The review concluded that blood cultures had minimal influence in clinical management in immunocompetent adults hospitalised with community acquired pneumonia and that pneumonia quality measures based on blood cultures needed to be re-assessed. Given the level of evidence presented, these conclusions are likely to be reliable.

To ascertain the influence of blood cultures on clinical management of adults with community acquired pneumonia that required hospitalisation and to determine the suitability of blood cultures as measures of quality of care.

MEDLINE, MEDLINE In-Process and Other Non-Indexed Citations, Cochrane Central Register of Controlled Trials (CENTRAL), Cochrane Database of Systematic Reviews, DARE and ACP Journal Club were searched from 1966 to 2007. Search terms were reported. Reference lists of relevant articles were examined for additional studies and experts in the area were consulted. Searches were restricted to studies in English.

Eligible studies (any design) needed to evaluate and report changes in clinical management (antibiotic therapy, duration of parenteral therapy or length of stay in hospital) as a consequence of blood cultures taken at or near admission to hospital in adults who presented with community acquired infection.

Sample size, setting and exclusion criteria varied across studies. Most studies required that diagnosis of pneumonia was confirmed with a chest X-ray. Blood cultures were mandatory in four studies (one subgroup). Reported antibiotic regimens were predominantly cephalosporin plus macrolide or fluoroquinolones or penicillin/penicillin derivatives. Reported outcomes included blood culture directed antibiotic narrowing or broadening, percentage true and false positive blood cultures and duration of parenteral therapy.

Two reviewers independently assessed each selected full-text article for inclusion. Disagreements were resolved by discussion and with the aid of a third reviewer. However, the authors stated neither how the initial papers were selected for the review nor how many reviewers performed the initial selection of studies from titles and abstracts.

Study quality was assessed according to methodological strengths and weaknesses. Individual study details (including design, sample size, blinding, definitions of inclusion/exclusion criteria and outcome measures, sources of potential bias, generalisability of conclusions and sensitivity analyses) were summarised in supplemental tables.

Two reviewers assessed study quality. The authors did not state how the validity assessment was performed.

In addition to general study details and denominator data (number of patients in whom blood cultures were performed), empiric antibiotic regimens, blood culture positivity (true and false), bacterial pathogens isolated in true positives, blood culture directed management changes (including antibiotic broadening ultimately associated with a resistant organism and antibiotic narrowing) and cost associated with blood cultures were extracted.

Two reviewers independently performed double data extraction.

Data were combined using characteristic summary tables and narrative synthesis. For the primary analysis, percentage positive blood cultures (true and false) and number of blood culture-directed management changes were reported for each individual study, together with a combined range. Studies that did not report the number of patients in whom blood cultures were performed were excluded from the primary analysis and reported in the secondary analysis. Secondary analysis also evaluated the cost of blood cultures and impact of blood cultures on outcomes other than antibiotic usage. Heterogeneity of inclusion and exclusion criteria and empiric antibiotic usage were qualitatively analysed.

Fifteen observational cohort studies (nine retrospective and six prospective) (n=4,260) were included in the review.

Study methodologies were adequate to evaluate key outcomes. Areas of potential bias both for and against blood culture usage were identified and were considered to be small.

Thirteen studies (seven retrospective and six prospective) (n=3,898) were included in the primary analysis. Blood cultures were true positive in 0% to 14% of cases resulting in antibiotic narrowing in 0% to 3% of cases and broadening ultimately associated with a resistant organism in 0% to 1%. The most common bacterial pathogen for true positive blood culture was Streptococcus pneumoniae (where reported 50% to 91%) with penicillin resistance found in 0% to 20%. The false positive rate ranged from 0% to 10% of cases where reported. Results also showed that physicians narrowed antibiotic coverage in 10% to 58% of eligible cases.

The direct cost of blood cultures and blood culture-directed antibiotic changes in management was evaluated in secondary analysis of six studies (n=1,931) and ranged from US dollar $1550 to $8000.

Blood cultures had limited use in immunocompetent adults hospitalised with community acquired pneumonia; therefore, there was a need to re-assess pneumonia quality measures that were based on bood cultures.

This review had clearly stated inclusion and exclusion criteria in terms of participants, study design and outcomes. The authors searched relevant databases and efforts were made to identify additional studies by reviewing reference lists from relevant literature and contacting experts in the field. Language restrictions were imposed and it was unclear whether unpublished studies were included and so biases may have been introduced. Appropriate efforts to minimise reviewer bias and error were taken while assessing study full texts and during the data extraction stage of the review process; it was unclear whether these were adhered to during initial study selection and validity assessment. Due to variations across studies a narrative synthesis was appropriate. Given the level of evidence presented, the authors conclusions are likely to be reliable within the review context. However, the studies included in the review were dated 2003 and before and (as the authors noted) may not reflect recent increases in prevalence of resistant pathogens (such as MRSA) in the community. The authors stated that these findings were not generalisable to immuno-compromised patients or other high-risk groups.

Practice: The authors stated that the findings of the review did not support acquisition of blood cultures for all or most adults hospitalised with community acquired pneumonia, which is also supported by 2007 guidance.

Research: The authors stated that quality measures for pneumonia that included blood cultures should be re-evaluated.

Not stated.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.