Population-based cohort study examining the association between splenectomy and empyema in adults in Taiwan

Abstract

ObjectiveThis study aimed to investigate the association between splenectomy and empyema in Taiwan.MethodsA population-based cohort study was conducted using the hospitalisation dataset of the Taiwan National Health Insurance Program. A total of 13 193 subjects aged 20–84 years who were newly diagnosed with splenectomy from 2000 to 2010 were enrolled in the splenectomy group and 52 464 randomly selected subjects without splenectomy were enrolled in the non-splenectomy group. Both groups were matched by sex, age, comorbidities and the index year of undergoing splenectomy. The incidence of empyema at the end of 2011 was calculated. A multivariable Cox proportional hazards regression model was used to estimate the HR with 95% CI of empyema associated with splenectomy and other comorbidities.ResultsThe overall incidence rate of empyema was 2.56-fold higher in the splenectomy group than in the non-splenectomy group (8.85 vs 3.46 per 1000 person-years). The Kaplan-Meier analysis revealed a higher cumulative incidence of empyema in the splenectomy group than in the non-splenectomy group (6.99% vs 3.37% at the end of follow-up). After adjusting for confounding variables, the adjusted HR of empyema was 2.89 for the splenectomy group compared with that for the non-splenectomy group. Further analysis revealed that HR of empyema was 4.52 for subjects with splenectomy alone.ConclusionThe incidence rate ratio between the splenectomy and non-splenectomy groups reduced from 2.87 in the first 5 years of follow-up to 1.73 in the period following the 5 years. Future studies are required to confirm whether a longer follow-up period would further reduce this average ratio. For the splenectomy group, the overall HR of developing empyema was 2.89 after adjusting for age, sex and comorbidities, which was identified from previous literature. The risk of empyema following splenectomy remains high despite the absence of these comorbidities.