Functional stroke / What would you do ?

[Nephro critical care]

I recently had this patient while working as a night shift hospitalist . On psych ward young patient has sudden onset of unilateral arm and leg weakness. Psych calls rapid response and I came. Patient does a a pretty good job of mimicking hemiplegia. She is 2 1/2 hours into symptoms . I get stat head CT and of course no abnormalities. She also conveniently mentions she vomited blood 3 days back . Of course no records of her vomiting blood . I called neurology who doesn't want to get out of bed ( it's midnight ) and tells me to give TPA if I clinically think this is an ischemic stroke. Now I am thinking my gut sense is that this is functional not a stroke . But if I am wrong I am gonna be screwed . An MRI will take about 2 hours to get and she will be out of the TPA window.

In the end I ended up not giving TPA and wrote I didn't think this was a stroke. MRI next day showed no stroke and patient was magically better on her own.

My experience is that this happens often . I have seen some patients come in multiple times to ED feigning hemiplegia end up getting TPA multiple times . One of them even started making a big show of feigning improvement of symptoms as soon as TPA bolus was given and making ED staff feel like heroes. She would then be admitted get an MRI and nothing would show up . I don't understand the secondary gain she was getting from getting TPA again and again .

---

Members don't see this ad.

 

[Ivos]

I've seen this a couple of times as well. Curious about how you guys handle it. Practicing in a less litigious country, we can stop tpa on clinical suspicion, but I still have that lingering doubt until a follow up scan is done.

 

[Hershalmike555]

Yah, I am also curious as well how to handle this situation. Will wait for more updates.

 

[TUGM]

Given the patient's self-reported history of GI bleeding/hemoptysis, I would not have given rt-PA regardless (this should be documented explicitly). Look for facial involvement (it's hard to fake a facial droop), although vascular territory may not always hit the face fibers. Sometimes when I suspected a functional exam, I apply sudden, noxious stimulus to gauge reaction and to see whether an arm or leg is truly plegic.

In true stroke patients, rt-PA has a 6-7% chance of causing a symptomatic intracranial hemorrhage. In non-stroke patients, this risk drops to about 2-3%. Not saying that this means t-PA away, but maybe you can use that data to help calm your anxiety until a follow up scan is performed.

 

[Hellothere111]

Usually, the threshold for offering IV TPA for posterior circulation strokes (even with NIHSS of 0) is lower. If the deficit is only sensory, we don't offer it. If it's motor, then it's a judgment call but invariably, the decision is to offer it as long as the patient understands the risks. Most non-physiological patients decline. There are those rare ones who accept but in those cases, often the argument for offering it is legal.


Sent from my iPhone using SDN mobile app

 

[neurodoc]

This is a very tough case and patients do present with these kinds of confounding behavioral/psych issues. Unfortunately the stakes here are a lot higher than when we have to deal with treating possible pseudoseizures. We could leave it to our psychiatric colleagues to sort out what kind of conscious or unconscious behavioral disorder the patient may be manifesting, but we have to do our best to decide whether or not the patient had a stroke, and there isn't that much time to decide.

You could rely on your neurological history and exam, which is what we're supposed to do. Sometimes the physical exam will be very reliable (e.g. facial and occulomotor abnormalities) but all stroke patients don't present such findings. And we all know that folks with epilepsy and migraine can present with apparent strokes, especially if they come to us without any reliable history.

This is exactly the kind of case for which having DWI/ADC MRI would be ideal. Let's say you have a patient like this who strikes you as presenting factitious stroke symptoms, though you have some lingering doubt. She might even have been followed by psychiatry and have a diagnosis of Munchausen Syndrome...but you know what...this time she does have a stroke! If you had access to DWI MRI you could actually rule in stroke and treat if otherwise indicated.

Munchausen Syndrome may not be a good example, because these and other highly manipulative patients are sure to catch on that they're going to get the MRI to r/o stroke anyway and they'll just keep coming in and wasting that resource. Hopefully our psychiatric colleagues will step up to the plate and cure them of their pathological behavior... Or maybe the patient will get treated by some other doctors who just have CT and a low threshold for thrombolyzing anything (including Bell Palsy and brachial plexopathy) that looks like it could be a stroke...and suffer a catastrophic hemorrhage...

 

[kchan99]

This is a very tough case and patients do present with these kinds of confounding behavioral/psych issues. Unfortunately the stakes here are a lot higher than when we have to deal with treating possible pseudoseizures. We could leave it to our psychiatric colleagues to sort out what kind of conscious or unconscious behavioral disorder the patient may be manifesting, but we have to do our best to decide whether or not the patient had a stroke, and there isn't that much time to decide.

Many patients with functional neurological disorders have a history of childhood maltreatment or other significant trauma, and may benefit from CBT directed to functional disorders, but some patients don't have current or prior stressors apparent to neurologists. Fortunately, DSM-V doesn't require stressors for diagnosis of conversion disorer/functional neurological symptom disorder. Unfortunately, it seems that there isn't too much interest from our psychiatric colleagues.

 

[Thama]

We have hyperacute MRI for these kinds of situations, we just run a DWI/ADC. FLAIR, SWI and COW MRA, takes about 15 minutes. Only used when the exam leads us to think stroke is unlikely but can't rule it out on other grounds, but would still be a tPA candidate if it were a stroke. I've given tPA in the scanner twice for situations like this, but it's usually negative and allows us to feel pretty good about withholding tPA from functional patients.

There's a school of thought that the vast majority of tPA complications occur in people with real stroke probably due to some degree of early hemorrhagic conversion, and so tPA in functional and otherwise healthy patients is actually pretty unlikely to be harmful. I like having imaging confirmation though.

 

[neglect]

You did a great job! You had no help from neurology on a very difficult case. That's a matter for your chief of staff. As mentioned above, gi bleed a few days ago (you gonna believe psych notes about a psych patient who may not have reported this?) excludes Tpa.

But I've resigned myself to giving it in conversion disorder. So far I've given it to three people I know were psych. One was excellent, really top performance, even had a soft field cut. I was still under the belief he had an MRI net stroke until his function started to differ from the exam dramatically (hard act to maintain). I e given it two other times to people I considered to have psych problems. And held it once that I recall in a psychogenic language problem - and one of the reasons is that she opted not to get it. So I documented just that, and I had her sign off on it, and diagnosed conversion disorder.

The bottom line on this is that you are unlikely to do harm with a neg ct without other Tpa contraindications. We know that you don't have time to eloquently discuss the patients life, socioeconomic status, and relevant psychology. You have to make a decision in minutes. So give it. I've seen plenty of weirdo strokes in the young, creating odd symtoms and odd elaborations. I gave it to a kid with a visual field cut who was freaked out: dense pca stroke. When in doubt and no contraindication: err on the side of giving it.

 

[typhoonegator]

Agree with neglect. If there isn't a strong indication to not give it, and you think there is a reasonable chance this is ischemic, then you should treat. Document your exam and decision making process, including the weighing of risks and benefits, and the patient's assent to treatment. All of us who regularly take acute stroke calls at big institutions will end up giving tPA to people who turn out to be obviously functional, and many more who never turn up an MRI abnormality who you're not really sure what was going on.

With telestroke, this is even more challenging because you're trying to figure all of this out over a video screen with a proxy examiner who is almost never a neurologist. Telemedicine is a real skill that takes time to develop and use properly.

All of that being said, the patient endorsed hemoptysis in the days prior. Even if you thought the symptoms were clearly true ischemia, it would be a difficult call to treat her. It certainly would be a very justifiable reason to exclude the tPA treatment option. In this case, it is convenient, because you can rely on that rather than your judgement that the symptoms were "probably not ischemic". Given that patients can have flaky "symptoms" that they elaborate on top of very real deficits, it can be very difficult to truly exclude any chance of ischemia.

 

[RenegadeSynapse]

When I can examine a patient and feel confident that the exam is functional, I've not offered tPA. Probably a hand full of times in residency (PGY4 atm). These were all very clear cases (i.e. staggeringly obvious Hoover's without confounding factors).

I've also given tPA to a patient the ED thought was functional. Had a Hoover's sign / giveaway weakness on the left and was "acting weird" but also had hemispatial neglect, lateralized agraphesthesia. CTA after the tPA showed a distal inferior MCA branch occlusion on the right.

I agree with Typhoon; if you take stroke call, you're gonna give these folks tPA a reasonable percentage of the time. I can't even begin to imagine sorting some of the flakier cases over telestroke.

 

[Nephro critical care]

About this psych patient I had a gut feeling that she was faking it so I documented about the hematemesis and my feeling that this was functional. I wonder though what I would do if I had a similar patient especially young with real symptoms and self reported blood in stool 2 weeks back. I would do a rectal exam and if negative , discuss with her about risks / benefits but lean towards giving TPA. And get a CT head / neck vessels and ship to interventional neuro center if large vessel clot.

 

[neurolddoc]

I too had one occasion to consult on a patient in the psych ward with hemiplegic symptoms. I was 99.9999% sure that the whole episode was functional., CT was negative, MRI not immediately available. I had a very explicit discussion with the patient and his mother that, while I thought the deficit was not due to anatomical disease, it was impossible to know with certainty. I discussed the risks and benefits of TPA and ultimately gave it. MRI later showed no stroke and there was no complication from the therapy.

The key is to carefully lay out the decision process with documentation of the risks and benefits explained to the patient/family and thoroughly documented in the record.

 

[Neurologo]

Interesting discussion. Few additional points I may add are the following:

- For the arm motor defect, looking for pronator drift can be helpful since most patients do not know about the pronation occurring due to the inhibitory UMN signal defect. But be aware that some true stroke cases can only show vertical drift without the pronation. Thus, if you see pronation, then you can be more confident that the deficit is real.

- About 14% of tPA cases worldwide turn out to be stroke mimics, and it appears to have little to no adverse events. Thus, it is ok to give it even if it turns out to be mimics including the somatoforms.

- MRI can be negative initially only to show the infarct later when repeated few days later. This is especially true for brainstem infarcts. So if your exam localizes to the brainstem, it is important to not miss true stroke just because the initial MRI was negative. If deficits persist, repeat MRI.

- True acute stroke cases with visible occlusive thrombus can end up showing no infarction on MRI done the next day thanks to tPA and/or thrombectomy. I've had 2 such cases with proven pathology of the extracted thrombi.

Hope these help managing ambiguous cases.

 

[Nephro critical care]

Interesting discussion. Few additional points I may add are the following:

- For the arm motor defect, looking for pronator drift can be helpful since most patients do not know about the pronation occurring due to the inhibitory UMN signal defect. But be aware that some true stroke cases can only show vertical drift without the pronation. Thus, if you see pronation, then you can be more confident that the deficit is real.

- About 14% of tPA cases worldwide turn out to be stroke mimics, and it appears to have little to no adverse events. Thus, it is ok to give it even if it turns out to be mimics including the somatoforms.

- MRI can be negative initially only to show the infarct later when repeated few days later. This is especially true for brainstem infarcts. So if your exam localizes to the brainstem, it is important to not miss true stroke just because the initial MRI was negative. If deficits persist, repeat MRI.

- True acute stroke cases with visible occlusive thrombus can end up showing no infarction on MRI done the next day thanks to tPA and/or thrombectomy. I've had 2 such cases with proven pathology of the extracted thrombi.

Hope these help managing ambiguous cases.

I appreciate all your points. Now thinking back I think I was a bit cavalier although it was a psych pt and turned out not to be a stroke. If a even a pt had presented to me with an ambigious exam and I wasn't sure I would err on side of giving TPA. The only reason to not give TPA would be out of window or blatant exclusion criteria. I pity our e- neuro ICU who have to make a call just looking in by camera.