Serologic Testing for Celiac Disease: Primum Non Nocere!
Article Outline
Abbreviations: BMI, Body mass index, CD, Celiac disease, NIH, National Institutes of Health, NASPGHAN, North American Society for Pediatric Gastroenterology, Hepatology and Nutrition, GFD, Gluten-free diet, TG, Transglutaminase, T1D, Type 1 diabetes
The decision to place a child on a strict gluten-free diet (GFD) for life should not be taken lightly. For the family, it involves a modification of their daily lives with significant additional expense. For the children, it has potential quality-of-life implications, setting them apart from their peers and forever impacting their social interactions. In symptomatic children with histologically confirmed celiac disease (CD), the benefits of maintaining a strict GFD are irrefutable. In the short term, there will be complete symptom resolution with repair of the intestinal mucosal damage. In the long term, there should be normal growth and development and the same expectations for health as in the general population. In contrast, the benefits of maintaining a GFD for those who are asymptomatic and identified with CD through a serological screening program are less clear.
See related article, p 461
There are certain groups of persons at increased risk for CD.1, 2 These include first-degree family members of an index case and those with specific autoimmune and non-autoimmune conditions. Of the autoimmune conditions associated with CD, type 1 diabetes (T1D) is best known, with up to 12% being affected. Many persons with CD in at-risk groups are asymptomatic at the time of diagnosis. Such cases were initially identified during studies using serological tests to screen for the condition, and, consequently, it has become common practice to test all persons in at-risk groups, even if they are asymptomatic. The North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) advocates such a policy in their Clinical Practice Guidelines on CD.1 The National Institutes of Health (NIH) Consensus Statement on CD also recommends screening asymptomatic at-risk persons but excluded those with T1D3 because there are no data to show that treatment of those with diabetes with asymptomatic CD has any benefit in the short term. Neither NASPGHAN nor the NIH advise mass screening of the general population for CD, even though as many as 1% may be affected, and this approach would probably not be cost effective.
The major reason for recommending testing of asymptomatic persons stems from the belief that early identification and treatment of CD will prevent long-term adverse health consequences. These include an excess in mortality rate, increased rates of malignancies (particularly intestinal lymphoma) and osteoporosis, and possibly onset of other autoimmune diseases.2, 4, 5, 6 In children, there is concern that delays in treatment may permanently stunt growth. Based on standardized mortality rates, there is an excess of deaths associated with CD.4 However, this seems confined to those with severe symptoms of malabsorption who had delayed diagnosis and poor compliance with treatment.4 Persons with mild symptoms or symptomless CD had no excess mortality.4 Intestinal malignancies are relatively more common in those with untreated CD, but recent data suggest the relative risk for such malignancies is lower than initially reported.5, 6 In terms of absolute numbers, intestinal lymphomas are relatively rare and account for a small proportion of all malignancies.5 Most malignancies in persons with CD occur in those with symptomatic disease who remain untreated, and there are little data demonstrating asymptomatic persons are at increased risk for cancer. Similarly, many patients with symptomatic CD have decreased bone mineralization at the time of diagnosis, but there are little data on the prevalence of osteoporosis in asymptomatic persons. Finally, the data suggesting that early diagnosis and treatment of CD can prevent onset of other autoimmune diseases are relatively weak.6 Although this is an attractive theory, there is need for more robust data before such a concept can be accepted.
The study by Simmons et al reported in this issue of The Journal attempts to address the need for testing children with T1D.8 By looking for antibodies to tissue transglutaminase (TG) in children with T1D they identified 71 TG+ subjects and compared them with 63 TG− controls matched for age, sex, and duration of diabetes. TG+ subjects were advised to undergo small intestinal biopsy to confirm the diagnosis of CD. The groups were compared for height, weight, body mass index (BMI), triceps skinfold thickness, and midarm circumference using age- and sex-specific Z scores. Additional comparisons included measures of bone mineral density and markers of bone turnover rates, HbA1C levels, records of insulin requirements, hypoglycemic events, and thyroid function tests. A subgroup analysis for these measurements compared those who agreed to have a biopsy and had characteristic histologic features of CD (n = 35) with TG− subjects matched for age, sex, and duration of diabetes. A symptom questionnaire was administered to all subjects.
TG+ subjects had lower weight, BMI, and midarm circumferences Z scores but no difference for height and skinfold thickness scores. Comparing those with biopsy-confirmed CD with TG− subjects; only the BMI scores were lower in the TG+ group, and there was no longer a difference in weight and midarm circumference scores. The groups were similar for bone mineral density, but the TG+ subjects had increased parathyroid hormone and urine n-telopeptides levels suggesting possible increased bone resorption. TG+ subjects had lower levels of free T4 and thyroid stimulating hormone, but the clinical significance of this finding is uncertain. There was no identifiable difference between the two groups for glucose control or frequency of hypoglycemic episodes. TG+ subjects reported an increased frequency of symptoms, but this finding is open to bias as the questionnaire was administered only after the subjects were identified as TG+ or TG−. There was no comparison between truly asymptomatic TG+ and TG− subjects.
Does this study clarify whether all children with T1D should be screened for CD? Unfortunately not, and the authors acknowledge this fact. On the other hand, the results also do not permit us to renounce the recommendations of NASPGHAN and other authorities in the field. In their discussion, the authors note that available data on growth effects of CD in children with T1D are conflicting, as are those relating to impact of CD on glycemic control.9, 10 One study has shown improved growth rates on a GFD over a 2-year period in children with T1D and CD.11 However, most of the subjects with CD in this report had symptoms before starting treatment, and therefore the results may not apply to the truly asymptomatic person. There are some who question the benefits of screening all children with T1D for CD.12 First, by labeling them with a second chronic illness requiring another significant lifelong diet change, we may be imposing an unnecessary psychological burden on some children. Second, because asymptomatic children identified through screening programs are less likely to be compliant with the GFD, subjecting them to an intestinal biopsy may place them at unnecessary risk. Similar concerns regarding the benefit of CD screening in children with Down syndrome have also been raised recently.13
As physicians, we constantly make treatment decisions that have the potential to significantly impact the lives of our patients. In doing so, we weigh the potential benefits against the possible risks. When the benefits clearly outweigh the risks our treatment decisions are justified. This approach embodies the principle of primum non nocere—first do no harm. With this in mind, perhaps it is time to reconsider the recommendations to screen asymptomatic persons for CD. We simply do not yet know the natural history of CD in such cases. Prospective studies comparing asymptomatic, screening-identified patients who maintain a strict GFD with those who do not are much needed. In addition, solid data are essential to justify the need for a strict lifelong GFD and to convince our patients to comply. A recent review and medical position statement issued by the American Gastroenterology Association (AGA) does not recommend routine testing of asymptomatic persons belonging to an at-risk group.14 Instead, the American Gastroenterology Association recommends testing when there are symptoms of CD. Physicians need to appreciate the variable clinical manifestations and know the conditions associated with CD. Testing those belonging to an at-risk group only when they have symptoms is a reasonable consideration until there is good evidence to recommend otherwise.7
References
- Guideline for the diagnosis and treatment of celiac disease in children: recommendations of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. J Pediatr Gastroenterol Nutr. 2005;40:1–19
- . Current approaches to diagnosis and treatment of celiac disease: an evolving spectrum. Gastroenterology. 2001;120:636–651
- National Institutes of Health Consensus Development Conference Statement on Celiac Disease, June 28-30, 2004. Gastroenterology. 2005;128:S1–S9
- Mortality in patients with coeliac disease and their relatives: a cohort study. Lancet. 2001;358:356–361
- Risk of non Hodgkin lymphoma in celiac disease. JAMA. 2002;287:1413–1419
- Ventura A, Magazzu G, Greco L. Duration of exposure to gluten and risk for autoimmune disorders in patients with celiac disease. Gastroenterology 1199;117:297-303.
- . Malignancy in coeliac disease: effect of a gluten free diet. Gut. 1989;30:333–338
- Impact of celiac autoimmunity on children with type 1 diabetes. J Pediat. 2007;150:461–466
- . Anthropometry, metabolic control, and thyroid autoimmunity in type 1 diabetes with celiac disease: a multicenter survey. J Pediatr. 2004;145:790–795
- . Coeliac disease in children and adolescents with type 1 diabetes: a study of growth, glycemic control, and experiences of families. Acta Paediatr. 2002;91:297–302
- Clinical benefit of a gluten free-diet in type 1 diabetic children with screening-detected celiac disease. Diabetes Care. 2006;29:2452–2456
- . Screening for celiac disease in children with type 1 diabetes (Two views of the controversy). Diabetes Care. 2003;26:1932–1939
- . Screening for celiac disease in asymptomatic children with Down syndrome: cost-effectiveness of preventing lymphoma. Pediatrics. 2006;118:816–817
- . American Gastroenterology Association (AGA) Institute Technical Review on the Diagnosis and Management of Celiac Disease. 2006;131:1981–2002
PII: S0022-3476(07)00232-6
doi:10.1016/j.jpeds.2007.02.061
© 2007 Mosby, Inc. All rights reserved.
Refers to article:
- Impact of Celiac Autoimmunity on Children with Type 1 Diabetes
