Study designs of evaluations included in the review
Prospective studies were eligible for inclusion.
Specific interventions included in the review
To be eligible for inclusion, studies had to compare discordant beta-lactam antibiotic monotherapy with concordant monotherapy with the same beta-lactam antibiotic. Therapy was defined as concordant when the isolated Streptococcus pneumoniae was fully susceptible to the beta-lactam administered, based on in vitro susceptibility testing (2002 Clinical and Laboratory Standards Institute Guidelines), and discordant when the strain was resistant. The antibiotics used in the included studies were: ceftriaxone (1 to 2 g intravenously every 24 hours), cefotaxime (1 to 2 g intravenously every 6 hours), amoxillin-clavulanate (1 to 2 g intravenously every 8 hours, or 875/125 mg orally every 12 hours, or 2,000/125 mg orally every 12 hours), penicillin-procaine (1,200,000 units intramuscularly every 12 hours). Doses were not reported in one study which used penicillin and cefuroxime.
Participants included in the review
Studies of adult patients with culture-proven pneumococcal pneumonia were eligible for inclusion. Disease severity varied between the included studies.
Outcomes assessed in the review
Studies that reported all-cause mortality, clinical success or bacteriological success were eligible for inclusion. Clinical success was defined as the improvement or resolution of signs or symptoms of pneumococcal pneumonia. Bacteriological success was defined as the absence of all initial pathogens or clinical evidence of eradication in the absence of samples.
How were decisions on the relevance of primary studies made?
Two reviewers independently assessed the studies for relevance. It was not stated how any differences were resolved.