Detection of gastro-oesophageal reflux in the neonatal unit

Aim: To determine whether a pH probe or multichannel intraluminal impedance (MII) more frequently detected gastro-oesophageal reflux and test the hypothesis that add reflux was associated with lower baseline impedance. Methods: A prospective study of infants in whom reflux was suspected and evaluated using combined pH and multichannel impedance. Studies were considered abnormal if the acid index was >10% or there were >79MII reflux events in 24 hours. The acid index was the percentage of total study time with a pH<four and the acid clearance time (ACT) the time from the pH falling below four to rising above four. Results: Forty-two infants [median gestational age 31 (range 23-42) weeks] were assessed. Only nine infants (21%) had abnormal studies, seven detected by pH monitoring, one by MII monitoring and one by both techniques (p= 0.04). After correcting for gestational age and post-natal age, baseline impedance remained negatively correlated with the acid index (r = -0.34, p = 0.038) and the maximum ACT (r = -0.44, p = 0.006). Conclusion: Clinical suspicion of reflux was frequently incorrect, and reflux was more frequently detected by a pH probe. The inverse relationship of acid reflux to baseline impedance suggests that mucosal disruption may result from acid reflux in this population.