|
Introduction References Case
Outcome Annotated
Bibliography Questions A First Class Emergency: Headache in Flight Case Presentation A 41 year-old flight attendant boarded a plane in London to return to Chicago, complaining of her typical migraine headache. A co-worker gave her a sumatriptan (Imitrex) tablet. Four hours later her headache was worse. A description of this headache was not available. A nurse on board opened the on-flight medical and administered 10mg of diazepam IM and phenergan. The patient was placed in the first class cabin and four hours later upon landing in Chicago, she was found unresponsive. Paramedics arrived in our ED stating the patient was responsive only to deep painful stimuli. She had a gag reflex; her airway was judged to be adequate at this time. The patient's past medical history was unknown except for the history of migraines. It was unclear what workup the patient had for headaches in the past. Besides working as a flight attendant, her social and family histories were unknown. Her medications and allergies were unknown as well. On exam, the patient's vital signs were temperature of 96.9F orally, heart rate of 63, blood pressure of 149/77 and respiratory rate of 16. Her pupils were reactive with her left pupil measuring 3mm and her right pupil 4mm. Her lung, heart and abdominal exams all appeared normal. There were no cutaneous findings and she appeared well hydrated. Her toes were downgoing bilaterally. To painful stimuli, she responded by localizing movements. She moved all extremities without any apparent focal deficits. An IV was started and a chemistry profile, complete blood count, blood cultures, serum and urine toxicology screens were sent. A CT scan of the brain without contrast was ordered. The CT was reviewed by the ED attending and radiologist. Her CBC revealed an elevated white blood cell count of 15.3, with 95% segmented neutrophils. Her hemoglobin, hematocrit and platelet count were unremarkable as was her chemistry profile. Urinalysis and urine and serum toxicology were obtained (and later reported as normal.) The attending
radiologist reviewed the head CT and noted hydrocephalus with early
cerebral edema. He also noted subtle increased density in the circle
of Willis and suggested that it could represent a low density subarachnoid
hemorrhage. A First Class Emergency: Headache in Flight Introduction Lumbar puncture and the risk of herniation Key Clinical Questions
Lumbar puncture was first described by Quincke in 1891, with complications from the procedure reported as early as 1896. Bleeding, post dural puncture headache, and infection are all well publicized complications. Herniation is by far the most feared complication, but controversy remains regarding the risk of herniation related to LP. Cerebral herniation occurs when intracranial pressure gradients lead to compression of the cranial contents against certain structures. Many different pathological processes can lead to herniation including cerebral edema, space occupying lesions (abscesses, tumors, blood) and hydrocephalus. Lumbar puncture may possibly cause herniation by altering pressures between compartments. By removing CSF and lowering the pressure in the spinal compartment, there can be downward movement of brain structures. The two common types of cerebral herniation are tentorial and tonsillar. Tentorial herniation occurs when the medial portion of the temporal lobe herniates into the tentorial notch. Clinical features include a dilated unreactive pupil due to compression of CN III and the oculomotor nucleus, usually on the side of the mass lesion and herniation. This typically begins on one side but can become bilateral. There is also often contralateral hemiparesis from the ipsilateral peduncle compressing against the tentorium. However, in up to 25% of patients, the contralateral peduncle may be forced against the opposite edge of the tentorial hiatus. Hemiparesis then occurs on the same side as the dilated pupil: a false localizing sign termed Kernohans notch syndrome. Tonsillar herniation is herniation of the cerebellar tonsils through the foramen magnum. Symptoms include neck stiffness and head tilt from impaction of the foramen magnum. Bradycardia and wide pulse pressure can be seen as well. Respiratory abnormalities can progress from Cheyne-Stokes respirations to complete respiratory arrest.
In the 1950's prior to the
development of computed tomography, two studies looked at herniation as
the result of lumbar puncture in patients with brain masses. Lubic and
Marotta1 reported on 401 patients with brain neoplasms who underwent lumbar
puncture, 32% of the patients noted to have papilledema. A complication
was noted in only one case, 12 hours after the procedure. Korein2 in 1959 looked at lumbar punctures performed on 129 patients with increased intracranial pressure or papilledema. 5 patients deteriorated neurologically and 3 died within 48 hours after lumbar puncture. These two studies concluded that there was a low risk of herniation after lumbar puncture even in patients with brain masses. This is no longer relevant with the advent of computed tomography. These studies, however, continue to be cited by future studies that prove CT is not helpful in trying to predict who will herniate. As stated previously, herniation is due to pressure changes: higher pressure above compressing down when the LP needle releases pressure from below. Unless brain shift is noted, it is very difficult to identify CSF pressure based on CT. Baker3 in 1994 compared opening pressures during 42 lumbar punctures to CT diagnosis. 13 of 42 had elevated pressures (>200mmH20.) 6 of those had normal CT's, 1 had hydrocephalus and 1 showed mass and midline shift. None of the patients developed complications 48 hours after LP. 29 of 42 patients had normal opening pressures, and yet 4 had mass lesions (2 with shift and hydrocephalus,) and again none had complications from their LP. In this study, almost half of the patients with increased intracranial pressure (6/13) had normal CT's, suggesting that a normal CT cannot rule out high pressure. The converse is also true; a CT showing mass effect may not have high pressures as well. These studies helped illustrate the point that some combination of increased intracranial pressures and mass effect/brain shift are responsible for causing cerebral herniation. A review in the Journal of Neurology in 2002 emphasizes the above point, by stating that it is important to perform CT before LP if clinical suspicion of brain shift exists and to specifically look for loss of differentiation of gray and white matter, effacement of cerebrospinal fluid spaces and additional displacement of brain structures.4
During bacterial meningitis, brain edema can lead to increased intracranial pressure. Lumbar puncture is considered the standard of care in diagnosing meningitis, yet may be detrimental in some cases. Rennick5 in 1993 studied whether the incidence of cerebral herniation was increased immediately after lumbar puncture in children with bacterial meningitis and whether children with herniation had normal CT's. He looked at 445 children with the diagnosis of bacterial meningitis. 19 of the 445 (4.3%) children herniated. CT was performed at the time of herniation in 14 patients, 5 of which were normal. Rennick concluded "that LP can cause cerebral herniation in bacterial meningitis and normal results on CT do not mean it is safe to do a LP in a child with meningitis. If meningitis is suspected in a child with decorticate or decerebrate posturing, focal neurological signs or no response to pain give antibiotics but do not perform an LP even when the CT looks normal."5 In a review article in Emergency Medicine in 2001, the authors repeated similar suggestions that LP should be delayed or avoided in patients with focal neurological signs, papilledema or a significantly altered mental status.6 In 2002, in a letter to the editor, Kastenbauer noted that among patients with pneumococcal meningitis, clinical and CT examinations failed to identify a substantial proportion of the patients who later herniated. He agreed that CT should precede LP in patients with focal neurological signs, seizures, or a reduced level of consciousness.7
Rennick's and Baker's studies, and many others like it, have shown that CT is not always helpful in identifying who is at risk for herniation. When the decision is made to CT prior to LP however, there are specific finding that should cause concern to the physician. These include any evidence of unequal pressures across the midline especially lateral shift of the midline structures. Loss of basilar cisterns is evidence of unequal pressures between the supratentorial and infratentorial compartments. Gower in 1987 mentioned that the strongest contraindication to LP is a posterior fossa mass. However, this is very difficult to see on CT, although obliteration of the 4th ventricle could be indirect evidence of a posterior fossa mass.
The general consensus has become fairly uniform in which patients should receive CT before LP. Archer in 1993 recommends CT first for patients who are unconscious, show focal neurological findings, papilledema or have other unusual clinical features such as immunocompromise.8 van Crevel in the Journal of Neurology repeats the same suggestions. He emphasized CT is to look for brain shift and not raised CSF pressure, which as stated previously is unreliable in CT and usually present in acute bacterial meningitis anyway. He recommends CT first in suspected meningitis in cases of coma, papilledema and/or hemiparesis.4 Rennick's study offers the most specific and perhaps surprising recommendations. In children with meningitis, he recommends not doing an LP in children with focal neurological, decerebrate or decorticate posturing or complete unresponsiveness, even if the CT is normal. He cites the same studies that CT is not reliable in detecting who is at risk for herniation.
The recommended treatment of intracerebral hypertension has not changed significantly. The mainstay of treatment is 20% mannitol IV - 1g/kg over 15 minutes. Its effects last about 4-6 hours. Corticosteroid treatment with dexamethasone IV 12-24 mg IVP is considered although Rosen states it has no proven benefit. Intubation and hyperventilation are also still frequently recommended although good evidence of beneficial effect is lacking.
CT is not 100% reliable in
identifying who will herniate from lumbar puncture. Patients with normal
CT's may still herniate, while those patients will grossly abnormal CT's
will not necessarily herniate. However, CT findings that should always
be excluded include loss of differentiation of gray and white matter,
effacement of CSF spaces, sulci, fissures, and ventricles, and displacement
of brain structures. Consider deferring LP in patients with focal neurologic
signs, papilledema, and unconsciousness. A First Class Emergency: Headache in Flight Reference
List
A First Class Emergency: Headache in Flight Outcome of Case Upon returning from CT, it was felt this patient could not adequately maintain her airway. She was intubated without any difficulty using midazolam, lidocaine, pancuronium and succinylcholine. She returned to radiology for a head CT with contrast which was read essentially the same as the previous study. Two grams of ceftriaxone were immediately given by vein along with 600 mg of acyclovir and 200 mg of methylprednisolone. An LP was performed with the opening pressure noted to be 20cmH20. The cell count was normal; the CSF glucose was 67 and protein 28. No CSF bacterial antigens were detected by CIE. After the LP, the patient became completely unresponsive. She was taken directly to the MICU and the neurosurgeon performed a ventriculostomy, noting a pressure of 40cmH20 from the ventriculostomy. Despite external
ventriculostomy drain placement, the patient never regained consciousness.
An MRI/MRA showed that the patient had a colloid cyst in the third ventricle.
Although it is not clear what role this cyst played in the patient's
rapid deterioration, it can be postulated that this cyst created a mechanical
obstruction at the level of the third ventricle which contributed to
her herniation.
Annotated Bibliography
A First Class Emergency: Headache in Flight Questions 1. Which of the following can be found during uncal herniation?
2. "Cushing" refers to which of the following?
3. What is the most effective way to initially control a rising intracranial pressure in a head trauma patient?
4. Which of the following herniation syndromes begins with Cheyne-Stokes breathing and small pupils, progresses to hyperventilation and midpoint pupils, and then moves to ataxic breathing, fixed pupils and death?
5. A 19 year old woman has a 6 week history of increasing headache that is worse in the morning and increases with coughing. CT of the head is negative. LP shows an opening pressure of 320 mmH20, with 0 white cells, 2 red cells, glucose of 60 and protein of 35. What would be the appropriate initial management?
Answers 1. E. Classically as the uncus herniates it compresses the ipsilateral 3rd cranial nerve, causing ipsilateral pupil dilatation and contralateral weakness because the pyramidal tract decussates below this level. However, occasionally the shift will be great enough to cause compression on both sides leading to a combination of the above findings. 2. D. This reflex may be a clue to increased intracranial pressure, however, it is often a late finding. 3. D. Maintaining mild hypocapnea is effective in decreasing the intracranial pressure through a vasoconstrictive effect on the cerebrovascular bed. (This point is debatable but still tested like this.) 4. A. These effects are seen with central transtentorial herniation. Tonsillar herniation does not produce the rostral caudal progression of symptoms. Uncal herniation produces asymmetry of symptoms, and subfacial herniation or herniation beneath the falx cerebri causes less CNS dysfunction. 5. D. This patient has pseudotumor cerebri, which is often treated with repeated therapeutic lumbar punctures. |