As an intern, one of the highest-yield schemas I learned was for persistent fevers despite treatment with abx.
I use it all the time as a resident, especially when working with cancer patients.
Let's go through it!
- Thread -
#MedTwitter #MedEd #FOAMed #tipsfornewdocs
The general ddx for persistent fevers in the hospital, despite abx:
1. Wrong bug
2. Wrong drug
3. Wrong process
4. No source control
5. Not enough time
Let's take them one at a time.
1. Wrong Bug
When giving abx, we assume it's bacterial, but after some time, you should also consider:
- viruses - ex: CMV, EBV, HBV, HCV
- fungi - ex: candida, aspergillus, PJP, etc
- atypical infections - tickborne, TB, etc.
In neutropenic fever, if a patient has been fevering for 4–7 days despite BSA, it may be reasonable to broaden to fungal coverage.
Voriconazole and posaconazole are preferred.
Fluconazole is saved for known candida, as it does not cover endemic mycoses or Aspergillus.
Some factors that may push you toward broadening to fungal coverage include:
- positive BDG
- severe sepsis
- CT with GGOs and/or nodules
- prolonged ANC <500
- TPN use
Though not part of the guidelines, one common misstep is not covering for atypicals.
Ex: A patient comes in sick and gets typical BSA with vanc/zosyn.
Only a few days later does someone bring up that our abx selection, though broad, has some gaps (i.e azithro or doxy)
If you really suspect an infectious process, you should go back to the drawing board and complete a full ID and social history intake.
See this sample template at @pointofcaremed
pointofcaremedicine.com/infectious-disease/id-social-history
2. Wrong Drug
This is most commonly due to MDROs.
Some examples and common treatments:
- ESBL GNRs - carbapenem
- VRE - daptomycin or linezolid
- Carbapenem-res PsA - many options (ceftaz-abi, mero-vabor, cefidericol, etc.)
- Candida Kruseii - voriconazole
It can be intimidating to reach for these big guns since most of us don't have much experience with them.
If you think you have an MDRO or if cultures show that you do, consult ID for advice on the best agent and antibiotic stewardship.
#IDTwitter - Any pearls on MDROs?
3. Wrong Process (not infectious)
Many things other than infections can cause fevers (especially in cancer patients).
Here are just a few:
- Drug fever
- Malignancy
- Transfusion reaction
- Thrombophlebitis
- Autoimmune Disease
In neutropenic fever, it's particularly unfair that cancer itself, the needed transfusions, clots (which all cancer patients are susceptible to), and the antibiotics meant to address the fevers in the first place, can all be the culprit.
twitter.com/ROKeefeMD/status/1617567957445672961
Some common drugs that can cause fevers:
- Antibiotics - penicillins, cephalosporins
- AEDS - ex: phenytoin, carbamazepine
- Anti-neoplastics
- Allopurinol
If the patient fits more into the "Fever of Unknown Origin (FUO)" bucket, it can be helpful to complete a full rheum review of systems.
See this @pointofcaremed template for some guidance on your HPI and exam.
pointofcaremedicine.com/rheumatology-immunology/rhematology-review-of-systems
That being said, FUO falls into its own separate ddx schema.
Be on the lookout for a future thread on this topic based on a recent case conference with @CoreyDWatts!
In the meantime, you can check out this template for working up FUO.
pointofcaremedicine.com/infectious-disease/fever-of-unknown-origin
4. No Source Control
This is a common reason for an infectious cause of FUO.
Imaging can be helpful for localizing a source.
In particular, you should think about:
- endocarditis
- CLABSI
- osteomyelitis
- liver/abdominal abscesses
TTE to look for endocarditis should be reserved for those with pos BCx.
The workup and management of endocarditis is its own beast, best saved for another time.
But you can learn more with this @pointofcaremed template.
pointofcaremedicine.com/infectious-disease/endocarditis
Do know that blood culture-negative endocarditis (BCNE) does happen, often due to getting abx before BCx draws.
Other causes:
- Coxiella burnetti
- Bartonella (quintana/henselae)
ncbi.nlm.nih.gov/pmc/articles/PMC5708915/
5. Not Enough Time
Even with the right treatment, fevers can last for a few days. This is especially true for diseases that are hard to treat and require prolonged courses of abx, like endocarditis or osteomyelitis.
Fevers caused by a virus can last for several days as the body goes through its natural process.
It can be challenging to sit on your hands in these situations, but it's often the right choice if the patient is stable!
Lastly, though not part of the schema, you should be mindful of fevers 2/2 a new infection from the iatrogenic spread of respiratory viruses!
If it seems new or the patient has URI symptoms, repeat that RVP!
In general, after thinking through the schema, the decision to broaden abx or go digging for a source is dependent on clinical judgment, the patient's status, and their risk of decompensation.
It's one of the many gray areas in medicine where there is no easy answer!
Check out the @GlassHealthHQ page I made on this topic!
glass.health/read/XhjZajixP9/
Let me know if there's anything you'd add to it!
@dereckwpaul
@ASanchez_PS
@EvonneMcArthur
@madellenaconte
Do you have any favorite pearls for fevers in these challenging patients?