Predictive value of Müller maneuver, cephalometry and clinical features for the outcome of uvulopalatopharyngoplasty. Evaluation of predictive factors using discriminant analysis in 30 sleep apnea patients.

The success rate of uvulopalatopharyngoplasty (UPPP) in the treatment of obstructive sleep apnea is generally only 50-60%. In order to improve this, various predictive factors for the outcome of UPPP were studied, including the Müller maneuver and cephalometry. Thirty unselected consecutive patients with obstructive sleep apnea (median apnea index (AI) 26 apneas/h, range 5-78) underwent UPPP with standard tonsillectomy except in the case of small tonsils and using CO2 laser for the palatal resection. Polysomnographic control 5 months after surgery showed that 63% of all patients had obtained a reduction in AI > or = 50% including an AI < 20 after surgery. Tonsillectomy had no influence on the outcome. Further, the Müller maneuver did not predict the outcome, whereas cephalometry proved to be of good predictive value. Lowered position of the hyoid bone, increased cranio-cervical angle and shortening of the maxilla length were significantly associated with poor results of UPPP, as were overweight, narrowing of the hypopharynx, the severity of sleep apnea, and hypersomnia. However, in a discriminant analysis containing all these variables, the three cephalometric measurements together with hypersomnia were most closely associated with the outcome of UPPP. A predictive model containing these four variables could correctly classify 83% of the patients in the present study. The model had a false positive rate of 7% in predicting a successful outcome. This indicates that cephalometry is mandatory in the selection of UPPP candidates and that a predictive model containing some or all of the four variables may substantially improve the success rate.