Background: pH monitoring is the standard diagnostic tool for gastroesophageal reflux in infants. However, this method does not document the reflux of all kinds of fluid from the stomach into the esophagus, but only documents acid material. The parameters that define reflux episodes by pH monitoring have been derived empirically from observations of many infants considered healthy and ill. Acid reflux is a continuum, some reflux is normal and doubt exists as to how much reflux is abnormal. In this study, one of the standardized protocols for analyzing pH recordings was evaluated and compared with simultaneously obtained intraesophageal impedance measurement (IMP), a pH-independent method of detecting bolus movement within the esophagus.
Methods: The esophagi of 50 infants with reflux symptoms were measured, using both standard pH probe and multiple-site impedance measurement. A standard protocol for analyzing esophageal pH records was used. The sampling rate for pH values was 15/min. Acid reflux was defined as pH less than 4.0 (threshold pH) for at least 15 seconds (minimal duration) with at least 30 seconds (latency time) between separate episodes. The software used could adjust independently or in combinations the sampling rate and these reflux criteria. Thereby it was determined whether changes in the criteria for acid reflux improved the sensitivity and predictive value of pH monitoring when compared with reflux episodes defined by IMP.
Results: During 318 hours of recording in 50 infants, 1,887 episodes of reflux occurred according to IMP. Only 282 (14.9%) of the IMP-determined episodes were acid reflux episodes. No alkaline reflux episodes occurred. Among the 270 pH probe-determined episodes using the standard criteria of acid reflux, only 153 (sensitivity, 54.3%; positive predictive value, 56.7%) were accompanied by unmistakable retrograde bolus movement using IMP measurements. Retrograde bolus movement did not accompany the other 117 episodes. Using a sampling rate of 15/min, a pH threshold of 4.0, a minimal duration of reflux episodes of 8 seconds, and a latency time of 60 seconds, the positive predictive value of pH probe results increased to 60.7%. Variations in the sampling rate or criteria for defining acid reflux did not significantly improve the accuracy of the pH probe results versus IMP-defined episodes.
Conclusions: Most reflux episodes that occur in infants are undetectable by standard pH probe monitoring. pH monitoring does not detect all reflux in the esophagus but is useful for detecting acidity in the esophagus and determining the duration of its presence. Combining pH monitoring with impedance measurement is a valuable diagnostic tool for gastroesophageal reflux in infants.