What Fraction of Adult Community-Acquired Pneumonia Is Caused by the Pneumococcus? New Insights from Spain

(See the Major Article by Torres et al on pages 1075–85.)

Adult community-acquired pneumonia remains a leading cause of global disease and death as well as a major public health concern. A number of pathogens can cause the disease. Yet, while a pathogen can be identified in a subset of patients, for the majority of them the cause remains unclear despite extensive microbiological clinical evaluations [1–3]. Given the well-known difficulties in establishing an individual cause at the time of pneumonia diagnosis, management is generally empirical, directed to the most likely pathogens [1, 2].

Although the prevalence of pathogens identified in patients with community-acquired pneumonia varies depending on patient characteristics, setting, geographic location, and other factors, there is general agreement that empirical regimens should cover Streptococcus pneumoniae, considered one of the leading bacterial causes of community-acquired pneumonia [1, 2]. Nevertheless, the precise contribution of the pneumococcus to the burden of adult community-acquired pneumonia is poorly defined and has likely evolved over time.

Before the use of antibiotics, pneumococcus was identified in 90%–95% of all pneumonias [4]. Since the beginning of widespread antibiotic use, the estimated proportion of pneumococcal community-acquired pneumonias has declined, a change that was further impacted by the introduction of pneumococcal vaccines. More specifically, the introduction and widespread use of pneumococcal conjugate vaccines (PCVs) have substantially changed the epidemiology of pneumococcal diseases, including but not restricted to pneumococcal pneumonia.

While pneumococcal pneumonia is commonly described as a single clinical entity, it can be caused by different pneumococcal serotypes. Currently, there are about 100 recognized distinct serotypes, and a number of them have known pathogenic potential. PCVs were originally designed to provide protection against the most prevalent serotypes causing invasive pneumococcal disease, but their impact has clearly extended to pneumonia, otitis media, and other pneumococcal diseases. Widespread use of PCVs has rapidly reduced the circulation of serotypes included in the vaccines, with remarkable reductions in pneumococcal diseases through both direct and indirect protection. Examining the impact of these vaccines has enabled an appreciation of how much disease is caused by the pneumococcal serotypes covered by the vaccines. Serotypes not covered by the vaccines, however, have replaced the niche left after removal of vaccine serotypes and are now leading causes of the residual burden of pneumococcal diseases. The intensity and ultimate impact of serotype replacement on the residual burden of pneumococcal diseases have varied across populations, and it remains a matter of extensive inquiry [5–7].

Quantifying the contribution of pneumococcal infections to the burden of community-acquired pneumonia is challenging. Standard microbiological assessments available as part of routine clinical care have limited ability to identify causes of pneumonia, especially bacterial causes [1–3]. Moreover, even when pneumococcal isolates are obtained, serotyping requires specialized laboratorial techniques, and it is not part of routine clinical care. Much of our understanding of the evolution of pneumococcal diseases caused by specific serotypes comes from large active population- and laboratory-based surveillance systems that proactively pursue identification and characterization of invasive isolates identified from cultures from sterile anatomic sites [8, 9]. However, it is well understood that a large fraction of pneumococcal infections are noninvasive and thus will pass undetected by surveillance that focuses only on invasive diseases [4].

The recent development of a novel serotype-specific urinary antigen detection (UAD) assay (UAD1) to accurately identify noninvasive infections caused by the serotypes covered by the 13-valent PCV (PCV13) revolutionized the identification of pneumococcal infections [10]. The use of this initial UAD1 assay unveiled a sizable and yet previously unrecognized burden of pneumococcal community-acquired pneumonia caused by serotypes included in PCV13 [11, 12]. More recently, a new UAD assay—UAD2—has been developed to identify 11 additional serotypes not covered in UAD1 (serotypes 2, 8, 9N, 10A, 11A, 12F, 15B, 17F, 20, 22F, and 33), including serotypes that would be covered in the experimental 20-valent PCV (PCV20; serotypes 2, 9N, 17F, and 20 are not included in PCV20) [13].

In this issue of Clinical Infectious Diseases, Torres and colleagues [14] report findings from a large multicenter, prospective, hospital-based study conducted in Spain to determine the contribution of specific pneumococcal serotypes to adult community-acquired pneumonia by using standard diagnostic methods supplemented with serotype-specific UAD assays (UAD1 and UAD2). The study encompassed >3100 adult patients (mean age, 66 years) admitted with community-acquired pneumonia to 4 tertiary-care hospitals from November 2011 through November 2018. PCV13 was added to the infant vaccination schedules in 2015–2016 at 3 sites and in 2011 at 1 site; all sites had >90% coverage by 2018. All sites introduced 23-valent pneumococcal polysaccharide vaccine for adults aged ≥65 years in 2000, and 1 site introduced PCV13 for those adults in 2017. Microbiological evaluations were based on each hospital’s clinical practice and at the discretion of the attending physician, including blood, pleural fluid, and sputum (high-quality specimens) cultures, standard urinary antigen tests for pneumococcus and Legionella, and other viral polymerase chain reaction and serological assessments for atypical bacteria. UAD1 was applied to all available urine samples, and UAD2 to samples collected since 2016.

Torres et al [14] report that based on all available tests, a pathogen was identified in 44% of study participants, and S. pneumoniae was detected in 28.8% of patients throughout the study period. The study focused mainly on pneumococcal pneumonia, and detailed information on other pathogens was not presented. For the period 2016–2018, when UAD2 was used, the proportion of community-acquired pneumonia due to S. pneumoniae was 33.9%. The majority of pneumococcal detections were done with urine antigen tests, but there were also several detections with blood and sputum cultures. While there are several challenges and differences in selection criteria, completeness of testing, and laboratory testing procedures among etiological studies [4], the findings from the current study suggest a much higher burden of pneumococcal disease than previously described [1–3, 12]. The serotype-specific tests also revealed activity of serotypes not covered by current PCVs.

The characterization and distribution of pneumococcal serotypes implicated in adult community-acquired pneumonia is of great interest, because newer high-valency vaccine formulations are in late stages of evaluation and may become available for use soon. These formulations include serotypes 22F and 33F in a 15-valent PCV (PCV15), and also serotypes 8, 10A, 11A, 12F, and 15B in PCV20. Torres and colleagues [14] estimated that 14.1%, 14.5%, and 23.8% of adult community-acquired pneumonia cases, respectively, would be covered by the current PCV13, PCV15, and PCV20 formulations.

While the impact of PCV infant and adult vaccination programs at the study sites was likely evolving, serotypes 3 and 8 were the most commonly detected serotypes during the study period. Importantly, these serotypes were associated with complications and more severe disease in exploratory analyses. Of note, serotype 3 is included in the current PCV13 formulation, whereas serotype 8 activity has been increasing during the last years and becoming more recognized as an emergent replacement serotype across many countries [5]. Additional characterizations of these serotypes and the potential impact of current and new vaccines on their activity would be informative.

In an elegant, pragmatic etiological study among adults hospitalized with community-acquired pneumonia in Spain, Torres and colleagues [14] demonstrated that the pneumococcus accounted for a large fraction of adult community-acquired pneumonia—a fraction much higher than previously recognized in previous similar studies. A sizable fraction of those pneumococcal pneumonias could be prevented through vaccination.

Notes

Financial support. This work was supported by the National Institute of Allergy and Infectious Diseases, National Institutes of Health (grant K24 AI148459).

Potential Conflicts of Interest. C. G. G. reports consultation fees from Pfizer, Merck, and Sanofi-Pasteur and grants from Sanofi, Campbell Alliance/Syneos Health, the Centers for Disease Control and Prevention, the National Institutes of Health, the Food and Drug Administration, and the Agency for Health Care Research and Quality, outside the submitted work. The author has submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.