BACKGROUND AGA IgA II and AGA IgG II have recently been suggested as reliable tools for celiac disease (CD) diagnosis. We compared their utility for diagnosis and monitoring CD in children with that of tTG IgA, an established CD marker. METHODS We studied a cohort of 161 CD and 129 control children in whom CD was histologically confirmed or ruled out. We followed 37 children with CD on a gluten-free diet for 12-84 months. In fasting sera, we measured AGA IgA II, AGA IgG II, and tTG IgA using ELISAs. RESULTS The best sensitivity (92.5%), specificity (97.6%), positive predictive value (98%), and negative predictive value (91.2%) were obtained using tTG IgA. AGA IgG II correctly identified 3 of 3 children with CD with total IgA deficiency who had negative AGA IgA II and tTG IgA results. In children <2 years old without total IgA deficiency, AGA IgG II and tTG IgA performed equally well (sensitivity 96.4% and specificity 100%). AGA IgA II, AGA IgG II, and tTG IgA concentrations diminished significantly (P < 0.0001) after 1 year of a gluten-free diet, reaching values below the cutoff in 87%, 70%, and 51% of cases, respectively. CONCLUSIONS The best available index for diagnosing CD in children was tTG IgA. In infants <2 years old, AGA IgG II performed as well as tTG IgA in cases without total IgA deficiency and allowed detection of CD when total IgA was <0.06 g/L. Gluten-free diet monitoring can be achieved using any of the studied serum markers.