The Unfolding Story of Celiac Disease Risk Factors
2013; Elsevier BV; Volume: 12; Issue: 4 Linguagem: Inglês
10.1016/j.cgh.2013.10.031
ISSN1542-7714
AutoresBenjamin Lebwohl, Jonas F. Ludvigsson, Peter H.R. Green,
Tópico(s)Eosinophilic Esophagitis
ResumoWhat causes celiac disease (CD)? On the surface, the answer is straightforward: dietary gluten. In susceptible individuals, ingested gluten is deamidated by tissue transglutaminase in the lamina propria of the small intestine and then bound by antigen-presenting cells to the HLA DQ2 or DQ8. This sets off an immune cascade that results in infiltration of the epithelium with lymphocytes and tissue destruction, leading to villous atrophy that is characteristic of CD.1Kupfer S.S. Jabri B. Pathophysiology of celiac disease.Gastrointest Endosc Clin N Am. 2012; 22: 639-660Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar The identification of dietary gluten as the environmental culprit was a major advance in the history of gastroenterology. Although diet was long considered to trigger or exacerbate CD, it was the Dutch pediatrician Willem Dicke who determined it was the protein fraction of wheat that made his patients ill. These observations were made before World War II, and Dicke's hypothesis was borne out during a blockade-induced famine in 1944, when patients with CD improved markedly during the period of severe wheat shortage, only to relapse on the reintroduction of wheat at the end of the famine.2Dicke W.K. Weijers H.A. Van De Kamer J.H. Coeliac disease. II. The presence in wheat of a factor having a deleterious effect in cases of coeliac disease.Acta Paediatr. 1953; 42: 34-42Crossref PubMed Scopus (310) Google Scholar The subsequent discovery that the HLA DQ2 or DQ8 haplotype was necessary for the development of CD3Sollid L.M. Lie B.A. Celiac disease genetics: current concepts and practical applications.Clin Gastroenterol Hepatol. 2005; 3: 843-851Abstract Full Text Full Text PDF PubMed Scopus (184) Google Scholar led to the present understanding that CD arises when gluten is introduced to the genetically susceptible individual. Yet this simple explanation is belied by the fact that this gene–environment combination is far more common than the prevalence of CD; gluten is a ubiquitous dietary staple, and the at-risk HLA haplotypes are present in 30%-40% of Western populations.4Kagnoff M.F. Celiac disease: pathogenesis of a model immunogenetic disease.J Clin Invest. 2007; 117: 41-49Crossref PubMed Scopus (297) Google Scholar Genome-wide association studies have identified dozens of additional genetic risk loci relating to the immune response, illustrating that CD is a complex, polygenic, immune-based disorder.5van Heel D.A. Franke L. Hunt K.A. et al.A genome-wide association study for celiac disease identifies risk variants in the region harboring IL2 and IL21.Nat Genet. 2007; 39: 827-829Crossref PubMed Scopus (535) Google Scholar, 6Trynka G. Hunt K.A. Bockett N.A. et al.Dense genotyping identifies and localizes multiple common and rare variant association signals in celiac disease.Nat Genet. 2011; 43: 1193-1201Crossref PubMed Scopus (531) Google Scholar But just as the genetic story of CD is more complicated than HLA inheritance, the environmental trigger of CD is about more than gluten. Ultimately, growing knowledge of the genetic determinants of CD will by itself not be adequate to understand why CD develops. Epidemics are triggered by environmental exposures, because genetic changes are too slow to drive these phenomena. We have now witnessed 2 epidemics of CD: one that was dramatic and limited, and another that, although less visible, is greater in scale and ongoing. The Swedish epidemic of CD of 1985–1994 has been extensively documented, and resulted in the development of hypotheses regarding environmental risk factors for this disorder.7Ivarsson A. Persson L.A. Nystrom L. et al.Epidemic of coeliac disease in Swedish children.Acta Paediatr. 2000; 89: 165-171Crossref PubMed Scopus (351) Google Scholar This epidemic was restricted to children younger than 2 years; in that age group, the incidence of diagnosed CD rose from 65 cases per 100,000 person-years to 198 cases per 100,000 person-years. In contrast, incidence data for older children were relatively flat during this period. The epidemic abruptly ended in 1995, although children born during the period of the epidemic have an ongoing increased risk of developing CD. Subsequent investigation led to the hypothesis that infant feeding practices affect the risk of CD in young children. The epidemic occurred during a period of relatively low rates of breastfeeding at the age of 6 months and during the same period of time, the quantity of gluten in infant formula greatly increased. Although it is difficult to separate the relative importance of each feeding practice, it seemed that high quantity of initial gluten intake without overlapping with breastfeeding was responsible for this epidemic. Although a systematic review of the issue has concluded that breastfeeding has not been definitively proved to be associated with risk of CD,8Szajewska H. Chmielewska A. Piescik–Lech M. et al.Systematic review: early infant feeding and the prevention of coeliac disease.Aliment Pharmacol Ther. 2012; 36: 607-618Crossref PubMed Scopus (87) Google Scholar subsequent research has indicated that the timing of gluten introduction is important in determining risk.9Norris J.M. Barriga K. Hoffenberg E.J. et al.Risk of celiac disease autoimmunity and timing of gluten introduction in the diet of infants at increased risk of disease.JAMA. 2005; 293: 2343-2351Crossref PubMed Scopus (361) Google Scholar PreventCD, a prospective randomized trial of infants with a family history of CD, is testing specifically whether the introduction of small quantities of gluten beginning at age 4 months of age will induce tolerance to gluten in this high-risk group.10Hogen Esch C.E. Rosen A. Auricchio R. et al.The PreventCD Study design: towards new strategies for the prevention of coeliac disease.Eur J Gastroenterol Hepatol. 2010; 22: 1424-1430PubMed Google Scholar The second epidemic is more diffusely spread over time and space. Studies from the United States and elsewhere have shown that the seroprevalence of CD (as defined by positive tissue transglutaminase and endomysial antibodies) has increased markedly in recent decades. An analysis of stored serum from military recruits at the Warren Air Force Base in the years spanning 1948–1954 found a CD seroprevalence of 0.2%, whereas 2 recent cohorts from Olmsted County (spanning the years 2006–2008) matched by year of birth and age at sampling found a seroprevalence of 0.9% and 0.8%, respectively.11Rubio–Tapia A. Kyle R.A. Kaplan E.L. et al.Increased prevalence and mortality in undiagnosed celiac disease.Gastroenterology. 2009; 137: 88-93Abstract Full Text Full Text PDF PubMed Scopus (562) Google Scholar An analysis of another cohort in this country found a doubling in seroprevalence during adulthood from 1974 (0.21%) to 1989 (0.45%).12Catassi C. Kryszak D. Bhatti B. et al.Natural history of celiac disease autoimmunity in a USA cohort followed since 1974.Ann Med. 2010; 42: 530-538Crossref PubMed Scopus (278) Google Scholar The mode of presentation of CD has changed in the past generation, with rising numbers of patients presenting without diarrhea.13Lo W. Sano K. Lebwohl B. et al.Changing presentation of adult celiac disease.Dig Dis Sci. 2003; 48: 395-398Crossref PubMed Scopus (207) Google Scholar Patients presenting with anemia may have more severe disease expression (as measured by the degree of villous atrophy and the presence of metabolic bone disease) than patients presenting with diarrhea.14Abu Daya H. Lebwohl B. Lewis S.K. et al.Celiac disease patients presenting with anemia have more severe disease than those presenting with diarrhea.Clin Gastroenterol Hepatol. 2013; 11: 1472-1477Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar Because most individuals in the United States with CD are undiagnosed,15Rubio–Tapia A. Ludvigsson J.F. Brantner T.L. et al.The prevalence of celiac disease in the United States.Am J Gastroenterol. 2012; 107: 1538-1544Crossref PubMed Scopus (518) Google Scholar this is largely a hidden epidemic, but there is no reason to believe that the prevalence has peaked. In Finland, which had a higher prevalence of CD than the United States to begin with, the seroprevalence of CD doubled between the years 1978 (1.05%) and 2000 (1.99%).16Lohi S. Mustalahti K. Kaukinen K. et al.Increasing prevalence of coeliac disease over time.Aliment Pharmacol Ther. 2007; 26: 1217-1225Crossref PubMed Scopus (569) Google Scholar It is not known whether this epidemic will subside or if the prevalence of CD will continue to rise to a new set-point. But given the morbidity associated with CD17Green P.H. Cellier C. Celiac disease.N Engl J Med. 2007; 357: 1731-1743Crossref PubMed Scopus (1470) Google Scholar and the cost and difficulty of the gluten-free diet18Lee A.R. Ng D.L. Zivin J. et al.Economic burden of a gluten-free diet.J Hum Nutr Diet. 2007; 20: 423-430Crossref PubMed Scopus (215) Google Scholar, 19Dorn S.D. Hernandez L. Minaya M.T. et al.The development and validation of a new coeliac disease quality of life survey (CD-QOL).Aliment Pharmacol Ther. 2010; 31: 666-675Crossref PubMed Scopus (86) Google Scholar these data have sparked interest in identifying the cause of this less visible epidemic. Certain infections (eg, rotavirus among infants20Stene L.C. Honeyman M.C. Hoffenberg E.J. et al.Rotavirus infection frequency and risk of celiac disease autoimmunity in early childhood: a longitudinal study.Am J Gastroenterol. 2006; 101: 2333-2340Crossref PubMed Scopus (385) Google Scholar and Campylobacter among adults21Riddle M.S. Murray J.A. Cash B.D. et al.Pathogen-specific risk of celiac disease following bacterial causes of foodborne illness: a retrospective cohort study.Dig Dis Sci. 2013; 58: 3242-3245Crossref PubMed Scopus (36) Google Scholar) have recently been shown to be associated with a increased risk of CD, but rates of these infections have not increased markedly, and so are not likely to be driving this epidemic. In contrast, a lack of exposure to certain microbes is a hallmark of modern times, and the increase in CD disease is congruent with the hygiene hypothesis, which states that decreased exposure to microbes may be driving the rise in autoimmune and atopic conditions. This hypothesis is particularly compelling in light of a recent study that found a dramatically different seroprevalence of CD in Finland (1.4%) and the Russian Karelia (0.6%), geographically proximate areas with a similar prevalence of HLA DQ2 and DQ8 but with major differences in economic development.22Kondrashova A. Mustalahti K. Kaukinen K. et al.Lower economic status and inferior hygienic environment may protect against celiac disease.Ann Med. 2008; 40: 223-231Crossref PubMed Scopus (109) Google Scholar Further evidence implicating the modern relationship with microbes is now accumulating. Children who were born by elective cesarean section are at increased risk of developing CD, whereas those born by emergent cesarean section (and may have had contact with the birth canal) are not.23Marild K. Stephansson O. Montgomery S. et al.Pregnancy outcome and risk of celiac disease in offspring: a nationwide case-control study.Gastroenterology. 2012; 142: 39-45.e3Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar In addition, there seems to be an inverse relationship between Helicobacter pylori colonization and CD.24Lebwohl B. Blaser M.J. Ludvigsson J.F. et al.Decreased risk of celiac disease in patients with Helicobacter pylori colonization.Am J Epidemiol. 2013; 178: 1721-1730Crossref PubMed Scopus (96) Google Scholar Drugs are another modern innovation that may be affecting the CD epidemic. Population-based studies from Sweden have shown that prescription of antibiotics25Marild K. Ye W. Lebwohl B. et al.Antibiotic exposure and the development of coeliac disease: a nationwide case-control study.BMC Gastroenterol. 2013; 13: 109Crossref PubMed Scopus (115) Google Scholar and proton pump inhibitors26Lebwohl B. Spechler S.J. Wang T.C. et al.Use of proton pump inhibitors and subsequent risk of celiac disease.Dig Liver Dis. 2014; 46: 36-40Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar are each associated with an increased risk of the subsequent development of CD. The associations identified in these studies are not necessarily causal. Studies of drug exposure in particular may be prone to protopathic bias, wherein early symptoms of the outcome of interest (CD) may lead to the prescription of the exposure (eg, antibiotics or proton pump inhibitors).27Horwitz R.I. Feinstein A.R. The problem of "protopathic bias" in case-control studies.Am J Med. 1980; 68: 255-258Abstract Full Text PDF PubMed Scopus (160) Google Scholar Confounding variables, such as another (unmeasured) microbe that correlates positively with H pylori and is protective against CD, would create a correlation of this exposure and outcome without direct causation. Although methodologic measures can be taken to address bias and confounding, ultimately these studies should be viewed as hypothesis-generating, mandating investigation to test the potential mechanisms by which exposures inherent to modern life may be contributing to this immune-based disorder. What about other medications, such as supplements? In this month's issue of Clinical Gastroenterology and Hepatology, Størdal and colleagues report on an association between maternal iron supplementation among pregnant women and the risk of the subsequent diagnosis of CD in their offspring.28Størdal K. Haugen M. Brantsæter A.L. et al.Association between maternal iron supplementation during pregnancy and risk of celiac disease in children.Clin Gastroenterol Hepatol. 2014; 12: 624-631Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar The authors found that the use of iron, whether as sole supplements or in combination in a multivitamin/mineral product, is associated with an increased risk of CD (odds ratio, 1.48), which remained significant after adjusting for maternal CD, gender of the child, age of the child, and the timing of gluten introduction. This is a methodologically rigorous study and the findings are intriguing. Strengths include its prospective design, in which exposure was measured before outcome; the large cohort size; long follow-up period; and the use of validated food frequency questionnaires to measure dietary iron. A significant limitation is the reliance on a combination of parental reporting and nonvalidated method of using the Norwegian Patient Register for case identification. To compensate for this, the investigators performed a sensitivity analysis in which 10% and 20% of cases were reclassified as control subjects, and found that the association between iron supplementation and CD persisted. This is reassuring, provided that misclassification caused by reliance on claims codes was random, and not differential. Although health-conscious behavior may be associated with iron supplement use and a CD diagnosis, the association between exposure and outcome was seen only in iron supplements and not for other dietary supplements. The finding of a dose-response relationship between estimated cumulative iron supplementation and the subsequent risk of CD also argues for causality. The mechanism by which iron supplementation could increase the risk of CD is uncertain, but there are a few promising leads. Proteins related to iron metabolism including hepcidin and lactoferrin influence innate immunity,29Johnson E.E. Wessling–Resnick M. Iron metabolism and the innate immune response to infection.Microbes Infect. 2012; 14: 207-216Crossref PubMed Scopus (165) Google Scholar and the host's iron status may influence the makeup of the microbiome.30Nairz M. Schroll A. Sonnweber T. et al.The struggle for iron: a metal at the host-pathogen interface.Cell Microbiol. 2010; 12: 1691-1702Crossref PubMed Scopus (270) Google Scholar, 31Zimmermann M.B. Chassard C. Rohner F. et al.The effects of iron fortification on the gut microbiota in African children: a randomized controlled trial in Cote d'Ivoire.Am J Clin Nutr. 2010; 92: 1406-1415Crossref PubMed Scopus (310) Google Scholar The notion that excess iron may lead to a loss of immune tolerance has precedent; a population-based case-control study found a greater than 2-fold risk of CD among individuals with hereditary hemochromatosis.32Ludvigsson J.F. Murray J.A. Adams P.C. et al.Does hemochromatosis predispose to celiac disease? A study of 29,096 celiac disease patients.Scand J Gastroenterol. 2013; 48: 176-182Crossref PubMed Scopus (8) Google Scholar A clear mechanism remains elusive at this time, especially because it is not established that exposure to increased iron in utero has downstream consequences on the development of the immune system after birth. However, it cannot be ruled out that the association between iron supplementation and CD is correlative but not causal. One important unmeasured confounding variable is undiagnosed maternal CD, which may be associated with both the exposure (iron supplementation) and is certainly associated with the outcome (increased risk of CD in the offspring). Anemia is strongly associated with undiagnosed CD.33Halfdanarson T.R. Litzow M.R. Murray J.A. Hematologic manifestations of celiac disease.Blood. 2007; 109: 412-421Crossref PubMed Scopus (273) Google Scholar In a population where approximately 10% of women had anemia, assuming that 3% of such individuals with anemia have undiagnosed CD33Halfdanarson T.R. Litzow M.R. Murray J.A. Hematologic manifestations of celiac disease.Blood. 2007; 109: 412-421Crossref PubMed Scopus (273) Google Scholar (compared with a general population rate of 1%), and that 10% of these mothers' children would develop CD, an association between iron supplementation and CD would be observed, with an odds ratio of 1.27, which is similar in magnitude to the effect shown in this study. One way to examine the potential role of undiagnosed CD in women with anemia is to analyze maternal iron supplementation only in women without a diagnosis of anemia to see if the association with offspring CD remains. Given the uncertainty regarding causality, it is premature to argue that iron supplementation during pregnancy should be avoided in individuals with CD whose offspring are at risk for developing this condition. Overall, although statistically significant and intriguing, the observed effect (an absolute risk increase of 0.15%) is modest in magnitude. This exposure is not driving the slow, ongoing epidemic of CD. But the signal uncovered by this study may uncover the mechanisms by which many more have been losing immune tolerance of gluten in recent years. Association Between Maternal Iron Supplementation During Pregnancy and Risk of Celiac Disease in ChildrenClinical Gastroenterology and HepatologyVol. 12Issue 4PreviewThe aim of our study was to determine whether the use of iron supplements during pregnancy affects the risk for celiac disease in children. Full-Text PDF Issue HighlightsClinical Gastroenterology and HepatologyVol. 12Issue 4PreviewRoutine use of iron supplements in pregnancy is recommended in affluent countries. This recommendation, however, has been challenged in this issue of Clinical Gastroenterology and Hepatology. In a prospective cohort study of 78,846 children in Norway by Størdal et al, it was found that using iron supplements during pregnancy was associated with a modest but significant increased risk of celiac disease (4.65/1000 children whose mothers took iron supplements during pregnancy compared to 3.15/1000 children whose mothers did not [odds ratio, 1.33; 95% confidence interval, 1.05–1.68; P = .019]). Full-Text PDF
Referência(s)