Obesity is a strong risk factor for the development and progression of sleep apnea. Responses to exercise by patients with obstructive sleep apnea syndrome (OSAS) are clinically relevant to reducing body weight and cardiovascular risk factors. This study aimed to clarify the aerobic and anaerobic exercise capacities and their possible relationships with other findings in patients with OSAS. Forty patients (30 males, 10 females) and 40 controls (30 males, 10 females) were enrolled in this study. Questionnaires (excessive daytime sleepiness, daytime tiredness, morning headache, waking unrefreshed, and imbalance), overnight polysomnography, indirect laryngoscopy, and aerobic and anaerobic exercise tests were performed. Triceps, subscapular, abdomen, and thigh skinfold thicknesses were measured. Subcutaneous abdominal fat (abdomen skinfold) was significantly higher in OSAS patients than in controls. Maximal anaerobic power and anaerobic capacity were not different significantly between the patients and controls. We found that aerobic capacity was significantly lower in OSAS patients than in controls. Aerobic capacity was negatively correlated with upper-body subcutaneous fat (triceps and subscapular skinfolds) but not correlated with subcutaneous abdominal fat in OSAS patients. In multivariate analyses using all patients, the apnea-hypopnea index remained a significant independent predictor of aerobic capacity after controlling for a variety of potential confounders including body mass index. Our data confirm that central obesity (subcutaneous abdominal fat) is prominent in patients with OSAS. Our results suggest that lower aerobic exercise capacity in patients with OSAS might be due to daily physical activity that is restricted by OSA itself. This study also suggests that the degree of subcutaneous abdominal fat cannot be used for predicting aerobic capacity level. We think that upper-body subcutaneous fat might be suitable for determining the physical fitness of patients with OSAS.