The topic of massive ischaemic stroke has been explored by the College in several previous SAQs:
Overall, it is a popular area for discussion. Beyond the entry in Oh's Manual, stroke can be explored in greater depth in the following literature:
Incidentally, one cannot call it stroke until the signs have been there for a minimum of 24 hours. Until a day has passed, that hemiplegia could still be just a TIA, right?....
Anyway. Of all the cerebrovascular catastrophes, ischaemic stroke is the most common, comprising about 85% of the total volume. If you are older than 85 when your stroke hits, your chances of surviving to discharge from hospital is around 65% - the other 35% die in hospital - and after discharge, you have around 8% chance of surviving another 5 years. Oh's Manual mentions that even among relatively young survivors about 30% remain disabled to the point of being fully dependent. Of course, all these statistics may be totally meaningless, as the disability and lethality of a stroke is wholely related to the neurological deficits in every individual case.
Causes of stroke in the elderly
Causes of stroke in the young patient
In general, one cannot escape having a CT - in order to rapidly exclude haemorrhage.
A CT venogram (with contrast) may also be in order, particularly if headache and gradual onset have been the dominant features.
Clinical evidence of stroke will then be enough to trigger local thrmbolysis protocols.
An MRI usually follows, as a means of confirming that there is in fact a stroke evolving - usually this is more relevant if posterior fossa pathology is suspected.
A neat breakdown of the management options for early presenting stroke is available elsewhere as a short tabulated summary of advantages and disadvantages, followed by a series of rambling digressions about the various management strategies.
A few key points:
These people should not come to ICU. The availability of a specialist acute stroke unit improves mortality and outcome. That's the sort of place you need to admit them to. The only reason for ICU admission is if the ABCs are threatened.
These people should avoid getting intubated. This topic is covered in greater detail in the chapter dealing with the merits and demerits of ICU admission for stroke.
The in-hospital mortality for mechanically ventilated patients with these conditions were as follows:
The out of hospital 1-year mortality varies from study to study, but seems to be about 50-90%; of the survivors, the vast majority (up to 80%) are severely disabled.
As an added disadvantage, these people seem to be at an increased risk of ventilator-associated pneumonia. However, sometimes you need to do it to protect their airway, particularly when the airway reflexes have been lost. Otherwise, there is only aspiration in their future.
Ventilation should aim for normocapnea. The AHA does not recommend hyperventilation for acute stroke. The level of evidence for this is not very strong, and the authors hasten to add that in this population there are no studies demonstrating harm. However, a low CO2 would cause cerebral vasoconstriction, and potentially do damage to the penumbra.
There is no strong data to suggest any specific blood pressure control target. The AHA/ASA guideline suggest that you should tolerate a systolic blood pressure under 220mmHg systolic, or 120mmHg diastolic. The rationale for this is that the stroke patient may need a high blood pressure to drive blood into their swollen brain, but systolic pressure any higher than 220mmHg is likely to result in haemorrhagic transformation. However, where did the AHA/ASA get those numbers? One hundred anaesthetised cats; an animal model of experimental stroke. Thus, the debate regarding optimal blood pressure management in stroke is far from settled.
At least in the field of acute intracerebral haemorrhage, the recent (2015) INTERACT2 trial found that "larger SBP reductions within the first hour after randomization were associated with lower risks of poor outcome"; the effect was greatest when an SBP reduction of ≥20 mmHg was achieved in the first hour and was then maintained for 7 days. This is somewhat concerning, given that the current recommendations apparently suggest that any systolic blood pressure under 180mmHg is satisfactory. A nice reapprasial of the same INTERACT2 evidence (Arima et al, 2015) found that the best functional outcome occurred in the group of patients whose systolic blood pressure was controlled at around 130-140 mmHg.
For this, unfortunately, there is no game-changing evidence. The topic is discussed at greater length in the chapter debating the utility of ICU admission for stroke patients - there, it is offered as an example of "non-beneficial care". In brief, most of what we know about this is not stroke-specific, but extrapolated from traumatic brain injury guidelines for management of raised intracranial pressure. Basic posture manoeuvres, deep sedation and the use of osmotherapy are the main treatments. There does not seem to be any point in monitoring the ICP; even in patients who herniate and die the ICP does not seem to rise much above 20mmHg. Displacement from mass effect is the mechanism, not loss of perfusion due to increased intracranial pressure. Still, occasionally EVDs need to be inserted to control the hydrocephalus which may develop as a consequence of cerebellar stroke.
There has been some suggestion that the prophylactic use of antiepileptic drugs in large-scale haemorrhagic stroke may have some beneficial effects. These people clearly have new structural abnormalities of the brain, and are often intubated and sedated which would depress clinical features of seizure and produce a hypothetical (and potentially large) population of patients with unrecognised seizures. Ergo, anticonvulsants might be useful, so they say. For example, levitiracetam has been thought to have some neuroprotective effect (Imai et al, 2020). However, a systematic review of data published up to 2019 (Angriman et al, 2019), analysing data from about 3200 patinets, determined that ultimately there is no survival benefit, nor any difference in neurological outcome in the long term. The 2010 AHA/ACC stroke guidelines also do not recommend the routine use of antiepileptics.
Hyperglycemia is associated with increased edema in patients with cerebral ischemia and with an increased risk of hemorrhagic transformation. This is backed by various rat studies, but in humans there is some supportive observation of the association of hyperglycaemia and poor stroke outcome.
However, aggressive BSl control is probably also bad - the unimaginatively named INSULINFARCT trial (2012) has demonstrated that infarct growth is greater if the BSL control is strict.
Thus, the BSL has to be kept within the high normal range.
There is no strong evidence to support the use of therapeutic hypothermia in stroke. Most authors, when pushed, will admit that maintaining normothermia (at least) is important.
There is no benefit from heparin infusion. The overall experience with anticoagulation has led the smart people to conclude that any ischaemia-relieveing benefit from anticoagulation is balanced by the murderous tendency towards intracranial haemorrhage.
In fact, subcutaneous thromboprophylaxis with anticoagulants might even be too dangerous in these people. The odds ratio of major bleeding was over 2. The larger the infarct, the more likely the hemorrhagic conversion. The PREVAIL study from 2007 found a statistically insignificant mortality increase due to clexane in comparision to unfractionated heparin, but recommended clexane anyway as a more convenient option.
This is a clinical entity characterised by a massive MCA stroke which gradually swells, creating mass effect and damaging the contralateral brain. Where this is concerned, we are in an evidence-free zone, awash in the warm froth of expert opinion.
Oh's Manual mentions decompresing craniectomy for these people. Some might argue that the increased chance of survival (NNT =2) is better than nothing. Other might argue that the sort of survival which results is not survival by any sane person's definition, as the patients tend to be fully dependent and frequently chronically unconscious. Younger patients (those under 60 years old) tend to recover better from decompressive craniectomy. This issue of decompressive craniectomy in MMCAS has become the topic of Question 2 from the second paper of 2014, and is discussed at length elsewhere.