Evidence suggests that celiac disease develops gradually from small bowel mucosal inflammation to crypt hyperplasia and eventually to overt villous atrophy. The concept of latent celiac disease is well recognized: a patient having a normal small bowel mucosal structure while on a gluten containing diet later develops typical mucosal villous atrophy (1-4). Thus, normal small bowel mucosal histology does not necessarily exclude celiac disease in the long term.
It is problematic to identify patients with celiac disease in its early stage. Minor villous shortening is difficult to interpret. An increased density of small intestinal intraepithelial lymphocytes (IELs) might predict forthcoming villous atrophy (3,5,6) but is considered unspecific (7). It remains obscure how common early-stage celiac disease is, who will eventually develop small bowel mucosal villous atrophy and how rapidly this will occur. The diagnosis of early or latent celiac disease is thus virtually always retrospectively verified only after mucosal deterioration has already taken place. This prompted us to evaluate the development of celiac-type villous atrophy in a large group of patients suspected of celiac disease 8 to 28 years earlier during childhood but found not to have the disease at the time by existing histologic criteria.
MATERIALS AND METHODS
From 1976 to 1992, 1158 patients suspected on clinical grounds of having celiac disease underwent small intestinal mucosal biopsy at the Department of Pediatrics in Tampere University Hospital. Of these, 178 (15.4%) had severe partial or subtotal villous atrophy with crypt hyperplasia consistent with celiac disease; the remaining 980 (84.6%) were did not have celiac disease because of apparently normal small bowel histology. The histology of the normal small bowel mucosal samples was re-evaluated in 1999; all 236 patients with minor changes comprised the case group and an equal number of sex and age-matched individuals showing unequivocally normal histology were selected as controls.
Altogether, 310 individuals (173 cases and 137 controls) could be traced and were invited to attend a follow-up study in 2001; 76 cases and 68 controls consented and 97 cases and 69 controls refused (Fig. 1). Symptoms or signs leading to a suspicion of celiac disease, possible adherence to a gluten-free diet and associated diseases were recorded by questionnaire. The emergence of celiac disease during follow-up was recorded from patient files; our hospital was the main (and virtually only) center for diagnostics and dietary counseling in celiac disease in children and adolescents.
The small bowel biopsies were obtained with a pediatric capsule at the first examination. At follow-up, biopsies were taken by during upper gastrointestinal endoscopy from the distal part of the duodenum from patients yielding positive serological screening results. Biopsies were processed and stained with hematoxylin-eosin and studied under light microscopy.
Small intestinal mucosal villous height/crypt depth ratio and density of IELs were measured in well-oriented biopsies as described (8). The reference values for these parameters were obtained in 169 subjects with normal small bowel biopsy specimens; mean +1 standard deviation was considered abnormal. A villous height/crypt depth ratio of 2.0 to 2.6 was defined as slightly decreased, a lower value being indicative of overt celiac disease and a higher value being indicative of normal mucosal architecture. IELs were counted per 100 enterocytes; inclusion criteria in our study for cases was >26 IELs per 100 enterocytes and for controls it was ≤26 IELs per 100 enterocytes. These values correspond well to those previously reported in control subjects (8,9). The cases had either a villous height/crypt depth ratio of 2.0-2.6 or increased IELs or both; in the control group both parameters were normal.
In the follow-up examination, two additional small bowel biopsies were freshly embedded in optimal-temperature compound (OCT, Tissue-Tec, Miles Inc, Elkhart, IN), snap frozen in liquid nitrogen and stored at −70°C. Five μm thick frozen sections were stained with monoclonal antibodies to detect αβ+ (βF1 antibody, Endogen, Woburn, MA) and γδ+ IELs (TCRγ, Endogen). The positive IELs were counted with a ×100 flat-field light microscope objective in randomly selected areas of surface epithelium; at least 30 fields of 1.6 mm epithelial length were counted and the density of IELs expressed as cells/ millimeter epithelium (10,11). Based on our previous study, reference values for αβ+ IELs were determined at 25 cells/mm and for γδ+ IELs at 4.3 cells/mm, respectively (12).
Serology and Chemical Analysis
Serum immunoglobulin (Ig)A class endomysial antibodies were evaluated by indirect immunofluorescence using human umbilical cord as substrate (13); a dilution of 1:≥5 was considered positive. Serum tissue transglutaminase IgA antibodies were determined by enzyme-linked immunosorbent assay (Inova Diagnostics, San Diego, CA) using guinea pig liver as antigen (14), a unit value (U) ≥20 being considered positive. Serum gliadin antibodies (IgA) were measured by enzyme-linked immunosorbent assay; the lower limit of positivity was 0.2 enzyme-linked immunosorbent assay units per millilitre (EU/mL) (14). Patients were also screened for IgA deficiency (IgA <0.05 g/L) by the nephelometric IgA method. Blood hemoglobin and erythrocyte folic acid concentrations were measured using routine laboratory methods in those who underwent endoscopy.
Human Leukocyte Antigen-DQ Typing
DQ typing was based on polymerase chain reaction with allele-specific primers identifying DQ2 and DQ8 and performed with a Dynal DQ “low resolution” SSp kit (Dynal AS, Oslo, Norway).
χ2 test with Yates correction was used in cross-tabulations; a P value of <0.05 was considered significant.
The study protocol was approved by the Ethical Committee of Tampere University Hospital. All subjects gave informed consent.
Four of the 236 subjects in the case group had been diagnosed with celiac disease after the first biopsy (Table 1). They had been re-examined despite the first apparently normal small bowel mucosal histology 6 months to 18 years after the first biopsy because of symptoms or family history. Celiac disease diagnoses were based on typical lesions in small bowel biopsies according to the ESPGHAN criteria (15). Clinical, histologic or serological improvement after the introduction of a gluten-free diet was seen in all four subjects.
Altogether, 144 individuals (76 cases and 68 controls) consented to join the follow-up study; none had selective IgA-deficiency. Three cases and seven controls were positive in one or several serological screening tests (Table 2). The blood hemoglobin was low in one case; in the remaining cases and in all seven controls hemoglobin and erythrocyte folic acid levels were normal. Nine of 10 seropositive individuals underwent endoscopy. Two new histologically verified celiac disease patients, one in each group, were found (Table 2). In the two new celiac patients, the density of IELs was increased, that of γ/δ+ IELs being 6.5 cells/mm (case) and 27 cells/mm (control), and of α/ß+ IELs 27 (case) and 50 (control). Both new patients also carried the human leukocyte antigen haplotype characteristic for celiac disease (Table 2). A gluten-free diet was administered, and serological and clinical improvement was evident in both.
In the remaining seven patients with available follow-up histology, villous height/crypt depth ratio and total IEL counts in the second biopsy were normal in all, but the density of γ/δ IELs was increased (6.5 cells/mm) in one control patient (patient number 5, Table 2). This patient also had human leukocyte antigen DQ8. Four of the seven who underwent a second biopsy because of positive serology and who had normal villous architecture did not have human leukocyte antigen DQ2 nor DQ8 (Table 2).
Upper gastrointestinal endoscopy was not performed in one control with Down syndrome (patient number 10, Table 2) for lack of co-operation. She was human leukocyte antigen DQ2 and DQ8 negative and therefore unlikely to have celiac disease.
All participating cases and controls were consuming a normal gluten-containing diet, and, according to their own estimate, none had made any changes in gluten intake during follow-up. Three cases and three controls had diabetes mellitus: two cases and one control inflammatory bowel disease.
To explore a potential bias resulting from the low participation rate, demographic data on the original series were assessed and are shown in Table 3. Cases participating in the follow-up study had significantly more frequent abdominal symptoms at baseline than did cases declining to participate or participating controls. Otherwise, there were no significant differences between those who refused and those who participated. At the follow-up 74% of the cases and 68% of the controls suffered from mostly slight abdominal symptoms (Table 3).
Previous studies have shown gradual development of villous atrophy in celiac disease but the numbers of patients have been small and the follow-up times relatively short (2,4,16,17). This study is the first long-term follow-up study in a large patient group to evaluate the development of celiac disease in patients with suspicion of celiac disease but normal or only equivocal small intestinal mucosal changes.
Intraepithelial lymphocytosis and a slight decrease in villous height/crypt depth ratio were not uncommon, and were found in 236 of the 980 biopsy specimens from patients suspected of celiac disease but without villous atrophy on biopsy. We found that progression from mild change to overt villous atrophy was rare even during a long follow-up. Four new celiac disease patients were detected on clinical grounds and subsequent screening found only two more. Furthermore, the screening results in cases with slight small-bowel mucosal abnormalities were similar to those in controls having normal mucosa. The minimum frequency of celiac disease was slightly higher (5 of 236, 2.1%) in the case group than in the control group (1 of 236, 0.4%), but the difference was not statistically significant, (P = 0.22).
One limitation of this study was the low rate of participation in follow-up screening tests. Once celiac disease had been excluded there was likely reluctance to consider repeated upper intestinal endoscopy. Table 3 shows, however, that patients who refused to participate did not significantly differ from those who participated except that they had somewhat fewer abdominal complaints. We may nevertheless presume that the frequency of celiac disease among those who did not participate would not be higher than in those who participated. We would thus have detected two more celiac patients in both groups among nonparticipants. The calculated maximum rate in such a hypothetical case would be 7 of 236 (3.0%) in cases and 3 of 236 (1.3%) in controls.
The present results indicate that follow-up screening of celiac disease in patients evincing minor small bowel mucosal changes does not, alone, yield significantly more positive cases than does screening in the general population (18,19). A gluten-free diet may, however, be of benefit in some patients with abdominal complaints and with only minor histologic lesions in their small bowel biopsy (20-22). Clearly, more specific methods are needed to extend surveillance or even gluten-free diet treatment to patients with minor mucosal lesions consistent with Marsh I or II to avoid overdiagnosis. A celiac-type human leukocyte antigen is mandatory in these cases, as the absence of both human leukocyte antigen DQ2 and DQ8 virtually excludes celiac disease (23,24). The treatment of symptomatic patients with minor small bowel mucosal changes is so far recommended in controlled studies. Gluten dependence should be further ascertained in these cases by the alleviation of symptoms, a decrease in antibodies and histologic recovery in IELs in general and in αβ+ and γδ+ IELs in particular (22).
In conclusion, although our results show that the mucosal damage in celiac disease may take several years, intraepithelial lymphocytosis and minor villous blunting is a nonspecific phenomenon. Only a minority of patients develop celiac disease even during a long follow-up. Routine surveillance is not necessary unless indicated for other reasons.
This study was supported by the Research Fund of the Finnish Coeliac Society, the Medical Research Fund of Tampere University Hospital and the Finnish Medical Foundation.
1. Mäki M, Holm K, Koskimies S, et al. Normal small bowel biopsy followed by coeliac disease. Arch Dis Child
2. Corazza GR, Andreani ML, Biagi F, et al. Clinical, pathologic, and antibody pattern of latent celiac disease
: Report of three adult cases. Am J Gastroenterol
3. Kaukinen K, Collin P, Holm K, et al. Small bowel mucosal inflammation in reticulin or gliadin antibody-positive patients without villous atrophy. Scand J Gastroenterol
4. Niveloni S, Pedreira S, Sugai E, et al. The natural history of gluten sensitivity: report of two new celiac disease
patients resulting from long-term follow-up of nonatrophic, first-degree relatives. Am J Gastroenterol
5. Troncone R, Greco L, Mayer M, et al. Latent and potential coeliac disease. Acta Paediatr Suppl
6. Wahab PJ, Crusius JBA, Meijer JWR, et al. Gluten challenge in borderline gluten-sensitive enteropathy. Am J Gastroenterol
7. Kakar S, Nehra V, Murray JA, et al. Significance of intraepithelial lymphocytosis in small bowel biopsy samples with normal mucosal architecture. Am J Gastroenterol
8. Kuitunen P, Kosnai I, Savilahti E. Morphometric study of the jejunal mucosa in various childhood enteropathies with special reference to intraepithelial lymphocytes. J Pediatr Gastroenterol Nutr
9. Hayat M, Cairns A, Dixon MF, et al. Quantitation of intraepithelial lymphocytes in human duodenum: What is normal? J Clin Pathol
10. Savilahti E, Arato A, Verkasalo M. Intestinal gamma/delta receptor-bearing T lymphocytes in celiac disease
and inflammatory bowel diseases in children. Constant increase in celiac disease
. Pediatr Res
11. Arranz E, Bode J, Kingstone K, et al. Intestinal antibody pattern of coeliac disease: Association with gamma/delta T cell receptor expression by intraepithelial lymphocytes, and other indices of potential coeliac disease. Gut
12. Järvinen TT, Kaukinen K, Laurila K, et al. Intraepithelial lymphocytes in celiac disease
. Am J Gastroenterol
13. Ladinser B, Rossipal E, Pittschieler K. Endomysium antibodies in coeliac disease: An improved method. Gut
14. Sulkanen S, Halttunen T, Marttinen A, et al. Autoantibodies in celiac disease
, importance of fibroblasts. J Pediatr Gasstroenterol Nutr
15. Walker-Smith JA, Guandalini S, Schmitz J, Schmerling DH, Visakorpi JK. Revised criteria for diagnosis of coeliac disease. Arch Dis Child
16. Collin P, Helin H, Mäki M, et al. Follow-up of patients positive in reticulin and gliadin antibody tests with normal small bowel biopsy findings. Scand J Gastroenterol
17. Iltanen S, Holm K, Ashorn M, et al. Changing jejunal gamma/delta T cell receptor (TCR)-bearing intraepithelial lymphocyte density in coeliac disease. Clin Exp Immunol
18. Mäki M, Mustalahti K, Kokkonen J, et al. Prevalence of celiac disease
among children in Finland. N Engl J Med
19. Fasano A, Berti I, Gerarduzzi T, et al. Prevalence of celiac disease
in at-risk and not-at-risk groups in the United States: A large multicenter study. Arch Intern Med
20. Cooper BT, Holmes GKT, Ferguson R, et al. Gluten-sensitive diarrhea without evidence of celiac disease
21. Arranz E, Ferguson A. Intestinal antibody pattern of celiac disease
: Occurrence in patients with normal jejunal biopsy histology. Gastroenterology
22. Kaukinen K, Mäki M, Partanen J, et al. Celiac disease
without villous atrophy. Revision of criteria called for. Dig Dis Sci
23. Kaukinen K, Partanen J, Mäki M, et al. HLA DQ typing in the diagnosis of celiac disease
. Am J Gastroenterol
24. Karell K, Louka AS, Moodie SJ, et al. HLA types in celiac disease
patients not carrying the DQA1*05-DQB1*02 (DQ2) heterodimer: results from the European genetics cluster on coeliac disease. Hum Immunol