Predictors for late tolerance development in food protein-induced allergic proctocolitis.

Background

Food protein-induced allergic proctocolitis (FPIAP) is a non-immunoglobulin E (IgE) mediated food allergy that typically presents with blood-mixed mucoid stool.

Objective

To identify the predictors that affect the tolerance development in infants with FPAIP and laboratory as well as clinical differences between patients with early and with late tolerance.

Methods

A total of 185 infants with FPIAP were included. The patients were grouped and analyzed based on laboratory tests and clinical characteristics.

Results

The median (interquartile range [IQR]) age of onset of symptoms was 2.0 months (1.0-3.0 months). Symptoms began in severe cases in patients (n = 23) at a younger median (IQR) age (1.5 months [0.7-2.0 months]) than the group with nonsevere presentation (median 2.0 months [IQR 1.5-3.0 months]) (p < 0.001). The frequency of neutropenia (<1500/mm³) (p = 0.045) and eosinophilia (450 mm³) (p = 0.018) was increased in severe cases. Concomitant IgE-related food allergy (odds ratio [OR] 3.595 [95% confidence interval {CI}, 1.096-11.788], p = 0.035), non-IgE-mediated multiple food allergy (OR 3.577 [95% CI, 1.595-8.018], p = 0.002), feeding with cow's milk-based formula (at least once during infancy) (OR 2.517 [95% CI, 1.188-5.333], p = 0.016), and late complementary feeding (OR 5.438 [95% CI, 2.693-10.981], p < 0.001) were the predictors for late tolerance development. The estimated optimal cutoff value for introduction of complementary foods for the resolution of allergy was 5.5 months, with 69.4% sensitivity, 74.4% specificity, and an area under the curve of 0.737 (95% CI, 0.626-0.812) (p < 0.001).

Conclusion

This study showed that the early introduction of complementary feeding accelerates tolerance development in FPAIP. A longer duration of an elimination diet has no impact on the resolution of allergy. Physicians should consider conservative avoidance measures and earlier introduction of complementary feeding in FPIAP.