Dig Dis Sci DOI 10.1007/s10620-014-3046-1
ORIGINAL ARTICLE
Children with Celiac Disease Are More Likely to Have Attended Hospital for Prior Respiratory Syncytial Virus Infection Anna Ro¨ckert Tjernberg • Jonas F. Ludvigsson
Received: 7 November 2013 / Accepted: 20 January 2014 Ó Springer Science+Business Media New York 2014
Abstract Background and Aim The purpose of this study was to examine the association between celiac disease (CD) and prior respiratory syncytial virus (RSV) infection or any viral bronchiolitis. Methods This was a retrospective case–control study. During 2006–2008 small intestinal biopsy data were collected from Sweden’s 28 pathology departments. We identified 3,835 children diagnosed with CD (villous atrophy, Marsh stage 3) before the age of 2 years in 1987 or later. Using conditional logistic regression we calculated odds ratios (ORs) for having a prior diagnosis of respiratory syncytial virus or other viral bronchiolitis compared to 19,102 age- and sex-matched controls. Results Of the 3,835 children with CD, 36 (0.9 %) had a prior diagnosis of RSV compared to 117/19,102 (0.6 %) matched controls. This corresponded to an OR of 1.46 (95 % CI 1.03–2.07). ORs were similar in girls and boys. The highest ORs were seen in children developing early CD (before 1 year of age (OR 1.82; 95 % CI 0.91–3.62). Prior record of any type of viral bronchiolitis was found in 3.4 % (132/3,835) of individuals with CD and in 2.0 %
A. R. Tjernberg (&) Department of Pediatrics, Kalmar County Hospital, 391 85 Kalmar, Sweden e-mail: [email protected] J. F. Ludvigsson ¨ rebro University Hospital, O ¨ rebro, Department of Pediatrics, O Sweden e-mail: [email protected] J. F. Ludvigsson Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, 171 77 Stockholm, Sweden
(390/19,102) of the matched controls corresponding to an OR of 1.60 (95 % CI 1.33–1.92). Conclusions Children with CD diagnosed\2 years of age were more likely to have attended hospital for a prior RSV infection or any viral bronchiolitis than other children. Keywords Autoimmunity Bronchiolitis Child Coeliac Infant Inflammation Respiratory syncytial virus Abbreviations CD Celiac disease CI Confidence interval OR Odds ratio RSV Respiratory syncytial virus VA Villous atrophy
Introduction Celiac disease (CD) is a chronic small intestinal immunemediated enteropathy precipitated by exposure to dietary gluten, which is a water unsoluble protein complex occurring in wheat, rye and barley but not in oats [1]. This disease has the characteristics of an autoimmune condition and tissue transglutaminase has been identified as its autoantigen [2]. CD occurs in 1–3 % of the western population, and its clinical presentation varies considerably. Children below 2 years of age often have intestinal symptoms and signs including failure to thrive, while older children and adults often have a more vague presentation [1]. Earlier data have shown that individuals with CD have an increased risk of respiratory tract infections such as influenza [3], invasive pneumococcal infections [4] and tuberculosis [5]. This could indicate increased susceptibility to these infections possibly due to altered mucosal
123
Dig Dis Sci
permeability [6]. Previous studies have also suggested that infections may contribute to the development of CD [7], and thus could be a trigger for autoimmune disease. One possible mechanism involves the involvement of toll-like receptor 3, which seems to activate tissue transglutaminase 2 (TG2) [8]. Respiratory syncytial virus (RSV) is one of the leading causes of morbidity and mortality in infants [9–11]. RSV is a paramyxovirus which can cause both upper and lower respiratory tract infections including severe bronchiolitis [9]. RSV has been linked to the development of both diabetic-specific autoantibodies and to manifest type 1 diabetes mellitus [12]. Meanwhile CD shares several genetic traits with type 1 diabetes mellitus [13], but data on RSV and CD are lacking. The aim of this study was to examine the association between viral bronchiolitis, especially RSV infection, and later CD.
Methods Through small intestinal biopsy reports from Sweden’s 28 pathology departments we identified children with villous atrophy (VA, equal to CD in this study). We then linked biopsy data to inpatient and hospital-based outpatient data on RSV and any viral bronchiolitis through the unique Personal Identity Number assigned to all Swedish residents [14]. RSV In this study, RSV was defined as having one of the following international classifications of disease (ICD) codes in the Swedish Patient Register (ICD9: 480B; and ICD10: J20.5, J21.0, B97.4, and J12.1). The Swedish Patient Register started in 1964 and initially only contained inpatient care. It became nationwide in 1987 (when the present study started), adding hospital-based outpatient visits in 2001 [15]. Viral Bronchiolitis The following ICD codes defined viral bronchiolitis (ICD9: 466; and ICD10: J20 and J21). Data were obtained from inpatient and hospital-based outpatient care through the Swedish Patient Register. In this paper ‘‘viral bronchiolitis’’ represents both bronchitis and bronchiolitis since the two terms are often used interchangeably, and it may be difficult to distinguish one from the other. Celiac Disease Through computerized biopsy reports obtained from all 28 Swedish pathology departments, we were able to identify patients with VA (histopathology stage Marsh 3) [16].
123
Biopsies from the duodenum/jejunum had been performed from 1969 through 2008, but we collected the data in the years 2006–2008. Local IT personnel collected all data, and then sent the following items to us the researchers: personal identity number, morphology (for a list of relevant morphology codes according to the Swedish SnoMed system, see ‘‘Appendix’’), topography (duodenum or jejunum), and date of biopsy. We did not have access to CD serology in patients with VA, but a prior validation has shown that 88 % of individuals with available data at time of biopsy were positive for CD serology [16]. A prior patient chart validation confirmed the celiac diagnosis in 108/114 randomly selected individuals with VA, indicating a positive predictive value for CD of 95 % [16]. Biopsy reports were on average based on three tissue specimens (and this should rule in about 95 % of all CD [17]) [18]. As part of the original data collection we also identified individuals with a small intestinal biopsy not showing VA. We divided these individuals into those with (a) inflammation but no VA and (b) those with normal mucosa. For the purpose of our initial studies on CD and risk of complications [19], only individuals with normal mucosa and positive CD serology (here defined as positive IgA/IgG TG2, EMA or gliadin antibodies) were kept in the dataset [20]. Detailed reviews of our biopsy data-collection [16, 18] (including the matching of CD serology data to individuals with normal mucosa [18] ) have been published elsewhere. After removal of duplicates and erroneous data we initially identified 29,096 individuals with CD, 13,306 with inflammation, and 3,719 with normal mucosa but positive CD serology. Controls We matched each individual with CD (and individuals with other small intestinal histopathology) with up to five controls for age, sex, county, and calendar year. Controls were randomly selected among individuals in the Swedish Total Population Register who had no prior record of small intestinal biopsy. Inclusion Criteria for Study Participants For the purpose of this study we restricted our data to individuals undergoing biopsy before the age of 2 years, since we were interested in the association between RSV and neartime CD. The vast majority of individuals in Sweden have their first RSV infection early in life. We chose the age limit of 2 years since it has been used in a number of earlier celiac studies and some suggest that risk factors for early CD may differ from that of CD with a later clinical onset [20]. Some 4,589 individuals with CD had been diagnosed before the age of 2 years (identical to that of our recent study on seasonal influenza and CD) [21]. Since RSV was first introduced as a
Dig Dis Sci Table 1 Characteristics of study participants Characteristic
Matched controls (%)
Celiac disease (%)
Total
19,102
3,835
Girls (%)
12,035 (63.0)
2,418 (63.1)
Boys (%)
7,067 (37.0)
1,417 (36.9)
a
Age \1 year (%)
4,640 (24.3)
935 (24.4)
14,462 (75.7)
2,900 (75.6)
1987–1992
4,941 (25.9)
990 (25.8)
1993–1996
8,203 (42.9)
1,647 (42.9)
1997–
5,958 (31.2)
1,198 (31.2)
Born in Nordic country
19,005 (99.5)
3,817 (99.5)
Prior RSV (%)
117 (0.6)
36 (0.9)
1 to \2 years (%) Calendar yeara
RSV respiratory syncytial virus a
At time of celiac disease diagnosis
diagnosis in the Swedish ICD system in 1987 we restricted our dataset to individuals born after January 1, 1987 so that they could potentially obtain a diagnosis of RSV from the first day of life (up until the date of biopsy, or corresponding date in matched controls). This left us with 3,835 children with CD (19,102 matched controls), 97 with inflammation (487 controls), and 280 children with normal mucosa but positive CD serology (1,397 controls). Statistics Conditional logistic regression was used to calculate odds ratios (ORs) for prior RSV infection, and in separate analyses also for viral bronchiolitis, in individuals with CD (and in separate analysis for individuals with other small intestinal histopathology). Our conditional approach (each individual with CD was first compared to his/her matched controls within the same stratum and then a summary estimate was calculated) minimizes the influence of age, sex, county and calendar year on our ORs. A priori sub-analyses included sex-, age- and calendar period-specific analyses. Age at CD diagnosis was divided into \1 year versus 1 to\2 years. The calendar periods were set up to mirror the so-called Swedish epidemic [22] since time-specific risk factors (such as nutritional recommendations) may have had a substantial effect over time. Our study period was divided into two high-risk periods (1987–1992 and 1993–1996) and one low-risk period (1997–2008). The study participants were fairly equally distributed in the three groups (Table 1). Furthermore we analysed the risk of CD within the first year after RSV infection, versus within 1–2 years after RSV infection. In a sub-analysis we excluded individuals born outside the Nordic countries. We also adjusted for education according to four a priori defined categories [23]. Similar analyses were carried out for prior viral bronchiolitis. Finally
we examined the association between RSV and later CD after adjustment for intrauterine growth retardation, cesarean section and maternal smoking since these variables have either been linked to the risk of CD [24] or RSV [9, 25] or may influence health seeking behaviour (e.g. maternal smoking). Data on these perinatal conditions were retrieved from the Medical Birth Registry [26] and available in 3,393 individuals with CD and 16,703 controls. We did not adjust for preterm birth since that has not been proven to be a risk factor for CD and therefore is no confounder [24]. We used SPSS 20 (SPSS, Inc. Chicago, IL, USA) for all analyses. ORs with 95 % confidence intervals that did not include one were regarded as statistically significant. Ethics The present study was approved by the Ethics Review board of Stockholm, Sweden. No individual informed consent was required since data were strictly registerbased.
Results Background Data Approximately 60 % of the participating children were girls (Table 1). One in four children had been diagnosed with CD in the first year of life. The majority, 70 % of children, were diagnosed with CD during the years of the celiac epidemic (up until 1996) [22]. The mean duration between RSV diagnosis and CD diagnosis was 300 days (duration between RSV and corresponding matching date in reference individuals was 310 days). Main Findings (RSV and CD) Of the 3,835 children with CD, 36 (0.9 %) had a prior diagnosis of RSV compared to 117/19,102 (0.6 %) matched controls. This corresponded to an OR of 1.46 (95 % CI 1.03–2.07). ORs did not change when we adjusted for education or restricted our data to children born in the Nordic countries (data not shown). We found the highest risk of CD within the first year after RSV infection (OR 1.63; 95 % CI 1.08–2.46) and beyond that there was no statistically significant risk increase for CD (OR 1.17; 95 % CI 0.63–2.17). Risk estimates did not differ by sex, age at CD diagnosis or calendar period (Table 2), but cases with RSV were rare in the first calendar period, and only in the last calendar period was the association with later CD statistically significant (OR 1.53; 95 % CI 1.03–2.29).
123
Dig Dis Sci Table 2 RSV and risk of later Celiac disease
Subgroup
RSV, N (%)
OR; 95 % CI
Celiac disease
Controls
Girls
21 (0.9)
56 (0.5)
1.71; 1.09–2.70
Boys
15 (1.1)
61 (0.9)
1.21; 0.70–2.06
P value
Sex
0.328 0.495
Agea
0.564 a
Age \1 year (%)
10 (1.1)
26 (0.6)
1.82; 0.91–3.62
0.089
1–\2 years (%)
26 (0.9)
91 (0.6)
1.38; 0.92–2.09
0.123
9 (0.2)
0 (0)
Not estimated
0.408
Calendar perioda 1987–1992 RSV respiratory syncytial virus a
At time of celiac disease diagnosis
Table 3 Viral bronchiolitis and risk of later Celiac disease
0.331
1993–1996
8 (0.5)
21 (0.3)
1.70; 0.83–3.48
0.148
1997–
28 (2.3)
87 (1.5)
1.53; 1.03–2.29
0.036
OR; 95 % CI
P value
1.54; 1.18–2.01 1.66; 1.29–2.15
0.01 \0.001
Subgroup
Viral bronchiolitis, N (%) Celiac disease
Controls
62 (2.6) 70 (4.9)
190 (1.6) 200 (2.8)
Sex
Agea
0.398 a
Age \1 year (%)
28 (3.0)
67 (1.4)
1.95; 1.28–2.95
0.002
1 to \2 years (%)
104 (3.6)
323 (2.2)
1.54; 1.25–1.90
\0.001
20 (2.0)
49 (1.0)
1.82; 1.15–2.89
0.011
Calendar perioda 1987–1992 At time of celiac disease diagnosis
0.624
1993–1996
41 (2.5)
116 (1.4)
1.63; 1.18–2.25
0.003
1997–
71 (5.9)
225 (3.8)
1.53; 1.18–1.98
0.001
Adjustment for intrauterine growth retardation, cesarean section and maternal smoking did not influence our risk estimates (adjusted OR 1.51; 95 % CI 1.05–2.19). RSV and Non-VA RSV was also associated with having a small intestinal biopsy without VA but without attaining statistical significance. Of 97 individuals with inflammation but no VA, two (2.1 %) had a prior record of RSV compared to 6/487 (1.2 %) matched controls. This corresponded to an OR of 1.60 (95 % CI 0.34–7.55). The OR was even higher for having a normal mucosa but with positive CD serology (OR 4.19; 95 % CI 2.00–8.77), based on 10/280 (3.6 %) prior RSV in those with a biopsy compared to 9/1,397 (0.6 %) of their matched controls. Additional Analyses: CD and Viral Bronchiolitis Some 3.4 % (132/3,835) of individuals with CD and 2.0 % (390/19,102) of matched controls had a prior record of any viral bronchiolitis, corresponding to an OR of 1.60 (95 %
123
P for interaction
0.681
Girls Boys
a
P for interaction
CI 1.33–1.92). Also for viral bronchiolitis the risk of a future CD diagnosis was highest in the first year after infection (OR 1.69; 95 % CI 1.36–2.09), and diminished somewhat thereafter (OR 1.39; 95 % CI 0.99–2.00). Table 3 shows the results of the subgroup analyses.
Discussion In this nationwide case–control study of more than 3,800 children with CD we found a positive association between RSV infection and later CD. We also found an increased risk of CD in children who had a prior episode of any viral bronchiolitis. The etiology of CD is considered multifactorial, and even though the presence of HLA DQ2/DQ8 is a prerequisite for CD, environmental factors such as infections [7, 27, 28] are likely to play an important role [12]. Sandberg-Bennich et al. investigated whether different perinatal factors could influence the risk of later CD, and the strongest risk factor was exposure to neonatal infections (OR 1.52; 95 % CI 1.19–1.95). A prior population-
Dig Dis Sci
based study from Southeast Sweden with prospectively collected data found a non-statistically significant association between infection at the time of gluten introduction and later CD (HR 1.8, 95 % CI 0.9–3.6), but that study only included 44 children with CD whereas the present study includes more than 3,800 children with biopsy-verified CD focused on a common respiratory tract infection rather than on infections in general. Infection with RSV is usually acquired early in life. The global incidence is high and the annual number of new episodes of RSV-associated lower respiratory tract infection in children \5 years has been estimated at more than 30 million [11]. American data suggest an annual RSV infection rate of 68.8/100 in children less than 1 year of age [29], and many children need medical care [10, 11]. There are several environmental and demographical risk factors for severe RSV and RSV hospital admission [25, 30]. We were able to adjust for several such risk factors including maternal smoking, intrauterine growth retardation and cesarean section, and this did not influence the risk estimates. However, we were unable to adjust for breastfeeding duration. Short breastfeeding duration has been linked to the risk of CD [31], although the largest prospective study from Sweden so far found no association between breastfeeding duration and CD [27]. However breastfeeding is believed to exert a protective role if continued during gluten introduction [32]. Whether breastfeeding protects against RSV infection is still debated [30]. Research has indicated that there is a difference in metabolic activity of intestinal microbial flora in children with CD compared to that in healthy controls [33]. Decker et al. [34] reported an association between caesarean section and CD, a result partly confirmed by Ma˚rild et al. [24] who found that elective (but not any) cesarean section was associated with CD. Potentially this could be due to alterations of the gut microbiota, a theory supported by our recent study that found a positive association between antibiotics use and later CD [35]. It is also possible that an early-life infection can alter the intestinal flora and thereby contributes to the development of CD. A few studies have evaluated vitamin-D status in celiac patients, and often levels have been low in untreated CD [36]. Low levels of vitamin D may predispose to respiratory tract infections [37], especially severe infections [38]. Belderbos et al. [39] quite recently suggested that low vitamin D levels in cord blood predisposes to RSV lower respiratory tract infections in the young child. Hence, also low vitamin D levels may link undiagnosed CD and RSV. Finally, we cannot rule out that individuals attending hospital due to RSV infection undergo investigations for a number of disorders, and that such surveillance bias has contributed to the increased OR for later CD. This latter explanation could also explain why children with RSV
infections were also at increased risk of having a later biopsy with normal mucosa but positive celiac serology. Likewise an Italian study showed that anti-transglutaminase antibodies can be produced temporarily during an infection [40] which also could explain the increased OR for RSV in individuals with positive serology but normal mucosa. To our knowledge this is the first study to examine the association between RSV and CD, and one of the largest so far examining any infectious disorder and the risk of later CD. Data on RSV were retrieved from the Swedish Patient Register where all inpatient and hospital-based outpatient care is recorded prospectively. This eliminates the risk of recall bias sometimes seen in retrospective case–control studies. The risk of misclassification of CD is low. We used biopsy data with villous atrophy (VA) from all Swedish pathology departments to identify a representative population of patients with CD. VA is the gold standard for CD diagnosis in Sweden. Our recent validation study in 2009 found that more than 95 % of Swedish gastroenterologists and pediatricians performed a small-intestinal biopsy before CD diagnosis, indicating that biopsy records have a high sensitivity for diagnosed CD [16]. This study has some limitations. We are not aware of any validation study of RSV or bronchiolitis in the Patient Register but generally some 85–95 % of all recorded diagnoses are correct [15]. During the early part of the study period, we only had access to inpatient data on RSV and viral bronchitis and not all children with these disorders are admitted to hospital. However, for most of the last calendar period (2001 and later) we had access to outpatient data as well and the OR during that period (OR 1.53) was very similar to the overall OR (1.46) arguing that inpatient data we€re valid in order to examine the association between RSV and CD. In conclusion this study found an association between RSV infection and later CD. While earlier evidence suggests that exposure to infectious disease predisposes for development of CD in genetically predisposed individuals, the relevance of RSV infection as a specific co-factor for CD development needs to be confirmed by other studies. Acknowledgments ART was funded by Kalmar County Council. JFL was supported by grants from The Swedish Society of Medicine, the Swedish Research Council—Medicine (522-2A09-195), the Swedish Celiac Society, and the Fulbright Commission. Conflict of interest
None.
Ethical standard This project (2006/633-31/4) was approved by the Research Ethics Committee of the Karolinska Institute, Sweden on June 14, 2006.
Appendix See Table 4.
123
Dig Dis Sci Table 4 Small intestinal histopathology classifications— a comparison
KVAST Kvalitets- och Standardiseringskommitte´n (English: Committee for Quality and Standardization) a
We have not included Marsh type 4 in this classification since such lesions are very rare (ref B) and cannot be identified through SnoMed Codes
b
Increased intraepithelial lymphocyte count (often [30/ 100 epithelial cells)
Classification used in this project
Normal
Inflammation
Villous atrophy
Marsh classificationa Marsh description
Type 0
Type 1
Type 2
Type 3a
Preinfiltrative
Infiltrative
Infiltrativehyperplastic
Flat destructive
Corazza et al. (ref A)
–
Grade A
Grade B1
SnoMed Codes
M0010, M0011
M40000, M41000, M42000, M43000, M47000, M47170
M58,
M58,
Grade B2 M58,
D6218,
D6218,
D6218,
M58005
M58006
M58007
KVAST/Alexander classification
Type 3c
I
II
III
IV
IV
Normal
Intraepithelial lymphocytosis (IEL)b
Partial VA
Subtotal VA
Total VA
Villous atrophy
-
-
-
?
??
??
IELb
-
?
?
?
?
?
Crypt hyperplasia
-
-
?
?
??
??
Characteristics
References 1. Ludvigsson JF, Green PH. Clinical management of coeliac disease. J Intern Med. 2011;269:560–571. 2. Dieterich W, Laag E, Schopper H, et al. Autoantibodies to tissue transglutaminase as predictors of celiac disease [see comments]. Gastroenterology. 1998;115:1317–1321. 3. Marild K, Fredlund H, Ludvigsson JF. Increased risk of hospital admission for influenza in patients with celiac disease: a nationwide cohort study in Sweden. Am J Gastroenterol. 2010;105:2465–2473. 4. Thomas HJ, Wotton CJ, Yeates D, Ahmad T, Jewell DP, Goldacre MJ. Pneumococcal infection in patients with coeliac disease. Eur J Gastroenterol Hepatol. 2008;20:624–628. 5. Ludvigsson JF, Sanders DS, Maeurer M, Jonsson J, Grunewald J, Wahlstrom J. Risk of tuberculosis in a large sample of patients with coeliac disease–a nationwide cohort study. Aliment Pharmacol Ther. 2011;33:689–696. 6. Sapone A, Lammers KM, Casolaro V, et al. Divergence of gut permeability and mucosal immune gene expression in two glutenassociated conditions: celiac disease and gluten sensitivity. BMC Med. 2011;9:23. 7. Stene LC, Honeyman MC, Hoffenberg EJ, et al. Rotavirus infection frequency and risk of celiac disease autoimmunity in early childhood: a longitudinal study. Am J Gastroenterol. 2006;101:2333–2340. 8. Jabri B, Sollid LM. Tissue-mediated control of immunopathology in coeliac disease. Nat Rev Immunol. 2009;9:858–870. 9. Tregoning JS, Schwarze J. Respiratory viral infections in infants: causes, clinical symptoms, virology, and immunology. Clin Microbiol Rev. 2010;23:74–98. 10. Hall CB. The burgeoning burden of respiratory syncytial virus among children. Infect Disord Drug Targets. 2012;12:92–97. 11. Nair H, Nokes DJ, Gessner BD, et al. Global burden of acute lower respiratory infections due to respiratory syncytial virus in young children: a systematic review and meta-analysis. Lancet. 2010;375:1545–1555. 12. Gutierrez-Achury J, Coutinho de Almeida R, Wijmenga C. Shared genetics in coeliac disease and other immune-mediated diseases. J Intern Med. 2011;269:591–603. 13. Smyth DJ, Plagnol V, Walker NM, et al. Shared and distinct genetic variants in type 1 diabetes and celiac disease. N Engl J Med. 2008;359:2767–2777.
123
Type 3b
14. Ludvigsson JF, Otterblad-Olausson P, Pettersson BU, Ekbom A. The Swedish personal identity number: possibilities and pitfalls in healthcare and medical research. Eur J Epidemiol. 2009;24:659–667. 15. Ludvigsson JF, Andersson E, Ekbom A, et al. External review and validation of the Swedish national inpatient register. BMC Public Health. 2011;11:450. 16. Ludvigsson JF, Brandt L, Montgomery SM, Granath F, Ekbom A. Validation study of villous atrophy and small intestinal inflammation in Swedish biopsy registers. BMC Gastroenterol. 2009;9:19. 17. Pais WP, Duerksen DR, Pettigrew NM, Bernstein CN. How many duodenal biopsy specimens are required to make a diagnosis of celiac disease? Gastrointest Endosc. 2008;67:1082–1087. 18. Ludvigsson JF, Brandt L, Montgomery SM. Symptoms and signs in individuals with serology positive for celiac disease but normal mucosa. BMC Gastroenterol. 2009;9:57. 19. Ludvigsson JF, Montgomery SM, Ekbom A, Brandt L, Granath F. Small-intestinal histopathology and mortality risk in celiac disease. JAMA. 2009;302:1171–1178. 20. Ivarsson A, Hernell O, Stenlund H, Persson LA. Breast-feeding protects against celiac disease. Am J Clin Nutr. 2002;75:914–921. 21. Lebwohl B, Green PH, Murray JA, Ludvigsson JF. Season of birth in a nationwide cohort of coeliac disease patients. Arch Dis Child. 2013;98:48–51. 22. Ivarsson A, Persson LA, Nystrom L, et al. Epidemic of coeliac disease in Swedish children [see comments]. Acta Paediatr. 2000;89:165–171. 23. Olen O, Bihagen E, Rasmussen F, Ludvigsson JF. Socioeconomic position and education in patients with coeliac disease. Dig Liver Dis. 2012;44:471–476. 24. Marild K, Stephansson O, Montgomery S, Murray JA, Ludvigsson JF. Pregnancy outcome and risk of celiac disease in offspring: a nationwide case-control study. Gastroenterology. 2012;142 (39–45):e33. 25. Nielsen HE, Siersma V, Andersen S, et al. Respiratory syncytial virus infection–risk factors for hospital admission: a case-control study. Acta Paediatr. 2003;92:1314–1321. 26. Cnattingius S, Ericson A, Gunnarskog J, Kallen B. A quality study of a medical birth registry. Scand J Soc Med. 1990;18: 143–148.
Dig Dis Sci 27. Welander A, Tjernberg AR, Montgomery SM, Ludvigsson J, Ludvigsson JF. Infectious disease and risk of later celiac disease in childhood. Pediatrics. 2010;125:e530–e536. 28. Sandberg-Bennich S, Dahlquist G, Kallen B. Coeliac disease is associated with intrauterine growth and neonatal infections. Acta Paediatr. 2002;91:30–33. 29. Glezen WP, Taber LH, Frank AL, Kasel JA. Risk of primary infection and reinfection with respiratory syncytial virus. Am J Dis Child. 1986;140:543–546. 30. Simoes EA. Environmental and demographic risk factors for respiratory syncytial virus lower respiratory tract disease. J Pediatr. 2003;143:S118–S126. 31. Akobeng AK, Ramanan AV, Buchan I, Heller RF. Effect of breast feeding on risk of coeliac disease: a systematic review and meta-analysis of observational studies. Arch Dis Child. 2006;91:39–43. 32. Szajewska H, Chmielewska A, Piescik-Lech M, et al. Systematic review: early infant feeding and the prevention of coeliac disease. Aliment Pharmacol Ther. 2012;36:607–618. 33. Tjellstrom B, Stenhammar L, Hogberg L, et al. Gut microflora associated characteristics in children with celiac disease. Am J Gastroenterol. 2005;100:2784–2788. 34. Decker E, Engelmann G, Findeisen A, et al. Cesarean delivery is associated with celiac disease but not inflammatory bowel disease in children. Pediatrics. 2010;125:e1433–e1440.
35. Marild K, Ye W, Lebwohl B, et al. Antibiotic exposure and the development of coeliac disease: a nationwide case-control study. BMC Gastroenterol. 2013;13:109. 36. Margoni D, Chouliaras G, Duscas G, et al. Bone health in children with celiac disease assessed by dual x-ray absorptiometry: effect of gluten-free diet and predictive value of serum biochemical indices. J Pediatr Gastroenterol Nutr. 2012;54:680–684. 37. Roth DE, Shah R, Black RE, Baqui AH. Vitamin D status and acute lower respiratory infection in early childhood in Sylhet. Bangladesh Acta Paediatr. 2010;99:389–393. 38. McNally JD, Leis K, Matheson LA, Karuananyake C, Sankaran K, Rosenberg AM. Vitamin D deficiency in young children with severe acute lower respiratory infection. Pediatr Pulmonol. 2009;44:981–988. 39. Belderbos ME, Houben ML, Wilbrink B, et al. Cord blood vitamin D deficiency is associated with respiratory syncytial virus bronchiolitis. Pediatrics. 2011;127:e1513–e1520. 40. Ferrara F, Quaglia S, Caputo I, et al. Anti-transglutaminase antibodies in non-coeliac children suffering from infectious diseases. Clin Exp Immunol. 2010;159:217–223.
123
ORIGINAL ARTICLE
Children with Celiac Disease Are More Likely to Have Attended Hospital for Prior Respiratory Syncytial Virus Infection Anna Ro¨ckert Tjernberg • Jonas F. Ludvigsson
Received: 7 November 2013 / Accepted: 20 January 2014 Ó Springer Science+Business Media New York 2014
Abstract Background and Aim The purpose of this study was to examine the association between celiac disease (CD) and prior respiratory syncytial virus (RSV) infection or any viral bronchiolitis. Methods This was a retrospective case–control study. During 2006–2008 small intestinal biopsy data were collected from Sweden’s 28 pathology departments. We identified 3,835 children diagnosed with CD (villous atrophy, Marsh stage 3) before the age of 2 years in 1987 or later. Using conditional logistic regression we calculated odds ratios (ORs) for having a prior diagnosis of respiratory syncytial virus or other viral bronchiolitis compared to 19,102 age- and sex-matched controls. Results Of the 3,835 children with CD, 36 (0.9 %) had a prior diagnosis of RSV compared to 117/19,102 (0.6 %) matched controls. This corresponded to an OR of 1.46 (95 % CI 1.03–2.07). ORs were similar in girls and boys. The highest ORs were seen in children developing early CD (before 1 year of age (OR 1.82; 95 % CI 0.91–3.62). Prior record of any type of viral bronchiolitis was found in 3.4 % (132/3,835) of individuals with CD and in 2.0 %
A. R. Tjernberg (&) Department of Pediatrics, Kalmar County Hospital, 391 85 Kalmar, Sweden e-mail: [email protected] J. F. Ludvigsson ¨ rebro University Hospital, O ¨ rebro, Department of Pediatrics, O Sweden e-mail: [email protected] J. F. Ludvigsson Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, 171 77 Stockholm, Sweden
(390/19,102) of the matched controls corresponding to an OR of 1.60 (95 % CI 1.33–1.92). Conclusions Children with CD diagnosed\2 years of age were more likely to have attended hospital for a prior RSV infection or any viral bronchiolitis than other children. Keywords Autoimmunity Bronchiolitis Child Coeliac Infant Inflammation Respiratory syncytial virus Abbreviations CD Celiac disease CI Confidence interval OR Odds ratio RSV Respiratory syncytial virus VA Villous atrophy
Introduction Celiac disease (CD) is a chronic small intestinal immunemediated enteropathy precipitated by exposure to dietary gluten, which is a water unsoluble protein complex occurring in wheat, rye and barley but not in oats [1]. This disease has the characteristics of an autoimmune condition and tissue transglutaminase has been identified as its autoantigen [2]. CD occurs in 1–3 % of the western population, and its clinical presentation varies considerably. Children below 2 years of age often have intestinal symptoms and signs including failure to thrive, while older children and adults often have a more vague presentation [1]. Earlier data have shown that individuals with CD have an increased risk of respiratory tract infections such as influenza [3], invasive pneumococcal infections [4] and tuberculosis [5]. This could indicate increased susceptibility to these infections possibly due to altered mucosal
123
Dig Dis Sci
permeability [6]. Previous studies have also suggested that infections may contribute to the development of CD [7], and thus could be a trigger for autoimmune disease. One possible mechanism involves the involvement of toll-like receptor 3, which seems to activate tissue transglutaminase 2 (TG2) [8]. Respiratory syncytial virus (RSV) is one of the leading causes of morbidity and mortality in infants [9–11]. RSV is a paramyxovirus which can cause both upper and lower respiratory tract infections including severe bronchiolitis [9]. RSV has been linked to the development of both diabetic-specific autoantibodies and to manifest type 1 diabetes mellitus [12]. Meanwhile CD shares several genetic traits with type 1 diabetes mellitus [13], but data on RSV and CD are lacking. The aim of this study was to examine the association between viral bronchiolitis, especially RSV infection, and later CD.
Methods Through small intestinal biopsy reports from Sweden’s 28 pathology departments we identified children with villous atrophy (VA, equal to CD in this study). We then linked biopsy data to inpatient and hospital-based outpatient data on RSV and any viral bronchiolitis through the unique Personal Identity Number assigned to all Swedish residents [14]. RSV In this study, RSV was defined as having one of the following international classifications of disease (ICD) codes in the Swedish Patient Register (ICD9: 480B; and ICD10: J20.5, J21.0, B97.4, and J12.1). The Swedish Patient Register started in 1964 and initially only contained inpatient care. It became nationwide in 1987 (when the present study started), adding hospital-based outpatient visits in 2001 [15]. Viral Bronchiolitis The following ICD codes defined viral bronchiolitis (ICD9: 466; and ICD10: J20 and J21). Data were obtained from inpatient and hospital-based outpatient care through the Swedish Patient Register. In this paper ‘‘viral bronchiolitis’’ represents both bronchitis and bronchiolitis since the two terms are often used interchangeably, and it may be difficult to distinguish one from the other. Celiac Disease Through computerized biopsy reports obtained from all 28 Swedish pathology departments, we were able to identify patients with VA (histopathology stage Marsh 3) [16].
123
Biopsies from the duodenum/jejunum had been performed from 1969 through 2008, but we collected the data in the years 2006–2008. Local IT personnel collected all data, and then sent the following items to us the researchers: personal identity number, morphology (for a list of relevant morphology codes according to the Swedish SnoMed system, see ‘‘Appendix’’), topography (duodenum or jejunum), and date of biopsy. We did not have access to CD serology in patients with VA, but a prior validation has shown that 88 % of individuals with available data at time of biopsy were positive for CD serology [16]. A prior patient chart validation confirmed the celiac diagnosis in 108/114 randomly selected individuals with VA, indicating a positive predictive value for CD of 95 % [16]. Biopsy reports were on average based on three tissue specimens (and this should rule in about 95 % of all CD [17]) [18]. As part of the original data collection we also identified individuals with a small intestinal biopsy not showing VA. We divided these individuals into those with (a) inflammation but no VA and (b) those with normal mucosa. For the purpose of our initial studies on CD and risk of complications [19], only individuals with normal mucosa and positive CD serology (here defined as positive IgA/IgG TG2, EMA or gliadin antibodies) were kept in the dataset [20]. Detailed reviews of our biopsy data-collection [16, 18] (including the matching of CD serology data to individuals with normal mucosa [18] ) have been published elsewhere. After removal of duplicates and erroneous data we initially identified 29,096 individuals with CD, 13,306 with inflammation, and 3,719 with normal mucosa but positive CD serology. Controls We matched each individual with CD (and individuals with other small intestinal histopathology) with up to five controls for age, sex, county, and calendar year. Controls were randomly selected among individuals in the Swedish Total Population Register who had no prior record of small intestinal biopsy. Inclusion Criteria for Study Participants For the purpose of this study we restricted our data to individuals undergoing biopsy before the age of 2 years, since we were interested in the association between RSV and neartime CD. The vast majority of individuals in Sweden have their first RSV infection early in life. We chose the age limit of 2 years since it has been used in a number of earlier celiac studies and some suggest that risk factors for early CD may differ from that of CD with a later clinical onset [20]. Some 4,589 individuals with CD had been diagnosed before the age of 2 years (identical to that of our recent study on seasonal influenza and CD) [21]. Since RSV was first introduced as a
Dig Dis Sci Table 1 Characteristics of study participants Characteristic
Matched controls (%)
Celiac disease (%)
Total
19,102
3,835
Girls (%)
12,035 (63.0)
2,418 (63.1)
Boys (%)
7,067 (37.0)
1,417 (36.9)
a
Age \1 year (%)
4,640 (24.3)
935 (24.4)
14,462 (75.7)
2,900 (75.6)
1987–1992
4,941 (25.9)
990 (25.8)
1993–1996
8,203 (42.9)
1,647 (42.9)
1997–
5,958 (31.2)
1,198 (31.2)
Born in Nordic country
19,005 (99.5)
3,817 (99.5)
Prior RSV (%)
117 (0.6)
36 (0.9)
1 to \2 years (%) Calendar yeara
RSV respiratory syncytial virus a
At time of celiac disease diagnosis
diagnosis in the Swedish ICD system in 1987 we restricted our dataset to individuals born after January 1, 1987 so that they could potentially obtain a diagnosis of RSV from the first day of life (up until the date of biopsy, or corresponding date in matched controls). This left us with 3,835 children with CD (19,102 matched controls), 97 with inflammation (487 controls), and 280 children with normal mucosa but positive CD serology (1,397 controls). Statistics Conditional logistic regression was used to calculate odds ratios (ORs) for prior RSV infection, and in separate analyses also for viral bronchiolitis, in individuals with CD (and in separate analysis for individuals with other small intestinal histopathology). Our conditional approach (each individual with CD was first compared to his/her matched controls within the same stratum and then a summary estimate was calculated) minimizes the influence of age, sex, county and calendar year on our ORs. A priori sub-analyses included sex-, age- and calendar period-specific analyses. Age at CD diagnosis was divided into \1 year versus 1 to\2 years. The calendar periods were set up to mirror the so-called Swedish epidemic [22] since time-specific risk factors (such as nutritional recommendations) may have had a substantial effect over time. Our study period was divided into two high-risk periods (1987–1992 and 1993–1996) and one low-risk period (1997–2008). The study participants were fairly equally distributed in the three groups (Table 1). Furthermore we analysed the risk of CD within the first year after RSV infection, versus within 1–2 years after RSV infection. In a sub-analysis we excluded individuals born outside the Nordic countries. We also adjusted for education according to four a priori defined categories [23]. Similar analyses were carried out for prior viral bronchiolitis. Finally
we examined the association between RSV and later CD after adjustment for intrauterine growth retardation, cesarean section and maternal smoking since these variables have either been linked to the risk of CD [24] or RSV [9, 25] or may influence health seeking behaviour (e.g. maternal smoking). Data on these perinatal conditions were retrieved from the Medical Birth Registry [26] and available in 3,393 individuals with CD and 16,703 controls. We did not adjust for preterm birth since that has not been proven to be a risk factor for CD and therefore is no confounder [24]. We used SPSS 20 (SPSS, Inc. Chicago, IL, USA) for all analyses. ORs with 95 % confidence intervals that did not include one were regarded as statistically significant. Ethics The present study was approved by the Ethics Review board of Stockholm, Sweden. No individual informed consent was required since data were strictly registerbased.
Results Background Data Approximately 60 % of the participating children were girls (Table 1). One in four children had been diagnosed with CD in the first year of life. The majority, 70 % of children, were diagnosed with CD during the years of the celiac epidemic (up until 1996) [22]. The mean duration between RSV diagnosis and CD diagnosis was 300 days (duration between RSV and corresponding matching date in reference individuals was 310 days). Main Findings (RSV and CD) Of the 3,835 children with CD, 36 (0.9 %) had a prior diagnosis of RSV compared to 117/19,102 (0.6 %) matched controls. This corresponded to an OR of 1.46 (95 % CI 1.03–2.07). ORs did not change when we adjusted for education or restricted our data to children born in the Nordic countries (data not shown). We found the highest risk of CD within the first year after RSV infection (OR 1.63; 95 % CI 1.08–2.46) and beyond that there was no statistically significant risk increase for CD (OR 1.17; 95 % CI 0.63–2.17). Risk estimates did not differ by sex, age at CD diagnosis or calendar period (Table 2), but cases with RSV were rare in the first calendar period, and only in the last calendar period was the association with later CD statistically significant (OR 1.53; 95 % CI 1.03–2.29).
123
Dig Dis Sci Table 2 RSV and risk of later Celiac disease
Subgroup
RSV, N (%)
OR; 95 % CI
Celiac disease
Controls
Girls
21 (0.9)
56 (0.5)
1.71; 1.09–2.70
Boys
15 (1.1)
61 (0.9)
1.21; 0.70–2.06
P value
Sex
0.328 0.495
Agea
0.564 a
Age \1 year (%)
10 (1.1)
26 (0.6)
1.82; 0.91–3.62
0.089
1–\2 years (%)
26 (0.9)
91 (0.6)
1.38; 0.92–2.09
0.123
9 (0.2)
0 (0)
Not estimated
0.408
Calendar perioda 1987–1992 RSV respiratory syncytial virus a
At time of celiac disease diagnosis
Table 3 Viral bronchiolitis and risk of later Celiac disease
0.331
1993–1996
8 (0.5)
21 (0.3)
1.70; 0.83–3.48
0.148
1997–
28 (2.3)
87 (1.5)
1.53; 1.03–2.29
0.036
OR; 95 % CI
P value
1.54; 1.18–2.01 1.66; 1.29–2.15
0.01 \0.001
Subgroup
Viral bronchiolitis, N (%) Celiac disease
Controls
62 (2.6) 70 (4.9)
190 (1.6) 200 (2.8)
Sex
Agea
0.398 a
Age \1 year (%)
28 (3.0)
67 (1.4)
1.95; 1.28–2.95
0.002
1 to \2 years (%)
104 (3.6)
323 (2.2)
1.54; 1.25–1.90
\0.001
20 (2.0)
49 (1.0)
1.82; 1.15–2.89
0.011
Calendar perioda 1987–1992 At time of celiac disease diagnosis
0.624
1993–1996
41 (2.5)
116 (1.4)
1.63; 1.18–2.25
0.003
1997–
71 (5.9)
225 (3.8)
1.53; 1.18–1.98
0.001
Adjustment for intrauterine growth retardation, cesarean section and maternal smoking did not influence our risk estimates (adjusted OR 1.51; 95 % CI 1.05–2.19). RSV and Non-VA RSV was also associated with having a small intestinal biopsy without VA but without attaining statistical significance. Of 97 individuals with inflammation but no VA, two (2.1 %) had a prior record of RSV compared to 6/487 (1.2 %) matched controls. This corresponded to an OR of 1.60 (95 % CI 0.34–7.55). The OR was even higher for having a normal mucosa but with positive CD serology (OR 4.19; 95 % CI 2.00–8.77), based on 10/280 (3.6 %) prior RSV in those with a biopsy compared to 9/1,397 (0.6 %) of their matched controls. Additional Analyses: CD and Viral Bronchiolitis Some 3.4 % (132/3,835) of individuals with CD and 2.0 % (390/19,102) of matched controls had a prior record of any viral bronchiolitis, corresponding to an OR of 1.60 (95 %
123
P for interaction
0.681
Girls Boys
a
P for interaction
CI 1.33–1.92). Also for viral bronchiolitis the risk of a future CD diagnosis was highest in the first year after infection (OR 1.69; 95 % CI 1.36–2.09), and diminished somewhat thereafter (OR 1.39; 95 % CI 0.99–2.00). Table 3 shows the results of the subgroup analyses.
Discussion In this nationwide case–control study of more than 3,800 children with CD we found a positive association between RSV infection and later CD. We also found an increased risk of CD in children who had a prior episode of any viral bronchiolitis. The etiology of CD is considered multifactorial, and even though the presence of HLA DQ2/DQ8 is a prerequisite for CD, environmental factors such as infections [7, 27, 28] are likely to play an important role [12]. Sandberg-Bennich et al. investigated whether different perinatal factors could influence the risk of later CD, and the strongest risk factor was exposure to neonatal infections (OR 1.52; 95 % CI 1.19–1.95). A prior population-
Dig Dis Sci
based study from Southeast Sweden with prospectively collected data found a non-statistically significant association between infection at the time of gluten introduction and later CD (HR 1.8, 95 % CI 0.9–3.6), but that study only included 44 children with CD whereas the present study includes more than 3,800 children with biopsy-verified CD focused on a common respiratory tract infection rather than on infections in general. Infection with RSV is usually acquired early in life. The global incidence is high and the annual number of new episodes of RSV-associated lower respiratory tract infection in children \5 years has been estimated at more than 30 million [11]. American data suggest an annual RSV infection rate of 68.8/100 in children less than 1 year of age [29], and many children need medical care [10, 11]. There are several environmental and demographical risk factors for severe RSV and RSV hospital admission [25, 30]. We were able to adjust for several such risk factors including maternal smoking, intrauterine growth retardation and cesarean section, and this did not influence the risk estimates. However, we were unable to adjust for breastfeeding duration. Short breastfeeding duration has been linked to the risk of CD [31], although the largest prospective study from Sweden so far found no association between breastfeeding duration and CD [27]. However breastfeeding is believed to exert a protective role if continued during gluten introduction [32]. Whether breastfeeding protects against RSV infection is still debated [30]. Research has indicated that there is a difference in metabolic activity of intestinal microbial flora in children with CD compared to that in healthy controls [33]. Decker et al. [34] reported an association between caesarean section and CD, a result partly confirmed by Ma˚rild et al. [24] who found that elective (but not any) cesarean section was associated with CD. Potentially this could be due to alterations of the gut microbiota, a theory supported by our recent study that found a positive association between antibiotics use and later CD [35]. It is also possible that an early-life infection can alter the intestinal flora and thereby contributes to the development of CD. A few studies have evaluated vitamin-D status in celiac patients, and often levels have been low in untreated CD [36]. Low levels of vitamin D may predispose to respiratory tract infections [37], especially severe infections [38]. Belderbos et al. [39] quite recently suggested that low vitamin D levels in cord blood predisposes to RSV lower respiratory tract infections in the young child. Hence, also low vitamin D levels may link undiagnosed CD and RSV. Finally, we cannot rule out that individuals attending hospital due to RSV infection undergo investigations for a number of disorders, and that such surveillance bias has contributed to the increased OR for later CD. This latter explanation could also explain why children with RSV
infections were also at increased risk of having a later biopsy with normal mucosa but positive celiac serology. Likewise an Italian study showed that anti-transglutaminase antibodies can be produced temporarily during an infection [40] which also could explain the increased OR for RSV in individuals with positive serology but normal mucosa. To our knowledge this is the first study to examine the association between RSV and CD, and one of the largest so far examining any infectious disorder and the risk of later CD. Data on RSV were retrieved from the Swedish Patient Register where all inpatient and hospital-based outpatient care is recorded prospectively. This eliminates the risk of recall bias sometimes seen in retrospective case–control studies. The risk of misclassification of CD is low. We used biopsy data with villous atrophy (VA) from all Swedish pathology departments to identify a representative population of patients with CD. VA is the gold standard for CD diagnosis in Sweden. Our recent validation study in 2009 found that more than 95 % of Swedish gastroenterologists and pediatricians performed a small-intestinal biopsy before CD diagnosis, indicating that biopsy records have a high sensitivity for diagnosed CD [16]. This study has some limitations. We are not aware of any validation study of RSV or bronchiolitis in the Patient Register but generally some 85–95 % of all recorded diagnoses are correct [15]. During the early part of the study period, we only had access to inpatient data on RSV and viral bronchitis and not all children with these disorders are admitted to hospital. However, for most of the last calendar period (2001 and later) we had access to outpatient data as well and the OR during that period (OR 1.53) was very similar to the overall OR (1.46) arguing that inpatient data we€re valid in order to examine the association between RSV and CD. In conclusion this study found an association between RSV infection and later CD. While earlier evidence suggests that exposure to infectious disease predisposes for development of CD in genetically predisposed individuals, the relevance of RSV infection as a specific co-factor for CD development needs to be confirmed by other studies. Acknowledgments ART was funded by Kalmar County Council. JFL was supported by grants from The Swedish Society of Medicine, the Swedish Research Council—Medicine (522-2A09-195), the Swedish Celiac Society, and the Fulbright Commission. Conflict of interest
None.
Ethical standard This project (2006/633-31/4) was approved by the Research Ethics Committee of the Karolinska Institute, Sweden on June 14, 2006.
Appendix See Table 4.
123
Dig Dis Sci Table 4 Small intestinal histopathology classifications— a comparison
KVAST Kvalitets- och Standardiseringskommitte´n (English: Committee for Quality and Standardization) a
We have not included Marsh type 4 in this classification since such lesions are very rare (ref B) and cannot be identified through SnoMed Codes
b
Increased intraepithelial lymphocyte count (often [30/ 100 epithelial cells)
Classification used in this project
Normal
Inflammation
Villous atrophy
Marsh classificationa Marsh description
Type 0
Type 1
Type 2
Type 3a
Preinfiltrative
Infiltrative
Infiltrativehyperplastic
Flat destructive
Corazza et al. (ref A)
–
Grade A
Grade B1
SnoMed Codes
M0010, M0011
M40000, M41000, M42000, M43000, M47000, M47170
M58,
M58,
Grade B2 M58,
D6218,
D6218,
D6218,
M58005
M58006
M58007
KVAST/Alexander classification
Type 3c
I
II
III
IV
IV
Normal
Intraepithelial lymphocytosis (IEL)b
Partial VA
Subtotal VA
Total VA
Villous atrophy
-
-
-
?
??
??
IELb
-
?
?
?
?
?
Crypt hyperplasia
-
-
?
?
??
??
Characteristics
References 1. Ludvigsson JF, Green PH. Clinical management of coeliac disease. J Intern Med. 2011;269:560–571. 2. Dieterich W, Laag E, Schopper H, et al. Autoantibodies to tissue transglutaminase as predictors of celiac disease [see comments]. Gastroenterology. 1998;115:1317–1321. 3. Marild K, Fredlund H, Ludvigsson JF. Increased risk of hospital admission for influenza in patients with celiac disease: a nationwide cohort study in Sweden. Am J Gastroenterol. 2010;105:2465–2473. 4. Thomas HJ, Wotton CJ, Yeates D, Ahmad T, Jewell DP, Goldacre MJ. Pneumococcal infection in patients with coeliac disease. Eur J Gastroenterol Hepatol. 2008;20:624–628. 5. Ludvigsson JF, Sanders DS, Maeurer M, Jonsson J, Grunewald J, Wahlstrom J. Risk of tuberculosis in a large sample of patients with coeliac disease–a nationwide cohort study. Aliment Pharmacol Ther. 2011;33:689–696. 6. Sapone A, Lammers KM, Casolaro V, et al. Divergence of gut permeability and mucosal immune gene expression in two glutenassociated conditions: celiac disease and gluten sensitivity. BMC Med. 2011;9:23. 7. Stene LC, Honeyman MC, Hoffenberg EJ, et al. Rotavirus infection frequency and risk of celiac disease autoimmunity in early childhood: a longitudinal study. Am J Gastroenterol. 2006;101:2333–2340. 8. Jabri B, Sollid LM. Tissue-mediated control of immunopathology in coeliac disease. Nat Rev Immunol. 2009;9:858–870. 9. Tregoning JS, Schwarze J. Respiratory viral infections in infants: causes, clinical symptoms, virology, and immunology. Clin Microbiol Rev. 2010;23:74–98. 10. Hall CB. The burgeoning burden of respiratory syncytial virus among children. Infect Disord Drug Targets. 2012;12:92–97. 11. Nair H, Nokes DJ, Gessner BD, et al. Global burden of acute lower respiratory infections due to respiratory syncytial virus in young children: a systematic review and meta-analysis. Lancet. 2010;375:1545–1555. 12. Gutierrez-Achury J, Coutinho de Almeida R, Wijmenga C. Shared genetics in coeliac disease and other immune-mediated diseases. J Intern Med. 2011;269:591–603. 13. Smyth DJ, Plagnol V, Walker NM, et al. Shared and distinct genetic variants in type 1 diabetes and celiac disease. N Engl J Med. 2008;359:2767–2777.
123
Type 3b
14. Ludvigsson JF, Otterblad-Olausson P, Pettersson BU, Ekbom A. The Swedish personal identity number: possibilities and pitfalls in healthcare and medical research. Eur J Epidemiol. 2009;24:659–667. 15. Ludvigsson JF, Andersson E, Ekbom A, et al. External review and validation of the Swedish national inpatient register. BMC Public Health. 2011;11:450. 16. Ludvigsson JF, Brandt L, Montgomery SM, Granath F, Ekbom A. Validation study of villous atrophy and small intestinal inflammation in Swedish biopsy registers. BMC Gastroenterol. 2009;9:19. 17. Pais WP, Duerksen DR, Pettigrew NM, Bernstein CN. How many duodenal biopsy specimens are required to make a diagnosis of celiac disease? Gastrointest Endosc. 2008;67:1082–1087. 18. Ludvigsson JF, Brandt L, Montgomery SM. Symptoms and signs in individuals with serology positive for celiac disease but normal mucosa. BMC Gastroenterol. 2009;9:57. 19. Ludvigsson JF, Montgomery SM, Ekbom A, Brandt L, Granath F. Small-intestinal histopathology and mortality risk in celiac disease. JAMA. 2009;302:1171–1178. 20. Ivarsson A, Hernell O, Stenlund H, Persson LA. Breast-feeding protects against celiac disease. Am J Clin Nutr. 2002;75:914–921. 21. Lebwohl B, Green PH, Murray JA, Ludvigsson JF. Season of birth in a nationwide cohort of coeliac disease patients. Arch Dis Child. 2013;98:48–51. 22. Ivarsson A, Persson LA, Nystrom L, et al. Epidemic of coeliac disease in Swedish children [see comments]. Acta Paediatr. 2000;89:165–171. 23. Olen O, Bihagen E, Rasmussen F, Ludvigsson JF. Socioeconomic position and education in patients with coeliac disease. Dig Liver Dis. 2012;44:471–476. 24. Marild K, Stephansson O, Montgomery S, Murray JA, Ludvigsson JF. Pregnancy outcome and risk of celiac disease in offspring: a nationwide case-control study. Gastroenterology. 2012;142 (39–45):e33. 25. Nielsen HE, Siersma V, Andersen S, et al. Respiratory syncytial virus infection–risk factors for hospital admission: a case-control study. Acta Paediatr. 2003;92:1314–1321. 26. Cnattingius S, Ericson A, Gunnarskog J, Kallen B. A quality study of a medical birth registry. Scand J Soc Med. 1990;18: 143–148.
Dig Dis Sci 27. Welander A, Tjernberg AR, Montgomery SM, Ludvigsson J, Ludvigsson JF. Infectious disease and risk of later celiac disease in childhood. Pediatrics. 2010;125:e530–e536. 28. Sandberg-Bennich S, Dahlquist G, Kallen B. Coeliac disease is associated with intrauterine growth and neonatal infections. Acta Paediatr. 2002;91:30–33. 29. Glezen WP, Taber LH, Frank AL, Kasel JA. Risk of primary infection and reinfection with respiratory syncytial virus. Am J Dis Child. 1986;140:543–546. 30. Simoes EA. Environmental and demographic risk factors for respiratory syncytial virus lower respiratory tract disease. J Pediatr. 2003;143:S118–S126. 31. Akobeng AK, Ramanan AV, Buchan I, Heller RF. Effect of breast feeding on risk of coeliac disease: a systematic review and meta-analysis of observational studies. Arch Dis Child. 2006;91:39–43. 32. Szajewska H, Chmielewska A, Piescik-Lech M, et al. Systematic review: early infant feeding and the prevention of coeliac disease. Aliment Pharmacol Ther. 2012;36:607–618. 33. Tjellstrom B, Stenhammar L, Hogberg L, et al. Gut microflora associated characteristics in children with celiac disease. Am J Gastroenterol. 2005;100:2784–2788. 34. Decker E, Engelmann G, Findeisen A, et al. Cesarean delivery is associated with celiac disease but not inflammatory bowel disease in children. Pediatrics. 2010;125:e1433–e1440.
35. Marild K, Ye W, Lebwohl B, et al. Antibiotic exposure and the development of coeliac disease: a nationwide case-control study. BMC Gastroenterol. 2013;13:109. 36. Margoni D, Chouliaras G, Duscas G, et al. Bone health in children with celiac disease assessed by dual x-ray absorptiometry: effect of gluten-free diet and predictive value of serum biochemical indices. J Pediatr Gastroenterol Nutr. 2012;54:680–684. 37. Roth DE, Shah R, Black RE, Baqui AH. Vitamin D status and acute lower respiratory infection in early childhood in Sylhet. Bangladesh Acta Paediatr. 2010;99:389–393. 38. McNally JD, Leis K, Matheson LA, Karuananyake C, Sankaran K, Rosenberg AM. Vitamin D deficiency in young children with severe acute lower respiratory infection. Pediatr Pulmonol. 2009;44:981–988. 39. Belderbos ME, Houben ML, Wilbrink B, et al. Cord blood vitamin D deficiency is associated with respiratory syncytial virus bronchiolitis. Pediatrics. 2011;127:e1513–e1520. 40. Ferrara F, Quaglia S, Caputo I, et al. Anti-transglutaminase antibodies in non-coeliac children suffering from infectious diseases. Clin Exp Immunol. 2010;159:217–223.
123
