The role of gram stain in reducing broad-spectrum antibiotic use: A systematic literature review and meta-analysis

The number of studies that verify whether Gram stain can help to reduce the use of broad-spectrum antibiotics is relatively limited compared to those evaluating its concordance with culture test results. Thereby, we aimed to evaluate the effectiveness of Gram staining in the reduction of broad-spectrum antibiotics and its impact on clinical outcomes. We systematically reviewed studies having used Gram stain to guide antibiotic selection and evaluated performance measures between 1996 and 2022. We extracted available data on broad-spectrum antibiotic use as a primary outcome of the studies in view of an exploratory meta-analysis designed to estimate the clinical effect of Gram stain. We also evaluated the clinical response and coverage rates of the initial antibiotic therapy. One randomized study and four non-randomized studies were eligible, all of which were conducted in tertiary care hospitals in Japan. Gram stain was associated with reduced broad-spectrum antibiotic use, including antipseudomonal antibiotics (odds ratio [OR], 0.05; 95% confidence interval [CI], 0.01-0.34), anti-methicillin-resistant Staphylococcus aureus antibiotics (OR, 0.21; 95% CI, 0.07-0.63), and carbapenems (OR, 0.07; 95% CI, 0.02-0.19), without impairing clinical outcomes, including clinical response rate (OR, 1.48; 95% CI, 0.95-2.31) and coverage rate of initial antibiotic therapy (OR, 0.70; 95% CI, 0.40-1.22) using random-effects models in our meta-analysis. In conclusion, Gram stain may be useful in guiding initial antibiotic selection without apparent adverse clinical outcomes. However, currently available studies evaluating the clinical usefulness of Gram stain are limited to specific clinical settings.