This study evaluated the efficacy of convalescent plasma therapy in treating patients with severe acute respiratory syndrome (SARS) at Prince of Wales Hospital in Hong Kong from March 20 to May 26, 2003. Eighty SARS patients were given convalescent plasma, and their outcomes were assessed based on discharge by day 22 following symptom onset or death/hospitalization beyond 22 days. Patients who received convalescent plasma before day 14 of illness had a significantly higher discharge rate (58.3% vs 15.6%; P < 0.001) and those who were PCR positive and seronegative for coronavirus at the time of plasma infusion had an even better outcome (66.7% vs 20%; P = 0.001). The overall mortality rate among patients given convalescent plasma was 12.5%, compared to 17% in the general SARS population. The study suggests that early administration of convalescent plasma may be beneficial in the early phase of SARS, but it also highlights several limitations, including the lack of a randomized control group and the potential risk of transfusion-transmitted infections. Despite these limitations, the findings indicate that further testing of convalescent plasma therapy in the early phase of SARS may be worthwhile.This study evaluated the efficacy of convalescent plasma therapy in treating patients with severe acute respiratory syndrome (SARS) at Prince of Wales Hospital in Hong Kong from March 20 to May 26, 2003. Eighty SARS patients were given convalescent plasma, and their outcomes were assessed based on discharge by day 22 following symptom onset or death/hospitalization beyond 22 days. Patients who received convalescent plasma before day 14 of illness had a significantly higher discharge rate (58.3% vs 15.6%; P < 0.001) and those who were PCR positive and seronegative for coronavirus at the time of plasma infusion had an even better outcome (66.7% vs 20%; P = 0.001). The overall mortality rate among patients given convalescent plasma was 12.5%, compared to 17% in the general SARS population. The study suggests that early administration of convalescent plasma may be beneficial in the early phase of SARS, but it also highlights several limitations, including the lack of a randomized control group and the potential risk of transfusion-transmitted infections. Despite these limitations, the findings indicate that further testing of convalescent plasma therapy in the early phase of SARS may be worthwhile.