The optimal therapy in the first episode of primary spontaneous pneumothorax remains still controversial. We aimed to identify high-risk primary spontaneous pneumothorax patients of recurrence and compared the recurrence rate between surgical and conservative treatments after the first episode of primary spontaneous pneumothorax. Between 2015 and 2019, 294 primary spontaneous pneumothorax patients were analyzed. Patients were divided two groups according to the first treatment: conservative group (n = 177, chest tube treatment or nasal oxygen) and surgery group (n = 117, chest tube treatment followed by surgery). The predictors for recurrence were investigated. There was a statistically significant difference between two groups in terms of pneumothorax volume (p < 0.001) and size (p < 0.001), presence of bullae/blebs (p < 0.001), and high-grade dystrophic severity score (p < 0.001). Recurrence rate was statistically higher in the conservative group (n = 62, 35.0%) than in the surgery group (n = 6, 5.1%) (p < 0.001). According to multiple logistic regression analysis, large pneumothorax (p = 0.008), bullae/blebs (p = 0.001), and continued smoking after the first episode of primary spontaneous pneumothorax (p < 0.001) were independent risk factors for recurrence in patients in the conservative group. In the surgery group, the rate of recurrence was significantly higher in patients who continued to smoke (13.2% versus 0%, p = 0.01). Patients who have a large pneumothorax and presence of a bulla/bleb may benefit surgery in the first episode of primary spontaneous pneumothorax. Efforts should be directed at smoking cessation after the first episode of primary spontaneous pneumothorax.