RESUMO
BACKGROUND: Although recurrence is included in surgical indications, there is no definitive guideline for managing recurrent spontaneous pneumothorax (SP) according to characteristics of recurrence, such as the type [primary SP (PSP) or secondary SP (SSP)] and recurrence-free interval (RFI). Actually, some patients with early cessation of air leak after closed thoracostomy tubes, a long RFI, or medical comorbidities are often managed conservatively without surgery. However, the validity of this management is unclear. The aims of the present study are to analyze treatment outcomes for recurrent SP according to the type of SP and RFI, and to check the validity of conservative treatment for patients according to type of SP and RFI. METHODS: We included 1,250 consecutive cases (624 right sided and 626 left side cases for the first episode) who were hospitalized and treated from January 2012 to June 2018. To investigate recurrence according to treatment modality (surgical or conservative treatment), we estimated RFI in each group during the observation period. RFI was measured from the completion of treatment to recurrence or last follow-up. We divided patients into two groups [the early (EG) and the late (LG) recurrence group] according to 1-year. Recurrence was defined as a subsequent episode of an ipsilateral SP, while a contralateral SP was regarded as an independent case in the present study. RFI between subgroups was compared using the Kaplan-Meier method with the log-rank test. A P value less than 0.05 (two-sided) was regarded as statistically significant. RESULTS: Recurrence occurred in 47 cases after surgical intervention for the first episode (585 cases). Recurrence occurred in 265 cases after conservative treatment for the first episode of SP (665 cases). For the first episode, the surgical group (SG) had a significantly longer RFI than the conservative group (CG), regardless of the type (both, P<0.001). Conservative treatment and surgical intervention for the second episode after conservative treatment for the first episode were performed in 98 and 167 cases, respectively. For the second episode after conservative treatment for the first episode, SG also had a significantly longer RFI than CG, regardless of the types (PSP P<0.001, SSP P=0.031). To check the validity of conservative treatment for patients with a long RFI, we analyzed recurrence by dividing patients into two groups according to one-year RFI. For PSP, the early recurrence group (EG, RFI ≤1 year) had 99 cases and the late recurrent group (LG, RFI >1 year) had 67 cases. SG had a significantly longer RFI than CG in both EG and LG (EG, P<0.001 and LG, P=0.001). In addition, for SSP, there were 67 cases of EG and 32 cases of LG, SG had significantly longer RFIs than CG in EG (P=0.007). However, there was no significant difference in RFI between SG and CG in LG (P=0.748). CONCLUSIONS: The present study revealed diversity of management outcomes according to characteristics of recurrence and provides some evidences of the validity of conservative treatment in recurrent SSP with a long RFI. Further large-scale prospective randomized trials are required to validate these findings.