Clinical Efficacy Comparison Between Piperacillin-tazobactam and Carbapenem in the Treatment of Infection Caused by Extended-spectrum β-lactamase-producing Escherichia Coli

OBJECTIVE To explore the clinical efficacy, safety and drug economy of piperacillin-tazobactam(PTZ) and carbapenem(CBP) in the treatment of infections caused by extended-spectrum β-lactamase(ESBL)-producing Escherichia coli, to provide a feasible solution to curb the growth of carbapenem resistant bacteria. METHODS The samples of ESBL-producing Escherichia coli strains were detected from September 2019 to January 2021 by retrospective analysis in Fuyang People's Hospital. According to the target treatment of PTZ or CBP, there were divided into PTZ group and CBP group, the clinical date of underlying disease, Charlson's comorbidity index, Pitt bacteraemia score, minimal inhibitory concentration(MIC) of PTZ and 14-d all-cause mortality were collected. Clinical effective rate and survival curve were used to evaluate effectiveness of 2 groups, and the incidence and severity of adverse reactions were used to evaluate safety of 2 groups, and the pharmacoeconomics of two groups was evaluated with cost-minimization analysis. Meanwhile, the risk factors of treatment failure for PTZ were explored. RESULTS A total of 268 patients were isolated, ESBL-producing Escherichia coli were detected from 4 specimen types including pucture fluid(84 cases), urine(64 cases), secretion(61 cases) and blood (59 cases). There was no statistically significant difference in the success rate(76.26% vs 82.86%, P=0.252), 14 d all-cause mortalit(6.06% vs 11.43%, P=0.138) and the incidence of adverse reactions (4.54% vs 8.57%, P=0.230) between PTZ group(n=198 cases) and CBP group(n=70 cases), however, the drug cost of PTZ therapy was lower than that of CBP group(2 035.05 yuan vs 3 250.19 yuan, P<0.001). Septic shock (OR 34.161, 95%CI 3.244-359.744, P=0.003), Charlson's comorbidity index>3 points(OR 1.260, 95% CI 1.019-1.558, P=0.033), CCr<30 mL·min^-1 (OR 3.159, 95% CI 1.068-9.349, P=0.038), MIC of PTZ>4 μg·mL^-1(OR 2.723, 95% CI 1.238-5.993, P=0.013) and bloodstream infections(OR 4.029, 95% CI 1.652-9.827, P=0.002) remained associated with risk of PTZ treatment failure. CONCLUSION PTZ is not associated with increased 14 d mortality and may be an effective drug against infections caused by ESBL-producing Escherichia coli, including bloodstream infection, which can obtain satisfactory curative effect and the best economic effect. However, if patient combined with one of the above five risk factors, it recommend to adjust CBP, or PTZ combined with respiratory quinolone or aminoglycosides.