The Frequency of Undiagnosed Celiac Disease in Youth with Type 1 Diabetes and Its Association with Diabetic Retinopathy: The SEARCH for Diabetes in Youth Study

Brady, Ryan P.; Jensen, Elizabeth T.; Rigdon, Joseph; Crimmins, Nancy A.; Mallon, Daniel; Dolan, Lawrence M.; Imperatore, Giuseppina; Kahkoska, Anna R.; Mottl, Amy K.; Honor, Ann; Pettitt, David J.; Merjaneh, Lina; Dabelea, Dana; Shah, Amy S.

doi:https://doi.org/10.1155/2023/9038795

Pediatric Diabetes

On this page

Abstract Introduction Materials and Methods Results Discussion Data Availability Disclosure Conflicts of Interest Authors’ Contributions Acknowledgments Supplementary Materials References Copyright Related Articles

Research Article | Open Access

Volume 2023 | Article ID 9038795 | https://doi.org/10.1155/2023/9038795

The Frequency of Undiagnosed Celiac Disease in Youth with Type 1 Diabetes and Its Association with Diabetic Retinopathy: The SEARCH for Diabetes in Youth Study

Ryan P. Brady,¹Elizabeth T. Jensen,²Joseph Rigdon,³Nancy A. Crimmins,¹Daniel Mallon,¹Lawrence M. Dolan,¹Giuseppina Imperatore,⁴Anna R. Kahkoska,⁵Amy K. Mottl,⁶Ann Honor,⁷David J. Pettitt,⁸and Lina Merjaneh⁹ et al.

Academic Editor: Farid H. Mahmud

Received05 Dec 2022

Revised07 Feb 2023

Accepted08 May 2023

Published29 May 2023

Abstract

Aims. Celiac disease (CD) in adults with type 1 diabetes has been associated with increased cardiovascular risk and the earlier occurrence of diabetes-associated complications. In the Search for Diabetes in Youth study, we aimed to assess the frequency of CD and the potential for undiagnosed CD among youth with childhood onset type 1 diabetes. In addition, we assessed the burden of cardiovascular risk factors and diabetes-associated complications in youth with type 1 diabetes by CD status and IgA tissue transglutaminase autoantibody (tTGA) levels. Methods. 2,444 youths with type 1 diabetes completed a CD questionnaire and underwent tTGA testing. Integrating the celiac disease questionnaire and tTGA results for this cross-sectional analysis, participants were categorized as follows: (1) reported CD; (2) seropositive for CD (no reported CD and seropositive tTGA); and (3) type 1 diabetes only (comparison group: no reported CD and seronegative tTGA). Subanalyses were performed on those with no reported CD and tTGA ≥10x ULN, designated potentially undiagnosed CD. Cardiovascular risk factors and diabetes-associated complications were evaluated by CD status and tTGA levels utilizing a Poisson model to estimate relative risk. Results. Reported CD in youths with type 1 diabetes was 7%. Seropositivity for tTGA with no reported CD was present in 4%, and 1.2% had potentially undiagnosed CD. Youths with potentially undiagnosed CD had a 2.69x higher risk of diabetic retinopathy than comparison group. In addition, CD with tTGA <0.05 (controlled CD) was associated with lower HbA1c. Conclusions. Undiagnosed CD is likely present in youths with type 1 diabetes and potentially undiagnosed CD is associated with a higher risk of diabetic retinopathy. These findings indicate the importance of routine screening for CD in type 1 diabetes in youths.

1. Introduction

Celiac disease (CD) is a chronic and inflammatory enteropathy triggered by the ingestion of gluten in genetically predisposed individuals [1]. CD and type 1 diabetes are both autoimmune diseases with shared genetic risk [2], and their association has been well described [3, 4]. While routine screening guidelines for CD in youth with type 1 diabetes have been published [5], given the often asymptomatic nature of CD in this population [6], it has the potential to remain undiagnosed.

CD in adults with type 1 diabetes is associated with higher cardiovascular risk [7–10], the earlier occurrence of diabetes-associated complications [11], and increased mortality [12]. As both disease processes often begin in childhood [13] and have been increasing in prevalence over time [14, 15], there is concern cardiovascular risk factors and the antecedents of diabetes-associated complications may occur in youth. Early diagnosis of CD and subsequent interventions therefore may reduce future cardiovascular burden.

The objectives of this study were to assess the frequency and characteristics of CD and the potential for undiagnosed CD in youths with type 1 diabetes who were enrolled in the Search for Diabetes in Youth study, a large observational cohort of youths with childhood onset diabetes in the United States. In addition, we assessed the burden of cardiovascular risk factors and diabetes-associated complications by CD status and IgA tissue transglutaminase autoantibodies (tTGA) levels. Developing a better understanding of the frequency and added disease burden of CD, particularly undiagnosed CD, will help guide existing screening practices in youth with type 1 diabetes. Our hypotheses were that CD may be undiagnosed in the youth-onset type 1 diabetes population and that CD is associated with cardiovascular risk factors and diabetes-associated complications at a young age.

2. Materials and Methods

2.1. Study Participants

Participants included in this study were enrolled in the Search for Diabetes in Youth study, a multicenter study covering 5 geographically distinct areas (California, Colorado, Ohio, South Carolina, and Washington) of the United States investigating the prevalence, incidence, and complications in participants with youth-onset diabetes (onset before 20 years of age). Comprehensive details of the SEARCH study have been published [16]. Individuals included in this analysis had a diagnosis of type 1 diabetes based on medical record review or referring physician [16] and/or had at least one diabetes autoantibody detectable in their serum at the time of their baseline SEARCH study visit.

Participants for the current study met the following additional inclusion criteria: (1) a baseline search visit in 2002–2006 or 2008 [16]; (2) a follow-up study visit in the years 2011 to 2019; and (3) duration of diabetes >5 years at the follow-up visit. At the follow-up visit, participants were assessed for the presence of CD, cardiovascular risk factors, and diabetes-associated complications and comorbidities [16]. The study was reviewed and approved by the local institutional review boards at each of the 5 SEARCH study sites and all participants and their parents/guardians provided informed assent and/or consent.

2.2. Clinical, Demographic, and Biochemical Assessment

Medical history and demographic information were obtained through questionnaires completed by study participants and/or their parents/guardians. Presence of autoimmune conditions (possible answers were “yes,” “no,” and “do not know”) including hyperthyroidism, hypothyroidism, Addison’s disease, and vitiligo were ascertained by questionnaire. Additional autoimmune conditions (e.g., arthritis) were not ascertained at SEARCH study visits and therefore not available. Sex, race, and ethnicity were self-reported, and race and ethnicity were categorized as Hispanic, non-Hispanic Black (NHB), non-Hispanic White (NHW), or other. Resting systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured 3 times using an aneroid sphygmomanometer and an appropriately sized cuff by trained study staff. Participants were seated for 5 minutes prior to measurements and the average of the 3 measurements was used. Venipuncture was performed after at least an 8-hour overnight fast. Biochemical measurements of HbA1c and lipids were performed as described previously [16].

2.3. Celiac Disease Assessment

The presence of CD [5] in the SEARCH study was assessed using questionnaire response. The questionnaire asked whether a health care provider ever informed the participant that they had CD (possible answers were “yes,” “no,” and “do not know”). Those who answered, “do not know” (n = 27) were excluded from further analysis. tTGA were measured for this study utilizing stored serum samples from the SEARCH biospecimen repository on the remaining 2,444 participants with type 1 diabetes. The stored serum samples used to assess tTGA levels were collected at the time of the study follow-up visit and therefore coincide temporally with all data (questionnaire, diabetes complication, anthropometric, and biochemical data) reported here. Radiobinding assay using recombinant tissue transglutaminase protein was used to detect tTGA [17, 18]. tTGA assays were performed at the Barbara Davis Center Autoantibody Core Facility (Denver, CO, USA), which is a Clinical Laboratory Improvement Amendments certified laboratory and has been designated as a National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases North America Autoantibody Core Laboratory [19]. The assay result is expressed as an index with an assay range of −0.100 to 2.000. A tTGA level of ≥0.05 was used as the cut-off for a positive test (100th percentile of normal controls) [13, 17], and all levels >0.030 were repeated using the same assay to confirm a positive result.

2.4. Celiac Disease Status

First, we integrated the questionnaire response for reported CD and the laboratory measured tTGA level for all participants with type 1 diabetes to establish the following three mutually exclusive categories for analysis of CD status:(1)Individuals with reported CD: participants who answered “yes” to the CD question in the questionnaire, regardless of tTGA levels(2)Individuals seropositive for CD: participants who answered “no” to the CD question but had seropositive tTGA (levels ≥0.05)(3)Individuals with type 1 diabetes only (serving as comparison group): participants who answered “no” to CD question and had seronegative tTGA (levels <0.05)

Of the individuals categorized as seropositive for CD (abovementioned group 2), subanalysis was performed on those with tTGA ≥0.50 (≥10x ULN). These individuals, who had no reported CD, were designated for the purpose of this analysis as those with “potentially undiagnosed CD” based on two prior studies demonstrating that all individuals in their studies with tTGA levels ≥10 ULN had diagnosed CD after intestinal biopsy [20, 21]. In addition, using the same tTGA assay performed here, one study found a single tTGA level ≥10 ULN at the time of intestinal biopsy had a 96% positive predictive value (PPV) for positive intestinal biopsy specimen [22].

Second, we also sought to examine differences in cardiovascular risk factors among participants by tTGA levels. For this, four mutually exclusive categories were established:(1)Controlled CD: participants who answered “yes” to CD question and had seronegative tTGA levels (levels <0.05)(2)Intermediate tTGA elevation: all participants who had seropositive tTGA levels between ≥0.05 and <0.50, regardless of CD questionnaire response(3)High tTGA elevation: all participants with seropositive tTGA levels ≥0.50 (10x ULN), regardless of CD questionnaire response(4)Individuals with type 1 diabetes only (serving as comparison group): participants who answered “no” to CD question and had seronegative tTGA (levels <0.05)

2.5. Cardiovascular Risk Factors and Diabetes-Associated Complications

Blood pressure (SBP z-score, DBP z-score), lipids, (HDL-C, LDL-C, and triglycerides), and HbA1c were used as cardiovascular risk factor outcomes for this analysis [23] and assessed at SEARCH follow-up visit.

Diabetes-associated complications were defined by assessment at SEARCH visit.(1)Diabetic retinopathy was assessed via 45° color digital fundus images acquired with a nonmydriatic camera (Visucam Pro N, Carl Zeiss Meditech, Dublin, Ca). The Wisconsin Ocular Epidemiology Reading Center graded all images [24]. Retinopathy severity was based on the worse eye using established grading parameters and categorized as none, minimal nonproliferative, mild to moderate non-proliferative, and proliferative [25]. Diabetic retinopathy (yes/no) was then defined as having either mild, moderate, or proliferative retinopathy in at least one eye based on these grading parameters.(2)Peripheral neuropathy (yes/no) was defined as a score of >2 out of a total score of 8 on the Michigan Neuropathy Screening Instrument [26].(3)Cardiovascular autonomic dysfunction was assessed by heart rate variability utilizing the SphygmoCor-Vx device (AtCor Medical). Indices of heart rate variability derived from the electrocardiographic R-R intervals included: standard deviation (SD) of the intervals, root mean square differences of successive intervals, normalized high-frequency power, normalized low-frequency power, and the low-to-high frequency ratio [24]. Cardiovascular autonomic dysfunction (yes/no) was defined as participants having at minimum 3 of 5 heart rate variability abnormalities based on ≤5^th or ≥95^th percentile observed in age-and sex-matched control participants of the SEARCH Cardiovascular Disease Study [27].(4)Pulse wave velocity in the carotid to femoral segment (PWVf, continuous variable) was measured with the SphygmoCor device (AtCor Medical). The average of three recordings of PWVf was used, with higher PWVf indicating greater arterial stiffness [27].(5)Diabetic kidney disease (yes/no) was defined as the presence of albuminuria (≥30 μg/mg of creatinine) or an estimated glomerular filtration rate ≤60 mL/min/1.73 m² using the CKD-Epi equations with serum creatinine and cystatin C [28]. Samples were run on first-morning urine void samples, if not available, a spot sample was used [24].

2.6. Statistical Analysis

Data are presented as mean (SD) for continuous variables or frequency (%) for categorical variables. Basic demographic and clinical characteristics were compared between reported CD, seropositive for CD, and potentially undiagnosed CD to the comparison group (Tables 1). For frequencies, we used chi-square test for difference in proportions; Fishers exact test was used when expected counts were <5. For continuous measures we used a two-sample t-test.

To examine the relationship between CD status or tTGA levels with cardiovascular risk factors and diabetes-associated complications (Tables 2 and 3), complete case analysis was utilized on study outcomes and covariates to allow comparison across models and across unadjusted and adjusted estimates; except in the case of diabetic kidney disease, in which separate analysis was performed on the subset of the study population that had available urine samples. Linear regression models were utilized to estimate ß-coefficients for continuous outcomes (SBP, DBP, HDL-C, LDL-C, triglycerides, HbA1c, and PWVf) and a Poisson model with robust error variance was utilized to estimate relative risk (RR) for dichotomous outcomes (retinopathy, peripheral neuropathy, cardiovascular autonomic dysfunction, and diabetic kidney disease). For lipid outcomes (LDL-C, HDL-C, and triglycerides) adjusted models controlled for race and ethnicity and age at follow-up. For HbA1c and diabetes-associated complication outcomes, models controlled for race and ethnicity, age at follow-up, and age at diabetes diagnosis were adjusted. All models were adjusted for clinic site. These covariates were selected based on directed acyclic graph modeling [29] that were developed a priori based on the literature.

3. Results

The clinical characteristics of the 2,444 SEARCH individuals with type 1 diabetes included in this study stratified by CD status are presented in Table 1. The frequency of reported CD was 7%. Compared with participants with type 1 diabetes only (comparison group), participants with reported CD were younger at age of diabetes diagnosis (8.3 vs. 9.8 years, ), more likely to be female (61.3% vs. 48.8%, ) and differed regarding race and ethnicity (), with a higher percentage of those with reported CD consisting of NHW individuals. In addition, participants with reported CD had a lower BMI z-score (0.44 vs. 0.62, ) and SBP z-score (−0.84 vs. −0.66, ).

The frequency seropositive for CD (participants who did not report CD and had seropositive tTGA) was 4%. Compared with participants with type 1 diabetes only, participants seropositive for CD were younger at age of diabetes diagnosis (8.2 vs. 9.8 years, ), more likely to be female (52.5% vs. 48.8%, ) and had lower HDL-C (52.4 vs. 56.1 mg/dL, ).

Of these youth seropositive for tTGA with no reported CD, subanalysis was performed on those with tTGA ≥0.50 (≥10x ULN), suggesting potentially undiagnosed CD (n = 27). The frequency of tTGA seropositivity ≥10x ULN among those who did not report CD (n = 2444–168 = 2,276) was 1.2% (27/2,276). Of these participants with potentially undiagnosed CD (n = 27), twenty-three (n = 23) had complete data to perform complete case analysis. Participants with potentially undiagnosed CD (n = 23) had lower HDL-C (48.2 vs. 55.7 mg/dL, ) compared with participants with type 1 diabetes only.

We also examined coexisting autoimmune conditions amongst participants with type 1 diabetes by CD status (Supplementary Table (available here)). The frequency of hyperthyroidism, hypothyroidism, and Addison’s disease were greater than two-fold higher in individuals with reported CD compared to the comparison group, youth with type 1 diabetes only (all ). There were no differences in the frequency of autoimmune conditions between seropositive for CD and potentially undiagnosed CD compared to those with type 1 diabetes only.

We examined the relationship of CD status with cardiovascular risk factors and diabetes-associated complications (Table 2). In unadjusted analyses, the presence of reported CD was associated with lower SBP z-score (unadjusted β: −0.23 (95% Confidence Interval (CI): −0.41, −0.05)) and lower PWVf (unadjusted β −0.19 (95% CI: −0.40, −0.03)). After adjustment for covariates, neither association retained statistical significance. Reported CD was not significantly associated with any diabetes-associated complications. However, potentially undiagnosed CD was associated with higher risk of diabetic retinopathy in both unadjusted and adjusted models (adjusted RR 2.69 (95% CI: 1.66, 4.36)) compared with youth with type 1 diabetes only.

Lastly, we sought to determine if the tTGA levels correlated with cardiovascular risk factors. tTGA <0.05 in individuals with reported CD was associated with significantly lower HbA1c in both unadjusted and adjusted models (adjusted β: −0.55 (95% CI: −1.09, −0.02)) (Table 3). tTGA levels were not significantly associated with the remaining cardiovascular risk factors.

4. Discussion

This study describes the frequency and characteristics of reported CD and potentially undiagnosed CD and their relationship to cardiovascular risk factors and diabetes-associated complications in a large, geographically, and racial and ethnically diverse US population of youth with type 1 diabetes. In addition, this study assessed the relationship of tTGA levels with cardiovascular risk factors in youth with type 1 diabetes. Results of this study found that undiagnosed CD is likely present in youth with type 1 diabetes and those with potentially undiagnosed CD had a higher risk of diabetic retinopathy. Furthermore, individuals with CD and tTGA levels <0.05, suggesting controlled CD, are associated with lower HbA1c.

Youth with type 1 diabetes are at an increased risk for CD compared with the general population [3] in part due to shared HLA genetic risk [2]. The reported prevalence of CD in type 1 diabetes varies widely based on different data sources (1–16.4%) [6], likely secondary to studies performed in homogenous populations [30] and varying CD diagnostic criteria utilized [4, 6]. The largest prevalence study to date of CD in youth with type 1 diabetes was a cross-sectional study combining data from four registries (Registries from the US, Germany/Austria, the UK and Australia) and found a prevalence of 3.5% [4]. A systematic review of the prevalence of CD in children and adolescents with type 1 diabetes found a pooled prevalence of 5.1% [6]. In comparison, our findings demonstrated a 7% frequency of reported CD in youth with type 1 diabetes [4, 6]. Our data showing a greater risk of developing CD in youth diagnosed with diabetes at a younger age and that risk differs by race and ethnicity are consistent with prior work [4, 31]. Individuals with reported CD were more likely to be female than those with type 1 diabetes only, similar to the higher female prevalence of CD in the general population [30]. Reports on sex distribution of CD in youth with type 1 diabetes have been mixed [4, 32, 33], which may reflect the geographic, race, and ethnic makeup of the populations studied. The results reported here offer insight into the characteristics of CD in a multiethnic and race and geographically diverse adolescent type 1 diabetes population in the US.

Prevalence and epidemiologic studies likely underestimate CD in youth with type 1 diabetes since a majority of youth with CD is asymptomatic at the time of diagnosis [6] and diagnosis depends on local screening practices. In addition, some patients and families may choose not to undergo endoscopy, the gold standard for diagnosis, contributing to undiagnosed, and therefore unreported, CD in biopsy-confirmed prevalence studies. Our study design allowed for the unique opportunity to evaluate for potentially undiagnosed CD based on tTGA levels. Our data show 4% of our study population with type 1 diabetes had tTGA levels ≥0.05 indicating they were seropositive for CD without a reported CD diagnosis. While tTGA seropositivity does not indicate a true diagnosis of CD, as tTGA levels can be transiently elevated in individuals with type 1 diabetes and spontaneously normalize [34], it does raise the possibility that some youth with type 1 diabetes have undiagnosed CD. To this end, we found that amongst those with no reported diagnosis of CD, 1.2% had tTGA levels ≥10x ULN. To date, two published studies show individuals with tTGA levels ≥10x ULN all have diagnosed CD after intestinal biopsy in their study population [20, 21]. Furthermore, an additional study demonstrated a tTGA level ≥10x ULN (using the same assay as this study) at the time of biopsy had a 96% PPV for positive intestinal biopsy specimens [22]. In addition, all our study participants had diabetes duration >5 years at the time of tTGA assessment, ensuring tTGA elevation was not occurring at the time of diabetes diagnosis. Cases of potentially undiagnosed CD could be secondary to how study sites operationalize current CD screening recommendations for youth with type 1 diabetes [5]. Further studies are needed to explore potential barriers to CD screening and diagnosis in this population.

While the presence of diabetes-associated complications in individuals with type 1 diabetes and CD have been demonstrated in the adult and young adult populations [11, 35, 36], to our knowledge, no study to date has examined the risk of diabetes-associated complications in youth with type 1 diabetes by CD and potentially undiagnosed CD status. While reported CD was not found to significantly increase risk of diabetes-associated complications at this young age, individuals with potentially undiagnosed CD were found to have greater than 2.5× higher risk of diabetic retinopathy compared to those with type 1 diabetes only (comparison group with no reported CD and normal tTGA levels). Higher risk of diabetic retinopathy has been reported in multiple studies in adults with type 1 diabetes and CD [11, 35, 36]. Leeds et al. found that adults with type 1 diabetes and newly diagnosed CD had a higher prevalence of advanced retinopathy and nephropathy [35]. Rohrer et al. found the adjusted risk for both retinopathy and nephropathy were higher in patients with type 1 diabetes and CD versus those without CD, and both conditions occurred earlier [11]. The increased risk for the microvascular complication of diabetic retinopathy in individuals with type 1 diabetes with both diagnosed, and now potentially undiagnosed, CD suggests that the concurrent disease process of CD may negatively impact the microvasculature in individuals with type 1 diabetes. The mechanisms behind type 1 diabetes and CD on the microvasculature is unclear, though inflammation [8, 37, 38] has been suggested to play a role. In addition, the higher risk specifically of the microvascular complication of retinopathy, as demonstrated here in youth with potentially undiagnosed CD and in adults with CD and type 1 diabetes [11, 35], remains undetermined. A potential contributing factor specifically for retinopathy, however, may be reduced vitamin A levels secondary to malabsorption in the setting of CD [39]. Vitamin A plays an important role in the retina [40] and malnutrition retinopathy has been described in CD [39], although this warrants further investigation. Lastly, studies specifically assessing outcomes in those with undiagnosed CD are limited, but in the general adult population, undiagnosed CD was associated with a nearly 4-fold increased risk of death over seronegative individuals [41]. Therefore, prior works and our study results add credence to the role of CD and potentially undiagnosed CD, as a potential added risk factor for increased morbidity [42], especially in the already higher risk type 1 diabetes population. Concerningly, this risk may begin at a young age. Future studies are needed to determine the mechanisms in which the combined disease process increases risk for these outcomes.

Finally, we sought to assess the relationship between tTGA levels with cardiovascular risk factors. CD [10] and type 1 diabetes [43] are both associated with increased cardiovascular risk, and combined disease appears to accelerate the atherosclerotic process [9]. In addition, the role tTGA levels and CD control plays in cardiovascular risk is not fully understood [35, 44]. In analysis focused on tTGA levels, we demonstrated that tTGA levels’ suggesting controlled CD was associated with lower HbA1c. Similar findings were demonstrated in another large study of youth with type 1 diabetes [44], which found that individuals with CD who had normalization of CD antibodies had lower HbA1c than individuals with CD who had positive CD antibody status and those with type 1 diabetes only. It is possible that any difference detected in HbA1c in those with controlled CD reflects confounding by an individual’s overall health behavior and/or medical adherence.

Strengths of this study include analysis of a diverse race and ethnic cohort, standardized tTGA measurements, and the ascertainment of tTGA levels on 2,444 youth with type 1 diabetes to assess the impact of tTGA levels on risk factors and identify potentially undiagnosed CD. An additional strength was the ability to simultaneously assess CD status, tTGA levels, and an array of cardiovascular risk factors and diabetes-associated complications in a large cohort of youth with type 1 diabetes. Limitations of this study are mainly centered on the data collected in SEARCH pertaining to CD. The original design of the SEARCH study did not include comprehensive assessment of CD or dietary data, including adherence to a gluten-free diet. As a result, CD status was based on report via questionnaire without the ability to verify if they have biopsy proven CD. However, to internally validate reported CD in our subset of the study population, we were able to review individuals who reported CD at a single site (Ohio) only for biopsy proven CD. We found that of the 21 who reported CD, 18 had biopsy proven CD, one is being worked up for CD, and one has not undergone a biopsy but avoids gluten altogether now, demonstrating reliability of self-report data. Lastly, IgA levels were not obtained on all individuals as the prevalence of IgA deficiency in the US ranges from 1 : 233–1 : 1,000 [45], indicating IgA deficiency in potentially 2–10 participants in this study. This frequency of IgA deficiency does not influence the statistical outcomes presented here and, if anything, would cause us to miss a CD case, underestimating our positive findings.

In conclusion, we demonstrate that there is a potential for undiagnosed CD in youth with type 1 diabetes. Furthermore, potentially undiagnosed CD was associated with a higher risk of diabetic retinopathy. In addition, tTGA levels may be associated with the cardiovascular risk factor HbA1c. These findings indicate the importance of routine screening for CD in type 1 diabetes in youth.

Data Availability

The datasets supporting the current study are available at the NIH and are available from the corresponding author on request.

Disclosure

This study includes data provided by the Ohio Department of Health, which should not be considered an endorsement of this study or its conclusions. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention and the National Institute of Diabetes and Digestive and Kidney Diseases. This work was completed in partial fulfillment of the Master of Science degree in Clinical and Translational research in the Division of Epidemiology, University of Cincinnati College of Medicine.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this article.

Authors’ Contributions

R. P. B. designed study, interpreted data, and wrote the manuscript. E. T. J. designed study, conducted analysis, interpreted data, and edited the manuscript. J. R. conducted analysis and edited the manuscript. N. A. C. designed study and edited the manuscript. D. M. designed study and edited the manuscript. L. M. D. designed study and edited the manuscript. G. I. designed study and edited the manuscript. A. R. K. interpreted data, contributed to discussion, and edited the manuscript. A. M. interpreted data, contributed to discussion, and edited the manuscript. D. P. interpreted data, contributed to discussion, and edited the manuscript. L. M. interpreted data, contributed to discussion, and edited the manuscript. D. D. designed study and edited the manuscript. A. S. S. designed study, interpreted data, and wrote and edited the manuscript. All authors approved the final version of the manuscript. R. P. B. is the guarantor of this work and, as such, had full access to the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Acknowledgments

The Search for Diabetes in Youth study is indebted to the many youths and their families, and their healthcare providers, whose participation made this study possible. Search 3/4: The authors wish to acknowledge the involvement of the Kaiser Permanente Southern California’s Marilyn Owsley Clinical Research Center (funded by Kaiser Foundation Health Plan and supported in part by the Southern California Permanente Medical Group); the South Carolina Clinical and Translational Research Institute, at the Medical University of South Carolina, NIH/National Center for Advancing Translational Sciences (NCATS), under Grant nos. UL1 TR000062 and UL1 Tr001450; Seattle Children’s Hospital and the University of Washington, NIH/NCATS, under Grant no. UL1 TR00423; University of Colorado Pediatric Clinical and Translational Research Center, NIH/NCATS, under Grant no. UL1 TR000154; the Barbara Davis Center at the University of Colorado at Denver (DERC NIH Grant no. P30 DK57516); the University of Cincinnati, NIH/NCATS, under Grant nos. UL1 TR000077 and UL1 TR001425; and the Children with Medical Handicaps program managed by the Ohio Department of Health. This publication was supported by an Institutional Clinical and Translational Science Award, NIH/NCATS 1ULTR001425. Grant Support (Search 4): The Search for Diabetes in Youth Cohort study (Grant nos. 1R01DK127208-01 and 1UC4DK108173) is funded by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, and supported by the Centers for Disease Control and Prevention. The Population Based Registry of Diabetes in Youth study (Grant nos. 1U18DP006131, U18DP006133, U18DP006134, U18DP006136, U18DP006138, and U18DP006139) is funded by the Centers for Disease Control and Prevention (DP-15-002) and supported by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases. Grant Support (Search 1, 2, 3): Search for Diabetes in Youth is funded by the Centers for Disease Control and Prevention (PA numbers 00097, DP-05-069, and DP-10-001) and supported by the National Institute of Diabetes and Digestive and Kidney Diseases. This study was also supported by Kaiser Permanente Southern California (Grant nos. U48/CCU919219, U01 DP000246, and U18DP002714), University of Colorado Denver (Grant nos. U48/CCU819241-3, U01 DP000247, and U18DP000247-06A1), Cincinnati’s Children’s Hospital Medical Center (Grant nos. U48/CCU519239, U01 DP000248, and 1U18DP002709), University of North Carolina at Chapel Hill (Grant nos. U48/CCU419249, U01 DP000254, and U18DP002708), Seattle Children’s Hospital (Grant nos. U58/CCU019235-4, U01 DP000244, and U18DP002710-01), and Wake Forest University School of Medicine (Grant nos. U48/CCU919219, U01 DP000250, and 200-2010-35171). Funding was also provided by the Endocrine Fellows Foundation and the Pediatric Endocrine Society. ARK is supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant no. KL2TR002490. Funding was also provided by the T32 ES010957 Molecular Epidemiology in Children’s Environmental Health Training Program (MECEH) Grant.

Supplementary Materials

Supplementary table: co-autoimmunity describes the frequency of coexisting autoimmune conditions amongst participants with type 1 diabetes by the CD status. (Supplementary Materials)

References

J. F. Ludvigsson, D. A. Leffler, J. C. Bai et al., “The Oslo definitions for coeliac disease and related terms,” Gut, vol. 62, no. 1, pp. 43–52, 2013.
View at: Publisher Site | Google Scholar
D. J. Smyth, V. Plagnol, N. M. Walker et al., “Shared and distinct genetic variants in type 1 diabetes and celiac disease,” New England Journal of Medicine, vol. 359, no. 26, pp. 2767–2777, 2008.
View at: Publisher Site | Google Scholar
A. Akirov and O. Pinhas-Hamiel, “Co-occurrence of type 1 diabetes mellitus and celiac disease,” World Journal of Diabetes, vol. 6, no. 5, pp. 707–714, 2015.
View at: Publisher Site | Google Scholar
M. E. Craig, N. Prinz, C. T. Boyle et al., “Prevalence of celiac disease in 52,721 youth with type 1 diabetes: international comparison across three continents,” Diabetes Care, vol. 40, no. 8, pp. 1034–1040, 2017.
View at: Publisher Site | Google Scholar
B. Draznin, V. R. Aroda, G. Bakris et al., “14. Children and adolescents: standards of medical care in diabetes—2022,” Diabetes Care, vol. 45, pp. S208–S231, 2022.
View at: Publisher Site | Google Scholar
A. Pham-Short, K. C. Donaghue, G. Ambler, H. Phelan, S. Twigg, and M. E. Craig, “Screening for celiac disease in type 1 diabetes: a systematic review,” Pediatrics, vol. 136, pp. 170–176, 2015.
View at: Publisher Site | Google Scholar
K. Warncke, S. Liptay, E. Fröhlich-Reiterer et al., “Vascular risk factors in children, adolescents, and young adults with type 1 diabetes complicated by celiac disease: results from the DPV initiative,” Pediatric Diabetes, vol. 17, no. 3, pp. 191–198, 2016.
View at: Publisher Site | Google Scholar
W. F. Tetzlaff, T. Meroño, M. Menafra et al., “Markers of inflammation and cardiovascular disease in recently diagnosed celiac disease patients,” World Journal of Cardiology, vol. 9, no. 5, pp. 448–456, 2017.
View at: Publisher Site | Google Scholar
D. Pitocco, S. Giubilato, F. Martini et al., “Combined atherogenic effects of celiac disease and type 1 diabetes mellitus,” Atherosclerosis, vol. 217, no. 2, pp. 531–535, 2011.
View at: Publisher Site | Google Scholar
L. Santoro, G. De Matteis, M. Fuorlo et al., “Atherosclerosis and cardiovascular involvement in celiac disease: the role of autoimmunity and inflammation,” European Review for Medical and Pharmacological Sciences, vol. 21, no. 23, pp. 5437–5444, 2017.
View at: Publisher Site | Google Scholar
T. R. Rohrer, J. Wolf, S. Liptay et al., “Microvascular complications in childhood-onset type 1 diabetes and celiac disease: a multicenter longitudinal analysis of 56,514 patients from the German-Austrian dpv database,” Diabetes Care, vol. 38, no. 5, pp. 801–807, 2015.
View at: Publisher Site | Google Scholar
K. Mollazadegan, D. S. Sanders, J. Ludvigsson, and J. F. Ludvigsson, “Long-term coeliac disease influences risk of death in patients with type 1 diabetes,” Journal of Internal Medicine, vol. 274, no. 3, pp. 273–280, 2013.
View at: Publisher Site | Google Scholar
E. Liu, F. Dong, A. E. Barón et al., “High incidence of celiac disease in a long-term study of adolescents with susceptibility genotypes,” Gastroenterology, vol. 152, no. 6, pp. 1329–1336.e1, 2017.
View at: Publisher Site | Google Scholar
E. J. Mayer-Davis, J. M. Lawrence, D. Dabelea et al., “Incidence trends of type 1 and type 2 diabetes among youths, 2002–2012,” New England Journal of Medicine, vol. 376, no. 15, pp. 1419–1429, 2017.
View at: Publisher Site | Google Scholar
C. Catassi, D. Kryszak, B. Bhatti et al., “Natural history of celiac disease autoimmunity in a USA cohort followed since 1974,” Annals of Medicine, vol. 42, no. 7, pp. 530–538, 2010.
View at: Publisher Site | Google Scholar
R. F. Hamman, R. A. Bell, D. Dabelea et al., “The SEARCH for diabetes in youth study: rationale, findings, and future directions,” Diabetes Care, vol. 37, no. 12, pp. 3336–3344, 2014.
View at: Publisher Site | Google Scholar
F. Bao, L. Yu, S. Babu et al., “One third of HLA DQ2 homozygous patients with type 1 diabetes express celiac disease-associated transglutaminase autoantibodies,” Journal of Autoimmunity, vol. 13, pp. 143–148, 1999.
View at: Publisher Site | Google Scholar
M. Li, L. Yu, C. Tiberti et al., “A report on the international transglutaminase autoantibody workshop for celiac disease,” American Journal of Gastroenterology, vol. 104, no. 1, pp. 154–163, 2009.
View at: Publisher Site | Google Scholar
G. M. Vaitaitis, D. Waid, D. H. Wagner, D. M. Waid, P. A. Gottlieb, and D. H. Wagner, “Biomarker discovery in pre-type 1 diabetes; Th40 cells as a predictive risk factor,” The Journal of Immunology, vol. 202, pp. 115.4–4142, 2019.
View at: Publisher Site | Google Scholar
C. Webb, F. Norström, A. Myléus et al., “Celiac disease can be predicted by high levels of anti-tissue transglutaminase antibodies in population-based screening,” Journal of Pediatric Gastroenterology and Nutrition, vol. 60, no. 6, pp. 787–791, 2015.
View at: Publisher Site | Google Scholar
M. Cerqueiro Bybrant, E. Udén, F. Frederiksen et al., “Celiac disease can be predicted by high levels of tissue transglutaminase antibodies in children and adolescents with type 1 diabetes,” Pediatric Diabetes, vol. 22, no. 3, pp. 417–424, 2021.
View at: Publisher Site | Google Scholar
E. Liu, F. Bao, K. Barriga et al., “Fluctuating transglutaminase autoantibodies are related to histologic features of celiac disease,” Clinical Gastroenterology and Hepatology, vol. 1, no. 5, pp. 356–362, 2003.
View at: Publisher Site | Google Scholar
J. Schofield, J. Ho, and H. Soran, “Cardiovascular risk in type 1 diabetes mellitus,” Diabetes Ther, vol. 10, no. 3, pp. 773–789, 2019.
View at: Publisher Site | Google Scholar
D. Dabelea, J. M. Stafford, E. J. Mayer-Davis et al., “Association of type 1 diabetes vs. type 2 diabetes diagnosed during childhood and adolescence with complications during teenage years and young adulthood,” JAMA, vol. 317, no. 8, pp. 825–835, 2017.
View at: Publisher Site | Google Scholar
R. Klein, B. E. Klein, Y. L. Magli et al., “An alternative method of grading diabetic retinopathy,” Ophthalmology, vol. 93, no. 9, pp. 1183–1187, 1986.
View at: Publisher Site | Google Scholar
E. L. Feldman, M. J. Stevens, P. K. Thomas, M. B. Brown, N. Canal, and D. A. Greene, “A practical two-step quantitative clinical and electrophysiological assessment for the diagnosis and staging of diabetic neuropathy,” Diabetes Care, vol. 17, no. 11, pp. 1281–1289, 1994.
View at: Publisher Site | Google Scholar
M. Jaiswal, E. M. Urbina, R. P. Wadwa et al., “Reduced heart rate variability is associated with increased arterial stiffness in youth with type 1 diabetes: the SEARCH CVD study,” Diabetes Care, vol. 36, no. 8, pp. 2351–2358, 2013.
View at: Publisher Site | Google Scholar
P. Bjornstad, D. Z. Cherney, and D. M. Maahs, “Update on estimation of kidney function in diabetic kidney disease,” Current Diabetes Reports, vol. 15, no. 9, pp. 57–69, 2015.
View at: Publisher Site | Google Scholar
I. Shrier and R. W. Platt, “Reducing bias through directed acyclic graphs,” BMC Medical Research Methodology, vol. 8, no. 1, p. 70, 2008.
View at: Publisher Site | Google Scholar
J. Y. Kang, A. H. Kang, A. Green, K. A. Gwee, and K. Y. Ho, “Systematic review: worldwide variation in the frequency of coeliac disease and changes over time,” Alimentary Pharmacology & Therapeutics, vol. 38, no. 3, pp. 226–245, 2013.
View at: Publisher Site | Google Scholar
F. Cerutti, G. Bruno, F. Chiarelli, R. Lorini, F. Meschi, and C. Sacchetti, “Diabetes Study Group of the Italian Society of Pediatric Endocrinology and Diabetology. Younger age at onset and sex predict celiac disease in children and adolescents with type 1 diabetes: an Italian multicenter study,” Diabetes Care, vol. 27, no. 6, pp. 1294–1298, 2004.
View at: Publisher Site | Google Scholar
K. Larsson, A. Carlsson, E. Cederwall et al., “Annual screening detects celiac disease in children with type 1 diabetes,” Pediatric Diabetes, vol. 9, pp. 354–359, 2008.
View at: Publisher Site | Google Scholar
A. Parkkola, T. Härkönen, S. J. Ryhänen et al., “Transglutaminase antibodies and celiac disease in children with type 1 diabetes and in their family members,” Pediatric Diabetes, vol. 19, no. 2, pp. 305–313, 2018.
View at: Publisher Site | Google Scholar
S. Castellaneta, E. Piccinno, M. Oliva et al., “High rate of spontaneous normalization of celiac serology in a cohort of 446 children with type 1 diabetes: a prospective study,” Diabetes Care, vol. 38, no. 5, pp. 760–766, 2015.
View at: Publisher Site | Google Scholar
J. S. Leeds, A. D. Hopper, M. Hadjivassiliou, S. Tesfaye, and D. S. Sanders, “High prevalence of microvascular complications in adults with type 1 diabetes and newly diagnosed celiac disease,” Diabetes Care, vol. 34, no. 10, pp. 2158–2163, 2011.
View at: Publisher Site | Google Scholar
K. Mollazadegan, M. Kugelberg, S. M. Montgomery, D. S. Sanders, J. Ludvigsson, and J. F. Ludvigsson, “A population-based study of the risk of diabetic retinopathy in patients with type 1 diabetes and celiac disease,” Diabetes Care, vol. 36, no. 2, pp. 316–321, 2013.
View at: Publisher Site | Google Scholar
H. Korkmaz, M. Sozen, and L. Kebapcilar, “Increased arterial stiffness and its relationship with inflammation, insulin, and insulin resistance in celiac disease,” European Journal of Gastroenterology and Hepatology, vol. 27, no. 10, pp. 1193–1199, 2015.
View at: Publisher Site | Google Scholar
W. Zhang, H. Liu, M. Rojas, R. W. Caldwell, and R. B. Caldwell, “Anti-inflammatory therapy for diabetic retinopathy,” Immunotherapy, vol. 3, no. 5, pp. 609–628, 2011.
View at: Publisher Site | Google Scholar
T. G. Martins, A. L. Costa, M. K. Oyamada, P. Schor, and A. M. Sipahi, “Ophthalmologic manifestations of celiac disease,” International Journal of Ophthalmology, vol. 9, no. 1, pp. 159–162, 2016.
View at: Publisher Site | Google Scholar
J. Sajovic, A. Meglič, D. Glavač, Š. Markelj, M. Hawlina, and A. Fakin, “The role of vitamin A in retinal diseases,” International Journal of Molecular Sciences, vol. 23, no. 3, p. 1014, 2022.
View at: Publisher Site | Google Scholar
A. Rubio-Tapia, R. A. Kyle, E. L. Kaplan et al., “Increased prevalence and mortality in undiagnosed celiac disease,” Gastroenterology, vol. 137, no. 1, pp. 88–93, 2009.
View at: Publisher Site | Google Scholar
U. Peters, J. Askling, G. Gridley, A. Ekbom, and M. Linet, “Causes of death in patients with celiac disease in a population-based Swedish cohort,” Archives of Internal Medicine, vol. 163, no. 13, pp. 1566–1572, 2003.
View at: Publisher Site | Google Scholar
C. Mameli, S. Mazzantini, M. Ben Nasr, P. Fiorina, A. E. Scaramuzza, and G. V. Zuccotti, “Explaining the increased mortality in type 1 diabetes,” World Journal of Diabetes, vol. 6, no. 7, pp. 889–895, 2015.
View at: Publisher Site | Google Scholar
K. Nagl, E. Bollow, S. Liptay et al., “Lower HbA1c in patients with type 1 diabetes and celiac disease who reached celiac-specific antibody-negativity-A multicenter DPV analysis,” Pediatric Diabetes, vol. 20, no. 8, pp. 1100–1109, 2019.
View at: Publisher Site | Google Scholar
L. Yel, “Selective IgA deficiency,” Journal of Clinical Immunology, vol. 30, no. 1, pp. 10–16, 2010.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2023 Ryan P. Brady et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies