A Novel Classification Scheme for Gastroparesis Based on Predominant-Symptom Presentation

AIM: Symptoms of gastroparesis are very diverse. Classifying patients by predominant symptom may improve management strategy. GOAL: To validate a new symptom-predominant classification for gastroparesis using symptom severity and quality-of-life measures. STUDY: Subjects with gastroparesis for >2 months were prospectively enrolled. A physician classified each subject into one of the following: vomiting-predominant, dyspepsia-predominant, or regurgitation-predominant gastroparesis. Subjects also classified themselves independently from the physician. Each subject completed a Patient Assessment of Gastrointestinal Disorders-Symptom Severity Index (PAGI-SYM) and SF-12v2 Health-Related Quality-Of-Life survey. Receiver operating characteristic curves were constructed with sensitivity and specificity of each PAGI-SYM subscale to differentiate subjects into symptom-predominant subgroups. Area under the curve (AUC) was used to compare the receiver operating characteristic curves. Analysis of variance, Cohen's kappa (kappa) statistic, student t test, and Pearson correlation (r) were used. RESULTS: One hundred subjects (87 females, mean 48 y) were enrolled. There was a 78% concordance between physician and subject's classification of gastroparesis with substantial correlation (kappa=0.64). PAGI-SYM nausea/vomiting subscale (AUC=0.79) and PAGI-SYM heartburn/regurgitation subscale (AUC=0.73) were the best in differentiating subjects into vomiting-predominant and regurgitation-predominant gastroparesis, respectively. No subscale was adequate to differentiate dyspepsia-predominant gastroparesis. SF-12v2 total scores significantly correlated with worsening of the total PAGI-SYM scores (r=-0.339 to -0.600, all P<0.001). CONCLUSIONS: There was a substantial agreement between physician and patient using a symptom-predominant gastroparesis classification. Results suggest that a predominant-symptom classification is a valid means to categorize subjects with vomiting-predominant and regurgitation-predominant gastroparesis. Patients with dyspepsia and delayed gastric emptying need further research.