Predictive factors of early autologous blood pleurodesis for postoperative air leak

Background: The usefulness of autologous blood pleurodesis for air leak after pulmonary resection is well known; however, factors predicting the therapeutic efficacy are poorly understood. Herein, we aimed to examine the predictors of early autologous blood pleurodesis for air leak following pulmonary resection. Methods: Patients who underwent pulmonary resection and autologous blood pleurodesis with thrombin for postoperative air leak between January 2016 and October 2022 were retrospectively analyzed. Patients received 50–100 mL of autologous blood and 20,000 units of thrombin on postoperative days 1–4. If necessary, the same procedure or pleurodesis with other chemical agents was repeated until the air leak stopped. Patients were divided into single-dose and multiple-dose groups based on the number of times pleurodesis had occurred before the air leak stopped and were statistically analyzed. Logistic regression analysis was performed to identify predictors of treatment efficacy. Results: Of the 922 patients who underwent pulmonary resection, 57 patients (6.2%) were included and divided into single-dose (n = 38) and multiple-dose (n = 19) groups. The amount of air leaks was identified as a significant predictor of multiple dosing, with a cutoff of 60 mL/min, in multivariate logistic regression analyses (odds ratio 1.13, 95% CI 1.03–1.24, p = 0.0065). The multiple-dose group showed a significantly higher recurrence of air leak (p = 0.0417). Conclusions: The amount of air leaks after pulmonary resection is the only significant factor predicting whether multiple autologous blood pleurodesis is required, and the recurrence rate of pneumothorax is significantly higher in such cases.