The Value of Legionella and Pneumococcal Testing in Pneumonia
Amesh A. Adalja, MD, FACP, FACEP, FIDSA, October 3, 2018
Community-acquired pneumonia is one of the most significant infectious diseases worldwide, and because of how common it is, it is a leading cause of death in many countries, including the United States. Though there is general consensus on the most common microorganisms responsible for this disease, many cases go without a specific microbiological etiology's being identified. In addition to sputum cultures, blood cultures (when indicated), and respiratory viral testing, urinary antigen testing for 2 specific etiologies – the pneumococcus (Streptococcus pneumoniae) and Legionella pneumophilia serogroup I – are frequently performed. Clinical treatment guidelines recommend the use of these tests in certain settings and are the subject of a new study published in Clinical Infectious Diseases.
The most recent (2007) community-acquired pneumonia guidelines issued by the Infectious Diseases Society of America (IDSA) and the American Thoracic Society (ATS) recommend the use of these antigen tests when certain patient characteristics are present. The indicators for pneumococcal urinary antigen testing are ICU admission, failure of outpatient therapy, leukopenia, active alcohol abuse, chronic severe liver disease, asplenia, and pleural effusion. Legionella urinary antigen testing is indicated for ICU admission, failure of outpatient antibiotic therapy, active alcohol abuse, travel within the past 2 weeks, and pleural effusion.
In this multicenter prospective surveillance study, Bellew et al evaluated 1,941 patients hospitalized with community-acquired pneumonia to determine the sensitivity and specificity of the guideline recommendations for urinary antigen testing and to determine the association between testing indicators and positive test results. All patients had both antigen tests performed.
No Associations Found for Pneumococcal Testing, Legionella Predictors
Of the study patients, 61% had an indication for pneumococcal antigen testing and 4.2% had a positive test. They found no statistical difference between rates of positivity of those with indications for testing and those without, giving the guideline indicators a sensitivity of 61% and a specificity of 39%. None of the indications for pneumococcal testing or any other variable was found to be significantly predictive on multivariate analysis.
Of the study patients, 64.8% had an indication for legionella antigen testing and 1.6% had a positive test. They found no statistical difference between rates of positivity of those with indications for testing and those without, giving the guideline indicators a sensitivity of 63% and a specificity of 35%. The only guideline indicator that was found to have value was recent travel, which conferred an odds ratio of 2.17 for positivity. In a multivariate analysis, hyponatremia, fever, and diarrhea were found to have significant predictive ability, with travel history nearly reaching statistical significance.
Additionally, if a viral etiology was identified, antigen testing for either pathogen was rarely positive. Conversely, higher procalcitonin levels were more commonly seen in those with positive antigen results.
Pandemic Preparedness and Public Health Significance
The authors make a compelling argument against the routine use of pneumococcal antigen testing based on cost, lack of clinical indicators of positivity, the incorporation of pneumococcal coverage in all community-acquired pneumonia antibiotic regimens, and the general absence of antibiotic narrowing with a positive result. They also discuss the value of legionella antigen testing for treatment guidance as well as public health investigation. They conclude that future guidelines should evaluate “which patient types and clinical scenarios would benefit most from a microbiological diagnosis.”
One patient type and clinical scenario in which the value of these tests could be pivotal is when a disease of public health significance might be present. Respiratory infections such as avian influenza, pneumonic plague, pneumonic tularemia, SARS, and MERS could present in a way similar to both Legionnaire’s disease and pneumococcal pneumonia, and, in the right patient, a positive or negative test result may be able to provide crucial information that could have cascading impacts.
Bellew S, Grijalva CG, Williams DJ, et al. Pneumococcal and Legionella urinary antigen tests in community-acquired pneumonia: prospective evaluation of indications for testing. Clin Infect Dis 2018; https://academic.oup.com/cid/advance-article-abstract/doi/10.1093/cid/ciy826/5108546?redirectedFrom=fulltext. Accessed October 2, 2018.