Validation of Esophageal Global Symptom Severity as a Patient-Reported Outcome for Evaluation of Reflux Symptoms

INTRODUCTION: Visual Analog Scales (VAS) are simple, easy for patients to comprehend, and require limited translation. We evaluated the value of esophageal global symptom severity (GSS) measured using VAS in assessing initial reflux symptom burden as compared with other validated questionnaires, esophageal symptom burden, and outcome after reflux management. METHODS: We analyzed pooled data from published historical cohorts of patients undergoing pH-impedance testing for reflux symptoms from 3 continents (North America, Europe, Asia). Univariate (Spearman correlation), multivariable (general linear regression), and receiver operating characteristic analyses were performed to compare GSS with validated symptom instruments including gastroesophageal reflux disease questionnaire (GERDQ), GERD health-related quality of life (GERD-HRQL), Reflux Symptom Index (RSI), and metrics from pH-impedance monitoring per Lyon Consensus 2.0. RESULTS: One thousand two hundred ninety-six patients (mean age 52.0 years, 61.9% female) were included: 937, 197, and 162 from North America, Europe, and Asia, respectively. GSS significantly correlated with GERDQ (R = 0.455), GERD-HRQL (R = 0.440), RSI (R = 0.491), acid exposure time (AET) (R = 0.158), and total reflux episodes (R = 0.161) ( P < 0.0001 for each comparison). The mean GSS was higher with abnormal GERDQ, GERD-HRQL, RSI, pathologic AET, and conclusive GERD per Lyon Consensus ( P < 0.0001 each comparison). On receiver operating characteristic analyses, GSS was noninferior to GERDQ, GERD-HRQL, and RSI in predicting pathologic AET and total reflux episodes, and conclusive GERD. Percentage improvement in GSS after antireflux treatment significantly correlated with change in GERDQ (R = 0.536, P < 0.0001) and treatment satisfaction (R = 0.532 P = 0.0002). On multivariable linear regression analyses, percentage change in GSS remained an independent predictor of both change in GERDQ (β = 0.813, P < 0.0001) and satisfaction with antireflux therapy (β = 1.90, P = 0.0006). DISCUSSION: GSS correlates with other validated reflux questionnaires and discriminates abnormal from normal reflux burden in patients with reflux symptoms. GSS change also reflects reflux treatment outcome and satisfaction. GSS is a useful addition to patient symptom assessment before and after GERD treatment.