Clinical and histopathological correlation of duodenal biopsy with IgA anti-tissue transglutaminase titers in children with celiac disease.

Abstract

Data correlating anti-tissue transglutaminase (tTG) antibody titers with severity of duodenal involvement is limited. The aim of this study was to correlate IgA anti-tTG antibody titers with symptoms, anthropometric parameters, and duodenal histopathology. Consecutively diagnosed patients of celiac disease as per modified ESPGHAN criteria presenting over a year were enrolled. Demographic data, symptoms, weight-for-age z score (WAZ), height-for-age z score (HAZ), IgA anti-tTG titer, and duodenal histopathology graded as per modified Marsh criteria were recorded. Spearman rank correlation test was used for association between TTG age, WAZ, and HAZ. Receiver operating curve (ROC), sensitivity, specificity, negative predictive value, and positive predictive value were used to obtain anti-tTG cutoff value predictive of Marsh grade 3. One hundred and forty-two patients with celiac disease were evaluated. tTG showed significant correlation with WAZ (r = 0.822, p = <0.001) and HAZ (r = 0.722, p = <0.001) but not with age (r = 0.202, p = 0.066). The median anti-tTG titers rose progressively with higher Marsh grade on histopathology (p = 0.001). The median anti-tTG titer was also significantly higher in patients with classic celiac disease as compared to non-diarrheal celiac disease (144u/mL vs. 27, p = 0.02). Anti-tTG titer of 62.5u/mL was strongly predictive of duodenal histology of Marsh grade 3a and higher with sensitivity, specificity, positive predictive value, and negative predictive value of 95.4 %, 98 %, 93.8 %, and 88.3% respectively. There is a significant correlation between IgA anti-tTG titers and anthropometric parameters and severity of duodenal histopathology. With further validation, strongly positive titers may be sufficient to predict severity of this disease.