide valuable information on the implications of differences in threshold for work-up colonoscopy. Intuitively, when aiming for colorectal cancer prevention by endoscopy screening and provided that on-going randomised trials come out favourably also in a public health perspective, we may end up recommending an “extended flexible sigmoidoscopy screening” using a colonoscope rather than a flexible sigmoidoscope and go as far as a limited bowel cleansing may permit, but offer “gold standard colonoscopy” for those motivated for a full bowel cleansing. 1. Peravali R, Kandiah K, Surah A, Murria P, Taniere P, Radley S. Retrospective analysis of pre- and peri-operative imaging in confirmed proximal colonic cancers--possible implications for screening flexible sigmoidoscopy. Colorectal Dis 2009;11:146-9. 2. Gondal G, Grotmol T, Hofstad B, Bretthauer M, Eide TJ, Hoff G. Grading of distal colorectal adenomas as predictors for proximal colonic neoplasia and choice of endoscope in population screening: experience from the Norwegian Colorectal Cancer Prevention study (NORCCAP). Gut 2003;52:398-40 3. Lewis JD, Ng K, Hung KE, Bilker WB, Berlin JA, Brensinger C, Rustgi AK. Detection of proximal adenomatous polyps with screening sigmoidoscopy: a systematic review and meta-analysis of screening colonoscopy. Arch Intern Med. 2003 Feb 24;163(4):413-20. Competing interests: None declared |
|||
|
|
|||
|
Hans-Hermann Dubben, Associate professor University of Hamburg, Institute of Primary Medical Care, Martinistrasse 52, 20246 Hamburg, Germany
Send response to journal:
|
In the interesting and important study on colorectal cancer screening by Hoff et al. it was reported that “no severe complications occurred during flexible sigmoidoscopy.” How were “severe complications” defined? What kind of complications occurred due to sigmoidoscopy and colonoscopy and how frequent were they? Yours sincerely,
Competing interests: None declared |
|||
|
|
|||
|
Geir Hoff, professor Cancer Registry of Norway, 0304 Oslo, Norway
Send response to journal:
|
Dear Professor Dubben, Thank you for asking these questions. We agree that complications, harms and risks should be clearly weighted against benefits of screening. This is particularly important once (or if) the expected benefit of endoscopy screening may become evident after a longer follow-up period. At this interim analysis we did not want to repeat too much information from the already published baseline data referred to in the present BMJ paper. In this publication (1) complications are described in detail. These baseline results also include the 50-54-year old add-on age cohort not yet included in our follow-up analyses. Briefly, there were no perforations, bleeding or other complications requiring hospitalization after flexible sigmoidoscopy. Out of 38 events not requiring hospitalization (0.2%), there were 26 cases of vasovagal reaction due to the on-site administration of enema or the flexible sigmoidoscopy examination itself. During colonoscopy work-up of screen-positives there were six perforations – one in 336 therapeutic colonoscopies and none in the 803 purely diagnostic colonoscopies. Additionally, four patients were admitted to hospital due to post-polypectomy bleeding, but none of them required transfusions or surgical intervention. There were 41 minor events not requiring hospitalization during the 2524 colonoscopies – including 24 cases of vasovagal reactions. For patients requiring surgery for screen-detected cancer (n=37) or complications at work-up colonoscopy (n=6) there was no post-surgical mortality, but two severe complications – one case of pulmonary embolism and one case of anastomotic leakage after resection. Yours sincerely,
1. Gondal G, Grotmol T, Hofstad B, Bretthauer M, Eide TJ, Hoff G. The Norwegian colorectal cancer prevention (NORCCAP) screening study: baseline findings and implications for clinical work-up in age groups 50-64 years. Scand J Gastroenterol 2003;38:635-42 Competing interests: None declared |
|||