Intestinal permeability in children: variation with age and reliability in the diagnosis of cow's milk allergy.

OBJECTIVE: To analyse to what extent age may alter intestinal permeability (IP) in children and to assess its reliability according to clinical manifestations in cow's milk allergy (CMA).
DESIGN: A routine prospective study was performed in 200 children (12.5+/-23 mo, 0.5-168 mo) presenting with clinical manifestations suggesting CMA. Controls (n = 105) were those with a negative cow's milk challenge, whereas CMA children (n = 95) fulfilled ESPGHAN criteria. Permeability was measured as a percentage of urinary excretion of lactitol (L, %) and mannitol (M, %) (0.1 g/kg for each, oral absorption after a 6 h fast, 5 h urine collection, analysis by gas chromatography) and determination of the L/M ratio (L/M, %).
RESULTS: In control children, L/M correlated negatively with age (r -0.33, p = 0.0006), whereas in those with CMA no correlation was found. Median L/M was significantly higher in CMA children (n = 95) than in controls (n = 105), 4.35+/-7.57% (95% CI 5.30-8.39%) vs 1.97+/-0.87% (95% CI 1.76-2.09%), (p = 0.0001). With a L/M cut-off value defined as mean + 2 SD of controls, in CMA IP exhibited a 68% sensitivity and a 77% negative predictive value. The highest sensitivity (70%) was seen at ages 6-12 mo and the lowest (58%) at age more than 3 y. An abnormal IP was seen in 80% of CMA children with digestive manifestations, in 43% with extra-digestive, 68% with mixed and 40% with anaphylactic manifestations.
CONCLUSION: IP correlates negatively with age in control children and is altered in children with CMA. The test is at its most accurate in the diagnosis of CMA when done at ages 6-12 mo, when there are digestive manifestations.