-

November 8th, 2012

Steroids for Bell’s Palsy and the ID Doctor

OK, let’s imagine you’ve just gotten a call/email/text from one of your colleagues about Bell’s palsy; he/she is a busy PCP who periodically asks you very reasonable ID questions.

I suspect it went something like this:

COLLEAGUE:  Hi Friendly ID Doctor, quick question — I have a patient with Bell’s palsy — wondering whether to give him steroids.
FRIENDLY ID DOCTOR:  Steroids? Absolutely not! Don’t you know that a host of infectious diseases cause Bell’s palsy, including Lyme, HSV, VZV, HIV, syphilis [and other obscure diagnoses I'm too tired or too forgetful to list]. Have you excluded these?
COLLEAGUE: Well, I just read that the neurologists are recommending steroids …
FRIENDLY ID DOCTOR: I’d get Lyme titers, and start the patient on doxycycline and high-dose valacyclovir. In addition, [insert boilerplate language about the limitations of curbside ID consults here].
COLLEAGUE: But the neurologists say antiviral therapy doesn’t really work. And doxycycline? Why?
LESS-FRIENDLY ID DOCTOR: Never mind — how about you call a neurologist and ask what to do?

Every so often, something comes along that shows that you are absolutely and completely and totally biased in your approach to a problem by your perspective. Whoppingly biased, if that’s a word. I confess that Bell’s palsy is one of those things.

(I didn’t even know that the preferred term is now “Bell palsy”, without the possessive. Sounds weird. I’m sticking with “Bell’s”.)

To me — and probably to most Infectious Diseases specialists — a patient with Bell’s is an ID problem worth solving.

But to the vast majority of the rest of the world, including these new guidelines and the great swath of primary care providers and neurologists, it’s “an acute, peripheral facial paresis of unknown cause.”

Emphasis mine.

And after reading this evidence-based review, I confess that the data supporting the use of steroids for Bell’s palsy are far stronger than those for antiviral agents or, with my New England bias, doxycycline, which was not even mentioned in the guidelines. Oh well.

Nonetheless, for providers who see a new case, especially in the warm months, please humor me — send a Lyme titer. And at least think about those other ID diagnoses that cause Bell’s.

I mean Bell.

7 Responses to “Steroids for Bell’s Palsy and the ID Doctor”

  1. In this Lyme endemic area of northern California, Bell’s must be considered to be neuroLyme – top of list. There is no test that reliably rules it out. Only in. Any doc who thinks there is a r/o test for Lyme is mistaken. Too many have lost their face around here. If you want to get a ‘Lyme titer’ (Elisa, WB, FTA etc), do it as a data point to be followed, not for diagnosis. Unless it’s +, of course.

  2. Marat says:

    I think there is a misunderstanding caused by inappropriate use of term “Bell’s palsy”. This article is about diagnostic approach to facial nerve paralysis, not about Bell’s palsy. Lyme disease (and other known diseases) cause facial nerve paralysis. On the other hand, when the etiology is unknown (when we have excluded possible causes), we call it Bell’s palsy. Hence, we should not use term “Bell’ palsy” in the case of Lyme disease, etc. Guidelines from American Academy of Neurology deal with the Bell’s palsy, not with the facial nerve paralysis.
    Also, according to Uptodate, “screening for antibodies to B. burgdorferi is not recommended among adults with seventh nerve palsy who do not have additional manifestations of Lyme disease.” “As an example, one study of 69 patients presenting with presumed idiopathic facial nerve palsy found that none of the patients had serologic evidence of Lyme disease” – http://www.ncbi.nlm.nih.gov/pubmed/1520085.

  3. “As an example, one study of 69 patients presenting with presumed idiopathic facial nerve palsy found that none of the patients had serologic evidence of Lyme disease” –

    Exactly!! – ‘serologic evidence of Lyme’ cannot be used to rule out LD!!! There are so many patients with negative serology who become +++ after treatment with antibiotics. It’s the nature of the bug vs our immune systems. Even the CDC states that serology is not to be used for diagnostic purposes. Only epidemiology. If you crave evidence of infection on paper, consider doing a CD57, C4a, or a real culture from Advanced Labs. But none will give a result before a decision to treat should be made.

  4. Marat says:

    Jon Sternhold
    When did CDC state that serology is not to be used for diagnostic purposes? So you will make a diagnosis of Lyme disease in patient with ONLY facial paralysis and no other symptoms, on basis of epidemiology only (living in endemic area)?
    Provide references, please.

  5. http://www.columbia-lyme.org/patients/ld_lyme_symptoms.html

    Scroll to cranial neuritis.

    And from http://www.angelfire.com/biz/romarkaraoke/lymeart.html

    “Neurologic problems, which occur in 15%-40% of patients (25), may be the presenting symptom. Early on, patients may experience headaches without any signs of inflammation in the CSF (18). Shortly thereafter, patients may develop meningitis, cranial neuritis,…”

    Those of us who work in Lyme endemic areas see this all the time. That’s not to say this is a cut and dried situation. With NO reliable Lyme test to r/o disease, a high prevalence of ECM negative disease, and a patient with a potentially devastating malady, the risk of abx use is overshadowed by the risk of not treating. Someday we’ll have a reliable, fast test. Until then, it’s a huge problem.

    Taking this further opens too large a can of worms for this thread. “The Lyme wars’ rage on, sadly.

  6. Terry Wuerz MD says:

    Would also appreciate references.

    I would add, getting better after antibiotics is often a serendipitous event. It does not prove cause and effect. See: viral URTI or most cases of sinusitis (also viral) as an example.

  7. Todd Ellerin says:

    I appreciate Marat’s comments that Lyme facial palsy should be in a distinct category from Bell palsy and this is not an issue of semantics as the treatment paradigm is different: doxy for Lyme palsy (assuming no meningitis) and steroids
    -/+ antiviral for Bell.

    I’d like to take an opposing role from Dr. Sterngold. I’ve found that lyme antibody is almost alway positive in patients with lyme facial palsy. Can’t quote the sensitivity but one would expect the lyme IgM western blot to be positive in stage 2 lyme. I’ve treated thousands of patients with lyme disease (various stages) and can’t remember a case where I thought the initial lyme test was a false negative in a patient who turned out to have a lyme facial palsy.

    IThe clinical features of these facial palsies are different: Bell palsy is an acute presentation with generally no prodrome whereas lyme palsy usually has features of stage 1 lyme which can include low grade temps, aches, EM rash, etc. I’ve found this to be particularly helpful in adults, perhaps less so in children.

    I agree that lyme testing should not be performed in all its who present with facial palsies but in warmer months in New England when pretest prob of lyme disease is high, a negative lyme serology makes me feel pretty comfortable that the dx is not lyme dz. I still don’t recommend sending lyme serologies in pts who present with acute onset of facial palsy w/o prodrome in summertime in NE b/c you get some pts who have a +screen f/b +IgG western blot (neg IgM) and this is likely a Bell palsy in someone exposed to lyme in the more distant past. History is still critical.

    That’s my take on it, at least for for now..

Leave a Reply

Note: This is a moderated forum. By clicking on the "Submit Comment" button below, you agree to abide by the NEJM Journal Watch Terms of Use.

Our physician bloggers cannot respond to requests for personal medical advice, and recommend patients discuss health issues with their individual physicians.

HIV Information: Author Paul Sax, M.D.

Paul E. Sax, MD

Contributing Editor

NEJM Journal Watch
Infectious Diseases

Biography | Disclosures | Summaries

Learn more about HIV and ID Observations.