What Is Known
- Coeliac disease is treated with a strict long-term gluten-free diet.
- Few data are available about the long-term clinical history and possible consequences of poor adherence to the diet.
What Is New
- Among patients with a long-term poor adherence to gluten-free diet, almost two-thirds showed no recurrence of villous atrophy at duodenal biopsies.
- The knowledge of (and adherence to) gluten-free diet rules may be partly lost during transition to adulthood in patients with coeliac disease diagnosis at childhood.
- Coeliac disease patients diagnosed more than 30 years ago do not have a higher mortality compared to the general population.
For patients with coeliac disease (CD) a lifelong gluten-free diet (GFD) represents the only effective treatment (1). GFD leads to the resolution of clinical symptoms and serological markers, the recovery of normal duodenal histology, and is thought to be effective in the prevention of complications (2). The lack of adherence to strict GFD is considered the main reason for a poorly controlled disease with an increased risk of intestinal lymphoma development (3–6). Despite the constant growth of the gluten-free food market and the increasing availability of specific products, GFD, however, still remains very restrictive and results in social burden and poor compliance, especially for adolescents and young adults (7). Thus, the transition from childhood to adulthood plays a key role in maintaining GFD adherence (8). On the other end, the concept of “gluten tolerance” has been introduced in the literature to identify subgroups of patients who can safely reintroduce gluten in their diet (9). Although some of the characteristics of CD have been widely investigated and established among the scientific community, other questions remain open. A study on the long-term history of CD can help clarify some of the shadows of this chronic condition. With these premises, we aimed to evaluate the evolution of CD in a cohort of patients mostly diagnosed at paediatric age, over a long follow-up period (at least 30 years). In particular, our aim was to assess the rate and accuracy of compliance by these patients to strict GFD, to verify the rate of their participation to regular clinical follow-up programs, and to compare clinical data and endpoints such as death, complications, symptoms between patients who stayed on strict GFD and patients who were not compliant to their diet. Furthermore, we conceived a questionnaire to be filled in by patients and parents/caregivers to evaluate GFD knowledge.
This monocentric cohort study was performed at the Center for the Prevention and Diagnosis of Celiac Disease (ie, a regional tertiary referral centre) of the Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico (Milan, Italy). The study was approved by the local Ethics committee (ref. no. 554/14).
All the patients with a diagnosis of CD before 1985, thus with a follow-up of at least 30 years, were eligible in the study. Due to the existence at the Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico of a CD dedicated service since the 1970s, we were able to collect the original clinical, laboratory, histologic, and genetic (HLA asset) data at diagnosis and during follow-up directly from the archive of clinical records available at our centre. Moreover, between October 2014 and May 2015, either during the annual clinical follow-up or by telephone call, we completed a case report form for every patient to collect data about adherence to GFD and current health status.
The clinical presentation type at CD onset was classified as: classic (diarrhoea, weight loss, longitudinal growth retardation), non-classic if mono-paucisymptomatic (dyspepsia, anemia, hypertransaminasemia, etc) or associated with the presence of dermatitis herpetiformis (DH) (10). The diagnosis of CD was confirmed according to the then-applicable guidelines (11,12). All duodenal biopsies were performed at the Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico and evaluated by an expert pathologist within our tertiary referral centre. Clinical and histological documentation was available in the archives of clinical records and pathology reports were all revised after the validation of the Marsh Oberhuber classification in 1999 (13).
The patients were divided into 3 groups according to their reported GFD compliance. The first group consisted of patients reporting strict lifelong GFD (lifelong-GFD group); the second group was made up of CD subjects reporting a past GFD discontinuation of at least 5 years (discontinued GFD group); the third group included patients with a chronic continuous gluten ingestion since CD diagnosis (no-GFD group).
Standardized Mortality Ratio Analysis
When information was unavailable, death and cause of death were retrieved from Italy's National Health Service central register, which collects and manages epidemiological data on the whole country population. Analysis was carried out on the overall mortality levels for all causes/pathologies. The comparison was made using the mortality tables of Lombardy's population (from Italy's National Institute of Statistics, ISTAT) between 2002 and 2015 (the same range of years when deaths occurred). SURVSOFT V.2.0 software was used for the analysis (Cancer Registry Bavaria, http://www.krebsregister-bayern.de/software_e.html). The standardized mortality ratio is defined as the ratio of the observed number of deaths in the population being studied and the expected number of deaths in a comparable group of individuals from the general population, matched with respect to the main factors affecting mortality, commonly age, sex, calendar period:
Gluten-free Diet Questionnaire
The patients were asked to fill in a 2-section questionnaire, meant to investigate their knowledge about GFD and dietary recommendations (see Supplementary Figure 1, Supplemental Digital content, http://links.lww.com/MPG/B357). The first section (“Rules for correct GFD”) consisted in 16 questions focusing on the knowledge of dietary and behavioural recommendations when on GFD. The patients were asked to report on their adherence to each recommendation by means of multiple-choice questions: a score ranging from 0 to 1 was assigned to each answer (“yes” = 1 point; “yes, for <5 years” = 0.75; “yes in the past, but not anymore” = 0.5; “no” = 0), thus the score for this section could range from 0 (minimal adherence) to 16 (optimal adherence). The second section (“Gluten content in food”) investigated the actual level of awareness about gluten content in 33 common gluten-containing and gluten-free foods. The score, ranging from 0 (no knowledge about gluten-free foods) to 33 (optimal knowledge), was calculated by assigning 1 point for each correct answer. Former caregivers were also asked to fill in a specular questionnaire. The results of the questionnaires were compared to clinical and demographic data in order to identify possible correlations between the knowledge by patients and caregivers about GFD rules and the prognosis of patients.
All the variables were expressed as number (percentages) or mean ± standard deviation (SD) or median and range. Continuous variables were analysed with the ANOVA one-way variance test or with the non-parametric Kruskal-Wallis test and the t test independent sample or Mann-Whitney's test. Categorical variables were compared using chi-square or Fisher's exact test and by multivariate analysis when appropriate. The Kolmogorov-Smirnov test was used to assess the Gaussian distribution of the data. Correlations were analysed by Pearson or Spearman's test in case of Gaussian or non-parametric variables. The P values lower than 0.05, 2-tailed, were considered statistically significant. All the analyses were carried out by statistical software SPSS ver. 18 (rel. 2009; SPSS Inc, Chicago, IL).
The Center for the Prevention and Diagnosis of Coeliac Disease has been actively involved in the management of CD since the 1970s and to date manages about 3000 CD patients. Among them, 337 coeliac patients who had been diagnosed before 1985 (220 women, median age at enrolment 40, range 31–81 years, median age at CD diagnosis 5, range 0–46 years) were identified. In order to ascertain the death rate and the causes of death in this group, mortality data were retrieved from Italy's National Health service central register: totalling 4 deaths (3 women died at 23, 32, and 53 years of road accident, hepatocarcinoma developed after HCV hepatitis, duodenal adenocarcinoma and a 69-year-old man died of adenorcarcinoma of the colon). With comparison to the general population, the standardised mortality ratio was 0.37 (confidence interval 0.10 to 0.94).
Among the 337 CD patients, 197 (58%, 127 women, median age at enrolment 41, range 31–65 years) were eligible for further analysis (see Flowchart of the study, Fig. 1) and were divided into 3 groups according to their GFD adherence: 133 (67%) CD patients reported to have kept up with strict lifelong GFD, 29 (15%) discontinued GFD and 35 (18%) are on gluten-containing diet (Table 1). Almost all (94%) patients had their CD diagnosis during paediatric age (0–18) with 65% below the age of 3 years. No significant differences in demographic data at diagnosis and at enrolment were found between patients for whom data on GFD adherence were available and the remaining 140 CD patients without such data. The 3 groups did not show any significant difference in age at CD diagnosis nor at follow-up. The onset of symptoms was mainly classical in the 3 groups. The diagnosis was confirmed according to the different recommendations and no differences were seen within the 3 groups as regards histologic findings at diagnosis. Other autoimmune diseases were diagnosed in all 3 groups independently from their adherence to diet.
During follow-up, 63 (32%) patients underwent at least 1 gastroscopy with biopsy (29/133 patients from the lifelong GFD group, 20/29 patients from the discontinued GFD group and 14/35 patients from the no GFD group, see Table 2). The follow-up endoscopy was performed more frequently in the discontinued-GFD group of patients compared to the other groups. For the patients on strict GFD, the most common indication was the presence of GI symptoms (62%), the remaining patients underwent endoscopy mainly for the persistence/recurrence of positive serology or gluten consumption. A non-atrophic (Marsh 0–1–2) duodenal mucosa was seen in 27/29 (93%) cases in the lifelong GFD group, in 12/20 (60%) cases in the discontinued GFD group and in 8/14 (57%) cases in the no GFD group. Interestingly, 20 patients out of 34 that declared past or present gluten consumption had normal duodenal histology. At multivariate analysis, no relevant clinical or demographic differences were found between patients with chronic gluten consumption with persistent atrophy and those without atrophy at follow-up biopsies. Furthermore, 11 out of 20 patients (55%) with normal histology while on a gluten-containing diet had HLA assessment, which was compatible with CD (all carrying HLA DQ2). For the remaining 9 patients HLA was not available.
When asked to fill the questionnaire in, 90 patients returned it (submission rate was 46%). The questionnaires were returned by 70/133 patients in the lifelong GFD group, 13/29 patients in the discontinued GFD group, 7/35 patients in the no GFD group. After dividing in 2 groups (lifelong-GFD vs discontinued-GFD plus no-GFD groups) we observed lower scores in the gluten-eating patients (Fig. 2A). This difference was significant for the total score (P = 0.033) and the first part on GFD rules (P = 0.028) but not for the second on food categories (P = 0.081). In 66 cases the caregivers returned the questionnaires. A significant moderate correlation was shown between the total scores of CD patients and those of the caregivers (correlation index 0.492, P < 0.001). No significant difference was seen in caregivers’ score if divided according to patients’ gluten consumption (Fig. 2B).
The investigation of the clinical characteristics of patients with a 30 years long history of CD confirmed that back in the 1980s CD was mostly a paediatric business, and that the poor adherence to GFD is the major predictor of the persistence of mucosal lesions at follow-up endoscopy. Our data, however, have also showed that in a subset of patients, a long-term gluten-containing diet did not cause any relapse of villous atrophy. Our long-term follow-up data showed that the incidence of CD complications in patients with a diagnosed CD is low. Furthermore, we were able to ascertain, through a specifically designed questionnaire, that a certain degree of GFD's knowledge is lost during the transition from childhood to adulthood.
The efficacy of GFD in terms of allowing complete mucosal recovery has been a matter of discussion after the publication of data suggesting that a certain degree of duodenal mucosal damage persists in spite of strict GFD (14–17). The actual requirement of GFD for all CD patients has been questioned on the basis of data suggesting that a subset of patients may develop a sort of quiescence with the maintenance of clinical and mucosal recovery after a gluten-containing diet (18,19). A study on gluten challenge in CD showed that a subset of patients does not develop duodenal mucosal changes after their re-exposure to gluten (20). Results that point to the same direction have been reached in the setting of potential CD patients, many of whom do not seem to develop CD over time despite a prolonged gluten intake, some of them even reverting to negative serology (21). In this study, the stratification of our cohort into different groups according to the reported GFD adherence has showed that nearly two-thirds of patients with poor GFD adherence who had been evaluated with gastroscopy had no symptoms relapse and no villous atrophy. The proportion is remarkable if compared to that 20% reported in the study by Matysiak-Budnik et al (18). As already reported by those Authors, the development of CD remission may be transient and a single follow-up biopsy is probably insufficient (18). Moreover, villous atrophy may not always be a reliable marker of disease activity and that the Marsh score alone may be insufficient to evaluate mucosal changes during follow-up (22). Although the subgroup of patients who underwent follow-up gastroscopy is relatively small (only 63 patients), these results confirm the possibility that a subgroup of CD patients may be suitable for a controlled gluten re-introduction after a period on GFD and without the risk of developing symptoms. Unfortunately, due to long-term and retrospective study design we are not able to assess the quantity of gluten ingested, bulb histology, persistence of normal biopsies, and immunological pattern or analysis of T cell reactivity toward gluten or gliadin peptides. For all these reasons we could refer to our patients as temporarily CD remission while on a gluten-containing diet. A larger prospective multicentre study could help to identify quantity of gluten allowed, duration of remission, and eventual development of permanent gluten tolerance.
Another noticeable data is a rate of CD complications apparently lower than what reported in earlier literature, with an overall incidence of 1.5% over 30 years (1 patient with small-bowel adenocarcinoma and 2 symptomatic patients with persistent villous atrophy despite strict GFD, therefore classified as refractory CD type 1) and a 0.3% mortality for CD-related complications, these data being similar to those reported in other studies (23). Moreover, no differences in the incidence of autoimmune diseases were noticed among patients with different dietary behaviours, with a general prevalence comparable with the literature (24,25). Despite the epidemiological data suggesting the efficacy of GFD against the development of autoimmune disorders and CD complications, such as refractory CD and intestinal lympoma (26–28), some authors question the association between incidence of autoimmune disorders or CD complications and gluten exposure (24,29–31). One should note that the results of this study are possibly partly limited by the fact that our cohort consisted primarily in patients with a long history of CD, but also a long history of GFD, while the literature data seem to infer a connection between the development of CD complications and the duration of undiagnosed CD (32,33).
Further interesting results have emerged from the analysis of the questionnaires administered to patients and their caregivers. The correlation observed between the scores of caregivers and patients confirms how crucial the role of the caregivers of paediatric coeliac patients is when it comes to adherence-to-treatment (34,35). As expected, the patients reporting a lifelong good adherence to GFD scored higher in both sections of the questionnaire as compared to the patients reporting a poor dietary adherence. On the other hand, the scores for the caregivers did not differ significantly between the different groups, probably because the poor adherence to GFD by the patients began after the childhood period, confirming the hypothesis that adherence to GFD may be partly lost during transition to adulthood (36).
To our knowledge, this is the first study to investigate the clinical history of CD over a follow-up period of more than 30 years with a medium-sized cohort of patients referred to a single tertiary centre. Other studies reporting data from long follow-up series included patients with a median of 10 to 20 years of CD history (9,18,37). Several limitations must be taken into account. First of all, the retrospective design of the study may have affected the results, especially on consideration of the proportion of patients who could not be reached during data collection. The possible bias as regards overall prognosis and complications was overcome by accessing Italy's national Health Service central register to ascertain the number and causes of deaths. This source allowed us to determine the overall mortality for our group of patients which was unexpectedly lower than for the sex- and age-matched general population. This result could be partially explained by the healthier life style and greater compliance to screening and surveillance programs in patients with an active disease, as recently suggested as an explanation for better life expectancy than the general population in a study on patients with in-situ ductal breast carcinoma (38).
Moreover, as regards the investigation on GFD rules the non-optimal rate of response to the questionnaires could affect the reliability of the results. It is also important to highlight that our questionnaire, was not previously validated. In fact, at the time of our study a validated questionnaire to survey GFD knowledge was unavailable, but we considered important to evaluate it more specifically in patients with a long history of CD. For this reason, we referred to the recommendations provided by the Italian Coeliac Association (Associazione Italiana Celiachia, AIC) to design our questionnaire. Our questionnaire, however, overlaps for a good part with that recently used by a team of Canadian gastroenterologists for the same purposes (39).
In conclusion, the availability of ample clinical data from a cohort of CD patients and the opportunity of interviewing a reasonable percentage of them has enabled us to draw a picture of the long-term history of diagnosed CD patients, which showed a low rate of evolution towards CD complications and a surprisingly high percentage of patients with a possible CD remission in spite of a gluten ingestion. The results of the questionnaire survey confirmed the need for nutritional counselling to young people in their transition into adulthood, a period where adherence to GFD could be lost. Another important point to take into consideration is that the access to original clinical records is fundamental during follow-up, especially for patients with a long history of CD. Considering the difficulty of designing long-term prospective studies, further data to confirm our observations can be gathered by carrying out a large multicentre study or establishing a national registry.
1. Rubio-Tapia A, Murray JA. Celiac disease. Curr Opin Gastroenterol
2. Bai JC, Fried M, Corazza GR, et al. World Gastroenterology Organisation global guidelines on celiac disease. J Clin Gastroenterol
3. Catassi C, Bearzi I, Holmes GK. Association of celiac disease and intestinal lymphomas and other cancers. Gastroenterology
2005; 128 (4 suppl 1):S79–86.
4. Ciacci C, Cirillo M, Cavallaro R, et al. Long-term follow-up of celiac adults on gluten-free diet: prevalence and correlates of intestinal damage. Digestion
5. Smedby KE, Akerman M, Hildebrand H, et al. Malignant lymphomas in coeliac disease: evidence of increased risks for lymphoma types other than enteropathy-type T cell lymphoma. Gut
6. Lebwohl B, Granath F, Ekbom A, et al. Mucosal healing and risk for lymphoproliferative malignancy in celiac disease: a population-based cohort study. Ann Intern Med
7. Errichiello S, Esposito O, Di Mase R, et al. Celiac disease: predictors of compliance with a gluten-free diet in adolescents and young adults. J Pediatr Gastroenterol Nutr
8. Elli L, Maieron R, Martelossi S, et al. Transition of gastroenterological patients from paediatric to adult care: a position statement by the Italian Societies of Gastroenterology. Dig Liver Dis
9. Hopman EG, von Blomberg ME, Batstra MR, et al. Gluten tolerance in adult patients with celiac disease 20 years after diagnosis? Eur J Gastroenterol Hepatol
10. Ludvigsson JF, Leffler DA, Bai JC, et al. The Oslo definitions for coeliac disease and related terms. Gut
11. McNeish AS, Harms HK, Rey J, et al. The diagnosis of coeliac disease. A commentary on the current practices of members of the European Society for Paediatric Gastroenterology and Nutrition (ESPGAN). Arch Dis Child
12. Revised criteria for diagnosis of coeliac disease. Report of Working Group of European Society of Paediatric Gastroenterology and Nutrition. Arch Dis Child
13. Oberhuber G, Granditsch G, Vogelsang H. The histopathology of coeliac disease: time for a standardized report scheme for pathologists. Eur J Gastroenterol Hepatol
14. Rubio-Tapia A, Rahim MW, See JA, et al. Mucosal recovery and mortality in adults with celiac disease after treatment with a gluten-free diet. Am J Gastroenterol
15. Bardella MT, Velio P, Cesana BM, et al. Coeliac disease: a histological follow-up study. Histopathology
16. Lee SK, Lo W, Memeo L, et al. Duodenal histology in patients with celiac disease after treatment with a gluten-free diet. Gastrointest Endosc
17. Lanzini A, Lanzarotto F, Villanacci V, et al. Complete recovery of intestinal mucosa occurs very rarely in adult coeliac patients despite adherence to gluten-free diet. Aliment Pharmacol Ther
18. Matysiak-Budnik T, Malamut G, de Serre NP, et al. Long-term follow-up of 61 coeliac patients diagnosed in childhood: evolution toward latency is possible on a normal diet. Gut
19. Mäki M, Lähdeaho ML, Hällström O, et al. Postpubertal gluten challenge in coeliac disease. Arch Dis Child
20. Lähdeaho ML, Mäki M, Laurila K, et al. Small-bowel mucosal changes and antibody responses after low- and moderate-dose gluten challenge in celiac disease. BMC Gastroenterol
21. Auricchio R, Tosco A, Piccolo E, et al. Potential celiac children: 9-year follow-up on a gluten-containing diet. Am J Gastroenterol
22. Elli L, Zini E, Tomba C, et al. Histological evaluation of duodenal biopsies from coeliac patients: the need for different grading criteria during follow-up. BMC Gastroenterol
23. Biagi F, Gobbi P, Marchese A, et al. Low incidence but poor prognosis of complicated coeliac disease: a retrospective multicentre study. Dig Liver Dis
24. Sategna Guidetti C, Solerio E, Scaglione N, et al. Duration of gluten exposure in adult coeliac disease does not correlate with the risk for autoimmune disorders. Gut
25. Elli L, Bonura A, Garavaglia D, et al. Immunological comorbity in coeliac disease: associations, risk factors and clinical implications. J Clin Immunol
26. Ludvigsson JF. Mortality and malignancy in celiac disease. Gastrointest Endosc Clin N Am
27. West J, Logan RF, Smith CJ, et al. Malignancy and mortality in people with coeliac disease: population based cohort study. BMJ
28. Corrao G, Corazza GR, Bagnardi V, et al. Mortality in patients with coeliac disease and their relatives: a cohort study. Lancet
29. Metso S, Hyytiä-Ilmonen H, Kaukinen K, et al. Gluten-free diet and autoimmune thyroiditis in patients with celiac disease. A prospective controlled study. Scand J Gastroenterol
30. Lohi S, Mäki M, Montonen J, et al. Malignancies in cases with screening-identified evidence of coeliac disease: a long-term population-based cohort study. Gut
31. Olén O, Askling J, Ludvigsson JF, et al. Coeliac disease characteristics, compliance to a gluten free diet and risk of lymphoma by subtype. Dig Liver Dis
32. Ventura A, Magazzù G, Greco L. Duration of exposure to gluten and risk for autoimmune disorders in patients with celiac disease. SIGEP Study Group for Autoimmune Disorders in Celiac Disease. Gastroenterology
33. Rubio-Tapia A, Kyle RA, Kaplan EL, et al. Increased prevalence and mortality in undiagnosed celiac disease. Gastroenterology
34. Anson O, Weizman Z, Zeevi N. Celiac disease: parental knowledge and attitudes of dietary compliance. Pediatrics
35. Charalampopoulos D, Panayiotou J, Chouliaras G, et al. Determinants of adherence to gluten-free diet in Greek children with coeliac disease: a cross-sectional study. Eur J Clin Nutr
36. O’Leary C, Wieneke P, Healy M, et al. Celiac disease and the transition from childhood to adulthood: a 28-year follow-up. Am J Gastroenterol
37. Hære P, Høie O, Schulz T, et al. Long-term mucosal recovery and healing in celiac disease is the rule—not the exception. Scand J Gastroenterol
38. Elshof LE, Schmidt MK, Rutgers EJ, et al. Cause-specific mortality in a population-based cohort of 9799 women treated for ductal carcinoma in situ. Ann Surg
2017; doi: 10.1097/SLA.0000000000002239. [Epub ahead of print].
39. Silvester JA, Weiten D, Graff LA, et al. Is it gluten-free? Relationship between self-reported gluten-free diet adherence and knowledge of gluten content of foods. Nutrition