Coeliac disease in primary care: case finding study
BMJ 1999; 318 doi: https://doi.org/10.1136/bmj.318.7177.164 (Published 16 January 1999) Cite this as: BMJ 1999;318:164All 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.
We read with interest the paper by Hin et al entitled "Coeliac
disease in primary care: case finding study" (1). As a result of their
study 30 patients were diagnosed with coeliac disease compared to 7 in the
local general hospital during the preceding 12 months. As a result of a
large-scale population screening survey we diagnosed 13 coeliacs and there
were 2 patients who presented clinically and were diagnosed prior to the
study follow up (2). Hin’s ratio of subclincal to clinical
coeliacs, 4.3:1, compares with our ratio (6.5:1) and gives some estimate
of the proportions of the "coeliac iceberg", at least, indicating
that the majority of patients with coeliac disease remain undiagnosed due
to oligosymptomatic or atypical forms. Our work confirms that of Hin in
that we have reported that coeliacs detected by screening are not "silent"
in the true sense of the word, but are simply unrecognised (3). With
reference to the prevalence of coeliac disease in the general adult
population, our work from a random sample of the Northern Ireland
population indicates that the condition has a prevalence of 1:122,
representing the highest reported prevalence of the disease in the
unselected adult population (2). We have no reason to believe that the
prevalence should be significantly different in the remainder of the
United Kingdom.
We would reiterate that an increased awareness of the various
manifestations of coeliac disease, coupled with a low threshold for
serological testing, will help to uncover many more coeliacs who could
benefit from treatment.
References
1 Hin H, Bird G, Fisher P, Mahy N, Jewell D. Coeliac disease in
primary care: case finding study. BMJ 1999;318:164-7.
2 Johnston SD, RGP Watson, SA McMillan, J Sloan, AHG Love. The
prevalence of coeliac disease in Northern Ireland. Lancet 1997;350:1370.
3 Johnston SD, RGP Watson, SA McMillan, J Sloan, AHG Love. Coeliac
disease detected by screening is not silent - simply unrecognised. Q J Med
1998;91:853-860.
Competing interests: No competing interests
We read with interest the paper by Hin et al on coeliac disease [1]
and agree that coeliac disease is under-diagnosed. Patients with
unexplained anaemia should certainly be investigated to exclude coeliac
disease. However, the authors screened their patients by using serum IgA
antiendomysial antibody (AEA) test. Antigliadin antibody (AGA) was
estimated in cases of IgA deficiency. Patients only had a small
intestinal biopsy if they were found to be AEA positive. The authors
conclude by recommending serological testing with AEA for patients at high
risk of having coeliac disease, including those patients with unexplained
anaemia. This may not be sufficient.
The sensitivity of AEA for coeliac disease ranges between 89 and
100%. It can be falsely negative in patients with selective IgA
deficiency (2 - 3% of patients with coeliac disease). There is also a
relative lack of sensitivity in children less than 2 years old where it
drops to 80% [2]. The sensitivity of serum AGA is much lower and it has
been reported to be positive in only 68% of untreated adults with coeliac
disease [3]. In an audit of the diagnosis of coeliac disease in our
hospital serving a population of 440,000, fifty nine new cases (median age
= 51 years; range 2 - 83) were diagnosed over 2 years (1997 - 98). All
had been anaemic or had haematinic deficiency, but only 80% of those
tested had a positive AEA.
We feel that all the patients with unexplained anaemia should be
referred for endoscopy and small intestinal biopsy, regardless of the
results of AEA or AGA,
(2)
which should be considered an essential part of investigations for
unexplained anaemia.(3)
References:
[1] Hin H, Bird G, Fisher P, Mahy N, Jewell D. Coeliac disease in
primary care: case finding study. BMJ 199; 318: 164-167.
[2] Volta U, Monlinaro N, Fusconi M, et al: IgA antiendomysial antibody
test: A step forwards in coeliac disease screening. Dig Dis Sci 1991; 36:
752-756.
[3] Volta U, Molinaro N, Fratangelo D, Bianchi FB. IgA subclass
antibodies to gliadin in serum and intestinal juices of patients with
coeliac disease. Clin Exp Immunol 1990; 80 (2) : 192 - 5.
Competing interests: No competing interests
EDITOR –
Hin et al's study highlights how serological screening for coeliac
disease in primary care can uncover a large number of patients with this
condition(1). The study was a case
finding , among nine surgeries serving a population on 70.000 (5,3 % less
than 10 years). In two years 1000 blood samples, from patients fulfilling
the entry criteria, were sent for serological screening (EMA). Their most
important findings were that coeliac disease is underdiagnosed and that
coeliac disease presents mainly with atypical simptoms such as anaemia and
feeling "tired all the time".
We performed a similar study, now submitted for publication, in a
paediatric primary care setting. 26 family paediatricians of a North-East
Italian region serving a population of 20.200 children (0-14 years of age)
selected, over two years, 240 children fulfilling the study entry
criteria. All the children were tested for EMA or AGA IgG if IgA deficent
(and biopsied if positive to the serological test). Unrecognized coeliac
disease prevalence in the 240 children selected was 7,5 %. After the
study coeliac disease prevalence among the registred population varied
form 1:2.020 to 1:721. The most frequent primary presentation was anaemia
(23,5 %), growth stature defect (7,6 %) without gastrointestinal symptoms.
As Hin et al, we want to stress the importance to recognize coeliac
disease in a primary care setting expecially in the paediatric population.
General practitioners must be aware that also children increasingly
presents with atypical symptoms(2). The study was a case finding , among
nine surgeries serving a population on 70.000 (5,3 % less than 10 years).
In two years 1000 blood samples, from patients fulfilling the entry
criteria, were sent for serological screening (EMA). Their most important
findings were that coeliac disease is underdiagnosed and that coeliac
disease presents mainly with atypical simptoms such as anaemia and feeling
"tired all the time".
We performed a similar study, now submitted for publication, in a
paediatric primary care setting: 26 family paediatricians of a North-East
Italian region serving a population of 20.200 children (0-14 years of age)
selected, over two years, 240 children fulfilling the study entry
criteria. All the children were tested for EMA or AGA IgG if IgA deficent
(and biopsied if positive to the serological test). Unrecognized coeliac
disease prevalence in the 240 children selected was 7,5 %. After the
study coeliac disease prevalence among the registred population varied
form 1:2.020 to 1:721. The most frequent primary presentation was anaemia
(23,5 %), growth stature defect (7,6 %) without gastrointestinal symptoms.
As Hin et al, we want to stress the importance to recognize coeliac
disease in a primary care setting expecially in the paediatric population.
General practitioner must be aware that also children increasingly
presents with atypical symptoms(2). The higly sensitive and
specific EMA serological test, simple and cheap, should be asked for in
any child with gastrointestinal and extraintestinal symptoms related to
coelic disease.
Alessandro Ventura, Professor of Paediatrics
Sergio Facchini, Research Fellow
Istituto per l'Infanzia "Burlo Garofolo",
Department of Clinical Paediatrics,
University of Trieste,
34100 Trieste,
Italy
Paediatric Research Association (A.P.R.), ASL 4,5,6 Vicenza, Italy
(Members : E Amantidu, MF Andreotti, A Andrighetto, A Baggiani, F Benetti,
S Bonati, I Brusaterra, M Capozzo, E Ciscato, F Cracco, G Ferrari, M
Fomale, F Fusco, E Laverda, M Marchioro, E Nicolussi, A Pasinato, D
Pittarello, E Pizio, R Salvadori, D Sambugaro, S Sassolino, V Spavenello,
C Trevisan, G Ziglio, V Zuffellato).
1)Hin H, Bird G, Fisher P, et al. Coeliac disease in primary care:
case finding study. BMJ 1999;318:164-167
2)Ventura A, Magazzù G, Greco G, et al. Autoimmune disorders in
coeliac disease: relationship with duration of gluten exposure. J Paediatr
Gastroenterol Nutr 1997;24:463
3)Cataldo F, Ventura A, Lazzari R, et al. Antiendomysium antibodies
and coeliac disease: solved and unsolved questions. An italian multicentre
study. Acta Paediatr 1995;84:1125
Competing interests: No competing interests
Sir
We agree with Hin et al. (BMJ January 16th), that occult coeliac
disease (CD) is common in cases of unexplained anaemia, and that testing
for CD by looking for IgA anti-endomysial antibodies (used to such good
effect by Hin et al.) is under-utilised. We have been studying blood donor
volunteers, turned away by The National Blood Authority (NBA) because
they are unexpectedly found to be anaemic in routine pre-donation
screening. Routine NBA procedure, is to contact the general practitioner
in all cases where females volunteers with a haemoglobin of less than 11.5
gm/l, and all males with a value of 12.5 gm/l. In our initial pilot study,
we found 7% (32) of 483 anaemic blood donors tested, to be IgA anti-
endomysial antibody seropositive, and therefore likely to have CD.
After a time delay of 3-6 months, we sent out questionnaires to all
of our endomysial antibody seropositive anaemic subjects. Of 22
responders, none had been considered for CD serology. Indeed, in 14/22, no
follow up tests of any sort had been organised, and in 18/22, oral iron
had been prescribed empirically with no attempt to explain the anaemia.
Hin et al (1) found 3% (30/1000) of a pre-selected, mostly symptomatic
study group had CD. Our "patients", all felt well enough to donate blood
and had no specific health complaints. Targeted serological testing,
particularly focusing on cases of unexplained anaemia, offers a simple
and worthwhile way to diagnose unsuspected CD in a large number of cases.
1) Hin H, Bird G, Fisher P, Mahy N, Jewell D. Coeliac disease in
primary care : case finding study. BMJ 1999;318: 164-167.
R HARVEY (MD FRCP)
Consultant Gastroenterologist
Frenchay Hospital
Bristol BS16 1LE
R.J. LOCK (BSc. MPhil)
Scientific Officer, Immunology
Southmead Hospital
Bristol BS10 5ND
D J UNSWORTH (BSc PhD MRCP MRCPath)
Consultant Immunologist
Southmead Hospital
Bristol BS10 5ND
We the above signed, declare that we have participated in the design,
execution, and analysis of the data summarised in our letter , and have
seen and approved the final version. We know of no conflict of interest
in connection with this letter. (February 98).
Competing interests: No competing interests
Editor
Hin et al state "the most likely presentation [of coeliac
disease] is a combination of microcytic anaemia ... and
feeling tired" (1). While agreeing with this assertion, we
have found that other haematological abnormalities, such as a persistent
macrocytosis, are often the main presenting
feature. Of 127 consecutive adult patients with
histologically proven coeliac disease presenting to our
hospital, 51 had haematological abnormalities as the only or
main presenting feature: 30 had chronic iron deficiency
anaemia, 7 had a macrocytic anaemia due to folate or vitamin
B12 deficiency, and 14 had an isolated persistent
macrocytosis, often accentuated during pregnancy. The
presence of even minor haematological abnormalities, if
persistent, should alert the clinician to the possibility of
coeliac disease.
Michael Mitchell,
Specialist Registrar,
Department of Medicine,
Institute of Clinical Science,
Royal Victoria Hospital,
Grosvenor Road
Belfast BT12 6BA
John Robinson,
Consultant Gastroenterologist,
Department of Medicine,
Craigavon Area Hospital,
Portadown Road,
Portadown
(1) Hin H, Bird G, Fisher P, Mahy N, Jewell D. Coeliac
disease in primary care: a case finding study. BMJ 1999; 318:
164-7.
Competing interests: No competing interests
This study has confirmed that positive endomysial antibody tests will
result in positive biopsies for celiac disease. This shows us that
positive blood tests are reliable. However, this study does not tell us
whether any of the participants who had negative blood tests had positive
biopsies. This would have showed us the reliability of negative
endomysial antibody tests.
Since your study did not do biopsies on all participants in the
study, you cannot say how reliable the negative blood tests were. I would
like to see you adjust your findings to show that positive blood tests are
reliable for diagnosing celiac, and that negative blood tests need to be
studied further.
Two well known Doctors in the United States, Dr Peter Green and Dr
Joseph Murray, have found in their private practices that up to 50% of
their symptomatic patients with positive biopsies, have negative blood
tests.
Competing interests: No competing interests
Editor- Hin et al demonstrated in this issue that primary care
screening by endomysial antibody is helpful in detecting previously
undiagnosed Coeliac disease. Positive endomysial antibody although has a
high sensitivity, is not always present in untreated Coeliac disease. We
have recently seen two elderly patients with Coeliac disease where
antiendomysial antibody was negative.
Case-1: A 62-year-old woman presented with unexplained weight loss of 10 Kg
for two years. Her BMI was 18, she was a known patient with primary
hypothyroidism and pernicious anaemia. Clinical examination, routine
biochemistry, short synacthen test, haematinics were unremarkable. Both
antigliadin and antiendomysial antibobdies were negative. Malabsorption
studies showed only abnormality in absorption of fat(Light scattering
index of seven, normal >20). Jejunal biopsy showed total villous
atrophy, on gluten withdrawal she gained eight Kg weight in next 12 months
and repeat jejunal biopsy showed normal villous structure.
Case-2: A 63-year-female presented with severe microcytic anaemia with a
haemoglobin of 5.6 gm/dl. Her serum ferritin was low ( 3, normal 8-186
mcg/L)and d-Xylose was abnormal(0.8, normal 1-2 gm). Antigliadin antibody
was positive but antiendomysial was not. She had subtotal villous atrophy
which responded to gluten withdrawal.
In conclusion, it is important to consider the possobility of Coeliac
disease even when the antobody screening is negative as a high index of
suspicion can lead to the correct diagnosis and treatment.
Competing interests: No competing interests
Hin et al. are correct in concluding from their study that the
specificity of the endomysial antibody test was 100%, but are wrong in
quoting the figure of 30/30. The figure of 30/30 refers to the test’s
positive predictive value. The specificity of a test is defined as the
proportion of the population without the condition who test negative.
Because we know that zero patients without the condition tested positive
the proportion must be 100%, even though the actual numbers are not known.
This set of circumstances, of there being zero false positive
results, is exceptional. Any other figure would mean that specificity
could not be calculated without ascertaining the true status, in this
example by biopsy, of all those tested.
Competing interests: No competing interests
Editor-
Hin et al demonstrate that primary care screening by endomysial antibody
(EmA) will uncover many patients with coeliac disease who might otherwise
go undiagnosed [1]. EmA testing has been available to general
practitioners in Northern Ireland since 1991.
We retrospectively reviewed
its
use by GPs in the west of Northern Ireland (catchment population 270,000)
from 1994-1996 inclusive [2]. Of 181 patients without known coeliac
disease and with data available (of a total of 239 tested), 20 (11%) were
EmA +ve. Nineteen subsequently had small bowel biopsy which confirmed
villous atrophy for all: the twentieth was started on a gluten-free diet
without biopsy after referral to a general
medical clinic. Only 44% (79/181) of patients tested and 35% (7/20) of EmA
+ve patients reported diarrhoea, while 14% (6/42) of patients with anaemia
and no diarrhoea had EmA. Over the same period, 67 patients were
diagnosed coeliac
by hospital doctors.
Our GPs are already aware that coeliac disease may
present without diarrhoea and are making an important contribution to the
identification of coeliac
disease, which is also underdiagnosed and misdiagnosed by hospital
doctors[3].
William Dickey,
Consultant Gastroenterologist,
Altnagelvin Hospital,
Londonderry BT47 6SB
Stanley McMillan,
Consultant Clinical Scientist,
Regional Immunology Laboratory,
Royal Group of Hospitals,
Belfast BT12 6BA
1. Hin H, Bird G, Fisher P, Mahy N, Jewell D. Coeliac disease in
primary care:case finding study. BMJ 1999; 318: 164-7 (16 Jan).
2. Dickey W, McMillan SA, Hughes DF. Identification of coeliac disease in
primary care. Scand J Gastroenterol 1998; 33: 491-3.
3. Dickey W, McConnell JB. How many hospital visits does it take before
celiac sprue is diagnosed? J Clin Gastroenterol 1996; 23: 21-3.
Competing interests: No competing interests
The relevance of age in screening CD
27th April 1999
The Editor
BMJ
Dear Sir,
We read with interest the paper by Hin et al. describing the prevalence of
coeliac disease (CD) in a population attending a general practice [1].
Although this paper will greatly help to diffuse among general
practitioners the concept that CD must not be suspected only in patients
presenting with a classical malabsorption syndrome, Hin et al., who
screened subjects aged between 6 months and 85 years, provided only little
information regarding the possibility that different age groups could have
different prevalences of CD.
In an adult general population we found four patients affected by CD
out of 2237 screened subjects [2]. Interestingly, none of the four
patients was aged over 40 years. Thus, in the adult general population
aged under 40 years, the prevalence of CD was 1/230 (95% Confidence
Interval 1.7-11.1‰), which is very similar to the prevalence shown by
Catassi in a neighbouring paediatric population [3]. On the other hand,
the prevalence of CD was much lower in the subjects aged both 40-59 years
(95% CI 0.0-4.8‰) and over 60 years (95% CI 0.0-7.5‰). We concluded that
the prevalence of CD is not the same in different age groups. It is likely
that, similarly to our results, the prevalence of CD was not uniformly
distributed in the sample studied by Hin. It would greatly help general
practitioners, willing to screen for CD the patients attending their
practice, to know whether different age groups have different prevalence
of CD.
It was a reason of concern to read that none of the patients
considered to be affected by irritable bowel syndrome (IBS) resulted to be
affected by CD. We showed that many patients affected by classical CD,
i.e. presenting with symptoms of frank malabsorption, had had relevant
diagnostic delay because of the fact that they had long been labeled as
affected by IBS [4]. Moreover, it emerges from Hin's paper that the
prevalence of CD among patients presenting with "malabsorption symptoms or
diarrhoea" was around 5%. Since we feel that the distinction between
patients affected by IBS or "malabsorption symptoms or diarrhoea" could be
not immediately recognised by general practitioners without a specific
interest in gastroenterology, it is crucial to know the diagnostic
criteria used by the authors to include the patients in either of the
groups in order to avoid perpetuating diagnostic delay in patients
affected by classical CD.
1. Hin H, Bird G, Fisher P, Mahy N, Jewell D. Coeliac disease in
primary care: case finding study. BMJ 1999;318:164-7.
2. Corazza GR, Andreani ML, Biagi F, Corrao G, Pretolani S, et al. The
smaller size of the "coeliac iceberg" in adults. Scand J Gastroenterol
1997;32:917-9.
3. Catassi C, Rätsch I-M, Fabiani E, Rossini M, Bordicchia F, Candela F,
et al. Coeliac disease in the year 2000: exploring the iceberg. Lancet
1994;343:200-3.
4. Corazza GR, Brusco G, Andreani ML, Biagi F, Di Stefano M, Gasbarrini G.
Previous misdiagnosis and diagnostic delay in adult celiac sprue. J Clin
Gastroenterol 1996;22:324-5.
Dr. Federico Biagi, MD
Prof. Gino R Corazza, MD
Gastroenterology Unit
University of Pavia
IRCCS Policlinico San Matteo
Piazzale Golgi, 5
27100 Pavia, Italy
Tel. 0039 0382 502974
Fax 0039 0382 502618
Competing interests: No competing interests