Prevalence, misclassification, and clinical consequences of the heteroresistant phenotype in Escherichia coli bloodstream infections in patients in Uppsala, Sweden: a retrospective cohort study.

Summary

Among 255 E. coli bloodstream infections, heteroresistance was common (gentamicin 43%, piperacillin-tazobactam 9%) and misclassified as susceptible in 96% of cases by routine testing. When patients received the implicated antibiotic, heteroresistance was associated with higher odds of intermediate/ICU admission and mortality. Length of stay was unaffected.

Key Findings

• Breakpoint-crossing heteroresistance was detected in 43% (gentamicin) and 9% (piperacillin-tazobactam) of E. coli isolates; <1% for cefotaxime.
• Routine clinical susceptibility testing misclassified 96% (120/125) of heteroresistant isolates as susceptible.
• In patients treated with the implicated drug, heteroresistance was associated with higher odds of intermediate care (piperacillin-tazobactam BCHR: OR 3.1, 95% CI 1.1–9.6) and ICU admission and mortality (gentamicin BCHR: ICU OR 5.6; mortality OR 7.1).
• Heteroresistance showed no association with length of hospital stay.

Clinical Implications

Consider heteroresistance in treatment failures; integrate population analysis or alternative detection algorithms for high-risk antibiotics; avoid monotherapy with agents showing BCHR when feasible.

Limitations

• Single-region retrospective cohort limits generalizability
• Focused on three antibiotics; broader spectrum agents not assessed

Future Directions

Develop rapid clinical assays for heteroresistance detection and evaluate stewardship interventions and outcome impacts in multicenter prospective studies.