Background: Manual aspiration as the initial management of a large pneumothorax in a clinically stable patient has been reported to be safe and effective. However, the effect with smaller needles, the number of aspiration, the indication other than spontaneous pneumothorax and failure factors are unknown. We assessed the effectiveness and failure risk factors of manual aspiration up to three using a 20-or 22-gauge (G) needle in patients with a large, clinically stable pneumothorax.Methods: We included 107 clinically stable patients with large pneumothorax. Patients who were unstable, required a ventilator, underwent chest tube drainage or had an observed small pneumothorax, bilateral pneumothorax, hemopneumothorax, or postoperative pneumothorax were excluded. Up to three aspirations were performed using 20-or 22-G intravenous needles. When the aspiration volume was ≥2,500 mL or lung expansion did not occur, a chest tube was placed.
Results:The first aspiration was successful in 57 patients (53.3%), the second in 16 patients (59.3%), and the third in eight patients (80.0%). No patient experienced any obvious complications or required emergent hospitalization or surgery after aspiration. Aspiration failure was correlated with an inter-pleural distance >20 mm at the level of the hilum (odds ratio: 4.93; 95% confidence interval: 1.49-22.71; P=0.0075), spontaneous secondary pneumothorax (odds ratio: 3.11; 95% confidence interval: 1.14-8.76; P=0.027), and ≤24 h from onset to presentation (odds ratio: 2.95; 95% confidence interval: 1.12-8.26; P=0.028). There were no significant differences in intrathoracic pressure after aspiration or plasma factor XIII levels between patients with resolved and persistent pneumothorax.Conclusions: Manual aspiration up to three times using a small needle might be one of a treatment option in clinically stable patients with any large pneumothorax. Aspiration failure was correlated with an inter-pleural distance >20 mm at the level of the hilum, spontaneous secondary pneumothorax, and ≤24 h from onset to presentation.