Treating neurological emergencies in the hospital? Avoid these 10 mistakes

WHEN APPROACHING neurological emergencies in the hospital, what do hospitalists often get wrong? At this fall’s management of the hospitalized patient conference at the University of California, San Francisco, S. Andrew Josephson, MD, chair of the UCSF department of neurology, was tasked with describing those common errors. Here are the top 10 mistakes that Dr. Josephson warned hospitalists to avoid.

Mistake No. 1: not ordering the right test for patients with bilateral weakness

When someone having a neurological emergency in the hospital presents with bilateral weakness in their extremities—bilateral legs, bilateral arms—”the first thing to do is have the respiratory therapist come down and do an FVC/MIF [forced vital capacity/mean inspiratory flow],” said Dr. Josephson. “It’s a cheap test, and it can be lifesaving.”

Abnormal FVC/MIF values indicate that “this is someone who may be developing ventilatory failure, and you’ve got to think of Guillain Barre syndrome, myasthenic crisis or other neuromuscular emergencies. These patients need to be monitored in the ICU, and you want to intubate them before they get in bad shape.”

“When someone is herniating, you want to let the blood pressure be high.”

S. Andrew Josephson, MD
University of California, San Francisco

Patients with Guillain Barre can have an unusual presentation. “People talk about tingling and numbness, but usually their sensory exam is normal,” Dr. Josephson explained. “Their weakness is not always ascending and while areflexia occurs, it’s often not present in the first week for many patients.” And while spinal fluid may have high proteins and normal red blood cells, “early in the disease, you might not see this pattern. You have to think about Guillain Barre empirically.”

Treatment is either IVIG or plasmapheresis. “They’re equal, they’re both effective and the earlier you give treatment the better,” he noted. Meanwhile, don’t give these patients steroids. “Steroids have the potential to Guillain Barre worse.”

And keep in mind that these patients have a high risk of thromboembolism. “We give them DVT prophylaxis, and Guillain Barre can also involve the autonomic nerves,” Dr. Josephson said. “We worry about them developing an ileus or cardiac arrhythmia.”

Mistake No. 2: not using the right exam screen for upper motor neuron weakness

When treating neurological emergencies in the hospital, any movement in any part of the body entails a two-neuron system.

“The first neuron starts in the brain, sending an axon all the way down to the spinal cord where it synapses on a second neuron, which then goes out to the muscle,” Dr. Josephson said. “We call the neuron that starts in the brain the upper motor neuron.”

The good news, Dr. Josephson added, is that when upper motor neurons are injured, that weakness produces a predictable pattern: The distal muscles are weaker than the proximal ones, and extensors are weaker than flexors in the arms (although dosiflexors are weaker at the ankle).

To figure out if a patient with weakness has an upper motor neuron (brain or spinal cord) problem, Dr. Josephson recommended a three-step exam. First, check for pronator drift. Have the patient stretch both arms out with their eyes closed, fingers extended and palms up. If one of the arms or palms turns inward and down, “the arm extensors are preferentially weak,” he said. “That’s pronator drift.”

Next, have the patient do fast finger movements with their thumb and first finger, first one hand and then the other.

“If one hand is slower than the other, that’s upper motor neuron weakness,” he said. Patients can also tap their feet to see if one foot is slower.

The third step: Test one muscle in each extremity. “I like doing finger extensors, so I ask patients to put their hands out and not let me push their fingers down.” Same with the feet: Have them point their toes up to their head, and “don’t let me pull them down.”

If all three tests are normal, “the patient’s weakness isn’t due to a problem in the brain or the spinal cord.” Unfortunately, the screens that are typically done—asking the patient to squeeze fingers or “step on the gas”—are really insensitive as they are targeting the wrong muscles.

“Those screens test big flexor muscles, and those are the wrong ones to test,” Dr. Josephson said.

Mistake No. 3: not looking for a neurologic etiology for falls

When patients have a history of falls and an otherwise thorough workup, Dr. Josephson said that neurologists consider three possible neurologic explanations.

The first is Parkinson’s disease, and examining a patient with Parkinson’s usually reveals tremor, rigidity and postural instability. Or “we think of neuropathy,” he said, which is very common in older patients as long nerves start to deteriorate. “You obviously see numbness and decreased ankle reflexes and this too can lead to falls.”

The third diagnosis neurologists consider is cervical myelopathy, a degenerative disease resulting from spinal cord compression in the neck. “With this,” Dr. Josephson said, “we might see spasticity or Lhermitte’s sign, and patients have increased ankle reflexes and an upper motor neuron pattern of weakness.”

To test for Lhermitte’s sign, “ask patients to touch their chin to their chest. If they feel a tingling or electrical sensation in the back of the neck that radiates down the back, that’s very sensitive for spine disease in the cervical area.”

All three diagnoses “are very treatable,” depending on their cause, Dr. Josephson noted. “Don’t dismiss falls as just something that happens to old people.”

Mistake No. 4: not getting an MRI in possible posterior fossa stroke/lesion

With patients having a neurological emergency present with acute vertigo, “95% have an inner ear problem due to peripheral vertigo,” said Dr. Josephson. “But 5% are going to have something really scary, and what we worry about is a brainstem or cerebellar lesion such as an ischemic stroke.”

Patients having a cerebellar stroke often start with mild symptoms. But within three to five days, the cerebellum starts to swell. “The posterior fossa is surrounded by bone and is very unforgiving,” he noted. “If it swells, it pushes into the brainstem and causes coma or sudden death.”

The posterior fossa is also very poorly visualized on a CT. “Make sure these patients aren’t discharged from the ED with just a CT,” Dr. Josephson cautioned. “You have to do an MRI.”

Patients with central vertigo may have asymmetry on their cerebellar exam. Have patients extend their arms on the finger-nose-finger exam, and “if one side is worse than the other, that’s a cerebellar lesion until proven otherwise.”

And because many cranial nerves live in the brainstem, any cranial nerve abnormality—slurred speech, a ptosis, pupillary asymmetry—is likewise a central lesion until proven otherwise.

Mistake No. 5: scanning the wrong level of the spine

When patients with neurological emergencies in the hospital have lower extremity weakness or weakness in all four extremities, take a pin and go up and down their chest, asking if they can feel the sensation.

“If you find a sensory level on the trunk, it has to be a spinal cord problem,” Dr. Josephson said. One mistake he often sees is clinicians failing to test for a sensory level.

Another mistake: not consulting a dermatome chart. “If the sensory level you find is at T10, that’s where the lesion likely is,” he said. The sensory level indicates which area of the spine to scan—but too often, clinicians scan only the lumbar spine, not the thoracic spine.

“That just doesn’t make any sense,” Dr. Josephson said. “If you’re getting a sensory level in the thoracic dermatome, that’s what you need to scan.”

Mistake No. 6: treating emergent ICP issues like cardiovascular collapse

You are called to the bedside of a 64-year-old with glioblastoma admitted for treatment of recurrent disease. On exam, the patient has his baseline hemiparesis, “but now he’s got a large, dilated, unreactive right pupil and is altered. This is someone who is herniating.”

What should you do? When treating neurological emergencies in the hospital, at least most of the time, “you lay patients down who are acutely ill,” Dr. Josephson pointed out. “But in neurology, when people are sick, we sit them up to at least 30 degrees. We want to get blood away from the brain to reduce their ICP [intracranial pressure]. Lying someone completely flat is the absolute wrong thing to do.”

Keep this equation in mind: Cerebral perfusion pressure equals the mean arterial pressure minus the intracranial pressure (CPP=MAP-ICP). “All that matters is the CPP,” he explained. “You can decrease your CPP by increasing your ICP or decreasing your MAP. When someone is herniating, you want to let the blood pressure be high.”

As for therapy, hyperventilate patients if they are intubated. But treatment is really osmotic agents, either mannitol or sometimes hypertonic sodium, which quickly removes brain water.

“You’ve got time to wait for the surgeon who’s going to put in an extraventricular drain or some other sort of intervention,” said Dr. Josephson. “But don’t treat emergent ICP issues like you treat cardiovascular collapse. Those are totally different diseases.”

Mistake No. 7: not starting the right empiric treatment for meningitis

Dr. Josephson described another patient having a neurological emergency in the hospital: a 75-year-old man brought in with altered mental status and a first-time seizure. His general exam is normal, but he’s disoriented on the neuro exam and he is threatening staff. His labs are normal as is his head CT, chest X-ray and urinalysis.

“This is somebody who needs a lumbar puncture,” Dr. Josephson said.

But even before you do the lumbar puncture, “start patients on empiric treatment if you think they have meningitis,” he added. “Most people remember that we give vancomycin and ceftriaxone: vanc for resistant strep pneumo and ceftriaxone for almost everything else.” Dose ceftriaxone at 2 grams Q12, which “is not the pneumonia dose of ceftriaxone. It’s four times that amount.”

But here’s what doctors often miss: If someone is immunosuppressed, older than age 60 or pregnant, “you also need to give ampicillin empirically” (Bactrim if patients have a penicillin allergy). Why? “Because vanco-ceftriaxone doesn’t cover Listeria,” he explained. “That is the No. 2 or No. 3 cause of meningitis in older patients.”

Also, “it is now standard of care to administer dexamethasone at the same time or before the antibiotics,” Dr. Josephson said. “That’s because steroids reduce inflammation and—at least in cases of strep pneumo and maybe Neisseria—reduce morbidity and mortality.”

Mistake No. 8: overusing steroids in neurological disorders

Patients with multiple sclerosis often present with a neurological emergency in the hospital with worsening weakness, reporting that it’s their typical flare—and asking to be admitted for IV steroids.

The first question you need to ask instead, said Dr. Josephson: Is this really an MS flare?

“This can be difficult to sort out,” he said. Patients with MS often have “pseudo flares” where their weakness does become worse—but it’s because they have a UTI, another systemic condition or a virus. It is “a tough diagnosis to make, particularly if their weakness is similar to a weakness they’ve had in the past.”

To sort it out, do a pseudo-flare workup. The gold standard is an MRI scan with gadolinium; enhancing lesions do indicate “active inflammation and an acute flare in the right clinical scenario.”

When treating flares, good evidence suggests that steroids don’t really change patient outcomes over time. “But they do speed recovery a bit,” Dr. Josephson added. As a result, “we reserve steroids for only those patients with severe disability, such as they can’t walk or they have a urinary retention problem or a severe visual symptom.”

Patients who can no longer walk do need to be admitted for IV steroids. But for those still ambulatory, “great data suggest that high-dose PO steroids are just as good and safe.” At UCSF, “this has saved a lot of admissions.”

Here’s the problem with just automatically administering steroids: “When we give steroids, inflammatory diseases can melt away in the neuro world, but then they come back, sometimes more severe.” Lymphoma is one example. “When it comes back, you’re many months behind in terms of appropriate treatment.”

The bottom line: “I almost never give empiric steroids until I have a diagnosis and understand what we’re treating,” Dr. Josephson noted.

Mistake No. 9: using short-acting opiates for headache syndromes like migraine

Patients presenting with a neurological emergency in the hospital who have a history of migraine aura often come in with severe headache, saying they have taken Percocet two to three times a day for the last week and now they’d really like some Dilaudid.

The real problem, Dr. Josephson pointed out: “Migraine is not treated effectively with opiates, and opiates make head pain worse due to rebound phenomena.” Instead, “we have many tools at our disposal for both abortive and preventive therapy,” including non-opiate “migraine cocktails” of Compazine and Benadryl.

For migraine patients taking opiates chronically, “good data support weaning opiates slowly,” he said. “Don’t do it cold turkey. Start a migraine preventive agent as you wean them off. “Sometimes a simple burst of steroids at the same time you’re weaning off opiates can be very effective.”

In the meantime, “I do tell patients that they’re going to have a rough couple of weeks as they get off opiates but we try to help them with other medications to get them through it.”

Mistake No. 10: assuming all jerks are due to seizure

A young man is brought to the hospital after losing consciousness. According to his family, his extremities were jerking when he lost consciousness, and the ED wants you to admit him for status epilepticus.

Instead, said Dr. Josephson, “myoclonic jerks are really common.” In the hospital, those jerks are most often associated with hepatic disease or patients being treated with opiates. “But myoclonic jerks happen when the brain has any type of hypoperfusion—so if you syncopize, you will have jerks.”

Don’t assume that someone had a seizure just because there was some jerking, he said. “Jerking happens all the time.” One way to tell a seizure from syncope: People having seizures are usually blue or ruddy, not pale.

Phyllis Maguire has been Executive Editor of Today’s Hospitalist since 2006. Based in Bucks County, Pa., her health care interests are hospital medicine and long-term care options. She also likes zydeco, hiking, and reading memoirs and romances.

Lumbar punctures and DVT prophylaxis

You suspect a patient may have meningitis and you’re ordering a lumbar puncture. Is that OK if the patient came into the hospital on antiplatelets or DVT prophylaxis?

According to S. Andrew Josephson, MD, chair of the neurology department at the University of California, San Francisco, antiplatelets shouldn’t be a problem in this scenario. Some controversy exists around performing a lumbar puncture on patients taking two antiplatelets, like aspirin and Plavix.

Still, “when we suspect meningitis, you have to carefully weigh the risks and benefits,” he said. “Meningitis is just too important to not miss.”

One trick that Dr. Josephson and his neurology colleagues do in their hospital in terms of DVT prophylaxis: They write for patients to receive Lovenox at bedtime, not during the day.

“Then the next day,” he pointed out, “you can do an LP nearly anytime. There probably is some concern in the first hours after taking DVT prophylaxis. But by the next day, you’re typically fine.”