Increase of the apnoea-hypopnoea index after uvulopalatopharyngoplasty: analysis of failure.

From 70 patients who had uvulopalatopharyngoplasty (UPPP) operation and a pre- and postoperative sleep registration, we could retrospectively determine the failures and the correlation between variables such as age, gender, body mass index (BMI), earlier or concomitant tonsillectomy, unilevel (uvula-palate-tonsil) or multilevel (base of tongue as well) obstruction during sleep endoscopy and treatment outcome. From 70 patients, the preoperative sleep registration classified 15 social unacceptable snorers and 55 obstructive sleep apnoea syndrome (OSAS) patients. In this study we focused on the OSAS patients. From the 55 OSAS patients, 32 were classified as successful after UPPP, because they had a decreased apnoea-hypopnoea index (AHI) after surgery (</=20). Eight patients had a decreased AHI, but more than 20 apnoeas/hypnoeas per hour. Fifteen patients were identified as UPPP failures with an equal or increased AHI and/or subjective deterioration of snoring. We were unable to find a statistically difference between the two groups with respect to variables such as age, BMI and AHI preoperative (P > 0.56) as between the level of obstruction(s) (P > 0.24). For earlier or concomitant tonsillectomy we found a statistically difference (P > 0.039), but a very small number in the high failure group (n = 8). We conclude that although sleep endoscopy adds to better patient selection and better results, paradoxically, the finding of obstruction on palate-uvula level during sleep endoscopy can still give UPPP failures.