During our morning session on multiple sclerosis, a stroke alert is called. The clerkship director and I walk over to the ED. A 66-year-old female is presenting for left-sided weakness (arm more than leg), but there is no facial droop. Her husband explains she was normal when they went to bed, but when they woke up at 7:30 am, she “just wasn’t right”. She has already gotten the imaging workup, but there is nothing to be done because she is well outside the 3-4.5-hour window for TPA (tissue plasminogen activator) and this is not a large infarct. (Even if we did know the time from initial event, she has been on oral anticoagulants for atrial fibrillation. These are difficult to reverse and a contraindication to TPA.) We put in admission orders to optimize her recovery, including blood pressure and sugar control. We also order an MRI to diagnose and prognosticate the extent of the infarct. The attending cancels some of the orders placed by the ED and the psychiatry PGY2 resident. “This is the tyranny of the order set [default groups within Epic, intended to save time and typing]! Why order a carotid duplex when we just got a better picture of it with the CTA already performed? We are just wasting hospital resources and Medicare dollars.”
[Editor: The hospital CFO may not consider it a “waste” when Medicare dollars are transferred to the hospital…]
We arrive at the Situation Room, a narrow office crammed with two computers and an old couch. The clerkship director, resident, and I hang out here until the next stroke alert. I am pimped on the types of strokes. I fail miserably, citing only two of the common sites of brain bleeds. There are two main types of strokes: intracerebral hemorrhage (ICH, brain bleed, rarely lethal) and ischemic (occlusion of an artery, potentially fatal due to increased intracranial pressure). This neurologist gave a great lecture on strokes during second year, so I pull up the slides on Blackboard and then UpToDate each topic for more information. A common cause of ICH is hypertension due to cocaine and meth use among the young and poorly controlled chronic conditions among the elderly. ICH can also be caused by anticoagulants and Alzheimer’s (amyloid angiopathy). “You can quickly figure out what is the cause by the location of the bleed. Hypertension is a deep brain bleed, in the basal ganglia, thalamus, pons, or cerebellum. Dementia patients bleed into the cortex.”
He asks me, “What kind of workup would you do for the patient we just saw in the ED?” I answer, “Well, she is out of the window for TPA, and not a candidate for endovascular therapies [clot in proximal artery].” I recommend ordering an echocardiogram, carotid duplex, and EKG. “Right, we need to rule out the preventable causes of ischemic strokes” These include cardioembolism (a result of, for example, atrial fibrillation, an infected heart valve from iv drug use, or a ventricular thrombus after a heart attack), carotid stenosis, and a patent foramen ovale or hole in the heart, that can allow a clot to pass from the venous circulation into systemic circulation). We check Epic and see that the MRI images are available, though without a radiologist’s read yet. He points out a small infarct in her posterior limb of the internal capsule. Nothing to do.
(I followed up with her over the next several days and her condition was unchanged. She’ll have a permanent limp and some arm weakness, but can live independently.)
We get a stroke alert for a 76-year-old diabetic female who had a breast cancer lumpectomy one year ago. Her husband reports returning from grocery shopping to find that she was slurring words and unable to walk. He promptly called 911 so we’re probably seeing her about two hours after the onset. Her blood pressure is 215/100, too high for TPA, so she’s on a nicardipine drip in hopes of bringing it down. The neurologist calmly examines her with standard techniques (“follow my hand with your gaze”) and some of his own design (“close your eyes and tell me what you feel” as he hands her objects such as a key or lighter). She has a left facial droop, dysarthria (speech disorder due to muscle weakness), right gaze preference, and a left hemianopsia (blindness). Like most of our stroke admits, she gets a CT perfusion scan (five minutes and reimbursed at $12,000 by Medicare) to see if she is a candidate for endovascular intervention, i.e., clearing out a plumbing clog with a drain snake. Her scan is among the 10 percent that suggest endovascular intervention: proximal (closer to the heart) clot surrounded by potentially viable tissue. Her clog is in the middle cerebral artery (MCA, the main artery of the brain).
She is carted off to the endovascular suite. I call Straight-Shooter Sally, who did not get to see an endovascular procedure on her week of stroke service. We meet up in the Interventional Radiology suite; endovascular procedures are split between interventional radiology and interventional neurology. We’re both excited, but the neurologist doesn’t say anything during the 45-minute procedure. “Well that was useless,” says Sally. We follow up with the patient the next day and she has almost no symptoms, except mild weakness in her right wrist.
(It seems obvious that cleaning out the pipes would work, but there are no good clinical trials to support the anecdotal evidence. A lot of patients who get endovascular therapy would likely have recovered on their own.)
During the 4:00 pm debrief in the “Situation Room”, I ask if all stroke patients should get a $12,000 CT perfusion scan. “It depends whom you ask,” responds my attending. “The people who designed our current protocols say, ‘Yes.’ But they mostly are not neurologists. Medicare doesn’t understand the purpose of the CT perfusion scan. Two out of three scans that they pay for are unnecessary in my opinion. Only a small percentage of strokes are amenable to endovascular therapy. And we are not an institution at the cutting edge doing research on other indications. There is no excuse except laziness and dipping into a free pot of gold.” I ask about the VAN score to screen for patients for a large proximal clot. If a patient does not have focal weakness and one of the following: Visual disturbance, Aphasia, or hemi-Neglect, it is extremely unlikely to be a large proximal clot amenable to endovascular therapy. My attending doesn’t disagree with the VAN system, but thinks it adds little to an experienced neurologist’s judgment. “Stroke centers are graded by the door-to-needle time [time to get a stroke patient administered TPA]. The ED is so focused on taking the thought out of medicine with protocols.” He noted that every stroke patient now goes through the same steps: (1) non-contrast CT brain to rule out brain bleed, (2) CT angiogram to look for a clot, and (3) CT perfusion scan to evaluate salvageable brain tissue. “Though lucrative, most of this is unnecessary and doesn’t change management. CMS hasn’t investigated us yet, but I hope they do.”
In his opinion, what would help more patients at a tiny fraction of the cost is simply speeding up radiology. “During nights and weekends we don’t have in-house radiologists. We use teleradiologists who are contracted to get back to us within 30 minutes. We need a 5-minute look at brain anatomy, but they take the full 30 minutes to give us a detailed report so that they can’t be sued for missing something. We get a report on spine, teeth, lungs, etc. The ED can’t read images, so the stroke patient is sitting there for 30 minutes without any therapy. A good neurologist reads his or her own films and a brave one will make the call without a radiologist.”
[Editor: Smaller hospitals are unable to do either the CT perfusion scans or the endovascular intervention (“thrombectomy”), so our near-octogenarian Presidential candidates might not want to spend too much romancing voters in small towns. See “A Breakthrough Stroke Treatment Can Save Lives—If It’s Available” (WSJ, February 6, 2018).]
Statistics for the week… Study: 7 hours. Sleep: 6 hours/night; Fun: 1 night. Burger and beers with live music. Mischievous Mary has already started looking for visiting away electives in cardiothoracic surgery.
The rest of the book: http://fifthchance.com/MedicalSchool2020