Objective: To systematically examine the available evidence regarding the effect of initial discordant therapy with beta-lactam antibiotics on mortality, clinical success, and bacteriological eradication in patients with pneumococcal pneumonia.
Methods: We analyzed prospective studies that compared the clinical effectiveness of concordant (active in vitro) beta-lactam monotherapy with discordant (inactive in vitro) monotherapy with the same beta-lactam in patients with pneumococcal pneumonia. Relevant studies were identified from searches of the PubMed database (1950 to November 2005) and references from articles. Outcomes between groups of patients who received concordant and discordant treatment were compared by simple pooling of data and by estimation of pooled odds ratios or risk difference (RD), when applicable.
Results: Six prospective studies were included in our analysis. No statistically significant difference was found in mortality of patients treated with beta-Iactam concordant and discordant therapy (51/275 [19%] vs 9/42 [21%]; P = .66; data from 6 studies; RD, -0.05; 95% confidence interval [CI], -0.23 to 0.12; data from 5 studies). In addition, no statistically significant difference was found regarding clinical success (37/42 [88%] vs 5/6 [83%]; P = .57; odds ratio, 2.57; 95% CI, 0.46 to 14.34; RD, 0.07; 95% CI, -0.36 to 0.50; data from 3 studies) or bacteriological success (24/30 [80%] vs 3/3 [100%]; P = .99; and RD, -0.18; 95% CI, -0.79 to 0.42; data from 2 studies) between concordant and discordant therapy.
Conclusion: The initial discordant treatment with beta-lactam antibiotics was not associated with a statistically significant Increase in mortality or clinical or bacteriological failure of therapy for pneumococcal pneumonia.