University of Arkansas for Medical Sciences Little Rock, Arkansas
Seemeen Hassan, MD, MS1, Zaheer Ahmed, MD2, Rajiv Chhabra, MD, MRCP3; 1University of Arkansas for Medical Sciences, Little Rock, AR; 2Medical University of South Carolina, Charleston, SC; 3Saint Luke's Hospital; University of Missouri Kansas City School of Medicine, Kansas City, MO
Introduction: We aimed to compare the frequency of acid reflux detection between impedance pH monitoring (IPM; catheter based in most cases) and ambulatory Bravo capsule pH monitoring (ACPM). We also aimed to assess the patient factors associated with the presence of gastroesophageal acid reflux.
Methods: The Clinical Outcomes Research Initiative (CORI) database version V4 administered by NIH/NIDDK was queried from 2008 to 2014. Esophageal pH monitoring procedures performed on symptomatic patients were identified and categorized based on the method of testing. (IPM versus ACPM). Presence of acid reflux was determined by DeMeester score (DS) > 14.72. Primary objective was to compare the mean DS and frequency of acid reflux detection between the two methods. The secondary objectives were to assess the correlation of DS with percentage of acid exposure (total, supine, upright) in cases of IPM, and to assess patient factors associated with presence of acid reflux in all procedures included in the study (table 1). The percentage of acid exposure was calculated using Johnson 1974 and Richter 1992 criteria. Mean DS were compared using the two samples t test. The correlation between DS and percentage of acid exposure was calculated using proportions shown in table 2. Odds ratio and 95% CI were calculated using a multivariate regression model to predict factors associated with the presence of acid reflux.
Results: A total of 704 IPM procedures (mean DS=23) and 264 ACPM procedures (mean DS=18) were included in the analysis. Acid reflux was diagnosed at a higher rate using IPM in comparison with ACPM (58% vs 32%, OR 0.33, 95% CI 0.24 – 0.46, p< 0.001). The correlation between presence of acid reflux by DS, acid exposure by Johnson criteria and acid exposure by Richter criteria is shown in table 2. There was a stronger agreement between DS and both the acid exposure criteria when acid reflux was absent. The correlation was weaker when acid reflux was present, especially when DS and Richter criteria were compared. The multivariate model suggests a statistically significant positive association of the detection of acid reflux with patient ages 30 – 39 years (vs ages 18 to 29), male sex (vs female), and IPM method (vs ACPM method).
Discussion: Impedance pH testing may be more likely to detect acid reflux in symptomatic patients compared with ambulatory capsule based pH testing. There is a strong correlation between DeMeester score and total percentage of acid exposure especially when acid reflux is absent.
Table 1. A list of patient related variables: Assessment of variable association with detection of acid reflux by DeMeester Score. All age categories are compared with age 18 – 29. Gender is compared with female sex. Race is compared with nonwhite race. pH testing is compared with impedance based test.
Table 2. Correlation between DeMeester score and percentage of acid exposure (Total, upright and supine). Percentage of acid exposure is based on Johnson and Richter Criteria. Proportion in cases of acid reflux was calculated by dividing the number of procedures with acid reflux present both by percentage of acid exposure and by DeMeester score / number of procedures with acid reflux present by DeMeester score alone. Proportion in cases of absent acid reflux was calculated by dividing the number of procedures without acid reflux both by percentage of acid exposure and by DeMeester score / number of procedures without acid reflux by DeMeester score alone.
Disclosures: Seemeen Hassan indicated no relevant financial relationships. Zaheer Ahmed indicated no relevant financial relationships. Rajiv Chhabra indicated no relevant financial relationships.