Little is known about pulmonary vascular complications in children with sickle cell disease (SCD). We hypothesized that transfer factor (diffusing capacity of the lung for carbon monoxide [D(LCO)] ) may be used as a surrogate for the size of the pulmonary vascular bed and that pulmonary vascular abnormalities in children with SCD may limit exercise capacity. Fifty stable patients with SCD aged 10 to 18 years and 50 healthy control subjects matched for race and age were recruited. Incremental ergometer cardiopulmonary exercise testing was performed using respiratory mass spectrometry for exhaled gas analysis. A rebreathing maneuver was used to measure functional residual capacity, effective pulmonary blood fl ow (Qpeff), and D(LCO), and helium dilution was used to calculate minute ventilation, oxygen consumption, and CO 2 production. In the 89 evaluable subjects, there were no ventilatory differences between SCD and control subjects. Qpeff was consistently 15% to 20% greater in subjects with SCD than control subjects at all stages, but D(LCO) corrected for both surface area and hemoglobin was only about 7% to 10% greater in subjects with SCD at all stages. As a result, the D(LCO)/Qpeff ratio was considerably lower in SCD at all stages. Arteriovenous oxygen content difference was about one-third less in SCD at all stages. Contrary to our hypothesis, failure to maintain a sufficient Qpeff to compensate for anemia led to exercise limitation. The ratio of pulmonary capillary blood volume to fl ow is reduced throughout, implying subtle pulmonary vascular disease; however, this was not a factor limiting exercise.