Background: Continuous positive airway pressure(CPAP) is doubtlessly using as a medical treatment of choice for patients with obstructive sleep apnea (OSA) syndrome. CPAP is effective in OSA patients as a physical "pneumatic pressure splint" mechanism. We have done this study for two purposes, first to seek for the factors to determine the optimal CPAP titer, second to predict the minimal CPAP titer using the determined factors. Methods: We studied a 72 OSA patients who were treated with CPAP. All of them were studied by using a two nights polysomnographic tests in hospital. We compared the patients requiring CPAP over 10cmH2O with those who required CPAP under 5cm H2O to determine the factors affecting the minimal CPAP titer. Results: The high CPAP group is characterized by a significantly higher body mass index(BMI), apnea index(AI) and apnea and hyponea index(AHI) and significantly lower lowest SaO2. Regression analysis using the optimal four variables resulted in the following prediction equation for CPAPtiter. CPAPtiter=8.382 + 0.064 x BMI + 0.077 x AI - 0.004 x AHI - 0.077 x lowest SaO2 When this regression equation was applied to the 72 patients, the mean CPAP titer as predicted by the above equation was 7.80+/-2.96 mmHg. Compared this value with actually determined CPAP titer, 7.93+/-4.00mmHg, there was no significant difference between the two values. Conclusion: Obesity, apnea severity and lowest SaO2 were strongly correlated with CPAP titer. Linear regression equation for CPAP titer using these indices predicted very closely the actually measured values in the sleep laboratory.