OBJECTIVE Although upper airway surgery in selected patients with obstructive sleep apnea (OSA) has been shown to be beneficial, its long-term effects have been questioned. The main objective was to evaluate whether results following surgery remain stable over time, both in objective and subjective terms. As a secondary aim, such stability was also measured in relation with the type of surgery performed. METHODS This work constitutes a retrospective study of OSA adult patients subjected to the following surgical procedures: different types of pharyngoplasties, tongue-base surgery, partial epiglottectomy or hyoid suspension. Those who exclusively underwent tonsillectomy or nasal surgery were excluded. Before surgery, a sleep study, and an assessment of the patients' sleepiness and quality of life were performed, which were repeated at 8, 34, and 48 months after surgery. A total of 153 patients was included. RESULTS Following surgery, the apnea-hypopnea index decreased from 34.84/h to 14.54/h and did not vary more than one point in subsequent controls (p = 0.01). The oxygen desaturation index changed from 31.02/h to 14.0/h and remained stable in the second (15.34/h) and third (11.43/h) controls (p = 0.01). Parameters measuring sleepiness and well-being demonstrated the maintenance of long-term benefits. New pharyngoplasties were observed to be more stable than classic pharyngoplasties in the long term (p = 0.04). Single-level surgeries were found to be more stable than multilevel surgeries, although a statistically significant difference was not observed (p = 0.07). CONCLUSIONS The benefits obtained remained stable in the long term. In our sample, modern pharyngoplasty techniques showed superiority over the classic ones regarding long-term stability.