#and that isnt even getting into other factors affecting treatment efficacy

[salt-baby]

What EDS treatments are effective?

you all responded really well to a previous medical education post I did, so I wanted to try something new: a series on research papers!

today I’m looking at “Ehlers-Danlos Syndrome: An Analysis of the Current Treatment Options”(2019) by Song et al

tldr:

when researchers (well, doctors) looked at the medical records of 98 patients with EDS of any subtype, from their physical medicine and rehabilitation (PMR) clinic, to find out which medications improved, did nothing to, and worsened symptoms, they found:

- complimentary and alternative medicine and opioids were most common for patients to try, with NSAIDS, physical therapy, and bracing/splinting close behind.

- the following improved symptoms in at least 60% of patients: massage, bracing/splints/orthotics, ibuprofen, a nerve block, platelet-rich plasma, prolotherapy, taking a combo of hydrocodone and acetaminophen, steroid injections, oral steroids, baclofen and botox.

- physical therapy improved symptoms in 43% of patients and heat in 58%.

- surgery, going to a chiropractor, acetaminophen alone, and drugs for nerve pain (ie duloxetine, gabapentin, tri-cyclic antidepressants) tended to have low efficacy (less than 30% of patients had an improvement in symptoms).

Note that there may have been a bit of bias in terms of who participated in the study, that it’s not certain how reliable the medical records are, and that there’s a bit of ambiguity in the results because they included multiple subtypes of EDS.

if you want me to walk you through the whole paper, and learn some skills for interpreting papers on your own, keep reading!

I chose this paper for a couple reasons: its open access, easy to read, and has very clearly reported and well defined results. it also has very simple statistics (which is great, because thats my least favorite part). 

the introduction provides some good background information on EDS, although a few things are of note. One is that they refer to a phenomenon called pain sensitization or central pain in paragraph three, which they don’t go into too much detail on. This topic is a candidate for another post in this series, since it’s surprisingly little known. in short, this has to do with the body adapting to being constantly in pain, and as a result, an abnormal pain response develops (3). I don’t want to go into it too much here, but if this part was confusing to you, that’s what it’s referring to. 

additionally, they mention the so-called three stages of hEDS, which I traced back to this paper (4). While this researcher has a lot of publications in heds, I find it hard to believe that a sample size of only 21 (very different!) patients is enough to firmly define three stages that all patients go through. Plus, their ages are never reported, and “stiffness” is defined as just “not hypermobile anymore”. IMO, not very strong evidence.

Looking at the methods, most of it is as I said above. This type of paper is a retrospective review, meaning they looked at data that already existed, did analysis on it, and drew conclusions on it. Reviews in general can be really useful when you’re first starting out on a specific topic, and I recommend people start there. 

Another thing about methods is they used a Mann-Whitney U test, which, fair warning statistics isn’t my strong point, but I’ll do my best. For more resources, look here and here (5) (sidenote, this whole college-level statistics textbook is open access and very practical, and the author is clearly very passionate about teaching.). For the purposes of this paper, just note that the P value in the last column of their results table indicates significance, ie whether the data they got was any better than random chance. Significance in research is usually defined as a P value less than 0.05. The lower the P value, the more likely the results weren’t just random chance. Any P value over 0.05 is generally considered just random chance. 

That’s really important for interpreting these really beautiful results tables. Make sure you pay attention to the little “Significant? Y/N” column. If there’s an N, there’s not enough evidence to prove anything about that treatment one way or another.

Note that when you’re interpreting the results, just because it didn’t work for most people in this study doesn’t mean it won’t work for you. I really like diclofenac gel for joint pain, but only 24% of these participants had the same experience, and that result had a P value of <0.0001!

There’s not much to say about what the authors wrote as their interpretation of the results, which I think was sound. 

Regarding limitations of the study, I think there’s a few. Overall, my opinion is that the paper is decently strong evidence of how effective certain treatments are, especially when combined with other, more specific papers. But for limitations, the sample size is a bit small, but for EDS, pretty good. 

There’s the issue of almost all participants being women, but that’s a product of EDS affecting female people more strongly than male people (1). To be specific, both afab and amab people have an equal chance of inheriting EDS, because although the exact gene hasn’t been discovered, the way it’s inherited has been (an autosomal dominant gene) (1). It’s possible that sociological factors are at play, but the author’s conclusion (and my own) is that it’s likely one of many biological factors that makes EDS worse in female people than male people(1). I’m using female/male instead of afab/amab because anecdotally, some ftm trans people with eds have found their symptoms improved with testosterone (which one person in this treatment efficacy study also found).

This study also mixed all the subtypes together, with only 76 of the 98 participants having diagnosed heds. 2 had other subtypes (ceds and cveds), and 20 were unspecified EDS. That muddies the results a bit, because although the subtypes have some things in common, they can also be pretty distinct. For that reason, I would’ve liked to know what happened when they separated by subtype and ran their significance tests, but it’s not the biggest deal to me that they didn’t.

Then there’s the issue of the sample itself, which was taken from a PMR clinic. Who goes to a PMR clinic? people who live within a reasonable distance and have the money to both get the referral and pay for the specialist. Additionally, people may be hesitant to tell their joint doc about EVERY treatment they tried, and that doc may not have written everything down. Plus, this study only includes those diagnosed with eds (which again requires money), and therefore may not represent EVERYONE with eds, diagnosed or not. In research, these things can be really difficult to control for, and in my opinion, there’s not much the researchers could’ve done to fix this without massively changing the study. 

Additionally, all these medical records would’ve been made by the same few doctors at this clinic, who may have had biases or writing quirks that impacted the data. Plus, the doctors who wrote the paper also probably wrote some of those medical records. That’s not unusual, but definitely not ideal.

I did notice cannabis was missing from the list of treatments (although cbd was not), likely reflecting it’s illegality and controversy. patients don’t want to tell their doctors that they did illegal drugs, and researchers don’t want to publish about illegal drugs either. Still, it’s worth pointing out that cannabis has promising efficacy for eds pain (2).

In conclusion, I do really like this paper, and I think it’s a great intro to both research and EDS treatments. From here, there’s plenty of more specific papers about specific treatments, and its a great way to get started.

This series will be navigable by the tag “salt baby reads”, and I really do mean to encourage learning and questions with this. If you don’t understand, or (respectfully) disagree, please feel free to send me an ask. I want to start discussions about this, because I really do think its important for people to be educated about the conditions they have, so they can make the medical decisions that are right for them. Let me know if you liked this!

Sources:

1.  “Ehlers–Danlos syndrome hypermobility type and the excess of affected females: Possible mechanisms and perspectives” (2010) Castori et al. 

2.  “Use of complementary and alternative medicine by patients with hypermobile Ehlers–Danlos Syndrome: A qualitative study” (2022) Doyle et al.

3. “Pain in the Ehlers–Danlos syndromes: Mechanisms, models, and challenges” (2021) Malfait et al. 

4. “Natural history and manifestations of the hypermobility type Ehlers–Danlos syndrome: A pilot study on 21 patients” (2010) Castori et al. 

5. “Handbook of Biological Statistics” (2014) John H. McDonald