The role of weight loss in the therapy of obstructive sleep apnea syndrome (OSAS) was investigated in 23 affected patients with various degrees of obesity (body mass index range 26.6-61.0) free of cranio-facial malformations. Weight loss resulted 18.5 +/- 14.7 (s.d.) kg and was significantly correlated with baseline BMI value (r = 0.94; P less than 0.0001). Weight loss significantly reduced the number of apneas + hypopneas per hour of sleep ((A + H)I) from 66.5 +/- 23.0 to 33.0 +/- 26.2 (P less than 0.0001) and improved the mean of oxygen desaturation peaks during apneas (mSaO2) from 81.9 +/- 6.9 to 87.6 +/- 3.9; P less than 0.001). A significant correlation was found between weight loss and changes in the (A + H)I (r = -0.55; P less than 0.01) and the mSaO2 (r = 0.46; P less than 0.05). The (A + H)I significantly improved in both patients who lost more than 10 kg (basal BMI: 42.3 +/- 10.0) and in those who lost less than 10 kg (basal BMI: 30.2 +/- 2.3), whereas the mSaO2 improved only in the former. Obese patients with moderate to heavy ORL pathological findings had worse pretreatment and final OSAS parameters than those with absent or mild ORL lesions. However, both groups showed a significant, although quantitatively different, improvement of the (A + H)I and mSaO2 after weight loss. Compared to those who were cured or improved after the treatment, patients who failed to obtain significant effects on OSAS clinical presentation also had a significantly higher prevalence of ORL pathology. It is concluded that: (1) weight loss improves parameters and clinical presentation of OSAS in the majority of affected obese patients; (2) a relationship exists between the entity of weight loss and that of improvement of the syndrome; (3) weight loss must be encouraged even in patients with mild to moderate overweight; (4) the presence of ORL pathology may represent a confusing factor in the interpretation of the results obtained after weight loss.