BACKGROUND: Data on Asian patients who present to primary care physicians with dyspepsia are limited. AIM: To determine predictors of response to a trial of proton pump inhibitor therapy. METHOD: One hundred ninety patients presenting to their general practitioners with dyspeptic symptoms but without alarm symptoms, underwent endoscopy and were subsequently treated with 20 mg of esomeprazole twice a day for 2 weeks. Possible predictors of response were assessed before treatment. Dyspeptic symptoms were scored at baseline and at the end of treatment. Excellent response and poor response were defined as end of treatment score of <or=1 and failure to achieve >or=50% reduction in symptom scores. RESULTS: On bivariate analysis male sex and having <or=2 concurrent dyspeptic symptoms predicted excellent response, whereas bloating as most bothersome symptom, >2 dyspeptic symptoms, anxiety, and irritable bowel syndrome predicted poor response. On multivariate analysis, male sex and <or=2 dyspeptic symptoms, and bloating remained significant predictors of excellent and poor responses, respectively. CONCLUSIONS: Predictors of response to proton pump inhibitor trial that could be easily assessed in a primary care clinic were identified. In our population the response was not as favorable as suggested by clinical trials. Frequent overlap with irritable bowel syndrome seems to be an important factor.
BACKGROUND: Data on Asian patients who present to primary care physicians with dyspepsia are limited. AIM: To determine predictors of response to a trial of proton pump inhibitor therapy. METHOD: One hundred ninety patients presenting to their general practitioners with dyspeptic symptoms but without alarm symptoms, underwent endoscopy and were subsequently treated with 20 mg of esomeprazole twice a day for 2 weeks. Possible predictors of response were assessed before treatment. Dyspeptic symptoms were scored at baseline and at the end of treatment. Excellent response and poor response were defined as end of treatment score of <or=1 and failure to achieve >or=50% reduction in symptom scores. RESULTS: On bivariate analysis male sex and having <or=2 concurrent dyspeptic symptoms predicted excellent response, whereas bloating as most bothersome symptom, >2 dyspeptic symptoms, anxiety, and irritable bowel syndrome predicted poor response. On multivariate analysis, male sex and <or=2 dyspeptic symptoms, and bloating remained significant predictors of excellent and poor responses, respectively. CONCLUSIONS: Predictors of response to proton pump inhibitor trial that could be easily assessed in a primary care clinic were identified. In our population the response was not as favorable as suggested by clinical trials. Frequent overlap with irritable bowel syndrome seems to be an important factor.