Empyema: A debilitating condition that warrants further research

Hippocrates (460–377 BC) was the first to describe empyema and the ‘succussion’ manoeuvre to detect fluid in the hemithorax. Fast forward to the 21st century, pleural infection continues to cause significant morbidity and mortality worldwide. Half of the 4 million patients with pneumonia develop effusions that can progress to empyema characterized by pus or bacteria in the pleural space. About 65 000 patients with pleural infections require hospitalization in the UK and USA, which accounts for the phenomenal cost of US$500 million per year. Empyema cases in adults and children have been increasing since 1990. In the UK, close to 30% of patients aged > 65 years, with comorbidity, die of pleural infection. Utah reported a sixfold increase in empyema death when two time periods, 2000–2004 and 1950–1975, were compared. Reasons for increase include clinical awareness and improved diagnostics, immunosuppressive therapy and transplantation, human immunodeficiency virus/acquired immune deficiency syndrome (HIV/AIDS), replacement phenomenon arising from multivalent pneumococcal vaccine, variability in practice and delay in treatment. Timely administration of appropriate antibiotics, coupled with the evacuation of infected pleural fluid/ pus, is pivotal in assuring good outcome. As adequacy of empirical antimicrobial therapy is acutely correlated with mortality, knowledge of the bacteriology of empyema is central to patient care. It is often polymicrobial and is dependent on host (adult/child/immunocompromised), source of infection (community-/ hospital-acquired) and geographical location. Clinicians must be cognizant of their local epidemiological data as bacterial isolates differ by geographical location. Before the advent of antibiotics and pneumococcal vaccination, Streptococcus pneumoniae, Streptococcus pyogenes and Staphylococcus aureus were common. Today, Streptococcus milleri accounts for up to 50% of adult community-acquired empyema cases in the UK, Canada, Scandinavia and New Zealand, with S. pneumoniae and anaerobes causing the rest. S. milleri empyema can be associated with comorbidities such as cancer and diabetes mellitus and accounts for 20% of mortality. S. aureus (including methicillin-resistant S. aureus (MRSA)) is more frequently found in hospital-acquired empyemas, while Klebsiella pneumoniae is the causative bacteria for both communityand hospital-acquired empyemas in Taiwan. CURB65, a composite score of confusion, uraemia, respiratory rate, BP and age > 65 years, predicts mortality well for individuals with pneumonia but underestimates mortality in those with parapneumonic effusions. Data demonstrated that patients with parapneumonic effusions had worse 30-day mortality and longer hospital stay compared to their pneumonia counterparts. Published in a recent issue of Respirology, Brims et al. investigated the bacteriology and clinical outcome of patients with empyema in Western Australia. This study represented the largest data set comprising of 601 cases of empyema (culture-positive pleural infection (CPPI)) over 6 years. Most were males (71%) with median age of 63 years (interquartile range (IQR): 50–74) and cancer as the major comorbidity. Hospitalacquired empyema (HA-CPPI) accounted for 66% of cases, community-acquired empyema (CA-CPPI) 27% and the rest from because of oesophageal rupture/leak. Although 54% of HA-CPPI were iatrogenic following thoracic surgery and pleural intervention, it cannot explain the low incidence of computed tomography (CT)-detected coexisting pneumonia in 27% of HA-CPPI and 44% of CA-CPPI. Moreover, Streptococcus viridans (Streptococcus anginosis), part of normal flora of the oropharynx and gastrointestinal tract and rarely associated with pneumonia, was the most common bacteria cultured from CA-PPI. HA-CPPI and empyema from oesophageal perforation were polymicrobial, requiring broad-spectrum antibiotics against S. viridians, S. aureus, Gram-negative bacilli and anaerobes. Hospital stay was longer for HA-CPPI compared with CA-CPPI (19 vs 15 days). Although inpatient mortality was similar for CA-CPPI and HA-CPPI at 15%, mortality at 1 year was higher in those with HA-CPPI (42%) compared with CA-CPPI (32%). These findings were in contrast to the widely quoted mortality of 20% for immunocompetent and 30% for immunocompromised patients with pleural infections. The predictors of mortality in the first year were increasing age, cancer, renal failure and no surgical intervention. The study by Brims et al. confirmed the findings of other studies that the bacteriology of pleural infection was different from the common causative pathogens of pneumonia. Moreover, pneumonia was not consistently detected by CT in these patients with empyema, thereby challenging the concept that empyema is an extension of infection of the underlying lung parenchyma. The incidence of HA-CPPI also increased over the 6-year period and carried a high mortality of 42%. It would be conceivable to expect that prompt administration of appropriate antibiotics guided by knowledge of local bacteriology as well as early referral for surgical intervention would improve 1-year survival. Most often than not, surgery is not an option for the elderly and infirmed. Research into pathophysiology, drug therapeutics, intrapleural fibrinolytics and minimally