Smoking Index and Risk Gradient for Spontaneous Pneumothorax: Evidence from 111 Indian Patients

Abstract

Background: Cigarette smoking is the single most modifiable risk factor for spontaneous pneumothorax (SP), yet South-Asian dose-response data remain sparse. Objectives: (i) Quantify the association between smoking index (SI) and pneumothorax type—primary (PSP) versus secondary (SSP); (ii) model SI cut-points that best predict SSP. Methods: Secondary analysis of a prospective cohort of 111 consecutive SP patients (November 2020–October 2021) at a North-Indian tertiary centre. SI was calculated as number of cigarettes smoked per day ×Total number of years of smoking and stratified: mild < 100, moderate 100–299, heavy ≥ 300. Logistic regression adjusted for age and sex estimated odds ratios (OR) for SSP versus PSP across SI strata. Receiver-operating-characteristic (ROC) analysis determined the optimal SI threshold for predicting SSP. Results: Overall, 80/111 (72.1 %) were ever-smokers; median SI 145 (IQR 45-295). Heavy SI occurred in 38/75 SSP (50.7 %) but in 0/36 PSP (p < 0.001). Adjusted OR for SSP rose stepwise: moderate SI = 5.6 (95 % CI 2.1-14.7); heavy SI = 14.8 (5.7-38.9). ROC analysis yielded area-under-curve 0.86; SI ≥ 220 pack-year-equivalents provided 78 % sensitivity and 81 % specificity for SSP. Conclusions: A clear SI dose-response is demonstrable: values ≥ 220 confer 15-fold higher odds of secondary pneumothorax. Incorporating SI into outpatient triage could trigger proactive imaging surveillance and smoking-cessation counselling in high-risk Indian smokers.