ED Management of Patients with Febrile Neutropenia: Guideline Concordant or Overly Aggressive? Baugh CW, Wang TJ, Caterino JM, et al. Academic Emergency Medicine. 2017;24(1):83-91 [paper]
Why I chose this article?
It’s always interesting to find something that is practice changing, reasonable, and not commonly done. This fits the bill. I was always taught to admit febrile neutropenic patients.
Febrile neutropenia (FN) is common, occurring in 10-50% of patients with solid tumors and >80% in hematologic malignancies. Up to 70% are never found to have a bacterial source of infection. Infectious Disease Society of America (IDSA), American Society of Clinical Oncology (ASCO), and National Comprehensive Cancer Network (NCCN) all recommend risk stratification of patients with febrile neutropenia. Low-risk patients can be considered for treatment as an outpatient with close follow up and ability to access medical facilities at all times.
Low-risk patients are classified as patients with a Multinational Association for Supportive Care in Cancer (MASCC) score ≥ 21 and no other indication of need for inpatient management. MASCC score has a PPV of 91%, sensitivity of 71%, and specificity of 68%.
Guideline concordant therapy for low-risk FN is discharge with oral antibiotics (typically a fluoroquinolone) and close follow up. High-risk is further stratified for choice of IV antibiotics based on guidelines. Vancomycin is used frequently and has specific guidelines for its use in FN.
Vancomycin should be used for
- Severe sepsis/hemodynamic instability
- Blood cultures positive for Gram positive organism
- Known resistance
- Soft tissue/skin infection present
- Suspected catheter related infection
- Radiographically apparent PNA
- Severe mucositis
Guideline concordance and clinical outcomes of patients with febrile neutropenia presenting to the ER at a tertiary care center associated with a comprehensive cancer center.
Retrospective cohort via chart review of patients with FN defined as ANC <1000 cell/uL and temp >100.4°F (38° C).
Patients were categorized as low or high risk based on a MASCC score of ≥21 and no high risk IDSA and ASCO criteria. Two criteria from the score are subjective, the burden of disease and the presence of dehydration requiring IV fluids. These were assigned based on level of distress in notes and vital signs for the burden of disease and by signs of dehydration documented in the notes, given IV fluids, and BUN/Cr ratio.
- Neutropenia < 1000 cells/uL
- Temp > 100.4°F
- Cancer treatment in past 6 months (active disease)
- Unknown etiology of fever
Primary: Guideline concordant management – disposition and route of antibiotic were in agreement with guidelines.
Secondary: Identification of bacteremia, sepsis induced hypotension or death within 30 days, and use of vancomycin.
930 cases of neutropenia were identified and 326 of these cases had a fever. 153 cases were excluded for a known etiology or without active cancer.
173 used for analysis with 44 classified as low-risk and 129 classified as high-risk.
52 patients were not given guideline recommended care (43 of low-risk and 9 of high-risk). All except 1 patient from the low-risk group were admitted with the majority of patients receiving IV antibiotics. Discordance of the high-risk group was secondary to inappropriate antibiotics (oral or none) and 2 patients were inappropriately sent for outpatient treatment.
For the secondary outcomes they noted 26 patients overall and 18 high-risk patients received vancomycin when not indicated. Only 1 (3.1%) patient of the low-risk was culture positive and none had sepsis induced hypotension or death in 30 days.
No patients were lost to follow up.
- Retrospective single site study.
- Small numbers especially in the low-risk group.
- Subjective aspect of the chart review regarding the MASCC score in regards to the burden of disease and dehydration.
Guidelines recommend the possibility of outpatient treatment of FN in low-risk patients. This paper points to ED management being overly aggressive and leading to unnecessary antibiotic use and admissions. We tend to forget that broad spectrum antibiotics and hospitalization are not without risk (C. Diff, resistant strains, cost of care) and neutropenic patients are at higher risk of iatrogenic complications. I think with strict criteria for ensuring outpatient follow up and selecting low-risk patients, this is a valid tool that can be used to prevent hospitalization unnecessarily in an undifferentiated FN patient. It is important to note that everyone (ED, oncology, patient, caregivers) should be on board with this and follow up should be arranged before discharge, as even low-risk patients can subsequently worsen. Keep in mind that low-risk doesn’t mean no-risk.
This is even more evident as they discuss how the MASCC score isn’t perfect by any means, but has been validated and in use for 16 years. The Clinical index for Stable Febrile Neutropenia score (CISNE – not sure where the F goes or the E comes from) is mentioned as a newer better tool but it hasn’t been adopted into the guidelines yet.
Another piece to keep in mind is local resistance patterns and choice of outpatient oral antibiotic therapy. A fluoroquinolone + amoxicillin-clavulanate is the recommended oral therapy, but options are limited in areas with high fluoroquinolone resistance. Discussions between the ED, ID, and oncology could establish standardized outpatient regimens with the plan to admit for parenteral antibiotics if the patient is unable to take the recommended oral antibiotics.
Observation admission is also a possibility for low-risk patients to blunt the risks of a full admission and it could allow for arrangement of home parenteral antibiotics if those were necessary.
One major take-away: don’t use vancomycin unless it is indicated. When starting antibiotics from the ED it is knee-jerk to include vancomycin to cover for MRSA. Sometimes we cannot avoid it as new sepsis protocols and guidelines have made it mandatory. Learning and documenting the rationale for using or not using vancomycin can help avoid overuse of this medication leading to increased resistance and hard to treat pathogens such as VRE.
Overall this article highlights a patient population that a small change in our practice could really benefit. It is surprising that these guidelines have been out for years but many people are not aware of them. Hopefully this article will shine some light on the topic and bring it more to the forefront.