Appropriateness of Empirical Antibiotic Therapy in Bacterial Culture-Positive Inpatients: A Retrospective Cohort Study

Abstract Objectives We aimed to evaluate the prevalence, relative factors, and outcomes of different empirical antibiotic therapy (EAT) prescriptions in infected inpatients. Methods We performed a retrospective cohort analysis of inpatients a tertiary hospital in China between October 1, 2019 and September 30, 2020. Bacterial culture-positive patients received EAT were enrolled in this study. We initially assessed the prevalence of different EAT prescriptions by univariate analysis, and then used an enter multivariable logistic regression model to calculate adjusted odds ratios for assessing relative factors and outcomes of them. Results 1,257 infected patients received EAT on the day of sampling culture. 31.7% (398/1257) received appropriate but unnecessarily broad-spectrum empirical antibiotic therapy (AUEAT), and 37.3% (469/1257) received inappropriate empirical antibiotic therapy (IEAT). Age was a correlated factor of receiving AUEAT and IEAT. The odds of receiving AUEAT and IEAT increased with age (adjusted OR 1.023 [95% CI, 1.013 ~ 1.032]; p < 0.001; 1.009 [95% CI, 1.001 ~ 1.018]; p = 0.033). Patients who received AUEAT and IEAT had higher rates of ICU care and increased hospital costs. AUEAT has a higher proportion of poor prognosis (29.4%, 117/398, p < 0.001). Respiratory tract was the most common infection site (418/1257, 33.3%). The common pathogens were Escherichia coli (315/1257, 25.1%), Klebsiella species (208/1257, 16.5%), Staphylococcus aureus (204/1257, 16.2%), and Pseudomonas species (167/1257, 13.3%). 45.3% (570/1257) patients were infected with multidrug-resistant organism (MDRO), of which 53.5% (305/570) received IEAT. Among 305 MDRO infected patients with IEAT, 71.5% (218/305) were infected with Gram-negative bacteria. The majority of Gram-negative bacteria was Enterobacteriaceae (178/218, 81.7%), among which E. coli (116/178, 65.2%) and Klebsiella (38/178, 21.3%) strains accounted for high proportions. Conclusions Inappropriate or unnecessarily broad-spectrum use of antibiotics is widely prevalent in hospital. AUEAT and IEAT were bound to increase antimicrobial resistance, rates of ICU care, and health care costs. AUEAT was associated with increased risk of poor prognosis. Near half of inpatients infected with MDRO, and these patients were more likely to receive IEAT. Early identification of infectious pathogens and resistance can provide the basis for rational use of antibiotics and improve the current situation of antibiotic abuse and antimicrobial resistance.