The role of Streptococcus pneumoniae in community-acquired pneumonia among adults in Europe: a meta-analysis

Abstract

The primary objective of this meta-analysis was to estimate the prevalence of adult community-acquired pneumonia (CAP) caused by Streptococcus pneumoniae in Europe, adjusted for possible independent covariates. Two reviewers conducted a systematic literature search using PubMed on English-language articles that involved human subjects with CAP during the period from January 1990 to November 2011 across European countries. A mixed-effects meta-regression model was developed and populated with 24,410 patients obtained from 77 articles that met the inclusion criteria. The model showed that the observed prevalence of S. pneumoniae in CAP significantly varies between European regions, even after adjusting for explanatory covariates, including patient characteristics, diagnostic tests, antibiotic resistance, and health-care setting. The probability of detecting S. pneumoniae was substantially higher in studies that performed more frequently a diagnostic polymerase chain reaction assay compared to all the other diagnostic tests included. Furthermore, S. pneumoniae was more likely to be confirmed as the cause of a CAP in studies with intensive care unit patients as compared to those with hospital- or community-treated patients. This study provides estimates of the average observed prevalence of S. pneumoniae, which could be used for projecting the health and economic benefits of pneumococcal immunization.

Background paper to the updated pneumococcal vaccination recommendation for older adults in Germany

“…The percentage of PP among all CAP cases in the included studies ranged from 0-67 % [33] and 12-85 % [34]. The causes of this gaping difference are more likely differences in the diagnostic tests conducted, case definitions, and patient collectives, rather than real differences between countries or studies [33].…”

“…The CAPNETZ study [8,32] is the largest prospective study on the microbiologi- [33,34]. The percentage of PP among all CAP cases in the included studies ranged from 0-67 % [33] and 12-85 % [34].…”

“…The percentage of PP among all CAP cases in the included studies ranged from 0-67 % [33] and 12-85 % [34]. The causes of this gaping difference are more likely differences in the diagnostic tests conducted, case definitions, and patient collectives, rather than real differences between countries or studies [33]. Based on a meta-analysis (random effects model) of 77 individual studies, with a total of 24,410 CAP patients, Rozenbaum et al [33] estimate the percentage of PP as 19.3 % (no CI is given) of all CAP cases in which a pathogen was detected.…”

“…Less than 1 % of all pneumonia cases were coded as PP (ICD10 code J13); most certainly a significant underestimation of the actual percentage. Based on the data reported by Rozenbaum et al [33] and from the CAPITA study, we estimate that pneumococci caused roughly 20 % of all cases coded with any ICD10 code for pneumonia (J12-J18 This statistic includes only those deaths that occurred during hospitalisation; the actual mortality is likely higher. In the CAPNETZ study, the 30-day case fatality of PP cases in people aged ≥60 years was somewhat lower, at 7.7 % (32 of 418 cases).…”

Review: Current and new generation pneumococcal vaccines

“…8,11 However, in a meta-analysis of studies from Europe it is clear that there are regional differences in the prevalence of the pneumococcus as a cause of CAP, depending on both the treatment setting and diagnostic approach. 12 However, since the pneumococcus is the most important pathogen in most situations, the data on the epidemiology of CAP to a large extent mirrors that of pneumococcal pneumonia. 8 A number of additional reviews confirm the enormous clinical and economic burden caused by CAP in North America, Europe, the Asia Pacific region, Latin America, and elsewhere, 2, 13-15 with pneumococcal disease per se having been documented to carry high economic costs in the United States and in Europe.…”