Headache
                     
Headache is a vexing complaint for EP's. It accounts for 1-3% of ED visits. If you worked 30y as an EP and saw the typical 2.5pts./h for 1,600h/y you would see ~120,000 patients. About 2,400 would need to be evaluated for headache. Of these perhaps at most 1 in 500 would have some kind of catastrophe like subarachnoid hemmorhage or bacterial meningitis. This means every five years you see about 20,000 patients and 400 of them have a headache. Out of this group you are looking for the one patient (at most) who has something significant wrong with them. And God forbid you miss that one!!! The goal of this lecture is to help you learn how to efficiently and accurately screen for intracranial catastrophes.

In order to be SURE of NOT missing that one out patient over five years you would need to do some combination of CT or CTA and/or LP in everyone. Completely impractical, actually it would be ridiculous. So you need to have some way of deciding who to work up beyond the H+PE and what workup to do. I am going to explain this to you by using the concept of "red flags" - facts from the H+PE that should lead to serious consideration of performaing a workup.

 

Chronic Headacher

First of all, a "green flag" - the chronic headacher. A patient who has chronic headaches most often carries a diagnosis of "migraine". This may or may not be justified. But it's very reassuring when a patient says in so many words "I am haivng another one of my headaches". The history should focus on identifiying this headache as substantially the same in character, location, associated symptoms, etc. as their prior headaches. The physical exam can be fairly brief - focus on the neuro exam and be sure to document the absence of meningismus or fever. If the H+PE show no surprises, symptomatic treatment is indicated (discussed below). If the treatment works - discharge the patient. If it does not, reassess.

We will come to the red flags soon, but first you must make a binary decision. You must answer the question "is this a primary headache syndrome or is the headache part of a broader symptom complex?" For example: a patient has a cough, stuffy nose, sore throat, feels feverish, etc. even though the triage complaint is "headache". Absent something really striking on the PE this history is NOT compatible with an intracranial catastrophe. Similarly, the "headache" may turn out to some obvious local pathology like a scalp abscess, cranial zoster, otitis externa, etc. But if the chief complaint is really headache (i.e. a "primary" headache), what should you look for?

Subarachnoid Hemmorhage
The grandaddy of headache red flags is the sudden onset of very severe headache reaching maximal intensity in seconds to minutes (thunderclap headache). Nearly all (97%) patients with aneurysmal subarachnoid hemmorhage (SAH) have the sudden onset of a very severe headache, classically described as "worst headache of my life". There should be no question that a patient with this history must be worked up for SAH. There are two acceptable ways to rule out SAH in the setting of a thunderclap headache.

 

CT/LP for the diagnosis of SAH

  • Noncontrast CT is ~97% sensitive for the diagnosis of SAH if performed w/in 12h of HA onset.
    • sensitivity decreases with time
    • to pick up cases of SAH missed by CT, an LP should follow a negative CT in cases of thunderclap headache
CT/CTA for the diagnosis of SAH
Our own Dr. McCormack published a superb review and analysis of this concept in Academic Emergency Medicine.
  • In that analysis CT followed by CT angio  (CTA) strategy had a sensitivity exceeding 99% to exclude aneurysmal SAH.
    • This is a reasonable strategy but will probably never be prospectively validated due to the huge sample size required. 
      • R. McCormack et al.  Can Computed Tomography Angiography of the Brain Replace LP in the Evaluation of Acute-onset Headache After a Negative Noncontrast Cranial Computed Tomography Scan.  AEM.  2010.
Meningitis in Adults
Bacterial meningitis is a very uncommon disease in adults. The incidence increases with age and with certain types of immunosupression, but overall is a bit more than ~1/100,000 people per year in the US. If each one presented to the ED this would mean that if you saw 120,000 patients in a 30y career in EM you could expect to see 1 or 2 cases. Most patients present with at least 2-3/4 of the following:
  • Headache (nearly all if they are awake enough to complain). The headache is typically severe and it should be obvious that it is not a "normal headache"
  • Fever (~95%)
  • Stiff neck (~90%) This is not a subjective complaint but a physical finding: flexion of the neck shows an inability to touch the chin to the chest.
  • Altered mental status: this may be subtle (inattention, slow to anwser questions, etc.) or profound and there may be focal neurological deficits.
Consider an LP for the following types of patients (to rule out meningitis (note the caveat)).
  • Severe HA with any of the other three (caveat: many febrile illnesses are associated with headache (ex. influenza). If the typical "viral syndrome" (abrupt onset of fever, headache, myalgia, and malaise accompanied by respiratory tract illness, such as non-productive cough, sore throat, and nasal discharge) an LP is clearly not needed). 
    • A non-validated clinical observation: if you suspect a simple viral syndrome in a febrile HA patient a benign cause is suggested if antipyretics relieve both the fever and the HA).
  • Severe HA and a focus of bacterial infection (such as UTI, pneumonia, otitis media, sinusitis).
  • Severe HA and recent CNS instrumentation.
  • Unusual headache in a non-immunocompetent patient. 
  • Headache, fever, and an unusual rash, especially petechial or purpuric (suggest meningococcal mengitis). 
Should a CT performed prior to LP?
  • There are lots of different viewpoints on this. Mine is that:
    • If the most likely serious diagnosis is meningitis AND there are NO focal deficits perform LP immediately. 
    • If meningitis is high but not first on your list, give empiric abx after blood cultures, get a CT and follow up with the LP. 
    • If menigngits is low on the list, get CT and if that is not revealing give empiric abx and then do the LP.

Rash in Meningococcal Mengitis

Headache and Focal Neuro Findings or Seizure

By far the easiest type of HA in the ED is the combination of HA and focal neurological findings or a seizure. This suggests a space occupying lesion such as tumor, abscess, intracranial hemorrhage or a stroke. All of these patients need imaging. There are various imaging protocols, here are my recommendations.
  • Presentation w/in 3h of onset: STAT non-contrast head CT
    • CT should be completed and read w/in 3h. to determine if there are no contraindications to thrombolysis
    • There is some evidence that thromblytics may be effective up to 4.5h in ischemic stroke
  • Presentation >3h from onset or unknown onset
    • Consider transfer to a stroke/endovascular center
      • If you are in a stroke center, there are probably protocols you should follow. The ACR reccomends the following.
        • STAT non-contrast head CT to r/o hemorrhage
        • STAT MRI with DWI or CT perfusion scan of the brain
          • These identify potentially viable areas of brain tissue in the "ischemic penumbra".
  • Unilateral HA/neck pain* and neurofindings suggests carotid dissection
    • CT/CTA brain, CTA neck
      • the other thing to think about is cervical epidural abscess (very rare). If the chief complaint is really neck pain (which often gets labelled as "headache") order MRI of the cervical and thoracic spine. If MRI is not available - CT with IV contrast.

Cervical Epidural Abscess with Contiguous Osteomyelitis
  • Focal findings, seizures, etc. also occur in SAH and meningitis but if these are the cause, other features should suggest the correct diagnosis and imaging is going to be performed anyway.

The "Unusual" Headache
 The "unusual" headache is either unusual for the patient or doesn't seem to fit any of the above clinical constructs. This is the most difficult type of HA to decide how to work up in the ED. The best way to approach unusual headaches is to use pattern recognition.
  • Throbbing HA with photophobia, N/V, a family history of migraine all suggest migraine. If the neuro exam is normal and red flags are absent you can treat empirically. 
  • Temporal HA in an older patient (at least 50) especially with vague systemic c/o and visual changes suggest temporal arteritis.
    • Check the ESR. If >50 start treatment with high dose prednisone and arrnage prompt follow up. Will need a temporal artery biopsy.
  • Exertional HA suggests carotid dissection
    • CT/CTA head and neck
  • HA with visual complaints
    • Check the eyes to rule out acute glaucoma
  • HA and anticoagulants, advanced liver disease, FH of SAH/aneurysm
    • CT head to r/o bleed
  • HA that involves a cluster of people in the same home, workplace in the heating season.
    • CO level to rule out carbon monoxide toxicity
  • HA in moderate to severe immunocompromised patient
    • MRI w/o contrast to r/o mass lesion
    • CT with and w/o IV contrast if MR not available/practical
  • HA needing imaging in a pregnant patient
    • MRI/MRV w/o contrast
  • unusual or severe HA and pro-thrombotic disorder/history
    • MRI/MRV to r/o cerebral venous sinus thrombosis
    • CT with contrast if MR not available
      • Discuss with your radiologist or techs so that the study is performed optimally
  • unusual or severe HA and cocaine or severely elevated BP
    • CT head to r/o hemorrhage

Non-Specific HA Treatment
My HA cocktail consists of three classes of meds given together. They work by different mechanisms and my anecdotal experience indicates at least 80% sucess. This meds cause drowsiness, so expect that the patient will sleeep for a while and wake up feeling better.
  • NSAID
    • IV Toradol, 30mg
  • Dopamine agonist (IV only, PO not effective)
    • Prochlorperazine (Compazine) 10mg IV or chlorpromazine, 10mg IV
  • Dipehnhydramine, 12.5mg IV pre-treatment to prevent akasthesia
ONLY if the above doesn't work do I resort to opiates and I try PO first.
  • Avoid narcotic prescriptions if at all possible