Caution about overinterpretation of symptom indexes in reflux monitoring for refractory gastroesophageal reflux disease.

BACKGROUND & AIMS: Symptom index (SI) and symptom association probability (SAP) are indexes used to analyze data collected from ambulatory pH and/or impedance monitoring and quantify the association between symptoms and reflux events. However, their characteristics are not well defined. We measured factors that affect SI and SAP values to determine their utility in assessing patients with refractory gastroesophageal reflux disease (GERD). METHODS: We conducted a cross-sectional study of 254 patients with poor responses to proton pump inhibitor (PPI) therapy. Participants underwent esophagogastroduodenoscopy and wireless pH (n = 127) or impedance/pH monitoring when they were not receiving PPI therapy (n = 41) or impedance/pH monitoring while they received twice-daily PPI therapy (n = 86). SI and SAP values were calculated individually; ranges of values for each cell in the 2 × 2 contingency table were determined. Monte Carlo simulation was conducted to determine how varying reflux and symptom rates within the contingency table impacted the expected value and variability in SI and SAP. RESULTS: At best, only 33% of patients who were refractory to PPI therapy had positive SI or SAP scores for acid or nonacid reflux events. Abnormal SAP (>95%) and SI (>50%) scores required high rates of reflux. At reflux rates less than 10%, observed in 70% of the studied population, SI and SAP values were largely determined by chance occurrences, rather than the relationship between symptoms and reflux. The values for each index varied significantly day-to-day. CONCLUSIONS: SI or SAP indexes can be overinterpreted, unless patients with gastroesophageal reflux disease who are refractory to PPI therapy have high rates of reflux.