Since the first description (1) of atypical clinical presentation of celiac disease (CD), several investigators in recent decades have reported a decreasing prevalence of the classic celiac triad—failure to thrive (FTT), diarrhea, and abdominal distension—in children with CD (2,3). The prevalence of a different clinical picture or of absence of symptoms (4) is rising, and the diagnosis may be delayed. Many individuals with CD may remain undiagnosed and are therefore exposed to the risk for long-term complications, such as osteoporosis, infertility, cancer, and higher mortality for intestinal lymphoma (5). To ascertain whether the clinical pattern and age at diagnosis of CD are still changing in children in more recent years, we analyzed data from all children receiving a diagnosis of CD in 2 towns of the same region in northern Italy in the past 20 years.
PATIENTS AND METHODS
A database was established in 1987 where data concerning children undergoing esophagogastroduodenoscopy (EGDS) from 1987 to 2006 were entered. EGDS was always performed by the same operator (G.O.). For the first period (1987–1995) data were collected in the Pediatric Endoscopy Unit in Torino, and for the second period (1996–2006) in Novara, where the endoscopist had moved. The database was structured and filled in the same way in the 2 towns in a northern Italian region (Piemonte) serving similar populations, so comparison between the 2 periods is methodologically possible. Data entered were as follows: name; sex; date of birth; age at diagnosis; symptoms at presentation; family history; anthropometric parameters such as weight, height, and percentiles; serological markers such as anti-gliadin antiendomysium and/or anti-tissue-transglutaminase antibodies (past 5 years); hemoglobin; serum iron; and data on endoscopic and histological findings of esophageal, gastric, and duodenal mucosa. For the second decade (1996–2006) data about interval between symptom onset and diagnosis, age of introduction of gluten, and duration of breast-feeding were added.
Of 2422 consecutive children undergoing EGDS in the past 20 years, a diagnosis of CD was made in 307 (12.6%, M/F 100/207) according to the criteria of the European Society of Paediatric Gastroenterology, Hepatology, and Nutrition and the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (6,7). The first 179 (median age, 4.2 years; range, 0.8–18 years) received diagnoses in the first period (1987–1995) and 128 (median age, 5.4; range, 0.9–18.3) in the second period (1996–2006, Table 1). All of the children were referred to the endoscopist for various symptoms and later received diagnoses of CD on the basis of biopsy findings.
Symptoms were divided in typical (FTT, chronic diarrhea, abdominal distension) and atypical (isolated anemia, short stature, recurrent abdominal pain, family history of CD without symptoms, constipation, others) CD. Associated diseases like type 1 diabetes mellitus (TIDM) and Down syndrome were recorded.
We then compared the prevalence of clinical characteristics and histological findings in children with CD divided in 2 groups according to the decades of diagnosis (1987–1995 vs 1996–2006) and then according to four 5-year periods.
Age at diagnosis increased significantly in the second decade from 4.2 to 5.4 years (P = 0.02), mainly because of the older age of children receiving diagnoses in the past 5-year period (P = 0.01) (Table 1 and Fig. 1). The prevalence of typical symptoms at presentation was significantly higher in 1987–1995 than in 1996–2006 (P < 0.0001 for FTT and diarrhea and P = 0.02 for abdominal distension), whereas association with TIDM (P = 0.02) and silent forms (ie, without symptoms or physical signs) were significantly more frequent in the later period (P = 0.01). Children with typical symptoms were significantly younger than children with atypical symptoms (P < 0.0001) in all of the studied periods. When analyzing the four 5-year periods, we found that the prevalence of symptoms at presentation, typical and atypical, in 66 children receiving diagnoses in 1987–1991 compared with 113 children receiving diagnoses in 1992–1995 was similar, but anemia was significantly higher in 1992–1995 (P = 0.001). By contrast, in 43 children receiving diagnoses in 1996–2000, the prevalence of FTT (P = 0.001) and diarrhea (P = 0.01) was significantly lower than in 85 children receiving diagnoses in 2001–2006, whereas the prevalence of recurrent abdominal pain (P = 0.02) and silent forms (P = 0.02) was significantly higher in 2001–2006 (Table 1). In most of the children presenting with typical symptoms, some associated symptoms were present, and for some of those the prevalence was different in the 2 decades and even in the past 5 years.
The prevalence of recurrent vomiting was higher in 1987–1995 (20% vs 11% in 1996–2006, P = 0.03), whereas the opposite was true for a family history of CD (2% vs 19%, P < 0.001) short stature (6% vs 19%, P < 0.001), irritability (1% vs 17%, P < 0.001), and constipation/irregular bowel habits (9% vs 18%, P = 0.04), the latter being more frequent in the past decade. Almost all of the children presenting with atypical symptoms were monosymptomatic except for recurrent abdominal pain that was associated with recurrent vomiting in 23% and irritability in 19% of children. Finally, when we considered children, instead of just symptoms, the prevalence of children with a typical form of CD (ie, with at least 2 or more typical symptoms) decreased every 5 years from 76% in 1987–1990 to 63%, 62%, and 44% in 2001–2006 (Fig. 1).
In children receiving diagnoses in 1987–1995, histological examination showed a significant higher prevalence of subtotal villous atrophy (70% vs 57% in 1996–2006, P = 0.02), with an increased frequency of partial villous atrophy in the latter period, and a significant higher prevalence of gastritis (55% vs 8% in 1996–2006, P < 0.001) and a lower prevalence of esophagitis (15% vs 38% in 1996–2006, P = 0.001).
The age at introduction of gluten (median, 6 months; range, 3–12 months) in the diet, and duration of breastfeeding (median, 3 months; range, 0–18 months), were similar in the most recent 5-year periods. The lag phase between gluten introduction and first symptom was significantly longer for children with atypical symptoms (6.8 years) than for those with typical symptoms (1.2 years, P < 0.001) for the older age at first symptom in children with atypical symptoms.
The clinical spectrum of CD has widened over the past decades, and clinical pattern and age at diagnosis of CD are still changing in our children. Indeed, our data show some significant changes both in clinical presentation and in mucosal damages seen at histological examination in children receiving diagnoses of CD in the past 10 years, mostly in the past 5 years. The prevalence of FTT, diarrhea, and abdominal distension significantly decreased. By contrast, the prevalence of atypical symptoms increased in the past decade, particularly so in the past 5 years, when more than half of the children with CD had an atypical presentation. This was the cause of the median older age at diagnosis because children presenting with typical symptoms were always younger and their median age at diagnosis was similar throughout the 20 years.
Particularly interesting is the increasing frequency of asymptomatic (silent) forms to 10% in the past 5 years. This was probably due to higher diffusion of familial screening—these asymptomatic children were siblings or sons/daughters of individuals with CD—and/or to a wider use of serology with the more sensitive test of anti-tissue transglutaminase antibodies. The increased prevalence of children with CD but with atypical symptoms may be due to a better awareness of family pediatricians looking for CD and performing serological screening more often, even in children with uncertain symptoms. The increased prevalence was not due to more widespread use of duodenal biopsy because it was always our policy to perform a duodenal biopsy in every child undergoing EGDS.
Other changes were seen at histology: prevalence of partial villous atrophy increased in the most recent period, suggesting an earlier diagnosis, when atrophy is not yet fully developed. The increased prevalence of esophagitis suggests in children with CD a similar behavior in the general population, where an increased prevalence of esophagitis (8) has been described in the past decade and is associated with a decreased prevalence of gastritis, possibly because of the decreasing frequency of Helicobacter pylori infection (9). In our series, gastritis was present in 55% of children seen in the first decade, and in 17% of them it was associated with H pylori infection, whereas in the past decade it was seen in 8% and the infection was present in only 1 child.
Our findings demonstrate that changes in clinical presentation of CD in children are still in progress in the past 5 years, and pediatricians—at least in northern Italy—seem to be well aware of them.
1. Maki M, Kallonen K, Lahdeaho ML, et al
. Changing pattern of childhood coeliac disease in Finland. Acta Paediatr Scand 1988; 77:408–412.
2. Steens RFR, Csizmadia CGDS, George EK, et al
. A national prospective study on childhood celiac disease in the Netherlands 1993–2000: an increasing recognition and a changing clinical picture. J Pediatr 2006; 147:239–243.
3. Fasano A. Clinical presentation of celiac disease in the pediatric population. Gastroenterology 2006; 128:68–73.
4. Bottaro G, Cataldo F, Rotolo N, et al
. The clinical pattern of subclinical silent celiac disease: an analysis of 1026 consecutive cases. Am J Gastroenterol 1999; 94:691–696.
5. Corrao G, Corazza GR, Bagnardi V, et al
. Mortality in patients with coeliac disease and their relatives: a cohort study. Lancet 2001; 358:356–361.
6. Walker-Smith JA, Guandalini S, Schmitz J. Revised criteria for diagnosis of coeliac disease. Report of Working Group of European Society of Paediatric Gastroenterology and Nutrition. Arch Dis Child 1990; 65:909–911.
7. Hill ID, Dirks MH, Liptak GS, et al
. Guideline for the diagnosis and treatment of celiac disease in children: recommendations of the North American Society for Paediatric Gastroenterology Hepatology and Nutrition. J Pediatr Gastroenterol Nutr 2006; 40:1–19.
8. El-Serag HB. Time trends of gastroesophageal reflux disease: a systematic review. Clin Gastroenterol Hepatol 2007; 1:17–26.
9. Unal S, Karakan T, Dogan I, et al
. The influence of Helicobacter pylori
infection on the prevalence of endoscopic erosive esophagitis. Helicobacter 2006; 1:556–561.