The unreasonable effectiveness of deep brain stimulation
Our best treatment for Parkinson’s disease, and why it shouldn’t work
There are two ways to treat Parkinson’s disease: take drugs, or stick a bloody long electrode deep into the brain, switch it on, and keep it on forever. The drugs are used first. The electrode second, if at all. The electrode can reduce, or even flat-out stop, the crippling symptoms of Parkinson’s disease. It works the instant you turn it on – a truly miraculous thing to watch.
But it should not work at all. The fact it does work tells us we know practically nothing about how the brain works.
What we do know is that the most obvious signs of Parkinson’s disease – the tremor of the arms, the difficulty of moving – follow the loss of neurons in the middle of the brain, neurons that contain dopamine. So the drug treatments for Parkinson’s mostly work by restoring some dopamine to the brain. But they do not work indefinitely. Higher and higher doses are needed over time as more neurons are lost, and as the receptors for dopamine habituate to the surge that occurs every time the drug is taken. Eventually, the dose gets so high that the drugs cause as many problems as they solve. These huge hits of dopamine, larger than our brains ever evolved to cope with, cause the brain to adapt in strange ways. Patients can end up with permanent involuntary spasms or contorted limbs. In the end, the drugs just add to the misery.
Enter the accidental saviour (*), sticking an electrode deep in the brain. Neurologists cunningly call this “deep brain stimulation”; for all their talents, sparkling wit is not a job requirement. But knowing where to put the electrode is. They don’t just shove it any old where, but lovingly and carefully position the tip into a cluster of brain regions, my beloved basal ganglia. Literally the ‘’flobbly bits at the bottom’’, the basal ganglia are crucial for controlling movement. They also receive lots and lots of dopamine from those pesky neurons that die in Parkinson’s disease. So it makes sense that doing something to the basal ganglia could treat Parkinson’s disease.
Doing what, exactly? The neurosurgeon lowers the electrode into one of the basal ganglia; they turn the electrode on, pulsing electricity through it at above 100 times a second. And, instantly, the tremor goes away.
(If the electrode is in the right place. If it’s not, there’s a fair chance the patient’s arm will shoot out and smack the surgeon across the face; or that the patient will burst into tears and have no idea why. And the other crucial bit is the pulsing above 100 times a second. Slower than that, and either nothing happens or the symptoms – especially the tremor – get worse).
It works! So why shouldn’t it work? Well, they stuck that electrode into the only bit of the basal ganglia that contains excitatory neurons, neurons that can make other neurons fire. And that should make things worse, not better.
Our best idea for what goes wrong in Parkinson’s disease is that the neurons in the basal ganglia are firing in the wrong patterns. In these wrong patterns, they bunch their spikes together, ultimately swamping the messages trying to get through the other bits of brain they connect to.
Now think about what deep brain stimulation does: it puts massive, constant, 100-or-more times a second pulses of electricity into that excitatory bit of the basal ganglia. That causes a train of constant excitation to pulse through the basal ganglia, and out to other bits of brain, again swamping the messages trying to get through them. It overwrites one set of wrong patterns with a different set of wrong patterns.
[For the neuro-hardcore: you know I’m talking about the subthalamic nucleus. Yes, there is some controversy over whether deep brain stimulation causes the neuron bodies to fire; but no controversy that it recruits their axons to fire. We have evidence from rats, monkeys, and humans that deep brain stimulation causes a dramatic change in firing rates and patterns in the brain regions that are targeted by the subthalamic nucleus. And these changes are not what we see in normal, healthy firing of these brain regions.]
Our best theories of the basal ganglia say that the role of this excitatory bit is to stop movement; to call a halt to what’s going on, so we may do something else. It does this by exciting brain regions that inhibit other regions all over the brain, that stop those other regions from firing. Our best theories predict that, if we stick this stimulation into the excitatory bit, then we should not be able to move at all. All movement should be blocked. Yet the exact opposite happens. Here, as in practically every part of the brain, we do not have the first clue as to what is going on.
So if deep brain stimulation doesn’t fix the basal ganglia – if anything it makes their output patterns worse compared to normal – how on earth does it work? We have no answer. But we do have ideas. One is that we have our understanding of the basal ganglia all wrong. That actually these new patterns of firing are just fine, as the gaps in between the 100-times-a-second pulses do allow enough messages to get through for the brain to function again.
Another is that we have the direction all wrong. The above is all about how deep brain stimulation causes the excitatory neurons to change their output, that the new signal goes outward from them. But what if the new signal actually goes backward?
The regions of cortex that control movement connect directly to the stimulated neurons. Those connections are made by axons coming out of the cortical neurons, along which they send their firing, their spikes. But axons are not a one-way cable. If you hit them with enough juice, spikes can travel back up an axon, go into their neuron, and change the firing that neuron sends to other neurons in cortex. And is deep brain stimulation enough juice? Yes, definitely. So it could be that deep brain stimulation works backwards: it drives activity back up the axons from cortex, and changes how cortical neurons fire.
We just don’t know.
That’s science. We’re finding out. And in finding out how deep brain stimulation works, we are learning two things at the same time. We are learning how to make the stimulation work better, how to tune it for each patient, how to reduce its side-effects. And we are also learning about how these regions of the brain work when they’re healthy.
At this point, you may be thinking: Hey, dopamine is the key, right? So why don’t we just put the electrode in the dopamine neurons, and release more dopamine?
The answer is simple. It’s because stimulating dopamine neurons is the single most addictive thing mankind has ever discovered. More addictive than Nicotine and Caffeine flavoured Pringles (once you pop, you will never ever stop!). More addictive than a Stardew Valley expansion for World of Warcraft. Take a happy, healthy rat. Put an electrode in its dopamine neurons. Then offer it a lever to push. Every time it pushes the lever, a single pulse of electricity goes down that electrode and makes the dopamine neurons fire. Does it push the lever? Hell yes. They will push it four times a second. A second. 240 times a minute. They will push it without pause, without food, water, or sleep, until they die from exhaustion. Happy, though. So, no, we cannot put the electrode in the dopamine neurons.
In deep brain stimulation we have a miracle treatment for Parkinson’s disease, discovered by accident, that we have no idea how it works. Does it matter that we don’t know how it works? Yes. It means that we do not have a good grasp of what causes the movement problems of Parkinson’s disease. And we do not have a good grasp of why the movement problems have anything to do with dopamine (though clues are emerging); nor why taking a drug that makes dopamine can temporarily fix movement problems. But thanks to deep brain stimulation we do have a good grasp of what we don’t know. And that’s a start.
Want more? Follow us at The Spike
Twitter: @markdhumphries
(*) Accidental saviour? Indeed. Like far-too-many medical treatments for neurological and psychiatric disorders, deep brain stimulation was discovered entirely by accident. It came from the older treatment of destroying small bits of the brain below cortex. This, bizarrely, worked: destroying a select small bit of brain on one side stopped the tremor of the limbs on the other side of the body. And how do you destroy a small cluster of neurons? You zap them with a high voltage coming from, you guessed it, a bloody long electrode. Dr Alim Benabid had a patient on the operating table with tremor in both arms. He zapped a small bit of brain on one side, and the tremor on the other side stopped.
Now he had a problem: the patient desperately wanted him to zap the same bit of brain on the other side, so that the tremor would stop in both arms. But Dr Benabid couldn’t do that, as destroying the same bit on both sides would stop the patient from controlling his arms at all. Instead, Dr Benabid wondered if just stimulating and not destroying would help. How did he make this leap? Because to make sure the electrode was in the right place, they used to stimulate the bits of brain they were passing through, and see the patient’s reaction (remember, in neurosurgery, the patient is awake in the vast majority of operations). And he’d noticed that this stimulation could change, and sometimes reduce, the tremor. But he’d never tested quite how far it could be reduced. So he inserted the electrode, turned down the voltage, cranked up the rate of electrical pulses as high as they would go. Luckily, his amplifier had a top rate of 100 pulses-a-second. When he flicked the switch, the tremor stopped. Voila, deep brain stimulation was born.
