Focus on stroke: How stroke affects the brain
A stroke can cause debilitating brain damage, costing patients control of their limbs, their ability to use language and other cognitive skills. Moheb Costandi meets the researchers who are gaining a better understanding of the affected brain areas and how they work, with the aim of informing the development of new treatments to limit and overcome the effects of stroke.
With a tap on the touchpad, Masud Husain plays a film clip on his laptop. It shows a woman who has suffered a large stroke on the right side of her brain, sitting up in a hospital bed and flicking through a newspaper.
“I’ve asked her to find some interesting stories,” says Husain, a professor of clinical neurology at the Institute of Neurology. “Her left arm is paralysed, and all her interest and attention is to the right. She’s not blind to information on the left, so if you show her this side she’ll read it, but very soon she naturally drifts over to the right.”
Another clip shows a patient with the same syndrome, sitting at a computer screen which displays a grid made up of the letters ‘C’, ‘O’ and ‘Q’. He has been asked to find and point out all of the ‘C’s, something that most of us could manage easily.
But for this patient, the task proves difficult. His right index finger wanders up and down the extreme right of the screen, where he repeatedly points to the same Cs, while completely neglecting the rest of the screen. And when the screen contains more distracting information, his neglect becomes even more severe.
This syndrome – referred to as hemispatial neglect – is the most common cognitive impairment following stroke, occurring in nearly three-quarters of patients who have had a right hemisphere stroke, but as yet there is no treatment for it.
Living with neglect
Husain and his colleagues have been investigating the neurological basis of hemispatial neglect.
In one study, they presented a person who had had a stroke with visual stimuli to the left visual field while using functional neuroimaging to examine the corresponding brain activity. This revealed that even when the person is unaware of stimuli on the left, the visual cortex, which receives inputs from the eyes, does registers them.1
“The eye and the visual parts of the brain see the stimulus on the left,” explains Husain, “but the patient doesn’t, because it’s not entering into conscious awareness.”
Patients with hemispatial neglect tend to have damage to parts of the right frontal, temporal or parietal lobes that are known to be involved in attention and to modulate visual processing by ‘top-down’ mechanisms.
“The term ‘hemispatial’ may actually be a misnomer,” says Husain, “because these patients aren’t neglecting half of space. There’s an imbalance between left and right attentional systems, and maybe a problem with sustaining attention over time.”
“We’ve also shown that working memory is extremely limited in patients like this,” he continues. “This has fundamental implications not only for stroke, but also for attention deficit disorder, and for how working memory operates within the brain.”
Animal studies have shown that a drug called guanfacine, which activates brain receptors for the neurotransmitter noradrenaline, enhances the performance of aged monkeys on attention and working memory tasks, possibly via its effects on the prefrontal cortex.
Husain and his colleagues tested the drug in a small study involving three patients with right hemisphere stroke, and found that it improved the performance of two of them on visual search tasks, apparently by enabling them to sustain their attention for longer.2
The third patient had extensive prefrontal cortex damage. Husain and his team are now trying to replicate the findings in a larger study, and to determine whether or not the extent of damage to this region could predict whether or not a patient would benefit from drug treatment.
Speaking my language
Elsewhere in the same building, Sophie Scott, a professor of cognitive neuroscience, uses functional neuroimaging to investigate how the brain produces and perceives language. She also collaborates with neurologists to try to understand the changes in the brain when damage causes problems with speech.
As speech deficits are the most common symptom of stroke, patients can provide valuable insights into the underlying brain mechanisms.
“It’s been estimated that something like two-thirds of strokes leave people with some kind of communication deficit,” says Scott. “The speech input and output systems in the left hemisphere both fall within the territory of the left cerebral artery, so it’s possible to lose both with one stroke.”
One particularly intriguing symptom of stroke is the so-called foreign accent syndrome. This appears to be rare, with less than 100 cases documented in the past century. It often occurs because of a small stroke that produces subtle changes in the way speech is produced.
“People with foreign accent syndrome can still produce speech,” says Scott, “but they have systematic difficulties with particular speech sounds, and they also have strange rhythm and inclination.”
This syndrome is partly due to problems with syntax, or the way in which words are combined to form sentences. “That’s one of the reasons they sound like a non-native speaker, but they probably don’t get a whole new accent. If you play their speech to a native speaker, they will hear someone with a speaking disorder, not someone with their accent.”
From a theoretical perspective, foreign accent syndrome is interesting because it could reveal how minor changes in the brain’s speech production mechanisms lead to alterations in the speech sounds produced.
Traditionally, the brain was thought to contain two distinct speech areas, each specialised for a particular function: Broca’s area, in the left frontal lobe, which is involved in speech production, and Wernicke’s area, in the left temporal lobe, which is involved in speech comprehension (see ‘Background’, below).
Scott emphasises that higher cognitive functions such as speech involve multiple brain regions acting together, and some of her own work suggests that foreign accent syndrome occurs because of a disconnection between the brain regions involved in planning the articulation of speech and the motor areas that control the production of speech sounds.3
“It’s not as simple as Broca’s and Wernicke’s areas,” says Dr Fatemeh Geranmayeh, who works in the Computational, Cognitive and Clinical Neuroimaging Laboratory at Imperial College London. “There are cases where isolated lesions in those areas have not resulted in the language deficits you’d expect, so there must be other factors involved.”
With her colleagues, Geranmayeh is investigating the function of networks of brain regions during language tasks, and how disruption to these networks affects the recovery of language functions following a stroke.
“My main interest is in a frontal-parietal-temporal lobe network that has previously been shown to be active during language tasks,” says Geranmayeh. “Nobody has looked at this specific network in stroke patients with aphasia.”
Damage to specialised brain regions such as Broca’s or Wernicke’s areas can have a huge impact on speech. The same speech deficits can, however, occur when the connections between these regions are disrupted, because of alterations to the input or output of each region.
On the right tract
Geranmayeh is using functional neuroimaging to examine how activity in each individual brain region is affected by stroke. In addition, she is employing another neuroimaging technique called diffusion tensor imaging to examine the brain’s white-matter tracts, which contain long-range connections between the specialised regions, and how stroke affects the integrity of these tracts.
One such tract, the superior longitudinal fasciculus, is of particular interest. This thick bundle of nerve fibres runs along the length of the brain in both hemispheres, connecting brain regions in all four lobes. Another is the anterior commissure, located towards the front of the brain, which connects the left and right temporal lobes.
Recovery of speech function following a stroke varies widely between patients, and Geranmayeh suggests that examining the integrity of the connectivity within the brain’s language network could be a useful way of predicting the extent of recovery.
“My hypothesis is that disruption to this network is critical,” she says. “Those with relatively intact connectivity will do better in terms of recovery of language function, but if a crucial white matter tract or node has been taken out, then recovery will be worse.”
The same techniques are being used by Dr Nick Ward, a reader in clinical neurology at the Institute of Neurology, to investigate how changes in the brain can lead to recovery of upper limb movement in people who have had a stroke.
“What a patient can or can’t do after stroke is a result of what’s happened to their brain,” says Ward. “Think of the brain as a wiring diagram – there’ll be wires and junction boxes which have been damaged, and this is what causes symptoms such as weakness or difficulty communicating.”
Weakness of limbs on the opposite side to the brain damage is a common symptom of stroke, and this is usually the result of damage to the primary motor cortex, or its connections with the spinal cord, that controls the muscles.
Ward and colleagues have shown how the brain tries to work around stroke-induced damage to the primary motor cortex, by using adjacent secondary motor areas instead to send signals to the spinal cord.4 This is possible because of a phenomenon called plasticity – the brain can reorganise itself – and it is this reorganisation that underlies the recovery of upper limb function.
This back-up system is, however, differently wired up from the way the damaged primary one was. The secondary motor areas connect to the spinal cord indirectly, probably via the brainstem, and the signals it sends are thought to reach groups of muscles rather than individual ones. So despite an overall improvement in limb function, recovering stroke survivors often have difficulty moving individual fingers.
Like recovery of speech, recovery of limb function is highly variable between patients, and Ward argues that detailed analyses of an individual’s brain could predict not only the extent to which they might recover limb function but also the best treatment option for them.
“I’m interested in the differences between stroke survivors in terms of how their brains are organised,” says Ward. “I want to be able to characterise an individual’s brain, in terms of which parts are left undamaged and what’s connected to what. There may a characteristic signature, or biomarker, that tells us whether or not they will respond to a given treatment.”
Although the advent of stroke units has been of huge benefit, he says there’s still much room for improvement, but it would require changes in health service infrastructure. He envisions large neurorehabilitation centres that provide rigorous assessment, diagnosis and bespoke treatment for every patient, and also do research into the mechanisms of recovery.
“Stroke is very common and there are over half a million people [in the UK] with significant impairments,” Ward says. “That’s a lot of people whose hands and arms don’t work, and the question is how much better could they be if they had received optimal treatment.”
“One of the challenges is how to provide treatment more effectively within the constraints of how the health services are set up. Stroke costs an estimated £8 billion a year, so investing in making neurorehabilitation more effective makes economic sense.”
Background: The brain’s language centres
Historically, the human brain was thought to contain two distinct areas, each of which is specialised for a different language function. Both regions are located in the left hemisphere, which is therefore said to be the ‘dominant’ hemisphere.
In 1861, the French anatomist and surgeon Pierre Paul Broca described two patients who had lost the ability to speak after suffering strokes. Broca had examined their brains upon their deaths, and found that both had damage to the same region on the outer surface of the left frontal lobe. The region subsequently came to be known as Broca’s area, and the speech production deficits observed as a result of stroke are often referred to as Broca’s aphasia.
In 1881, the German neurologist Carl Wernicke described several patients with damage to another part of the brain, located towards the back of the left temporal lobe. In contrast to Broca’s patients, these patients had great difficulty understanding spoken language, and Wernicke’s area, as it came to be known, was therefore associated with speech comprehension.
The emergence of modern techniques, particularly functional magnetic resonance imaging (fMRI), has enabled neuroscientists to refine their understanding of the location and function of these two areas, and it is now clear that the historical view of the brain’s language centres is too simplistic.
In 2007, for example, Nina Dronkers of the University of California, Davis and colleagues used fMRI to scan the preserved brains of Broca’s original patients. They found that the most extensive damage was located slightly further forward than the region usually designated as Broca’s area, and that the damage was not restricted to the outer surface, but extended deeper.5
We now know that Broca’s area is largely involved in controlling the muscles involved in producing speech sounds. It is also involved in hand movements, and may, like Wernicke’s area, also contribute to language comprehension.
Earlier this year, Iain DeWitt and Joseph Rauschecker of Georgetown University published a meta-analysis of 100 imaging studies of speech perception, and concluded that the area most frequently activated is located three centimetres farther forward than Wernicke’s area.6
The researchers were quoted in numerous news stories as saying that neuroscience textbooks will have to be re-written in light of their findings. In fact, Sophie Scott and her colleagues had already identified this region, and showed that it responds to and processes speech sounds differently from the way Wernicke’s area does. In a study published in 2000, they concluded that this region is part of a separate, parallel pathway, which processes only those speech sounds that are intelligible.7
- Rees G et al. Unconscious activation of visual cortex in the damaged right hemisphere of a patient with extinction. Brain 2007;123(8):1624-33.
- Malhotra PA et al. Noradrenergic modulation of space exploration in visual neglect. Annals of Neurology 2005;59(1):186-90.
- Scott SK et al. Foreign accent syndrome, speech rhythm and the functional neuronatomy of speech production. Journal of Neurolinguistics 2006;19(5):370-84.
- Ward N. Assessment of cortical reorganisation for hand function after stroke. Journal of Physiology 2011;589(23):5625-32.
- Dronkers NF et al. Paul Broca’s historic cases: high resolution MR imaging of the brains of Leborgne and Lelong. Brain 2007;130(5):1432-41.
- DeWitt I, Rauschecker JP. Phoneme and word recognition in the ventral auditory stream. Proceedings of the National Academy of Sciences 2012;109(8):E505-14.
- Scott SK et al. Identification of a pathway for intelligible speech in the left temporal lobe. Brain 2000;123(12):2400-6.
- Moheb Costandi is a freelance writer based in London.
Wellcome Trust funding
Professor Masud Husain is a Wellcome Trust Senior Research Fellow in Clinical Science. Professor Sophie Scott is a Wellcome Trust Senior Research Fellow in Basic Biomedical Science. Dr Fatemeh Geranmayeh is a Wellcome Trust Clinical Research Fellow. Dr Nick Ward’s group receives funding from the Wellcome Trust.
This article is part of the Wellcome Trust’s Focus on stroke, a series of articles, interviews and videos running throughout May 2012, which is the Stroke Association’s Action on Stroke Month.
For more information on stroke, visit the Stroke Association’s site or call its helpline on 0303 303 3100. If you or someone with you is suspected of having a stroke, call the emergency services immediately.