Table 3.
Summary of the included studies evaluating TG2 deposits in potential celiac disease (PCD).
| Reference (first author) | Age, median (range) | PCD (n) | PCD: definition | TG2 IgA deposits: prevalence | Diagnostic value (%) | Intensity | Results of follow-up studies and comparison with other markers |
|---|---|---|---|---|---|---|---|
| Tosco (5) | 6 years, 8 months (18 months–16 years) | 106 | TTG or EMA+, Marsh 0–1 | 66/102 | SE: 64.7 SP: ne | – | At 4 years follow-up, 30.8% develop overt CD. TG2 deposits in the first biopsy was the only marker to predict evolution toward villous atrophy |
| Salmi (13) | 40 years (16–81) | 25 | EMA+, Marsh 0–1 | 12/12 | SE: 93 | – | TG2 had the best SE and SP for detecting early developing CD |
| SP: 93 | |||||||
| Maglio (17) | 6 years, 4 months (6 months–16 years) | 109 | TTG+, Marsh 0–1 | 74/109 | SE: 67.9 SP: 87.9 | Intensity significantly weaker in PCD compared to CD | – |
| Tosco (19) | 7 years (2–17) | 28 | TTG+, Marsh 0–1 | 19/28 | SE: 68 | – | – |
| SP: 80 | |||||||
| Kurppa (20) | 11 years (4–17) | 17 | EMA+, Marsh 0–2 | 17/17 | SE: 100 | Less intense in PCD than CD | Seven of eight on a GCD develop overt CD within 2 years |
| SP: 100 | |||||||
| Tosco (30) | 7 years, 1 month (9 months–17 years, 11 months) | 39 | EMA or TTG+, normal histology | 33/39 | SE: 85 SP: 91 | – | – |
SE, sensitivity; SP, specificity; PPV, positive predictive value; NPV, negative predictive value; EMA, anti-endomysium antibodies; TTG, anti-transglutaminase antibodies; PCD, potential CD; ne, not evaluable (no control group).