Improving outcomes in community-acquired pneumonia

Community-acquired pneumonia (CAP) has an annual incidence of 2 to 10‰ that increases with advancing age. Depending on pathogen- and host-related factors, the course of CAP varies from a mild affection treated in the ambulatory setting, to a devastating form complicated by septic shock, respiratory failure, and death. A number of pathogens cause CAP, the most frequent being typical bacteria (Streptococcus pneumoniae, Hemophilus influenza), atypical bacteria (Mycoplasma pneumoniae, Legionella sp.), and viruses. However, the infecting agent is rarely known at the onset of the disease and is not identified at all in half to two-third of cases, so the antibiotic treatment is mostly empiric. Because of its frequency and occasional severity, CAP carries a substantial burden, both for affected individuals and for health systems. CAP is also one of the most frequent reasons for antibiotic prescriptions, and policy decisions concerning the recommended empiric treatment can have a major impact on antibiotic resistance. Unsettled issues relate to the need for covering atypical bacterial pathogens in all patients, and multidrug-resistant bacteria in patients with specific risk factors. No breakthrough therapy has emerged since the introduction of antibiotics eighty years ago, though progresses in life sustaining techniques have probably improved the prognosis of the most severely affected patients. Blunting the host inflammatory response to reduce associated damages without compromising the clearance of the infecting pathogen is one of the most researched issues. Adjunctive corticosteroid therapy is promising and hastens the resolution of symptoms, but safety concerns have not been completely ruled out. Future research efforts should lead to better understanding of the role of lung microbiome in the development and evolution of the disease; in better definition of the potential of modifying the inflammatory response to improve short- and long-term prognosis; and in the development of refined diagnosis tools and algorithms to avoid under- and overdiagnosis or under- and overtreatment