- Email: [email protected]
High Prevalence of Celiac Disease Among Screened First-Degree Relatives
ORIGINAL ARTICLE
High Prevalence of Celiac Disease Among Screened First-Degree Relatives Shilpa S. Nellikkal, MBBS; Yamen Hafed, MD; Joseph J. Larson, BS; Joseph A. Murray, MD; and Imad Absah, MD Abstract Objective: To investigate the prevalence of first-degree relatives (FDRs) with celiac disease detected at screening and diagnostic significance of anti-tissue transglutaminase (anti-TTG). Patients and Methods: We performed a retrospective cohort study of 104 patients with a diagnosis of celiac disease and their FDRs, collecting data from electronic records of Mayo Clinic and celiac disease registry from December 20, 1983, to May 22, 2017. We collected demographics, presenting symptoms, indication for testing, family history, number of other family members screened, biopsy reports, and results of serologic tests. Results: Of 477 FDRs identified, 360 were screened (mean screening rate per family, 79%25%) and 160 FDRs (44.4%) were diagnosed with celiac disease, at a mean age 31.921.6 years (62% female). All diagnosed FDRs had positive anti-TTG titers. Clinical features were documented in 148 diagnosed FDRs, of those 9 (6%) had classic, 97 (66%) had non-classic symptoms, and 42(28%) had no reported symptoms. Histology reports were available from 155 FDRs: 12 (8%) had Marsh 1, 77 (50%) had Marsh 3a, and 66 (43%) had Marsh 3b. A level of anti-TTG greater than or equal to 2.75 of the upper limit of normal identified FDRs with villous atrophy with 87% sensitivity, 82% specificity, and a positive predictive value of 95%. Conclusion: In a retrospective cohort study of patients diagnosed with celiac disease, we found a high prevalence of celiac disease among screened FDRs. High anti-TTG titers associated with villous atrophy on small bowel biopsies, irrespective of symptoms. ª 2019 Mayo Foundation for Medical Education and Research
C
eliac disease (CD) is a common immune-mediated disorder with an estimated prevalence of 1% in the general population.1 Modes of presentation of adults and children with CD can vary from classic (diarrhea predominant) or non-classic with individuals either presenting with complications of CD, extra-intestinal manifestations, as well as truly asymptomatic (or subclinical) individuals picked up through screening high-risk groups.2,3 Recent data suggest that the majority of CD patients are asymptomatic or have non-classic presentation,4,5 confirming the need of having a low screening threshold for CD. In the current guidelines, diagnosis of CD involves screening using highly sensitive serologic markers such as antietissue transglutaminase antibodies (anti-TTG) followed by an endoscopic
n
Mayo Clin Proc. 2019;94(9):1807-1813
assessment and obtaining confirmatory small bowel biopsy (SBB) specimens.6 Furthermore, the 2012 European Society for Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) guidelines suggested that SBB can be omitted in a specific subgroup of symptomatic children with highly positive antiTTG titers,7,8 highlighting the significance of highly sensitive serologic markers in diagnosing and managing CD.9 It is known that certain groups of the population are at higher risk for CD, for example, patients with specific chromosomal disorders such as trisomy 21 and Turner syndrome up to 9%,10,11 autoimmune disorders such as autoimmune thyroiditis, diabetes mellitus type 1 approximately 7% to 10%,12,13 and first-degree relatives (FDRs) of patients with CD up to 11% in previous
Mayo Clin Proc. n September 2019;94(9):1807-1813 n https://doi.org/10.1016/j.mayocp.2019.03.027 www.mayoclinicproceedings.org n ª 2019 Mayo Foundation for Medical Education and Research
From the Division of Pediatric Gastroenterology and Hepatology (S.S.N., Y.H., I.M.); the Division of Gastroenterology and Hepatology (J.A.M., I.A.); and the Division of Biomedical Statistics and Informatics (J.J.L.), Mayo Clinic, Rochester, Minnesota.
1807
MAYO CLINIC PROCEEDINGS
studies. The relative risk of CD in FDRs is higher in siblings, parents, and children with odds ratios of 2.5, 1.6, and 1, respectively.14 There is disagreement between the pediatric and adult societies’ guidelines in regard to screening at-risk asymptomatic individuals including FDRs. The US Preventive Services Task Force latest recommendations support only screening symptomatic FDRs due to paucity of data to support screening in asymptomatic FDRs.15,16 This contradicts the belief of providers who should be vigilant and actively screen for CD in light of a changing presentation from classic to silent or non-classic presentation. A better detection rate of CD could conceivably prevent long-term complications such as nutritional deficiencies, development of new autoimmune conditions, and small bowel malignancy known to occur in undiagnosed CD cases.12,17,18 Previous studies among FDRs report CD prevalence rates of 5.5%, 8.3%, and 10.9%, and the majority were asymptomatic.19-21 Hence, screening only symptomatic FDRs can decrease the detection rate of CD. Easy accessibility of gluten-free diet (GFD) to the general population may lead high-risk groups such as FDRs of patients with CD to follow a GFD without formal testing or diagnosis. Hence, an alternative, highly sensitive, noninvasive test such as anti-TTG titers may result in more compliance with formal testing. Therefore, in this study the primary aim was to assess the prevalence of CD diagnosis among screened FDRs using anti-TTG serologic markers. A secondary goal was to assess the sensitivity of the anti-TTG serologic markers compared with SBB procedures. PATIENTS AND METHODS Study Design After obtaining study approval from the Mayo Clinic Institutional Review Board using a retrospective cohort of CD subjects between December 20, 1983, and May 22, 2017, as a basis, we identified first-degree familial relationships. The first subject within each family diagnosed was termed index case patient 1808
Mayo Clin Proc.
n
(ICP), whereas the remaining family members were termed FDRs. Data were extracted from two sources: (1) the Mayo Clinic electronic health records and (2) the CD registry. Part of the CD registry data was funded by the National Institutes of Health14 between 2003 and 2006 to collect data on newly diagnosed CD patients and their families, containing both prospectively and retrospectively collected data. Adult patients were included if they met the American College of Gastroenterology guidelines, and children were included if they met North American Society for Pediatric Gastroenterology, Hepatology and Nutrition or ESPGHAN guidelines. Eleven ICPs were diagnosed by either endomysial antibodies immunoglobulin A (IgA) (EMA) or anti-gliadin antibodies. Only five FDRs and one ICP were diagnosed as per the ESPGHAN guidelines. Histologic findings were classified into Marsh1 (>30 intraepithelial lymphocytes [IELs] with normal crypts and villi), Marsh 3a (>30 IELs, crypts hyperplasia, and mild villous atrophy [VA]) and Marsh 3b (>30 IELs, crypts hyperplasia, and marked VA).22 Patients with Marsh 1 (IELs) were included if they had more than one positive celiac serology marker (anti-TTG and EMA) and resolution of their symptoms and serologic markers in response to GFD. Patients were grouped based on age and classified as a child (<18 years old) or adult (18 years old). Only patients with appropriate consent based on their source of identification were included. The following data were abstracted in both ICPs and FDRs: date of birth, sex, date of CD diagnosis (biopsy date), age at diagnosis of CD, presenting symptoms, indication for testing, family history, number of other family members screened with their serology results, biopsy reports, and human leukocyte antigen positivity (if tested). Symptoms were classified as classic if reports of weight loss and diarrhea or failure to thrive in children and weight loss and diarrhea in adults were identified. Symptoms were considered nonclassic if reports of nonspecific abdominal pain, bloating, flatulence, constipation, headache, poor dental enamel, anemia, and osteoporosis were identified in the medical records.
September 2019;94(9):1807-1813
n
https://doi.org/10.1016/j.mayocp.2019.03.027 www.mayoclinicproceedings.org
DIAGNOSING CELIAC DISEASE AMONG FIRST-DEGREE RELATIVES
Patients were classified as asymptomatic if there were no reported symptoms at the time of CD screening. Statistical Analysis Patient characteristics were summarized for ICPs and FDRs, both overall and by age or relationship to the ICP. Age, number of FDRs, and percent of FDRs screened were presented as mean (SD). Different kits of anti-TTG (human recombinant TTG Elisa and Red Cell TGG, both from Inova, San Diego, CA) with different cutoffs were used in this period of the study, therefore the number of times anti-TTG titers were above the upper limit of normal (ULN) was presented using median (interquartile range [IQR]), whereas categoric findings were presented using counts and percentages. The KruskalWallis test was used for differences in the number of times anti-TTG titers were above the ULN across various groups. Categoric factors were evaluated using the c2 test of independence. Among FDRs, the receiver operating characteristic (ROC) method was used to classify CD positivity according to the number of times the anti-TTG titers were above the ULN. The ROC method helps assess the discriminatory ability of a continuous measure based on total area under the curve. We also used the ROC method to identify a single anti-TTG titer threshold that optimized the sensitivity and specificity with and without IgA-deficient patients. We used a method of estimating the optical threshold which minimized the Euclidean distance relative to the point denoting perfect classification P values less than .05 were considered statistically significant. Analyses were performed using SAS version 9.4M5 (SAS Institute Inc; Cary, NC) and R version 3.4.2 (Vienna, Austria).
ICPs (n=104) All underwent SBB except 1
All FDRs (n=477) 117 FDRs did not undergo CD screening Screened FDRs (n=360) 182 FDRs tested negative on anti-TTG serology Anti-TTG positive FDRs (n=178) 18 anti-TTG positive FDRs had negative SBB reports
Diagnosed FDRs (n=160) All underwent SBB except 5
RESULTS Patients We identified 104 CD ICPs, with mean age (SD) of 30.8 (19.7) years (8.2 [4.9] years among children; 43.4 [11.8] years among adults). From these 104 ICPs, we identified Mayo Clin Proc. n September 2019;94(9):1807-1813 www.mayoclinicproceedings.org
477 FDRs (all FDRs) of whom 360 (75.5%) were screened for CD (screened FDRs). Among the 360 screened FDRs, 145 (40.3%) were siblings, 137 (38.1%) offspring, and 78 (21.7%) parents. Three hundred fifty-five of 360 (98.6%) FDRs were screened because of a family history of CD. On average, we identified 4.6 (2.8) FDRs per family but only 3.4 (2.3) were screened, which equates to a mean per family screening rate of 79% (25%). Among the 360 screened FDRs, 160 (44.4%) FDRs were diagnosed with CD, which are termed diagnosed FDRs (Figure 1). Among the 153 diagnosed FDRs with available data, age at diagnosis was 31.9 (21.6) years (9.0 [4.4] years in 54 children; 44.3 [16.5] years in 99 adults) (Table). Both ICPs and diagnosed FDRs had a similar predominance of female patients with 64 of 104 ICPs (61.5%) and 99 of 160 FDRs (61.9%). The median period of
n
FIGURE 1. First-degree relative (FDR) distribution throughout the screening process. CD ¼ celiac disease; ICP ¼ index case patient; SBB ¼ small bowel biopsy; TTG ¼ tissue transglutaminase.
https://doi.org/10.1016/j.mayocp.2019.03.027
1809
MAYO CLINIC PROCEEDINGS
TABLE. Demographics and Clinical Characteristics of ICPs and CD-Diagnosed FDRsa Demographics and clinical characteristics
Index (n¼104)
Diagnosed FDRs (n¼160)
b
Age, mean (SD), y
31.9 (21.6) (n¼153)b
30.8 (19.7) (n¼92)
Female
64 (61.5)
99 (62)
Symptoms Classic Non-classic Asymptomatic
b
(n¼88) 10 (11.4) 67 (76.1) 11 (12.5)
(n¼148)b 9 (6.1) 97 (65.5) 42 (28.4)
Diagnostic serology assessment Anti-TTG positive Other serologyc
(n¼93)b 82 (88.2) 11 (11.8)
(n¼160)b 160 (100) 0
Anti-TTG titer ULN, median (IQR)
12.4 (4.2-16.7) (n¼73)b
9.5 (5.0-12.5) (n¼143)b
b
Histology Marsh 1 Marsh 3a Marsh 3b
(n¼155)b 12 (7.7) 77 (49.7) 66 (42.6)
(n¼93) 11 (11.8) 37 (39.8) 45 (48.4)
anti-TTG ¼ anti-tissue transglutaminase; CD ¼ celiac disease; FDR ¼ first-degree relative; ICP ¼ index case patient; IQR ¼ interquartile range; ULN ¼ upper limit of normal. b Data on this characteristic was not available on all subjects so we have provided the sample size for this specific characteristic. c Other serology such as endomysial antibodies immunoglobulin A or anti-gliadin antibodies. a
Values are presented n (%) unless otherwise stated.
Clinical Features Among ICPs and Diagnosed FDRs Symptom records were available among 88 ICPs and 148 FDRs (Table). Among the ICPs, classic presentation was reported in 10 (11.4%) patients, non-classic in 67 (76.1%), and 11 (12.5%) were silent (asymptomatic). Family history was the indication for CD screening in the majority of the screened FDRs. Clinical presentation was documented in 148 of 160 (93%) of the diagnosed FDRs, of whom many recalled symptoms during the screening for CD diagnosis: 9 (6.1%) had classic presentation, 97 (65.5%) had nonclassic presentation, and 42 (28.4%) were asymptomatic. No follow up data was available on the screened FDRs who had negative screening to compare their symptoms with the diagnosed FDRs. Anti-TTG Serology Of the 360 patients screened, 178 FDRs (49.4%) had positive anti-TTG serology. Of 178 patients who were positive for antiTTG serology, 160 were diagnosed FDRs with CD (149 tested positive for anti-TTG 1810
Mayo Clin Proc.
n
IgA and 11 were found to be IgA-deficient but anti-TTG IgG positive). The numerical value of the anti-TTG titers was available in 143 of 160 diagnosed FDRs and 73 of 104 ICPs, which allowed us to calculate the TTG value above the ULN (Figure 2). We found that diagnosed FDRs had a median
1.0
0.8
Sensitivity
CD diagnosis between ICPs and diagnosed FDRs was 192 days (IQR, 34.0-634.0 days).
0.6
0.4
0.2 AUC=0.901 (95% CI: 0.838-0.964) 0.0 1.0
0.8
0.6 0.4 Specificity
0.2
0.0
FIGURE 2. Receiver operating curve for predicting partial or total villous atrophy using the number of times beyond upper limit of normal for tissue transglutaminase value. AUC ¼ area under the curve.
September 2019;94(9):1807-1813
n
https://doi.org/10.1016/j.mayocp.2019.03.027 www.mayoclinicproceedings.org
DIAGNOSING CELIAC DISEASE AMONG FIRST-DEGREE RELATIVES
ULN value of 9.5 (IQR, 5.0-12.5) whereas among the ICPs the median ULN value was 12.5 (IQR, 4.2-16.7). When comparing children and adult diagnosed FDRs, we noted a significantly higher median anti-TTG ULN titer among children compared with adults (12.5 versus 8.5; P¼.003). Histology Findings Among ICPs and Diagnosed FDRs Histology reports were available in 93 of 104 ICPs (10 had no pathology reports and 1diagnosed without SBB as per ESPGHAN guidelines) and all FDRs who underwent confirmatory SBB (155 of 160; 5 patients were diagnosed without SBB as per ESPGHAN guidelines). Among the ICPs, we found 11 (11.8%) had Marsh 1, 37 (39.8%) had Marsh 3a and 45 (48.4%) had Marsh 3b. The FDR group had a similar biopsy results distribution as the ICP group: 12 (7.7%) had Marsh 1, 77 (49.7%) had Marsh 3a, and 66 (42.6%) had Marsh 3b. On comparing children to adult-diagnosed FDRs, distributions of histologic findings were not different across the groups: 4 (8.2%) had Marsh 1, 27 (55.1%) had Marsh 3a, and 18 (36.7%) had Marsh 3b; adults had 7 (7.1%), 48 (48.5%), and 44 (44.4%), respectively (P¼.67). Among the diagnosed FDRs, using the ROC method, we found that anti-TTG (IgA and IgG) levels of 2.75 greater than or equal to ULN predicts VA with 87% sensitivity, 82% specificity, 95% positive predictive value, and 61% negative predictive value (area under the curve, 0.90) (Figure 2). Similar results were noted after excluding 10 ICPs and 11 FDRs who were IgA deficient (88% sensitivity, 85% specificity, 96% positive predictive value, and 65% negative predictive value with an area under the curve of 0.90). We also compared the mean anti-TTG ULN values among patients with VA (Marsh 3a and 3b) and without VA (Marsh 1) in two groups: ICPs and diagnosed FDRs. Among ICPs, no difference in median antiTTG ULN value was seen between those without VA (4.5; IQR, 2.0-12.5) and those with VA (12.5; IQR, 5.4-18.5; P¼.11). However, the diagnosed FDRs with VA had a Mayo Clin Proc. n September 2019;94(9):1807-1813 www.mayoclinicproceedings.org
n
higher median anti-TTG ULN value (9.5; IQR, 5.0-12.5) than those without VA (0.8; IQR, 0.3-2.7; P<.001). Symptoms did not associate with the degree of VA in either ICPs (P¼.66) or diagnosed FDRs (P¼.41). DISCUSSION Our study found high prevalence of CD among screened FDRs with 160 of 360 (44.4%), when more than two-thirds of FDRs were screened overall. An FDR was diagnosed in a median period of approximately 6 months of a diagnosed ICP. Among the diagnosed FDRs, a correlation between high anti-TTG titers and VA independent of their symptoms was noted and a majority showed VA on SBB. Non-classic and silent (asymptomatic) presentation predominated among diagnosed FDRs. We also found that high positive anti-TTG titers at approximately 2.75 ULN were a predictor of VA on SBB in our cohort. Children had higher antiTTG ULN titers than adults. Histologic findings among ICPs and FDRs (both children and adults) were similar. The new recommendations of the US Preventive Services Task Force suggest that only symptomatic FDRs should be tested.16 In our study, we found a high prevalence of CD in screened FDRs (44.4%) with more than one-quarter being asymptomatic (28%), which is higher than the prevalence reported by previous studies ranging from 5.5% to 11%.19e21,23 This higher prevalence of diagnosed FDRs with CD in our cohort could be due to tertiary center referral bias and proactive testing among FDRs which was higher in our cohort (75.5%) compared with previous studies. Increasing incidence of CD among high-risk CD genotypes was recently reported, which supports the high prevalence noted in our study.24 We observed a larger number of symptomatic than asymptomatic cases among CD-diagnosed FDRs, which is likely due to the recall bias on further questioning of symptoms before CD diagnosis. However, overall asymptomatic presentation was more common in FDRs when compared with ICPs, suggesting earlier diagnosis due to early and proactive screening. Our results
https://doi.org/10.1016/j.mayocp.2019.03.027
1811
MAYO CLINIC PROCEEDINGS
are consistent with previous studies among FDRs that showed higher rates of asymptomatic CD than symptomatic CD25 and lack of association between symptoms and the degree of VA, whereas higher anti-TTG titer correlated with a higher degree of VA.26 Several studies have shown that antiTTG is a test of choice for CD screening among symptomatic and asymptomatic CD patients.27e29 In our study, an anti-TTG value of 2.75 ULN predicted the presence of VA with sensitivity and specificity of 87% and 82%, respectively. Based on high prevalence of positive anti-TTG titers between screened FDRs noted in our cohort and previous studies, we suggest that FDRs must be screened regardless of symptoms.30 FDRs with highly positive anti-TTG titers (>3 UNL) will more likely have VA on SBB. Furthermore, gastroenterologists and general practitioners should ask about family history of CD among their patients’ FDRs and consider screening them for CD if present. The limitations of our study include the retrospective nature of our study, referral bias associated with a tertiary care center, and limited follow-up data on patients who were lost to follow-up or elected to seek medical care locally. Hence, prospective studies are required to confirm the predictive value of anti-TTG titers as a diagnostic test among FDRs and the possibility of foregoing the need for confirmatory SBB. The strengths of the study are that data were partly prospectively collected and the median period of an FDR diagnosed with CD was within 6 months of identifying an ICP. CONCLUSION In a retrospective cohort study of patients diagnosed with CD, we found a high prevalence of CD among screened FDRs. High anti-TTG titers associated with VA on SBB, irrespective of symptoms. We also found a positive association between high anti-TTG titers and the presence of VA, supporting the need for screening all FDRs, including those who are asymptomatic. Larger multicenter prospective studies are required to confirm anti-TTG titers as a diagnostic test among FDRs. 1812
Mayo Clin Proc.
n
Abbreviations and Acronyms: anti-TTG = antietissue transglutaminase; CD = celiac disease; EMA = endomysial antibodies immunoglobulin A; ESPGHAN = European Society for Pediatric Gastroenterology, Hepatology and Nutrition; FDR = first-degree relative; ICP = index case patient; IEL = intraepithelial lymphocyte; Ig = immunoglobulin; IQR = interquartile range; ROC = receiver operating characteristic; SBB = small bowel biopsy; ULN = upper limit of normal; VA = villous atrophy
Potential Competing Interests: The authors report no competing interests. Correspondence: Address to Imad Absah, MD, Division of Pediatric Gastroenterology and Hepatology, Mayo Clinic, 200 First Street SW, Rochester, MN 55905 (absah.imad@ mayo.edu).
REFERENCES 1. Rubio-Tapia A, Ludvigsson JF, Brantner TL, Murray JA, Everhart JE. The prevalence of celiac disease in the United States. Am J Gastroenterol. 2012;107(10):1538-1544. quiz 1537, 1545. 2. Fasano A. Clinical presentation of celiac disease in the pediatric population. Gastroenterology. 2005;128(4 suppl 1):S68-S73. 3. Green PHR. The many faces of celiac disease: clinical presentation of celiac disease in the adult population. Gastroenterology. 2005;128(4 suppl 1):S74-S78. 4. Ludvigsson JF, Rubio-Tapia A, van Dyke CT, et al. Increasing incidence of celiac disease in a North American population. Am J Gastroenterol. 2013;108(5):818-824. 5. Almallouhi E, King KS, Patel B, et al. Increasing incidence and altered presentation in a population-based study of pediatric celiac disease in North America. J Pediatr Gastroenterol. 2017; 65(4):432-437. 6. Hill ID, Dirks MH, Liptak GS, et al. 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):1-19. 7. Husby S, Koletzko S, Korponay-Szabo IR, et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines for the diagnosis of coeliac disease. J Pediatr Gastroenterol Nutr. 2012;54(1):136-160. 8. Lock RJ, Pitcher MC, Unsworth DJ. IgA anti-tissue transglutaminase as a diagnostic marker of gluten sensitive enteropathy. J Clin Pathol. 1999;52(4):274-277. 9. Klapp G, Masip E, Bolonio M, et al. Celiac disease: the new proposed ESPGHAN diagnostic criteria do work well in a selected population. J Pediatr Gastroenterol Nutr. 2013;56(3): 251-256. 10. Mårild K, Størdal K, Hagman A, Ludvigsson JF. Turner syndrome and celiac disease: a case-control study. Pediatrics. 2016;137(2): e20152232. 11. Mazurek D, Wyka J. Down syndromedgenetic and nutritional aspects of accompanying disorders. Rocz Panstw Zakl Hig. 2015; 66(3):189-194. 12. Canova C, Pitter G, Ludvigsson JF, et al. Celiac disease and risk of autoimmune disorders: a population-based matched birth cohort study. J Pediatr. 2016;174:146-152.e1. 13. Mahmud FH, Murray JA, Kudva YC, et al. Celiac disease in type 1 diabetes mellitus in a North American community: prevalence, serologic screening, and clinical features. Mayo Clin Proc. 2005;80(11):1429-1434. 14. RubioeTapia A, Van Dyke CT, Lahr BD, et al. Predictors of family risk for celiac disease: a population-based study. Clin Gastroenterol Hepatol. 2008;6(9):983-987.
September 2019;94(9):1807-1813
n
https://doi.org/10.1016/j.mayocp.2019.03.027 www.mayoclinicproceedings.org
DIAGNOSING CELIAC DISEASE AMONG FIRST-DEGREE RELATIVES
15. Hill ID, Fasano A, Guandalini S, et al. NASPGHAN clinical report on the diagnosis and treatment of glutenrelated disorders. J Pediatr Gastroenterol Nutr. 2016; 63(1):156-165. 16. US Preventive Services Task Force, Bibbins-Domingo K, Grossman DC, et al. Screening for Celiac disease: US Preventive Services Task Force recommendation statement. JAMA. 2017;317(12):1252-1257. 17. Theethira TG, Dennis M, Leffler DA. Nutritional consequences of celiac disease and the gluten-free diet. Expert Rev Gastroenterol Hepatol. 2014;8(2):123-129. 18. Rubio-Tapia A, Kyle RA, Kaplan EL, et al. Increased prevalence and mortality in undiagnosed celiac disease. Gastroenterology. 2009;137(1):88-93. 19. Farre C, Humbert P, Vilar P, et al. Serological markers and HLADQ2 haplotype among first-degree relatives of celiac patients. Dig Dis Sci. 1999;44(11):2344-2349. 20. Högberg L, Fälth-Magnusson K, Grodzinsky E, Stenhammar L. Familial prevalence of coeliac disease: a twenty-year follow-up study. Scand J Gastroenterol. 2003;38(1):61-65. 21. Bonamico M, Ferri M, Mariani P, et al. Serologic and genetic markers of celiac disease: a sequential study in the screening of first degree relatives. J Pediatr Gastroenterol Nutr. 2006; 42(2):150-154. 22. Oberhuber G, Granditsch G, Vogelsang H. The histopathology of coeliac disease: time for a standardized report scheme for pathologists. Eur J Gastroenterol Hepatol. 1999;11(10):11851194.
Mayo Clin Proc. n September 2019;94(9):1807-1813 www.mayoclinicproceedings.org
n
23. Hoffenberg EJ, Bao F, Eisenbarth GS, et al. Transglutaminase antibodies in children with a genetic risk for celiac disease. J Pediatr. 2000;137(3):356-360. 24. Liu E, Dong F, Barón AE, et al. High Incidence of celiac disease in a long-term study of adolescents with susceptibility genotypes. Gastroenterology. 2017;152(6):1329-1336. 25. Tommasini A, Not T, Kiren V, et al. Mass screening for coeliac disease using antihuman transglutaminase antibody assay. Arch Dis Child. 2004;89(6):512-515. 26. Hawamdeh H, Al-Zoubi B, Al Sharqi Y, Qasrawi A, Abdelaziz Y, Barbar M. Association of tissue transglutaminase antibody titer with duodenal histological changes in children with celiac disease. Gastroenterol Res Pract. 2016;2016. https:// doi.org/10.1155/2016/6718590. erratum 2019;2019:2607194. 27. Troncone R, Maurano F, Rossi M, et al. IgA antibodies to tissue transglutaminase: an effective diagnostic test for celiac disease. J Pediatr. 1999;134(2):166-171. 28. Sulkanen S, Halttunen T, Laurila K, et al. Tissue transglutaminase autoantibody enzyme-linked immunosorbent assay in detecting celiac disease. Gastroenterology. 1998;115(6):1322-1328. 29. Hopper AD, Hadjivassiliou M, Hurlstone DP, et al. What is the role of serologic testing in celiac disease? A prospective, biopsyconfirmed study with economic analysis. Clin Gastroenterol Hepatol. 2008;6(3):314-320. 30. Trovato CM, Montuori M, Anania C, et al. Are ESPGHAN “biopsy-sparing” guidelines for celiac disease also suitable for asymptomatic patients? Am J Gastroenterol. 2015;110(10): 1485-1489.
https://doi.org/10.1016/j.mayocp.2019.03.027
1813
High Prevalence of Celiac Disease Among Screened First-Degree Relatives Shilpa S. Nellikkal, MBBS; Yamen Hafed, MD; Joseph J. Larson, BS; Joseph A. Murray, MD; and Imad Absah, MD Abstract Objective: To investigate the prevalence of first-degree relatives (FDRs) with celiac disease detected at screening and diagnostic significance of anti-tissue transglutaminase (anti-TTG). Patients and Methods: We performed a retrospective cohort study of 104 patients with a diagnosis of celiac disease and their FDRs, collecting data from electronic records of Mayo Clinic and celiac disease registry from December 20, 1983, to May 22, 2017. We collected demographics, presenting symptoms, indication for testing, family history, number of other family members screened, biopsy reports, and results of serologic tests. Results: Of 477 FDRs identified, 360 were screened (mean screening rate per family, 79%25%) and 160 FDRs (44.4%) were diagnosed with celiac disease, at a mean age 31.921.6 years (62% female). All diagnosed FDRs had positive anti-TTG titers. Clinical features were documented in 148 diagnosed FDRs, of those 9 (6%) had classic, 97 (66%) had non-classic symptoms, and 42(28%) had no reported symptoms. Histology reports were available from 155 FDRs: 12 (8%) had Marsh 1, 77 (50%) had Marsh 3a, and 66 (43%) had Marsh 3b. A level of anti-TTG greater than or equal to 2.75 of the upper limit of normal identified FDRs with villous atrophy with 87% sensitivity, 82% specificity, and a positive predictive value of 95%. Conclusion: In a retrospective cohort study of patients diagnosed with celiac disease, we found a high prevalence of celiac disease among screened FDRs. High anti-TTG titers associated with villous atrophy on small bowel biopsies, irrespective of symptoms. ª 2019 Mayo Foundation for Medical Education and Research
C
eliac disease (CD) is a common immune-mediated disorder with an estimated prevalence of 1% in the general population.1 Modes of presentation of adults and children with CD can vary from classic (diarrhea predominant) or non-classic with individuals either presenting with complications of CD, extra-intestinal manifestations, as well as truly asymptomatic (or subclinical) individuals picked up through screening high-risk groups.2,3 Recent data suggest that the majority of CD patients are asymptomatic or have non-classic presentation,4,5 confirming the need of having a low screening threshold for CD. In the current guidelines, diagnosis of CD involves screening using highly sensitive serologic markers such as antietissue transglutaminase antibodies (anti-TTG) followed by an endoscopic
n
Mayo Clin Proc. 2019;94(9):1807-1813
assessment and obtaining confirmatory small bowel biopsy (SBB) specimens.6 Furthermore, the 2012 European Society for Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) guidelines suggested that SBB can be omitted in a specific subgroup of symptomatic children with highly positive antiTTG titers,7,8 highlighting the significance of highly sensitive serologic markers in diagnosing and managing CD.9 It is known that certain groups of the population are at higher risk for CD, for example, patients with specific chromosomal disorders such as trisomy 21 and Turner syndrome up to 9%,10,11 autoimmune disorders such as autoimmune thyroiditis, diabetes mellitus type 1 approximately 7% to 10%,12,13 and first-degree relatives (FDRs) of patients with CD up to 11% in previous
Mayo Clin Proc. n September 2019;94(9):1807-1813 n https://doi.org/10.1016/j.mayocp.2019.03.027 www.mayoclinicproceedings.org n ª 2019 Mayo Foundation for Medical Education and Research
From the Division of Pediatric Gastroenterology and Hepatology (S.S.N., Y.H., I.M.); the Division of Gastroenterology and Hepatology (J.A.M., I.A.); and the Division of Biomedical Statistics and Informatics (J.J.L.), Mayo Clinic, Rochester, Minnesota.
1807
MAYO CLINIC PROCEEDINGS
studies. The relative risk of CD in FDRs is higher in siblings, parents, and children with odds ratios of 2.5, 1.6, and 1, respectively.14 There is disagreement between the pediatric and adult societies’ guidelines in regard to screening at-risk asymptomatic individuals including FDRs. The US Preventive Services Task Force latest recommendations support only screening symptomatic FDRs due to paucity of data to support screening in asymptomatic FDRs.15,16 This contradicts the belief of providers who should be vigilant and actively screen for CD in light of a changing presentation from classic to silent or non-classic presentation. A better detection rate of CD could conceivably prevent long-term complications such as nutritional deficiencies, development of new autoimmune conditions, and small bowel malignancy known to occur in undiagnosed CD cases.12,17,18 Previous studies among FDRs report CD prevalence rates of 5.5%, 8.3%, and 10.9%, and the majority were asymptomatic.19-21 Hence, screening only symptomatic FDRs can decrease the detection rate of CD. Easy accessibility of gluten-free diet (GFD) to the general population may lead high-risk groups such as FDRs of patients with CD to follow a GFD without formal testing or diagnosis. Hence, an alternative, highly sensitive, noninvasive test such as anti-TTG titers may result in more compliance with formal testing. Therefore, in this study the primary aim was to assess the prevalence of CD diagnosis among screened FDRs using anti-TTG serologic markers. A secondary goal was to assess the sensitivity of the anti-TTG serologic markers compared with SBB procedures. PATIENTS AND METHODS Study Design After obtaining study approval from the Mayo Clinic Institutional Review Board using a retrospective cohort of CD subjects between December 20, 1983, and May 22, 2017, as a basis, we identified first-degree familial relationships. The first subject within each family diagnosed was termed index case patient 1808
Mayo Clin Proc.
n
(ICP), whereas the remaining family members were termed FDRs. Data were extracted from two sources: (1) the Mayo Clinic electronic health records and (2) the CD registry. Part of the CD registry data was funded by the National Institutes of Health14 between 2003 and 2006 to collect data on newly diagnosed CD patients and their families, containing both prospectively and retrospectively collected data. Adult patients were included if they met the American College of Gastroenterology guidelines, and children were included if they met North American Society for Pediatric Gastroenterology, Hepatology and Nutrition or ESPGHAN guidelines. Eleven ICPs were diagnosed by either endomysial antibodies immunoglobulin A (IgA) (EMA) or anti-gliadin antibodies. Only five FDRs and one ICP were diagnosed as per the ESPGHAN guidelines. Histologic findings were classified into Marsh1 (>30 intraepithelial lymphocytes [IELs] with normal crypts and villi), Marsh 3a (>30 IELs, crypts hyperplasia, and mild villous atrophy [VA]) and Marsh 3b (>30 IELs, crypts hyperplasia, and marked VA).22 Patients with Marsh 1 (IELs) were included if they had more than one positive celiac serology marker (anti-TTG and EMA) and resolution of their symptoms and serologic markers in response to GFD. Patients were grouped based on age and classified as a child (<18 years old) or adult (18 years old). Only patients with appropriate consent based on their source of identification were included. The following data were abstracted in both ICPs and FDRs: date of birth, sex, date of CD diagnosis (biopsy date), age at diagnosis of CD, presenting symptoms, indication for testing, family history, number of other family members screened with their serology results, biopsy reports, and human leukocyte antigen positivity (if tested). Symptoms were classified as classic if reports of weight loss and diarrhea or failure to thrive in children and weight loss and diarrhea in adults were identified. Symptoms were considered nonclassic if reports of nonspecific abdominal pain, bloating, flatulence, constipation, headache, poor dental enamel, anemia, and osteoporosis were identified in the medical records.
September 2019;94(9):1807-1813
n
https://doi.org/10.1016/j.mayocp.2019.03.027 www.mayoclinicproceedings.org
DIAGNOSING CELIAC DISEASE AMONG FIRST-DEGREE RELATIVES
Patients were classified as asymptomatic if there were no reported symptoms at the time of CD screening. Statistical Analysis Patient characteristics were summarized for ICPs and FDRs, both overall and by age or relationship to the ICP. Age, number of FDRs, and percent of FDRs screened were presented as mean (SD). Different kits of anti-TTG (human recombinant TTG Elisa and Red Cell TGG, both from Inova, San Diego, CA) with different cutoffs were used in this period of the study, therefore the number of times anti-TTG titers were above the upper limit of normal (ULN) was presented using median (interquartile range [IQR]), whereas categoric findings were presented using counts and percentages. The KruskalWallis test was used for differences in the number of times anti-TTG titers were above the ULN across various groups. Categoric factors were evaluated using the c2 test of independence. Among FDRs, the receiver operating characteristic (ROC) method was used to classify CD positivity according to the number of times the anti-TTG titers were above the ULN. The ROC method helps assess the discriminatory ability of a continuous measure based on total area under the curve. We also used the ROC method to identify a single anti-TTG titer threshold that optimized the sensitivity and specificity with and without IgA-deficient patients. We used a method of estimating the optical threshold which minimized the Euclidean distance relative to the point denoting perfect classification P values less than .05 were considered statistically significant. Analyses were performed using SAS version 9.4M5 (SAS Institute Inc; Cary, NC) and R version 3.4.2 (Vienna, Austria).
ICPs (n=104) All underwent SBB except 1
All FDRs (n=477) 117 FDRs did not undergo CD screening Screened FDRs (n=360) 182 FDRs tested negative on anti-TTG serology Anti-TTG positive FDRs (n=178) 18 anti-TTG positive FDRs had negative SBB reports
Diagnosed FDRs (n=160) All underwent SBB except 5
RESULTS Patients We identified 104 CD ICPs, with mean age (SD) of 30.8 (19.7) years (8.2 [4.9] years among children; 43.4 [11.8] years among adults). From these 104 ICPs, we identified Mayo Clin Proc. n September 2019;94(9):1807-1813 www.mayoclinicproceedings.org
477 FDRs (all FDRs) of whom 360 (75.5%) were screened for CD (screened FDRs). Among the 360 screened FDRs, 145 (40.3%) were siblings, 137 (38.1%) offspring, and 78 (21.7%) parents. Three hundred fifty-five of 360 (98.6%) FDRs were screened because of a family history of CD. On average, we identified 4.6 (2.8) FDRs per family but only 3.4 (2.3) were screened, which equates to a mean per family screening rate of 79% (25%). Among the 360 screened FDRs, 160 (44.4%) FDRs were diagnosed with CD, which are termed diagnosed FDRs (Figure 1). Among the 153 diagnosed FDRs with available data, age at diagnosis was 31.9 (21.6) years (9.0 [4.4] years in 54 children; 44.3 [16.5] years in 99 adults) (Table). Both ICPs and diagnosed FDRs had a similar predominance of female patients with 64 of 104 ICPs (61.5%) and 99 of 160 FDRs (61.9%). The median period of
n
FIGURE 1. First-degree relative (FDR) distribution throughout the screening process. CD ¼ celiac disease; ICP ¼ index case patient; SBB ¼ small bowel biopsy; TTG ¼ tissue transglutaminase.
https://doi.org/10.1016/j.mayocp.2019.03.027
1809
MAYO CLINIC PROCEEDINGS
TABLE. Demographics and Clinical Characteristics of ICPs and CD-Diagnosed FDRsa Demographics and clinical characteristics
Index (n¼104)
Diagnosed FDRs (n¼160)
b
Age, mean (SD), y
31.9 (21.6) (n¼153)b
30.8 (19.7) (n¼92)
Female
64 (61.5)
99 (62)
Symptoms Classic Non-classic Asymptomatic
b
(n¼88) 10 (11.4) 67 (76.1) 11 (12.5)
(n¼148)b 9 (6.1) 97 (65.5) 42 (28.4)
Diagnostic serology assessment Anti-TTG positive Other serologyc
(n¼93)b 82 (88.2) 11 (11.8)
(n¼160)b 160 (100) 0
Anti-TTG titer ULN, median (IQR)
12.4 (4.2-16.7) (n¼73)b
9.5 (5.0-12.5) (n¼143)b
b
Histology Marsh 1 Marsh 3a Marsh 3b
(n¼155)b 12 (7.7) 77 (49.7) 66 (42.6)
(n¼93) 11 (11.8) 37 (39.8) 45 (48.4)
anti-TTG ¼ anti-tissue transglutaminase; CD ¼ celiac disease; FDR ¼ first-degree relative; ICP ¼ index case patient; IQR ¼ interquartile range; ULN ¼ upper limit of normal. b Data on this characteristic was not available on all subjects so we have provided the sample size for this specific characteristic. c Other serology such as endomysial antibodies immunoglobulin A or anti-gliadin antibodies. a
Values are presented n (%) unless otherwise stated.
Clinical Features Among ICPs and Diagnosed FDRs Symptom records were available among 88 ICPs and 148 FDRs (Table). Among the ICPs, classic presentation was reported in 10 (11.4%) patients, non-classic in 67 (76.1%), and 11 (12.5%) were silent (asymptomatic). Family history was the indication for CD screening in the majority of the screened FDRs. Clinical presentation was documented in 148 of 160 (93%) of the diagnosed FDRs, of whom many recalled symptoms during the screening for CD diagnosis: 9 (6.1%) had classic presentation, 97 (65.5%) had nonclassic presentation, and 42 (28.4%) were asymptomatic. No follow up data was available on the screened FDRs who had negative screening to compare their symptoms with the diagnosed FDRs. Anti-TTG Serology Of the 360 patients screened, 178 FDRs (49.4%) had positive anti-TTG serology. Of 178 patients who were positive for antiTTG serology, 160 were diagnosed FDRs with CD (149 tested positive for anti-TTG 1810
Mayo Clin Proc.
n
IgA and 11 were found to be IgA-deficient but anti-TTG IgG positive). The numerical value of the anti-TTG titers was available in 143 of 160 diagnosed FDRs and 73 of 104 ICPs, which allowed us to calculate the TTG value above the ULN (Figure 2). We found that diagnosed FDRs had a median
1.0
0.8
Sensitivity
CD diagnosis between ICPs and diagnosed FDRs was 192 days (IQR, 34.0-634.0 days).
0.6
0.4
0.2 AUC=0.901 (95% CI: 0.838-0.964) 0.0 1.0
0.8
0.6 0.4 Specificity
0.2
0.0
FIGURE 2. Receiver operating curve for predicting partial or total villous atrophy using the number of times beyond upper limit of normal for tissue transglutaminase value. AUC ¼ area under the curve.
September 2019;94(9):1807-1813
n
https://doi.org/10.1016/j.mayocp.2019.03.027 www.mayoclinicproceedings.org
DIAGNOSING CELIAC DISEASE AMONG FIRST-DEGREE RELATIVES
ULN value of 9.5 (IQR, 5.0-12.5) whereas among the ICPs the median ULN value was 12.5 (IQR, 4.2-16.7). When comparing children and adult diagnosed FDRs, we noted a significantly higher median anti-TTG ULN titer among children compared with adults (12.5 versus 8.5; P¼.003). Histology Findings Among ICPs and Diagnosed FDRs Histology reports were available in 93 of 104 ICPs (10 had no pathology reports and 1diagnosed without SBB as per ESPGHAN guidelines) and all FDRs who underwent confirmatory SBB (155 of 160; 5 patients were diagnosed without SBB as per ESPGHAN guidelines). Among the ICPs, we found 11 (11.8%) had Marsh 1, 37 (39.8%) had Marsh 3a and 45 (48.4%) had Marsh 3b. The FDR group had a similar biopsy results distribution as the ICP group: 12 (7.7%) had Marsh 1, 77 (49.7%) had Marsh 3a, and 66 (42.6%) had Marsh 3b. On comparing children to adult-diagnosed FDRs, distributions of histologic findings were not different across the groups: 4 (8.2%) had Marsh 1, 27 (55.1%) had Marsh 3a, and 18 (36.7%) had Marsh 3b; adults had 7 (7.1%), 48 (48.5%), and 44 (44.4%), respectively (P¼.67). Among the diagnosed FDRs, using the ROC method, we found that anti-TTG (IgA and IgG) levels of 2.75 greater than or equal to ULN predicts VA with 87% sensitivity, 82% specificity, 95% positive predictive value, and 61% negative predictive value (area under the curve, 0.90) (Figure 2). Similar results were noted after excluding 10 ICPs and 11 FDRs who were IgA deficient (88% sensitivity, 85% specificity, 96% positive predictive value, and 65% negative predictive value with an area under the curve of 0.90). We also compared the mean anti-TTG ULN values among patients with VA (Marsh 3a and 3b) and without VA (Marsh 1) in two groups: ICPs and diagnosed FDRs. Among ICPs, no difference in median antiTTG ULN value was seen between those without VA (4.5; IQR, 2.0-12.5) and those with VA (12.5; IQR, 5.4-18.5; P¼.11). However, the diagnosed FDRs with VA had a Mayo Clin Proc. n September 2019;94(9):1807-1813 www.mayoclinicproceedings.org
n
higher median anti-TTG ULN value (9.5; IQR, 5.0-12.5) than those without VA (0.8; IQR, 0.3-2.7; P<.001). Symptoms did not associate with the degree of VA in either ICPs (P¼.66) or diagnosed FDRs (P¼.41). DISCUSSION Our study found high prevalence of CD among screened FDRs with 160 of 360 (44.4%), when more than two-thirds of FDRs were screened overall. An FDR was diagnosed in a median period of approximately 6 months of a diagnosed ICP. Among the diagnosed FDRs, a correlation between high anti-TTG titers and VA independent of their symptoms was noted and a majority showed VA on SBB. Non-classic and silent (asymptomatic) presentation predominated among diagnosed FDRs. We also found that high positive anti-TTG titers at approximately 2.75 ULN were a predictor of VA on SBB in our cohort. Children had higher antiTTG ULN titers than adults. Histologic findings among ICPs and FDRs (both children and adults) were similar. The new recommendations of the US Preventive Services Task Force suggest that only symptomatic FDRs should be tested.16 In our study, we found a high prevalence of CD in screened FDRs (44.4%) with more than one-quarter being asymptomatic (28%), which is higher than the prevalence reported by previous studies ranging from 5.5% to 11%.19e21,23 This higher prevalence of diagnosed FDRs with CD in our cohort could be due to tertiary center referral bias and proactive testing among FDRs which was higher in our cohort (75.5%) compared with previous studies. Increasing incidence of CD among high-risk CD genotypes was recently reported, which supports the high prevalence noted in our study.24 We observed a larger number of symptomatic than asymptomatic cases among CD-diagnosed FDRs, which is likely due to the recall bias on further questioning of symptoms before CD diagnosis. However, overall asymptomatic presentation was more common in FDRs when compared with ICPs, suggesting earlier diagnosis due to early and proactive screening. Our results
https://doi.org/10.1016/j.mayocp.2019.03.027
1811
MAYO CLINIC PROCEEDINGS
are consistent with previous studies among FDRs that showed higher rates of asymptomatic CD than symptomatic CD25 and lack of association between symptoms and the degree of VA, whereas higher anti-TTG titer correlated with a higher degree of VA.26 Several studies have shown that antiTTG is a test of choice for CD screening among symptomatic and asymptomatic CD patients.27e29 In our study, an anti-TTG value of 2.75 ULN predicted the presence of VA with sensitivity and specificity of 87% and 82%, respectively. Based on high prevalence of positive anti-TTG titers between screened FDRs noted in our cohort and previous studies, we suggest that FDRs must be screened regardless of symptoms.30 FDRs with highly positive anti-TTG titers (>3 UNL) will more likely have VA on SBB. Furthermore, gastroenterologists and general practitioners should ask about family history of CD among their patients’ FDRs and consider screening them for CD if present. The limitations of our study include the retrospective nature of our study, referral bias associated with a tertiary care center, and limited follow-up data on patients who were lost to follow-up or elected to seek medical care locally. Hence, prospective studies are required to confirm the predictive value of anti-TTG titers as a diagnostic test among FDRs and the possibility of foregoing the need for confirmatory SBB. The strengths of the study are that data were partly prospectively collected and the median period of an FDR diagnosed with CD was within 6 months of identifying an ICP. CONCLUSION In a retrospective cohort study of patients diagnosed with CD, we found a high prevalence of CD among screened FDRs. High anti-TTG titers associated with VA on SBB, irrespective of symptoms. We also found a positive association between high anti-TTG titers and the presence of VA, supporting the need for screening all FDRs, including those who are asymptomatic. Larger multicenter prospective studies are required to confirm anti-TTG titers as a diagnostic test among FDRs. 1812
Mayo Clin Proc.
n
Abbreviations and Acronyms: anti-TTG = antietissue transglutaminase; CD = celiac disease; EMA = endomysial antibodies immunoglobulin A; ESPGHAN = European Society for Pediatric Gastroenterology, Hepatology and Nutrition; FDR = first-degree relative; ICP = index case patient; IEL = intraepithelial lymphocyte; Ig = immunoglobulin; IQR = interquartile range; ROC = receiver operating characteristic; SBB = small bowel biopsy; ULN = upper limit of normal; VA = villous atrophy
Potential Competing Interests: The authors report no competing interests. Correspondence: Address to Imad Absah, MD, Division of Pediatric Gastroenterology and Hepatology, Mayo Clinic, 200 First Street SW, Rochester, MN 55905 (absah.imad@ mayo.edu).
REFERENCES 1. Rubio-Tapia A, Ludvigsson JF, Brantner TL, Murray JA, Everhart JE. The prevalence of celiac disease in the United States. Am J Gastroenterol. 2012;107(10):1538-1544. quiz 1537, 1545. 2. Fasano A. Clinical presentation of celiac disease in the pediatric population. Gastroenterology. 2005;128(4 suppl 1):S68-S73. 3. Green PHR. The many faces of celiac disease: clinical presentation of celiac disease in the adult population. Gastroenterology. 2005;128(4 suppl 1):S74-S78. 4. Ludvigsson JF, Rubio-Tapia A, van Dyke CT, et al. Increasing incidence of celiac disease in a North American population. Am J Gastroenterol. 2013;108(5):818-824. 5. Almallouhi E, King KS, Patel B, et al. Increasing incidence and altered presentation in a population-based study of pediatric celiac disease in North America. J Pediatr Gastroenterol. 2017; 65(4):432-437. 6. Hill ID, Dirks MH, Liptak GS, et al. 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):1-19. 7. Husby S, Koletzko S, Korponay-Szabo IR, et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines for the diagnosis of coeliac disease. J Pediatr Gastroenterol Nutr. 2012;54(1):136-160. 8. Lock RJ, Pitcher MC, Unsworth DJ. IgA anti-tissue transglutaminase as a diagnostic marker of gluten sensitive enteropathy. J Clin Pathol. 1999;52(4):274-277. 9. Klapp G, Masip E, Bolonio M, et al. Celiac disease: the new proposed ESPGHAN diagnostic criteria do work well in a selected population. J Pediatr Gastroenterol Nutr. 2013;56(3): 251-256. 10. Mårild K, Størdal K, Hagman A, Ludvigsson JF. Turner syndrome and celiac disease: a case-control study. Pediatrics. 2016;137(2): e20152232. 11. Mazurek D, Wyka J. Down syndromedgenetic and nutritional aspects of accompanying disorders. Rocz Panstw Zakl Hig. 2015; 66(3):189-194. 12. Canova C, Pitter G, Ludvigsson JF, et al. Celiac disease and risk of autoimmune disorders: a population-based matched birth cohort study. J Pediatr. 2016;174:146-152.e1. 13. Mahmud FH, Murray JA, Kudva YC, et al. Celiac disease in type 1 diabetes mellitus in a North American community: prevalence, serologic screening, and clinical features. Mayo Clin Proc. 2005;80(11):1429-1434. 14. RubioeTapia A, Van Dyke CT, Lahr BD, et al. Predictors of family risk for celiac disease: a population-based study. Clin Gastroenterol Hepatol. 2008;6(9):983-987.
September 2019;94(9):1807-1813
n
https://doi.org/10.1016/j.mayocp.2019.03.027 www.mayoclinicproceedings.org
DIAGNOSING CELIAC DISEASE AMONG FIRST-DEGREE RELATIVES
15. Hill ID, Fasano A, Guandalini S, et al. NASPGHAN clinical report on the diagnosis and treatment of glutenrelated disorders. J Pediatr Gastroenterol Nutr. 2016; 63(1):156-165. 16. US Preventive Services Task Force, Bibbins-Domingo K, Grossman DC, et al. Screening for Celiac disease: US Preventive Services Task Force recommendation statement. JAMA. 2017;317(12):1252-1257. 17. Theethira TG, Dennis M, Leffler DA. Nutritional consequences of celiac disease and the gluten-free diet. Expert Rev Gastroenterol Hepatol. 2014;8(2):123-129. 18. Rubio-Tapia A, Kyle RA, Kaplan EL, et al. Increased prevalence and mortality in undiagnosed celiac disease. Gastroenterology. 2009;137(1):88-93. 19. Farre C, Humbert P, Vilar P, et al. Serological markers and HLADQ2 haplotype among first-degree relatives of celiac patients. Dig Dis Sci. 1999;44(11):2344-2349. 20. Högberg L, Fälth-Magnusson K, Grodzinsky E, Stenhammar L. Familial prevalence of coeliac disease: a twenty-year follow-up study. Scand J Gastroenterol. 2003;38(1):61-65. 21. Bonamico M, Ferri M, Mariani P, et al. Serologic and genetic markers of celiac disease: a sequential study in the screening of first degree relatives. J Pediatr Gastroenterol Nutr. 2006; 42(2):150-154. 22. Oberhuber G, Granditsch G, Vogelsang H. The histopathology of coeliac disease: time for a standardized report scheme for pathologists. Eur J Gastroenterol Hepatol. 1999;11(10):11851194.
Mayo Clin Proc. n September 2019;94(9):1807-1813 www.mayoclinicproceedings.org
n
23. Hoffenberg EJ, Bao F, Eisenbarth GS, et al. Transglutaminase antibodies in children with a genetic risk for celiac disease. J Pediatr. 2000;137(3):356-360. 24. Liu E, Dong F, Barón AE, et al. High Incidence of celiac disease in a long-term study of adolescents with susceptibility genotypes. Gastroenterology. 2017;152(6):1329-1336. 25. Tommasini A, Not T, Kiren V, et al. Mass screening for coeliac disease using antihuman transglutaminase antibody assay. Arch Dis Child. 2004;89(6):512-515. 26. Hawamdeh H, Al-Zoubi B, Al Sharqi Y, Qasrawi A, Abdelaziz Y, Barbar M. Association of tissue transglutaminase antibody titer with duodenal histological changes in children with celiac disease. Gastroenterol Res Pract. 2016;2016. https:// doi.org/10.1155/2016/6718590. erratum 2019;2019:2607194. 27. Troncone R, Maurano F, Rossi M, et al. IgA antibodies to tissue transglutaminase: an effective diagnostic test for celiac disease. J Pediatr. 1999;134(2):166-171. 28. Sulkanen S, Halttunen T, Laurila K, et al. Tissue transglutaminase autoantibody enzyme-linked immunosorbent assay in detecting celiac disease. Gastroenterology. 1998;115(6):1322-1328. 29. Hopper AD, Hadjivassiliou M, Hurlstone DP, et al. What is the role of serologic testing in celiac disease? A prospective, biopsyconfirmed study with economic analysis. Clin Gastroenterol Hepatol. 2008;6(3):314-320. 30. Trovato CM, Montuori M, Anania C, et al. Are ESPGHAN “biopsy-sparing” guidelines for celiac disease also suitable for asymptomatic patients? Am J Gastroenterol. 2015;110(10): 1485-1489.
https://doi.org/10.1016/j.mayocp.2019.03.027
1813