Community-Acquired Pneumonia: Causes and Frequency of Hospitalizations

Andrew F. Shorr, MD, MPH


October 07, 2015

This feature requires the newest version of Flash. You can download it here.

This is Andy Shorr from George Washington University in Washington, DC, with the Pulmonary and Critical Care Literature Update. Today, I want to discuss an article[1] in the July 30 issue of the New England Journal of Medicine by the EPIC investigators from the Centers for Disease Control and Prevention. This study focused on the microbiology and epidemiology of community-acquired pneumonia.

Community-acquired pneumonia remains a leading reason for hospital admission and results in huge direct and indirect costs to the healthcare system. We have not had a good epidemiology study of community-acquired pneumonia in almost 2 decades, which makes this effort to examine the current causes and frequency of admissions to the hospital for community-acquired pneumonia particularly important.

Investigators in Nashville, Tennessee, and Chicago, Illinois, looked at patients who were sequentially admitted to the hospital with a clinical diagnosis of community-acquired pneumonia. All patients were required to have radiographic evidence of an infiltrate and signs and symptoms of pneumonia. The evaluation included length of hospital stay, severity of illness, and the microbiology of the patients' pneumonias, using appropriate urinary antigens, serum antigen testing, various serologic tests, and, of course, nasal and sputum cultures. They looked at the burden of illness and the distribution of pathogens in terms of severity of illness and across the various patient age groups.

Study Population

Approximately 70% of patients who were eligible for the study participated, which raises some concerns about generalizability and bias. The investigators found an isolate or organism in about 38% of the 2320 patients who were included in the analysis. Of note, viral pathogens, predominantly rhinovirus and influenza, were more common than bacterial pathogens; not surprisingly, the number one bacterial pathogen was pneumococcus.

The authors specifically excluded patients who were severely immunosuppressed, patients who had percutaneous endoscopic gastrostomy feeding tubes or tracheostomies, patients who had been recently hospitalized, and patients who were coming from nursing homes because of disability. Thus, patients who would meet criteria for what we call healthcare-associated pneumonia were excluded. Thus, this study looked at classic community-acquired pneumonia, and the findings are consistent with earlier trials.

Study Results and Analysis

Overall, an organism is isolated in one third to 70% of patients with community-acquired pneumonia, whether it be viral, bacterial, fungal, or microbacterial. The most common bacterial isolate is pneumococcus. In part, this reflects the fact that we have tests that are good at diagnosing pneumococcus, such as urinary antigen. It also reflects the fact that infections such as influenza vary based on the seasonality of the study. This study was strong in that it encompassed a 30-month period with two respiratory winter seasons. They saw that during the change in season, flu peaked when you would expect flu to peak; and, not surprisingly, following that, the incidence of Staphylococcus aureus as a pathogen peaked, confirming the known relationship between S aureus superinfection and influenza infection generally. That pattern helps confirm the generalizability and face-validity of these studies.

Of interest, the study found that as patients aged, they were more likely to require a critical care unit (CCU) level of care. In addition, pathogens were more commonly isolated in older patients. Thus, if you were an older patient with pneumonia, it was much more likely that a pathogen would be isolated, whether bacterial or viral, than if you were younger. Not surprisingly, isolates such as Enterobacteriaceae and S aureus were more common in the CCU than the regular hospital unit. In fact, S aureus and Enterobacteriaceae were only rare causes of pneumonia in patients outside the CCU.

On average, about 1 in 5 patients did require CCU-level care, and about 1 in 20 patients required mechanical ventilation. The overall mortality for the cohort in hospital was about 2%. These authors estimated that the incidence of community-acquired pneumonia was about 25 cases per 10,000 patients per year. That is very similar to the incidence noted 10-15 years ago.

These data therefore confirm earlier investigations, and more importantly they benchmark the epidemiology of community-acquired pneumonia, which may change in the future given the introduction of the vaccine for pneumococcus. As more adults are vaccinated, that may modify the distribution of pneumococci that we see, both the serotypes of pneumococci and the prevalence of pneumococci overall.

For everyone who manages the care of patients with community-acquired pneumonia, which means anyone who works in pulmonary and critical care medicine, I urge you to look at this study in the July 30 issue of the New England Journal of Medicine.

This is Andy Shorr from Washington, DC.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: