What are your thoughts on Focused Ultrasound as a future non-invasive therapy?

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

someqsaboutstuff

Full Member
2+ Year Member
Joined
Oct 28, 2019
Messages
74
Reaction score
32
I recently learned about fUS for tremor and PD and also saw that there are many clinical trials for a wide variety of neurologic and psychiatric disorders. I was wondering what the current opinion on this is. Does it look like a therapy that will be implemented in many neuro clinics within the next 20 years, or are the results (thus far) from clinical trials not that promising? I think it would be interesting to perhaps perform this therapy for patients with psychiatric illness as well as movement disorders and even epilepsy.

Members don't see this ad.
 
Most movement docs prefer DBS in my experience. Non-invasive is a silly term too. They're basically melting part of your brain with ultrasonic waves.
 
  • Like
Reactions: 1 user
Movement subspecialist with a focus in neuromodulation here.

Focused ultrasound (FUS) is kind of a blast from the past. Going back to the early and mid 20th century, neurosurgeons were doing thalamotomies and pallidotomies for tremor and Parkinson's disease, just using a craniotomy and a wire to induce enough heating from RF to burn a hole in the brain. That was effective too, but there's a reason DBS took over. Not only is DBS safe to use bilaterally (which lesioning therapies generally aren't), it also offers reversibility in the case of side effects or inaccurate placement and the ability to adjust location, field size, and electrical parameters. It's infinitely more flexible and in the case of diseases that evolve and progress over time, vastly superior.

FUS just skips the crani and wire, but still amounts to burning a hole in the brain - an irreversible hole. Side effects? Hope you get over them. Not beneficial enough? Revisions aren't usually a thing. All at the altar of "non-invasiveness". 99/100 patients who want FUS are just poorly educated on the matter and the use case for FUS over DBS is virtually non-existent outside of the occasional case where surgical risk makes the difference.

As far as expansion to a variety of neurological and psychiatric disorders - DBS is headed that direction, though much more slowly than many had predicted. The abject failure of the DBS depression trials put a damper on the field, and use for things like OCD and Tourette's, while more clearly supported by literature, is still a messy process that frequently doesn't produce as much benefit as people expect. You'll likely see FUS kind of tailing along behind the DBS field, with Insightec using marketing to convince geriatric laypeople that magical ultrasound waves will cure whatever disease we've figured out how to treat reliably with DBS.
 
Last edited:
  • Like
Reactions: 3 users
Members don't see this ad :)
Movement subspecialist with a focus in neuromodulation here.

Focused ultrasound (FUS) is kind of a blast from the past. Going back to the early and mid 20th century, neurosurgeons were doing thalamotomies and pallidotomies for tremor and Parkinson's disease, just using a craniotomy and a wire to induce enough heating from RF to burn a hole in the brain. That was effective too, but there's a reason DBS took over. Not only is DBS safe to use bilaterally (which lesioning therapies generally aren't), it also offers reversibility in the case of side effects or inaccurate placement and the ability to adjust location, field size, and electrical parameters. It's infinitely more flexible and in the case of diseases that evolve and progress over time, vastly superior.

FUS just skips the crani and wire, but still amounts to burning a hole in the brain - an irreversible hole. Side effects? Hope you get over them. Not beneficial enough? Revisions aren't usually a thing. All at the altar of "non-invasiveness". 99/100 patients who want FUS are just poorly educated on the matter and the use case for FUS over DBS is virtually non-existent outside of the occasional case where surgical risk makes the difference.

As far as expansion to a variety of neurological and psychiatric disorders - DBS is headed that direction, though much more slowly than many had predicted. The abject failure of the DBS depression trials put a damper on the field, and use for things like OCD and Tourette's, while more clearly supported by literature, is still a messy process that frequently doesn't produce as much benefit as people expect. You'll likely see FUS kind of tailing along behind the DBS field, with Insightec using marketing to convince geriatric laypeople that magical ultrasound waves will cure whatever disease we've figured out how to treat reliably with DBS.
Well this is a shame xD. I find DBS interesting but feel that it isn’t worth going into NSGY just for a single subspecialty, and my impression is that it wouldn’t be as satisfying being a neurologist that helps manage DBS vs being the neurosurgeon who actually is the “the guy” who implants the electrode. I also was imagining a scenario where fUS could be a procedure frequently billed by neurologists, potentially increasing salary by a lot.

Can you comment on how involved you are in DBS therapy as a neurologist, how it’s satisfying and how my impression (helping NSGY w DBS) may be wrong?
 
Well this is a shame xD. I find DBS interesting but feel that it isn’t worth going into NSGY just for a single subspecialty, and my impression is that it wouldn’t be as satisfying being a neurologist that helps manage DBS vs being the neurosurgeon who actually is the “the guy” who implants the electrode. I also was imagining a scenario where fUS could be a procedure frequently billed by neurologists, potentially increasing salary by a lot.

Can you comment on how involved you are in DBS therapy as a neurologist, how it’s satisfying and how my impression (helping NSGY w DBS) may be wrong?
Implanting the electrode is hardly a skilful task. The neurologist decides on “where exactly” to plop the electrode, and the neurosurgeon just does that. Mapping is the hard part in DBS (done by neurologist), not the actual implantation.

fUS is a rather hotly debated topic. Thama is right in that by all metrics, fUS is worse than DBS, and less reversible. However, Michael Fox at Harvard in this video (link below) talks about an interesting aspect where most patients seem to rather prefer a fUS over hardware. It’s mostly because patients aren’t aware of the risks of fUS, but humans inherently don’t want invasive surgery into their brains.



If you enjoy the surgical aspect of putting neuromodulation devices in, one thing you could consider from neuro is doing NeuroIR and hoping the stentrode trial (Thomas Oxley) shows positive results. It’s somewhat of a pipe dream though.
 
  • Like
Reactions: 1 users
fUS is a rather hotly debated topic. Thama is right in that by all metrics, fUS is worse than DBS, and less reversible. However, Michael Fox at Harvard in this video (link below) talks about an interesting aspect where most patients seem to rather prefer a fUS over hardware. It’s mostly because patients aren’t aware of the risks of fUS, but humans inherently don’t want invasive surgery into their brains.

I get referrals for FUS all the time as our institution has a machine and a neurosurgeon willing to use it. The natural impulse to want something less invasive serves as a way to get people in the door and thus allows us to educate them.

100% of people referred to me that have wanted FUS over DBS have no idea that FUS creates a lesion - when I tell them "it burns a hole in your brain" they immediately recoil. What they've been led to believe through slick marketing is something like magical organic kosher ultrasound waves will cure their Parkinson's or tremor, which is the new way to avoid having PROBES LEFT IN YOUR BRAIN THAT A DOCTOR CONTROLS.

Following a frank discussion of what both procedures entail and what data and experience suggest they can expect as an outcome, some decide to get DBS instead, and the rest decide not to get any procedure for now. I've yet to have someone want FUS after being properly educated.
 
  • Like
Reactions: 3 users
Well this is a shame xD. I find DBS interesting but feel that it isn’t worth going into NSGY just for a single subspecialty, and my impression is that it wouldn’t be as satisfying being a neurologist that helps manage DBS vs being the neurosurgeon who actually is the “the guy” who implants the electrode. I also was imagining a scenario where fUS could be a procedure frequently billed by neurologists, potentially increasing salary by a lot.

Can you comment on how involved you are in DBS therapy as a neurologist, how it’s satisfying and how my impression (helping NSGY w DBS) may be wrong?

Functional and stereotactic neurosurgery is actually a great neurosurgical subspecialty, and encompasses a lot more than DBS. Really, DBS is the easiest thing that they do - they really aren't all that involved except briefly as a technician. We point, they shoot, we do the rest.

Let me give you an overview of a typical DBS process that's an amalgam of the process at multiple institutions where I've trained and practiced.

Patient is referred to neurology (movement disorders) for DBS evaluation. Sometimes they're mistakenly referred to neurosurgery instead, and if they catch this they'll just refuse the appointment and tell them to see us for evaluation. We evaluate them for appropriateness of DBS, confirm diagnosis, order preoperative imaging, refer to neuropsychology for cognitive testing, schedule any other necessary testing (like ON/OFF testing in PD), and send them to neurosurgery for preop evaluation. NSGY has a brief visit to recap and places any skull fiducials needed (if using a frameless system). Patient receives pre-op MRI and CT. We make the surgical plan, including target and trajectory - sometimes the surgeon will be involved here, but not usually.

On day of surgery, the surgeon makes the craniotomy and fixes the frame or frameless apparatus, then sets up the microdrive with recording electrode. We drive the recording electrode and perform microelectrode recordings (MER) to map the target nucleus. Based on MER, we move the drive to the location where we want the permanent electrode, and we tell the surgeon to switch the electrodes out. We then test benefit and side effect threshold at each contact to ensure we have a working electrode. If the placement appears to be poor, we and the surgeon typically mutually decide on a direction to move the electrode, usually with an intraoperative CT guiding us. Surgeon then anchors the electrode to the skull and closes, bringing the patient back 1-2 weeks later for battery implantation.

Surgeon follows up 2-3 weeks later to check surgical site, and then wishes the patient well. We take over and program the electrode over several visits, typically 3-4 in the first 6 months, and going down to every 6-12 months once we've optimized them. When the patient needs a battery 4-5 years later (10-15 in the case of rechargeables) or has a hardware issue, we send them back to the surgeon.

90-95% of DBS care is done by neurologists, and actually in the OR it's at least an even split or better in our favor. If your impression is that neurology "[helps] NSGY w DBS" then you have it precisely backwards.

Regarding billing, FUS has been marketed as something a neurologist could bill for, but from what I've seen it's the opposite - insurances are fine paying for a neurologist in the OR for DBS, but balk at us billing for our time for FUS, and so FUS is more neurosurgery-driven than DBS by far. DBS actually bills really well for neurologists - our OR time bills significantly higher RVUs per hour than clinic time would, DBS programming bills reasonably well (and is a good spot to integrate a midlevel with close supervision), and even creating the surgical plan is billable time. Training in both DBS and botox is kind of the secret way to become a largely procedural neurologist through a movement fellowship.
 
  • Like
Reactions: 6 users
I get referrals for FUS all the time as our institution has a machine and a neurosurgeon willing to use it. The natural impulse to want something less invasive serves as a way to get people in the door and thus allows us to educate them.

100% of people referred to me that have wanted FUS over DBS have no idea that FUS creates a lesion - when I tell them "it burns a hole in your brain" they immediately recoil. What they've been led to believe through slick marketing is something like magical organic kosher ultrasound waves will cure their Parkinson's or tremor, which is the new way to avoid having PROBES LEFT IN YOUR BRAIN THAT A DOCTOR CONTROLS.

Following a frank discussion of what both procedures entail and what data and experience suggest they can expect as an outcome, some decide to get DBS instead, and the rest decide not to get any procedure for now. I've yet to have someone want FUS after being properly educated.

Have long term outcomes been studied with FUS? Seems to me that you run a risk of have damage outside the intended area.

Good point about where the patients are coming from. They read about magic ultrasounds in rats that break apart proteinopathies, so that's understandable I guess. There's such a huge chasm between what we know and they know that it is sometimes difficult to even understand their misunderstandings.
 
Have long term outcomes been studied with FUS? Seems to me that you run a risk of have damage outside the intended area.

Good point about where the patients are coming from. They read about magic ultrasounds in rats that break apart proteinopathies, so that's understandable I guess. There's such a huge chasm between what we know and they know that it is sometimes difficult to even understand their misunderstandings.

The RCT for FUS thalamotomy included outcomes out to 12 months, which is reasonable in the field (the 2 major DBS RCTs in PD went to 6 and 24 months, respectively). There's some hint that tremor starts to return toward the end of that period in their data but would need longer term outcomes to be sure. We see breakthrough tremor all the time in DBS for ET after a year or 3, but we have the option to gradually modify stimulation parameters to compensate. FUS hasn't been widespread long enough to really see those outcomes en masse yet.

FUS pallidotomy doesn't have an RCT, though the main preliminary study appears to show stable benefit out to 12 months. Interestingly FUS subthalamotomy does have a small RCT, and while effect appears to be persistent out to 12 months, the side effect profile is kind of... yikes. Over 20% of treatment group with persistent lesion-related deficits at 12 months that if they had DBS and we couldn't improve with programming, we would consider lead revision. Not good at all.

I would consider FUS thalamotomy in the rare case of a patient with truly disabling tremor who was a poor surgical candidate - say an elderly person living alone with a history of CHF and DVT with PE on anticoagulation that they can't stop. The use case for FUS pallidotomy is far more narrow, and it's probably going to be truly non-existent sometime this year when Abbvie finally gets approval for their subcutaneous levodopa pump.
 
  • Like
Reactions: 1 users
The RCT for FUS thalamotomy included outcomes out to 12 months, which is reasonable in the field (the 2 major DBS RCTs in PD went to 6 and 24 months, respectively). There's some hint that tremor starts to return toward the end of that period in their data but would need longer term outcomes to be sure. We see breakthrough tremor all the time in DBS for ET after a year or 3, but we have the option to gradually modify stimulation parameters to compensate. FUS hasn't been widespread long enough to really see those outcomes en masse yet.

FUS pallidotomy doesn't have an RCT, though the main preliminary study appears to show stable benefit out to 12 months. Interestingly FUS subthalamotomy does have a small RCT, and while effect appears to be persistent out to 12 months, the side effect profile is kind of... yikes. Over 20% of treatment group with persistent lesion-related deficits at 12 months that if they had DBS and we couldn't improve with programming, we would consider lead revision. Not good at all.

I would consider FUS thalamotomy in the rare case of a patient with truly disabling tremor who was a poor surgical candidate - say an elderly person living alone with a history of CHF and DVT with PE on anticoagulation that they can't stop. The use case for FUS pallidotomy is far more narrow, and it's probably going to be truly non-existent sometime this year when Abbvie finally gets approval for their subcutaneous levodopa pump.
What’s going on with the neuroderm l-dopa pump? I didn’t know that Abbie was developing another pump.

Will there really be two different pumps? I guess options are good.
 
What’s going on with the neuroderm l-dopa pump? I didn’t know that Abbie was developing another pump.

Will there really be two different pumps? I guess options are good.

Both Abbvie and Neuroderm have been developing their pumps, but Abbvie appears to be winning the race. The major complication from these pumps is the tendency to develop subcutaneous nodules after prolonged use - nothing dangerous, but undesirable. Neuroderm seems to have had more trouble with this with their formulation than Abbvie has, leading to their pump needing 2 simultaneous injections sites to Abbvie's 1. Abbvie applied to the FDA in May 2022 and those I know involved with their trials expect them to receive approval sometime this year. Neuroderm just finished their phase 3 and are probably a year or so behind.

Duopa sucks and Abbvie is eager to move on, I think.
 
Last edited:
  • Like
Reactions: 1 user
Functional and stereotactic neurosurgery is actually a great neurosurgical subspecialty, and encompasses a lot more than DBS. Really, DBS is the easiest thing that they do - they really aren't all that involved except briefly as a technician. We point, they shoot, we do the rest.

Let me give you an overview of a typical DBS process that's an amalgam of the process at multiple institutions where I've trained and practiced.

Patient is referred to neurology (movement disorders) for DBS evaluation. Sometimes they're mistakenly referred to neurosurgery instead, and if they catch this they'll just refuse the appointment and tell them to see us for evaluation. We evaluate them for appropriateness of DBS, confirm diagnosis, order preoperative imaging, refer to neuropsychology for cognitive testing, schedule any other necessary testing (like ON/OFF testing in PD), and send them to neurosurgery for preop evaluation. NSGY has a brief visit to recap and places any skull fiducials needed (if using a frameless system). Patient receives pre-op MRI and CT. We make the surgical plan, including target and trajectory - sometimes the surgeon will be involved here, but not usually.

On day of surgery, the surgeon makes the craniotomy and fixes the frame or frameless apparatus, then sets up the microdrive with recording electrode. We drive the recording electrode and perform microelectrode recordings (MER) to map the target nucleus. Based on MER, we move the drive to the location where we want the permanent electrode, and we tell the surgeon to switch the electrodes out. We then test benefit and side effect threshold at each contact to ensure we have a working electrode. If the placement appears to be poor, we and the surgeon typically mutually decide on a direction to move the electrode, usually with an intraoperative CT guiding us. Surgeon then anchors the electrode to the skull and closes, bringing the patient back 1-2 weeks later for battery implantation.

Surgeon follows up 2-3 weeks later to check surgical site, and then wishes the patient well. We take over and program the electrode over several visits, typically 3-4 in the first 6 months, and going down to every 6-12 months once we've optimized them. When the patient needs a battery 4-5 years later (10-15 in the case of rechargeables) or has a hardware issue, we send them back to the surgeon.

90-95% of DBS care is done by neurologists, and actually in the OR it's at least an even split or better in our favor. If your impression is that neurology "[helps] NSGY w DBS" then you have it precisely backwards.

Regarding billing, FUS has been marketed as something a neurologist could bill for, but from what I've seen it's the opposite - insurances are fine paying for a neurologist in the OR for DBS, but balk at us billing for our time for FUS, and so FUS is more neurosurgery-driven than DBS by far. DBS actually bills really well for neurologists - our OR time bills significantly higher RVUs per hour than clinic time would, DBS programming bills reasonably well (and is a good spot to integrate a midlevel with close supervision), and even creating the surgical plan is billable time. Training in both DBS and botox is kind of the secret way to become a largely procedural neurologist through a movement fellowship.
This is fascinating! Thank you for the detailed explanation. I definitely have to shadow the DBS neurologists at my institution.

Do you have an opinion on adaptive DBS? I have a lot of research experience in machine learning, and it would be really cool to help design algorithms for DBS as well as manage patients with related issues.
 
This is fascinating! Thank you for the detailed explanation. I definitely have to shadow the DBS neurologists at my institution.

Do you have an opinion on adaptive DBS? I have a lot of research experience in machine learning, and it would be really cool to help design algorithms for DBS as well as manage patients with related issues.
Adaptive DBS means a lot of different things to different people. The stuff likely to be available in the foreseeable future (closed-loop systems modulating ON/OFF or amplitude in response to pathological signal) is probably a lot of fluff. The stuff that might be really cool (i.e. AI-driven ongoing automated programming) is a long way off.
 
Adaptive DBS means a lot of different things to different people. The stuff likely to be available in the foreseeable future (closed-loop systems modulating ON/OFF or amplitude in response to pathological signal) is probably a lot of fluff.
Curious why you think adaptive DBS is more fluff than substance?
 
Curious why you think adaptive DBS is more fluff than substance?
So, we already have a sensing system on the market. It's pretty much useless as currently constructed because a) the feedback it provides is less useful than a clinical history and exam, and b) the sensing itself drains battery at an alarming rate. People that leave it on all the time are coming back in less than a year for battery replacement in a device that should last 3 to 5.

Recognizing this, Medtronic is putting this system into a rechargeable. This will mostly remove the battery drain issue, but then you have to ask what it actually does. It's slightly convenient in terms of initial programming, and then there's little utility beyond that.

Using the signal from a sensing electrode to reduce or turn off stimulation during times when it's not needed is the simplest form of adaptive DBS, with the theoretical benefit of improving battery life. But if you need a rechargeable battery just to drive the sensing component, well, you can see how circularly useless this approach is. This is the kind of adaptive DBS that we are close to.

AI-driven modifications of programming parameters in response to physiological signal are much farther away. Not only because we don't have the AI yet, but because we don't understand what signals to listen for that would indicate we are encountering off-target effects, which are the primary consideration when programming a patient in the real world. We don't even have the input to train a simple classifier yet.
 
My experience with Medtronic's beta band sensing is that it's essentially useless.
 
Both Abbvie and Neuroderm have been developing their pumps, but Abbvie appears to be winning the race. The major complication from these pumps is the tendency to develop subcutaneous nodules after prolonged use - nothing dangerous, but undesirable. Neuroderm seems to have had more trouble with this with their formulation than Abbvie has, leading to their pump needing 2 simultaneous injections sites to Abbvie's 1. Abbvie applied to the FDA in May 2022 and those I know involved with their trials expect them to receive approval sometime this year. Neuroderm just finished their phase 3 and are probably a year or so behind.

Duopa sucks and Abbvie is eager to move on, I think.

OOF!

 
OOF!

FDA wants more info but doesn't want them to collect more data. Kind of odd - hopefully a brief hiccup.
 
FDA wants more info but doesn't want them to collect more data. Kind of odd - hopefully a brief hiccup.

Yeah, in re-reading it could be anything. From the device, to the tubing, needle, or drug manufacturing.
 
Top