Risk in primary care of colorectal cancer from new onset rectal bleeding: 10 year prospective study
BMJ 2006; 333 doi: https://doi.org/10.1136/bmj.38846.684850.2F (Published 06 July 2006) Cite this as: BMJ 2006;333:69
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Digital Rectal Examination is relevant in every medical practice.
Digital palpitation of the rectum can often find abnormal growths
which may require further testing or commonplace hemorrhoids. It is a
critical initial clinical test and is important in the assessment of the
size and location of tumors.
DRE has also been used as a screening tool for prostate cancer. It
seems to be very effective for larger masses found in the prostate and
correlated well with higher prostate-specific antigens.
More gastroenterologists are recommending that pediatricians and family
physicians perform DRE on pediatric patients exhibiting chronic
constipation before those patients are referred to intestinal specialists.
The pediatrician or family physician could identify fecal compaction and
treat it themselves, and then only refer patients who have a specific
abnormality to gastroenterologists.
DRE is here to stay.
Competing interests:
None declared
Competing interests: No competing interests
I read this article with great interest.Being in surgical practice in
a community based setting and teaching undergraduate medical students
partime, I am very concerned after reading this article.
It is very laudale indeed for this study to be undertaken in the primary
care setting in the rural area.There so much clinical material in the
rural setting which could contribute to the development of knowledge in
the clinical practice of any community.
It would be more useful if the authors categorize the colorectal tumours
detected in their setting according to the anatomic location in relation
to the distance from the anal verge.
There is no mention in the methodology of this study about Digital Rectal
Examination (DRE).This is rather surprising since this clinical
examination should have been done prior to any anorectal investigation
especially in the primay care setting.
We have been teaching the undergraduate students that "if one doesn't put
the finger in the rectum, one might end up putting the foot in".One
wonders whether this adage is still relevant in the modern practice of
medicine where flexible endoscopy is easily available even in the rural
primary care setting?
Endoscopy entails cost and it is not without its fare share of morbidity
and even mortality.Availabilty of endoscopy should not have been an excuse
for a doctor not to do DRE.It is understandable that DRE is very aversive
to the patients and doctors alike.
It is however still mandatory for all patients with rectal bleeding who
are not going to be subjected to some form of endoscopy soon.
This paper has made Practice guideline by NICE as the standard of clinical
practice to measure by.This is perhaps another example whereby clinical
practice guidelines which based its recommendation largely from
quantitative studies could give rise to potential error in clinical
practice on individual patient.
Competing interests:
None declared
Competing interests: No competing interests
Unfortunately, du Toit and colleagues do not provide any detail of
the 15 cancers and 13 adenomas discovered by investigating fully every
patient presenting with new onset rectal bleeding of any type,
irrespective of other symptoms
Consequently it is not possible to determine how many of the positive
findings were serendipitous (the target of the imminent national screening
programme) and how many were actually responsible for the presenting
symptoms. This important distinction appears to have been missed by the
authors of both the original paper and the accompanying editorial.
I would not wish to deter any patient with recent onset bleeding from
presenting to their GP and being referred on for appropriate investigation
but we should be clear about the aims of the investigation - opportunistic
or symptom - diagnostic. Only then can a rational (health economics-
based) decision be taken on whether funding should be directed at
detecting patients with haemorrhoids (and thus denying presymptomatic
diagnosis to those without).
Competing interests:
None declared
Competing interests: No competing interests
Du Toit and colleagues 1 on the basis of a 10-year study in a single
practice on 265 patients presenting with rectal bleeding (27 per year)
amongst whom 15 had cancer (1 to 2 per year) suggested that all patients
over the age of 45 presenting with new onset rectal bleeding should be
referred and investigated in hospital. Overall 5.7% had cancer falling to
2.4% (3/126) in patients below the age of 64 years. Only 2 of the 15
cancer patients had ‘diarrhoea’ but no information was given on how
comprehensive a history was taken. These findings are rather different to
other studies 2-4 which have shown that over 75% of cancer patients
presenting with rectal bleeding have an associated change in bowel habit
and over 65% do not have anal symptoms, which suggests their subset of
patients may not be typical.
In a prospective study of 4046 patients referred to hospital with
rectal bleeding 5 the predictive value for cancer in patients aged between
45 and 64 without the NICE Guidelines was 2% (38/1869; 95% CI 1.4 to 2.8).
The non-selective policy proposed by du Toit is likely to result in
the referral of large numbers of patients to the advantage of those who do
not have cancer at the expense of those that do. We need larger more
representative studies to confirm whether their findings should lead to a
change in the current NICE guidelines, which advocate a selective policy.
Competing interests:
None declared
Competing interests: No competing interests
Editor,
We read with interest the recent paper by du Toit et al detailing
risk of colorectal cancer from rectal bleeding within primary care.
Interestingly, although the title refers to colon cancer risk their
conclusion is based on colonic neoplasia. Of particular interest was their
statement that one in 10 (10%) of patients aged 45 or more with new onset
rectal bleeding had colonic neoplasia. This figure seems to be a gross
underestimate and is in contrast to our findings[1,2] and others[3,4].
Furthermore in this paper by du Toit et, al all but 2 patients reported
rectal bleeding alone
i.e. isolated rectal bleeding. There is little doubt that patients
presenting with isolated rectal bleeding require investigations given that
the risk of colorectal cancer is between 5-11% [1,3-4]. As is often the
problem in most studies involving rectal bleeding, defining rectal
bleeding has been difficult but more importantly the inclusion of other
accompanying symptoms in patient selection can often be misleading.
Moreover since most cancers and adenomatous polyps arise in the left
colon, the more pertinent question is what the most appropriate
investigation should be.
Our experience within a teaching hospital setting (where just over 50
% were direct referrals from primary care) has been quite different.
Endoscopic data on patients attending for colonoscopy was collected
prospectively over 39 months. 6300 colonoscopies were performed during
this period of which 1445 were performed for isolated rectal bleeding (55%
female; mean age 58 years (range 18 - 94)). Colorectal cancer was
identified in 85 patients (6%) whilst in those less than 50 years, there
were only 4 cases (0.2%). Adenomatous polyps were found in 212 patients
(15%), the majority of whom were over 50 years[1] (see Table 1). Thus in
our series, one in five (22%) patients over 50 years developed colonic
neoplasia. These results are comparable to the equally large American
cohort[3] (1766 patients) and even the smaller Northern England primary
care study[4] (99 patients).
Table 1 (N=1445)
Colorectal cancer: > 50yr = 81 (5.6%); < 50yr = 4 (0.2%)
In-situ carcinoma: > 50yr = 12 (0.8%); < 50yr = 0
Adenomatous polyp: > 50yr = 175 (12%); < 50yr = 37 (2.6%)
IBD: > 50yr = 47 (3.3%); < 50yr = 28 (1.9%)
It is notoriously difficult to predict the source of rectal bleeding
(luminal/colonic or outlet/anal) bleeding from the clinical history alone.
The Australian primary care study by Goulston et al[5] showed that
Gastroenterologists and Colorectal surgeons were only slightly better than
general practitioners at predicting source of bleeding in those over 40
years. As clinical history cannot be relied on solely, further
investigations are necessary especially in those over 50 years.
We have shown that investigation of patients presenting with isolated
rectal bleeding is justified especially in those over 50 years[1-2]. The
variation in reported[4-6] cancer rates in those presenting with isolated
rectal bleeding may be explained by small sample sizes as well as the
inclusion of patients with other accompanying symptoms such as diarrhoea.
The conundrum however is not whether to investigate isolated rectal
bleeding in those over 45 or 50 years but rather, how best to investigate
i.e. to confine examination to just the left colon (Flexible
Sigmoidoscopy) or the whole colon (Colonoscopy). The implications of this
are far reaching given the advent of colonoscopy screening programmes
coupled by budget constraints within the NHS.
References:
1.Arasaradnam RP, Gopal ST, Donnelly MT et al. Polyp prevalence in a
cohort of 1445 patients with Isolated Rectal Bleeding (IRB) – What are we
missing with the Flexible Sigmoidoscope (FS)?
Gastrointestinal Endoscopy April 2005 Volume 61: (5);247
2.Arasaradnam RP, Donnelly MT & Skelly MM. Isolated Rectal
Bleeding (IRB) – The Right & Left Conundrum. Gut 2005;54:A12(043)
3.Mulcahy HE, Patel RS, Postic G et al. Yield of colonoscopy in
patients with non-acute rectal bleeding: Multicentre database study of
1766 patients. Am J Gastroenterol 2002 Feb;97(2):223-5
4.Metcalf JV, Smith J, Jones R et al. Incidence and causes of rectal
bleeding in general practice as detected by colonoscopy. Br J Gen Prac
1996;46:161-164
5.Goulston KJ, Cook I, Dent OF. How important is rectal bleeding in
the diagnosis of bowel cancer and polyps? Lancet 1986 Aug 2;2(8501):261-5
6.Ellis BG, Thompson M. Factors identifying higher risk rectal
bleeding in general practice. Br J Gen Prac 2005;55:949-55
.
Competing interests:
None declared
Competing interests: No competing interests
Dear Editor,
We congratulate du Toit et al. for their research emphasising the
importance of rectal bleeding in primary care and trying to address this
important management dilemma. A standardised approach to investigation and
care is to be encouraged as is the refusal to accept the first apparent
diagnosis of ‘haemorrhoids’ as the source of bleeding if significant
pathologies are not to be missed.
It seems harsh to criticise such a fine prospective primary care
study, but one potential weakness was its assumption that bleeding was
caused by any neoplasia found, rather than more common causes. The
character of rectal bleeding (bright red versus dark) was not mentioned,
nor was the coexistence of piles ; probably as rigid sigmoidoscopy rather
than the necessary proctoscopy was carried out in study methodology. In
addition the presence of anal symptoms, usually due to piles, which is a
protective factor to the finding of neoplasia was not said to be recorded.
As Professor Wellers editorial in the same issue hints, the specific
characteristics of the bleeding is important to record; if the colour
shade of the blood was recorded it may help plan investigations. Bright
red bleeding is well investigated by flexible sigmoidoscopy but it is
generally agreed that darker bleeding usually requires colonoscopy. The
paper did not report why three different modes of investigation were
employed, nor the rationale for choosing between the three.
The location of the cancers was not stated but should have been, it is
unlikely that caecal cancer would cause bright rectal bleeding. Finally
the size and number of adenomas are necessary information to estimate
whether these lesions are significant in a study population largely over
65 years old.
If the nature of the bleeding and presence of bleeding piles is to be
ignored and a one-off bleed results in booking an endoscopy then up to 20%
of the adult population (the widely accepted incidence within a population
not those presenting) would need an endoscopy every year. Luckily most
patients rather than having new rectal bleeding have chronic bleeding and
in reality neoplastic causes persist whereas benign causes abate, which
allows some sorting into who to investigate with what urgency
Competing interests:
None declared
Competing interests: No competing interests
Sir,
More than 300 years ago, John Locke warned of the effects of the misuse of
language: “… the errors and obscurity, the mistakes and confusion that is
spread in the world by an ill use of words…” [1] If the evidence of the
article by du Toit et al. [2] is anything to go by, the philosopher’s
message appears to be in need constant reinforcement.
The context is set by the title of their paper which explicitly
refers to “colorectal cancer”. [2] However, the conclusion of the abstract
– later repeated in the discussion and the summary box – is that one in
ten patients aged 45 years or more with new onset rectal bleeding had
“colonic neoplasm” or “colorectal neoplasia”. Although less than 6% of
patients had cancer, substituting the combined figure for both cancer and
adenomas yields a more impressive “one in ten”. But, in their eagerness to
present their results in the best light, du Toit et al. simply end up by
being misleading. And, the proof of this is in the very same edition of
the BMJ. Weller, [3] in the accompanying leading article, stated that “The
study found that about one in 10 patients with new onset rectal bleeding
had cancer”. This is obviously untrue. If a reviewer, whose job it is to
study the paper in question carefully, can make such a mistake, then
questions must be raised about the way in which du Toit et al. presented
their data.
References
[1] Locke J. An Essay Concerning Human Understanding, 1690. Book III,
Chapter XI.
[2] Du Toit J, Hamilton W, Barraclough K. Risk in primary care of
colorectal cancer from new onset rectal bleeding: 10 year prospective
study. BMJ 2006;333;69-70.
[3] Weller D. Colorectal cancer in primary care. BMJ 2006; 333;54-5.
Competing interests:
None declared
Competing interests: No competing interests
Dear Sir,
Dr. du Toit and colleagues have highlighted an important issue in
this week’s BMJ.[1] Rectal bleeding is a common presentation of many
benign colorectal pathologies but also malignancy. The finding of a one in
ten chance of patients over the age of 45 presenting with rectal bleeding
and subsequently being diagnosed with colorectal cancer is comparable to
other recent studies[2], however these latest results are from data
collected from a single rural practice in the UK, including less than 300
patients. A recent a population-based, case-control study consisting of
over 1,500 patients has demonstrated that residence in a rural area was
associated with increased colon cancer risk (odds ratio, 1.4; 95%
confidence interval, 1.1-1.8)[3] However a large epidemiological study of
approximately 500,000 patients suggests that black men who reside in
metropolitan areas have a higher risk of colorectal cancer than black men
who reside in rural areas[4]. Both these studies have evaluated
populations in the USA, however UK populations are similar. It is
therefore important to consider both urban and rural cohorts for such
studies for precise positive predictive values that may be useful for the
population in general.
Furthermore, du Toit et al.’s findings also bring into question which
referral pattern to utilise for these young patients with rectal bleeding
only – should they all be referred under the two week wait rule? This
would greatly increase the numbers of patients referred as urgent cases to
gastroenterologists and colorectal specialists however may increase the
yield of cancers diagnosed. It may also reduce the time to treatment as it
has been shown that referral lag time may be a significant area that needs
to be reduced for quicker treatment for colorectal cancers in the UK.[5]
With targets such as eighteen weeks and the sixty-two day rule for
treatment and with screening for colorectal cancer starting this year, the
burden on the endoscopy services is potentially huge. However even in a
rationed system such as the NHS that is no reason for reduced, or
substandard diagnostics. However results from UK based studies such as du
Toit et al.’s work should encourage the powers that be to consider further
organisation of lower GI endoscopy along with collaboration with
radiological diagnostics such as CT colonography to ensure that efficient
diagnosis of colorectal cancer continues in the UK as referrals increase.
References
1.Toit JD, Hamilton W, Barraclough K. Risk in primary care of
colorectal cancer from new onset rectal bleeding: 10 year prospective
study. BMJ. 2006 Jun 21
2.Ellis BG, Thompson MR. Factors identifying higher risk rectal
bleeding in general practice. Br J Gen Pract. 2005 Dec;55(521):949-55.
3.Kinney AY, Harrell J, Slattery M, Martin C, Sandler RS. Rural-urban
differences in colon cancer risk in blacks and whites: the North Carolina
Colon Cancer Study. J Rural Health. 2006 Spring;22(2):124-30.
4.Coughlin SS, Costanza ME, Fernandez ME, Glanz K, Lee JW, Smith SA,
Stroud L, Tessaro I, Westfall JM, Weissfeld JL, Blumenthal DS. CDC-funded
intervention research aimed at promoting colorectal cancer screening in
communities. Cancer. 2006 Jun 26; [Epub ahead of print]
5.Flashman K, O'Leary DP, Senapati A, Thompson MR. The Department of
Health's "two week standard" for bowel cancer: is it working? Gut. 2004
Mar;53(3):387-91.
Competing interests:
None declared
Competing interests: No competing interests
Dear Sir,
The article 'Risk in primary care of colorectal cancer from new onset
rectal bleeding: 10 year prospective study' made an interesting read.
It is well known that rectal bleeding can be a symptom of colonic
neoplasm, however the number of people who have rectal bleeding and have a
colonic neoplasm is thought to be low.
The incidence tends to rise with age. The number of people we see in
our clinics with rectal bleeding particularly under the urgent category
are usually elderly, but slowly the number of young people are rising as
more IBD cases are being picked up and also as the threshold of referral
falls.
Most of the young people with IBD however have diarrhoea in addition
to bleeding PR but some can present with simple blood in stool
particularly if the inflammation is confined to lower rectum. Some simply
have piles.
The view that bowel investigations should be offered to everyone with
rectal bleeding over age of 45 years will be difficult to justify on
present evidence and also to implement in terms of manpower and costs.
There is not enough evidence presently to suggest the above
recommendation but it highlights the need for further studies.
It may also mean many people will be investigated who do not have to
be investigated exposing them to unnecessary risk(both colonoscopy and
sigmoidoscopy have procedural risks in addition to anaesthetic risks) and
discomfort plus distress(these procedures are at times in best hands
uncomfortable)
The better way to proceed is probably take a holistic view of the
patient keeping in view their age,other symptoms ,lab tests etc and
develop a scoring and referral system in the practices.This can be made
after consultation with the specialists in the field in respective trusts
which will be based on national guidelines modified to suit local needs.
I agree that it is very difficult to pick up patients with cancer in
primary care as most people present with nonspecific symptoms.But
guidelines are of immense help here.The situation becomes more difficult
for younger people as we know incidence of cancer is less but if it is
missed it can have devastating consequences.
Lastly we all come across that odd patient who does not fufill any
referral criteria but we know something is wrong and we feel the need to
investigate.Guidelines can never replace experience and a good clinical
eye but augment it.
Competing interests:
None declared
Competing interests: No competing interests
Authors' response
We recognise the limitations of our study and appreciate the
need not to overwhelm endoscopic/diagnostic services or investigate patients
needlessly. However our study is one of a number of studies in primary care
that have shown a significant association of new onset rectal bleeding with
colorectal carcinoma (see table).
Study
Number patients
Age
% with CRC and rectal bleeding
Goulston et al 19861
145
> 40
10.3
Fitjen et al 19952
81
136
> 50
> 40
11.0
6.6
Metcalf et al 19963
99
> 40
8.1
Norrelund & Norrelund 19964
208
> 40
15.4
Wauters et al 20005
245
> 50
10.6
Ellis & Thomson 20056
319
> 35
3.4 †
du Toit et al 20067
265
> 45
5.7
† % with CRC and rectal bleeding 5.2% for patients > 60years
age (no numbers given).
As yet there has been little discussion about the level of
risk that necessitates investigation, either urgently or non
urgently. Most of the studies above appear to suggest that a patient with any
rectal bleeding over the age of 45-50 years who presents in primary care has more
than a 5% chance of having colorectal carcinoma. We consider that this level of
risk warrants investigation.
Reference List
1. Goulston
K,.Dent O. How important is rectal bleeding in the
diagnosis of bowel cancer or polyps? The
Lancet 1986;261-5.
2. Fijten
G, Starmans R, et al. Predictive value of signs and symptoms for colorectal
cancer in patients with rectal bleeding in general practice. Family Practice 1995;12:279-87.
3. Metcalf
J et al. Incidence and causes of rectal bleeding in general practice as
detected by colonoscopy. British Journal
of General Practice 1996;46:161-4.
4. Norrelund
N. et al Colorectal cancer and polyps in patients aged 40 years and over who
consult a GP with rectal bleeding. Family
Practice 2006;13:160-5.
5. Wauters
H, Van Casteren V, Buntinx F. Rectal bleeding and colorectal cancer in general
practice: diagnostic study. BMJ 2000;321:998-9.
6. Ellis
B,.Thompson M. Factors identifying higher risk rectal
bleeding in general practice. British
Journal of General Practice 2005;55:949-55.
7. du Toit J, Hamilton W, Barraclough K. Risk in
primary care of colorectal cancer from new onset rectal bleeding: 10 year
prospective study. BMJ 2006;333:69-70.
Competing interests:
None declared
Competing interests: Response