Empiric Therapy for Febrile Neutropenia in a Penicillin-Allergic Patient
A 34-year-old woman with acute myeloid leukemia is day 12 post-induction chemotherapy. ANC is 80 cells/mcL and falling; temperature is 38.6°C. She has a documented 'penicillin allergy' consisting of a childhood rash that resolved without treatment. She is hemodynamically stable with no focal signs. Blood cultures are pending. She has a history of Clostridioides difficile infection 6 months ago, now resolved. Renal function is normal. The local antibiogram shows that 18% of E. coli isolates are ESBL-producing and that Pseudomonas aeruginosa susceptibility to piperacillin-tazobactam is 82%.
1. How do you classify her penicillin allergy, and how does that affect your empiric antibiotic choice?
Read the worked answer
A childhood rash that resolved without treatment is most consistent with a benign, non-severe delayed (or possibly viral) exanthem, not an IgE-mediated reaction or severe cutaneous adverse reaction. This is a low-risk label and does not preclude beta-lactam use. Given febrile neutropenia is a high-risk condition requiring prompt broad-spectrum coverage, I would use an anti-pseudomonal beta-lactam-piperacillin-tazobactam or cefepime-with appropriate monitoring, rather than avoiding beta-lactams and accepting inferior Gram-negative coverage. If there were any concern, a graded challenge or allergy consultation could be arranged, but the risk here is low and the benefit of prompt appropriate coverage is high.
2. Given the local ESBL rate and Pseudomonas susceptibility, which empiric regimen do you choose and why? What role, if any, does the prior C. difficile history play?
Read the worked answer
The 18% ESBL rate is high enough that I would want to know whether the patient has prior ESBL colonization, but for a first febrile neutropenia episode, standard guidance is an anti-pseudomonal beta-lactam (cefepime or piperacillin-tazobactam). Piperacillin-tazobactam susceptibility for Pseudomonas is only 82%, which is borderline; cefepime may be preferable if local cefepime susceptibility is higher. Carbapenem-sparing is generally preferred to limit resistance, but if the patient is unstable or has prior ESBL, a carbapenem would be reasonable. The prior C. difficile history argues against unnecessarily broad or prolonged anaerobic coverage; piperacillin-tazobactam has anaerobic activity, so if C. difficile risk is a concern, cefepime (no anaerobic coverage) may be preferred, with attention to C. difficile testing if diarrhea develops. I would avoid adding anaerobic coverage without an indication.
3. What would make you add vancomycin or an antifungal, and what monitoring would you institute?
Read the worked answer
I would add vancomycin if there is suspected catheter-related infection (e.g., exit-site erythema, tunnel tenderness), skin/soft tissue infection, pneumonia, or hemodynamic instability, per guidelines. Antifungal coverage would be considered if fever persists beyond 4-7 days of appropriate antibacterial therapy with no identified source, or if there are risk factors (prolonged neutropenia, prior invasive fungal infection, mold exposure). Monitoring: daily CBC with differential until ANC recovery, renal function, and cultures; reassess at 48-72 hours-if afebrile and cultures negative, narrow based on results. I would also monitor for C. difficile if diarrhea occurs and avoid unnecessary prolonged therapy.