Conjugate eye deviation in acute stroke

While poorly understood, conjugate eye deviation (CED) is really quite frequent: about 20% of stroke patients, about a third of thrombolyzed patients and roughly half of thrombectomy patients have it.

In this modern era where neurological workup of stroke patients is reduced to the barest minimum for the sake of door-to-needle and door-to-groin time, the door-to-neurologist’s-brain interval and ratio suffers just as much as the patient, inflating the role of the only diagnostic measure that is properly carried out: neuroimaging. Therefore radiological signs of severe impairment gain more attention and conjugate eye deviation is one of the easy and seemingly objective signs (while, say, a Babinski is hard to capture in an image of the brain). In contrast to the rest of the neurological exam you can also find CED on CT (CT-CED) retrospectively, so there are quite a few papers on it.

Clinical aspects

In the neurological exam one would require the following semiquantitative information:

  • Any eye deviation
  • Significant eye deviation (> 15°) – increases interrater reliability of CED
  • Severe eye deviation (> 30°) – if supratentorial, highly associated with spatial neglect
  • With head deviation – usually with severe eye deviation. Note that head-on-trunk deviation cannot be seen on CT as the patient’s head is fixed in the machine.
  • Fixed: cannot be overcome by contralateral visual stimulation – this might hint at accompanying spatial/visual neglect or hemianopia
  • Supranuclear: can be overcome by VOR/OCR – if not, obviously a brainstem stroke should be suspected
  • With nystagmus, usually in the direction of the eye deviation – this is ictal mainly. In theory it could be pontine but I have never observed the combinaton of CED + nystagmus in that case and could not find any case reports.

Since the FEF is quite near to the precentral gyrus and large strokes tend to damage the pyramidal tract as well, we expect a contralateral hemiparesis to accompany the ipsilesional eye deviation in most cases of damage to the frontal eye field.

CT versus clinician?

It is noteworthy that mild forms of conjugate eye deviation may actually only be seen with eyes closed (those of the patient, silly!) – this is much harder to examine clinically than radiologically, although it requires that the CT technician consistently ask patients to keep their eyes closed and that patients be able to follow this command.

This reduces correlation between the clinical examination (as reported in the gaze palsy item of the NIHSS) and the CT finding; e.g. in this 2017 study, a kappa of only 0,34 was achieved, while the interrater reliability of CT-CED was very good (even without a predefined threshold for the degree of eye deviation). UCSD seems to be better at fitting the clinical examination to CT results, achieving a kappa of 0,89 but they have the advantage of having a cool name for CT-CED: the “DEYECOM sign”.


Usually, CED and especially severe and/or fixed CED points to large hemispheric strokes. Still there is a well documented 2002 report of a smaller cortical stroke in the region of the frontal eye field (FEF) that caused ipsilesional CED (see below for comments about the localization of the FEF) and we have seen such ourselves.

Neglect versus CED

A long-standing controversy surrounds the relation between CED and spatial neglect. Obviously a patient that ignores, say, the left world, will have his eyes or even his head turned right. Vice versa, if you cannot turn your eyes to the left, you tend to not see things on the left (although you will still recognize that something is happening on the left – something a real neglect patient would not). So if in some studies only visual cues have been used to diagnose neglect, there should be a strong correlation between the two and the more severe the neglect the worse the eyes (and head) deviate. There are definite cases of CED without neglect and the two phenomena have a different half life, with CED usually subsiding after hours to days, while neglect can take much longer. So there is a relationship but they are not married.

Wrong-way CED

There are some case series and reports on contraversional CED in supratentorial strokes, sometimes called wrong-way deviation. This can be explained by:

  • ictal CED – stimulation of the frontal eye field (see below) causes contralesional eye deviation, often with nystagmus (so-called epileptic nystagmus which really is quite specific). If in the setting of acute stroke or intracerebral hemorrhage this requires that the FEF is not in the core but in the border region of the stroke (the outer penumbra, where cells are still viable, semi-functioning but easily excitable), for instance in M2 occlusions of the temporoparietal trunc,
  • bilateral stroke – not necessarily recognized in the acute images; can hint at proximal embolic source, vasculitis and much more,
  • herniation: the well-known falsely localizing sign of ipsilesional hemiparesis in space-occupying lesians such as malignant MCA stroke or large ICH with transtentorial herniation – compression of the contralateral crus cerebri against the cerebellum through midline shift of the mesencephalon (producing Kernohan’s notch),
  • hemorrhage with contralateral lesion,
  • parietal stroke: some authors mention that spatial neglect from right parietal stroke can lead to contraversive eye deviation. I have to admit that I don’t understand it, but there you have it.

Where is the FEF is located?

This question has bothered generations of researchers, starting with neurosurgical observations of patients in ancient times, somewhat later primate studies, then human semi-invasive studies (including direct cortical stimulation) and more recently PET, SPECT and fMRI research. The conclusion of this detailled review is that the precise localization depends on the setting of the experiment (which stimulus is used? what is required from the patient to do?), but it should be somewhere between the dorsal third of the midfrontal gyrus, the deeper parts of the precentral sulcus, and maybe the frontal parts of the precentral gyrus around that region.

Our best guess for the frontal eye field (blue area): in the dorsal third of the middle frontal gyrus, down into the precentral sulcus (green), perhaps to the anterior part of the precentral gyrus. Note the left central sulcus in yellow, the left superior frontal sulcus in pink and the Omega for the hand knob on the right side (we did not put it on the left as not to further confuse the reader).

Finding it on CT/MRI

Remember that one of the various (and not always successful) ways to localize the central sulcus is by walking along the easily recognized superior frontal sulcus from front to back, parallel to the falx, until you hit the precentral gyrus.

Laterally you should find the hand knob (with the Omega sign), more medially and inferiorly the paracentral lobule which closes the ring formed around the central sulcus: precentral gyrus -> paracentral lobule -> postcentral gyrus -> subcentral gyrus and back again. The frontal eye field should be slightly lateral to the corner of the L formed by the superior frontal sulcus and the precentral sulcus (see Image).

Since the FEF is near the precentral gyrus and large strokes tend to injure the pyramidal tract as well, we expect a contralateral hemiparesis to accompany the ipsilesional eye deviation in most cases of damage to the frontal eye field.

Subcortical causes

The most common pattern of caudate stroke today: s/p thrombectomy M1 right – note the hypodense demarcation in the striatum as well.

Besides the cortical structures the caudate nucleus, thalamus und the subinsular basal ganglia can cause neglect and thereby conjugate eye deviation, thus contradicting the dogma that CED is a strictly cortical sign. But remember that this holds true not only for neglect and CED but also for aphasia. While most of the acute cases with these so-called “cortical deficits” that end up with only a subcortical stroke do have cortical hypoperfusion in the acute phase  there are definite (if rare) cases of isolated strokes in either of the three regions with either of the three symptoms. Perhaps we should call them the gray (substance) deficits.

Significance of cortical signs

Why bother? Because cortical signs accompanying a hemiparesis strongly suggest a large vessel occlusion, thus necessitating CT angiography.

As an aside, many strokeologists count visual field disturbances as cortical signs. I hesitate to do so, as most quadrantanopias are probably caused by MCA strokes in the temporal or parietal subcortical visual radiation and then there is also the case of the anterior choroidal artery strokes causing hemianopia.

Differential for acute stroke

If taken as a sign of cortical dysfunction the differential for CED in the acute setting of presumed stroke is narrow: stroke/ICH, postictal (or ictal if ipsilateral to the paresis), dysglycemic. The other classic differentials for acute CNS focal deficit (such as migraine, MELAS, …) have not been reported to cause clinically consistent CED. Remark that if only the radiological sign of CED is used then a lot of false positives (50% in this series) will arise, because patients can and will look around at times during CT.


Ocular eye movements are mediated through various brainstem pathways, nuclei and cranial nerves. Lesions of these structures lead to nystagmus (gaze evoked, upbeat or downbeat), diplopia, internuclear ophthalmoplegia etc. CED is among the more rare signs of brainstem strokes:

  • Sometimes pontine lesions to the medial longitudinal fascicle (near median pontine, as in this case report) lead to contraversion of the eyes rather than the more frequent sign of internuclear ophthalmoplegia.
  • There are also some reports of medullary strokes (both dorsolateral – aka Wallenberg – and dorsomedial) with CED, in which case malfunction of the olivocerebellar pathways, the nucleus prepositus hypoglossi (taking care of gaze holding) or the MLF have been implicated, with ipsilesional eye deviation.
  • A non-systematic study of cerebellar strokes found subclinical (only radiographic) CED in about a third of patients, mostly contraversional.
  • In my personal experience, bilateral lesions to the pyramidal tract in the medulla have led to protraced (> 1 year) conjugate eye and head deviation.

Take home messages

  • Note eye and gaze position clinically in all your acute stroke patients – it is a sign of badness.
  • Note eye and gaze position on CT and try to verify it clinically.
  • In gaze deviation think 1) cortical 2) brainstem 3) subcortical 4) contralateral.

Did I hear dizziness?

Dizzying foto from Flickr, Creative Commons

We recently suffered through an M&M meeting about a patient that presented with vertigo and hearing loss. While clinically a typical vestibular neuritis, the importance of sudden hearing loss was grossly understated, with arguments like “this is all the more a peripheral disease” and “belongs to EENT then, anyway”. I certainly grew up with a standard operating procedure that allowed to turf “vertigo + hearing loss” without further workup, unless there were signs or symptoms of CNS involvement.

In this blog entry, I try to work through the numbers, invoking some recent publications.

Unilateral sudden hearing loss (USHL)

(bilateral hearing loss is a different animal altogether)

  • more than 30 dB hearing reduction in < 3d (in at least three frequencies)
  • quite frequent ~ 27/100.000
  • usually middle aged
  • about a third have vertigo, depending on how strictly you define it


As in so many other cranial nerve dysfunctions we don’t really know the etiology of your average run-of-the-mill USHL. It usually seems to be benign, because otherwise people would be dying all over the place, so depending on your prejudices you can suspect anything from viral to vascular.

It is really quite surprising that these etiologically occult cases have been ascribed by some to vascular causes (without any proof except for risk factors) as if this were benign in any way, similar to cases of mononeuropathic cranial nerve dysfunctions such as trochlear, abducens or facial palsy where the “microvascular” theory is built on the epidemiologic association with risk factors such as hypertension and diabetes. If those entities were “microvascular”, we ought to treat them it with tPA, GP-IIb/IIIa inhibitors (analogously to microvascular stroke #lacune), or at least include them into a risk factor reduction program and ASS them.

The viral theory (which seems to be accepted for the case of bell’s palsy) is also difficult to establish – at least antivirals don’t seem to work.

There are plenty of cool differential diagnoses, such as Cogan’s or Syphilis (see this article for a table), but of course you cannot exclude all the zebras in the average case.

Differential diagnosis

So what is in the differential that is acutely endangering the patient and, in particular, what changes if vertigo is added to the symptoms?

  • Vertebrobasilar stroke – see below.
  • Acoustic neuroma: this needs to be recognized, but not acutely.
  • Migraine – not dangerous, but eminently treatable.
  • Blood problems – sickle cell, leukemia, … – it seems that an ESR and a CBC are enough.
  • Real infections: syphilis mainly, maybe HIV?

Then there are things that usually only EENT specialists care for, such as Meniere’s and perilymphatic fistulas, cholesteatomas and many more. All these seem to me to be not that urgent and therefore safe to send away, if symptoms have been covered.

The literature, being written by neurootologists (a quite intellectual bunch), stresses the differentiation between central and peripheral causes, with multiple sclerosis, migraine and stroke among the most frequent central causes. As an ED physician I would rather stress stroke as the main cause to be identified acutely, deferring the decision when to MRI vertigo patients to the clinic.

Acute vestibular syndrome plus hearing loss

About 3-15% of vertigo patients turn out to have a vascular problem, either a stroke or a TIA due to serious stenosis. Now in the past, neurologic myth had it that hearing symptoms (SHL or tinnitus) reduce this risk, so that it is safe to turf the patient to EENT. It turns out that the opposite is true – nearly double the amount: 5,5%1, at least in this recent population based survey, which should all but overestimate the risk.

I will try to give an explanation for this fact in a second.

First let us consider the practical implications. We have learned to filter acute vertigo patients (it is superbly unclear what acute means but for the sake of simplicity I assume this means < 3 weeks), using a good history and physical/neuro exam, including (but not limited to) the head thrust maneuver (aka head impulse test). Some would want to simplify this to HINTS or INFARCT, which amounts to reducing the rest of the neuro exam to nystagmus (direction changing?) and cover test (vertical disconjugation?), which in my humble opinion is too brief in the hands of a Neurologist. But even if you limit yourself to HINTS you miss at most 1 percent of strokes and thus improve on MRI (which misses up to 20-30% of strokes in the acute phase).

Why is that? The only way that a stroke mimics vestibular neuritis with a positive Halmagyi sign (Halmagyi-positive pseudoneuritis), is that it hits the nerve root entry zone around the fourth ventricle. And this region is highly collateralized as it can derive it’s blood supply from small perforators, AICA, PICA and even dural and petrosal arteries.

On the other hand, a true vestibular neuritis that causes permanent vertigo usually will cause the horizontal vestibuloocular reflex (hVOR) to suffer, except for the rare case of inferior vestibular neuritis (as opposed to your standard superior neuritis which affects nerve fibres to the horizontal canal) – this occurs in only 9 of 703 cases of vestibular neuritis overall according to this retrospective case series. So most of the patients should have impaired hVOR, yet not all of them are discovered during the clinical head impulse test, because covert saccades (fixation-correcting saccades during the thrust rather than afterwards) are impossible to see without videooculography and sometimes the hVOR impairment is not serious enough to cause correcting saccades at all. We don’t really know the percentage of covert saccades in real patients, but in dizziness clinics it can be up to 30%. Personally, I see way more covert correcting saccades with the slow motion app of my iphone, which has not been properly studied but definitely improves the clinical HIT as it makes it objective. One problem with all the investigations into sensitivities of neurootological testing vs. HIT is the lack of a gold standard – there is no other biomarker for vestibular neuritis than VOR (plus other sophisiticated vestibular lab tests) and this is defined by gain reduction (i.e. eye movement divided by ear movement) to 0,7 or less in the vestibular lab in order to preserve test quality indicators, so we might miss some cases with only slight VOR reduction.

Pseudolabyrinthitis: Acute vestibular syndrome with hearing loss due to ischemia

This corresponds to a stroke in the territory of the labyrinth artery, which usually (98%) derives from the AICA or the basilar artery (in the remaining cases it branches off the PICA, but this is exceedingly rare – see this 2015 case report) and does not have collaterals, so that it serves as a Letzte Wiese of the vertebrobasilar system. For instance, a high grade basilar artery stenosis (or equivalently a vertebral artery stenosis in the frequent case of hypoplastic contralateral supply) might cause temporary hearing loss and vertigo with nystagmus. Although the whole labyrinth is expected to suffer, the resulting nystagmus at least in part will be horizontal as in a status post labyrinthectomy.

Now we don’t know the percentage of AVS+HL patients that have a real vascular cause but the above mentioned case series suggests taking it even more seriously – this has lead to the so-called HINTS+ scheme, adding hearing loss to your AVS workup as harbinger for possible vascular (if not necessarily central) cause, without reducing specificity of your workup.

The odds

  • Between 5-25% of AVS pt. have a vascular cause
  • About 15-20% of vascular cause AVS are AICA in origin
  • 68% of AICA strokes have auditory symptoms
  • 50% of AICA strokes have a “falsely peripheral” HIT

Transient audiovestibular symptoms

What about the patients that had a serious vertigo spell but recover before you see them in the ER? Some have dubbed this entity ATVS – acute transient vestibular syndrome. It hasn’t been studied much yet, but from what has been said above we should conclude that it merits a proper TIA workup (even if vertigo itself does not count as a TIA symptom in most guidelines), including (at least clinical) vestibular testing – see below – and an MRI, possibly with MR-perfusion if you read this recent Stroke 2017 paper.

My vertigo workup

  1. Do a proper history, asking explicitly about auditory symptoms and craniocervical pain. Take your time for that.
  2. Do a proper neuro exam.
  3. Do a proper neurootological base exam
    1. Frenzel screen for spontaneous, gaze-evoked and head shake nystagmus
    2. Extraocular eye movement tests including vergence, saccades, smooth pursuit
    3. VOR cancellation test
    4. tragus pressure test
    5. Romberg’s and Unterberger’s
  4. Do some simple vestibular testing using devices everyone should have in their ER
    1. HIT (using iphone’s slow motion camera)
    2. hearing test (using an iphone app such as MiMi)
    3. bucket test or subjective vertical iphone app
    4. dynamic visual acuity (have them read something while turning their head)
    5. Dix-Hallpike, head roll test, head hanging test

Recommended reading


Neuropathic pain in Guillain-Barre-Syndrome

We use the case of a mildly affected GBS patient with severe neuropathic pain to discuss the latter’s pathophysiology in general and the treatment for neuropathic pain in particular.
There is an excellent review article in BMJ 2014 which hints at the various drugs on the horizon, including some known substances whose application in neuropathic pain seems feasible (see below).

Neuropathic pain

  • Prophylaxis: in order to reduce the occurrence of neuropathic pain there is a good rationale for prophylactic treatment in certain cases, such as amputation, Zoster and nerve surgery, although not much good data has been published.
  • Classify: distinguish NP with autonomic features from that without, central versus peripheral (the latter often with autonomous signs)
  • Diagnosis: recognize the core features of pain character with good PPV (radiation, allodynia, hyperalgesia, autonomous sign if present, distribution along affected nerve structures)
  • Basic treatment
    • Stick to the basics: Despite the fact, that we Neurologists tend to think that NP is so special, we forget that basic nonsteroidal analgesics do work in NP; there is even a good pathophysiologic reason for that, since local cytokine production is influenced by at least the antiinflammatory substances (not acetaminophen, though).
    • Opiates: next is opiates, most often as a bridge to other approaches – still highly effective and reasonably well tolerated
    • alpha 2 delta blockers (gabapentinoids) work on the dorsal root ganglion and have been shown to be effective in many prototypical diseases
    • Na blockers (carbamazepine, phenytoin, lidocaine, lamotrigine)
    • NASRIs including tri- and tetracyclics and mirtazapin, as well as venlafaxine and duloxetine
  • Local treatment: only if the pathophysiological proces is focal – capsaicin and lidocain patch
  • Advanced treatment: for more advanced NP, ketamine seems to be the agent of choice as an NMDA blocker with all the problems arising. Further ideas could be: Baclofen, Clonidin. Many anticonvulsants (apart from those above) have been tested, not many survived, except in some diseases (trigeminal neuralgia for instance -> topiramate)
  • Experimental or future therapies: Allopurinol (ADP antagonist), Aprepipant (NK1 blocker), Memantine, Amantadine (both mild NMDA blockers), Cannabinoids (good preclinical data, moderate effect), cytokine inhibitors, NGF blockers

Pain in Guillain-Barré-Syndrome

I hesitate to repeat all the information to be found in the literature. Still it has to be said that pain is very prevalent, often preceding and also often following GBS, can take several forms (backache, interscapular, distal, skin, myalgia, …) and is often severe enough to run through the above list. Here is a good review article in Neurologia on this from 2015. The best case series has been published in Neuroloy in 2010.

On Spencer’s curve

2000px-znak_a-1-svg1In our recent ultrasound refresher course, I tried to give a talk on the vagaries of stenosis graduation (mainly) for extracranial stenoses. The gist of the talk is outlined in the following notes.

The Bernoulli principle

While Bernoulli’s equation is rather intricate, the underlying principle of conservation of flow along a stenosed tube is simple. Consider a tube with a short stenosis with laminar flow of a Newton fluid. Then the bigger area A1 multiplied with the flow  velocity v1 (take psv for simplicity, although this is not really correct) should be the same as the smaller area A2 multiplied with v2. Thus the increase of flow is proportional of A1 / A2, so that the reduction to a third of the area leads to an increase by the factor 3 of the flow velocity, a reduction to a tenth leads to ten times the flow velocity in the stenosis.img_7579

Adding friction

Since we usually don’t observe velocities higher than 5 m/sec, there must be a limiting principle and this is the resistance offered by the stenosis, which reduces flow in the whole vessel. This resistance can be approximated by the law of Hagen-Poiseuille and is proportional to the length of the stenosis, the inverse of r^4 and the inverse of viscosity. Again, this only holds for laminar flow and the case of blood offers some more complications, but the core message is: the longer the stenosis the higher the flow reduction. Also the flow reduction grows much more with decreasing vessel diameter than the flow velocity increase by Bernoulli’s principle can compensate. Very tight and long stenoses show a flow velocity reduction despite there high grade. If you  find a psv of only 2,4m/sec this might therefore mean either a mere medium grade stenosis or a very tight stenosis (near occlusion).img_7578

Spencer’s curve

Taking these two principles into account, Spencer and Reid (in their brilliant 1979 stroke article) deduced the famous curve now known as Spencer’s curve (see Alexandrov’s papers for a more detailled exposition).

Since at the time duplex sonography was technically not feasible, the Spencer curve is based on the theoretical assumption of a 2 mm stenosis and thus does not correct for the length of the stenosis (as well as the other factors mentioned below). This explains why the cw-doppler-data in their paper does not really fit the theoretical model. Still it is the best approximation we have to a theoretical foundation of stenosis quantification.

Measuring diameters

Diameter instead of area

Most of the studies have been done with angiographic imaging of stenoses or ultrasound measurements of the stenosis diameter rather than the area. Of course, the area could easily be calculated from the diameter (r^2 pi) if it were a circle, but then it isn’t. In ultrasound we could measure the area itself (if no shadowing artifacts are present), yet no one does really. Therefore you should remember that most of calculations of the stenosis area from the diameter are systematically invalid.

Where, when and how to measure diameters

For some absurd reason, Europeans kept to the local stenosis degree (i.e., diameter of perfused lumen divided by the original vessel diameter), at least in their ECST trial, while the NASCET trial used the more reasonable distal stenosis degree (i.e., minimal lumen diameter divided by non-stenosed distal ICA diameter). Since the ICA bulb varies in its bulbiness and distends with age and blood pressure (and with stenosis degree), the local stenosis degree is usually a shot in the dark. The distal (NASCET) degree suffers from pseudoocclusion, i.e., collapse of the distal vessel in very high grade stenosis. All attempts to calculate the NASCET stenosis degree from ECST and vice versa are irrational.

Being faithful to both traditions, I measure all the lumina (minimal lumen, original vessel lumen, distal lumen). The only really interesting number is the residual diameter (combined with the length of the minimal lumen), because this determines the hemodynamic compromise.

Other important factors


Neither is blood a Newton fluid, nor is all the viscosity (rarely measured today) explained by the hematocrit alone. Yet the hematocrit is an important number to factor into your interpretation of ultrasound data. You should note it.


Every vessel wall abnormality leads to small perturbations of flow and thus turbulence. Turbulence reduces anterograde flow and thus reduces the distal pressure after a stenosis. While very hard to quantify it is essential to mention turbulent flow when you see it. Remember that to distinguish retrograde (i.e. turbulent) flow from flow increase with aliasing you need to look at the color bar on the side of your duplex image, noting that flow increase jumps over the upper limit of the color spectrum while retrograd flow (usually) passes the black zero flow region.


decreases over the lifetime, even more if severe hypertension or calcification of vessel walls is present. This, again, is very hard to quantify, but often easy to recognize qualitatively in your duplex image, when you recognize the pulsation of the vessel wall. Reduced elasticity has to lead to increased flow velocities.


Since blood flow is not continuous but pulsatile and this in varying shapes, we should really be using mean flows in our stenosis calculations. This has historically not been done. As a consequence, valve abnormalities (aortic stenosis, insufficiency) have to be factored in, when we try to calculate the stenosis degree from peak systolic velocities.

Blood pressure and atrial fibrillation

The (pulsating) blood pressure is the driving force of cerebral blood flow, trying to overcome venous and intracerebral pressure as well as the distal blood pressure offered by collaterals (see below). At least, you should note the blood pressure and relativize your stenosis graduation in cases of extreme values. When the patient has atrial fibrillation, you probably should use an “average” heartbeat rather than the extreme values. But bear in mind that absolute arrhythmia is a risk factor for arterioarterial embolization in itself.

The geometry of the stenosis

usually is far from being that of a tube. Rather, the blood flow curves around plaques, rotating and hitting small plaques on the distal wall. Again, the effects are impossible to quantify, but at least the geometry should be noted. The shape of the stenosis area should be remarked upon, if it isn’t circular.

The role of collaterals

The pressure difference along a stenosis is not only determined by the resistance of the stenosis itself, but also by the collateral blood flow which leads to an increase of distal pressure (mostly but not only in diastole). This leads to a reduction of the blood flow velocity in the case of good collateralization, thus also reduced flow velocities.

The danger of a stenosis

In the neurovascular clinic, I try to estimate four risks of a stenosis: the hemodynamic risk (what happens if the stenosis were to increase?), the embolic risk (how high is the risk of an embolic event from the stenosis?), risk factors and other risks.

Hemodynamic risk

The hemodynamic risk is determined by

  • the current hemodynamic compromise (jet flow velocity, CCA flow vs. ICA flow, pulsatilities, MCA flow, CO2 reserve)
  • the dynamics of the stenosis (how long has this been going on? was there time to develop collaterals?)
  • the completeness of the circle of Willis (variations such as A1 hypoplasia, Pcomm hypoplasia)
  • secondary stenoses in the collateral circulation.

The problem is that we cannot foresee whether the stenosis will be slowly progressive or suddenly close up (as in plaque rupture). At least in asymptomatic stenoses I require CT- or MR-angiography to determine the completeness of the circle of Willis.

Embolic risk

The embolic risk is determined by

  • Plaque morphology and
  • Plaque type
  • Whether the atherosclerotic process is active or burnt out.
    As in coronaries, it is not reasonable to revascularise every severe asymptomatic stenosis. But in a patient where the overall atherosclerotic process is currently active (after an NSTEMI, say), we can expect the plaques to rupture.
  • Previous embolic events can be noted on MRI.
  • Emboli detection
  • Is the anti-platelet medication working? Multiplate or similar tests.

Risk factors

  • Did the patient stop smoking? How long ago?
  • Can we use high dose statins in this patient? Statins are highly effective against plaque deterioration, but also have serious side effects (less exercise tolerance, diabetes, muscle problems), especially in the high doses we like to use for severe stenosis.
  • Is the patient’s blood pressure controllable? Note that severe stenosis lead to very labile blood pressures as one of the most important sensors of the system is damped.
  • Exercise? Although this has not been studied properly in carotid artery stenosis, I surmise that health by fitness should improve the prognosis of carotid artery stenosis.

Other risks

  • Central or mixed sleep apnea syndrome – very prevalent among ICA stenosis patients, leading to a bag of systemic problems, not the least being poor blood pressure control.
  • Bad blood pressure control (see above)
  • Development of secondary stenoses in the collateral vessels (contralateral, ECA, …)
  • Development of a collateral rete with its danger of bleeding


I don’t see any better physical theories coming. Also, we can never expect better data than NASCET and this is a bad foundation. Therefore you have to tackle all the complexities outlined above and refrain from simplifying an ICA stenosis to a mere number (always the worst approach).




An algorithm for starting oral anticoagulants after stroke

Once you identified the heart as the emboligenic source of your stroke unit patient’s stroke, the question arises of why, when and how you institute anticoagulation. This hasn’t gotten any easier with all the new drug options, Big Pharma push and the resulting trust we are supposed have in DOACs.

In this short blog entry, I will list my 6-step program for starting oral anticoagulants after an ischemic event. Thanks to the great acronym creator, here is the mnemonic for it: SHuTOFF DOAC.

  • Stroke risk
  • Hemorrhage risk
  • Timing
  • Oral agents
  • Formulary
  • Follow-up

Stroke risk

Calculate the risk of recurrent stroke, if you find data.

  • Atrial fibrillation: 12% vs 1-3%/a under OAC, use a CHADSVASC-calculator for individualized data, bearing in mind that most of the underlying studies were in a primary prevention setting.
  • Atrial thrombus: An atrial thrombus is essentially just a sign of (possibly undetected) afib and insufficient anticoagulation, although it can occur in otherwise bad hearts (see this huge collection of TEE+ pts). Still, we would like to know how more acute the danger of recurrent stroke is, if you find an atrial thrombus on TEE. Or – as increasingly happens – on CTA, when you stumble on a left atrial appendage filling deficit by chance. Does it double or triple? Or stay the same? No proper data found on this. 
  • Ventricular thrombus: Apart from the fact that those are easier to find (TTE suffices) and that ventricular thrombi are due to bad hearts (large MI, severe cardiomyopathy) in general, no data can be found on the rate of acute stroke recurrence in this setting. In the long run (1/2a) it is very impressive (50%) according to very old studies, seemingly lowered by anticoagulation (to 30% in this analysis).
  • Mechanical heart valves: Few studies exist, since everyone thinks these patients absolutely have to be anticoagulated. Only part of the embolic risk is due to the valve itself, the rest comes from afib, especially with mitral valve replacement.
    This 1994 review finds a risk of 4 per 100 pt. years without anticoagulation, reduced to 2,2 by ASS and 1 by VKA. This newer analysis of pts. with St. Jude valves finds similar rates of embolism with OAC.
  • Bioprosthetic heart valves carry a significantly lower risk, about half the number of embolic events seems a good estimate.
  • Low EF: Although we left routine anticoagulation for low EF in primary prophylaxis after the WATCH and WARCEF studies (where a reduction in embolic strokes was offset by the increased bleeding risk for OAC as compared with ASS), a cardioembolic stroke in the setting of severely reduced EF and sinus rhythm should probably trigger oral anticoagulation. I could not find proper data for the stroke risk after an embolic event happened.
  • PFO plus/minus ASA: The risk is extremely low, if no proof of the paradoxical mechanism can be established (no pulmonary embolism, no DVT). Otherwise the risk should be roughly the same as the risk of recurrent venous thrombosis (determined by genetics, mobility, triggers and so on) times the cross-embolism-factor (how many of those embolisms cross over through the PFO, can be measured semiquantitatively in the bubble test, this is my personal invention :-)).

Hemorrhage risk

Risk for spontaneous ICH under OAC

Find and optimize risk factors for hemorrhagic complications under OAC, in particular ICH. For atrial fibrillation there is the simplified  HASBLED-score, but some particular risk factors might benefit from more intensive workup.

  • The A4F complex of the elderly
    • Age
    • Alzheimer’s
    • Apolipoprotein ε2, ε4
    • Amyloid angiopathy
    • Falls
  • Alcohol
  • Altered coagulation (cirrhosis and the like, low platelets)
  • Adherence problems
  • Diabetes
  • Hypertension
  • Interacting medication
    Obviously, the more drugs a patient takes, the higher the risk. With antiplatelet comedication, you double the risk with monotherapy and triple it with dual therapy (or even worse with the newer antiplatelets ticagrelor, cangrelor)
  • Liver and kidney problems
  • MRI markers: leukoaraiosis (no proper gold standard for quantification, may use Fazekas score or volume of white matter hyperintensities, no cutoff established) and number of microbleeds (no proper cutoff because the sensitivity depends on the MRI sequence used and field strength)

Risk for hemorrhagic transformation of the stroke

The acute setting of a stroke raises the obvious concern of bleeding into the stroke (either minor as hemorrhagic transformation or as parenchematous hematoma). Since the detection of hemorrhagic transformation is a matter of the sensitivity of your imaging technique (proportional to the field strength of your scanner, SWI outperforming T2*), only parenchematous hematoma (as can be seen with any old CT or even on ultrasound) should be used to judge the danger of anticoagulation and the reported rates vary between 10 and 30% of cardioembolic strokes. It is unclear, though, whether this rate increases with oral anticoagulation (it does with heparin and LMWH, but those are way more intense) and for how long the danger persists – see next step.


The 1-3-6-12 rule

There is practically no data on when to start OACs after stroke, so we use the guidelines (ESC 2016) with their practical 1-3-6-12d rule (TIA/NIHSS 0, NIHSS < 8, 8-15, > 15), although they don’t regulate the case of hemorrhagically transformed or parenchymatous hematoma in stroke.


And in the meantime?

  • It is unclear whether ASS makes sense for bridging until OAC in afib – it seems to not hurt much after cardioembolic stroke, though. The guidelines recommend ASS bridging.
  • Heparin or OAC bridging (at least with Warfarin) confers no benefit in the huge studies analyzed in Sandercock’s Cochrane analysis. Whether (lower dosed) DOACs could be used in this setting (analogous to, say, half-hearted LMWH treatment) is also an open question.

Bear in mind that the risk for recurrent stroke during the inpatient period is extremely low (around 3% in simple afib patients – old data, but replicated in modern case series such as this one). It is a daunting question when to bridge in the higher risk groups, such as mechanical heart valves. Most reviews recommend 7-14 days (without proper data to back that recommendation).

Oral anticoagulant

In the next step you have to choose among the 5 options. Consider the acronym DOAC:

  • Drugs: what other drugs is the patient on, what are the possible interactions?
  • Organs: is the patient at risk for renal insufficiency or liver failure?
  • Age: since the lightweight elderly are prone to renal failure even with borderline creatinine, age is a risk factor for all DOACs.
  • Compliance: DOACs aren’t very forgiving, if you forget some doses, VKA usually are.


Lookup the dose according to all of the above (most of the DOACs have a low and high dose regime, according to risk factors among the above)


Do you have to ensure lab checks? This is obvious and well established for VKAs. It should also be reasonable to check creatinines in patients at risk for renal failure. Rarely, if ever, are drug levels or anti-Xa-activity needed for standard therapy. When it comes to acute surgery, tPA or in case of bleeding, having drug levels at hand is, well, handy.

Drop attacks

splash-water-1362224788t1gMedical terminology knows 5 reasons for people to fall unaided: common fall, syncope, collapse, seizure and drop attack.

A drop attack consist of the loss of lower extremity tone, leading to collapse, but decidedly without disturbance of consciousness (as opposed to syncopes) and without accompanying neurological or other signs or symptoms, in particular without dizziness or faintness, diplopia etc. Peculiarly, the attacks occur while walking, not standing or sitting. They seem to be quite prevalent, constituting a significant percentage of falls in the elderly.

The concept of drop attacks is very old and yet, there is not much published about it. As far as I got in my literature review, a 1986 series in Neurology has the most modern data (an astonishing 108 patients!). Apart from that you have to work through case reports and chapters in neurology textbooks, such as Neurological Differential Diagnosis – a cased-based approach.

The prototypical patient is a woman over forty who reports falling forward while hurriedly walking on the pavement, as if someone had pushed her, without warning, the legs giving way. She might even have injured herself. Once down, she could get up again after a few seconds, without feeling dizzy, nauseus or unsteady.

The differential diagnosis is huge, since so many diseases have been associated with drop attacks, and in some cases falsely so.

Hemodynamic ischemia

Take vertebrobasilar ischemia, for instance. Before the advent of MRI and CTA, many anomalies in the posterior circulation were interpreted as evidence of pathology, such as hypoplastic posterior communicating arteries, asymmetry of one vertebral artery or hypoplastic V4 segments. In practice, it seems to be nearly impossible to get isolated drop attacks (without vertigo!) from a hemodynamic basilar compromise. In a series of 83 proven basilar artery occlusions from basilar stenoses, prodromi included “drop attacks” in only 4 cases and these were accompanied by vertigo in 3 (Ferbert, Stroke 1990). Similarly rare, yet pathophysiologically more reasonable, is the case of a high grade carotid artery stenosis with contralateral hypoplastic A1-segment – when the compromised ICA supplies both anterior cerebral arteries.

It is interesting to note that the stroke rate of people with drop attacks was not increased as compared to age-matched controls in the 1986 series.

Systemic hypoperfusion (aka syncope without dizziness)

The classical mechanisms of syncope (orthostatic, neurocardiogenic etc., aortic stenosis) practically always lead to disturbance of consciousness or at least dizziness. There is just one exception: rhythmogenic drop attacks (Adam-Stokes attacks). In the above mentioned case series this constituted a sizable percentage (13%). Although I would think that a careful reevaluation reduce that number considerably, I concede that an event recorder is a reasonable investment for recurring drop attacks, not the least, because the gadgets have become so simple to implant.

A special case is carotid hypersensitivity syndrome, where a vagal mechanism due to head rotation or local pressure is usually hypothesized. To be honest, I haven’t seen many cases of this, despite the fact that I am working in a neurovascular lab much of my spare time, so it can’t be that frequent.


Atonic (or astatic) seizures are well-known phenomena in pediatric neurology, arising in Lennox-Gastaut-syndrome, Doose syndrome and other epilepsies. It is rare as a manifestation of adult onset epilepsy, all the less in the elderly, yet the classical temporal lobe epilepsy can lead to temporal lobe syncopes or temporal lobe drop attacks in this age group as in any. 

In these modern times of weird autoimmune encephalitis variants, LGI1-antibody encephalitis has been reported to cause drop attacks even before it’s more typical facio-brachio-crural dystonic seizures.

Movement disorders

In (advanced) Parkinson’s you can be attacked by drops, usually with polypharmacy and fluctuating clinical course (on/off phenomena, freezing). Patients with Progressive Supranuclear Palsy tend to fall backward rather than forward, yet this can be described as a drop attack as well. Both diseases should present with clinical hints at the movement disorder.

Paroxysmal kinesiogenic dystonia has been proposed as an imitator of epilepsy and you could assume that this can lead to drop attacks as well, although I could not find a case report of this. At any rate of occurrence, a family history should help.

Negative myoclonus

This can be an expression of epilepsy (particularly, if focal as in benign partial epilepsy, see above) or a more generalized encephalopathy such as hepatic or toxic, leading to Asterixis (think of Pregabalin, Oxcarbazepine and toxic doses of any central acting drug). History, a hunt for the “flapping tremor” and lab works should rule this out.

Vestibular drop attacks

An acute and temporary disturbance in otolith function can lead to drop attacks. This has been eponymized by Tumarkin who coined the term otolith crisis in the thirties. The attacks are not the correlate of an acute Meniere’s endolymphatic hydrops, but due to unstable otolith function. In contrast to most other drop attacks there ought to be a sensation of vertigo, i.e., of movement of the outer world, yet only few patients can actually report this.

Theoretically, other vestibular disorders, in particular superior canal dehiscence syndrome, should be able to provoke vestibular drop attacks as well, yet there are no case reports.

(Cranio-)Cervical dysfunction

Quite a few diseases of the cervical myelon and the craniocervical junction can lead to temporary compression or dysfunction of either the pyramidal tract or the dorsal column afferent fibres, thus leading to either loss of tone or loss of feeling in the legs, hence the drop attack.

  • Posterior fossa tumors
  • Subacute combined degeneration (Vitamin B12)
  • Chiari Type I
  • Cervical spinal canal stenosis and other causes of cord compression

Other rare causes

  • Third ventricle tumors (colloid cyst, pineal cyst) – usually with postural headache
  • Isolated cataplexy as an abortive variant of narcolepsy
  • Coffin-Lowry-syndrome – stimulus-induced drop events

How to do a presentation

Academically, Iclimbing_through_the_yellow_band_mt-_everest_-may_2007_a was raised in Mathematics, where a lecture is in fact a development of ideas on the black- or whiteboard. In the Mathematics research community things then went downhill with presentations in LaTeX trying to imitate Powerpoints horrible themes.

In Medicine, the situation has been disastrous even before Powerpoint – medical talks have always been boring with lots of facts and no investment into the choreography of the talk itself. Prezi took this even further, emphasizing the structure of the content rather than the talk.

For me it was the fascination with SMACC talks such as Vic Brazil’s famous Timing, Tribes and STEMIs and of course Cliff Reid’s impressive resuscitation talks, that led me first to the theory (Presentation zen, Nancy Duarte and the like) and then the practice (TED talks) of modern presentation creation.

This blog entry serves to collect the most helpful references for creating presentations. It also captures the basic steps I recommend for giving a small 20 min talk on a current topic in our departments’ continued education rounds.

5 steps for creating presentations

While three seems to play an important role in structuring you talk, I prefer rules of 5 for mnemonics, so here are my presentation creation 5.

  1. What
  2. So what?
  3. How
  4. Practice
  5. Present


Choose a topic that interests you and your audience, priority on the latter. Don’t let the topic be dictated. Read up and become an expert. Helpful is Scott Weingarts tip to create a folder with a ppt-template, a note-file (eg mindmap), lots of subfolders (images, articles, videos, recordings) way ahead of the talk to let your subconscious work on the talk all those weeks, collecting articles, noting ideas.

So what?

Why should this topic be interesting for your audience? How do you want to change them? Write an elevator pitch for that.


Choose one of three formats to structure your talk:

  •  Problem/solution: most frequent in medicine
  •  Narrative: most effective – use drama theory, suspense curves and the like
  •  Chronologic: most diffficult – take the audience through a historic tour

Now write the basic steps for Intro, Story, Closing on yellow postits and thus create the structure as a modifiable lattice.


I recommend 5 rehearsals for simple and 6 rehearsals for important talks

  1. Ad hoc: starting from your postits, talk the talk standing in front of an empty chair, then note necessary changes to your structure, write down good formulations as segues in differently coloured postits, fill in the missing data, examples etc.
  2. W/o slides: run through the talk again immediately and note at which point you need visual aids, the flipchart, videos etc to improve the talk. Use differently coloured postits as placeholders. Use the following weeks’ break to search for this stuff and take your time, letting it stew.
  3. To a person: give the talk with slides to a good friend or colleague, gathering as much feedback as you can – this is the first time you give the talk to an audience; make sure you achieve what you planned for, change the talk radically otherwise and start with 2 again. Work through each formulation, each slide. It helps to record the talk on video.
  4. Preparation: give the talk again in private at least a week ahead of the date. You might record this (audio only) Then give your subconscious time to improve the talk even further.
  5. Rehearsal: This is the final rehearsal 1-2 days before your talk. If you give the talk to a big audience, you should repeat this as a dress rehearsal to a small group, such as your peers, gathering feedback.


Annouce and advocate your talk. Go to the venue and check everything out, if possible, 1 day before the talk. Get a clicker (not a pointer). Get to the talk on time. Be well-dressed and smart. Have someone announce your talk to make sure everyone is quiet. Stand before the audience and talk to them, not to the slides, looking them into the eyes, reacting. After the talk, get feedback and improve your talk.

5 steps for creating slides

  1. Not a teleprompter
  2. Not a handout
  3. Don’t distract
  4. Good design
  5. Good images

Not a teleprompter

Using your practice runs, you don’t need the slides as notes. Whenever the information on the slide is just for you to remember what to say, move it to the notes. Only citations may be read verbatim to the audience (it is always so cool, if you can cite by heart facing the audience).

Not a handout

If you feel the audience needs information to take home, prepare a handout. Your slides are not your handout. Every word on the slide needs to be checked whether it is inspirational or handout material. Long lists are handout material.

Don’t distract

Remove everything, every word, every object, every colour, every footer, every logo, everything that doesn’t serve your story. No template. Black background. No animations. No transitions.

Each slide should contain one thing, either a supporting image (fotos for emotions, graphs for illustration) or a slogan, maybe both.

If you absolutely need lists, keep them short, so that the eye can grasp the whole list in one glance (max. 5-6 entries)

Good design

Use all the theory of design (golden cut, font design, clear lines, few words) to improve your slides. Use a professional, if possible. Make it simple. Look at Apple’s advertisements and presentations to get the idea.

If you feel you have to excuse a busy slide, throw it out, redesign it. Most complex slides are better developed on the whiteboard.

Good images

Images are used to create emotions. Use good real fotos in high resolution, not artificial cartoons. You judt have to evoke the feeling. Make sure you have the right to use the images.


Obviously, I did not develop these concepts on my own. Here is an annotated list of useful links and references.