Phenotyping empyema by pleural fluid culture results and macroscopic appearance: an 8-year retrospective study.

Background: The clinical impact of phenotyping empyema is poorly described. This study was designed to evaluate clinical characteristics and outcomes based on the two readily available parameters, pleural fluid culture status and macroscopic fluid appearance.
Methods: A retrospective study was conducted on patients with empyema hospitalised between 2013 and 2020. Empyema was classified into culture-positive empyema (CPE) or culture-negative empyema (CNE) and pus-appearing empyema (PAE) or non-pus-appearing empyema (non-PAE) based on the pleural fluid culture status and macroscopic fluid appearance, respectively.
Results: Altogether, 212 patients had confirmed empyema (CPE: n=188, CNE: n=24; PAE: n=118, non-PAE: n=94). The cohort was predominantly male (n=163, 76.9%) with a mean age of 65.0±13.6 years. Most patients (n=180, 84.9%) had at least one comorbidity. Patients with CPE had higher rates of in-hospital mortality (19.1% versus 0.0%, p=0.017) and 90-day mortality (18.6% versus 0.0%, p=0.017) and more extrapulmonary sources of infection (29.8% versus 8.3%, p=0.026) when compared with patients with CNE. No significant difference in mortality rate was found between PAE and non-PAE during the in-hospital stay and at 30 days and 90 days. Patients with PAE had less extrapulmonary sources of infection (20.3% versus 36.2%, p=0.010) and more anaerobic infection (40.9% versus 24.5%, p=0.017) than those with non-PAE. The median RAPID (renal, age, purulence, infection source, and dietary factors) scores were higher in the CPE and non-PAE groups. After adjusting for covariates, culture positivity was not independently associated with mortality on multivariable analysis.
Conclusion: Empyema is a heterogeneous disease with different clinical characteristics. Phenotyping empyema into different subclasses based on pleural fluid microbiological results and macroscopic fluid appearance provides insight into the underlying bacteriology, source of infection and subsequent clinical outcomes.
Competing Interests: Conflict of interest: The authors declare that there is no conflict of interest.
(Copyright ©The authors 2023.)