Associated disorders in celiac disease

To Elinor & David

Örebro Studies in Medicine 27

Peter Elfström

Associated disorders in celiac disease

© Peter Elfström, 2009 peter.elfstrom@orebroll.se

Title: Associated disorders in celiac disease.

Publisher: Örebro University 2009 www.publications.oru.se

Editor: Maria Alsbjer maria.alsbjer@oru.se

Printer: Intellecta Infolog, V Frölunda 01/2009

issn 1652-4063 isbn 978-91-7668-649-2

AbstractAbstractAbstractAbstract

Peter Elfström (2009): Associated disorders in celiac disease.

Örebro Studies in Medicine 27, 83 pages.

Background: Celiac disease (CD) is an autoimmune disorder that affects genetically

susceptible individuals and is induced by dietary gluten. Treatment consists of a life-

long gluten-free diet. CD is common and affects about 1% of the general population.

The classic symptoms include diarrhea and malabsorption, but many patients have

only mild symptoms or no symptoms at all. The proportion of individuals presenting

with atypical symptoms or discovered only when investigating an associated condi-

tion of CD is increasing.

Aims: The aim of this thesis was to investigate the risk of possible associated disor-

ders through Swedish population-based registers. The objective was to gain more in-

formation on the consequences of having CD and to identify high risk groups where

screening may be considered.

Materials and methods: We used the Swedish hospital discharge register to examine

the risk of liver disease, autoimmune heart disease, Addison’s disease and thyroid

disorders in a cohort of about 14,000 individuals with CD and an age and sex-

matched reference population of 70,000 individuals. In the last study we used all re-

gional pathology registers and the cancer registry to examine the risk of hematopoi-

etic cancer, including lymphoma in three different cohorts: I) 28,810 individuals with

CD; II) 12,681 individuals with small intestinal mucosal inflammation but without

villous atrophy; and III) 3552 individuals with latent CD (a positive serology test for

CD with a normal small intestinal biopsy).

Results: CD is statistically significantly associated with an increased risk of liver dis-

ease, Addison’s disease, thyroid disease and lymphoma. We also found an increased

risk of lymphoma in individuals with small intestinal mucosal inflammation. There

was no statistically significant association between autoimmune heart disease or

leukemia and CD. Latent CD was not associated with any hematopoietic cancers.

Conclusion: This thesis found a positive association between CD and a number of

autoimmune and inflammatory disorders. Clinicians need to have a high awareness

of this association and to test for these conditions when symptoms appear.

Keywords: Addison; autoimmune; biopsy; celiac; child; cohort study; cancer; heart;

liver; lymphoma; thyroid.

List of papersList of papersList of papersList of papers

I) JF Ludvigsson, P Elfström, U Bromé, A Ekbom, S Montgomery.

Celiac disease and risk of liver disease - A general population based study.

Clinical Gastroenterology Hepatology. 2007 Jan; 5(1): 63-69.

II) P Elfström, A Hamsten, S Montgomery, A Ekbom, JF Ludvigsson.

Cardiomyopathy, pericarditis and myocarditis in a population-based cohort

of inpatients with coeliac disease.

Journal of Internal Medicine. 2007 Nov; 262(5): 545-54.

III) P Elfström, S Montgomery, O Kämpe, A Ekbom, JF Ludvigsson.

Risk of primary adrenal insufficiency in patients with celiac disease.

Journal of Clinical Endocrinology and Metabolism. 2007 Sep; 92(9): 3595-8.

Copyright 2007, The Endocrine Society.

IV) P Elfström, S Montgomery, O Kämpe, A Ekbom, JF Ludvigsson.

Risk of thyroid disease in individuals with celiac disease.

Journal of Clinical Endocrinology and Metabolism. 2008 Oct; 93(10): 3915-

21. Copyright 2008, The Endocrine Society.

V) P Elfström, F Granath, K Ekström Smedby, SM Montgomery, J Askling,

A Ekbom and JF Ludvigsson.

Hematopoietic cancer including lymphoma in celiac disease according to

Marsh criteria 0-3.

Manuscript.

All previously published papers were reproduced with permission from the publisher.

List of abbreviationsList of abbreviationsList of abbreviationsList of abbreviations

CD Celiac disease

HR Hazard ratio

OR Odds ratio

AGA Antigliadin antibodies

tTGA tissue Transglutaminase Antibodies

EMA Endomysial antibodies

VA Villous atrophy

GFD Gluten-free diet

AD Addison’s disease

TD Thyroid disease

PSC Primary sclerosing cholangitis

PBC Primary biliary cirrhosis

T1DM Type 1 Diabetes Mellitus

NHL Non Hodgkin’s Lymphoma

HL Hodgkin’s Lymphoma

ContentsContentsContentsContents

Background ........................................................................................................... 11

Introduction to celiac disease.............................................................................. 11

Prevalence .......................................................................................................... 12

Pathogenesis ....................................................................................................... 13

Environmental factors..................................................................................... 14

Genetics in celiac disease................................................................................. 15

Immunological mechanisms ............................................................................ 15

Clinical presentation........................................................................................... 18

Diagnosis of celiac disease .................................................................................. 19

Biopsy............................................................................................................. 20

Serology.......................................................................................................... 23

Genetic testing ................................................................................................ 25

Screening ........................................................................................................ 25

Treatment........................................................................................................... 26

Achieving and monitoring compliance with treatment ........................................ 28

Protective effect of a gluten-free diet ................................................................... 29

Associated disorders ........................................................................................... 30

Diabetes Mellitus ............................................................................................ 31

Refractory sprue ............................................................................................. 31

Dermatitis herpetiformis ................................................................................. 31

Associated disorders studied in this thesis ........................................................... 33

Study I Liver disease........................................................................................ 33

Study II Autoimmune heart disease ................................................................. 33

Study III Addison’s disease.............................................................................. 33

Study IV Thyroid disease ................................................................................ 34

Study V Hematopoietic malignancy ................................................................ 34

Aims ...................................................................................................................... 37

Subjects and methods ............................................................................................. 39

Setting ................................................................................................................ 39

Data sources....................................................................................................... 39

Study design ....................................................................................................... 40

Studies I-IV ..................................................................................................... 40

Study V........................................................................................................... 40

Brief background of statistics used in this thesis.................................................. 41

Statistical analyses .............................................................................................. 42

Ethics ................................................................................................................. 42

Gender issues...................................................................................................... 43

Results ................................................................................................................... 45

Study I Liver disease ........................................................................................... 45

Study II Autoimmune heart disease..................................................................... 46

Study III Addison’s disease ................................................................................. 46

Study IV Thyroid disease .................................................................................... 47

Study V Hematopoietic malignancy.................................................................... 47

Discussion.............................................................................................................. 49

Methodological considerations ........................................................................... 49

Selection bias .................................................................................................. 49

Surveillance bias ............................................................................................. 50

Misclassification ............................................................................................. 50

Confounding factors ....................................................................................... 51

Findings and implications ................................................................................... 51

Study I Liver disease........................................................................................ 52

Study II Autoimmune heart disease ................................................................. 52

Study III Addison’s disease.............................................................................. 53

Study IV Thyroid disease ................................................................................ 54

Study V Hematopoietic malignancy ................................................................ 54

Implications for future research .......................................................................... 57

Conclusions ........................................................................................................... 59

Sammanfattning på svenska ................................................................................... 61

Acknowledgements ................................................................................................ 65

References.............................................................................................................. 67

Papers I-V .............................................................................................................. 85

11

BBBBackgroundackgroundackgroundackground

Introduction to celiac diseaseIntroduction to celiac diseaseIntroduction to celiac diseaseIntroduction to celiac disease

Celiac disease (CD), also known as gluten intolerance, gluten-sensitive enteropathy

and celiac sprue, is a prevalent autoimmune disorder that is triggered by the inges-

tion of wheat gluten and related proteins of rye and barley in genetically susceptible

individuals1. The classic presentation with malabsorption was first described in 1888

by Samuel Gee2, but it was not until the late 1940s that Dicke3, a Dutch pediatrician,

first recognized that the ingestion of wheat was responsible for manifestation of the

disease. More recent milestones in the history of CD are the identification of an as-

sociation between HLA DQ2/DQ8 and CD in 1989 by Sollid et al4 and the identifi-

cation of tissue transglutaminase as an autoantigen of CD5 that has lead to introduc-

tion of serological testing for IgA anti transglutaminase6.

In recent years, CD has been shown to be far more common than previously

thought and it is now considered to be one of the most common chronic disorders

affecting approximately 1% within the US and European populations7. Today the

picture of CD has shifted towards milder symptoms8, 9. The proportion of individuals

presenting with classic symptoms of diarrhea, abdominal distention, poor growth

and failure to thrive is decreasing when compared with those presenting with milder

symptoms such as abdominal discomfort, bloating, indigestion, or non-

gastrointestinal symptoms. This has lead to an upward shift of age at diagnosis and

CD is now diagnosed at any age9. In addition to the intestinal symptoms, CD is asso-

ciated with various extra-intestinal complications and may present with autoimmune

disease10, liver disease11 or other associated disorders12.

The diagnosis of CD requires typical histological changes (intraepithelial lympho-

cytosis, crypt hyperplasia, and villous atrophy) on intestinal biopsy while the patient

consumes a gluten-containing diet, together with symptomatic improvement on die-

tary gluten removal13-16. Serological testing has evolved into an invaluable tool when

identifying individuals that should undergo a biopsy and is responsible for the recog-

nition that CD is no longer considered a rare disease. However, positive CD serology

alone is not sufficient for the diagnosis of CD since such individuals may have a

normal mucosa15, 17. Treatment consists of a lifelong gluten-free diet (GFD), in which

food products containing wheat, rye or barley are avoided. Oat and purified wheat

starch is generally considered safe if it is not contaminated by other cereals during

the growing, transportation, and milling processes1, 18, 19. In patients who do not re-

12

spond to treatment it must be clarified if the diagnosis is correct and if a GFD is

really maintained.

The pathogenesis of CD is not fully understood. It involves interactions between

genetic, immunological and environmental factors1. The disease is unique because

the triggering factor, gluten, is known. The presence of HLA DQ2/DQ8 is necessary

for developing CD, although it is not enough20. The immunological reactions trig-

gered by gluten involve both the adaptive as well as the innate immune response and

key elements are the transglutaminase dependent activation of CD4+ T cells and the

activation of intraepithelial T cells21. The timing and amount of gluten introduction

as well as the number of infections at the time of gluten introduction has also been

suggested to influence the risk of CD22-24.

The purpose of this thesis was to assess the risk of CD in associated disorders,

predominantly of autoimmune origin, using Swedish population-based registers. The

objective was to gain more information on the consequences of having CD and to

identify high risk groups where screening may be considered.

PrevalencePrevalencePrevalencePrevalence

CD represents a wide spectrum of clinical features with many individuals showing

mild or no symptoms at all1. Because of the diversity of symptoms many affected

people remain undiagnosed. Therefore CD used to be considered as a rare disease.

Until the 1980s, the prevalence varied between 0.03% and 0.1% individuals25-28.

However, most early studies of CD prevalence were based on retrospective searches

of already identified cases in determined geographic areas. The increasing availability

of serological screening tests has made it possible to screen for CD in the general

population29. These screening studies have shown that CD now is one of the most

common chronic disorders in children30, 31 as well as in adults32-34 and the elderly35.

These studies have indicated that CD is an underdiagnosed condition.

In the United States the prevalence in a screening study of both children and adults

was estimated to be 0.75% (1:133)33. Similar estimates were found in Italy36 and in

the UK32. Tommasini et al36 reported a prevalence of 0.94% in a pediatric population

in Italy and West et al32 reported a prevalence of 1.2% in an adult population in the

UK. Apart from Europe and North America a high prevalence of CD has been re-

ported in India37, South America38 and Australia39. Although CD is believed to be

13

rare in Africa a study of Saharawi children40 in North Africa found the highest preva-

lence of endomysium positivity reported up to date of 5-6%.

The highest prevalence of biopsy proven CD has been found in Scandinavia. A re-

cent Swedish general population screening study found a CD prevalence of 3%31 in

12 year old school children. In Finland studies have shown a prevalence in children

of 1%30 and 2% in adults34 and the elderly35, indicating a high prevalence of CD in

all ages. The reason for the high prevalence of CD and why it has increased over the

last decades in the Scandinavian countries is not known34, 41. The rise in the preva-

lence of diagnosed CD could be explained by a greater awareness of the disease and

the availability of cheap and reliable serologic tests. However, this can not explain

the increase in screening studies. Lohi et al34 found an increased prevalence of

screened CD when comparing cohorts from 1978-80 and from 2000-2001. This in-

crease can not be explained by a changing detection rate. A high genetic risk of CD

in the Scandinavian countries is a possible but not very likely explanation since the

same genetic traits are also seen in other countries with a lower prevalence and could

not explain the increased numbers. Another explanation might be that the prevalence

of autoimmune diseases in general is increasing42. Environmental factors such as the

timing and amount of gluten exposure22, 23, perhaps in relation to breastfeeding43 or

the number of infections22, 44 could also contribute to disease occurrence.

PathogenesisPathogenesisPathogenesisPathogenesis

CD is a complex inflammatory disorder that can be best described as an autoimmune

disease with a multifactorial etiology. The disease is unique because the triggering

factor, gluten, is known. The expression of CD is strictly dependant on intake of glu-

ten and most patients go into complete remission when they are put on a gluten-free

diet, and they relapse when gluten is reintroduced in the diet. Since the disease is pre-

cipitated by the ingestion of gluten it could also be argued that the disease is a food

intolerance disorder. It does not involve IgE antibodies and mast cell degranulation

and should therefore not be considered as an allergy45. However, autoimmunity re-

mains the most probable cause since CD has several autoimmune features such as

highly disease-specific autoantibodies, strong association with HLA DQ2/DQ8,

autoreactive intraepithelial lymphocytes and that several autoimmune diseases are

commonly presented by individuals with CD20. The pathogenesis of disease involves

interactions between environmental, genetic, and immunologic factors. The interplay

14

between these factors makes it difficult to identify the underlying pathogenic mecha-

nisms.

Environmental factorsEnvironmental factorsEnvironmental factorsEnvironmental factors

CD development is induced by dietary gluten. Gluten is derived from wheat, barley

and rye1 and can be described as the protein fraction that remains after flour is

washed free from starch. The term “gluten” is usually used for any protein that is

active in CD. But strictly speaking it is the name only for the disease-activating pro-

teins in wheat. Gluten contains two major protein fractions, the gliadins and the glu-

tenins, both of which contain disease-activating proteins. There are closely related

proteins in barley and rye that also can activate the disease. Wheat, rye, and barley

have a common ancestral origin in the grass family. Oats which does not activate CD

is more distantly related to wheat, rye, and barley. Gluten has a high glutamine and

proline content, which is important in disease pathogenesis. Due to its high proline

content, which is resistant to intestinal proteases, gluten is only partially digested in

the upper gastrointestinal tract. This results in the presence of relatively large pep-

tides in the small intestine46. There is no known difference in digestion of gluten be-

tween healthy individuals and those susceptible to developing CD and what it is

about gluten that allows it to pass through the epithelial barrier of the intestine and

trigger the immune system is not known45, 46.

Apart from gluten, the interaction of environmental factors in CD is poorly under-

stood. Studies have suggested that there is a crucial time window when tolerance to

gluten occurs, and that there might be a reduced risk of CD if the child was still be-

ing breastfed at the time of gluten introduction. In Sweden the incidence of CD in-

creased when gluten was introduced in comparatively large amounts22 and if gluten

was introduced when children were still being breastfed reduced the risk43. To start

with gluten earlier than 3 months or later than 6 months of age was found to be a

risk factor in a study of a cohort of children in Denver, Colorado23. The number of

infections at the time of gluten introduction has also been suggested to increase the

risk of CD22, 44, but this association has not been confirmed in a recent study by We-

lander et al24.

15

Genetics in celiac diseaseGenetics in celiac diseaseGenetics in celiac diseaseGenetics in celiac disease

Genetics clearly play a key role in the pathogenesis of CD. About 10% of first-degree

relatives of individuals with CD are affected by the disease7, 33, 47-49 and Greco et al50

in a study of Italian twins, showed a concordance rate of 75% of CD in monozygotic

twins.

CD is a multigenetic disorder with its strongest association with human leukocyte

antigen (HLA). The association between CD and HLA-DQ2 was first discovered by

Sollid in 19894. The majority of CD patients carry HLA-DQ2 (DQA1*05/DB1*02)

and a minority carry HLA-DQ8 (DQA1*03/DB1*0302)20. In Western populations

about one third of all individuals express HLA-DQ2 or -DQ8. The majority of these

HLA-DQ2 or -DQ8 positive individuals in the general population will, however,

never develop CD even though they are exposed to gluten. Thus the role of HLA

DQ2/DQ8 genes as necessary, but not sufficient, factors for CD development sug-

gests that they are involved in the immune response to gluten leading to CD51 but

they do not explain the whole genetic component in CD. Other CD susceptibility

genes, such as T lymphocyte-associated antigen 4 (CTLA4), also contribute to the

genetic risk but the functions and importance of these genes need to be confirmed in

further studies. CTLA4 is a T cell surface molecule involved in control of T cell acti-

vation and has been reported to be associated with CD52-54 as well as other autoim-

mune diseases such as diabetes, hypothyroidism and Graves disease55, 56.

Immunological mechanismsImmunological mechanismsImmunological mechanismsImmunological mechanisms

Information on the mechanisms of the autoimmune response in CD is emerging, al-

though many aspects remain unclear. Both the adaptive (the immune response medi-

ated by B and T cells) as well as the innate (nonspecific response to antigens) im-

mune response are involved in CD development.

The adaptive immune response

Incompletely digested peptides of gluten cross the epithelium and enter the lamina

propria in DQ2 or DQ8 positive individuals under certain conditions. What causes

this to happen is not known. Increased intestinal permeability during gastrointestinal

infections or other stress factors have been suggested. Gluten peptides are then

deamidated by tissue transglutaminase and form a gluten/transglutaminase complex

that binds to HLA DQ2 and DQ8 on antigen presenting cells. The antigen presenting

16

cells then present these complexes to intestinal CD4+ T cells that recognize them as

foreign. The CD4+ T cells become activated and start to release inflammatory media-

tors. This process leads to inflammation, villous atrophy (VA), and crypt hyperplasia

in the small intestine20, 21, 46, 57, 58.

The innate immune response

This process involves an increased IL-15 expression and activation of intraepithelial

T cells that also cause destruction of epithelial cells and mucosal damage but these

mechanisms are not fully understood45. The lymphocytes are thought to interact with

the enterocyte. They are then stimulated by IL15 secreted by the enterocyte and in-

duce a cytotoxic inflammatory response. It is not clear if this response is restricted to

patients with CD or whether it also can occur in other disorders.

Some questions still remain

Why do not all individuals that express HLA molecules DQ2 and DQ8 develop the

disease?

Why is it only gluten, of the many food proteins, that elicits this type of harmful

immune response?

How do gluten peptides cross the epithelial barrier and enter the intestinal mucosa?

Does the autoantibody response to tissue transglutaminase have a pathogenic role of

its own?

17

Copyright © [2002] Massachusetts Medical Society. All rights reserved

From Farrell59

18

Clinical presentationClinical presentationClinical presentationClinical presentation

The spectrum of clinical presentation in CD is broad and varies according to age.

While some patients suffer from severe malabsorption, others have minimal symp-

toms, or are completely asymptomatic. The classic symptoms of CD in children are

diarrhea, abdominal distention, poor growth and failure to thrive. Other symptoms

might be vomiting, irritability, anorexia or constipation. In older children and ado-

lescents short stature, abdominal pain, anemia, constipation, pubertal delay and ano-

rexia are common symptoms60. In adults diarrhea, steatorrhea, weight loss, fatigue,

and anemia have been the classic signs of CD9.

Today the picture of CD has shifted towards milder symptoms8, 9. The proportion

of individuals presenting with dramatic symptoms of diarrhea, severe weight loss,

and malnutrition is decreasing when compared with those presenting with milder

symptoms. This has lead to an upward shift of age at diagnosis and CD is now diag-

nosed at any age. Many individuals with CD have no gastrointestinal symptoms but

present with extra-intestinal manifestations or associated conditions. This type of

presentation is called the atypical or silent form61 and include anemia, fatigue, osteo-

porosis, short stature, dermatitis herpetiformis, subfertility, neurologic problems or

presence of other autoimmune diseases. There is a high prevalence of anemia in indi-

viduals diagnosed with CD62, 63 and elevated liver enzyme levels are also found more

often64, 65. Weight loss as a symptom of CD is now less common and individuals can

also be overweight at the time of diagnosis66. Since atypical presentations are increas-

ingly found to predominate, CD is now considered to be a multisystem disorder

rather than a mainly gastrointestinal one.

Today CD in children and adolescents manifest itself mostly with minor, often

chronic, abdominal symptoms or extra-intestinal manifestations, such as short stat-

ure, delayed puberty, or anemia45, 67. Adult patients with CD are often completely

without symptoms or have had symptoms for a long time before they are diagnosed1.

The shift in clinical presentation could be explained by a change in the natural his-

tory of the disease. However, a more likely explanation would be that the ability to

make the diagnosis has improved with the presence of accurate serological markers

of the disease and the use of endoscopic biopsy techniques. This results in a broader

spectrum of individuals being investigated for CD and consequently being diag-

nosed9.

19

Diagnosis of Diagnosis of Diagnosis of Diagnosis of celiac diseaseceliac diseaseceliac diseaseceliac disease

Current diagnostic criteria for CD are based on revised guidelines proposed by the

European Society for Paediatric Gastroenterology and Nutrition (ESPGHAN)13,

which have been extrapolated to adults14-16. According to these guidelines, the diag-

nosis of CD is made on the basis of an abnormal duodenal biopsy. CD is present if

typical histological changes (intraepithelial lymphocytosis, crypt hyperplasia, and

VA) are found on intestinal biopsy while the patient consumes a gluten-containing

diet, together with symptomatic improvement on dietary gluten removal. The pres-

ence of positive serologic tests at diagnosis and normalization of the test when on a

GFD are supportive of the diagnosis but not necessary13. The original ESPGHAN

statement also required a demonstration of normalized histology following a gluten-

free diet. This is no longer needed for a definitive diagnosis of CD14, but is still often

performed in children. It is important to note that all diagnostic tests need to be per-

formed while the patient is on a gluten-containing diet.

In Sweden, a recent survey showed that all pediatric clinics perform a small bowel

biopsy at presentation in children with suspected CD, and the majority of clinics per-

form a second biopsy when the child is on a gluten-free diet68.

The diagnostic approach to detecting CD has undergone changes in recent years.

Serological testing has evolved into an important tool when identifying individuals

who need to undergo an intestinal biopsy examination to diagnose CD. The best

available tests are the IgA antihuman tissue transglutaminase (TTG) and IgA endo-

mysial antibody immunofluorescence (EMA)14. However, the positive predictive val-

ues of these tests are lower than reported when applied in general population screen-

ing17. Therefore, the diagnosis of CD should not solely be based on serologic find-

ings. Serology tests do not attain 100% sensitivity. Hence symptomatic individuals

with strong clinical evidence for CD should undergo an intestinal biopsy examina-

tion despite negative serological tests15.

Gluten challenge is not considered necessary for CD diagnosis any more, but

should be considered in children who are less than two years old, or if the original

diagnosis was not based on a small intestinal biopsy69. A gluten challenge is reintro-

ducing dietary gluten and taking a new biopsy after 3 to 6 months or earlier if symp-

toms arise13.

The diagnosis of CD is not always clear-cut. This is the case in those with minimal

histological findings or negative serologic test results with strong clinical evidence for

CD. Individuals with a positive serology test for CD, but with a normal small intes-

20

tinal biopsy, are defined by National Institute of Health as having latent CD14, 70,

while individuals with duodenal/jejunal inflammation including intraepithelial lym-

phocytosis but without VA are considered as borderline gluten-sensitive45, 71. Silent

CD, where the person with CD has no clinical signs are often present in risk groups

such as diabetes, first degree relatives60 undergoing screening for CD.

How to manage latent CD and borderline mucosal changes is difficult and there is

no consensus. If a GFD is begun it is difficult to know how long it should be contin-

ued and a gluten challenge needs to be performed. A second option is to continue on

a gluten containing diet and repeat the biopsy in 6-12 months. A third possibility is

to test for HLA-type and if negative CD can be excluded.

Copyright © [2008] Elsevier Limited. All rights reserved. From Briani72

Biopsy Biopsy Biopsy Biopsy

Small intestinal biopsy with VA is the gold standard for establishing the diagnosis of

CD13, 14, and has been so for the last 30-40 years. The indication for a biopsy is most

often a positive serology test. A biopsy might also be done in cases of negative serol-

21

ogy but high clinical suspicion. The biopsy is usually performed with grasp biopsy

forceps during endoscopy or using a suction capsule.

The classic celiac lesion is often located in the proximal part of the small intestine

and was described by Marsh73 as a progression through milder forms to complete

VA. Marsh classified the mucosa pathology into stages. Marsh grade 0 is normal

duodenal mucosa, grade 1 is the presence of a raised intraepithelial lymphocyte

count (>30 lymphocytes per 100 epithelial cells), and grade 2 is raised intraepithelial

lymphocytes and crypt hyperplasia. Marsh grade 1 and grade 2 lesions are consid-

ered to be early changes in patients who are likely to develop CD. Marsh grade 3 is

raised intraepithelial lymphocytes and crypt hyperplasia with progression of the in-

flammation to VA. Marsh grade 3 is subdivided into Marsh 3a - partial VA; 3b -

subtotal VA; and 3c - total VA. Marsh 4 (hypoplastic) describes a rare histological

finding of a flat atrophic mucosa thought to signify irreversible injury caused by

chronic inflammation.

In Sweden, all small intestinal biopsies are classified according to the SnoMed sys-

tem. For duodenal/jejunal biopsies, the SnoMed system is based on the same histopa-

thology parameters as the Marsh classification73.

Small intestinal histopathology classifications – a comparison

Classification Normal Inflammation Villous atrophy

Marsh

Classification

Type 0 Type 1 Type 2 Type 3a Type 3b Type 3c

Marsh

Description

Pre-

infiltrative

Infiltrative Infiltrative-

hyperplastic

Flat-destructive

SnoMed

Codes

M0010,

M0011

M40000, M41000, M42000,

M43000, M47000, M47170

M58,

D6218,

M58005

M58,

D6318,

M58006

M58,

D6218,

M58007

KVAST /

Alexander

classification

I

Normal

II

Intraepithelial

Lymphocytosis (IEL) #

III

Partial VA

IV

Subtotal

VA

IV

Total VA

Characteristics

Villous

atrophy

- - - + ++ ++

IEL # - + + + + +

Crypt

hyperplasia

- - + + ++ ++

22

Mucosal changes can be patchy. Therefore, it is important to take multiple endo-

scopic biopsy specimens (ideally 4–6 biopsy specimens) from the proximal small in-

testine, including a duodenal bulb biopsy, to ensure that VA is detected74. Biopsy

specimens should be of sufficient size, carefully oriented, and mounted villous side

up to enable cross sectioning rather than tangential sectioning, because the latter can

lead to misleading interpretations15. The risk of poorly oriented samples is minimized

when taking multiple biopsies.

Endoscopic signs of CD have also been described in patients with CD. Endoscopic

features of CD include the absence of folds, scalloped folds, visible mucosal blood

vessels and a mosaic pattern of the mucosa between the folds. These features may

only be seen in individuals with subtotal /total VA. Endoscopic evaluation without

biopsies is therefore inadequate to confirm or exclude a diagnosis since endoscopic

findings are not sufficiently sensitive for CD75.

There is good evidence that VA (Marsh type 3) is a characteristic histopathologic

feature of CD76, 77. The specificity of VA in small intestinal biopsies was assessed in

Sweden and found to be high (90% of all biopsies with VA were correctly classified

in a blinded assessment of Swedish pathologists)77. Ludvigsson et al77 also showed

that in a randomly selected subsample of 121 individuals with VA, VA biopsies are

consistent with clinical CD in 95% of the cases.

Small intestinal inflammation without VA is a heterogeneous condition. It does

not fulfill the criteria for CD and these individuals do not receive a gluten-free diet.

A number of potential causes of this condition have been suggested (listed below).

However, Ludvigsson et al77 found co-morbidity other than CD in only a minority of

these biopsies when they reviewed 1534 biopsy reports in a validation study of VA

and inflammation in Swedish biopsy registers. One explanation for this might be that

some of the other causes of small intestinal inflammation that was seen in other stud-

ies such as giardiasis78 and helicobacter pylori infection 79 are rare in Sweden.

Conditions other than CD, where an increased density of small intestinal intraepithelial

lymphocytes is possible:80, 81

Tropical sprue Microscopic colitis

Autoimmune diseases Bacterial overgrowth

Non-steroidal anti-inflammatory drugs Helicobacter pylori

cow’s milk protein intolerance postinfectious enteritis

Crohn’s colitis Gastric metaplasia

23

SerologySerologySerologySerology

Serological testing has evolved into an important tool when identifying individuals

that should undergo a biopsy. The serological testing has also made it obvious that

CD is a more common disease than previously thought. In clinical practice the avail-

able tests are endomysial antibodies (EMA), tissue transglutaminase antibodies

(tTGA) and antigliadin antibodies (AGA). EMA and TTGA are the most sensitive

and specific serologic tests for identifying individuals who need to undergo an intes-

tinal biopsy examination to diagnose CD29.

Antibodies against gliadin (AGA) have been used since the beginning of the

1980s82. They are measured by quantitative enzyme-linked immunosorbent assay

(ELISA). AGA has a sensitivity and specificity of about 80%, and the use of this test

is no longer recommended in adults. However, in children under the age of 18

months a combination of AGA and tTGA is feasible since a large proportion of

younger children with CD lack EMA and tTGA83.

The use of tests for endomysial antibodies (EMA) also came into practice in the

1980s84. The EMA test uses either monkey esophagus or human umbilical cord as

substrate. The EMA test has a high specificity estimated at 99% and sensitivity over

90%29. The test uses an immunofluorescence staining technique and needs manual

evaluation of the results21.

In 1997 Dieterich et al5 identified tissue transglutaminase as the antigen for EMA.

This has lead to introduction of serological testing for IgA anti transglutaminase6.

The first generation of ELISA that was used for measuring tTGA used guinea pig

liver as the substrate. Now most laboratories use human recombinant tTGA that has

further improved the diagnostic accuracy. The sensitivity and specificity of tTGA is

about 95%29. From an analytical perspective, tTGA has advantages over EMA: it is

quicker, quantitative, less expensive and does not require the use of primate tissue85.

IgA deficiency is more common in CD than in the general population86 and levels

of total serum IgA needs to be determined when using IgA based antibody tests. In-

dividuals with IgA deficiency will have negative serology tests based on IgA antibod-

ies regardless of CD status and in these individuals measurement of IgG tTGA/EMA

and IgG anti-gliadin antibodies can be used instead21.

The optimal serologic tests for the detection of CD and follow-up assessment re-

main controversial. Many centers recommend a two step approach. First a tTGA test

and, if positive, followed by EMA. Only those with double positive serology tests

proceed to small intestinal biopsy85. However, reviews of the literature does not indi-

24

cate a statistically significant difference between the tTGA and EMA tests15 and out

of these two tests human recombinant tTGA is often recommended for initial testing

for CD76.

CD serology has a high specificity29, 87 for CD and a high negative predictive value.

CD serology is therefore well suited to rule out the presence of CD. In contrast, the

positive predictive value of CD serology has been disappointing. Although the speci-

ficity for CD serology is high29, 87, the fact that the prevalence of CD is about 1% in

the general population7 has important implications for the positive predictive value

of antibody testing. The positive predictive value is the risk of having CD if you have

a positive serology test. This value is dependent on the prevalence of CD in the popu-

lation being tested. If the specificity of a test is 99%, one person out of one hundred

persons tested will have a false positive test. If the prevalence of the same disease is

1%, one person out of one hundred persons tested will have the disease. This means

that if you test 100 people, two tests will be positive. One out of these two positive

tests will be falsely positive and the positive predictive value of the test will in this

case be 50%. Hence the low positive predictive values in the general population. The

positive predictive value is very important for the clinician, since he/she needs to de-

termine whether or not the patient has the disease, given the results of a test. While

the negative predictive value of TTGA is 98-100%, the positive predictive value of

TTGA was only 28.6% in the study by Hopper et al17 (3.9% prevalence of CD), and

76% in the study by Hadithi et al88. In their paper, Hopper et al therefore suggest

that serological markers can support but not confirm a diagnosis of CD17. There is

also seronegative CD. This is more common in lesser degrees of atrophy due to the

fact that the sensitivity of the tests is related to the grade of histological damage in

CD15, 89, 90. With that conclusion, CD should be confirmed by small intestinal bi-

opsy13, 14.

There are now available rapid antibody tests with a sensitivity of around 80%87.

These tests can be performed on a drop of whole blood and allow a visual reading of

the result within a few minutes. However, there is a risk that these tests can lead to

self-diagnosis and initiation of a GFD without a prior intestinal biopsy.

Also, new antibody tests are evolving which might increase the diagnostic accu-

racy of CD in children91.

25

Genetic testingGenetic testingGenetic testingGenetic testing

The role of HLA DQ2 or DQ8 status in the assessment of CD lies in its high negative

predictive value (i.e., if individuals lack the relevant disease-associated alleles, CD is

virtually excluded). The vast majority of all CD patients carry the HLA-DQ2 and/or

HLA-DQ8 alleles and the presence of those HLA alleles provides a sensitivity of

close to 100% for CD15. However, these HLA types are not specific for CD, because

they have also been found in 20–30% of the general population. Hence, a positive

test will be of little value and only 3% of individuals carrying these alleles will de-

velop CD92.

Genotyping for HLA can be used to rule out CD in individuals that are already on

a GFD but have not performed a small intestinal biopsy. It can also be used when the

diagnosis remains uncertain or in the presence of IgA deficiency or in first-degree

relatives of a confirmed case of CD93, 94.

It is important to note that the low specificity of the HLA test means that a posi-

tive result is not sufficient to diagnose CD and that the test is not suitable for disease

screening. Even the presence of HLA-DQ2 or HLADQ8 in patients with positive se-

rologic test results is suggestive but not pathognomonic for CD88.

ScreeningScreeningScreeningScreening

CD fulfills many of the WHO screening criteria stated below95. It is a common dis-

order affecting about 1 % of the general population7 and up to two thirds of indi-

viduals with CD are asymptomatic96. There is also an available treatment for the dis-

ease.

WHO criteria for mass screenings97

(1) Early detection of the disease could be difficult on a clinical basis.

(2) The disease must be a common disorder causing significant morbidity in the general

population.

(3) The screening tests must be highly sensitive and specific for the target disease.

(4) A treatment for the disease must be available.

(5) If not recognized, the disease could result in severe complications difficult to manage.

In spite of the many points favoring mass screening of CD the benefits of screening

are debatable. The ultimate questions in considering population screening are

26

whether this approach is of more benefit or harm, and what the costs of the ap-

proach are. Some other doubts of the usefulness of screening are98-100:

That the natural history of CD is not fully known, especially in milder forms of the

disease.

The expected low compliance in screening detected CD.

The difficulty in determining the optimal time to screen.

Diagnostic criteria of CD13 are based on clinically suspected cases and requires an

improvement on a GFD for a definitive diagnosis.

The low positive predictive value of serological testing in the general population will

lead to a number of false positive results and endoscopic procedures in healthy indi-

viduals17.

And does intervention reduce the risk of future complications?

At present mass screening for CD is not recommended98, 99. An alternative to mass

screening could be screening high risk groups of CD and a case finding strategy.

There is support for the screening of high risk groups101, such as children with type 1

diabetes mellitus in whom the prevalence of CD is about 5-10%102, 103. Other high

risk groups include first degree relatives7, 33, 47-49, iron deficient anemia62, Down’s syn-

drome104 and IgA deficiency86. Case finding of symptomatic patients and high risk

groups is an effective strategy in detecting undiagnosed CD105.

There is also an ethical difference between screening and case finding. If a patient

seeks medical help, then the physician is trying to diagnose the underlying condition.

This would be classified as case finding and clearly it is the patient who has initiated

the consultation and in some sense is giving consent for investigation. However, in-

dividuals who have been identified through a screening programme may have con-

sidered themselves to be “well,” and it is the physician or healthcare system that is

defining them as potentially ill.

TreatmentTreatmentTreatmentTreatment

Treatment of CD consists of a life-long GFD. This involves the elimination of wheat,

rye and barley from the diet. Rice, corn, and potatoes are the most commonly used

substitutes for gluten-containing grains. Also, a number of other seeds, grains and

nuts offer alternatives to gluten. Improvement of symptoms is generally seen within

days to weeks after the initiation of gluten-free diet, while full mucosal recovery usu-

27

ally takes longer106. For children with symptomatic CD, rapid effects of a GFD are

usually seen with normalization of nutritional measures, improved growth in height

and weight and resolution of gastrointestinal symptoms.

What constitutes a GFD differs in different parts of the world. The inclusion of

oats and wheat starch in the diet is controversial. A number of studies have shown

that oats can be added to the diet without adverse effects in both adults18, 107 and

children108, 109. The addition of oats in the diet is also preferred by most patients107.

However, use of oats in the GFD is not widely recommended in many countries be-

cause of concerns of unacceptable high levels of cross contamination during the

growing, transportation, and milling processes1. In contrast to a naturally GFD

wheat starch-derived gluten-free products are included in the diet in many European

countries. Wheat starch-derived gluten-free products still contain, even after purifica-

tion, trace amounts of gluten. A randomized study comparing wheat starch-based

gluten-free to naturally gluten-free products showed that the morphological and

clinical responses were equally good by both treatments19.

There is also a debate on the safe threshold for gluten-contamination110, 111. A diet

entirely free of gluten contamination would of course be ideal, but since even natu-

rally gluten-free products may contain amounts of gluten, it is unrealistic111. Taking

into account that too strict limits might lead to the poor availability of gluten-free

products, which would hamper overall dietary compliance112 it is important to try to

ascertain a safe limit of gluten contamination in gluten-free products. At present

there are two different thresholds for gluten-free products in Sweden. Naturally glu-

ten-free products can contain up to 20 mg gluten /kg and wheat starch-derived glu-

ten-free products can contain up to 200 mg gluten/kg. New guidelines have been put

forward by the WHO Codex Alimentarius commission in July 2008113 stating that

the thresholds should be 20 mg gluten /kg for naturally gluten-free and 100 mg glu-

ten/kg for wheat starch-derived gluten-free products.

The diet and gluten-free products are often low in B vitamins, calcium, vitamin D,

iron, zinc, magnesium, and fiber. Few gluten-free products are enriched or fortified,

and vitamin deficiencies may occur. Screening for nutritional deficiencies and osteo-

porosis should be considered due to its high prevalence in CD62. Vitamin B deficiency

has been detected in CD patients who have been on a GFD for as long as 10 years114.

Therefore vitamin supplementation should be given whenever vitamin deficiency is

suspected. Because of the nutritional risks associated with CD, a dietitian is an im-

portant member of the health care team that monitors CD patients.

28

Future therapeutic options may include intake of enzymes that digest gluten115,

thereby increasing the safe threshold for gluten intake. Other potential treatments

could be therapies that interfere with the immune response to gluten and also the in-

volvement of environmental factors in the development of CD could indicate an op-

portunity for primary prevention116.

Achieving anAchieving anAchieving anAchieving and monitoring compliance with treatmentd monitoring compliance with treatmentd monitoring compliance with treatmentd monitoring compliance with treatment

Complete removal of gluten from the diet in a patient with CD should result in

symptomatic, serologic, and histological improvement. However, the time it takes

for the small intestinal mucosa to heal on a GFD varies and many patients show in-

complete histological recovery even 24 months after starting a GFD106, 117, 118. Com-

pliance with the gluten-free diet is also extremely challenging and adherence to the

treatment requires ongoing education and support from a multidisciplinary health

team. There are many barriers to compliance with the diet such as ambiguous labels

on prepared foods, which do not indicate whether the product contains wheat and if

gluten is present as an additive or contaminant119.

The highest rates of compliance are reported in patients who are diagnosed as

young children120, whereas adults and those diagnosed via mass serologic screening

are less compliant121, 122. Many patients with CD have minimal symptoms and there-

fore in this group of patients, symptoms cannot be used to assess response. Compli-

ance with the diet can be expected to be poor, especially in patients with minimal

symptoms.

Compliance to the GFD is compromised by a number of factors, including a lack

of education and continued support by a physician and dietitian118. Other factors as-

sociated with adherence to a GFD are the presence of other food intolerances, con-

cern over cost, ability to follow a GFD outside the home, and the ability to follow

the GFD irrespective of mood and stress123. The most important factor for maintain-

ing a GFD seems to be understanding of the gluten-free diet123.

Patients with CD should be evaluated at regular intervals by a health care team in-

cluding a physician and a dietician. During these visits assessment of symptoms,

growth, physical examination and adherence to a GFD should be monitored and

promoted. Clinicians also need to have a high awareness of common complications,

including osteoporosis and autoimmune disorders and to test for these conditions

when symptoms appear119.

29

Beyond this, there is little evidence on the most effective means of monitoring pa-

tients with CD76. The assessment of dietary compliance by interview has been put

forward as the best marker of CD control118, but this may not correlate with intesti-

nal damage. Antibody titers are often used in assessing compliance of a GFD and are

reported to revert to normal within one year of a GFD. It is important to note that

this is not completely reliable124 and a decline in antibody titers can be falsely secure

markers of dietary compliance. Many CD patients on a GFD with negative tTGA or

EMA still have manifest mucosal inflammation. This is because seroconversion of

tTGA during GFD in many cases occurs in the absence of mucosal improvement122,

124, 125. On the other hand, positive serological values seem to indicate ongoing in-

flammation125. With the reduced number of repeat biopsy, there is a risk of underes-

timating the number of individuals with chronic inflammation in CD and these indi-

viduals might be at an increased risk of long term complications.

Neither serologic markers nor dietary inquiries are to be regarded as reliable pre-

dictors of intestinal outcome after introduction of a GFD. Therefore, biopsy still re-

mains the best means of ascertaining mucosal recovery126. However, repeat biopsies

are unpleasant for the patients and are not routinely used anymore.

A more practical way to monitor compliance could include an assessment of die-

tary compliance by a registered dietician, in combination with measurement of serum

antibodies119 with the understanding of the shortcomings of this method.

Even if compliance to a GFD is maintained the treatment still has an impact on

quality of life. These include the difficulties of dining out, travel, and impact on fam-

ily life127. Hallert et al128 also showed that celiac women perceive the burden of the

disease worse than men do and that excessive tiredness tend to interfere with daily

activities.

What should be done if a patient does not respond to the gluten-free diet?

First a review of the diagnosis and to rule out other conditions should be under-

taken72. Next step would be a careful reassessment of the diet since lack of adherence

to strict gluten-free diet is the main reason of poorly controlled disease in adults118.

Protective effect of a glutenProtective effect of a glutenProtective effect of a glutenProtective effect of a gluten----free dietfree dietfree dietfree diet

Several reports have suggested that the prevalence of autoimmune disorders and ma-

lignancy is related to the duration of gluten exposure. Ventura et al10 showed that the

longer an individual had consumed a gluten-containing diet before the diagnosis of

30

CD, the higher the prevalence of autoimmune disorders later in life. This indicated a

protective effect of a GFD, although the direct effect of a GFD was not investigated.

A decreased risk of subsequent autoimmune disease in individuals following a GFD

has also been seen in a French study129. On the contrary, other studies130, 131 have

shown that gluten withdrawal in individuals with CD did not protect them from

autoimmune diseases, and that the overall risk of autoimmune diseases was in-

creased.

A number of studies have shown that the increased risk of malignancy and mortal-

ity was restricted to the first years after diagnosis132-134 and have suggested that this

would be due to a protective effect of a GFD. Other studies showed no decrease in

mortality on GFD135 and that the increased risk of lymphoma remained after follow-

ing a GFD for at least 5 years136. Silano137 found that the association between CD

and lymphoma was related to poor adherence to a GFD. Similar results were also

seen by Hervonen et al138and Olen et al139.

Dietary treatment has also been reported to prevent progression to hepatic failure

in individuals with CD and liver disease140.

In summary, strict adherence to the GFD seems to be protective of the develop-

ment of associated disorders but the direct evidence of a protective effect of a GFD is

sparse.

Associated disordersAssociated disordersAssociated disordersAssociated disorders

There is a positive association between pediatric as well as adult CD and autoim-

mune diseases. Individuals with CD are also at an increased risk of a number of

other diseases, such as osteoporosis, lymphoma, liver disease and subfertility. Some

of these associations such as anemia and osteoporosis are caused by nutritional defi-

ciencies. In other associated disorders the mechanisms are more complex and not

fully understood. Shared genetic traits of CD and associated disorders are one poten-

tial explanation. Untreated CD leading to the onset of other diseases is another ex-

planation. This might be due to an ongoing intestinal inflammation and increased

permeability of pro-inflammatory substances and undefined triggers of autoimmu-

nity that subsequently lead to complications of CD. This latter explanation implies a

possibility to prevent and perhaps reverse complications of CD with a GFD.

Some of the most common and clinically important associated disorders in CD are

discussed below.

31

Diabetes MellitusDiabetes MellitusDiabetes MellitusDiabetes Mellitus

Both children and adults with CD are at an increased risk of type 1 diabetes mellitus

(T1DM)141. The prevalence of CD in T1DM is 4-9%. This association may partly be

due to shared genetic factors associated with both T1DM and CD103. CD can occur

before T1DM142 but in the majority of patients, CD is diagnosed after the clinical on-

set of T1DM143. Many children with T1DM and CD have silent CD, and the disease

may in those cases only be detected through serologic screening144. Routine screening

for CD in children with T1DM is already in use in many centers, but the appropriate

interval between reexaminations is unclear. A recent study by Larsson et al144 sug-

gests screening at the onset of T1DM and annually for a minimum of at least 2

years.

Refractory sprueRefractory sprueRefractory sprueRefractory sprue

Refractory sprue is defined as persistent symptoms and VA despite adherence to a

strict gluten-free diet and not accounted for by other causes than CD. Persistent VA

in adult CD, even in the absence of symptoms, has been shown to carry an increased

risk of a number of severe complications145. These patients usually have classic symp-

toms including diarrhea, weight loss, malabsorption, abdominal pain and anemia146.

Two subtypes of refractory sprue have been suggested. Type 1, in which there is a

normal intraepithelial lymphocyte phenotype, or type 2, in which there is a clonal

expansion of an aberrant intraepithelial lymphocyte population. This latter form car-

ries a highly increased risk of enteropathy-associated T-cell lymphoma and has been

described as a cryptic T-cell lymphoma1.

Dermatitis herpetiformisDermatitis herpetiformisDermatitis herpetiformisDermatitis herpetiformis

Dermatitis herpetiformis is regarded as a skin variant of CD. It is a blistering skin

disease usually located on elbows, knees, buttocks and scalp. The diagnosis is based

on skin biopsy and most patients also have small intestinal mucosal changes. Treat-

ment consists of a GFD initially in combination with Dapsone since it takes several

weeks to months for the rash to respond to the diet147.

32

Disorders associated with celiac diseaseDisorders associated with celiac diseaseDisorders associated with celiac diseaseDisorders associated with celiac disease

Autoimmune disorders Reference Malignancy Reference

Diabetes 102, 103,

141, 143, 144 Lymphoma 132, 133, 135,

136, 148-157

Addison’s disease 172-175, 237 Liver 148

Thyroid disease 10, 26, 130,

131, 135,

158-171, 238

Esophagus 132, 148

Juvenile idiopathic

arthritis

176 Papillary thyroid

cancer

177

Liver disease Reference Neurological

disorders

Reference

Primary biliary

cirrhosis

11, 178-180 Ataxia 72, 181

Primary sclerosing

cholangitis

11, 180-182 Seizures 181

Autoimmune

hepatitis

11, 183-184 Depression 181, 185

Liver failure 11, 140 Cerebral palsy 186

Hematological Reference Other Reference

Anemia 32, 62, 63 Dermatitis

herpetiformis

147

IgA deficiency 86, 187 Osteoporosis /

risk of fractures

32, 62, 188-190

Subfertility 191-193

Infections Reference Down’s syndrome 194

Tuberculosis 195 Dental enamel defects 196

Sepsis 197

33

Associated disorders studied in this thesisAssociated disorders studied in this thesisAssociated disorders studied in this thesisAssociated disorders studied in this thesis

Study IStudy IStudy IStudy I Liver d Liver d Liver d Liver diseaseiseaseiseaseisease

Previous studies have suggested an association between CD and liver disease140, 198, 199,

including hepatitis183, 184and hypertransaminasemia64, 65, 200. Autoimmunity associated

with CD might be one of the mechanisms by which CD is associated with such dis-

ease. Despite this, data on CD and autoimmune liver disease, other than primary bil-

iary cirrhosis (PBC)178-180, for which there is an established positive association, are

sparse. Although several studies have suggested a positive relationship between CD

and primary sclerosing cholangitis (PSC)180, 182, 201-203, these studies are limited by

small numbers of patients, and surveillance bias has not been taken into account. CD

might also be associated with more severe liver disease, including end-stage liver dis-

ease140, 203. In a Finnish study, 8 of 185 (4.3%) patients who had undergone liver

transplantation had CD140. Swedish data suggest that CD is associated with a 2-fold

to 3-fold risk increase for primary liver cancer and a 7-fold to 8-fold risk increase for

death from liver cirrhosis148, 157. There is little information on other forms of severe

liver disease in the general CD population.

Study II Study II Study II Study II Autoimmune heart diseaseAutoimmune heart diseaseAutoimmune heart diseaseAutoimmune heart disease

In 2002 Frustaci et al204 found an increased prevalence of CD among 187 patients

with myocarditis. CD may also be positively associated with cardiomyopathy (any

cardiomyopathy or idiopathic dilated cardiomyopathy)205-210. All these studies were,

however, based on few positive events, and with the exception of the prospective

Danish study205, they were cross-sectional and did not take into account whether CD

preceded or ensued heart disease. To our knowledge, research on CD and pericardi-

tis is limited to case-reports211-213. Individuals with CD are also at increased risk of

death from ischemic heart disease157, 214.

Study III Addison’s diseaseStudy III Addison’s diseaseStudy III Addison’s diseaseStudy III Addison’s disease

Several studies have indicated an increased risk of CD in patients with Addison’s dis-

ease (AD). O’Leary et al172 screened 41 patients with AD, and found an increased

prevalence of CD (5/41 = 12.2%). Similar findings have since been reported by Ital-

ian174, 175 and Norwegian researchers173. There is also an increased risk of CD in indi-

34

viduals with more than one autoimmune endocrinological disorder162. All earlier

studies on CD and AD were, however, limited in size, based on screening for CD in

individuals with endocrinological disease162, 172-175, and none of the authors estimated

risk of AD in CD162, 172-175. The rarity of AD makes such analyses difficult without

access to large datasets.

Study IVStudy IVStudy IVStudy IV Thyroid disease Thyroid disease Thyroid disease Thyroid disease

Graves’ disease and different forms of thyroiditis are often referred to as autoim-

mune thyroid disease (TD) and have previously been suggested to be associated with

CD10, 26, 130, 131, 135, 158-171. The majority of earlier studies have however been cross-

sectional and few of them have included children131, 163, 166, 169. Several studies have

shown a positive association between CD and an increased prevalence of autoim-

mune thyroid antibodies160, 169 but not clinically overt disease. Both Italian168, and

Finnish167 data suggest an increased prevalence of hypothyroidism in individuals with

CD. Sategna-Guidetti et al130 have also presented an increased risk of overall TD in

individuals with CD. Most other positive studies have either lacked controls26, 159, 170

or not attained statistical significance10, 131, 135, 163. To our knowledge no studies have

until now reported a statistically significant risk increase of hyperthyroidism in CD.

We only know of one longitudinal study examining the risk of future TD in indi-

viduals with CD131. This study131 indicated an increased incidence of TD in CD but

results did not reach statistical significance.

The prevalence of CD is likely increased among individuals with TD158, 160, 161, 164,

165. However, although Spadaccino et al171 recently reported an increased prevalence

of CD in Italian patients with TD, that study did not use a matched control group.

Study VStudy VStudy VStudy V Hematopoietic malignancy Hematopoietic malignancy Hematopoietic malignancy Hematopoietic malignancy

The association between CD and lymphoma has been the focus of a number of stud-

ies132, 133, 135, 136, 148-157. The most common type of lymphoma to be associated with CD

is non-Hodgkin’s lymphoma (NHL) with risk estimates from previous studies vary-

ing from no increased risk135 to a 42-fold increased risk of NHL132 and 69-fold rela-

tive risk of death from of NHL133. More recent data suggest a 3-6-fold increased risk

of lymphoproliferative disease136, 148-151, 153-155, 157, and that not only T-cell lymphoma

but also B-cell Non-Hodgkin lymphoma is associated with CD152. A recent review215

35

did not observe an association between CD and NHL overall, but the risk was ele-

vated among patients diagnosed at age >33 years and in specific T-cell NHL sub-

types. In a large cohort study of individuals with NHL, Gao et al156 found a 5-fold

increased risk of NHL but no increased risk of HL in individuals with a hospital dis-

charge of CD. The risk estimates of NHL in that study declined over time (first cal-

endar period (1965-74): OR=13.2; last calendar period (1995-2004): OR=3.84).

Many previous reports on CD and lymphoma have been limited by potential selec-

tion bias as restriction to referral centers132, 136, 150, 151, 154 or inpatients with CD148, 152,

157. With the exception of the studies by Askling et al148 and Gao et al156, most studies

have based their risk estimates of later lymphoma in CD on less than 10 individuals

with both CD and lymphoma132, 136, 150, 153, 155, 216. Previous reports have shown no as-

sociation between leukemia and CD148, 156. Although, there is an increased risk of

leukemia in another inflammatory disorder, rheumatoid arthritis217.

Little is known of the risk of malignancy in inflammation without VA and in la-

tent CD. Four NHL cases with positive antiendomysial antibodies, but no histologi-

cal lesion were found in a study by Mearin et al154. However, that study did not pre-

sent relative risk estimates for lymphoma in latent CD. There have also been reports

of no association between malignancy and positive CD serology where biopsy has

not been performed216, 218.

37

AAAAimsimsimsims

The aim of this thesis was to examine the association between CD and associated

disorders through Swedish population-based registers. CD is one of the most com-

mon chronic disorders among Swedish children. Sweden is also well suited for epi-

demiological research because of the existence of high quality national registers

based on the unique personal identification numbers assigned to all Swedish citizens.

This provided us with a sufficient data to investigate these associations. With this

thesis we aim to gain more information on the consequences of having CD and to

identify high risk groups where screening may be considered. In the long run we

hope that our research will decrease the risk of undetected complications in CD and

to improve the quality of life in individuals with CD.

Specific aims were to examine

If individuals with CD are at an increased risk of liver disease.

If there is an association between autoimmune heart disease and CD.

If there is an increased risk of Addison’s disease in CD.

If there is an association between CD and hypothyroidism, hyperthyroidism and thy-

roiditis.

If individuals with CD are at an increased risk of hematopoietic malignancies such as

lymphoma or leukemia.

39

SSSSubjects and methodsubjects and methodsubjects and methodsubjects and methods

SettingSettingSettingSetting

The studies in this thesis were all epidemiological studies carried out using Swedish

health registers. Sweden offers exceptional opportunities for epidemiological re-

search219. The reason for this is found in the structure of its health system, the exis-

tence of nationwide registers and the systematic use of unique personal registration

numbers.

Data sourcesData sourcesData sourcesData sources

The Swedish national identification number - Linkage was made possible through

the use of the unique personal identification number. This number is a ten-digit

number assigned to more than 99.9% of all Swedish residents at birth220. The first

six digits are the date of birth (year - month – day), followed by three digits that

identify the individual and the last digit is a control digit. The identification number

is used in all health care and population registers220.

The total population register – includes information on basic demographic statistics

such as personal identification number, area of residence, sex, age civil status, and

dates of emigration in all Swedish residents. The register was computerized in 1968,

and data are collected and updated by the local tax offices221.

The hospital discharge register – The hospital discharge register has been used to

identify individuals with CD as well as the diagnosis of interest (outcome measure) in

studies I-IV. This register contains individual-based information on inpatient care

countywide since 1964. The register has covered 85% of Sweden since 1978, and all

of Sweden since 1987. For every hospital discharge, information on diagnoses and

surgical procedures are recorded according to the International Classification of Dis-

eases (ICD)222. The reliability of diagnoses from the register is high223.

The Swedish cancer registry – Malignancy data from the cancer registry was used in

study V. Since 1958 all physicians and pathologists/cytologists in Sweden are obliged

by law to report each incident case of cancer that they diagnose and/or treat to the

centralized nationwide Swedish Cancer registry. 99% of all malignancies are re-

ported to the cancer registry. Information includes personal identification number,

40

type and location of cancer, date of diagnosis, whether the cancer is the first cancer

reported for each individual and if the cancer was first discovered at autopsy224.

Study designStudy designStudy designStudy design

StudiesStudiesStudiesStudies I I I I----IVIVIVIV

These were all historical cohort studies with prospective registration of data. The

prospective data collection is important since this means that the information on CD

was recorded regardless of disease status. In the cohort studies individuals with a di-

agnosis of CD (n = 15,000) were identified using the hospital discharge register. The

government agency for demographic statistics, Statistics Sweden, then used the total

population register to match each individual with CD with five reference individuals

at time of CD diagnosis. Reference individuals were matched for age, sex, calendar

year, and county of residence. All study participants were then linked to the hospital

discharge register. Follow-up time started one year after study entry (equal to the

date of first inpatient diagnosis of CD or corresponding date in the matched refer-

ence population) and ended on date of first discharge diagnosis of the disease of in-

terest, emigration, death, or end of the study period (December 31, 2003), whichever

occurred first.

We also performed a case-control study from the original cohort of individuals

with CD. We used conditional logistic regression to estimate the odds ratio for CD in

individuals with a prior diagnosis of the disease of interest. We performed this analy-

sis to evaluate whether the positive association between CD and the disease of inter-

est was restricted to patients with CD preceding the disease of interest. The end of

follow-up was defined as date of first CD diagnosis and the same date in the

matched reference population without CD. Those with 1 yr or less between the date

of the disease of interest and the CD diagnosis, and corresponding date in reference

individuals, were excluded.

StStStStudy Vudy Vudy Vudy V

This cohort study examined the risk of hematopoietic malignancy, including lym-

phoma in three cohorts of CD (n = 28,810); small intestinal inflammation (n =

12,681); and latent CD (n = 3,552). We did so through linkage between small intes-

tinal biopsy data from all pathology departments in Sweden and malignancy data

from the Swedish national cancer registry224. We also collected serology data from 8

41

university hospital biochemistry laboratories on individuals who had undergone

small intestinal biopsy at any hospital in the catchment areas of these biochemistry

laboratories. All study participants were matched with five reference individuals

from the total population register.

Follow-up time started at first biopsy of CD, small intestinal inflammation or

normal small intestinal biopsy with positive CD serology (or corresponding date in

the matched reference population). Follow-up time ended on date of first lymphoma

diagnosis, emigration, death or end of the study period (31st December 2006),

whichever occurred first. Cox regression was used to calculate hazard ratios for he-

matopoietic malignancy in CD.

A case-control study similar to that described in studies I-IV was also performed.

Brief background of statistics used in this thesisBrief background of statistics used in this thesisBrief background of statistics used in this thesisBrief background of statistics used in this thesis

Hazard ratio (HR) – Hazard ratio is the central output of Cox regression and can be

described as the ratio between the predicted hazard for a member of one group and

that for a member of the other group, holding everything else constant. The main

difference between hazard ratio and relative risk is that the Cox regression takes 1)

time to event and 2) exclusion of study participants into account225.

Odds ratio (OR) – The odds ratio is the ratio between the odds of two groups and

can be used to estimate the relative risk in a case-control study225.

Person-years – the sum of all years contributed in a study by subjects at risk of a dis-

ease225.

Incidence rate – number of new cases, often per 100,000 person-years225.

Kaplan-Meier curve – is a way of graphically presenting survival over time, and to

assess the cumulative incidence (incidence proportion) of an event. The probability

of surviving to any point in time is estimated from the cumulative probability of sur-

viving each of the time intervals that preceded it. Each step in a Kaplan-Meier curve

represents one or more “deaths”226.

Matching – Matching is a method used to reduce the risk of confounding and means

that controls are chosen to match cases with respect to gender, age or any factor that

is likely related to the disease under investigation but not intended to be analyzed in

the particular study227.

Post-hoc analysis – refers to analyses made after the original hypothesis (a priori) has

been tested.

42

Cox regression – is a multivariate technique which implements the proportional haz-

ards model. Cox regression is designed for analyses of time until an event or time be-

tween events. The classic example would be time from diagnosis with a terminal ill-

ness until the event of death (hence survival analysis). Cox regression provides a

convenient way of controlling for time in the analysis of follow-up studies225.

Conditional logistic regression – is a variant of logistic regression that is used when

the cases and controls are matched. The original parameters are odds parameters.

The most important application of a logistic regression is a case-control study. The

central statistical output is the odds ratio225.

Statistical analysesStatistical analysesStatistical analysesStatistical analyses

The information on associated disorders of CD was assessed through cohort studies

and case control studies. We used a matched reference population. Individuals were

matched for sex, age, calendar year and area of residence. This enabled internal

stratification which means that analyses were conditioned on risk-set so that an indi-

vidual with CD was compared only with his or her matched reference individuals.

All statistical tests were two sided and hazard ratios (HR) and odds ratios (OR)

with a 95% confidence interval (CI) excluding 1.0 and p-values <0.05 were consid-

ered statistically significant. Statistics were calculated using SPSS 11.0 or SPSS 15.0

(studies I-IV) and SAS (study V). Results were given as HR, OR and incidence rates.

EthicsEthicsEthicsEthics

Studies I-IV: This project (04-030/1) was approved by the Research Ethics Commit-

tee of the Karolinska Institutet, Stockholm, on March 18th, 2004. None of the par-

ticipants was contacted. Patient information was made anonymous prior to the

analyses and individual patient charts could not be traced.

Study V: This study was approved by the Regional Ethical Committee of the Karo-

linska Institutet, Stockholm, on June 14th, 2006 (2006/633-31/4), with additional

amendments (2007/747-32 and 2008/257-32). This study is based on registry linkage

and none of the participants was contacted. Subject information was made anony-

mous before the analyses.

43

Gender issuesGender issuesGender issuesGender issues

The majority (about 60%) of individuals in the studies in this thesis are female. The

female predominance in CD is similar to that in many other autoimmune disorders20.

We present both combined results for males and females as well as separate results to

see if the disease burden is different according to sex. We found no statistically sig-

nificant difference in risk estimates according to sex in the studies in this thesis.

45

RRRResultsesultsesultsesults

Study IStudy IStudy IStudy I Liver disease Liver disease Liver disease Liver disease

CD was statistically significantly associated with acute and chronic hepatitis, PSC,

fatty liver, liver failure, liver cirrhosis/fibrosis, and PBC but not with ascites, hepa-

tomegaly, and liver transplantation. There was no statistically significant difference

in risk estimates for liver disease according to sex or age at diagnosis of CD.

This study found a four fold increased risk of PSC in individuals with CD. When

restricting the diagnosis of PSC to those with simultaneous inflammatory bowel dis-

ease, there remained a statistically significant association between CD and PSC (HR

= 5.65; 95% CI, 1.72–18.53). The risk estimate increased when we included the first

year of follow-up (HR = 8.46; 95% CI, 2.83–25.26). When we excluded individuals

with a diagnosis of inflammatory bowel disease at any time before end of follow-up,

CD was still positively associated with PSC (HR = 4.12; 95% CI, 2.12– 8.00). This

study also found an increased risk of liver failure. We found no increased risk of liver

transplantation in individuals with CD. Only one individual with CD (and no liver

disease before study entry) had a liver transplantation during the follow-up time.

Post hoc analyses found positive associations of CD with all liver outcomes except

fatty liver and LT when compared to inpatient reference individuals. In these analy-

ses we included the first year follow-up. This was deemed reasonable since reference

individuals had been admitted to hospital and were therefore at the same risk of be-

ing investigated for liver disease due to hospital admission as individuals with CD.

Outcome Positive events

CD / ref pop

Hazard ratio Confidence

interval (95 %)

P – value

PSC 26 / 32 4.46 2.50 – 7.98 <0.001

PBC 7 / 3 10.16 2.61 – 39.49 0.001

Acute hepatitis 8 / 9 5.21 1.88 – 14.40 0.001

Chronic

hepatitis

18 / 16 5.84 2.89 – 11.79 <0.001

Liver failure 45 / 71 3.30 2.22 – 4.88 <0.001

Liver

transplantation

1 / 4 1.07 0.12 - 9.62 0.954

Ref pop = reference population; CD = celiac disease; PSC = primary sclerosing cholangitis;

PBC = primary biliary cirrhosis

46

Study II Autoimmune heart diseaseStudy II Autoimmune heart diseaseStudy II Autoimmune heart diseaseStudy II Autoimmune heart disease

There was a statistically significant association between CD and any cardiomyopathy

but not with pericarditis or myocarditis. Since 1987, the hospital discharge register

contains ICD-codes for dilated cardiomyopathy. Restricting our outcome measure to

heart diagnoses in 1987–2003 (i.e. diagnosed with an ICD-9 or ICD-10 code) re-

corded in a department of cardiology or internal medicine, we found no statistically

significant associations between adulthood CD, myocarditis, dilated cardiomyopathy

and pericarditis. In the pediatric group, the incidence for heart disease was generally

lower in individuals with CD than in reference individuals, although this association

was not statistically significant.

First year excluded* First year included* Outcome

All ICD 9-10 All

Cardiomyopathy 2.2; 1.4-3.4 1.7; 0.4-6.5 2.1; 1.4-3.2

Pericarditis 1.6; 1.0-2.7 1.5; 0.5-4.0 1.7; 1.1-2.6

Myocarditis 0.9; 0.4-2.1 2.1; 0.4-11.7 0.9; 0.4-2.0

Hazard ratio; 95 % Confidence interval

ICD = International classification of disease

Study III Addison’s diseaseStudy III Addison’s diseaseStudy III Addison’s diseaseStudy III Addison’s disease

There was a statistically significantly positive association between CD and subse-

quent AD (HR = 11.4; 95% CI = 4.4-29.6). The HR for AD was 24.1 in individuals

entering the study before the age of 16 years, and 8.9 in individuals entering the

study as adults. However, this difference was not statistically significant (formal in-

teraction test p=0.274). When we included the first year of follow-up, the HR for

subsequent AD increased to 19.4.

Restricting our outcome measure to AD listed as the main diagnosis (HR = 9.2) or

AD diagnosed in departments where AD is usually diagnosed (HR = 9.1) only af-

fected our risk estimates marginally. When we limited reference individuals to inpa-

tients, the risk estimate decreased but nevertheless remained statistically significantly

increased (adjusted HR = 4.6; 95% CI = 1.9-11.4; p = 0.001; based on 25 and 6

positive events in 15,425 individuals with CD and 14,370 inpatient reference indi-

viduals respectively.

47

Study IVStudy IVStudy IVStudy IV Thyroid disease Thyroid disease Thyroid disease Thyroid disease

CD was associated with hypothyroidism (HR = 4.4; 95% CI = 3.4-5.6; p < 0.001),

thyroiditis (HR = 3.6; 95% CI = 1.9-6.7; p < 0.001) and hyperthyroidism (HR = 2.9;

95% CI = 2.0-4.2; p < 0.001). The highest risk estimates were found in children (hy-

pothyroidism HR = 6.0; 95% CI = 3.4-10.6, thyroiditis HR = 4.7; 95% CI = 2.1-

10.5 and hyperthyroidism HR = 4.8; 95% CI = 2.5-9.4). Individuals with CD were

also at increased risk of later hyperthyroidism requiring surgery of the thyroid gland

(HR = 2.4; 95% CI = 1.4-4.3; p = 0.002; based on 19 individuals with CD and 37

reference individuals).

Conditional logistic regression showed an increased risk of CD in individuals with

prior hypothyroidism (Odds Ratio = 3.8; 95% CI = 2.8-5.2; p < 0.001), thyroiditis

(OR = 4.0; 95% CI = 2.2-7.2; p < 0.001) and hyperthyroidism (OR = 2.0; 95% CI =

1.5-2.8; p < 0.001).

Children Adults Outcome

Positive events

(CD / ref pop)

HR; 95 % CI Positive events

(CD / ref pop)

HR; 95 % CI

Hypothyroidism 27 / 22 6.0; 3.4 – 10.6 100 / 169 4.1; 3.1 – 5.4

Thyroiditis 12 / 12 4.7; 2.1 - 10.5 5 / 12 2.2; 0.8 – 6.5

Hyperthyroidism 17 / 17 4.8; 2.5 – 9.4 31 / 83 2.4; 1.5 – 3.7

CD = celiac disease; Ref pop = reference population; HR = hazard ratio; CI = confidence in-

terval.

Study V Hematopoietic malignancyStudy V Hematopoietic malignancyStudy V Hematopoietic malignancyStudy V Hematopoietic malignancy

CD and subsequent risk of hematopoietic malignancy

There was an increased risk of malignant lymphomas (of all types) in CD (HR=3.18;

95% CI=2.63-3.83), based on 175 observed positive events. The relative risks were

similar in men and women and did not change over calendar period. There was no

increased risk of malignant lymphoma or leukemia in children with CD (HR=1.08;

95% CI=0.41-2.82 and HR=1.08; 95% CI=0.41-2.82 respectively). We also found

an increased risk for developing non-Hodgkin’s lymphoma (NHL) (HR=4.81; 95%

CI=3.81-6.07; based on 132 NHL) and Hodgkin’s lymphoma (HL) (HR=4.39; 95%

CI=2.59-7.45; based on 25 HL). This study found no association between leukemia

and CD.

48

Small intestinal inflammation and subsequent risk of hematopoietic malignancy

There was an elevated risk of malignant lymphomas (HR=1.66; 95% CI=1.28-2.17),

based on 72 positive events among individuals with biopsy-verified inflammation.

Although risk estimates for NHL (HR=1.65; 95% CI=1.15-2.38) and HL (HR=1.48;

95% CI=0.60-3.62) were similar, only the association between inflammation and

NHL attained statistical significance. This study found no association between leu-

kemia and small intestinal inflammation.

Latent CD and subsequent risk of hematopoietic malignancy

There was no increased risk of lymphoma (of all types) in latent CD (HR=1.04; 95%

CI=0.44-2.43). This study also found no association between leukemia and latent

CD.

Celiac Disease Small intestinal

inflammation

Latent Celiac

Disease

Outcome

Hazard Ratio; 95% Confidence Interval

Any lymphoma 3.18; 2.63-3.83 1.66; 1.28-2.17 1.04; 0.44-2.43

NHL* 4.81; 3.81-6.07 1.65; 1.15-2.38 1.79; 0.74-4.34

HL# 4.39; 2.59-7.45 1.48; 0.60-3.62 1.08; 0.13-9.00

Leukemia 1.35; 0.86-2.14 1.64; 0.93-2.91 1.58; 0.45-5.52

* NHL = Non-Hodgkin’s Lymphoma

# HL = Hodgkin’s Lymphoma

49

DDDDiscussioniscussioniscussioniscussion

Methodological considerationsMethodological considerationsMethodological considerationsMethodological considerations

Selection biasSelection biasSelection biasSelection bias

Selection bias is when groups that are being studied are not completely comparable,

most often due to inappropriate selection of controls226, 227. Not all individuals with

CD are identified through the hospital discharge register since this register only con-

tains information on those admitted to hospital. Using data restricted to inpatients

with CD would be a major drawback if all individuals were identified today, how-

ever in our study most individuals were identified in the 80’s and early 90’s when

admittance to hospital during diagnostic procedures were more common27. They ma-

jority was also young children at the time of diagnosis68. The median age at study

entry was below two years of age for most of the study period. Children often need

general anesthesia during diagnostic procedures and are therefore admitted to hospi-

tal. Therefore, we believe that our cohort is somewhat representative of most indi-

viduals with CD during much of the study period.

We used two different sets of controls in these studies. Both of which has its ad-

vantages and drawbacks. First we used controls selected from the general popula-

tion. This enabled a close matching process but the diseases under study (outcome

measures) are more likely to be diagnosed among inpatients compared to popula-

tion-based controls and this might lead to somewhat inflated risk estimates due to

selection and surveillance bias. We also used inpatient controls. Then we do not have

the same problem with surveillance bias since both CD patients and controls have

been admitted to hospital and have an equal chance of being investigated for other

diseases due to hospital admission. It is important to note that inpatient controls

may have been admitted to hospital for diseases that are more severe than CD, for

example heart disease, cancer or diabetes. This would mean that they have a higher

risk of receiving any other diagnosis compared to CD. Using inpatients as controls

might lead to an underestimation of the true risk estimates.

Our belief is that the true estimates of associated disorders in CD are somewhere

between the risk estimates using general population controls and the risk estimates

using inpatients controls.

Our CD cohort contains both individuals identified through diagnostic or thera-

peutic procedures, as well as those admitted to hospital for other medical conditions.

50

This might influence the generalizability of our results and the results may not be

applicable to individuals with asymptomatic CD diagnosed through screening.

Surveillance biasSurveillance biasSurveillance biasSurveillance bias

There is also a risk that individuals with an inpatient diagnosis of CD stand a better

chance of having other diseases detected due to surveillance. To reduce the risk of

surveillance bias we excluded the first year after study entry in studies I-IV. This will

rule out any surveillance bias during the first year. There is however also a downside

in excluding the first year after study entry. A protective effect of a GFD could give

misleading results if the first year is excluded when there is still ongoing inflamma-

tion. By doing this there is a risk of underestimating the strength of real associations.

In this thesis we chose to exclude the first year after study entry to avoid presenting

false associations.

MisclassificationMisclassificationMisclassificationMisclassification

There are no validation studies of the discharge diagnosis of CD. However, in a re-

cent study Olen et al139 confirmed the discharge diagnosis of CD by reviewing the

medical records with an estimated positive predictive value of 77%. Individuals in

this study were all adults and the positive predictive value should be even higher in

pediatric CD. Another study152 found that >85% of individuals with a discharge di-

agnosis of CD and concomitant lymphoma were correctly classified with regard to

CD.

Misclassification of CD is also uncommon in the hospital discharge register due to

the well-established practice to carry out a small-bowel biopsy before the CD diag-

nosis68. In addition, Nilsson et al223 have made a systematic analysis of the general

reliability of diagnosis of ICD-8 in the hospital discharge register and found that 88–

90% of listed main diagnoses were correct. In ICD-9 and -10 this percentage should

be even higher.

The use of small intestinal biopsy began in the early 1970s but the hospital dis-

charge register dates back to 1964. We therefore suspect that incorrect diagnoses of

CD are more common in the early time period (1964-73). Notably is that only few

study participants entered the study during this time period (500 out of total 15,000

participants). Also this time period has the lowest risk estimates, which indicates that

51

a low specificity of CD in the 1960s and 1970s should not lead to overestimated

relative risks.

It is also possible that the reference population contains a number of individuals

with CD (if they have not been admitted to hospital at any time). The effects of these

false negative controls are likely to be more conservative risk estimates.

The classification of diseases changes over time, partly as a result of better and

new diagnostic techniques leading to new diagnostic entities such as dilated cardio-

myopathy and hepatitis C (although the diagnosis of CD has remained the same over

time). The hospital discharge register also contains no data on most investigative

procedures; we were therefore unable to verify whether individuals with liver cirrho-

sis/fibrosis had undergone liver biopsy or if individuals with heart disease had under-

gone endomyocardial biopsy.

To reduce the risk of differential misclassification of our outcome measures, we

increased the specificity of our diagnosis through estimating the risk of having our

outcome measure listed as main diagnosis, or having at least two hospital discharge

diagnoses of our outcome measure and receiving the diagnosis in departments where

the degree of misclassification is supposed to be low. Restricting our outcome meas-

ure in this way did not influence our risk estimates.

Confounding factorsConfounding factorsConfounding factorsConfounding factors

Confounding by age was dealt with by matching. We also presented stratified results

for potential confounding factors such as age, sex and calendar period. We did not

have any information on smoking. However, earlier studies have suggested a nega-

tive association228 or no association at all between CD and smoking229. In studies I-IV

we also carried out additional analyses where we adjusted for diabetes and also so-

cioeconomic index (in a subset of individuals where we had available data on socio-

economic index) and this did not change our risk estimates notably.

Findings and implicationsFindings and implicationsFindings and implicationsFindings and implications

We used the Swedish hospital discharge register to examine the risk of liver disease,

autoimmune heart disease, Addison’s disease and thyroid disorders and in the last

study in this thesis we used all Swedish pathology registers and the national cancer

registry to examine the risk of cancer in individuals with CD. We chose to investigate

52

these conditions since they are of clinical importance and of both autoimmune and

inflammatory origin. Addison’s disease is a known autoimmune disorder and an as-

sociation between CD and Addison’s disease could support the theory of a generally

increased risk of autoimmunity in CD. This thesis also contributes in defining the

risk of associated disorders in CD and help to identify high risk groups where screen-

ing for CD is justified.

Study I Liver diseaseStudy I Liver diseaseStudy I Liver diseaseStudy I Liver disease

This study suggests that individuals with CD are at increased risk of both prior and

subsequent liver disease. The risk of liver disease was increased both when we com-

pared with reference individuals from a general population cohort and when we used

inpatient reference individuals. Despite the increased risk of liver failure, this study

found no increased risk of liver transplantation in individuals with CD. This is likely

due to few positive events and insufficient power. An alternative but not mutually

exclusive explanation is that long-term treatment with gluten-free diet (in this study,

more than 1 year) reduced the need for liver transplantation in individuals with se-

vere liver disease, as suggested by earlier research140, 230. For these reasons we cannot

exclude the possibility that individuals with untreated CD are at a greater risk of

liver transplantation.

The association between CD and liver disease could partly be explained by shared

autoimmunity, and it is well known that CD is a risk factor for autoimmune dis-

ease10, 135. This is supported by the fact that the highest risk estimates in our study

were seen in autoimmune liver disease or liver disease in which autoimmunity

probably plays an important role, such as PSC231, PBC179, acute hepatitis, and

chronic hepatitis.

Study II Study II Study II Study II Autoimmune heart diseaseAutoimmune heart diseaseAutoimmune heart diseaseAutoimmune heart disease

When individuals were analyzed for the whole study period, we found a statistically

significantly positive association of CD with later cardiomyopathy but not with later

myocarditis or pericarditis. We also found a positive association of prior cardio-

myopathy and pericarditis with later CD. Our findings of a positive association of

CD with cardiomyopathy are consistent with earlier data204-209, but add to these stud-

ies through larger numbers. The strengths of this study include the large number of

53

positive events that exceeds that of all previous studies on CD and cardiomyopathy

taken together and also that we were able to adjust for important confounders in-

cluding ischemic heart disease.

However this study was hampered by changing diagnostic criteria in heart disease

and limitations of heart diseases registered in ICD 7-8. Cardiomyopathy can be di-

vided into dilated, hypertrophic and restrictive cardiomyopathy. We were unable to

study dilated cardiomyopathy before 1987 since the IPR did not distinguish between

various forms of cardiomyopathy in the ICD7–8 classifications. We found no asso-

ciation between adulthood CD and heart disease when we restricted our analyses to

heart disease diagnosed in departments of cardiology and medicine from 1987 and

onwards (i.e. diagnosed with an ICD-9 or ICD-10 code). The lack of association in

this, our most clinically relevant analysis, indicates that the modest associations seen

between adulthood CD and heart disease diagnosed since 1964 are most likely due

to surveillance bias and ⁄ or low specificity of the outcome measures. Neither did this

study find any evidence of a positive association between CD diagnosed in childhood

and later heart disease. Taking into account the changes in the classification of heart

disease in the last decades, this association needs further studying.

Study IIIStudy IIIStudy IIIStudy III Addison’s disease Addison’s disease Addison’s disease Addison’s disease

Our result of an increased risk of Addison’s disease in CD confirms earlier studies on

CD and AD172-175. These studies were, however, limited in size and lack data on po-

tential confounders172-175. In contrast, our population-based study included more than

14,000 patients with CD and 70,000 reference individuals, and was matched for age,

sex, calendar year and county. This gives the study considerable power to identify

infrequent events and enabled us to examine the risk of AD in various strata. Strati-

fied analyses revealed that risk estimates were similar in males and females. High risk

estimates for AD were seen both in childhood CD and in CD diagnosed in adult-

hood.

The positive association between CD and AD was seen both prior to and after di-

agnosis of CD; and we do not believe that one disease causes the other. Instead we

suggest that our findings might be explained by the presence of shared genetic traits.

Addison’s disease is a disease of known autoimmune origin and the positive associa-

tion between CD and Addison’s disease found in this study supports the theory of a

general increased autoimmunity in CD.

54

Study IVStudy IVStudy IVStudy IV Thyroid disease Thyroid disease Thyroid disease Thyroid disease

We found a statistically significantly increased risk of hypothyroidism, thyroiditis

and hyperthyroidism in individuals with CD. The highest risk estimates were found

in children. Our study is the largest longitudinal study of the risk of TD in individu-

als with CD up to date. The large number of individuals with CD and the fact that

we used a matched reference population enabled us to calculate reliable risk esti-

mates.

Our study confirms the positive association between CD and TD. Although, ear-

lier studies have suggested a positive association between CD and TD10, 26, 131, 135, 158-

171, the majority of these studies have either been cross-sectional10, 26, 135, 158-171, lacked

controls26, 159, 170, 171 or failed to attain statistical significance10, 135, 163. In 2005 Vilja-

maa et al131 conducted a longitudinal study of TD in a cohort of individuals with

CD. This study indicated an increased incidence of TD in CD but results did not

reach statistical significance.

To the best knowledge of the authors, our study is the first study to present a sta-

tistically significant risk increase of hyperthyroidism in individuals with CD.

This study concludes that CD is associated with hypothyroidism and thyroiditis as

well as hyperthyroidism. These associations were seen regardless of temporal se-

quence, and we therefore suggest that the increased risk of TD in CD is an expres-

sion of a more general increase in autoimmunity that characterizes many individuals

with CD. The positive association between CD and TD may be due to shared genetic

or immunological traits. Clinicians need to have a high awareness of this association

and to test for these conditions when symptoms of thyroid disease or CD are sus-

pected.

Study VStudy VStudy VStudy V Hematopoietic malignancy Hematopoietic malignancy Hematopoietic malignancy Hematopoietic malignancy

Celiac disease

This is the largest study of lymphoma and CD up to date, and was based on 28,810

individuals with biopsy-verified CD. Among them were 268 cases of lymphoma. In

addition, it is the first study of lymphoma in individuals with small intestinal in-

flammation (Marsh 1-2) and latent CD (Marsh 0). The large number of individuals

55

with latent CD (n=3,552) allowed us to examine if positive CD serology alone can

predict the risk of future lymphoma.

CD is associated with an increased risk of lymphoma of both NHL and HL sub-

type. Previous studies of CD and lymphoma have shown varying risk estimates with

lower risk estimates in more recent papers. Our results are in line with a recent study

where the risk of cancer in CD was studied in a large cohort of individuals with

NHL156. The main advantage between our study and previous studies is that we used

biopsy data to identify CD. Individuals identified through biopsy are unselected and

therefore likely to be representative of the average individual with CD. The small in-

testinal biopsy also provides us with an exact date of diagnosis and hence com-

mencement of gluten-free diet.

There was no association between CD and leukemia in our study. The lack of as-

sociation between CD in children and subsequent lymphoma or leukemia may be a

consequence of insufficient follow-up (low attained age at end of follow-up) in these

individuals. However, our results rule out highly increased risks.

Small intestinal inflammation

To our knowledge this is the first study to investigate the risk of cancer in individuals

with small intestinal inflammation. The risk of overall lymphoma was elevated in in-

dividuals with inflammation, although the relative risk was lower than in individuals

with CD. There was an increased risk of NHL and the risk of HL was similar but did

not attain statistical significance. Other hematopoietic malignancies such as CLL,

leukemia or myeloma were not positively associated with inflammation.

Small intestinal inflammation is a heterogeneous condition. It is a mixture of both

acute and chronic inflammation. This might explain the lower risk estimates found

in individuals with small intestinal inflammation compared to individuals with CD,

where chronic inflammation is present until diagnosis. Small intestinal inflammation

can be the initial presentation of CD but can also be caused by a number of other

conditions such as helicobacter pylori infection, giardiasis, lymphoma or gastric

metaplasia80, 232-235. Inflammation without VA does not fulfill the criteria for CD and

these individuals do not receive a gluten-free diet. Despite the heterogeneity of the

condition Ludvigsson et al77 found co-morbidity other than CD in only a small mi-

nority of these biopsies when they manually reviewed 1534 biopsy reports in a vali-

dation study of villous atrophy and inflammation in Swedish biopsy registers. One

explanation for this might be that some of the other causes of small intestinal in-

56

flammation that was seen in other studies such as giardiasis78 and helicobacter pylori

infection 79 are rare in Sweden.

Latent Celiac disease

Latent CD is defined by National Institute of Health as individuals with a positive

serology test for CD, but with a normal small intestinal biopsy14, 70. The natural his-

tory of this condition has not previously been investigated. Although, there have

been reports of no association with malignancy in individuals with positive CD se-

rology216, 218, the researchers have not investigated the mucosa at time of positive CD

serology and it is likely that most such patients have either inflammation or VA.

With the increased use of serological testing in individuals with mild or no symptoms

this condition is increasing90, 93. Given that many patients with positive CD serology

have a normal mucosa (positive predictive value of TTGA in a recent British study

was <30%17), more information on the long-term consequences of latent CD is

needed.

We found no increased risks of lymphoma or any other hematopoietic cancers in

our study. This contrasts with the elevated risk estimates of lymphoma found in CD

and small intestinal mucosal inflammation. The cohort with latent CD in our study

is smaller than those of CD and inflammation and this might have affected the pos-

sibility to find a positive association. However, we nevertheless had 3,552 individu-

als with latent CD and this should enable us to rule out highly increased risks. We

suggest that morbidity in CD is dependant on the histopathological appearance of

the small intestine, and that individuals without tissue damage are at lower risk of

lymphoma. Our findings indicate that positive CD serology alone cannot be used to

predict future risk of lymphoma in CD and that small intestinal biopsy is needed in

determining the risk of lymphoma in CD.

57

ImplicaImplicaImplicaImplications for future researchtions for future researchtions for future researchtions for future research

Further research needs to be conducted to

Identify which patients with CD are most at risk of developing complications.

Identify high risk groups where screening for CD should be considered.

Determine if a high compliance to a GFD reduces the risk of adverse outcome in CD.

Further explore the natural history of CD, especially in individuals with milder forms

of CD, intestinal inflammation or latent CD. Do they have an increased risk of ma-

lignancy or autoimmune diseases.

Determine what mechanisms cause associated disorders in CD.

Determine if gluten ingestion affects the risk of autoimmunity.

Determine if pediatric and adult CD is the same disease.

Determine how serious CD is in children.

59

CCCConclusionsonclusionsonclusionsonclusions

This thesis concludes that individuals with CD are at an increased risk of a number

of autoimmune and inflammatory disorders. We suggest that this association is an

expression of a more general increase in autoimmunity that characterizes many indi-

viduals with CD. Patients with CD are increasingly identified due to associated con-

ditions. The widespread availability of serologic tests for CD allows the diagnosis to

be considered by any physician. This suggests that clinicians need to have a high

awareness of these associations and to test for these conditions when symptoms ap-

pear.

Specifically, we conclude that

Individuals with CD are at an increased risk of acute and chronic hepatitis, primary

sclerosing cholangitis, fatty liver, liver failure, liver cirrhosis/fibrosis and primary bil-

iary cirrhosis.

We found no association between CD and dilated cardiomyopathy, pericarditis and

myocardititis.

We found a highly increased risk of Addison’s disease, a disease of known autoim-

mune origin, in individuals with CD.

There is a statistically significantly increased risk of hypothyroidism, thyroiditis and

hyperthyroidism in individuals with CD, with the highest risk estimates in pediatric

CD.

We found increased risks of both NHL and HL in individuals with CD. The inci-

dence and mortality in NHL is rising236 and the positive association between NHL

and a potentially treatable disorder CD is therefore of clinical significance. This

study is the first to estimate the risk of malignancy in individuals with small intesti-

nal inflammation and individuals with latent CD. While individuals with inflamma-

tion were at an increased risk of lymphoma, the risk of lymphoma in latent CD was

not increased. This indicates that positive CD serology alone cannot be used to pre-

dict future risk of lymphoma in CD and that small intestinal biopsy is needed in de-

termining the risk of lymphoma in CD.

61

SSSSammanfattninammanfattninammanfattninammanfattning på svenskag på svenskag på svenskag på svenska

Celiaki eller glutenintolerans, som sjukdomen heter på svenska, är en av de vanligas-

te kroniska sjukdomarna och drabbar ca 1 % av alla barn och ungdomar i Sverige. I

allt högre utsträckning diagnostiseras även sjukdomen hos vuxna. Behandlingen be-

står av en livslång glutenfri kost.

Varför man får glutenintolerans är inte helt klarlagt. Det är många faktorer som

samverkar. Det man vet är att sjukdomen utlöses av glutenexponering och att den

förekommer företrädesvis hos individer med vissa genetiska förutsättningar (HLA-

DQ2 eller DQ8-positiva). Gluten är ett protein som finns i vete. Glutenliknande äm-

nen finns även i råg och korn. Hos dem som utvecklar glutenintolerans leder intag av

gluten till en inflammation av tunntarmen och att tarmluddet försvinner. Detta leder

till svårigheter för kroppen att ta upp viktiga näringsämnen. Vanliga symptom på

glutenintolerans är viktnedgång, trötthet, diarré, buksmärtor och blodbrist (anemi).

Många har dock endast lindriga eller inga symptom alls och riskerar att gå oupp-

täckta under lång tid och därmed öka risken för bristtillstånd och komplikationer.

Tidigare forskning har visat att patienter med glutenintolerans löper en ökad risk för

en rad andra sjukdomar bla diabetes och benskörhet. Diagnosen glutenintolerans

ställs genom att ta ett vävnadsprov från tunntarmen. Är tarmluddet borta är detta

tecken på glutenintolerans. Ofta har man före vävnadsprovet taget blodprov för att

se om det finns antikroppar mot glutenintolerans. Blodprov med hög tillförlitlighet

har funnits att tillgå sedan 1990-talet. Att enbart använda blodprov räcker dock inte

för att ställa diagnos utan det krävs ett vävnadsprov från tunntarmen med påvisad

slemhinneskada.

Målsättningen med mitt avhandlingsarbete är att belysa risken hos individer med

glutenintolerans att utveckla associerade sjukdomar. För att göra pålitliga riskbe-

dömningar krävs stora patientmaterial. I Sverige finns flera nationella dataregister av

mycket hög kvalitet och med hjälp av personnumret går denna information att kopp-

las samman och användas vid forskning. Vår forskning kan bidra till att identifiera

risken för annan sjuklighet vid glutenintolerans samt att identifiera riskgrupper där

screening för glutenintolerans kan bli aktuellt. På sikt hoppas vi att vår forskning ska

leda till ökad kunskap om associerade sjukdomar vid glutenintolerans och därmed

minska risken för att patienter går med oupptäckta och obehandlade följdsjukdo-

mar. Avhandlingen består av fem delarbeten som belyser risk för leversjukdom,

62

hjärtsjukdom, Addisons sjukdom, sköldkörtelrubbningar samt cancer i vita blod-

kroppar hos patienter med glutenintolerans.

I studie 1-4 har min forskargrupp via slutenvårdsregistret identifierat alla patienter

som fått diagnosen glutenintolerans under perioden 1964-2003. För varje individ

med har statistiska centralbyrån identifierat 5 referensindivider matchade för kön,

ålder, och län. Via slutenvårdsregistret har även information om annan sjuklighet

hos samtliga individer hämtats in. Därefter har statistiska riskberäkningar gjorts för

respektive sjukdom utifrån materialet. Studierna är baserade på ca 15000 individer

med glutenintolerans och ca 75000 referensindivider.

I den första studien har vi undersökt risken för olika typer av leversjukdomar hos

individer med glutenintolerans. Vår studie visade på en ca 5 gånger ökad risk för he-

patit (leverinflammation), 4 gånger ökad risk för primär skleroserande kolangit, 10

gånger ökad risk för primär biliär cirrhos, 6 gånger ökad risk för leverförfettning och

ca 3 gånger ökad risk för leversvikt.

Därefter undersökte vi risken för autoimmuna hjärtsjukdomar vid glutenintolerans

då detta var bristfälligt studerat sedan tidigare. Studien kunde inte påvisa någon

ökad risk för hjärtmuskelsjukdom (kardiomyopati), hjärtmuskelinflammation (myo-

kardit) eller hjärtsäcksinflammation (perikardit) trots att man sett ett samband i tidi-

gare mindre studier. Vår studie komplicerades dock av att diagnoskriterierna för

hjärtmuskelsjukdom med tiden ändrats och att diagnosen delats upp i undergrupper.

Addisons sjukdom är en sjukdom där binjurarnas funktion försämras vilket leder

till kortisolbrist. I Sverige är Addisons sjukdom en ren autoimmun sjukdom som

lämpar sig väl ifall man vill studera en eventuell autoimmun koppling vid gluteninto-

lerans. Vi fann en kraftigt förhöjd risk för Addisons sjukdom vid glutenintolerans på

ca 11 ggr ökad risk. Fynden bekräftar att glutenintolerans bör anses som en auto-

immun sjukdom.

I nästa studie jämförde vi individer med glutenintolerans och referensindivider

med avseende på risk för sköldkörtelrubbningar. Glutenintolerans var kopplat till en

ökad risk av underfunktion av sköldkörteln (hypothyreos) med en ca 4 ggr ökad risk

för vuxna och ca 6 ggr ökad risk när studien begränsades till barn. För sköldkörtel-

inflammation (thyroidit) var den ökade risken ca 2 ggr ökad för vuxna och 5 ggr

ökad för barn. Studien var även den första hittills att kunna visa på en statistiskt sä-

kerställd riskökning för en överfunktion av sköldkörteln (hyperthyreos) med en risk-

ökning på ca 2 ggr ökad risk för vuxna och 5 ggr ökad risk för barn.

63

I den sista studien undersökte vi risken för personer med glutenintolerans att ut-

veckla lymfom och leukemi. Här har vi kopplat samman data från alla tunntarms-

undersökningar som gjorts i Sverige och därmed identifierat 28810 personer som

saknar tarmludd (celiaki) och 12681 personer som har tunntarmsinflammation men

där tarmluddet finns kvar. Tunntarmsinflammation kan vara ett förstadium till glu-

tenintolerans men behandlas inte med glutenfri kost. Vidare har vi identifierat 3552

personer med normal slemhinna men där blodprov talar för glutenintolerans.

Studieindividerna har sedan länkats till cancerregistret inför analys. Risken för

cancer hos personer med celiaki/inflammation kan sedan jämföras med den i nor-

malbefolkningen.

Studien visar på en ca 3 gånger ökad risk för lymfom vid glutenintolerans och en

fördubblad risk vid tunntarmsinflammation. Det fanns ingen ökad risk för lymfom

om man enbart hade förhöjda blodprover men ett normalt slemhinneprov. Detta in-

nebär att ett vävnadsprov krävs ifall man ska uttala sig om risken för lymfom vid

glutenintolerans. Studien påvisade ingen ökad risk för leukemi.

Slutligen kan det konstateras att det finns en ökad risk för en rad autoimmuna och

inflammatoriska tillstånd vid glutenintolerans. Denna kunskap är av vikt för att

kunna diagnostisera och behandla dessa patienter på bästa sätt.

65

AAAAcknowledgementscknowledgementscknowledgementscknowledgements

I wish to express my warm and sincere gratitude to:

Associate Professor Jonas Ludvigsson, my friend and tutor, for initiating and guiding

me in the field of research, for your joy and enthusiasm for our research, for your

never-ending support and encouragement. I could not imagine a better tutor.

Professor Anders Ekbom, co-tutor, for supporting me and giving me advice along the

way, for your laughter and spreading of a positive atmosphere. It has been a pleasure

to work with you.

Professor Scott Montgomery, co-author, for sharing your extensive knowledge and

advice in epidemiology in many stimulating discussions.

Professor Olle Kämpe, co-author, for generously contributing with your time, exper-

tise and knowledge in the field of endocrinology, and for letting me participate in a

very interesting day in Uppsala.

Fredrik Granath, for sharing your excellence in cancer statistics.

My co-authors Karin Ekström-Smedby and Johan Askling, for contributing with

your knowledge in cancer epidemiology. It is a privilege to work with you.

Professor Anders Hamsten, for your expertise in cardiology.

The late Ulrika Bromé, for her excellent contribution on the paper on celiac disease

and liver disease.

Lars Stenhammar and Curt Tysk, for your comprehensive and constructive criticism

at my half time seminar and also for your great knowledge in gastroenterology.

All my fellow colleagues at the department of Pediatrics, Örebro University Hospital,

for supporting me and being great colleagues and friends.

66

The research council at Örebro University Hospital and the Örebro University for

supporting my research.

The Medical Library at Örebro University hospital, and Margaretha Landin in par-

ticular, for excellent bibliographic service and information about ENDNOTE.

My friend and neighbour Eva Möller for your support and practical help.

My parents Gustaf and Margaretha, for your encouragement and for always being

there for me.

Finally, I would like to express my deepest gratitude to my beloved family, Elinor

and David, for all your love, support and inspiration. This thesis could not, in any

way, have been possible without you.

67

RRRReferenceseferenceseferenceseferences

1. Green PH, Cellier C. Celiac disease. N Engl J Med 2007;357(17):1731-43.

2. Gee S. On the coeliac affection. St. Bartholomew's Hospital Reports 1888;24:17.

3. Dicke WK, Weijers HA, Van De Kamer JH. Coeliac disease. II. The presence in

wheat of a factor having a deleterious effect in cases of coeliac disease. Acta Paediatr

1953;42(1):34-42.

4. Sollid LM, Markussen G, Ek J, Gjerde H, Vartdal F, Thorsby E. Evidence for a pri-

mary association of celiac disease to a particular HLA-DQ alpha/beta heterodimer. J

Exp Med 1989;169(1):345-50.

5. Dieterich W, Ehnis T, Bauer M, Donner P, Volta U, Riecken EO, et al. Identification

of tissue transglutaminase as the autoantigen of celiac disease [see comments]. Nat

Med 1997;3(7):797-801.

6. Dieterich W, Laag E, Schopper H, Volta U, Ferguson A, Gillett H, et al. Autoanti-

bodies to tissue transglutaminase as predictors of celiac disease [see comments]. Gas-

troenterology 1998;115(6):1317-21.

7. Dube C, Rostom A, Sy R, Cranney A, Saloojee N, Garritty C, et al. The prevalence

of celiac disease in average-risk and at-risk Western European populations: a system-

atic review. Gastroenterology 2005;128(4 Suppl 1):S57-67.

8. Ludvigsson JF, Ansved P, Falth-Magnusson K, Hammersjo JA, Johansson C, Ed-

vardsson S, et al. Symptoms and Signs Have Changed in Swedish Children With

Coeliac Disease. J Pediatr Gastroenterol Nutr 2004;38(2):181-6.

9. van Heel DA, West J. Recent advances in coeliac disease. Gut 2006;55(7):1037-46.

10. Ventura A, Magazzu G, Greco L. Duration of exposure to gluten and risk for auto-

immune disorders in patients with celiac disease. SIGEP Study Group for Autoim-

mune Disorders in Celiac Disease. Gastroenterology 1999;117(2):297-303.

11. Ludvigsson JF, Elfstrom P, Broome U, Ekbom A, Montgomery SM. Celiac disease

and risk of liver disease: a general population-based study. Clin Gastroenterol Hepa-

tol 2007;5(1):63-9 e1.

12. Collin P, Kaukinen K, Valimaki M, Salmi J. Endocrinological disorders and celiac

disease. Endocr Rev 2002;23(4):464-83.

13. Revised criteria for diagnosis of coeliac disease. Report of Working Group of Euro-

pean Society of Paediatric Gastroenterology and Nutrition. Arch Dis Child

1990;65(8):909-11.

14. National Institutes of Health Consensus Development Conference Statement on Ce-

liac Disease, June 28-30, 2004. Gastroenterology 2005;128(4 Suppl 1):S1-9.

68

15. Rostom A, Murray JA, Kagnoff MF. American Gastroenterological Association

(AGA) Institute technical review on the diagnosis and management of celiac disease.

Gastroenterology 2006;131(6):1981-2002.

16. Fasano A, Araya M, Bhatnagar S, Cameron D, Catassi C, Dirks M, et al. Federation

of International Societies of Pediatric Gastroenterology, Hepatology, and Nutrition

Consensus report on celiac disease. J Pediatr Gastroenterol Nutr 2008;47(2):214-9.

17. Hopper AD, Cross SS, Hurlstone DP, McAlindon ME, Lobo AJ, Hadjivassiliou M, et

al. Pre-endoscopy serological testing for coeliac disease: evaluation of a clinical deci-

sion tool. Bmj 2007;334(7596):729.

18. Janatuinen EK, Pikkarainen PH, Kemppainen TA, Kosma VM, Jarvinen RM, Uusi-

tupa MI, et al. A comparison of diets with and without oats in adults with celiac dis-

ease [see comments]. N Engl J Med 1995;333(16):1033-7.

19. Peraaho M, Kaukinen K, Paasikivi K, Sievanen H, Lohiniemi S, Maki M, et al.

Wheat-starch-based gluten-free products in the treatment of newly detected coeliac

disease: prospective and randomized study. Aliment Pharmacol Ther

2003;17(4):587-94.

20. Sollid LM, Jabri B. Is celiac disease an autoimmune disorder? Curr Opin Immunol

2005;17(6):595-600.

21. Alaedini A, Green PH. Narrative review: celiac disease: understanding a complex

autoimmune disorder. Ann Intern Med 2005;142(4):289-98.

22. Ivarsson A. The Swedish epidemic of coeliac disease explored using an epidemiologi-

cal approach--some lessons to be learnt. Best Pract Res Clin Gastroenterol

2005;19(3):425-40.

23. Norris JM, Barriga K, Hoffenberg EJ, Taki I, Miao D, Haas JE, et al. Risk of celiac

disease autoimmunity and timing of gluten introduction in the diet of infants at in-

creased risk of disease. Jama 2005;293(19):2343-51.

24. Welander A, Röckert-Tjärnberg A, Montgomery SM, Ludvigsson J, Ludvigsson JF.

Infectious disease and risk of later celiac disease in childhood. Manuscript 2008.

25. Hallert C, Gotthard R, Jansson G, Norrby K, Walan A. Similar prevalence of coeliac

disease in children and middle-aged adults in a district of Sweden. Gut

1983;24(5):389-91.

26. Midhagen G, Jarnerot G, Kraaz W. Adult coeliac disease within a defined geographic

area in Sweden. A study of prevalence and associated diseases. Scand J Gastroenterol

1988;23(8):1000-4.

27. Ascher H, Krantz I, Kristiansson B. Increasing incidence of coeliac disease in Sweden.

Arch Dis Child 1991;66(5):608-11.

28. Swinson CM, Levi AJ. Is coeliac disease underdiagnosed? Br Med J

1980;281(6250):1258-60.

69

29. Rostom A, Dube C, Cranney A, Saloojee N, Sy R, Garritty C, et al. The diagnostic

accuracy of serologic tests for celiac disease: a systematic review. Gastroenterology

2005;128(4 Suppl 1):S38-46.

30. Maki M, Mustalahti K, Kokkonen J, Kulmala P, Haapalahti M, Karttunen T, et al.

Prevalence of Celiac disease among children in Finland. N Engl J Med

2003;348(25):2517-24.

31. Myleus A, Ivarsson A, Webb C, al. E. Celiac Disease Revealed in 3% of Swedish 12-

year-olds Born During an Epidemic. J Pediatr Gastroenterol Nutr 2008;In press.

32. West J, Logan RF, Hill PG, Lloyd A, Lewis S, Hubbard R, et al. Seroprevalence, cor-

relates, and characteristics of undetected coeliac disease in England. Gut

2003;52(7):960-5.

33. Fasano A, Berti I, Gerarduzzi T, Not T, Colletti RB, Drago S, et al. Prevalence of ce-

liac disease in at-risk and not-at-risk groups in the United States: a large multicenter

study. Arch Intern Med 2003;163(3):286-92.

34. Lohi S, Mustalahti K, Kaukinen K, Laurila K, Collin P, Rissanen H, et al. Increasing

prevalence of coeliac disease over time. Aliment Pharmacol Ther 2007;26(9):1217-

25.

35. Vilppula A, Collin P, Maki M, Valve R, Luostarinen M, Krekela I, et al. Undetected

coeliac disease in the elderly: a biopsy-proven population-based study. Dig Liver Dis

2008;40(10):809-13.

36. Tommasini A, Not T, Kiren V, Baldas V, Santon D, Trevisiol C, et al. Mass screening

for coeliac disease using antihuman transglutaminase antibody assay. Arch Dis Child

2004;89(6):512-5.

37. Sood A, Midha V, Sood N, Avasthi G, Sehgal A. Prevalence of celiac disease among

school children in Punjab, North India. J Gastroenterol Hepatol 2006;21(10):1622-

5.

38. Gomez JC, Selvaggio GS, Viola M, Pizarro B, la Motta G, de Barrio S, et al. Preva-

lence of celiac disease in Argentina: screening of an adult population in the La Plata

area. Am J Gastroenterol 2001;96(9):2700-4.

39. Hovell CJ, Collett JA, Vautier G, Cheng AJ, Sutanto E, Mallon DF, et al. High

prevalence of coeliac disease in a population-based study from Western Australia: a

case for screening? Med J Aust 2001;175(5):247-50.

40. Catassi C, Doloretta Macis M, Ratsch IM, De Virgiliis S, Cucca F. The distribution

of DQ genes in the Saharawi population provides only a partial explanation for the

high celiac disease prevalence. Tissue Antigens 2001;58(6):402-6.

41. Cavell B, Stenhammar L, Ascher H, Danielsson L, Dannaeus A, Lindberg T, et al.

Increasing incidence of childhood coeliac disease in Sweden. Results of a national

study. Acta Paediatr 1992;81(8):589-92.

70

42. Ludvigsson J. Why diabetes incidence increases--a unifying theory. Ann N Y Acad Sci

2006;1079:374-82.

43. Ivarsson A, Hernell O, Stenlund H, Persson LA. Breast-feeding protects against celiac

disease. Am J Clin Nutr 2002;75(5):914-21.

44. Stene LC, Honeyman MC, Hoffenberg EJ, Haas JE, Sokol RJ, Emery L, et al. Rotavi-

rus infection frequency and risk of celiac disease autoimmunity in early childhood: a

longitudinal study. Am J Gastroenterol 2006;101(10):2333-40.

45. Jabri B, Kasarda DD, Green PH. Innate and adaptive immunity: the Yin and Yang of

celiac disease. Immunol Rev 2005;206:219-31.

46. Kagnoff MF. Overview and pathogenesis of celiac disease. Gastroenterology

2005;128(4 Suppl 1):S10-8.

47. Gudjonsdottir AH, Nilsson S, Ek J, Kristiansson B, Ascher H. The risk of celiac dis-

ease in 107 families with at least two affected siblings. J Pediatr Gastroenterol Nutr

2004;38(3):338-42.

48. Hervonen K, Hakanen M, Kaukinen K, Collin P, Reunala T. First-degree relatives

are frequently affected in coeliac disease and dermatitis herpetiformis. Scand J Gas-

troenterol 2002;37(1):51-5.

49. Hogberg L, Falth-Magnusson K, Grodzinsky E, Stenhammar L. Familial prevalence

of coeliac disease: a twenty-year follow-up study. Scand J Gastroenterol

2003;38(1):61-5.

50. Greco L, Romino R, Coto I, Di Cosmo N, Percopo S, Maglio M, et al. The first large

population based twin study of coeliac disease. Gut 2002;50(5):624-8.

51. Sollid LM. Coeliac disease: dissecting a complex inflammatory disorder. Nat Rev

Immunol 2002;2(9):647-55.

52. Holopainen P, Arvas M, Sistonen P, Mustalahti K, Collin P, Maki M, et al.

CD28/CTLA4 gene region on chromosome 2q33 confers genetic susceptibility to ce-

liac disease. A linkage and family-based association study. Tissue Antigens

1999;53(5):470-5.

53. Djilali-Saiah I, Schmitz J, Harfouch-Hammoud E, Mougenot JF, Bach JF, Caillat-

Zucman S. CTLA-4 gene polymorphism is associated with predisposition to coeliac

disease. Gut 1998;43(2):187-9.

54. Popat S, Hearle N, Hogberg L, Braegger CP, O'Donoghue D, Falth-Magnusson K, et

al. Variation in the CTLA4/CD28 gene region confers an increased risk of coeliac

disease. Ann Hum Genet 2002;66(Pt 2):125-37.

55. Marron MP, Raffel LJ, Garchon HJ, Jacob CO, Serrano-Rios M, Martinez Larrad

MT, et al. Insulin-dependent diabetes mellitus (IDDM) is associated with CTLA4

polymorphisms in multiple ethnic groups. Hum Mol Genet 1997;6(8):1275-82.

71

56. Kotsa K, Watson PF, Weetman AP. A CTLA-4 gene polymorphism is associated with

both Graves disease and autoimmune hypothyroidism. Clin Endocrinol (Oxf)

1997;46(5):551-4.

57. Sollid LM. Molecular basis of celiac disease. Annu Rev Immunol 2000;18:53-81.

58. Stenberg P, Roth EB, Sjoberg K. Transglutaminase and the pathogenesis of coeliac

disease. Eur J Intern Med 2008;19(2):83-91.

59. Farrell RJ, Kelly CP. Celiac sprue. N Engl J Med 2002;346(3):180-8.

60. Fasano A. Clinical presentation of celiac disease in the pediatric population. Gastro-

enterology 2005;128(4 Suppl 1):S68-73.

61. Hopper AD, Hadjivassiliou M, Butt S, Sanders DS. Adult coeliac disease. Bmj

2007;335(7619):558-62.

62. Murray JA. Celiac disease in patients with an affected member, type 1 diabetes, iron-

deficiency, or osteoporosis? Gastroenterology 2005;128(4 Suppl 1):S52-6.

63. Hin H, Bird G, Fisher P, Mahy N, Jewell D. Coeliac disease in primary care: case

finding study. Bmj 1999;318(7177):164-7.

64. Bardella MT, Fraquelli M, Quatrini M, Molteni N, Bianchi P, Conte D. Prevalence

of hypertransaminasemia in adult celiac patients and effect of gluten-free diet. Hepa-

tology 1995;22(3):833-6.

65. Novacek G, Miehsler W, Wrba F, Ferenci P, Penner E, Vogelsang H. Prevalence and

clinical importance of hypertransaminasaemia in coeliac disease. Eur J Gastroenterol

Hepatol 1999;11(3):283-8.

66. Olen O, Montgomery SM, Marcus C, Ekbom A, Ludvigsson JF. Celiac disease and

Body Mass Index - A study of two Swedish general population based registers.

Manuscript 2008.

67. Maki M, Kallonen K, Lahdeaho ML, Visakorpi JK. Changing pattern of childhood

coeliac disease in Finland. Acta Paediatr Scand 1988;77(3):408-12.

68. Stenhammar L, Hogberg L, Danielsson L, Ascher H, Dannaeus A, Hernell O, et al.

How do Swedish paediatric clinics diagnose coeliac disease? Results of a nationwide

questionnaire study. Acta Paediatr 2006;95(11):1495-7.

69. Wahab PJ, Crusius JB, Meijer JW, Mulder CJ. Gluten challenge in borderline gluten-

sensitive enteropathy. Am J Gastroenterol 2001;96(5):1464-9.

70. Dewar DH, Ciclitira PJ. Clinical features and diagnosis of celiac disease. Gastroen-

terology 2005;128(4 Suppl 1):S19-24.

71. Collin P, Wahab PJ, Murray JA. Intraepithelial lymphocytes and coeliac disease. Best

Pract Res Clin Gastroenterol 2005;19(3):341-50.

72. Briani C, Samaroo D, Alaedini A. Celiac disease: From gluten to autoimmunity.

Autoimmun Rev 2008;7(8):644-50.

72

73. Marsh MN. Gluten, major histocompatibility complex, and the small intestine. A

molecular and immunobiologic approach to the spectrum of gluten sensitivity ('celiac

sprue'). Gastroenterology 1992;102(1):330-54.

74. Hopper AD, Cross SS, Sanders DS. Patchy villous atrophy in adult patients with sus-

pected gluten-sensitive enteropathy: is a multiple duodenal biopsy strategy appropri-

ate? Endoscopy 2008;40(3):219-24.

75. Dickey W. Endoscopic markers for celiac disease. Nat Clin Pract Gastroenterol

Hepatol 2006;3(10):546-51.

76. Hill ID, Dirks MH, Liptak GS, Colletti RB, Fasano A, Guandalini S, et al. Guideline

for the diagnosis and treatment of celiac disease in children: recommendations of the

North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. J

Pediatr Gastroenterol Nutr 2005;40(1):1-19.

77. Ludvigsson JF, Brandt L, Montgomery SM, Granath F, Ekbom A. Validation study

of villous atrophy and small intestinal inflammation in Swedish biopsy registers.

Manuscript 2008.

78. Gardiasis incidence in Sweden. Stockholm: The Swedish Institute for Infectious Dis-

ease Control (SMI); 2008 [cited 2009 4 Jan ]. Available from: http://www.smi.se/in-

english/statistics/giardiaiosis/?base=domestic.

79. Thjodleifsson B, Asbjornsdottir H, Sigurjonsdottir RB, Gislason D, Olafsson I, Cook

E, et al. Seroprevalence of Helicobacter pylori and cagA antibodies in Iceland, Esto-

nia and Sweden. Scand J Infect Dis 2007;39(8):683-9.

80. Kakar S, Nehra V, Murray JA, Dayharsh GA, Burgart LJ. Significance of intraepithe-

lial lymphocytosis in small bowel biopsy samples with normal mucosal architecture.

Am J Gastroenterol 2003;98(9):2027-33.

81. Dickson BC, Streutker CJ, Chetty R. Coeliac disease: an update for pathologists. J

Clin Pathol 2006;59(10):1008-16.

82. Unsworth DJ, Manuel PD, Walker-Smith JA, Campbell CA, Johnson GD, Holborow

EJ. New immunofluorescent blood test for gluten sensitivity. Arch Dis Child

1981;56(11):864-8.

83. Lagerqvist C, Dahlbom I, Hansson T, Jidell E, Juto P, Olcen P, et al. Antigliadin im-

munoglobulin A best in finding celiac disease in children younger than 18 months of

age. J Pediatr Gastroenterol Nutr 2008;47(5):428-35.

84. Chorzelski TP, Beutner EH, Sulej J, Tchorzewska H, Jablonska S, Kumar V, et al.

IgA anti-endomysium antibody. A new immunological marker of dermatitis herpeti-

formis and coeliac disease. Br J Dermatol 1984;111(4):395-402.

85. Hill PG, McMillan SA. Anti-tissue transglutaminase antibodies and their role in the

investigation of coeliac disease. Ann Clin Biochem 2006;43(Pt 2):105-17.

86. Collin P, Maki M, Keyrilainen O, Hallstrom O, Reunala T, Pasternack A. Selective

IgA deficiency and coeliac disease. Scand J Gastroenterol 1992;27(5):367-71.

73

87. Korponay-Szabo IR, Szabados K, Pusztai J, Uhrin K, Ludmany E, Nemes E, et al.

Population screening for coeliac disease in primary care by district nurses using a

rapid antibody test: diagnostic accuracy and feasibility study. Bmj

2007;335(7632):1244-7.

88. Hadithi M, von Blomberg BM, Crusius JB, Bloemena E, Kostense PJ, Meijer JW, et

al. Accuracy of serologic tests and HLA-DQ typing for diagnosing celiac disease. Ann

Intern Med 2007;147(5):294-302.

89. Abrams JA, Diamond B, Rotterdam H, Green PH. Seronegative celiac disease: in-

creased prevalence with lesser degrees of villous atrophy. Dig Dis Sci

2004;49(4):546-50.

90. Rostami K, Kerckhaert J, Tiemessen R, von Blomberg BM, Meijer JW, Mulder CJ.

Sensitivity of antiendomysium and antigliadin antibodies in untreated celiac disease:

disappointing in clinical practice [see comments]. Am J Gastroenterol

1999;94(4):888-94.

91. Agardh D. Antibodies against synthetic deamidated gliadin peptides and tissue trans-

glutaminase for the identification of childhood celiac disease. Clin Gastroenterol

Hepatol 2007;5(11):1276-81.

92. Liu E, Rewers M, Eisenbarth GS. Genetic testing: who should do the testing and

what is the role of genetic testing in the setting of celiac disease? Gastroenterology

2005;128(4 Suppl 1):S33-7.

93. Green PH, Rostami K, Marsh MN. Diagnosis of coeliac disease. Best Pract Res Clin

Gastroenterol 2005;19(3):389-400.

94. Kaukinen K, Partanen J, Maki M, Collin P. HLA-DQ typing in the diagnosis of ce-

liac disease. Am J Gastroenterol 2002;97(3):695-9.

95. Fasano A. European and North American populations should be screened for coeliac

disease. Gut 2003;52(2):168-9.

96. Cranney A, Rostom A, Sy R, Dube C, Saloogee N, Garritty C, et al. Consequences of

testing for celiac disease. Gastroenterology 2005;128(4 Suppl 1):S109-20.

97. WHO. Screening criteria. WHO; 2008 [cited 2008 15 Nov]. Available from:

http://www.who.int/en/.

98. Hoffenberg EJ. Should all children be screened for celiac disease? Gastroenterology

2005;128(4 Suppl 1):S98-103.

99. Collin P. Should adults be screened for celiac disease? What are the benefits and

harms of screening? Gastroenterology 2005;128(4 Suppl 1):S104-8.

100. Kumar PJ. European and North American populations should be screened for coeliac

disease. Gut 2003;52(2):170-1.

101. Sjoberg K, Carlsson A. [Screening gor celiac disease can be justified in high-risk

groups]. Lakartidningen 2004;101(48):3912, 5-6, 8-9.

74

102. Carlsson AK, Axelsson IE, Borulf SK, Bredberg AC, Lindberg BA, Sj#berg KG, et al.

Prevalence of IgA-Antiendomysium and IgA-Antigliadin Autoantibodies at Diagnosis

of Insulin-Dependent Diabetes Mellitus in Swedish Children and Adolescents. Pediat-

rics 1999;103(6 Pt 1):1248-52.

103. Barker JM. Clinical review: Type 1 diabetes-associated autoimmunity: natural his-

tory, genetic associations, and screening. J Clin Endocrinol Metab 2006;91(4):1210-

7.

104. Carlsson A, Axelsson I, Borulf S, Bredberg A, Forslund M, Lindberg B, et al. Preva-

lence of IgA-antigliadin antibodies and IgA-antiendomysium antibodies related to ce-

liac disease in children with Down syndrome. Pediatrics 1998;101(2):272-5.

105. Berti I, Della Vedova R, Paduano R, Devetta M, Caradonna M, Villanacci V, et al.

Coeliac disease in primary care: evaluation of a case-finding strategy. Dig Liver Dis

2006;38(7):461-7.

106. Lee SK, Lo W, Memeo L, Rotterdam H, Green PH. Duodenal histology in patients

with celiac disease after treatment with a gluten-free diet. Gastrointest Endosc

2003;57(2):187-91.

107. Janatuinen EK, Kemppainen TA, Julkunen RJ, Kosma VM, Maki M, Heikkinen M,

et al. No harm from five year ingestion of oats in coeliac disease. Gut

2002;50(3):332-5.

108. Hogberg L, Laurin P, Falth-Magnusson K, Grant C, Grodzinsky E, Jansson G, et al.

Oats to children with newly diagnosed coeliac disease: a randomised double blind

study. Gut 2004;53(5):649-54.

109. Hoffenberg EJ, Haas J, Drescher A, Barnhurst R, Osberg I, Bao F, et al. A trial of

oats in children with newly diagnosed celiac disease. J Pediatr 2000;137(3):361-6.

110. Catassi C, Fabiani E, Iacono G, D'Agate C, Francavilla R, Biagi F, et al. A prospec-

tive, double-blind, placebo-controlled trial to establish a safe gluten threshold for pa-

tients with celiac disease. Am J Clin Nutr 2007;85(1):160-6.

111. Collin P, Thorell L, Kaukinen K, Maki M. The safe threshold for gluten contamina-

tion in gluten-free products. Can trace amounts be accepted in the treatment of

coeliac disease? Aliment Pharmacol Ther 2004;19(12):1277-83.

112. Collin P, Maki M, Kaukinen K. Safe gluten threshold for patients with celiac disease:

some patients are more tolerant than others. Am J Clin Nutr 2007;86(1):260; author

reply -1.

113. CodexAlimentariusCommission. Guidelines for gluten-free products. WHO; 2008

[cited 2009 6 Jan]. Available from:

http://www.codexalimentarius.net/web/archives.jsp?year=08.

114. Hallert C, Grant C, Grehn S, Granno C, Hulten S, Midhagen G, et al. Evidence of

poor vitamin status in coeliac patients on a gluten-free diet for 10 years. Aliment

Pharmacol Ther 2002;16(7):1333-9.

75

115. Gass J, Bethune MT, Siegel M, Spencer A, Khosla C. Combination enzyme therapy

for gastric digestion of dietary gluten in patients with celiac sprue. Gastroenterology

2007;133(2):472-80.

116. Olsson C, Hernell O, Hornell A, Lonnberg G, Ivarsson A. Difference in celiac disease

risk between Swedish birth cohorts suggests an opportunity for primary prevention.

Pediatrics 2008;122(3):528-34.

117. Tursi A, Brandimarte G, Giorgetti GM, Elisei W, Inchingolo CD, Monardo E, et al.

Endoscopic and histological findings in the duodenum of adults with celiac disease

before and after changing to a gluten-free diet: a 2-year prospective study. Endoscopy

2006;38(7):702-7.

118. Ciacci C, Cirillo M, Cavallaro R, Mazzacca G. Long-term follow-up of celiac adults

on gluten-free diet: prevalence and correlates of intestinal damage. Digestion

2002;66(3):178-85.

119. Pietzak MM. Follow-up of patients with celiac disease: achieving compliance with

treatment. Gastroenterology 2005;128(4 Suppl 1):S135-41.

120. Hogberg L, Grodzinsky E, Stenhammar L. Better dietary compliance in patients with

coeliac disease diagnosed in early childhood. Scand J Gastroenterol 2003;38(7):751-

4.

121. Fabiani E, Taccari LM, Ratsch IM, Di Giuseppe S, Coppa GV, Catassi C. Compli-

ance with gluten-free diet in adolescents with screening-detected celiac disease: A 5-

year follow-up study [In Process Citation]. J Pediatr 2000;136(6):841-3.

122. Vahedi K, Mascart F, Mary JY, Laberenne JE, Bouhnik Y, Morin MC, et al. Reliabil-

ity of antitransglutaminase antibodies as predictors of gluten-free diet compliance in

adult celiac disease. Am J Gastroenterol 2003;98(5):1079-87.

123. Leffler DA, Edwards-George J, Dennis M, Schuppan D, Cook F, Franko DL, et al.

Factors that Influence Adherence to a Gluten-Free Diet in Adults with Celiac Disease.

Dig Dis Sci 2007.

124. Hopper AD, Hadjivassiliou M, Hurlstone DP, Lobo AJ, McAlindon ME, Egner W, et

al. What is the role of serologic testing in celiac disease? A prospective, biopsy-

confirmed study with economic analysis. Clin Gastroenterol Hepatol 2008;6(3):314-

20.

125. Kaukinen K, Sulkanen S, Maki M, Collin P. IgA-class transglutaminase antibodies in

evaluating the efficacy of gluten-free diet in coeliac disease. Eur J Gastroenterol

Hepatol 2002;14(3):311-5.

126. Sategna-Guidetti C, Grosso S, Bruno M, Grosso SB. Reliability of immunologic

markers of celiac sprue in the assessment of mucosal recovery after gluten with-

drawal. J Clin Gastroenterol 1996;23(2):101-4.

127. Lee A, Newman JM. Celiac diet: its impact on quality of life. J Am Diet Assoc

2003;103(11):1533-5.

76

128. Hallert C, Granno C, Grant C, Hulten S, Midhagen G, Strom M, et al. Quality of life

of adult coeliac patients treated for 10 years. Scand J Gastroenterol 1998;33(9):933-

8.

129. Cosnes J, Cellier C, Viola S, Colombel JF, Michaud L, Sarles J, et al. Incidence of

autoimmune diseases in celiac disease: protective effect of the gluten-free diet. Clin

Gastroenterol Hepatol 2008;6(7):753-8.

130. Sategna Guidetti C, Solerio E, Scaglione N, Aimo G, Mengozzi G. Duration of gluten

exposure in adult coeliac disease does not correlate with the risk for autoimmune

disorders. Gut 2001;49(4):502-5.

131. Viljamaa M, Kaukinen K, Huhtala H, Kyronpalo S, Rasmussen M, Collin P. Coeliac

disease, autoimmune diseases and gluten exposure. Scand J Gastroenterol

2005;40(4):437-43.

132. Holmes GK, Prior P, Lane MR, Pope D, Allan RN. Malignancy in coeliac disease--

effect of a gluten free diet. Gut 1989;30(3):333-8.

133. Corrao G, Corazza GR, Bagnardi V, Brusco G, Ciacci C, Cottone M, et al. Mortality

in patients with coeliac disease and their relatives: a cohort study. Lancet

2001;358(9279):356-61.

134. Lewis HM, Renaula TL, Garioch JJ, Leonard JN, Fry JS, Collin P, et al. Protective

effect of gluten-free diet against development of lymphoma in dermatitis herpeti-

formis. Br J Dermatol 1996;135(3):363-7.

135. Collin P, Reunala T, Pukkala E, Laippala P, Keyrilainen O, Pasternack A. Coeliac

disease--associated disorders and survival. Gut 1994;35(9):1215-8.

136. Green PH, Fleischauer AT, Bhagat G, Goyal R, Jabri B, Neugut AI. Risk of malig-

nancy in patients with celiac disease. Am J Med 2003;115(3):191-5.

137. Silano M, Volta U, Vincenzi AD, Dessi M, Vincenzi MD. Effect of a Gluten-free Diet

on the Risk of Enteropathy-associated T-cell Lymphoma in Celiac Disease. Dig Dis

Sci 2007.

138. Hervonen K, Vornanen M, Kautiainen H, Collin P, Reunala T. Lymphoma in pa-

tients with dermatitis herpetiformis and their first-degree relatives. Br J Dermatol

2005;152(1):82-6.

139. Olen O, Askling J, Ludvigsson JF, Hildebrand H, Ekbom A, Ekström Smedby K. Ce-

liac Disease, compliance to a gluten free diet and risk of lymphoma by subtype - a

nested case-control study. Manuscript 2008.

140. Kaukinen K, Halme L, Collin P, Farkkila M, Maki M, Vehmanen P, et al. Celiac dis-

ease in patients with severe liver disease: gluten-free diet may reverse hepatic failure.

Gastroenterology 2002;122(4):881-8.

141. Ludvigsson JF, Ludvigsson J, Ekbom A, Montgomery SM. Celiac Disease and Risk of

Subsequent Type 1 Diabetes: A general population cohort study of children and ado-

lescents. Diabetes Care 2006;29(11):2483-8.

77

142. Cronin CC, Shanahan F. Insulin-dependent diabetes mellitus and coeliac disease.

Lancet 1997;349(9058):1096-7.

143. Iughetti L, Bulgarelli S, Forese S, Lorini R, Balli F, Bernasconi S. Endocrine aspects of

coeliac disease. J Pediatr Endocrinol Metab 2003;16(6):805-18.

144. Larsson K, Carlsson A, Cederwall E, Jonsson B, Neiderud J, Jonsson B, et al. Annual

screening detects celiac disease in children with type 1 diabetes. Pediatr Diabetes

2008;9(4 Pt 2):354-9.

145. Kaukinen K, Peraaho M, Lindfors K, Partanen J, Woolley N, Pikkarainen P, et al.

Persistent small bowel mucosal villous atrophy without symptoms in coeliac disease.

Aliment Pharmacol Ther 2007;25(10):1237-45.

146. Trier JS. Celiac sprue. N Engl J Med 1991;325(24):1709-19.

147. Reunala TL. Dermatitis herpetiformis. Clin Dermatol 2001;19(6):728-36.

148. Askling J, Linet M, Gridley G, Halstensen TS, Ekstrom K, Ekbom A. Cancer inci-

dence in a population-based cohort of individuals hospitalized with celiac disease or

dermatitis herpetiformis. Gastroenterology 2002;123(5):1428-35.

149. West J, Logan RF, Smith CJ, Hubbard RB, Card TR. Malignancy and mortality in

people with coeliac disease: population based cohort study. Bmj

2004;329(7468):716-9.

150. Catassi C, Fabiani E, Corrao G, Barbato M, De Renzo A, Carella AM, et al. Risk of

non-Hodgkin lymphoma in celiac disease. Jama 2002;287(11):1413-9.

151. Silano M, Volta U, Mecchia AM, Dessi M, Di Benedetto R, De Vincenzi M. Delayed

diagnosis of coeliac disease increases cancer risk. BMC Gastroenterol 2007;7:8.

152. Smedby KE, Akerman M, Hildebrand H, Glimelius B, Ekbom A, Askling J. Malig-

nant lymphomas in coeliac disease: evidence of increased risks for lymphoma types

other than enteropathy-type T cell lymphoma. Gut 2005;54(1):54-9.

153. Card TR, West J, Holmes GK. Risk of malignancy in diagnosed coeliac disease: a 24-

year prospective, population-based, cohort study. Aliment Pharmacol Ther

2004;20(7):769-75.

154. Mearin ML, Catassi C, Brousse N, Brand R, Collin P, Fabiani E, et al. European

multi-centre study on coeliac disease and non-Hodgkin lymphoma. Eur J Gastroen-

terol Hepatol 2006;18(2):187-94.

155. Goldacre MJ, Wotton CJ, Yeates D, Seagroatt V, Jewell D. Cancer in patients with

ulcerative colitis, Crohn's disease and coeliac disease: record linkage study. Eur J

Gastroenterol Hepatol 2008;20(4):297-304.

156. Gao Y, Kristinsson SY, Goldin LR, Bjorkholm M, Caporaso NE, Landgren O. In-

creased Risk for Non-Hodgkin Lymphoma in Individuals with Celiac Disease and a

Potential Familial Association. Gastroenterology 2008.

78

157. Peters U, Askling J, Gridley G, Ekbom A, Linet M. Causes of death in patients with

celiac disease in a population-based Swedish cohort. Arch Intern Med

2003;163(13):1566-72.

158. Collin P, Salmi J, Hallstrom O, Reunala T, Pasternack A. Autoimmune thyroid dis-

orders and coeliac disease. Eur J Endocrinol 1994;130(2):137-40.

159. Counsell CE, Taha A, Ruddell WS. Coeliac disease and autoimmune thyroid disease.

Gut 1994;35(6):844-6.

160. Sategna-Guidetti C, Bruno M, Mazza E, Carlino A, Predebon S, Tagliabue M, et al.

Autoimmune thyroid diseases and coeliac disease. Eur J Gastroenterol Hepatol

1998;10(11):927-31.

161. Cuoco L, Certo M, Jorizzo RA, De Vitis I, Tursi A, Papa A, et al. Prevalence and

early diagnosis of coeliac disease in autoimmune thyroid disorders. Ital J Gastroen-

terol Hepatol 1999;31(4):283-7.

162. Kaukinen K, Collin P, Mykkanen AH, Partanen J, Maki M, Salmi J. Celiac disease

and autoimmune endocrinologic disorders. Dig Dis Sci 1999;44(7):1428-33.

163. Ventura A, Neri E, Ughi C, Leopaldi A, Citta A, Not T. Gluten-dependent diabetes-

related and thyroid-related autoantibodies in patients with celiac disease. J Pediatr

2000;137(2):263-5.

164. Berti I, Trevisiol C, Tommasini A, Citta A, Neri E, Geatti O, et al. Usefulness of

screening program for celiac disease in autoimmune thyroiditis. Dig Dis Sci

2000;45(2):403-6.

165. Volta U, Ravaglia G, Granito A, Forti P, Maioli F, Petrolini N, et al. Coeliac disease

in patients with autoimmune thyroiditis. Digestion 2001;64(1):61-5.

166. Larizza D, Calcaterra V, De Giacomo C, De Silvestri A, Asti M, Badulli C, et al. Ce-

liac disease in children with autoimmune thyroid disease. J Pediatr 2001;139(5):738-

40.

167. Hakanen M, Luotola K, Salmi J, Laippala P, Kaukinen K, Collin P. Clinical and sub-

clinical autoimmune thyroid disease in adult celiac disease. Dig Dis Sci

2001;46(12):2631-5.

168. Sategna-Guidetti C, Volta U, Ciacci C, Usai P, Carlino A, De Franceschi L, et al.

Prevalence of thyroid disorders in untreated adult celiac disease patients and effect of

gluten withdrawal: an Italian multicenter study. Am J Gastroenterol 2001;96(3):751-

7.

169. Ansaldi N, Palmas T, Corrias A, Barbato M, D'Altiglia MR, Campanozzi A, et al.

Autoimmune thyroid disease and celiac disease in children. J Pediatr Gastroenterol

Nutr 2003;37(1):63-6.

170. Hadithi M, de Boer H, Meijer JW, Willekens F, Kerckhaert JA, Heijmans R, et al.

Coeliac disease in Dutch patients with Hashimoto's thyroiditis and vice versa. World

J Gastroenterol 2007;13(11):1715-22.

79

171. Spadaccino AC, Basso D, Chiarelli S, Albergoni MP, D'Odorico A, Plebani M, et al.

Celiac disease in North Italian patients with autoimmune thyroid diseases. Autoim-

munity 2008;41(1):116-21.

172. O'Leary C, Walsh CH, Wieneke P, O'Regan P, Buckley B, O'Halloran DJ, et al.

Coeliac disease and autoimmune Addison's disease: a clinical pitfall. Qjm

2002;95(2):79-82.

173. Myhre AG, Aarsetoy H, Undlien DE, Hovdenak N, Aksnes L, Husebye ES. High fre-

quency of coeliac disease among patients with autoimmune adrenocortical failure.

Scand J Gastroenterol 2003;38(5):511-5.

174. Biagi F, Campanella J, Soriani A, Vailati A, Corazza GR. Prevalence of coeliac dis-

ease in Italian patients affected by Addison's disease. Scand J Gastroenterol

2006;41(3):302-5.

175. Betterle C, Lazzarotto F, Spadaccino AC, Basso D, Plebani M, Pedini B, et al. Celiac

disease in North Italian patients with autoimmune Addison's disease. Eur J Endocri-

nol 2006;154(2):275-9.

176. Stagi S, Giani T, Simonini G, Falcini F. Thyroid function, autoimmune thyroiditis

and coeliac disease in juvenile idiopathic arthritis. Rheumatology (Oxford)

2005;44(4):517-20.

177. Kent L, McBride R, McConnell R, Neugut AI, Bhagat G, Green PH. Increased risk of

papillary thyroid cancer in celiac disease. Dig Dis Sci 2006;51(10):1875-7.

178. Logan RF, Ferguson A, Finlayson ND, Weir DG. Primary biliary cirrhosis and

coeliac disease: an association? Lancet 1978;1(8058):230-3.

179. Sorensen HT, Thulstrup AM, Blomqvist P, Norgaard B, Fonager K, Ekbom A. Risk

of primary biliary liver cirrhosis in patients with coeliac disease: Danish and Swedish

cohort data. Gut 1999;44(5):736-8.

180. Lawson A, West J, Aithal GP, Logan RF. Autoimmune cholestatic liver disease in

people with coeliac disease: a population-based study of their association. Aliment

Pharmacol Ther 2005;21(4):401-5.

181. Bushara KO. Neurologic presentation of celiac disease. Gastroenterology 2005;128(4

Suppl 1):S92-7.

182. Volta U, Rodrigo L, Granito A, Petrolini N, Muratori P, Muratori L, et al. Celiac

disease in autoimmune cholestatic liver disorders. Am J Gastroenterol

2002;97(10):2609-13.

183. Volta U, De Franceschi L, Molinaro N, Cassani F, Muratori L, Lenzi M, et al. Fre-

quency and significance of anti-gliadin and anti-endomysial antibodies in autoim-

mune hepatitis. Dig Dis Sci 1998;43(10):2190-5.

184. Villalta D, Girolami D, Bidoli E, Bizzaro N, Tampoia M, Liguori M, et al. High

prevalence of celiac disease in autoimmune hepatitis detected by anti-tissue tranglu-

taminase autoantibodies. J Clin Lab Anal 2005;19(1):6-10.

80

185. Ludvigsson JF, Reutfors J, Osby U, Ekbom A, Montgomery SM. Coeliac disease and

risk of mood disorders--a general population-based cohort study. J Affect Disord

2007;99(1-3):117-26.

186. Stenberg R, Schollin J. Is there a connection between severe cerebral palsy and in-

creased gluten sensitivity? Acta Paediatr 2007;96(1):132-4.

187. Korponay-Szabo IR, Dahlbom I, Laurila K, Koskinen S, Woolley N, Partanen J, et al.

Elevation of IgG antibodies against tissue transglutaminase as a diagnostic tool for

coeliac disease in selective IgA deficiency. Gut 2003;52(11):1567-71.

188. Stenson WF, Newberry R, Lorenz R, Baldus C, Civitelli R. Increased prevalence of

celiac disease and need for routine screening among patients with osteoporosis. Arch

Intern Med 2005;165(4):393-9.

189. Ludvigsson JF, Michaelsson K, Ekbom A, Montgomery SM. Coeliac disease and the

risk of fractures - a general population-based cohort study. Aliment Pharmacol Ther

2007;25(3):273-85.

190. West J, Logan RF, Card TR, Smith C, Hubbard R. Fracture risk in people with celiac

disease: a population-based cohort study. Gastroenterology 2003;125(2):429-36.

191. Collin P, Vilska S, Heinonen PK, Hallstrom O, Pikkarainen P. Infertility and coeliac

disease. Gut 1996;39(3):382-4.

192. Ludvigsson JF, Montgomery SM, Ekbom A. Celiac disease and risk of adverse fetal

outcome: a population-based cohort study. Gastroenterology 2005;129(2):454-63.

193. Meloni GF, Dessole S, Vargiu N, Tomasi PA, Musumeci S. The prevalence of coeliac

disease in infertility. Hum Reprod 1999;14(11):2759-61.

194. Cohen WI. Current dilemmas in Down syndrome clinical care: celiac disease, thyroid

disorders, and atlanto-axial instability. Am J Med Genet C Semin Med Genet

2006;142C(3):141-8.

195. Ludvigsson JF, Wahlstrom J, Grunewald J, Ekbom A, Montgomery SM. Coeliac dis-

ease and risk of tuberculosis: a population based cohort study. Thorax

2007;62(1):23-8.

196. Pastore L, Carroccio A, Compilato D, Panzarella V, Serpico R, Lo Muzio L. Oral

manifestations of celiac disease. J Clin Gastroenterol 2008;42(3):224-32.

197. Ludvigsson JF, Olen O, Bell M, Ekbom A, Montgomery SM. Coeliac disease and risk

of sepsis. Gut 2008;57(8):1074-80.

198. Bardella MT, Valenti L, Pagliari C, Peracchi M, Fare M, Fracanzani AL, et al.

Searching for coeliac disease in patients with non-alcoholic fatty liver disease. Dig Li-

ver Dis 2004;36(5):333-6.

199. Bardella MT, Vecchi M, Conte D, Del Ninno E, Fraquelli M, Pacchetti S, et al.

Chronic unexplained hypertransaminasemia may be caused by occult celiac disease.

Hepatology 1999;29(3):654-7.

81

200. Hagander B, Berg NO, Brandt L, Norden A, Sjolund K, Stenstam M. Hepatic injury

in adult coeliac disease. Lancet 1977;2(8032):270-2.

201. Gillberg R, Dotevall G, Ahren C. Chronic inflammatory bowel disease in patients

with coeliac disease. Scand J Gastroenterol 1982;17(4):491-6.

202. Hay JE, Wiesner RH, Shorter RG, LaRusso NF, Baldus WP. Primary sclerosing

cholangitis and celiac disease. A novel association. Ann Intern Med

1988;109(9):713-7.

203. Al-Osaimi AM, Berg CL. Association of primary sclerosing cholangitis and celiac

disease: a case report and review of the literature. Dig Dis Sci 2004;49(3):438-43.

204. Frustaci A, Cuoco L, Chimenti C, Pieroni M, Fioravanti G, Gentiloni N, et al. Celiac

disease associated with autoimmune myocarditis. Circulation 2002;105(22):2611-8.

205. Fonager K, Sorensen HT, Norgard B, Thulstrup AM. Cardiomyopathy in Danish

patients with coeliac disease. Lancet 1999;354(9189):1561.

206. Curione M, Barbato M, De Biase L, Viola F, Lo Russo L, Cardi E. Prevalence of

coeliac disease in idiopathic dilated cardiomyopathy. Lancet 1999;354(9174):222-3.

207. Chimenti C, Pieroni M, Frustaci A. Celiac disease in idiopathic dilated cardiomyopa-

thy. Ital Heart J 2001;2(9):658-9.

208. Prati D, Bardella MT, Peracchi M, Porretti L, Cardillo M, Pagliari C, et al. High fre-

quency of anti-endomysial reactivity in candidates to heart transplant. Dig Liver Dis

2002;34(1):39-43.

209. Not T, Faleschini E, Tommasini A, Repetto A, Pasotti M, Baldas V, et al. Celiac dis-

ease in patients with sporadic and inherited cardiomyopathies and in their relatives.

Eur Heart J 2003;24(15):1455-61.

210. De Bem RS, Da Ro Sa Utiyama SR, Nisihara RM, Fortunato JA, Tondo JA, Carmes

ER, et al. Celiac disease prevalence in brazilian dilated cardiomyopathy patients. Dig

Dis Sci 2006;51(5):1016-9.

211. Dawes PT, Atherton ST. Coeliac disease presenting as recurrent pericarditis. Lancet

1981;1(8228):1021-2.

212. Faizallah R, Costello FC, Lee FI, Walker R. Adult celiac disease and recurrent peri-

carditis. Dig Dis Sci 1982;27(8):728-30.

213. Laine LA, Holt KM. Recurrent pericarditis and celiac disease. Jama

1984;252(22):3168.

214. Ludvigsson JF, de Faire U, Ekbom A, Montgomery SM. Vascular disease in a popula-

tion-based cohort of individuals hospitalised with coeliac disease. Heart

2007;93(9):1111-5.

215. Ekstrom Smedby K, Vajdic CM, Falster M, Engels EA, Martinez-Maza O, Turner J,

et al. Autoimmune disorders and risk of non-Hodgkin lymphoma subtypes: a pooled

analysis within the InterLymph Consortium. Blood 2008;111(8):4029-38.

82

216. Lohi S, Maki M, Montonen J, Knekt P, Pukkala E, Reunanen A, et al. Malignancies

in cases with screening-identified evidence of coeliac disease: a long-term population-

based cohort study. Gut 2008.

217. Askling J, Fored CM, Baecklund E, Brandt L, Backlin C, Ekbom A, et al. Haemato-

poietic malignancies in rheumatoid arthritis: lymphoma risk and characteristics after

exposure to tumour necrosis factor antagonists. Ann Rheum Dis 2005;64(10):1414-

20.

218. Anderson LA, McMillan SA, Watson RG, Monaghan P, Gavin AT, Fox C, et al. Ma-

lignancy and mortality in a population-based cohort of patients with coeliac disease

or "gluten sensitivity". World J Gastroenterol 2007;13(1):146-51.

219. Calltorp J, Adami HO, Astrom H, Fryklund L, Rossner S, Trolle Y, et al. Country

profile: Sweden. Lancet 1996;347(9001):587-94.

220. Lunde AS, Lundeborg S, Lettenstrom GS, Thygesen L, Huebner J. The person-

number systems of Sweden, Norway, Denmark, and Israel. Vital Health Stat 2

1980;2(84):1-59.

221. Johannesson I. Total population register of Statistics Sweden. New possibilities and

better quality. Örebro: Statistics Sweden; 2005 [cited 2008 2 Sept]. Available from:

http:www.

222. Epidemiology TCo. The National Patient Register. Stockholm: The National Board

of Health and Welfare 2008 [cited 2008 31 Oct]. Available from:

http://www.socialstyrelsen.se/en/Statistics/statsbysubject/The+Swedish+Hospital+Disc

harge+Register.htm.

223. Nilsson AC, Spetz CL, Carsjo K, Nightingale R, Smedby B. [Reliability of the hospi-

tal registry. The diagnostic data are better than their reputation]. Lakartidningen

1994;91(7):598, 603-5.

224. Cancer Incidence in Sweden 2005. Stockholm: The National Board of Health and

Welfare 2007 [cited 2008 31 Oct]. Available from:

http://www.socialstyrelsen.se/NR/rdonlyres/FD7B695E-A55F-41F4-B539-

4C63D8199601/7192/2007423.pdf.

225. Bland M. An Introduction to Medical Statistics. 3 ed. Oxford: Oxford University

Press; 2000.

226. Fletcher R, Fletcher S. Clinical epidemiology - The Essentials. 4th ed ed. Philadelphia:

Lippincott Williams and Wilkins; 2005.

227. Rothman KJ, Greenland S, Lash T. Modern Epidemiology. 3rd edition ed. Philadel-

phia: Lippincott Williams and Wilkins; 2008.

228. Snook JA, Dwyer L, Lee-Elliott C, Khan S, Wheeler DW, Nicholas DS. Adult coeliac

disease and cigarette smoking [see comments]. Gut 1996;39(1):60-2.

229. Ludvigsson JF, Montgomery SM, Ekbom A. Smoking and celiac disease: a popula-

tion-based cohort study. Clin Gastroenterol Hepatol 2005;3(9):869-74.

83

230. Ojetti V, Fini L, Zileri Dal Verme L, Migneco A, Pola P, Gasbarrini A. Acute crypto-

genic liver failure in an untreated coeliac patient: a case report. Eur J Gastroenterol

Hepatol 2005;17(10):1119-21.

231. Saarinen S, Olerup O, Broome U. Increased frequency of autoimmune diseases in pa-

tients with primary sclerosing cholangitis. Am J Gastroenterol 2000;95(11):3195-9.

232. Lahdeaho ML, Kaukinen K, Collin P, Ruuska T, Partanen J, Haapala AM, et al. Ce-

liac disease: from inflammation to atrophy: a long-term follow-up study. J Pediatr

Gastroenterol Nutr 2005;41(1):44-8.

233. Brown I, Mino-Kenudson M, Deshpande V, Lauwers GY. Intraepithelial lymphocy-

tosis in architecturally preserved proximal small intestinal mucosa: an increasing di-

agnostic problem with a wide differential diagnosis. Arch Pathol Lab Med

2006;130(7):1020-5.

234. Chang F, Mahadeva U, Deere H. Pathological and clinical significance of increased

intraepithelial lymphocytes (IELs) in small bowel mucosa. Apmis 2005;113(6):385-

99.

235. Mahadeva S, Wyatt JI, Howdle PD. Is a raised intraepithelial lymphocyte count with

normal duodenal villous architecture clinically relevant? J Clin Pathol

2002;55(6):424-8.

236. Bosetti C, Levi F, Ferlay J, Lucchini F, Negri E, La Vecchia C. Incidence and mortal-

ity from non-Hodgkin lymphoma in Europe: the end of an epidemic? Int J Cancer

2008;123(8):1917-23.

237. Elfström P, Montgomery SM, Kämpe O, Ekbom A, Ludvigsson JF. Risk of Primary

Adrenal Insufficiency in Patients with Celiac Disease. J Clin Endocrin Metabolism;

2007;92(9):3595-98.

238. Elfström P, Montgomery SM, Kämpe O, Ekbom A, Ludvigsson JF. Risk of Thyroid

disease in Individuals with Celiac Disease. J Clin Endocrin Metabolism;

2008;93(10):3915-21.

Publikationer i serien Örebro Studies in Medicine

1. Bergemalm, Per-Olof (2004). Audiologic and cognitive long-term sequelae from closed head injury.

2. Jansson, Kjell (2004). Intraperitoneal Microdialysis. Technique and Results.

3. Windahl, Torgny (2004). Clinical aspects of laser treatment of lichen sclerosus and squamous cell carcinoma of the penis.

4. Carlsson, Per-Inge (2004). Hearing impairment and deafness. Genetic and environmental factors – interactions – consequences. A clinical audiological approach.

5. Wågsäter, Dick (2005). CXCL16 and CD137 in Atherosclerosis.

6. Jatta, Ken (2006). Inflammation in Atherosclerosis.

7. Dreifaldt, Ann Charlotte (2006). Epidemiological Aspects on Malignant Diseases in Childhood.

8. Jurstrand, Margaretha (2006). Detection of Chlamydia trachomatis and Mycoplasma genitalium by genetic and serological methods.

9. Norén, Torbjörn (2006). Clostridium difficile, epidemiology and antibiotic resistance.

10. Anderzén Carlsson, Agneta (2007). Children with Cancer – Focusing on their Fear and on how their Fear is Handled.

11. Ocaya, Pauline (2007). Retinoid metabolism and signalling in vascular smooth muscle cells.

12. Nilsson, Andreas (2008). Physical activity assessed by accelerometry in children.

13. Eliasson, Henrik (2008). Tularemia – epidemiological, clinical and diagnostic aspects.

14. Walldén, Jakob (2008). The influence of opioids on gastric function: experimental and clinical studies.

15. Andrén, Ove (2008). Natural history and prognostic factors in localized prostate cancer.

16. Svantesson, Mia (2008). Postpone death? Nurse-physician perspectives and ethics rounds.

17. Björk, Tabita (2008). Measuring Eating Disorder Outcome – Definitions, dropouts and patients’ perspectives.

18. Ahlsson, Anders (2008). Atrial Fibrillation in Cardiac Surgery.

19. Parihar, Vishal Singh (2008). Human Listeriosis – Sources and Routes.

20. Berglund, Carolina (2008). Molecular Epidemiology of Methicillin-Resistant Staphylococcus aureus. Epidemiological aspects of MRSA and the dissemination in the community and in hospitals.

21. Nilsagård, Ylva (2008). Walking ability, balance and accidental falls in persons with Multiple Sclerosis.

22. Johansson, Ann-Christin (2008). Psychosocial factors in patients with lumbar disc herniation: Enhancing postoperative outcome by the identification of predictive factors and optimised physiotherapy.

23. Larsson, Matz (2008). Secondary exposure to inhaled tobacco products.

24. Hahn-Strömberg, Victoria (2008). Cell adhesion proteins in dif-ferent invasive patterns of colon carcinoma: A morphometric and molecular genetic study.

25. Böttiger, Anna (2008). Genetic Variation in the Folate Receptor- and Methylenetetrahydrofolate Reductase Genes as Determinants of Plasma Homocysteine Concentrations.

26. Andersson, Gunnel (2009). Urinary incontinence. Prevalence, treatment seeking behaviour, experiences and perceptions among persons with and without urinary leakage.

27. Elfström, Peter (2009). Associated disorders in celiac disease.