Okay, so I've been tasked with talking about intrathecal pumps, which I think you'll end up dealing with at some point, and they can actually be incredibly helpful. Some of my most satisfied patients are IT pump patients, but they're also patients that if you have a pump failure, you've got a real problem. So these are my disclosures. My aims, we're going to try to understand which patients might be good candidates, understand appropriate surgical technique, know the results for the different indications, know how to troubleshoot pump related complications, and know how to manage baclofen overdose or withdrawal. So the general concept is you're getting drug to a place that is going to be more effective with less side effects. For most of these drugs. Now, there is one drug pre-alt that cannot be given systemically, or IV, or PO. But in general, let's say with baclofen, you can obtain the same CSF concentration with 1,500th the dose by giving it intrathecally. There are three FDA approved drugs for intrathecal drug therapy. Morphine, baclofen, and zirconatide, the product of the cone snail is the most poisonous substance on the planet. But the reality is that people put all sorts of things in pumps, and you have to know what's in the pump, especially if they come to your practice or show up in your ER, and they're not your patient, they're not in your system. So people put all sorts of things in general, these are made up in compounding pharmacies, which has its own special potential pain. So in terms of indications, indications include spasticity, secondary dystonia, cancer pain in general, we have a good randomized control trial that shows benefit for cancer pain, but if you're an insurer, you don't really want to pay for this if the patient is going to die within the next month. So in general it tends to be cost effective if there's life expectancy more than three months. If it's less than three months in general, our pain guys will place an intrathecal catheter and connect it to subcutaneous reservoir or a long tunnel catheter hooked up to an infusion pump to give the patient relief. And then it gets used for a lot of other things where it's really not clear, there isn't great evidence. So be that somatic pain, failed spine surgery, syndrome, whatever that is, CRPS, postherpetic neuralgia, peripheral neuropathy. And as I said intrathecal is the only route for zirconatide. So in general, you want to have side effects from the drug. So there's nothing that I find more frustrating, well, there are, but I do find it frustrating when a patient comes to me on 20 milligrams of PO baclofen three times a day and has no side effects and they send the patient to me for an intrathecal trial. That makes no sense. Go up on the dose. Same as a patient who gets sent to me for DBS who's not on adequate cinema. But intrathecal baclofen can be very effective for patients who get intolerable side effects. So in general, patients with spasticity or secondary dystonia, they're not a candidate for Botox and in general unacceptable side effects at effective doses for all baclofen. In general, for patients who have an injury, say a stroke or spinal cord injury, we tend to wait about a year to sort of see where they are because things change enough that you don't want to necessarily commit someone to a pump and all the work that's involved with maintaining that. I would say, you know, my intrathecal baclofen patients, these are the patients that I routinely give my cell phone out to. These are the patients that I've actually made house calls for. So if you're going to take this on, you know, it is a big thing to take on, but even if you don't take it on, these patients can show up in your ER and you do need to know how to deal with them. And one thing that I think is important, you know, for post-stroke patients, just because they're asymmetric doesn't mean it's not going to work. So there is still a role for a patient with unilateral spasticity. Conjudications are the things you would expect. Infection, need for anticoagulants. Sometimes, you know, for small kids or people who are super skinny, it can be an issue because of the size of the pump. You know, you have a little old lady where there just isn't much room between the pelvis and the ribs. So every time they lean forward, the pump is digging into one or the other. Or if there's some contraindication to the drug in that patient. So appropriately, all our insurers require an intrathecal trial. I'm very careful in counseling the patients on this, that this is like a light switch. It goes on and off. So often you may be overdosing the patient on the intrathecal trial. You just want to see, do you get benefit? But then I explain to the patients, once you have the pump in, we can adjust it. It's like a dimmer. And we can mess with the dosing in terms of time of day and scheduling. And this can be done as a bolus or continuing infusion. We always measure opening pressure, especially for ITB, because you don't want to miss hydrocephalus, which could be exacerbating the spasticity or is going to increase your risk of CSF leak after pump placement. So there are two pumps that are being sold on the market in the U.S. In terms of dealing with these patients, if they roll into the ER, you need to know which one it is. The 99% of what's sold in the U.S. is the Medtronic Synchromed II. Occasionally you'll see a full one. And I can't say I've seen one of these in a number of years, but J&J had this 3,000 pump that is no longer sold in the U.S. And the problem is that this technology works really beautiful, except that the complication rate is quite high. And those complications can be catheter-related, pump-related, infection or CSF leak. And I think if you look at these numbers from this Japanese series, they're not terrible in terms of CSF leak. About 3%. You can get migration. But I think there is some concern with the newer catheter from Medtronic. I've heard some people say that they, you know, especially Albert Lee in Indiana who's managing about 600 pumps, Mike Turner's practice, that his CSF leak and catheter failure rate is around 30%. So it can be quite high. And complications are all over the place. It seems to be lower if you're using consensus techniques for placement. In general somewhere in the 8 to 25% range. And the problem is that if you stop giving these drugs, patients will go into withdrawal, especially with opioids and especially with baclofen, that can be life-threatening. And this is a real source of inconvenience and actually threatens morbidity and mortality. And it diminishes the cost-effectiveness and adoption of the therapy as frankly it should. With kids, infection is more of a concern. There's some evidence that sub-fascial placement of the pump reduces that risk. But in this paper that's almost 20 years ago from Ken Follett from a Medtronic-sponsored clinical study, you can see complication rates are all over the place. So basically they assembled a group of experts to look at kind of best practices. And I think this has helped. So I think the important things are number one, don't place the catheter in the midline. If you place it through the interspinous ligament, every time the patient extends, you're risking shearing that catheter. You have to come from a paramedian approach. And then you have to have loops behind the pump. In general, the way I do this is I work with the patient in the holding area as to where we're going to put the pump. Because you have the patient sitting in a position that's going to be normal for them. If you don't do that and you get the patient in a position in the OR and you place the pump, you're going to end up with the pump someplace you don't want it. Okay. So I'm going to talk a little bit about troubleshooting and I'm going to focus on the Medtronic pump. And this is probably the most important thing in terms of dealing with these when they come in the door. So you have to understand that there are three spaces in pump design. There's the space between the reservoir where you put the drug, that's in the center at the fill port, to the rollers. It's a peristaltic pump with this system. The Codman and Fluonix are not. Then there's the distance between the roller to the catheter's tip. And you can access that via this catheter access port. That's distal to the rollers. So you can pull off drug that's in the catheter or between the rollers and CSF and you can pull off CSF. And then there's the intrathecal space. And it's really important when you put these in and when you program them, you do program the memory of the device that can be picked up by anyone interrogating that pump, the length of the catheter. And that length, you have to know the internal pump tubing and the catheter volume. So you have to put in the catheter. So the reason this is important is that, you know, let's say you're giving 100 micrograms a day of intrathecal baclofen and the concentration of what's in your pump is 500 micrograms per milliliter, it's going to take 24 hours for the drug to hit the CSF if a bolus isn't given. So that means if you pull off from, let's say you don't prime the pump, you're replacing the pump, let's say, and you pull CSF to make sure the catheter is working and you hook up a new pump, if you don't bolus that patient, that patient is going to be in your ER and then your ICU in acute baclofen withdrawal within 24 hours. So you have to know what's in the drug. Similarly, you don't want to access the catheter access port, do a die study without withdrawing the drug first because then you have an acute overdose, the patient goes floppy and needs to be intubated, drops their blood pressure, potentially dies. So you have to understand what you're dealing with and make sure that you, that the patient is going to get drug when it's appropriate. So this is unfortunately a fairly common scenario. Patient isn't getting desired effect in spite of increasing flow rate or dosage. And this is sort of how I approach this problem. Is there a break or kink in the tubing? Did the tubing migrate? You can get plain radiographs, although the new catheter is frankly quite hard to see. You can see the tip but not most of the catheter and sometimes you end up having a CT to try and get a better idea. And then the question is, is the pump delivering what's programmed to deliver? So you need to make sure the programming is accurate. So this system with the synchro med, it does not actually measure or sense what's in the reservoir. It calculates based on the rate of the peristaltic pump what you expect to have in the reservoir. So to really know if it's delivering, you have to withdraw what's in the reservoir and see if that's what you expect to have in there. So you should have some accuracy. If it's totally off and there's only, you know, your programmer says there should be 10 mls in the pump and there's 15 or two, you've got a problem. There's a big discrepancy. And then you have to do either die study or roller study. We'll talk about that later. So one question is, is the tubing in the wrong place occluded or micro fractured. You can perform a die study. This does not have 100% sensitivity. So you have to be careful. And just because your die study is normal does not mean that the catheter isn't broken. So you turn off the pump. So you don't overdose the patient. Pull one or two cc's off the catheter access port. So that's at least two times the internal pump tubing and catheter volume. And then you just inject OmniPake via the catheter access port and get CT. And then, again, once you've done that, you need to clear that volume with drugs. You have to program a bolus into the pump so that you're not sending the patient home and they're back in your ER in 14 hours in acute withdrawal. So this would be normal myelogram that you would expect. It can get subdural and not intrathecal. And obviously you can also get occluded. One caveat, Rick Abbott more than 10 years ago reported a patient with an intradural but not intrathecal placement of the catheter, flushed the catheter and the high pressure pushed the drug into the intrathecal space causing an overdose. So this is very unlikely if the patient is getting reasonable results from intrathecal infusion, but it's something to keep in the back of your mind. You can also do what's called a roller study where you program to deliver a certain amount of drug over a certain period of time and then you can get x-rays and show that the rollers change position. The problem is that even if you do all this, when push comes to shove, if it doesn't make sense and it looks like the patient is not getting drug and you're putting huge amounts of baclofen or morphine, sometimes you're not going to see a break in the catheter or a CSF leak and frankly we end up exploring those patients if there's a high level of suspicion. When you replace the catheter or replace your system, it's really difficult and you have to keep those patients in house. You can't just send them home. Because you have a patient that you've been delivering let's say 500 micrograms a day and you find the thing isn't working, what do I give this patient? Do I give them 250? Do I give them 100? It's a bit of guess work and if you guess wrong, that patient will be either overdosing or underdosing. You can deal with that, but you want to deal with it in the hospital, not with the patient at home. So the other common scenario is the patient calling saying she has swelling at the lumbar wound, on exam you see a seroma. Sometimes you can manage this conservatively with an abdominal binder and watch while waiting. It will tend to declare itself as to whether it's a seroma, CSF leak, infection, et cetera. You got to know whether it's a drug so you can do all those diagnostics I talked about. If it's CSF, sometimes it will get better, sometimes it won't, especially with this new catheter, there's some concern and then you have to move the catheter to another level. Again, always think about does the patient potentially have hydrocephalus that you're missing and that's why they're at higher risk of CSF leak. Hopefully when you perform the trial there's an opening pressure, but if they came from another practice, all bets are off. And then sort of normal signs of pus, is it erythematous, is it warm, make sure you go through warning signs with the family and patient of meningitis or epidural abscess. So if you think it's infection, you got to be careful. The last thing you want to do is access the catheter access port because then you take a patient with a local infection and give them meningitis. So be careful with that. This can be an issue if the reservoir is low. One of the things about these pumps is if the pump runs dry, the catheter tends to occlude. So we have it set up so it actually, you schedule a patient, you give them a printout of their programming at the end of their visit that says when their refill date is and frankly, the thing starts beeping if it's getting really low. You set up alarms so the patient's abdomen starts beeping. This is also a concern if you have a patient with a pump and the pump dies, you don't want the coffin beeping. So they got to call you so you can turn off the alarms and turn off the system. So anyway, so be really careful about low reservoirs. There are a few reports of saving infected pockets around pumps. I can't say that I've done this. Sometimes with antibiotic infusions or muscle flaps. In general on a patient where we're concerned about meningitis, I'll have our IR guys do an LP under fluoro so they don't shear the catheter with a spinal needle before starting broad spectrum antibiotics. If the gram standard culture is positive, you are not going to salvage it. This is not something that you necessarily need to deal with in the middle of the night. It's best to deal with in the light of day. You need to really make sure that the patient's not going to go into withdrawal. So there are a couple things you can do. You can do a fast taper and give orals for at least baclofen and opioids and see if you can prevent withdrawal that way. Albert Lee has not published this yet, but I saw him give a talk a week or two ago where he talked about what's called a Turner exchange where you place a new pump at a different level and take out the old one. Because the reality is the majority of these infections are in the pocket. They're not up at the spinal level. So patient comes in for withdrawal and with intrathecal baclofen, the mnemonic is itchy, bitchy and twitchy. So they get very irritated. They have pruritus and agitation. They can go into rhabdo. And so those patients can be quite sick. And so if you're considering a patient who's in your ER that's gone through withdrawal, you want to know is the pump empty. So interrogate the pump. And you can always call one of the companies once you figure out what it is to help walk you through it and send a rep if you don't have a programmer. If the pump is dead, well, if it's completely dead, you know, the patient or family says, well, it was beeping for about a week and then it stopped, then you've got a problem. You can use the programmer to interrogate. And then you have to figure out is the patient going to withdraw because of the pump or catheter or broken. So x-rays, may need a pump study, et cetera. In general, if you are unsure, call for help. Medtronic and the other companies actually employ not me but some of my colleagues that can help walk you through a sick patient. But this is actually a great technology, but it is not, if you're not afraid of an intrathecal baclofen pump, you should be, as Yoda would say. Thank you. Thank you. Thank you, Jason. Our next speaker is Jeffrey Arley. Yes, go ahead. I have a couple of questions. Yeah. Yeah. I think it's variable. You know, I know there are some people who are very experienced that, you know, if the patient's doing great and is at a reasonable dose, then they're fine. I tend to, you know, always think the worst. Especially that patient, you know, that gets sent to you and is on some outrageous, outrageous dose of morphine, you know, my hackles are raised. We tend to do, you know, my partner, Eleanor and I, we tend to do all these patients, even for pump replacement, lateral decubitus, we are always prepared to replace the pump. I do tend to withdraw from the sutureless connector that attaches to the pump, one or two cc's, to make sure I can get CSF. That may be overkill, but that's what we do and then reprime the catheter. So we're sort of always prepared to replace it. And you can just take a small 23-gauge needle. It's preferable if you have a blunt one, but if you don't, you can just take it carefully and pull it up from the catheter. And then just two other points. From the catheter access port. Or from the catheter itself, you can pull out if you are disconnecting them, if you have that connector piece, if you wanted to double check, but then you have to know how much you're cutting or splicing. One other point is if you are integrating this into your practice, so Jason gave a great talk with all the nuances of it and everyone should be aware of that, but you could have a lot of pain guys. I don't manage any of the pump settings, which is why I don't mind doing pumps, because otherwise I wouldn't want to do them. But if you have a really savvy pain guy who's happy to manage the pumps for you, as long as you're aware of the nuances, because you'll probably be the one coming in in the night, but if there are issues, you can just be the technician and then they can manage it. So you don't have to be scared of these if you have good pain guys. The other thing just to note is granulomas. Do you want to talk briefly about granulomas? Yeah, so what that does is it gets into more troubleshooting. You can get, especially with morphine and high concentrations and some of the compounding, you can get granuloma at the tip. And so usually the tip is a T8 or something, so the patient becomes myelopathic. If that happens, if you can drop down the rate as much as possible, it will often resolve and then, of course, you need to check for withdrawal. The patient's really decompensating. The general recommendation these days is to do an expansile duroplasty, as opposed to trying to dig into granuloma at the spinal cord where you can really hurt someone. So that, you know, turn it down, it'll usually get better on its own, but if you have to do something, just do an expansile duroplasty. Don't try and tear it off the spinal cord. But you can often do, well, that's a whole other discussion. Yeah, but you can sometimes pull it out safely, but if there's any resistance, you stop and everything. But the granuloma, as you can see, with high doses of morphine, fentanyl, high doses of opioids by themselves, you tend to see it, and they tend to present with patients. You're going up on the dose, going up on the dose, they're not getting relief, and then suddenly they start to have a neurological deficit or myelopathy, and then that's, you get an MRI and you see sort of a granuloma at the tip, and usually that resolves if you just fill the pump with saline and give them time to get better, or you can try to remove it. Just a word of caution about MRIs, SynchroMed 2 is MRI compatible. That Fluonix thing, do not get an MRI with that in place. It'll, like, discharge the drug. So, like, for Fluonix, if you need to get an MRI, you've got to pull everything out of that reservoir, get your MRI, refill, and reprime. So just be a little careful. I'm not sure about MR compatibility with the Codman. In general, it is so easy, if you can, to, you know, as long as you're not afraid of infection, to access the catheter access port and do a myelogram without having to stick them in the back. So that's what we tend to do if there's no suspicion of infection.

intrathecal pumps

patient treatment

surgical techniques

pump-related complications

drug delivery

follow-up care