PlayPlay Podcast downloadDownload Podcast Take Post TestTake Post Test

Transcription:

Dr. Matthew Delaney: Charles, we are back. This is episode two of Evidence-Based Medicine Breakdown. It's been a little bit of time. We've had a lot of stuff going on, but I am pumped to be back for another month of some fast-paced, high-yield papers.

Dr. Charles Khoury: Yeah. I'm pumped, too. I'm excited to be here. It's funny, I was looking at these papers the other day and a lot of them are related to back pain. As I'm getting older, things are starting to get a little worse. I woke up this morning, getting ready to be here and, you know, once a week or so, my back starts hurting. So I popped open the ibuprofen and took some ibuprofen and I've got some Tylenol in my pocket right now. But it kind of got me thinking, as you get older, you almost take for granted all the times that your back doesn't hurt. So I think for the next week, whenever my back doesn't hurt, I'm going to wake up and say, "Well, I'm glad my back doesn't hurt" because days like today, I'm reminded of why we're going to talk about thoracic spine injuries and skeletal muscle relaxants. So I'm excited about this.

Dr. Matthew Delaney: You speak like you're 74 years old. But no, I mean, I think it's right that as we age, things change. And you don't know this about me, but since we last recorded, I actually had a brief period where I started to wear makeup.

Dr. Charles Khoury: Yeah. You know, I think we chatted about that the other day and I didn't really get the whole story and I just kind of took it as you now wear makeup. So do you want to tell me a little more of the story?

Dr. Matthew Delaney: Yeah. So, I don't currently wear makeup, but I have worn makeup. So I'll set the stage. There's a great conference called Essentials of Emergency Medicine. And I've been asked to speak there a couple years and this year was a different format. So normally, it's just a normal in-person conference. And with COVID stuff, this year they rented out a sound stage in Hollywood. And I'll tell you, Hollywood is not nearly as fancy as I thought, but the sound stage is super nice. And they get like a professional film crew to film and you just give your lecture under the lights. And so I got there the first day, there are a bunch of like union Hollywood-type people setting up like a cheese plate and they're like, "Oh, hair and makeup is over there. Before you give your lecture..." And I'm giving a lecture about neuropathy. This is not the Top Gun sequel. But they're like, "Delaney, be sure you get your makeup done." And so I was like, "Well, look, I've got a face for radio, if anything, so I'll be fine. I've not worn makeup in the 41 years that have preceded this." And so I get ready and I go up to give my talk and they're like, "Hey, did you put your makeup on?" And I was like, I was nervous, so I said, "Yes, of course I've put my makeup on." And they're like, "Well, it doesn't seem like you have. You need to go back and get makeup done." So I like shuffle back over and this woman with a cigarette in one hand and makeup in the other starts to powder my face. And I go back up and they're like, "Well, hey, you're kind of wasting our time. You clearly haven't put makeup on." I was like, "Well, I just came." And so then they start going on the radio and they get this woman, she comes back out and she's up there and she's just furiously working on me in front of everybody. And after a minute, I was like, "Can I just ask what's wrong with me that you have to keep coming and putting makeup on me?" She just looks at me and she goes, "Oh, you just have a really shiny head."

Dr. Charles Khoury: So, is this going to be a thing now? Are you going to start wearing makeup to shift and to the office?

Dr. Matthew Delaney: No. I mean, I've been a little more cognizant of the fact that I think I do have a bit of a shiny head. But the problem was by the time they got my head to be ready for camera, I looked like Bernie from Weekend at Bernie's or like I was involved in a really bad accident and my family told the funeral home like, "Just make him look like he used to look again." So it's not a great look for me. But I'm with you that, as we age, we find ourselves in certain situations doing certain things that younger me would've never thought about.

Dr. Charles Khoury: But does your back hurt today?

Dr. Matthew Delaney: It did earlier, but it's better now. So I think it's shifted over to you right now.

Dr. Charles Khoury: All right. So as we start, I'm excited because all my ibuprofens are in the right place. My work ibuprofen is at work, it's not in my bag. And my bag ibuprofen's not in my car. Everything's in the right place. I'm ready to go. I've got my NSAIDs ready for the day. So let's do it.

Dr. Matthew Delaney: All right. We've got five papers. So these are kind of wide ranging, but a lot of them do hit on things that can make our back hurt. So Charles, first paper looks at thoracic spine Fractures in the panscan era. The paper is by Bizimungu et al, and it's called Thoracic Spine Fracture in the Panscan Era. It's from Annals of Emergency Medicine 2020. And just to set the stage a little bit, over the past 10 to 15 to 20 years, there's been this shift away from selective CT scanning, "You're in a trauma, we're going to image what hurts" towards "Let's do a panscan," so CT head, cervical spine, thoracic spine, chest, lumbar, abdomen, pelvis with some angios thrown in there, depending on kind of how you define a panscan.

And what's very clear from the literature is that when we do the panscan approach and image everything, we find a lot more things. We find injuries, we find incidental findings. But it's not clear from the evidence, and there was a trial a couple years ago, the REACT trial that randomized folks to panscan versus selective scan, and it basically found that mortality was the same. So it's not clear that panscanning improves mortality or significant morbidity. And I think that there are strong arguments to be made for a panscan. I think there are settings where more information is better, but I don't think that applies to every patient who's in a trauma who could potentially be injured. The question they're trying to hone in on is if we're looking at patients who have been in a trauma, kind of what's the incidence that they have at T-spine fracture and what does that mean for the patient going forward? What's the associated morbidity? What's the mortality? And they also looked kind of at, "Is x-ray good or should we be CAT scanning these patients?" So basically, a very big, broad look at what do we do with T-spine fractures? How worried should we be if a patient potentially has a T-spine fracture?

Dr. Charles Khoury: So here's what they did, they enrolled about 11,400 patients. Two-hundred seventeen of them or 1.9% of these patients had thoracic spine fractures. We found that x-ray was not great at all. In fact, 91.4% who had both chest radiograph and CT had their thoracic spine fracture observed on CT only. So that's a huge number missed by x-ray. They also found that 62% of patients with T-spine fractures had associated thoracic injuries. So most commonly, rib fractures, about 45% of them had rib fractures, pneumothorax in 36%, clavicle fractures in 18% and then scapular fractures and hemothoraxes. So a lot of these patients with T-spine injuries also had associated other injuries other than T-spine fractures. Twenty-two percent of patients with T-spine fractures also had concomitant cervical spine fracture, and 25% had lumbar spine fracture.

Dr. Matthew Delaney: So when you look at that overall, patients who had these T-spine fractures did worse than patients who didn't have T-spine fractures and you look at things like admission rates were higher, 89% versus 47%; mortality was 6.3 versus 4; longer median length of stay at nine days versus six days. So you look at that and you say, "Gosh, a T-spine fracture must be a marker for badness." But it's interesting, they looked and said, if you had a T-spine fracture without other thoracic injury, your mortality was similar to what they saw on patients who didn't have a T-spine fracture.

So first, they say as a whole T-spine fractures are problematic. They are associated with worse outcomes for patients. But if it's just isolated, that mortality increase seems to go away. Another piece of good news is that most of these fractures arguably weren't that significant clinically. So the authors say that 47.4% were clinically significant. That number jumps out. But if you look, what they meant was 40% of the patients needed thoracolumbar sacral orthosis bracing, a TLSO brace, and the evidence for TLSO bracing is weak. Ten percent needed surgery, so that's important, and about 4% had an associated neurologic deficit. So I think it's interesting that while, as a whole, patients with T-spine fractures do have an increased risk of badness. If we're looking just at isolated T-spine fractures, that doesn't seem to really move the needle in terms of risk to the patient.

Dr. Charles Khoury: All right. So you're talking about all these isolated fractures. What is your approach clinically here?

Dr. Matthew Delaney: So I think there's several ways that I would go. If you're in a system, specifically in a trauma system where the patients typically get panscanned, I would say, do what your system needs to do. There are system-based benefits potentially to doing a panscan. But if you're in a system where it's your call, you see a trauma patient, you'll decide who you image, this study is really helpful in the world of T-spine fractures. So this was pulled out of the derivation of this NEXUS chest decision instrument. And we talk about NEXUS for C-spine. You know, it's a risk stratification tool and a lot of us aren't using this chest decision instrument. But I think when we're saying, "What am I going to do to try to figure out if a patient could have a thoracic spine fracture?" I like this decision aid. And I'll simplify it here. If you have an abnormal chest x-ray, don't use the aid. I would scan that patient. I would CT that. If you have a distracting injury, so we think about the big stuff, do you have a long bone fracture? Do you have other spine fractures? Do you have a spinal cord injury? And a lot of this is just use common sense. If you think you're not getting a good exam because it's a distracting injury, don't use this aid.

Here's where I think the real power is, chest wall tenderness. If you have any chest wall tenderness, don't use the decision aid. And that includes sternal tenderness, scapular tenderness, or T-spine tenderness. But if you have a normal chest x-ray, if you don't have a distracting injury and nothing in the chest or the spine hurts when you push on it, you could apply this rule. And in fact, in this study, all of the T spine injuries would've been caught had they used this rule. So if I'm in a system where I'm going to selectively scan a patient, I think this is a valuable set of data points I can use to what seems like appropriately risk stratify patients of, "Hey, you have a risk for a T-spine fracture. I'm going to CT you" versus "I think things are okay based on this decision instrument."

Dr. Charles Khoury: So with a potential overuse of CT, are you worried about cost or are you worried about radiation or are you worried about incidental findings? Tell me what you're thinking through here.

Dr. Matthew Delaney: So I used to really worry about the radiation impact on patients. And as CT scanning software has gotten better, we're seeing lower and lower doses of radiation. So, I worry some about that, but less. I think cost is a bit nebulous. Certainly, a panscan is not no cost to the patient, but it's kind of hard to figure out what the insurance is going to cover. I think for me, the incidental findings are real and we have other studies from trauma patients that show up to a quarter of patients who get a CT of their chest, abdomen and pelvis will have incidental findings. And Charles, you know what happens is we find something and somebody's going to follow up on that. And it's hard once you spot some abnormality, whether it's important or not to stop that diagnostic ball from rolling.

So if I'm looking at a patient, I'm going to use this NEXUS chest. If you don't hurt, if I can get a good exam, I'm hesitant to scan you just to get more information, because I think the risk of significant injury is low. And I think more information is not necessarily better.

Dr. Charles Khoury: So the clinical bottom line is that literature indicates that thoracic spine fractures are not common and they're not an independent predictor of worse outcomes in blunt trauma patients. However, most T-spine fractures are associated with other thoracic injuries. We know that neurological injury is pretty rare and that surgery's uncommon for these. If clinicians suspect thoracic spine fracture, chest x-ray alone is not an effective screening and CT should really be strongly considered.

Dr. Matthew Delaney: All right. Charles, we're going to circle back to things that hurt our backs. But you we an interesting paper here that I've reacted strongly against initially, because it looks at kind of performance metrics in the ED and specifically things like length of stay and efficiency of clinicians. I actually liked this paper, but give me your pitch on why you included this paper, because this is a little outside of what I would normally come across.

Dr. Charles Khoury: Yeah. You know, in emergency medicine specifically, and I'm sure in quite a few other specialties, length of stay sometimes gets put out there as this surrogate for high quality care. And the idea is basically if you're efficient and you're getting patients through the system quicker, then your length of stay is going to be lower per patient. And this really was just a hypothesis, but a lot of hospital systems have started looking at this as a metric. And these authors don't really subscribe to this hypothesis and have a different hypothesis that length of stay is more about the types of patients seen rather than an individual clinician's practice patterns.

And the authors also came up with the second hypothesis that the order in which you see the patients on shift impacts length of stay. And this is based on a few smaller studies that tell us that early in the shift, we tend to pick up more complicated patients. And then as the shift wraps up, we tend to pick up the simpler patients who naturally will have a lower length of stay. So basically, they're looking at does the type of patient matter and also does the timing of when we see that patient on our shift have an impact on overall average length of stay.

Dr. Matthew Delaney: The paper is by Chang et al, Association Between Emergency Physician Length of Stay Rankings and Patient Characteristics from Academic Emergency Medicine in 2020. This is a retrospective look at a single ED. So they looked at 62 clinicians who saw over 264,000 patients over a five-year period. Now, the wording of the paper honestly feels a little bit clunky, but there's some really good takeaways. So the primary outcome is variation in characteristics of patients seen by physicians across different length of stay quartiles.. So they look at the physician group and the group including advanced practice providers and said, "Is there a difference in who you're seeing, based on what quartile you are when it comes to speed?" So things like age, sex, emergency severity index when you check in, comorbidities and chief complaints. And they looked for statistically significant differences in the characteristics of the patients. So basically, are the groups different? Are fast people seeing different patients than the slower clinicians? And then they looked with a lot of complicated math to say, "Does when you saw the patient on shift seemed to impact your overall efficiency by looking at length of stay?"

So what they found was that, yes, patient characteristics vary between the four quartiles. And they said that in the longest length of stay quartile, so the clinicians who were labeled as being slowest because their length of stay was longer, generally folks in that quartile saw patients who were older who had more comorbidities and had a higher ESI, so were sicker or thought to be sicker when they checked into triage compared to those in the shortest length of stay quartile, which is a little weird, right? That the most efficient folks saw patients who were overall less complicated and the allegedly least efficient saw these more complex patients.

They also found that the order in which you saw the patients or the shift order made a difference. So at the end of the shift, the length of stay was on average 36 minutes shorter, it's 15% shorter than they saw at the start of the shift. So kind of getting at that idea of, "Do you finish up your shift with quicker or easier patients?"

Now, it's interesting in the end, you know, they break these clinicians into quartiles, and I would feel bad if I'm in the bottom quartile, I would feel like I'm much slower, much less efficient potentially than somebody in the top quartile. And this is great, they said that the differences really were very small. So when they take out the different characteristics of patients and they do a lot of mathematical modeling here, but basically they say it's all about the same. It's the patient type is what matters, not necessarily the clinician seeing the patient. So they said when they even things out, the rankings get all kind of scrambled. So 66% of the time, the efficiency score changed by 10 points. And in 41% of the cases, it changed by over 25%. So I think that that's really interesting that we have these quartiles. We get feedback on this. I make changes to my practice based on this. But when you even things out, the quartiles get almost completely rearranged.

Dr. Charles Khoury: Yeah, I think the study gives us some interesting insights on practice patterns more than anything. It reinforces this concept that it's particularly important in shift work to continue to drive care forward. So we talk about this a lot that. When you're on shift or you're working in clinic, wherever you are, you need to be the one that's driving care forward. There's this tendency to make the mistake of seeing a few sick patients at once, then you sit down and then you wait for the results to come back on those patients. But if anything, for me, the study reinforces the idea that that time could be used to drive forward the care on a few not so sick patients. To use a football analogy, you should always be advancing the ball. You don't want the drive to stall out. If you have time, if you've put in orders on quite a few maybe sicker patients, you've followed up on everybody, you've got this lull of 15, 20, 25 minutes while you're waiting on results, you should probably go see a few of those lower acuity patients and keep driving the care forward.

Dr. Matthew Delaney: Yeah. And if I think about really fast, efficient clinicians that I've worked with, that's what they do, right? It's the doctor who's picked up two or three sick ones. I look at them, and I'm like, "Gosh, you have to be underwater," yet they'll say like, "All right, I'm going to run to fast track and grab a couple more." And again, when you're mixing in the quick ones with your sicker ones, you're going to drive your overall length of stay down. But Charles, I look at this and I think the bottom line here is that, obviously, there are going to be some of us who are faster than others. Obviously, some of us are going to be slower. It sounds like from this paper that the actual difference in speed is more to do with the patients we choose to see, rather than a kind of effort or lack of effort on anyone's part.

You know, I wear two hats. I have my doctor hat. And when I wear my doctor hat, I know that kids get sick often and they're constantly getting these upper respiratory tract infections. It's almost always a viral process. And so when I have my doctor hat on, I can go in a room, I can look at a kid who's got cough, cold, congestion and I know that that kid almost certainly doesn't need antibiotics. The literature very clearly tells us that most of the time, an upper respiratory tract infection in a kid is going to be viral. And that despite knowing that this is almost always going to be viral, we all, no matter where we practice, have been overprescribing antibiotics. We're throwing antibiotics at things that have very little chance of getting better with antibiotics. There's been this shift in the past decade or so towards either just straight up not giving antibiotics, saying "This is a virus. Go home" or giving delayed antibiotic prescriptions. And we'll talk about that.

But I'm all for the idea that when we look at a kid and think you have a virus that we should either not give antibiotics or try to convince the parent, "Hey, let's wait this out because if it's a virus, it's going to get better on its own." That's my doctor hat. My other hat is my dad hat. And you know me, I'm not a helicopter parent. But I'll tell you, man, when I wear my dad hat, my kids are sick and when they were in daycare, I swear that one of my children was febrile basically every day. They're not happy. I am not happy with my dad hat on because I'm having to pick them up from daycare. I've got a kid that feels bad. They're crying, they're not sleeping. And with the dad hat on my head, I would do anything I could to make my kid, number one, feel better, but also go to sleep. And I've taken my kids to their pediatrician and we've been offered antibiotics for what I'm pretty sure is a viral process. And I don't want to cause a fight with the pediatrician and they just hand me this prescription and I say, "Thanks." And it's really hard for me as the dad to not go fill that prescription.

And when I see parents in the ED or in urgent care and fast track with sick kids, this idea of delayed antibiotics has always been a little bit tricky to me. It always feels a little bit like it's a way of making myself as the clinician feel better and giving the parents some choice. But I always assume if I'm doing delayed antibiotics, that they're just being friendly. Like they're going to leave the emergency department and go fill that prescription. And I would guess that 0% of the patients that I give a delayed prescription to, delay the prescription at all. I assume that that's just a thing we do to feel better about ourselves. We feel like we're doing evidence-based care and then they run out and just get that prescription filled.

Dr. Charles Khoury: Yeah. So the paper we're discussing was published in Pediatrics in 2021. It's by Mas-Dalmau, et al. And it is titled Delayed Antibiotic Prescription for Children with Respiratory Infections: A Randomized Trial. This was a multicenter randomized clinical trial, comparing three antibiotic prescription strategies in children with acute uncomplicated respiratory tract infections from 39 primary care centers in Spain.

The three antibiotic strategies, so there was the first one, which was the delayed antibiotic prescription. And this was a prescription that was given to parents with the guidance and a recommendation to only consider administering the antibiotic if the child didn't start feeling better. And there were a few criteria there, but pretty much it was four days for acute otitis media, seven days for pharyngitis, 15 days for rhinosinusitis and 20 days for acute bronchitis. There were also some specifics of if the child had a temperature greater than 39 degrees Celsius after 24 hours or a temp of between 38 and 39 after 48 hours, or if the child felt much worse, so all these things in the delayed antibiotic prescription leg of the trial. The next one was immediate antibiotic prescription. This one is pretty easy. The physician or provider wrote a prescription to be taken from the day of consultation. And then, finally, there was this no antibiotic prescription arm of it, which obviously no antibiotics were prescribed at all.

Dr. Matthew Delaney: They enrolled 436 children. Most of them were between two and ten. If you look ,just over half the kids had acute otitis media, about a third had a pharyngitis. And from a symptom standpoint, it was kind of cough, cold, congestion, fever, general malaise, pain. And if you look at time from onset of symptoms to when did they show up, it was about 2.5 days. So these aren't the parents that are rushing in, but it's, "Yeah, my kid feels bad. They got a sore throat. Their ear hurts. I've waited a day or two. It's not better." When they looked at these three groups, whether you got immediate antibiotics, whether you were told to do delayed antibiotics, or whether you were told you're not getting antibiotics, there was no difference in the mean duration of any symptom until it went away completely. So the groups are same, same, same. I think that's good, but the devil is in the details here. So we're talking about mean duration of severe symptoms. And if you look, the duration ranged from 10 days to 12.4 days. Again, there's no statistical difference between the groups, but that's a really crummy pep talk. When I'm looking at this kid, they've had symptoms for almost three days, and I'm going to say, "The good news is no matter what we do, your kid's going to get better. But it's going to be almost 12 days until their symptoms go away."

They also looked at antibiotic cues. And this is interesting, because I assume that delayed antibiotics is just a thing we do that parents don't buy into. And actually, that doesn't seem to be true, at least in this study in Spain. So in the delayed group where they said, "Hey, you can use this, but we want you to wait," only 25% of the kids actually ended up getting an antibiotic. So the vast majority just didn't take that delayed prescription. In the immediate group, 96% took it, makes sense. In the no antibiotic group, 12% somehow scored some doxycycline or Augmentin on the street. When they looked at other things like complications, did you have to go back and see your primary care doctor? All the groups are about the same. The only thing that jumped out is if you got immediate antibiotics, there were more GI adverse events, because you treat an ear infection, your kid's for sure going to get diarrhea.

Dr. Charles Khoury: I suspect our pediatrics colleagues are already doing this, but studies like this make me think that we need to continue to evolve the way we message the reason we avoid antibiotics. Maybe we need to provide our patients with more information or handouts upfront, maybe in the waiting room, or perhaps we need to sit down and counsel them even more about why we're avoiding antibiotics specifically and why we believe they have a viral infection. But I know that it's tough to fight a wave of misinformation about antibiotics and viral infections in one short visit.

Dr. Matthew Delaney: Yeah, I think you're right, that messaging played a really big role here. And one thing they did that's interesting is they had some scripting. And so they said, "Look, your kid is sick. If it's otitis media, expect four days of symptoms, seven days for pharyngitis, 15 days for rhinosinusitis and 20 for acute bronchitis." And those numbers hold up when you look at other studies. But I think that's an important piece that I haven't been telling parents, is, "Hey, this may last 15 to 20 days." So they're at least setting the stage of, "Oh, this is normal. They're still feeling bad, but Delaney said that they would feel bad."

And you know, I'm probably too hard on parents in terms of how much they love antibiotics. I think there's going to be some cultural differences potentially between where we are and in Spain. But in general, parents seemed kind of up for whatever. So in that delayed antibiotic group, I think it's really telling that 75% of the parents were like, "I think that that's a good plan and I, in fact, will delay to the point of never giving them antibiotics." It's interesting when you look at kind of the opinions of the parents that were randomized to the three groups. Overall, the patient satisfaction and parent satisfaction was about 95%. So they liked it, but they looked to see how pro-antibiotic are you. Now, remember you were randomized to these groups, but in the immediate group, they said 81% of the time, "I'm pro-antibiotic." In the delayed, it was 42%. And then, the no antibiotic group, it was 29%. So there does seem to be something there with this discussion about antibiotics that does make parents say, "Hey, you know, I'm pro-antibiotic if you're going to offer me an antibiotic. I'm less pro if you're going to offer me this delayed approach." That being said, I think that this is hopeful that parents are much more savvy kind of consumers and utilizers of antibiotics than I had previously thought.

Dr. Charles Khoury: So the clinical bottom line, although the author state there's no statistically significant difference in symptom duration or severity in children with uncomplicated respiratory infections who receive delayed compared to no antibiotic therapy or immediate antibiotic therapy strategies, it seems to me the no antibiotic strategy was the clear winner, less antibiotics taken compared to delayed or immediate in patients with diagnoses that are known to be mostly viral with no difference in duration of symptoms, severity of symptoms or complications.

Dr. Matthew Delaney: All right, Charles. It is time for the hot take and we're back in the world of back pain here. And here's the hot take, is that I am getting back into the muscle relaxant game. I had not prescribed these for years, but I am shifting to use these more. I think they're imperfect, but that is a bit of a hot take, because these are medications that I know that you don't typically reach for when you're seeing a patient with low back pain.

Dr. Charles Khoury: Yeah, I totally agree. I've actually been prescribing more methocarbamol with low back pain in addition to Naproxen. And one of the reasons is because of this paper.

Dr. Matthew Delaney: Yeah. One of the issues with how we treat, just kind of run the mill, acute, make you feel bad, wake up in the morning, need to do something about it low back pain is that we don't have a lot of good tools for patients. I think very clearly the literature says that things like acetaminophen and ibuprofen or other NSAIDs are effective. But look, the patients come in, they've maybe taken that stuff already and they need something. I can sympathize with the patient that their back really hurts. And I used to write for opioids a lot. And I think it's very clear now that opioid are often a problematic medicine to use for this acute, what's going to be ultimately be self-limited back pain. But I'm looking at the patient, they're miserable, they've tried the over-the-counter stuff and I want to reach for something else.

And muscle relaxants have not been on my radar screen, but I started to shift back. And I think honestly, this whole category of medication is a little confusing. So I went to the pharmacology literature and they get into this very intense debate about what this category of medication, these skeletal muscle relaxants actually do when it comes to receptors or pain pathways. You know, clinically, the evidence to date has been pretty messy. It's very clear if you give patients muscle relaxants that you will alter their sensorium and patients will even say like, "Yeah, I kind of felt a little bit high." But when we look more closely into the literature, it's not clear for relaxing the muscles at all or are we just sedating the patient to the point that they're like, "I don't know, man. I feel a little bit better"? So the clinical evidence for muscle relaxants has been messy. The pharmacologic basis for why we'd use these medications is weird. And this quote is great, Charles, and it's an author describing the category of skeletal muscle relaxants. And they say, at best skeletal muscle relaxants could be described as "an enigmatic collection of agents." So in terms of what these things are, it's honestly a little bit unclear.

Dr. Charles Khoury: Yeah. The other thing is that while we may not know the exact mechanism of these skeletal muscle relaxants, it's very clear that a lot of us feel very comfortable writing for skeletal muscle relaxants. In the year 2000, this survey found that of the 44 million prescriptions written for the 24 million patients with both acute and chronic low back pain, about 18.5% of them were for skeletal muscle relaxants, making this drug the most commonly prescribed for MSK disorders.

Dr. Matthew Delaney: It's crazy. I guess maybe this isn't that hot of a take if people are already writing for these. But for me, this is a change and this is a hot take based on this paper. So the paper by Abril, et al, is the Relative Efficacy of Seven Skeletal Muscle Relaxants, an Analysis of Data from Randomized Studies. This is in the Journal of Emergency Medicine, April of 2022. And the basic question here is do skeletal muscle relaxants work and do they work on certain patients more than others? The methods here are robust but complicated.

So basically, Charles, this was a single center that did four different randomized placebo-controlled studies, looking at patients who had acute non-radicular low back pain and they ran it sequentially. So to simplify it, they looked at the efficacy and the side effect profiles of seven different skeletal muscle relaxants, compared to placebo and then looked to say, "Does age, sex, baseline kind of functional impairment matter?" So they would enroll patients in the ED and followed up with them a week later. And they used this Roland-Morris disability questionnaire to look at how did you feel at discharge and then at one week. This is not a questionnaire I'm familiar with. Is this something that you've used before?

Dr. Charles Khoury: Yeah, I've actually got it pulled up in another tab. And I've taken the survey myself a few times. I think it's kind of interesting. I'm not entirely sure right this moment that I don't have back pain disability based on this. My favorite question from the survey, "Because of my back, I try to get other people to do things for me," which is like-- I don't know. You take that for what you will. But one question that's not on the survey that I would've loved to see is how many different bottles of ibuprofen do you have? Like I was saying earlier, I've got this work ibuprofen and I've got this home ibuprofen. And I was actually talking to a nurse the other day that told me that she keeps ibuprofen in her bedroom versus the kitchen, so she doesn't have to get out of bed to get her kitchen ibuprofen. And I think that's kind of an entirely new frontier there.

Dr. Matthew Delaney: It's a lot of bottles floating around. So we're talking low back pain, so to get into this study, you had to have two weeks or less of back pain between the scapula and the gluteal folds. So not chronic, it's got to be acute. And then, you had to have non-traumatic, non-radicular low back pain, and that was determined by the attending emergency physician. And then, you had to have enough scores on your disability survey to be like, "Yeah, this seems like your back really hurts." You could not be in the study if you had radicular pain, if your pain had been there for more than two weeks, if you had had trauma in the previous month, or if you'd had back pain on average more than several times per year. Charles, you would probably be out based on the tales of woe you're spinning and the number of times you try to get me to do things for you because your back hurts.

So this is a great patient population. It's exactly what you showed up here to record with. Again, remember multiple sequential randomized control trials, so to not get lost in the details, everybody that's enrolled got NSAIDs plus one of the following, either placebo, Baclofen, metaxalone, tizanidine, diazepam, orphenadrine, methocarbamol and cyclobenzaprine. Those are all of the muscle relaxants that I knew of plus like three or four others. So everybody's getting NSAIDs, you're randomized to one of those different medications. We can put the different kind of formulations in the show notes, but that's the basic setup here. So, Charles, what did they find when they randomized you to NSAIDs plus something else?

Dr. Charles Khoury: Yeah. So the primary outcome was improvement in the RMDQ survey between ED discharge and the one-week followup and a five-point improvement on this scale is generally considered a clinically significant improvement. They also looked at patients who reported that their pain intensity was either none or mild at one week. And at one week, 887 patients were enrolled and they actually followed up with about 97% of them at one week. And I'm not sure how they got such a great response rate, but kudos to them. At one week, 60% to 68% of patients had no or only mild pain. And these numbers were basically the same across the various groups, including the placebo. Everybody who still had pain got about 10 points better. The mean improvement was a 10-ish point decrease on the RDMQ.

Interestingly, at one week, placebo performed pretty much as well as skeletal muscle relaxants. Sex, age, baseline, RMDQ score and history of previous episodes of low back pain didn't meaningfully impact the association between the skeletal muscle relaxants, placebo and the one-week outcomes. There were more adverse effects with cyclobenzaprine over placebo. It was about 35% versus 16% with a significant P value. It was about 35% adverse effects with cyclobenzaprine versus 16% otherwise. And these were mostly drowsiness, dry mouth, dizziness and nausea. Across all agents, patients reported more adverse events if they took the medication more frequently and more of the adverse events were reported in women. Most skeletal muscle relaxants seem to be pretty well-tolerated.

Dr. Matthew Delaney: I like this study. I think it's really interesting. One of the things that I love about it is how many people got better at one week. And we've talked about how historically, if you look at patients with acute low back pain at six weeks, 94% will be back to their baseline function. They won't be asking friends to do stuff for them because of their back pain. But this study is even better. It says that one week, there's a 60% chance that your back pain is going to have gotten better. So I can now put that into my pep talk here.

You know, Charles, this idea of treating back pain with something else beyond NSAIDs is really interesting. And when I first went through this paper, I got the wrong conclusion, I think. Because you look and it's like, "Well, placebo doesn't beat a skeletal muscle relaxant." So we should just do NSAIDs if placebo is as good as these other agents. But I think it's important to remember that the placebo effect, to give someone a pill and say, "This is a pill to treat your symptoms," the placebo effect is real. And in this study we find that it's actually as real as these skeletal muscle relaxants. And I don't think we should move past that too quickly. Patients show up, they have back pain. We say, "Here's an NSAID, and we're also going to give you something else." And whether that something else is a skeletal muscle relaxant or placebo, it's that doing something else and harnessing that placebo effect, and I know I sound kind of woowy woo here, like I'm going to sell you a crystal, but this is real. And I think this is yet another study that says the standard therapy plus something probably does get patients feeling better. And so, ethically, we can't prescribe patients placebos. And if we're looking at what's the other thing I can do, I would say that opioids are not great. I think skeletal muscle relaxants aren't perfect. But this study tells us that, you know, if that's your something else, patients do feel better and it probably doesn't matter what something else you pick, just do something in addition to just the standard therapy, the over-the-counter stuff.

Dr. Charles Khoury: Yeah, this study kind of reinforces what I've been doing for moderate to severe acute low back pain. I like writing a strong NSAID, like naproxen, in addition to a muscle relaxant like Robaxin or Xanaflex for a couple of days.

Dr. Matthew Delaney: That's a hot take for us. Apparently, 20% of us have been writing these already. But yeah, that's certainly a shift for me. All right. We're going to close up with a pitch. The pitch here is that we need to refine what we're doing when we're doing a lumbar puncture on a patient. So to set the stage, it's very clear from literature, we're doing fewer and fewer LPs. One of the things is we have vaccines for meningitis. So in kids and in young adults, we're not having to tap as many people, which is great. And the other thing that's really changed over the course of my career is CT scan has become a much more effective tool to look for subarachnoid. So when we trained, you got a negative CT, you're getting an LP if we're worried about an aneurysmal bleed. And we're doing less and less of that, not to get into the weeds of whether we should or when we should do that. But the bottom line is in my practice, I'm doing a whole lot fewer lumbar punctures than I was doing even five years ago. And I always worry when there's a procedure that I know how to do, but I just don't do often. Do I get a little kind of subclinical skill atrophy? Do I lose a little bit of finesse?

There are still patients, I did in LP the other day. We're still doing this procedure. We still need to be able to do this procedure. And we need to do it well, because while it's a generally safe procedure, the morbidity associated with it, and specifically that this risk of a post-LP headache is real-- now, I don't know how real it is. The studies say it happens between three and 33% of the time. And when I was trained, I was told that when you're doing an LP, there are things you can do to prevent or decrease the chance they could have a post-LP headache, things like you got to pick the right patient, technique mattered. I mean, attendings would have these like weird things. If I twisted to the right 90 degrees or I twisted the left 90 degrees, I always tell them to do this when they go home. But the general idea that there are factors we can control that would decrease the chance the patient would get a post-LP headache. We've said these things for years, but we haven't really looked at them from a scientific standpoint. So the paper by Cognat et al is Preventing Post-Lumbar Puncture Headache in Annals of Emergency Medicine 2021.

Dr. Charles Khoury: Yeah, I totally agree with you, Delaney. We are doing less LPs nationally. The data indicates that. I think that we have more advanced modalities to figure things out, except for in the case of meningitis, in which case, obviously an LPs going to be necessary. But this particular literature review addressed 19 frequently asked questions regarding post lumbar puncture, headache risk factors, and prevention.

Dr. Matthew Delaney: The authors asked a lot of questions, but run through at rapid fire, which I love. So kind of putting them in categories, are some patients more at risk of post-LP headaches? Can we select patients? So, are women more at risk? Actually, they might be. It's debatable. Body habitus, this is really interesting. Thinner patients probably have a higher risk of post-LP headache because the thought is the higher your BMI, the higher your opening pressure, so it's less likely you'll get a headache. So women, debatable; thinner patients, debatable. Children and newborns have a higher risk of post-LP headache. Interestingly, the older you are, the less likely that you'd have a post-lp headache. And I think it's like you get some brain atrophy and you just have more room for things to slosh around. Patients who have chronic headaches might have a higher risk, that kind of fits. And then when, they looked broadly at underlying diseases or medications, none of those were associated with higher risk of post-LP headaches.

Dr. Charles Khoury: So what about needle types? Which needle should I use and how should I handle it to minimize the risk of post-LP headache?

Dr. Matthew Delaney: This is the thing that I was fussed at the most in residency. I'd be halfway through the LP, and the attending of the day would come in and be like, "I would've done this differently." So there's this whole debate about, do you use cutting or traumatic versus kind of more blunt-tipped or atraumatic needles? Really, what that's getting at is when you're going into the spinal canal, you're going through tissue, you've got arachnoid, you've got matter there. And these cutting needles will just cut it and kind of leave a path for you to go in, theoretically, for CSF to come out after the procedure to cause the headache, whereas the blunt-tipped needles just kind of push it to the side. So, it seems very clear from the literature that these blunt-tipped atraumatic needles are associated with a low risk of post-LP headache.

Other things that we think, does the diameter of the needle matter? Doesn't seem to, it's not totally clear. Reinserting the stylet, again, the idea being here that you don't want to pull this tissue out with you. So if you put the stylet, just going to leave it there or cut it off, that doesn't really seem to matter.

Now, I'll tell you, I don't love these atraumatic needles. I find them to be harder to use. And I think it's just because I was trained on the cutting needles, but these authors look at a couple of studies and say that my kind of argument here is wrong, that in fact atraumatic needles, these blunt-tipped needles do not seem to make the procedure more difficult.

Dr. Charles Khoury: So should I perform an LP in a specific way to prevent post-lumbar puncture headache?

Dr. Matthew Delaney: Maybe, but it's not super clear. So patient position, laying down verse seated, there's not enough there to say definitively. So it probably doesn't matter. Higher intervertebral spaces, where are you typically going in?

Dr. Charles Khoury: L4, L5.

Dr. Matthew Delaney: I thought you were supposed to go there. The study say, if you go at L1, there's a lower incidence of post-LP headache. That might be true in the study. That seems really high up for me to go, but there you go. Difficult lumbar punctures, they found no association with difficult lumbar punctures in the incidence of post-LP headaches, which if I think about a few I've done in the past, that's good for the patient, because I have sweated through some of these. Limiting CSF volume or aspirating CSF, basically, "Oh, we'll get a little fluid, not a lot of fluid," that doesn't seem to be associated with decreasing the incidence of headache.

Dr. Charles Khoury: So I've always been taught that there are things that we can do after LP to reduce post-LP headache occurrence. What does this review say about that?

Dr. Matthew Delaney: This is awesome. So I used to tell patients like, "Don't get out of bed for a day. You got to push fluids," turns out that doesn't seem to be real. So better rest after LP, nothing. Fluid supplementation, nope. Caffeine, I mean, if you ever try to get a blood patch for a patient, the interventional radiologist or anesthesiologist ask you a million questions about like the formulation of the coffee they drink. This study found caffeine does not seem to decrease the incidence of post-LP headache. Other pharmacologic agents, basically like if we throw pain medicine at patients, will that help? Maybe, the evidence is really low quality. All right. Charles, so that was a rapid fire run through many, many different things. To break it down, tell me things that are associated with a lower risk of LP headache, things that are associated with a higher risk and then what are things we're doing that are just generally unhelpful?

Dr. Charles Khoury: Yeah. Let's review. So the things that are associated with a lower risk of post-LP headache are older age, higher BMI, use of an atraumatic needle and lateral decubitus positioning. Things that are associated with a higher risk of post-LP headache, history of headaches, bedrest following LP, and a distal disc space. And things that are probably unhelpful are IV fluids and caffeine.

Dr. Matthew Delaney: Bottom line here for the pitch is if we're looking at a patient and we're going to need to do an LP, there are some modifiable things. I can't change your age. I can't change your BMI. I can reach for an atraumatic needle. I can work on positioning some. There are patients, there are techniques that are going to be associated with an increased or decreased risk of post-LP headache. But there are some things that are just going to be unmodifiable.

Dr. Charles Khoury: All right. So I've learned a ton today. I am full of caffeine and ibuprofen. I feel better. I'm ready to take on the day. So this has been a good review.

Dr. Matthew Delaney: Yeah. So five papers, five easy bottom lines. Thank you for listening. If you like what you heard, go to your favorite podcast platform, give us a five-star review. Shoot us an email. Let us know if there are topics that you'd like us to discuss. If we discuss a topic that you didn't like, or you have a different take on it, reach out to us, let us know. But this has been episode two of Evidence-Based Medicine Breakdown. We will be back next month with more tales of woe and makeup. So thanks for listening. Talk to you next month.