Clinical phenotypes of secondary spontaneous pneumothorax (SSP) in different underlying lung diseases: a retrospective study and literature review on the clinical heterogeneity in this single entity.

BACKGROUND

The clinical outcome of secondary spontaneous pneumothorax (SSP) was mostly evaluated in chronic obstructive pulmonary disease (COPD) or interstitial lung disease (ILD) patients whereas data were limited in non-COPD/ILD-related pneumothorax. This study aimed to describe the clinical outcome of pneumothorax secondary to different pulmonary conditions.

METHODS

A retrospective observational study was conducted on SSP patients in a Hong Kong hospital. Clinical outcome of pneumothorax secondary to different underlying lung diseases including persistent air leak (PAL), 2-year post-discharge pneumothorax recurrence and all-cause mortality were evaluated. Logistic regression and cox regression analysis were applied for covariates adjustment.

RESULTS

Of the 223 SSP patients enrolled, 32 (14.3%) had acute chest infection and 191 (85.7%) had chronic lung diseases. The acute chest infection group had higher risk of PAL >14 days (40.6% vs. 17.8%; P=0.003), longer hospitalization (24 vs. 13 days; P=0.001) and higher all-cause mortality rate (25.0% vs. 12.0%; P=0.049) than chronic disease group. Among those with chronic lung diseases, COPD/ILD-related pneumothorax led to higher PAL risk (>2 days: OR: 2.14, 95% CI: 1.01-4.54, P=0.046; >14 days: OR: 4.02, 95% CI: 1.49-10.88, P=0.006) and prolonged hospitalization (14 vs. 12 days; P=0.02) but similar post-discharge all-cause mortality and recurrence rate at two years compared to non-COPD/ILD-related pneumothorax. Higher proportion of COPD/ILD group was considered unfit for surgical pleurodesis. One-year recurrence rate was lower in asthma (8.3%) and bronchiectasis (0%) patients than other disease groups (15.8-33.3%).

CONCLUSIONS

Our study demonstrated heterogeneous clinical phenotypes among different lung diseases with poorer outcome in acute chest infection and COPD/ILD population. Differentiation between infective and non-infective, COPD/ILD-related and non-COPD/ILD-related SSP may enable more personalized management.