Celiac Disease in the New World

It seems that in the past few years an explosion of new interest in celiac disease (CD) is occurring, both in the Old World, which has been traditionally at the cutting edge of research into and scientific excitement regarding this disease, and, finally and very recently, also in the New World. The article by Araya et al. published in this issue is yet another example of the resurgence of interest in this fascinating disease. This is one of the very first reports published on Latin American celiac disease. Does celiac disease exist in South America? Sure it does. A very recent study—to my knowledge the first ever to address the issue of CD prevalence in South America—estimates that in Brazil CD is found in at least 1 in 681 individuals (1). In fact, of 2045 sera from unselected blood donors, 3 anti-endomysium (EMA)–positive sera were found, and 3 more were both immunoglobulin (Ig)G and IgA-antigliadin antibody (AGA)–positive. Also, previous reports from Venezuela (2), Brazil (3), Argentina (4,5), Uruguay (6), and Chile from the same group (7) have shown in various ways that CD is present throughout Latin America. Araya et al. now report an epidemiologic investigation from Chile intended to provide some insight into the genetic composition of patients with CD compared with their relatives, their adherence to gluten-free diet, and their clinical–nutritional status. We thus learn some interesting facts. First, all the patients, interviewed (at mean age 18 years) “years after” the original diagnosis, described themselves as “healthy,” although less than one fourth of them are maintaining a strict gluten-free diet. Are Amerindians with CD luckier than their European counterparts? Are they really “healthy” after years of gluten-containing diet? That this is not the case is not surprising. In fact, we learn that 21% of them had a short stature, in evident contrast with the expected normal distribution of heights among their first-degree relatives. Additionally, in spite of their generic claim of well-being, in response to a specific questionnaire we also learn that 26% of them have some symptoms potentially related to CD, such as decreased appetite, weight loss, diarrhea, and dyspepsia. Still, the majority of the patients had eaten some gluten and felt well. These findings (frequent noncompliance associated with minor morbidity) are not new in the celiac community. Indeed, imperfect gluten avoidance is a very common finding in studies dealing with long-term follow-up of celiac patients. Minimal or more substantial transgressions are found in percentages of teenagers and adults varying between 19% and 55%(8–11). Furthermore, dietetic compliance is reportedly even lower in patients with CD diagnosed after mass screening (12). As in the study by Araya et al., the majority of the patients who are not following a strict gluten-free diet do not report symptoms. In prior studies about CD and dietetic compliance mainly from Europe (9,13), mild malnutrition and mild to severe histologic abnormalities in jejunal biopsy specimens have been frequent findings. The crucial question here is whether they should stay on a gluten-free diet. Patients with celiac disease who are consuming a gluten-containing diet, even if apparently asymptomatic, seem to be at higher risk of complications. Remaining on a strict gluten-free diet prevents most, and possibly all, of the known complications of CD. This has been shown for the bone density changes reported in CD (14–16), for short stature (11), and for the common association with reactive lymphoid follicles in the gastric mucosa (17). No effect of dietary compliance could be shown only for another disorder associated with CD, depression (18). Also, we can infer indirectly that a preventative effect of gluten-free diet on the development of autoimmune disorders associated with CD is likely. This stems from the data produced by the Italian large multicenter study demonstrating an association between length of prediagnostic phase (and thus of the gluten-containing diet) and incidence of autoimmune disorders (19). Of course, malignancies, among them particularly lymphoma of the small intestine, represent the most severe of all complications of CD (20–22). Although their incidence is certainly significantly higher than in the general population, the role of the gluten-free diet in prevention of CD-associated malignancies, first suggested by Holmes (23) in a famous epidemiologic study in 1989 and by other investigators (24), awaits a definitive confirmation. A large epidemiologic investigation sponsored by the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) is currently ongoing, and its results will be greatly appreciated by the scientific community. The other interesting aspect covered by Araya et al. in their study is the search for human leukocyte antigen (HLA) alleles in patients with CD and in their relatives. It is a well-known fact that CD tends to run in families (25). The disease prevalence among healthy first-degree relatives varies from 8% to 18%. Twin studies have also reinforced the evidence for a genetic basis of CD. In fact, the concordance rate between identical twins is 70% and may actually be higher, because individuals thought to be discordant have later been shown to be concordant. Subsequently, it has become clear that susceptibility to CD is conferred by some HLA haplotypes. Celiac disease has in fact been shown to be associated with the HLA class II extended haplotypes DR3-DQ2 or DR5/7-DQ2. Approximately 95% of patients with CD carry such a DQ2 heterodimer encoded by the HLA DQA1*0501 and DQB1*02 alleles, and most of the remaining patients carry the DQ8 heterodimer (encoded by the alleles DQA1*0301 and DQB1*0302). Among the DQ2 carriers, the risk of developing the disease has been shown to be increased in individuals homozygous for DQβ1*0201. A recent, fascinating development of the search for the pathogenesis of CD has shown what is the likely role of this molecule in the pathogenesis of CD. Gliadin peptides, once deamidated in a specific and orderly fashion by tissue transglutaminase, the newly recognized autoantigen of CD, bind with high affinity to the peptide-binding groove of DQ2 and DQ8 molecules. This binding is considered a critical factor in triggering the T-cell response to gliadin. For recent excellent reviews see Schuppan (26) and Sollid (27). Araya et al. find, in this study and in their previous one (7), that their Amerindian population of patients with CD, unlike their European counterparts, most commonly carry the DQ8 HLA haplotypes. This does not come as a surprise, because it reflects the strong Spanish genetic influence of that population. Also, the high incidence of DR4 haplotypes is in part the result of the Mapuche Indian genetic component. The authors also put forward an interesting hypothesis involving a possible “protective” effect by some genetic alleles. Observing that the frequency of the allele DQA1*0101 (and of other DQB1 alleles) is significantly higher among relatives of patients with CD than in their affected relatives, and that the DQA1-0101/DQB1-0501 haplotype has been frequently found in European populations not affected by CD, they speculate that these alleles may actually have a protective rather than neutral effect on CD development. Of course, this is presently merely a fascinating speculation, and no proof whatsoever is available at this point. Finally, it should not be forgotten that the now well-identified “susceptibility” haplotypes for CD (DQ2, DQ8) do not tell the whole genetic story. In fact, they are also carried by as many as 25% to 30% of the general population. In other words, less than 1% of the carriers of this allele have CD. It is thus clear that other genetic factors altering mucosal susceptibility are necessary. To determine the localization of other genetic risk factors a systematic screening of the genome was undertaken on 110 affected sibling pairs and their parents to test linkage. In addition to HLA loci, a risk factor in 5qter was suggested in both symptomatic and silent forms of CD (28). More recently, the chromosomal region of the CD28/CTLA4 genes on 2q33, which encode receptors regulating the T-lymphocyte activation and are linked to the susceptibility to many autoimmune diseases, was tested for linkage to CD. This region was indicated as a possible novel susceptibility locus for CD, supporting the now widely held belief that CD has a strong immune-mediated component (29). In summary, much investigation has been undertaken on the genetic aspects of CD, and the exciting future in this area most likely lies in the search for and identification of factors outside the HLA region. The rapidly advancing understanding of the sophisticated molecular biology underlying CD should not, however, lead us to forget the everyday practical issue of making the diagnosis as early as possible. In this regard, the Old World remains light years ahead of the New World. In this area, countries usually at the forefront, such as the United States, are lagging far behind, as currently pitilessly pointed out to the world Internet community by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) of the National Institutes of Health, Bethesda, Maryland. In its Website (http://www.niddk. nih.gov/health/digest/pubs/celiac/index.htm) the NIDDK currently states that: “In the United States, the time between the first symptoms and diagnosis [of celiac disease] averages about 10 years.” Thus, it appears that the fast pace of bench research in this field is not followed as it should be by a parallel increase in awareness of celiac disease in this part of the world—a remarkable example of how research and education do not necessarily feed each other as they should for the best interest of our patients.