Small bowel mucosa atrophy in a cohort of pediatric patients with serum anti-transglutaminase low titer and negative anti-endomysial antibodies

Background. Celiac disease (CD) is classically characterized by the serological presence of autoantibodies such as IgA anti-tissue transgutaminase 2 antibodies (anti-tTG2 Abs), IgA endomysial antibodies (EMA) and by different degrees of intestinal atrophy, irrespective of symptoms. Anti-tTg2 Abs are more sensitive but slightly less specific than EMA. Currently, based on the most recent ESPGHAN guidelines, there is a simplified diagnostic algorithm for patients with elevated antibody titers (>10 upper normal limit). However, the diagnostic work-up in patients with positive, but persistently non-elevated serum titers, however, is not unambiguously supported by guidelines and is often left to the judgment of the single physicians. EMA are considered a fundamental milestone of diagnosis in patients showing positive anti-tTG2. Aims. We aimed at evaluating the small bowel (SB) mucosa histological status in a cohort of pediatric patients showing positive, but persistently non-elevated anti- tTG2, normalized in upper limit normal (ULN) (< 5 ULN). The presence of EMA was also assessed. Results: 281 subjects (mean age: 8.5±3.9 yrs) have been enrolled. Of them, 162 (group A) had positive but persistently non-elevated anti-tTG2 Ab titers (< 5 ULN): 142 (87.7%) (group A1) showed SB mucosa atrophy (Marsh-Oberhuber 3a, 3b, 3c), compatible with CD; 29 among A1 group were repeatedly negative for serum EMA. In 20 patients (group A2) with < 5 ULN anti-tTG2 titers, SB mucosa was normal (Marsh-Oberhuber 0): they were classified as potential CD; in 5 of them serum EMA were negative. In 62 children (Group B) (mean age: 9.2±3.8 yrs) anti-tTG2 titers were high (> 5 ULN): all of them had SB mucosa changes compatible con CD (Marsh-Oberhuber 3a, 3b, 3c), while EMA were negative only in 2. Prevalence of EMA positivity was reported to be higher in high-titer group compared to low-titer patients (Fisher test p=0.0007). Both low and high titer patients with SB mucosa damage were grouped together (CD group: 204 patients): in 31 of them EMA had been repeatedly negative. Conclusions: Children investigated for suspected CD, with persistently non-elevated anti-tTG2 antibody titers, should undergo EGDS as a critical step in their diagnostic work up, irrespective of EMA status