Katia R. Schesquini-Roriz ^1,2* Gloria Maria Fraga Rodríguez ³ Jocelyn Cristina B. Castellanos ³ Laura Martinez-Martinez ⁴ Susana Boronat Guerrero ³ Carlos Rodrigo ⁵ Isabel Badell ³

• ¹ Federal University of Rondonia, Porto Velho, Brazil
• ² Autonomous University of Barcelona, Barcelona, Catalonia, Spain
• ³ Department of Pediatrics, Hospital de la Santa Creu i Sant Pau, Autonomous University of Barcelona, Barcelona, Spain
• ⁴ Department of Immunology, Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de Barcelona, Barcelona, Spain
• ⁵ Pediatric Service, Hospital Germans Trias i Pujol, Badalona, Spain

Introduction: Celiac disease (CD) is an autoimmune enteropathy triggered by gluten ingestion in genetically susceptible individuals. The haplotypes HLA-DQ2 and DQ8, transglutaminase (TGA) antibodies, and biopsy findings are the main tests performed in the evaluation and CD diagnosis. The objective was to establish possible correlations between transglutaminase levels, genetic markers tests, and qualitative intestinal biopsy findings (modified Marsh classification) at the diagnosis. Methods: A retrospective cohort study. The selection criteria were confirmed CD cases with genetic tests performed. Statistical analysis was done mainly through One-way ANOVA, Kendall’s correlation coefficient (T), and linear regression. Results: The study included 112 patients, with a mean age of 6±4 years. All cases were tested to HLA-DQ2, and it was positive in 93%. HLA-DQ8 was tested in 73% of cases and it was positive in 61%. The percentage of negative genetic markers (DQ2/DQ8) was 4.5% for patients tested to both haplotypes. A comparison of DQ2/ DQ8 (positive and negative) with clinical findings and tests performed did not identify any differences for most of the parameters analyzed. Cases of type I diabetes presented significant negative expression for DQ2 (-); p=0.05 and positive expression for DQ8(+); p=0.023. The TGA antibody levels ranged from 18 to 36,745U/ml. An inverse correlation was found between age and TGA-L level (p = 0.043). In 23% of the cases, the TGA levels were greater than 1000 U/ml and presented a moderate positive correlation with the atrophy biopsy profile (T=0.245). Patients with an atrophic biopsy profile (Marsh III) had a moderate positive correlation with growth failure (T=0.218) but a negative correlation with constipation (T= -0.277). Conclusion: In terms of diagnosis tests for CD, transglutaminase levels and age presented an inverse correlation, with the level decreasing as age increased. A moderately positive correlation was found between mean transglutaminase with intestinal atrophy and growth retardation. The genetic test DQ2 was positive for 93% and negative genetic markers (DQ2/DQ8) represented 4.5% of cases studied.