Objective The aim of this study was to evaluate the association between genioglossus activity during sleep onset and the outcome of uvulopalatopharyngoplasty (UPPP) in patients with obstructive sleep apnea. Study Design Case series with planned data collection. Setting Sleep medical center. Subjects and Methods Forty-four patients with obstructive sleep apnea underwent overnight polysomnography with synchronous genioglossus electromyography (GGEMG) with intraoral electrodes. In addition, all patients underwent revised UPPP with uvula preservation and were followed up with polysomnography at least 3 months after surgery. Results Twenty-five patients (56.8%) were responders. Multiple regression analysis revealed that increasing tonsil size (odds ratio [OR], 0.086; P = .038) and higher sleep-onset GGEMG (OR, 0.664; P = .04) were significant predictors for surgical success. The area under the receiver operating characteristic curve was 0.942 (OR, 0.040; P < .001) for those predictors, 0.884 for GGEMG, and 0.848 for tonsil size. Moreover, all patients were divided into 4 groups according to tonsil size and sleep-onset GGEMG. The success rate of patients with tonsil size III or IV and sleep-onset GGEMG >11.20% (maximal GGEMG) was optimal (92.9%, 13 of 14), while the success rate of patients with tonsil size I or II and sleep-onset GGEMG <= 11.20% was 0% (0 of 10). Conclusions Sleep-onset GGEMG and tonsil size are both important in deciding outcomes of UPPP. Patients with tonsil size III or IV and higher sleep onset may be more suitable candidates for UPPP because of the higher probability of surgical success.