Risk factors for Cefepime Resistance in Bacteria Causing Gram-negative Bloodstream Infections in a Children’s Cancer Center

Abstract

Abstract Background Strategies that seek to provide timely and effective therapy for febrile neutropenia (FN) in children with cancer or hematopoietic cell therapy (HCT) are threatened by increasing rates of antimicrobial resistance (AR). At our institution clinically stable patients with FN are treated with cefepime but this carbapenem-sparing strategy results in delays in the time to effective antimicrobial therapy and poorer clinical outcomes in patients with Gram-negative bloodstream infection (GNBSI) caused by cefepime nonsusceptible (CN) bacteria. Identification of risk factors for CN may allow providers to individualize treatment and improve outcomes. Methods This retrospective single-center study included patients <22 years of age who had received immunosuppressive therapy within 3 months and had a laboratory confirmed GNBSI between 11/1/21 and 2/28/25. Demographic and clinical data were abstracted from patients’ medical records. Univariate and multivariable logistic regression analyses were employed to identify risk factors associated with cefepime nonsusceptibility. Variables with a p-value below 0.1 in the univariate analysis were included in the Least Absolute Shrinkage and Selection Operator (LASSO) analysis to identify variables with meaningful associations with the outcome. Variables retained by LASSO were then included in the multivariable logistic regression model. A p-value <0.05 in the final model was considered statistically significant. Results Among 209 episodes of GNBSI, 89 (43%) were caused by cefepime-nonsusceptible (CN) organisms and 120 (57%) by cefepime-susceptible (CS) organisms (Table 1). CN episodes involved patients who were older and were more common among those of Hispanic ethnicity and with leukemia. Body mass index was higher and time since cancer diagnosis shorter in those with CN. Patients with CN isolates were less likely to have received prior medical care at another US medical center and more likely to have received care at a foreign center. Prior colonization or infection with an ESBL-expressing, carbapenem-resistant bacteria or vancomycin-resistant Enterococcus were all significantly associated with CN infection. Receipt of cefepime and fluoroquinolone or trimethoprim-sulphamethoxazole prophylaxis in the 30 days prior to the GNBSI were also more common in the CN group (Table 1). In multivariable analysis, prior ESBL+ bacteria (adjusted OR 5.45; 95% CI, 2.35-12.64), cefepime prophylaxis (adjusted OR 3.07; 95% CI, 1.08-8.72), fluoroquinolone prophylaxis (adjusted OR 2.80; 95% CI, 1.31-5.98), and outpatient onset (adjusted OR 0.30; 95% CI, 0.14-0.62) remained significantly associated with CN (Table 2). Conclusion Consideration of individual risk factors for infection with CN GBSI may allow providers to modify empiric antimicrobial therapy in high-risk groups, improving infection outcomes.