ORIGINAL ARTICLE: GASTROENTEROLOGY
Celiac Disease Alone and Associated With Type 1 Diabetes Mellitus
Alexandra Tsouka, Farid H. Mahmud, and yMargaret A. Marcon
ABSTRACT Objectives: The aim of the study was to evaluate complication screening and follow-up patterns in a population with type 1 diabetes mellitus and celiac disease (T1D/CD) in relation to a matched cohort with celiac disease (CD) alone at our center. Methods: We retrospectively reviewed the health charts of 41 children with T1D and biopsy-proven CD and compared anthropometrics and complication of screening within 2 years from CD diagnosis. Follow-up patterns were determined 3 years postdiagnosis. This population was then compared with a population with both symptomatic and asymptomatic CD matched for age and sex. Results: In comparison with T1D/CD, patients with CD alone had a significantly lower height, weight and body mass index (BMI z score 0.01 vs 0.81, P ¼ 0.001) and higher rates of screening for anemia (95% vs 71%, P ¼ 0.003) and bone health (49% vs 29%, P ¼ 0.05). Minimal rates of laboratory abnormalities were observed in either group, irrespective of symptoms at presentation, but CD alone more often presented with anemia than T1D/CD. Repeat serology testing was significantly more frequently performed in the T1D/CD group. Follow-up was equally variable with a median of 3 (range 0–4) visits with a gastroenterologist during the first 3 years postdiagnosis. Conclusions: These results suggest that patients with T1D/CD represent a distinct and possibly milder phenotype from CD alone. Complication screening was variable and negative for the majority of the patients. Guidelines for follow-up may need to be tailored to specific groups to standardize evaluation and complication screening, especially with regard to bone health. Key Words: celiac disease, pediatrics, type 1 diabetes mellitus
(JPGN 2015;61: 297–302)
C
eliac disease (CD) is an immune-mediated chronic disease characterized by intestinal inflammation that resolves with dietary exclusion of gluten. Because of autoimmune conditions, CD and type 1 diabetes mellitus (T1D) share a common genetic basis and CD is observed at a higher frequency in patients with T1D, with prevalence rates reported between 5% and 10% compared with 1% Received July 31, 2014; accepted March 10, 2015. From the Division of Endocrinology and Gastroenterology, and the yDivision of Hepatology and Nutrition, Department of Pediatrics, The Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada. Address correspondence and reprint requests to Margaret A. Marcon, MD, Division of Gastroenterology, Hepatology and Nutrition, Hospital for Sick Children, 555 University Ave, Toronto, Ontario M5G 1X8, Canada (e-mail: [email protected]). The authors report no conflict of interest. Copyright # 2015 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition DOI: 10.1097/MPG.0000000000000789
JPGN
Volume 61, Number 3, September 2015
What Is Known
Crohn disease (CD) occurs with increased frequency in type 1 diabetes mellitus (T1D). With the advent of serologic screening tests, children with T1D are being screened for CD. There are no guidelines for follow-up of children with both T1D and CD.
What Is New
Children diagnosed as having TID who develop CD may have a milder phenotype with better preserved growth at presentation. There may be a higher incidence of autoimmune thyroid disease in T1D/CD then TID alone. We propose an algorithm for follow-up of those with T1D/CD.
prevalence in the general population (1). As a consequence, many clinics have adopted a universal CD screening strategy in their diabetes population. Screening has also resulted in an increasing number of screen-positive but asymptomatic patients who lack ‘‘classic’’ symptoms that may include impaired growth, gastrointestinal symptoms, anemia, or unexplained hypoglycemia. This group of patients has not been well described with regard to complication screening and follow-up. This is relevant because this population is still prone to develop complications related to diabetes and has the potential to develop complications related to CD including anemia, vitamin deficiencies, and low bone mass (2–7). In the context of children with multiple autoimmune conditions, such as T1D and CD, care is often shared between pediatric subspecialties; in this case endocrinologists, who manage diabetesrelated complications and metabolic control, and gastroenterologists whose focus is on celiac-related complications such as anemia and liver dysfunction. Existing guidelines from pediatric gastroenterology and endocrinology societies do not include specific recommendations on the follow-up of patients with CD and T1D (8,9). This is highlighted by a survey of Canadian adult and pediatric gastroenterologists, which describes significant variability on practice with regard to complication screening including routine laboratory testing, nutritional evaluation, serology, and bone health assessment as well as clinical follow-up (10). The aim of this study was to describe our institutional experience of children with T1D and CD, in regards to rates and outcomes of complication screening, anthropometrics, and followup patterns in a population with T1D and CD and compare with a group with CD alone. We also compared the symptomatic subgroup
297
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
Marcon et al
JPGN
of the T1D/CD population to a population with symptomatic CD only and the asymptomatic subgroup of the T1D/CD population to a population with asymptomatic CD only (patients screened through family screening because a first-degree relative tested positive for CD). We assessed anthropomorphics, complication screening, and follow-up patterns to delineate whether these 2 groups of patients with CD, with or without T1D really are different with respect to their CD manifestations.
METHODS Between 2005 and 2011, 771 tissue transglutaminase (TTG) serology screens were sent from the Diabetes Clinic at the Hospital for Sick Children (SickKids). Forty-three screens were positive: 41 children underwent biopsy, which was confirmatory for CD, and 2 had undulating serology and were just studied. We retrospectively reviewed the health charts of 96 children: 41 with T1D and CD, 41 with CD alone, and 14 with asymptomatic CD, all diagnosed between 2005 and 2011 who were observed within the gastroenterology clinic (Fig. 1).
CD Comparison Groups The CD-alone group was randomly selected from a database of patients with biopsy-confirmed (Marsh II or higher) CD to match age, sex, and CD duration with a ratio of 1 control subject to 1 case subject. Exclusion criteria for matching were coexistent diagnosis of metabolic or other disorders affecting growth (growth hormone deficiency, Turner syndrome, trisomy 21, or Wilson syndrome) or diagnosis through screening or other significant health conditions. These individuals were all symptomatic. The second comparison population (asymptomatic CD only) was selected from a database of patients who underwent family screening for CD because of the diagnosis of another member and were found to have CD with positive serology and biopsy. All of these individuals were asymptomatic with regard to gastrointestinal symptoms or growth impairment at the time of diagnosis as assessed by review of diabetes clinician, dietitian, and gastrointestinal consultation notes. All of the patients were followed up in the gastroenterology clinic at SickKids. Celiac serology is assessed using anti-TTG immunoglobulin A (IgA) and immunoglobulin G antibodies (Celikey Tissue Transglutaminase IgA Antibody Assay, Phadia, Uppsala, Sweden [reference range: positive >8 U/mL]) and in the instance of a positive test, confirmatory endomysial antibody testing is performed on the same sample. Serum IgA was assessed in subjects to rule out IgA deficiency and minimize false-negative results. Patients with positive serology were referred to a pediatric gastroenterologist for consultation and biopsy. Biopsy specimens are classified according to the Marsh-Oberhuber classification (11) (Marsh score >II is consistent with CD). Only patients who were biopsy-positive for CD were included in the present review.
Data Collected For all of the patients identified, the following data were collected: sex and age at diagnosis of CD, weight, height, and body mass index (BMI) at the first gastroenterology visit and 1 year later, symptoms at diagnosis including abdominal pain, bloating, diarrhea, constipation, vomiting and, acid reflux. Thyroid function tests and family history of CD, gluten-related dermatitis herpetiformis and inflammatory bowel disease were also reviewed. Laboratory testing at diagnosis and during the 2 years following the CD diagnosis included anti-TTG IgA and immunoglobulin G and endomysial antibody antibodies and serum IgA,
298
Volume 61, Number 3, September 2015
hemoglobin, iron, ferritin, liver panel (alanine transaminase, aspartate transaminase, alkaline phosphatase), albumin. Bone health laboratory tests included calcium, vitamin D, parathyroid hormone, dual-energy x-ray absorptiometry, thyroid function tests (thyroid-stimulating hormone, T4), small intestine biopsy results. Adherence to a gluten-free diet (GFD) as reported in the clinic notes by a registered dietician with CD and GFD-related experience during follow-up visits and/or follow-up TTG serology using food records and normalization of follow-up testing of TTG IgA. Anemia screening was considered abnormal for boys and girls of age
Celiac Disease Alone and Associated With Type 1 Diabetes Mellitus
Alexandra Tsouka, Farid H. Mahmud, and yMargaret A. Marcon
ABSTRACT Objectives: The aim of the study was to evaluate complication screening and follow-up patterns in a population with type 1 diabetes mellitus and celiac disease (T1D/CD) in relation to a matched cohort with celiac disease (CD) alone at our center. Methods: We retrospectively reviewed the health charts of 41 children with T1D and biopsy-proven CD and compared anthropometrics and complication of screening within 2 years from CD diagnosis. Follow-up patterns were determined 3 years postdiagnosis. This population was then compared with a population with both symptomatic and asymptomatic CD matched for age and sex. Results: In comparison with T1D/CD, patients with CD alone had a significantly lower height, weight and body mass index (BMI z score 0.01 vs 0.81, P ¼ 0.001) and higher rates of screening for anemia (95% vs 71%, P ¼ 0.003) and bone health (49% vs 29%, P ¼ 0.05). Minimal rates of laboratory abnormalities were observed in either group, irrespective of symptoms at presentation, but CD alone more often presented with anemia than T1D/CD. Repeat serology testing was significantly more frequently performed in the T1D/CD group. Follow-up was equally variable with a median of 3 (range 0–4) visits with a gastroenterologist during the first 3 years postdiagnosis. Conclusions: These results suggest that patients with T1D/CD represent a distinct and possibly milder phenotype from CD alone. Complication screening was variable and negative for the majority of the patients. Guidelines for follow-up may need to be tailored to specific groups to standardize evaluation and complication screening, especially with regard to bone health. Key Words: celiac disease, pediatrics, type 1 diabetes mellitus
(JPGN 2015;61: 297–302)
C
eliac disease (CD) is an immune-mediated chronic disease characterized by intestinal inflammation that resolves with dietary exclusion of gluten. Because of autoimmune conditions, CD and type 1 diabetes mellitus (T1D) share a common genetic basis and CD is observed at a higher frequency in patients with T1D, with prevalence rates reported between 5% and 10% compared with 1% Received July 31, 2014; accepted March 10, 2015. From the Division of Endocrinology and Gastroenterology, and the yDivision of Hepatology and Nutrition, Department of Pediatrics, The Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada. Address correspondence and reprint requests to Margaret A. Marcon, MD, Division of Gastroenterology, Hepatology and Nutrition, Hospital for Sick Children, 555 University Ave, Toronto, Ontario M5G 1X8, Canada (e-mail: [email protected]). The authors report no conflict of interest. Copyright # 2015 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition DOI: 10.1097/MPG.0000000000000789
JPGN
Volume 61, Number 3, September 2015
What Is Known
Crohn disease (CD) occurs with increased frequency in type 1 diabetes mellitus (T1D). With the advent of serologic screening tests, children with T1D are being screened for CD. There are no guidelines for follow-up of children with both T1D and CD.
What Is New
Children diagnosed as having TID who develop CD may have a milder phenotype with better preserved growth at presentation. There may be a higher incidence of autoimmune thyroid disease in T1D/CD then TID alone. We propose an algorithm for follow-up of those with T1D/CD.
prevalence in the general population (1). As a consequence, many clinics have adopted a universal CD screening strategy in their diabetes population. Screening has also resulted in an increasing number of screen-positive but asymptomatic patients who lack ‘‘classic’’ symptoms that may include impaired growth, gastrointestinal symptoms, anemia, or unexplained hypoglycemia. This group of patients has not been well described with regard to complication screening and follow-up. This is relevant because this population is still prone to develop complications related to diabetes and has the potential to develop complications related to CD including anemia, vitamin deficiencies, and low bone mass (2–7). In the context of children with multiple autoimmune conditions, such as T1D and CD, care is often shared between pediatric subspecialties; in this case endocrinologists, who manage diabetesrelated complications and metabolic control, and gastroenterologists whose focus is on celiac-related complications such as anemia and liver dysfunction. Existing guidelines from pediatric gastroenterology and endocrinology societies do not include specific recommendations on the follow-up of patients with CD and T1D (8,9). This is highlighted by a survey of Canadian adult and pediatric gastroenterologists, which describes significant variability on practice with regard to complication screening including routine laboratory testing, nutritional evaluation, serology, and bone health assessment as well as clinical follow-up (10). The aim of this study was to describe our institutional experience of children with T1D and CD, in regards to rates and outcomes of complication screening, anthropometrics, and followup patterns in a population with T1D and CD and compare with a group with CD alone. We also compared the symptomatic subgroup
297
Copyright 2015 by ESPGHAN and NASPGHAN. Unauthorized reproduction of this article is prohibited.
Marcon et al
JPGN
of the T1D/CD population to a population with symptomatic CD only and the asymptomatic subgroup of the T1D/CD population to a population with asymptomatic CD only (patients screened through family screening because a first-degree relative tested positive for CD). We assessed anthropomorphics, complication screening, and follow-up patterns to delineate whether these 2 groups of patients with CD, with or without T1D really are different with respect to their CD manifestations.
METHODS Between 2005 and 2011, 771 tissue transglutaminase (TTG) serology screens were sent from the Diabetes Clinic at the Hospital for Sick Children (SickKids). Forty-three screens were positive: 41 children underwent biopsy, which was confirmatory for CD, and 2 had undulating serology and were just studied. We retrospectively reviewed the health charts of 96 children: 41 with T1D and CD, 41 with CD alone, and 14 with asymptomatic CD, all diagnosed between 2005 and 2011 who were observed within the gastroenterology clinic (Fig. 1).
CD Comparison Groups The CD-alone group was randomly selected from a database of patients with biopsy-confirmed (Marsh II or higher) CD to match age, sex, and CD duration with a ratio of 1 control subject to 1 case subject. Exclusion criteria for matching were coexistent diagnosis of metabolic or other disorders affecting growth (growth hormone deficiency, Turner syndrome, trisomy 21, or Wilson syndrome) or diagnosis through screening or other significant health conditions. These individuals were all symptomatic. The second comparison population (asymptomatic CD only) was selected from a database of patients who underwent family screening for CD because of the diagnosis of another member and were found to have CD with positive serology and biopsy. All of these individuals were asymptomatic with regard to gastrointestinal symptoms or growth impairment at the time of diagnosis as assessed by review of diabetes clinician, dietitian, and gastrointestinal consultation notes. All of the patients were followed up in the gastroenterology clinic at SickKids. Celiac serology is assessed using anti-TTG immunoglobulin A (IgA) and immunoglobulin G antibodies (Celikey Tissue Transglutaminase IgA Antibody Assay, Phadia, Uppsala, Sweden [reference range: positive >8 U/mL]) and in the instance of a positive test, confirmatory endomysial antibody testing is performed on the same sample. Serum IgA was assessed in subjects to rule out IgA deficiency and minimize false-negative results. Patients with positive serology were referred to a pediatric gastroenterologist for consultation and biopsy. Biopsy specimens are classified according to the Marsh-Oberhuber classification (11) (Marsh score >II is consistent with CD). Only patients who were biopsy-positive for CD were included in the present review.
Data Collected For all of the patients identified, the following data were collected: sex and age at diagnosis of CD, weight, height, and body mass index (BMI) at the first gastroenterology visit and 1 year later, symptoms at diagnosis including abdominal pain, bloating, diarrhea, constipation, vomiting and, acid reflux. Thyroid function tests and family history of CD, gluten-related dermatitis herpetiformis and inflammatory bowel disease were also reviewed. Laboratory testing at diagnosis and during the 2 years following the CD diagnosis included anti-TTG IgA and immunoglobulin G and endomysial antibody antibodies and serum IgA,
298
Volume 61, Number 3, September 2015
hemoglobin, iron, ferritin, liver panel (alanine transaminase, aspartate transaminase, alkaline phosphatase), albumin. Bone health laboratory tests included calcium, vitamin D, parathyroid hormone, dual-energy x-ray absorptiometry, thyroid function tests (thyroid-stimulating hormone, T4), small intestine biopsy results. Adherence to a gluten-free diet (GFD) as reported in the clinic notes by a registered dietician with CD and GFD-related experience during follow-up visits and/or follow-up TTG serology using food records and normalization of follow-up testing of TTG IgA. Anemia screening was considered abnormal for boys and girls of age
