Pneumothorax as a Poor Prognostic Indicator in COVID-19 in Turkey: A Propensity Score Matching Analysis

M Özgel,I Gülçek,M Ağar,H Ulutaş
DOI: https://doi.org/10.4103/njcp.njcp_785_23
2024-08-01
Abstract:Background: Spontaneous pneumothorax in COVID-19 occurs infrequently but in up to 15% of patients dependent on mechanical ventilation (MV). Pneumothorax-related deaths account for 1% of all COVID-19-related deaths. Aim: To determine factors associated with pneumothorax in COVID-19 patients and the effect of pneumothorax on early survival. Methods: This was a retrospective study of 4799 COVID-19-positive hospitalized patients. The groups were homogenized using propensity score matching (PSM) in two groups comprising 67 COVID-19 patients each. The prevalence of pneumothorax was determined. Multiple logistic regression was used to determine factors associated with pneumothorax. P value < 0.05 was taken as significant. Results: The prevalence of pneumothorax in COVID-19 patients was 1.6%. Lung disease, comorbidities, and oxygen support, which were significantly different between the two groups before PSM, were homogenized after PSM. In a univariate analysis, symptom duration (P ˂ 0.001), neutrophilia (P ˂ 0.001), lymphopenia (P ˂ 0.001), neutrophil-lymphocyte ratio (P = 0.003), ferritin levels (P = 0.012), D-dimer levels (P = 0.011), MV support (P ˂ 0.001), antibiotherapy (P ˂ 0.001), length of hospital stay (P = 0.009), and death (P = 0.002) differed significantly between the groups. Pneumothorax had a significant negative effect on survival (32.8% vs. 59.7%, P = 0.01). In a multivariate regression model, factors associated with pneumothorax were duration of symptoms (Adjusted Odds ratio (AOR) 1.68; 95% Confidence Interval (CI): 1.26-2.25; P = 0.001), mechanical ventilation (AOR 23.92; 95% CI: 4.12-138.72; P = <0.001), dual antibiotics (AOR 8.28; 95% CI: 1.56-43.86; P = 0.013), neutrophilia (AOR: 1.08; 95% CI: 1.02-1.14; P = 0.011), and lymphopenia (AOR: 0.92; 95% CI: 0.86-0.90; P = 0.022). Conclusion: The presence of pneumothorax was associated with poor survival in COVID-19 patients. Patients with a prolonged time from symptom onset to treatment and those dependent on mechanical ventilation in intensive care were in the high risk group for the development of pneumothorax.
What problem does this paper attempt to address?