Allergic proctocolitis is a major cause of bloody stool in early infancy. This study was aimed at ascertaining the clinical courses, sigmoidoscopic and histologic findings of allergic proctocolitis. We also analyzed the relationship between peripheral eosinophilia, the age at symptom onset, and sigmoidoscopic and histologic findings.

We reviewed 25 infants retrospectively who had sigmoidoscopy and biopsy performed with a clinical diagnosis of allergic proctocolitis from April 2003 to April 2007.

The mean age at symptom onset was 15.2±13.2 weeks. Fourteen infants (56.0%) were breast fed, one (4.0%) was formula fed, six (24.0%) were on combined formula, and four (16.0%) were on a weaning diet. Peripheral eosinophilia (≥250/mm³) was seen in eighteen infants (75.0%), but total serum IgE was increased only in six (24.0%). Sigmoidoscopic findings were variable from normal (8.0%), erythema or edema (20.0%), lymphoid hyperplasia (8.0%), erosion (12.0%), hemorrhage and ulcer (4.0%) to lymphoid hyperplasia with erosion, hemorrhage, or ulcer (48.0%). Histologic findings showed focal infiltration of eosinophils in lamina propria (96.0%) and crypt epithelium (96.0%). In twenty four infants (96.0%), the number of eosinophils in mucosa was increased by a more than 60/10 high power field. There was a negative correlation between peripheral eosinophilia and the age at symptom onset. Among the twelve breast fed infants, bloody stool disappeared in ten (83.0%) with a maternal elimination diet of major food groups, but two improved spontaneously.

Allergic proctocolitis should be considered as one of the major causes of bloody stool in healthy appearing infants. To confirm the diagnosis it is necessary to perform sigmoidoscopy and biopsy but histologic findings are more informative than sigmoidoscopic findings. Peripheral eosinophilia was prominent in the infants with an early onset of symptoms. Most infants experienced benign courses and recovered with the elimination of causative foods but did not need exclusive food restrictions.