Background: Prolonged air leaks (PALs) after lung resection are one of the most common complications in thoracic surgery. Several options are available to treat PALs. The autologous blood patch pleurodesis is commonly used but has not been thoroughly investigated. Materials and methods: We conducted a prospective randomized study including all consecutive patients with PALs after pulmonary resections. Patients were randomized to either having received pleurodesis by injecting 100 mL autologous blood at d 5 and 6 (Group A) or being placed under observation (Group B). Patients from either group undergoing revisions were further investigated by a post hoc analysis and formed Group C. Results: A total of 24 patients were included: 10 patients were randomized to group A and 14 to group B. Six patients (3 from each group) underwent surgical revision and were included in Group C. Groups A and B did not differ in baseline characteristics. The median time to drainage removal was 9 d (range: 5-23 d) in Group A; 9 d (range: 2-20 d) in Group B; and 6 d in Group C (range: 3-10 d), (A/B versus C, P < 0.04; A versus B was not significant). Conclusions: There is no evidence indicating a benefit for blood patch pleurodeses in patients undergoing lung resections and presenting with postoperative PALs for more than 5 d. An early operative closure of postoperative air leakage seems to be more effective.