A Detailed Look at Chiari Malformation and Syringohydromyelia
I recently (10-04) had the chance to attend a seminar on syringomyelia at the Cavalier King Charles Spaniel -- USA club show in St. Louis. The seminar was offered by a neurosurgeon (Dr. Coates) and neurologist (Dr. O'Brien), both associated with the prestigious veterinary hospital at the U of MO at Columbia.
There was some very interesting and important information offered in this seminar, including an important take-home message for breeders, about which more later.
First: syringomyelia is a clinical state caused at least in part by a skull deformity similar, but not identical to, what is called a Type 1 Chiari ("kee-ar-ee") malformation of the skull when it occurs in humans. This malformation results in a shortening of the back of the skull, leaving less room for the back end of the dog's brain and particularly for the cerebellum. The cerebellum is pressed inward and indented by the skull and may press outward through the hole at the back of the skull, the foramen magnum.
The foramen magnum is the hole through which the spinal cord leaves the skull and enters the spine. Pressure exerted by the cerebellum through the foramen magnum varies with the level of excitement of the dog: when excited, the brain sort of pulses. With the cerebellum pressing rhythmically into the foramen magnum, spinal fluid is forced into the spinal cord at an accelerated rate. This creates areas in which the central canal within the spinal cord becomes greatly enlarged. Each such area is known as a syrinx, and it is these hollowed-out fluid-filled areas that are visible on an MRI (along with the indented cerebellum). The interference all this causes with nerve transmission results in the clinical signs of scratching, pain, apathy, and poor coordination we call "syringomyelia."
Now, does your dog have syringomyelia?
Here is what Dr. Coates indicated during the seminar, and also to me individually afterwards:
1. Nearly all dogs with SM experience periods of dull behavior in which they appear to be "having a terrible headache" or otherwise to be in pain. This is clearly noticed by the owners and clearly observable by experienced vets (experienced with this condition). Dogs with SM often have more subdued, quiet personalities.
2. A large percentage, but not all, of affected dogs show abnormal scratching behavior. This behavior involves persistent scratching motions directed by a back leg towards the neck or shoulder region, motions that often do not make physical contact with the neck or shoulder. Most often this behavior is directed only towards one side of the neck. Scratching behavior is most commonly elicited by petting or rubbing the neck region, by pressure from a collar on the neck, by sudden changes of neck position, and by excitement. Affected dogs frequently show this behavior while barking. They frequently show it while moving and most typically do not stop moving in order to scratch, but make scratching motions while trotting. Dogs may curl their bodies into a "C" while trotting because they are persistently making scratching motions while moving. A dog may turn in circles and rub its face on the floor. The behavior is described as "paroxysmal" or "involuntary," which implies it is very difficult to interrupt scratching or redirect the dog's attention. Scratching episodes may last for extended periods of time -- more than an hour. A dog may scratch a raw spot on its neck. Severity of this behavior varies a great deal from dog to dog.
3. Dogs often but not always show evidence of neck pain. They often resent being touched, petted, or groomed near the affected area. An affected dog may hold its head down at an odd angle or tuck its head in towards its chest. It may hold its head at an abnormal angle to eat or sleep.
4. Dogs may yip or scream for no apparent reason. Pain may be severe (perhaps 20% of affected animals?) and in this case the dog may show extended bouts of screaming or may go into a corner and shiver. If more minor, the dog may simply acquire a reputation as a "baby" or a "wimp."
5. Dogs may have seizures (about 20%?).
6. Dogs frequently show balance problems. Clinically, they respond abnormally if lifted to stand on their hind legs, if "wheelbarrowed" (rear lifted and dog pushed forward to walk on front legs only), or if pushed sideways. They frequently show a sitting posture in which the front legs are very close together and the feet turned slightly outwards.
7. Dogs may not seem to know where their feet or legs are. If a front leg is lifted and the foot turned under, the dog may be willing to stand that way, with the foot turned under, as though it does not recognize that it is not standing normally. The limbs may be weak. Probably a combination of these things cause the dog to stumble and fall frequently. Dogs affected in this way are often lame, but the lameness is not typical of pain in the leg or foot -- rather of confusion about where the limb is. The gait may be strange -- wobbly or rolling.
8. If a dog has shown a lot of scratching behavior at a young age, it may develop scoliosis from the constant bending in a circle.
9. There may be hearing loss. This may be common -- it is not usually evaluated, but is not unusual with humans who have the Chiari malformation.
10. Pain medications and drug therapies (glucocorticoids) do not usually produces satisfactory long term amelioration of severe or progressing symptoms, and never resolve the problem. Surgical treatment, in which part of the backskull is removed to enlarge the foramen magnum and give the brain more space, and part of the first cervical vertebra is also removed in order to provide still more room for the previously compressed brain, is quite successful in producing long-term improvement, although it cannot return an affected dog to a normal state. The surgery is not considered difficult and as it becomes more recognized and available, the euthanasia rate for this problem (currently about 30%) should decline.
Now. How do you know for sure, if your dog shows only one or two minor expressions of these signs? Perhaps she sometimes shows scratching motions that do not make contact with the body and has a quiet personality. Perhaps she seems clumsy and occasionally you wonder if she might "look like she has a headache." Probably she occasionally stumbles. No doubt she now and then yips in pain for no reason you can see (my dogs yip when they have a spiky burr in their feet!). Perhaps the dog is limping and you thought it had pulled a muscle, but now . . .
SM info sites will tell you that this problem can only be diagnosed for sure by an MRI.
This is not quite accurate. Here is what we seem to be seeing in Cavaliers:
Most, perhaps the vast majority, of Cavaliers seem to have a Chiari-like malformation of the backskull. This is probably typical of most or all brachycephalic breeds, such as Pugs, Pekingese, English Toy Spaniels, etc. It may also be true of minimally-brachycephalic breeds, such as Am. Cocker Spaniels. The malformation is apparently also not rare in Chihuahuas and Maltese, and perhaps other toy dogs.
The painfully few normal Cavaliers that have been MRIed so far often or usually show a Chiari malformation, usually with at least small syrinxes (syrinces?). Some have no clinical signs of syringomyelia other than perhaps a quiet personality. Some have no signs at all.
Karen Ostmann (Sheeba Cavaliers) referred to a study in which 100 Cavaliers were MRIed and 97 showed a Chiari malformation -- the other three were checked using inferior equipment and cannot be definitively cleared. I did not have a chance to get more information about this study.
The vast majority of Cavaliers do not show clinically abnormal behavior consistent with syringomyelia. If most or virtually all have a Chiari malformation of the skull, then this malformation turns into a skull trait of brachycephalic dogs. If most have subclincial syrinces that do not interfere with their behavior, then the MRI is not a useful diagnostic tool. It provides accurate but useless information.
Saying that a dog with syringomyelia has a Chiari malformation and syrinces appears to be like saying that a lame dog has four toes on its back foot. The statement is true but pointless: all normal dogs, whether lame or not, have four toes on each back foot. All Cavaliers . . .
What this means for the pet owner whose dog shows possible SM symptoms, and who is therefore wondering what to do, is that an MRI may not be useful under any circumstances, unless a neurosurgeon refuses to operate without one, in which case it is merely a bureaucratic requirement, and a very damned expensive one.
What this means for the breeder is that there is at present no diagnostic tool that can separate affected dogs from unaffected. All the breeder has to go on is the behavior of the dog and its siblings, and its parents and their siblings . . . there is no point in going to the expense of MRIing a dog whose sibling has clinical syringomyelia, because the knowledge that the dog has a Chiari malformation does not add to the breeder's knowledge about that dog. The important information is that the dog's sibling is affected. One may then proceed on the assumption that the dog presents a breeding risk in the same way that the she would if her sibling were affected by any polygenic problem.
The next step:
It seems obvious to me that it is vitally important to MRI entire families of Cavaliers, along with Pugs, Chihuahuas, Pekes, Cockers, and perhaps German Shepherds or some other mesocephalic breed, in order to get a real idea of the prevalence of the Chiari malformation in severely-brachycephalic, moderately-brachycephalic, and normal-headed dogs.
It seems obvious to me that these families should include animals showing a spectrum of syringo-consistent behaviors, and also include a lot of clinically normal dogs. It seems quite clear that if the Chiari malformation itself is not sufficient to cause clinical presentation of syringomyelia symptoms, then a really serious effort needs to take place aimed at finding out what does separate clinically-healthy dogs from clinically-affected animals. Causes could include genetic, developmental, and / or environmental factors. Shy, nervous animals are a lot more likely to bloat than calm, confident animals. How about personality factors in this case? There are a lot of possibilities, and somebody needs to do some brainstorming and then some research.
DNA samples need to be taken from these families in which clinically-normal and clinically-affected animals can be found. The "normal" animals assessed first with MRIs and then assigned, via a rigorous method of evaluation, to some useful category: clearly completely normal, probably normal, borderline normal. The "affected" animals can be assigned to categories of their own: minimally, moderately, or severely affected. Then molecular geneticists and others can try to sort out what traits and genes and environmental factors correlate with clearly normal versus clearly affected phenotypes.
I'd probably design categories something like this: Completely normal -- past six years of age; vigorous, active, confident attitude relative to the average Cavalier; never makes scratching motions without making contact with body; no association between scratching behavior and excitement or neck pressure; no evidence of neck pain; clinically normal balance; sound movement. In contrast, a probably normal animal might be more like this -- same as above, but a quiet or subdued personality. A borderline animal might be like this -- same as above, but with a quiet or subdued personality and occasional scratching behavior that might be consistent with SM (or might be normal). Affected categories would probably be easier to define. What's important is to compare perfectly normal with clearly affected animals when trying to first figure out what separates the two.
And in the meantime, a note of extra caution for breeders, if it turns out that virtually all Cavaliers have the Chiari malformation:
There are two ways to make sure the cerebellum has ample room in the backskull, if avoiding the Chiari malformation becomes a (possibly premature) breed-wide goal. The first is to extend the backskull, changing the head shape. This will require breeding towards a different style of head than is currently preferred, and possibly campaigning to change the standard of the breed.
The second is to shrink the cerebellum. I suspect this would not come without a cost. I would not like to see what might happen if breeders breed hard away from the Chiari malformation without also accepting a significant change in head type. I believe that at the moment, evidence suggests that the breeders should focus on clinical signs, not on MRIs, when making their breeding decisions. I also suspect this problem should be treated as polygenic for the purpose of making breeding choices.
There are more links below, but here are a couple to start with.
Here's a more extensive description: http://www.thecavalierclub.co.uk/syringo/syringom.html
And here's a case history with links: http://website.lineone.net/~malburley/