Craniocervical Instability in Hypermobility and EDS: What the Evidence Actually Says

• About
• Latest Posts

Adam Foster

Director at The Fibro Guy

Adam Foster is the founder of The Fibro Guy and one of the leading voices in hypermobility, EDS, and chronic pain rehabilitation. A former Reconnaissance Soldier, he developed chronic pain after an IED blast in Afghanistan and built his approach from the ground up when the medical system had no answers. Since 2009 he has worked with thousands of people across eight countries, drawing on pain neuroscience, motor learning, and sensorimotor rehabilitation. He co-authored "No Pain, No Pain" with trauma psychologist Dr Mairi Harper, directed the documentary "Invisible Monster," and has been featured in Forbes, BBC Radio 4, ITV News, LiveScience, and The Telegraph.

Latest posts by Adam Foster (see all)

• Fibromyalgia and Central Sensitisation, What Is It in Plain English? - 8 October 2026
• Which supplements are worth taking when you’re hypermobile - 4 October 2026
• Hypermobile Wrists: Why They Hurt, Which Brace to Use, and How to Build Control - 4 October 2026

Craniocervical instability is one of those phrases that, once you’ve met it, is very hard to un meet. It usually turns up with a frightening diagram, a testimonial from somebody who says a fusion gave them their life back, and a link to a clinic that’ll scan your neck in an upright machine and tell you what’s wrong. If you’ve got hypermobility and a neck that aches, feels heavy, or gives you the distinct sense that your head isn’t quite screwed on properly, that combination is built to hold your attention, and it does.

So it’s worth going through slowly. Craniocervical instability is a real diagnosis, it can be serious, and a small number of people with connective tissue disorders genuinely have it and genuinely do better after surgery. It’s also, at the same time, one of the most oversold ideas in the whole hypermobility world. The distance between how often it gets talked about and how often it actually turns out to be the problem is really quite enormous.

We’re not going to tell you whether to have surgery or to avoid it, as that decision belongs to you and a neurosurgeon who has properly examined you and your imaging. We’re also leaving Chiari malformation, tethered cord and intracranial hypotension mostly alone here, even though they orbit the same conversation. This one’s about the neck, the measurements, and the much larger group of people whose neck is hypermobile and thoroughly fed up without being unstable at all.

What craniocervical instability actually is

The top of your spine is a properly clever bit of engineering. Your skull sits on the first vertebra, the atlas, which sits on the second, the axis. The joint between the skull and the atlas is what lets you nod. The joint between the atlas and the axis is what lets you turn your head, and that one does a great deal of the work, as it provides roughly half of your neck’s total rotation. The whole region is called the craniocervical junction, and because it has to be mobile and stable at the same time, it doesn’t rely much on bone locking into bone, it relies on ligaments.

The alar ligaments and the tectorial membrane tie the skull to the upper vertebrae. The transverse ligament holds the atlas snug against the peg of bone sticking up from the axis, which is called the dens. Between them, those straps stop the skull and the top two vertebrae sliding too far relative to each other, and relative to the brainstem and spinal cord running straight through the middle of the lot.

Craniocervical instability, or CCI, is when those straps allow more movement than they should, to the point where the movement itself becomes the problem. Atlantoaxial instability is the same idea one level down, between the atlas and the axis specifically. And the theory in connective tissue disorders is straightforward enough. If your ligaments are more lax throughout your body, the ligaments holding your head on aren’t exempt from that. So a subset of people with conditions like Ehlers-Danlos will have a craniocervical junction that moves too much, and in the worst cases that lets bone press on brainstem or cord.

Now, a hypermobile neck isn’t the same thing as an unstable neck, and that distinction is where most of this topic quietly falls apart. Hypermobility means more range of movement than average, which is the sort of thing the Beighton score tries to capture. Instability means movement that’s genuinely pathological, producing real mechanical compression or real neurological signs. Generalised hypermobility is built into what hEDS means, so the first one is common by definition, and the second one really isn’t.

It’s also worth noticing that the Beighton score never looks at the neck at all. So a hypermobile neck is usually something assumed from the rest of you, rather than something anybody has gone and measured.

Somebody has been through everything published on how CCI in Ehlers-Danlos gets diagnosed and operated on, and the short version is that there’s no agreement on it. No agreement on which measurements to use, none on who should be operated on, and none on what counts as a good result, with even the choice of measurement inconsistent across most of the papers ^[1]. The verdict was that treatment indications remain a matter of debate, and that agreed guidelines and care pathways are still needed ^[1].

How often it’s actually the problem

Welsh health records were used to count everybody carrying a diagnostic code for Ehlers-Danlos or joint hypermobility syndrome. The rate came out at roughly one in every five hundred people, most of them women ^[2]. That’s not a small group. What is small, sitting inside that already large group, is craniocervical instability that would genuinely warrant surgery. The two get quietly conflated online, until an hEDS diagnosis starts to feel like carrying a ticking time bomb at the top of your spine.

It isn’t one.

The rare cases get a lot of airtime for reasons that have nothing to do with how common they are. People who’ve had dramatic surgery and improved are, understandably, vocal about it. The clinics that operate publish their own results. And a genuinely frightening symptom, like the sense that your head might come off your neck, travels through online communities considerably faster than a boring one does. None of that tells you how common the underlying problem is though, it tells you how compelling the story is.

There’s a selection effect stacked on top of that as well. A lot of the figures you’ll see quoted, this percentage of hEDS people also have tethered cord, that percentage also have instability, come out of groups who’d already been referred to a neurosurgery service. They were symptomatic, and being considered for an operation, before anybody counted them ^[3]. It’s a bit like standing in an A and E waiting room and concluding that most people have broken bones. The number is real for that room, it just isn’t the number for everyone with hypermobility.

Dizziness, brain fog and headaches

Open any page on craniocervical instability and you’ll find a symptom list: occipital headache, dizziness, a bobble head feeling, brain fog, visual disturbance, tinnitus, fatigue, racing heart on standing, difficulty swallowing. Read that while feeling rough and it’s pretty much impossible not to tick most of the boxes. The trouble is that the list isn’t specific to CCI at all. Nearly every item on it also turns up in migraine, in the general symptom load that comes with hypermobility, and in dysautonomia, which is the automatic side of your nervous system not regulating things properly. And every one of those is a great deal more common.

In a retrospective cohort of people with Ehlers-Danlos, autonomic dysfunction was the single most common systemic problem, sitting alongside high rates of migraine and headache, gut symptoms and mast cell issues ^[4]. A much larger survey of people with hEDS found the same thing, with multiple overlapping problems being the norm rather than the exception, and gut problems, chronic pain and joint subluxations dominating the list. Dysautonomia, anxiety and migraine were all more common than in most comparison groups ^[5]. So the dizziness and the brain fog and the racing heart are already very well explained by things we understand perfectly well in this group. You don’t need a mechanically unstable skull to account for any of it.

That matters because the treatments could not be more different. If your lightheadedness is your body struggling with being upright, which is what orthostatic intolerance means, the answer is fluids, salt, graded conditioning, and the rest of the boring but effective stuff. We cover it in our work on the POTS, EDS and MCAS overlap, and in the first step most people miss with POTS and exercise. If your head aches at the back of the skull, that pattern is common enough in this group that we wrote a whole piece on headaches and migraines in hypermobility. It overlaps heavily with the coat hanger pain that tracks blood pressure rather than bone position. If it’s brain fog, our article on brain fog in EDS, POTS and long COVID walks through the far more likely drivers. Not one of those gets fixed by fusing your neck.

None of which is us waving your symptoms away. If your head genuinely feels unsupported, if you get electric shocks down your arms, if you’ve got new weakness or clumsiness, all of that deserves proper assessment. We’ll come to the genuine red flags at the end. The point is narrower than that. A symptom checklist that matches your experience isn’t a diagnosis. Where the same list gets produced by a common problem and a rare one, the common one is nearly always the answer, and it usually responds to something considerably less drastic than an operation.

Why a scan can mislead you

Craniocervical instability gets diagnosed on imaging, usually with a set of measurements taken off the bones. The ones you’ll see named are the clivo axial angle, the Grabb Mapstone Oakes measurement (which sounds like a firm of provincial solicitors), the basion axial interval, and how far the atlas turns on the axis ^[1]. Each one carries a threshold meant to separate normal from abnormal, and two of them come up far more than the others.

The clivo axial angle is the angle your head sits at on top of your spine, and anything under about 135 degrees gets taken as a sign that the brainstem might be getting bent. The Grabb Oakes measurement is how far a bit of bone bulges into the space your brainstem and cord run through, and it’s supposed to come in at just under a centimetre ^[1].

On paper, that sounds precise. In practice there are two problems with it, and somebody has now gone and measured the first one properly. A group in Australia took upright MRI scans of fifty adults who’d been sent for imaging for something else entirely, mostly a trapped nerve or ordinary neck pain. Anybody with an inherited connective tissue disorder had been deliberately screened out ^[6]. Then they applied the standard cut offs to those scans, and what came back depended a great deal on which position the head happened to be in. Up to about one in seven of those entirely ordinary necks crossed the Grabb Oakes line. And a handful came in at or below the clivo axial angle that’s supposed to signal a deformed brainstem ^[6]. The authors called those what they are, false positives, and concluded that the usual cut off may overestimate the risk of compression of the brainstem and upper cord ^[6]. Two crossed both lines at once, and they’d been sent in for a trapped nerve ^[6].

Those are the lines that are supposed to mean something is wrong with your skull. They were being crossed by people nobody suspected of anything, in work that had gone out of its way to exclude connective tissue disorders. Not one person in it was hypermobile.

The same round up that catalogued these measurements graded the evidence behind them as low to moderate quality. It also noted that several of the key papers came from the same author, and partly reused the same people, which introduced selection bias and pulled the level of evidence down ^[1]. Where your reference range gets built largely out of the people you already decided to operate on, it will tend to agree that people like them need operating on.

The second problem is movement itself. A lot of the selling happens around upright and dynamic imaging. So you get scans taken while you sit, or while your neck bends forwards and back, on the promise that they show up instability an ordinary lying down scan would miss. There’s a real logic to imaging under load, in fairness. But a hypermobile neck comes with a fairly obvious catch, as it moves more than average by definition. Stack that on top of cut offs that already misfire in people who aren’t hypermobile at all, and the odds of a number coming back that looks alarming go up considerably. Without a single thing having changed inside your neck.

None of this means imaging is useless, or that every private scan is a con. It means a measurement on its own doesn’t settle anything. A number that crosses a threshold, in a person whose neck was always going to cross thresholds, in a scan done by a service that also sells the treatment, is a piece of information and nothing more. It needs weighing up carefully, by somebody independent.

What the surgery evidence actually shows

The operation for craniocervical instability is occipitocervical fusion, where the skull is fixed to the upper spine with rods, screws and a bone graft, sometimes with a reduction to correct the angle first. It’s major surgery and you can’t undo it, and the honest summary is that it can help carefully selected, genuinely severe cases, while the quality of the evidence sitting behind it is low.

Across everything anybody has managed to find, only 78 people with Ehlers-Danlos have actually had the operation ^[1]. That whole body of evidence sits at the bottom half of the scale, with nothing randomised or controlled anywhere in sight ^[1]. The conclusion drawn from it was cautious and specific. Surgery should only be undertaken where the scan clearly shows instability and the symptoms and signs match it, and the field needs agreed guidelines it doesn’t yet have ^[1]. That’s a long way from the confident promise you meet online.

Now, look at the individual outcome papers and you can see why the caution is warranted. In one of the more recent ones, roughly three quarters of the people followed up after a fusion said they felt better overall, and their headache and neck pain scores came down ^[7]. Encouraging, until you notice the rest of it. Nobody was looking forward in time, there was no comparison group, most of the outcomes came from the people themselves, and they were only followed for a bit over a year on average ^[7]. And that isn’t our verdict, it’s theirs: looking back at people they’d already operated on, with nobody to compare them against, doesn’t validate the reasons given for operating, and what’s needed instead is a proper trial set up in advance, across several centres ^[7]. When the surgeons who ran the series ask for a better one, take them at their word.

The series with the longest follow up tracked twenty people, over five years ^[8]. It’s worth laying out what it found, as the headline and the small print point in rather different directions.

– The bones: genuinely corrected. The clivo axial angle came up into the normal range, and every one of them came back normal on the Grabb Oakes measurement too ^[8].

– Still better at five years: headaches, dizziness, balance and walking ^[8].

– Not better at five years: neck pain. Nor muscle pain, fatigue, memory or vertigo ^[8].

– The hardware: 40 percent of them asked to have it taken back out again, a year or more after the fusion, because the tissue over the screws had thinned and it hurt ^[8].

That last one is worth reading properly, as it isn’t the fusion failing. It’s the hardware sitting too close to the surface. The same surgeons have since moved to lower profile instrumentation, set as low on the skull as they can manage, so there’s more tissue over the top of it ^[8].

None of that tends to appear on a testimonial page.

Some of the studies don’t even measure whether people felt any better. One comparison of two fixation techniques reported only the angles on the scan, and never captured a single symptom score from the people themselves. So it can tell you the bones ended up in a similar position either way, and absolutely nothing about whether the surgery helped anybody ^[9].

The biggest recent paper, the one you’ll see cited as a 347 person series, isn’t really about outcomes at all. It’s a screening protocol, describing how one surgical service decides who to operate on. Only a minority of those referred met the full surgical criteria, fewer again went on to have surgery, and the results rest on how satisfied people said they were ^[10]. To their credit, those authors state outright that without a sham controlled comparison they can’t rule out placebo and expectancy effects, and that their work is preliminary ^[10]. When even the surgeons publishing the thing say that, be very suspicious of anybody quoting 347 at you as though it were a success rate.

Genuine, severe, mechanically demonstrable craniocervical instability with matching neurological signs is a real thing, and for those specific people surgery by an experienced team can be the right call and can genuinely change a life. For everybody else, the evidence base is thin, and prone to exactly the biases that make weak treatments look strong. Selected groups, no comparison, outcomes people report themselves, and a very natural tendency to feel better after finally being taken seriously and having something done about it. That doesn’t make the surgeons villains, it makes the evidence what it is, and with surgery you can’t undo on the table, weak evidence is a reason to move slowly rather than quickly.

One more thing worth naming, as it colours a lot of what you’ll read online. A good chunk of the content out there is produced by clinics that both diagnose and treat, whether the treatment is a fusion or injections marketed as a way of tightening the ligaments up. Where the same business performs the scan, sets the threshold and sells the fix, the incentives aren’t neutral. The injection therapies in particular tend to arrive with very confident before and after images, so ask what trial sits behind them and see what you get handed. We approach every treatment the same way we do in our piece on pain medication for hypermobility and EDS: what does it actually do, what does it not do, and who’s doing the telling.

Cervical dizziness

A close cousin of the CCI conversation is the idea that your neck is directly causing your dizziness, which gets called cervical dizziness, or cervicogenic dizziness. It’s worth a short detour, as the same pattern of overclaiming applies, and the top experts genuinely don’t agree it’s a diagnosis anybody can confirm.

The Bárány Society, which is the international body for vestibular research, published a position statement about as blunt as these documents ever get. There’s no accepted consensus diagnosis for cervical dizziness, there’s no agreed diagnostic test, and the society declined to recommend any specific diagnostic criteria or any specific therapy ^[11]. Others have reached the practical version of the same conclusion. Cervicogenic dizziness is a diagnosis of exclusion, meaning you only land on it after ruling everything else out, and a lot of the treatment literature is poor quality ^[12]. So if a clinician tells you with total confidence that your neck is the definite cause of your dizziness, and that they’ve got the definite fix for it, they’re a good deal more certain than the whole field is.

There’s a real mechanism sitting in the background though, and it’s the one that leads into the useful part. Your neck is stuffed with position sensors, and the brain leans on them heavily to work out where your head is and which way is up. That link isn’t fringe, it’s basic neurology, and it’s exactly why the neck can influence balance without anything being mechanically unstable.

The guarded, poorly mapped neck

Your neck is unusually well wired for position sense. The main sensors up there are the muscle spindles inside the neck muscles rather than the joint capsules, and their density is much higher than in the muscles of the shoulder or the thigh ^[14]. Those spindles tell your brain where your head is and how fast it’s moving. That information doesn’t only help you turn to look at something, it feeds straight into your balance and your sense of which way is up. In healthy volunteers, simply disturbing neck proprioception with muscle vibration is enough to shift where people think they’re moving in space, nudging their whole sense of orientation off course ^[13]. So the neck really does help steer the ship, and it does that through sensory information as much as through holding bones in place.

Hypermobility bites right here, though not quite in the way it usually gets told. The popular version goes: lax ligaments, therefore poor feedback. That’s too quick, as the neck’s main position sensors sit in the muscles rather than in the ligaments ^[14]. What has actually been measured is that people with chronic neck pain have moderately impaired joint position sense compared with people who don’t, and that the size of the error tracks the intensity of the pain ^[14]. That’s the finding. What comes next is our reading of it rather than a result, by the way. A neck that’s spent years braced is a neck whose muscles report from a narrow, unvarying range, and a brain working off a narrow signal has a great deal less to go on.

It’s the same territory we cover for the rest of the body in our articles on brain maps and proprioception and on how the nervous system builds reliable movement. Both build on the basics in hypermobility and exercise part 1. And stiffening is what bodies tend to do when they aren’t sure. Brace the area, recruit the big superficial muscles, clamp the lot down, as a joint that’s clamped can’t surprise you. In the neck that shows up as the constant ache, the tightness, and the feeling that your head is heavy and your shoulders have ended up somewhere near your ears.

This is the same tight and weak paradox we get into in our piece on why hypermobile muscles feel tight and weak at once. The muscles aren’t tight because they’re strong, they’re tight because they’re working overtime to make up for information the brain isn’t getting cleanly. Neck position sense is measurably worse where there’s neck pain, through a mix of pain switching off the deep stabilising muscles and altered signals from the tissues. The standard advice that flows out of that research is to retrain the sensory and motor control, rather than only strengthen or hold still ^[14].

There’s an emotional layer stacked on top of all this, and it isn’t woolly at all, it’s mechanical. A nervous system acting on noisy information from a neck that feels unstable runs hot, stays braced and stays alert. That feeds directly into the anxiety loop we describe in our work on why hypermobile bodies are wired for anxiety and on fear of movement. Being told your skull might be dangerously unstable pours petrol on exactly that fire. The guarding gets worse, the symptoms get worse, and “my neck is broken” starts quietly confirming itself.

What actually helps most people

If the problem for most people is a lax, poorly mapped, over guarded neck rather than a mechanically unstable one, then the job is reasonably clear. Those with hypermobility need to give the brain better information, and teach it to trust the area again. For the neck that runs in the same three stages we use everywhere else: sensory clarity first, then tone regulation, then motor learning.

Sensory clarity means feeding the brain cleaner input before you ask it for hard work. Slow, controlled head movements where you actually pay attention to where your head is. Light touch and tactile cues around the neck and jaw. Gentle repositioning tasks, eyes open and then eyes closed, so the brain has to lean on the neck sensors rather than defaulting to vision. It isn’t glamorous and it’ll never trend on social media, but it’s the foundation, as you can’t control a joint your brain can’t locate.

Tone regulation is the switch off before the switch on. A neck that’s spent years bracing needs to learn to let go before it can learn to work properly, otherwise every exercise just adds more clamping on top of the clamping that’s already there. This is where a lot of well meaning strengthening goes wrong. Load a guarded neck with heavy work, without first calming the guarding, and you tend to reinforce the very pattern you were trying to change.

Then comes motor learning, which is the retraining of the deep control muscles and the coordination between them. One of the most studied versions of that for the neck is craniocervical flexion training, the gentle nod that biases the deep neck flexors instead of the big superficial ones that love to take over. Pooling the neck exercise trials together gives moderate certainty evidence that exercise of this kind produces real neural change in the neck muscles. That includes reduced activity in sternocleidomastoid, which is one of the superficial muscles doing so much of the guarding ^[15]. That finding is about muscle activation though, and not about pain, which changes what you should expect from it.

On the wider question of exercise for neck problems, the evidence is genuinely helpful and genuinely modest at the same time. The big Cochrane round up of exercise for mechanical neck disorders concluded that specific strengthening work across the neck, shoulder and shoulder blade does reduce pain and improve function. Its own summing up was rather less comfortable than that sounds though. No high quality evidence was found, so there is still real uncertainty about how effective exercise is for neck pain ^[16].

Proprioceptive training has a reasonable track record across a lot of different populations. Pooling dozens of separate trials of it gives average gains of a bit under a half in proprioceptive function, with motor performance improving by much the same amount ^[17]. Most of those trials used active movement rather than something done to the person while they lay there, although the reviewers were careful to add that no single approach stood out as the best one ^[17]. None of it was specific to hypermobility, so take it as a principle worth applying rather than a promise about your neck. The principle does line up completely with what we see though: give the nervous system clearer input and make it work for that input, and control improves.

A quick word on collars, as they come up constantly. A soft or hard collar can feel like instant relief, and for a genuine acute injury or a confirmed severe case there’s a place for short term support. Living in one, for a neck that’s guarded rather than injured, is a very different proposition (and this next bit is our clinical experience rather than anything from a trial, by the way). What has been shown is that these muscles change how they fire in response to what you train them to do ^[15]. What we see in practice is that the longer somebody lives in a collar, the rougher the first couple of weeks without it tend to be. It’s the neck equivalent of the trap we describe with KT tape, useful as a temporary sensory cue and counterproductive as a permanent crutch. The same goes for endless end range stretching to chase the tightness, which for a lax neck is usually pouring effort into the wrong hole. It’s a theme we go into properly in our guide to exercise and rehabilitation for hypermobility.

If you want to see how the same logic plays out elsewhere, the approach is identical whether we’re talking about the neck, the knee, the core or the ribs. Give the brain a clearer map, calm the guarding, then rebuild control in a graded way. It isn’t as dramatic as an operation, which is probably exactly why it tends to work.

Red flags, and when it genuinely is serious

Everything above is about the large group of people who have a hypermobile, unhappy neck rather than a dangerous one. It’d be irresponsible to leave it there though. A small number of people do have something that needs urgent attention, and telling them to go away and do slow neck nods would be every bit as wrong as telling everybody else to get fused.

Genuine compression of the spinal cord, whatever the cause, produces a recognisable pattern that’s quite different from ache and fatigue. It tends to be progressive, and it shows up in lost function rather than in comfort alone. In degenerative cervical myelopathy, which is the age related kind, a large share of the people managed without surgery go on to get worse over the following few years. The reported range runs from a fifth of them to nearly two thirds ^[18]. That’s why genuine myelopathy gets treated as serious rather than as something to sit on.

Two caveats though, and both of them cut in your favour. That research is on age related degenerative disease rather than on Ehlers-Danlos, so the numbers don’t transfer to you. And in the same literature, roughly a quarter of a healthy population turn out to have cord compression visible on a scan and no myelopathy at all ^[18]. Compression in a picture isn’t the same thing as a damaged cord, so the useful part here is the clinical picture rather than the image.

The signs that mean stop reading and go and get assessed properly are these. New or worsening weakness in the arms or legs. Increasing clumsiness of the hands, dropping things, struggling with buttons or handwriting. Changes in walking or balance that are getting worse over weeks and months. Problems with bladder or bowel control. Electric shock sensations running down the spine or into the limbs when you move your neck. Difficulty swallowing, or new changes to your voice that persist, which we touch on in our piece on hypermobility and throat problems. Any of those, and particularly if they’re progressing, warrants prompt medical assessment rather than a wait and see approach.

That isn’t fearmongering, it’s the flip side of not overreacting. Take the rare serious presentation seriously, and stop treating the common harmless one as though it were that.

Where this leaves you

Craniocervical instability is real, and for a small number of people it’s genuinely serious and surgery is the right answer. It’s nothing like as common as the internet makes it feel though. The diagnosis rests on measurements that struggle to separate a hypermobile neck from a dangerous one, and the surgical evidence is weak enough that even the surgeons publishing it are asking for better studies. Meanwhile the far more common story is a lax, poorly mapped, over guarded neck sitting on top of a dysautonomic body. That explains most people’s symptoms a great deal better, and it responds to something considerably kinder than an operation.

So if you’ve got red flags, go and get properly assessed. If you haven’t, the work is slow and unglamorous, and it genuinely helps. Clearer sensory input, less guarding, better control, and a nervous system that slowly learns it can trust the top of your spine again. And if you want somewhere to start on the wider picture, try our guides on how hypermobility and EDS are diagnosed and on what gets mistaken for hypermobility. Both are a considerably calmer read than an upright MRI clinic’s testimonial page.

The Fibro Guy

References

[1] Lohkamp LN, Marathe N, Fehlings MG. Craniocervical instability in Ehlers-Danlos syndrome: a systematic review of diagnostic and surgical treatment criteria. Global Spine Journal 2022;12(8):1862-1871. doi: 10.1177/21925682211068520

[2] Demmler JC, Atkinson MD, Reinhold EJ, Choy E, Lyons RA, Brophy ST. Diagnosed prevalence of Ehlers-Danlos syndrome and hypermobility spectrum disorder in Wales, UK: a national electronic cohort study and case-control comparison. BMJ Open 2019;9(11):e031365. doi: 10.1136/bmjopen-2019-031365

[3] Gensemer C, Daylor V, Nix J, Norris RA, Patel S. Co-occurrence of tethered cord syndrome and cervical spine instability in hypermobile Ehlers-Danlos syndrome. Frontiers in Neurology 2024;15:1441866. doi: 10.3389/fneur.2024.1441866