PDA

View Full Version : Question?



frecklesmom
2nd November 2008, 12:52 AM
I keep trying to wrap my mind around SM. Sometimes have more questions than necessary :rolleyes:.
In the percents-is this correct or have I gone off.
Using 100 Cavaliers
98 will have OH/CM 98%
30-70 will have syringomelia 30-70%
Of the 30-70, 60% will be symptomatic (18-42)

Karlin
2nd November 2008, 02:39 PM
No, those would be a bit high.

The lowest number for CM seen in MRI samples is 85% (Geoff Skerritt) and Clare Rusbridge says at least 90%, but the figure is most likely higher given the enormous difficulty they have had in finding cavaliers without CM. One of the reasons a neurologist panel is being convened in the UK is to address issues over whether there is CM on a given scan, as some neurologists are saying there's none whereas others feel that these are not dogs clear of CM. It can be very hard for neurologists less familiar with the malformation to diagnose it. Also the panel will resolve disputes over the readings of scans/grades. It is a fact that researchers were unable to find enough dogs clear of CM to serve as a clear control group for the genome scan in Montreal and had to go to an entirely different breed that also has SM showing up but also has many fully CM/SM-clear dogs and lines (griffons).

The estimate of 30-70% with SM is correct -- 30% (from the North Carolina study) is actually low as it excluded dogs with what they termed hydromyelia but many other researchers would view that as the start of SM so the division is a little controversial. Add in those dogs and the figure pushes up around 40% but researchers have gone with their conservative 30%. However these were all dogs under 5. Clare Rusbridge had 70% with SM in her initial sample, which included older dogs. However researchers tend to use the minimum percentage as an estimate.

Clare Rusbridge says of the dogs she sees with SM, 30% are symptomatic I believe -- but I will check that figure. I may be recalling it incorrectly.

Karlin
3rd November 2008, 11:36 PM
Found the reference -- 35% of dogs which MRI with SM show clinical signs:

http://sm.cavaliertalk.com/diagnosing/diagnosing/infosheet.html


Recent studies suggest 35% of SM-affected dogs have clinical signs of the condition. The youngest reported dogs with SM have been 12 weeks old. Dogs may be presented at any age although the majority of dogs (approximately 45%) will develop first signs of the disease within the first year of life and approximately 40 % of cases have first signs between 1 and 4 years old. As many as 15% develop signs as mature dogs with the oldest reported case first developing signs of disease aged 6.8 years. Due to the vague nature of signs in some cases and lack of awareness about the disease there is often a considerable time period (mean 1.6 years) between the onset of signs and confirmation of a diagnosis.

frecklesmom
3rd November 2008, 11:57 PM
Thank you, Karlin :). You can find anything :thmbsup:

So, of 100 Cavaliers

85-90 will have OH/CM 85-90%
30-70 will have syringomelia 30-70%
Of the 30-70, 35% will be symptomatic (10.5-24.5)

Now on to understanding CSF flow :*gh:
It would certainly be easier if the human brain and "accessories" were completely understood.

CORRECTED 11/4/08 thanks,Karlin

Jan Bell
4th November 2008, 10:41 AM
:thmbsup: Thanks for this thread. Very useful and concise information.

Karlin
4th November 2008, 11:53 AM
98 will have OH/CM 98%

No, 85-90% are the current estimates. Clare Rusbridge uses the figure of a minimum of 90% and possibly higher.

Mark
5th November 2008, 12:35 AM
Is it correct that the North Carolina study was only of 59 dogs? Not sure any real conclusions can be based on such a small study.

Karlin
5th November 2008, 01:39 PM
Yes, obviously a single study too isn't a complete picture. But that is only one of many peer reviewed samples and the results there also roughly fit what Geoff Skerritt has seen in his MRI sample of over 1200 cavaliers.

It is unfortunate that the clubs and KCs have not pushed for nor funded larger studies but all researchers can work with are the samples they are funded for. One of the main US national clubs funded that NC study so it would be worth people asking them why they chose to stick with a sample that size perhaps?

Each of these studies has not had as its goal, a measure of the rate of affectedness alone, but have tried to get a sense of the affectedness of the sample. Each has had a specific research objective. I would recommend going and actually reading the abstracts for this study. The goal was to better understand CSF flow and its possible relation to syrinx development and pain. The affectedness levels were a byproduct but absolutely consistent with all studies.

There's a growing list of peer reviewed papers in prominent journals and papers presented at leading vet conferences, in which samples as in NC, were chosen where the majority of dogs showed no clinical signs of SM, but have all returned at minimum, 30% of dogs already with syrinxes -- full blown SM -- even in the NC sample where the dogs were quite young. Every single dog in the NC sample had skull abnormalities for example. The odds of selecting the only 59 cavaliers with abnormalities seems very odd indeed. This study was the first real wake-up call for US breeders because it was funded by their own major national club, and numerous breeders put dogs in for this study on the assumption that they would come back with clear dogs (breeders posting to the L-list over the months of the study made it clear this was their assumption). Most went away with MRIs showing things they did not expect -- at the very least, the documented skull abnormalities, high degree of cerebellar herniation (the brain forced out of the opening into the spinal cord), etc.

The fact that not a single study has been done that shows LESS that a 30 affected rate, and that these high levels are consistent across every study, suggests these are not isolated quirky pockets. In addition neurologists themselves have seen a steadily rising number of cavaliers, far beyond any other breed, presenting with the pain and discomfort of SM. I have spoken recently to Geoff Skerritt, who has now MRId 1200+ cavalirs, and nearly 9 in 10 he would diagnose with malformation of the skull and cerbellar herniation, with again significant numbers with syrinxes as well. He has stated in a letter that he finds the state of SM in the breed from when he first began working as a neurologist decades ago to be absolutely salarming. He has said his figures for MRIs with that massive sample are right in line with all the others.

As the KC rightly has noted, exact percentages are not the issue however -- it is the fact that so many dogs are affected. Even the most skeptical breeders that I have come across do not believe the levels are as low as the 1% of the 'health survey' by the KC . Nonetheless, how odd that the KC chooses to use an unscientific, self reporting owners survey done a few years ago in which almost no dogs were MRId as the basis for its accepted percentage, rather than the year after year now of scientific, peer reviewed studies in which enormous levels of affectedness within MRId dogs, mostly with no clinical signs, are repeated over and over? Why might it choose the non-studies rather than the studies, some funded within the clubs, to come up with its lower percentage?.

It is also worth reading more on the genome study. There are strong indications that the genes are extremely widespread and that breeding must only go forward with extreme care. The genes don't lie. Some 10,000+ pedigrees went into the genome research and hundreds of MRIs of a wide range of dogs. From talking to the researchers I know how difficult it has been to find ANY completely clear cavaliers -- clear of skull malformation/herniation; clear of syrinxes. It was relatively easy to find griffons -- not just individual dogs but entire families and lines -- fully clear of both. I am sure the researchers -- and breeders -- would be thrilled if a single indication could be found that there are whole clear lines of cavaliers and that the research samples over the past decade have all been anomolies, but they have not been able to find them despite a long search now.

I know neither researchers, nor breeders, feel they havethe luxury right now of finding exact percentages across a huge rage of cavaliers. They are far more concerned with the urgent issue of finding ways of addressing this alarming problem through better breeding practice -- before this is no longer possible to do. None of the geneticists and other researchers who have looked at this issue in this breed are in doubt that there is only a limited timeframe in which it can be addressed.

Mark
5th November 2008, 08:06 PM
I am one of these peer reviewers for scientific journals and it is remarkable how some studies get published. I have full text access to all vet/medical & neurological journals so am able to make a pretty reasonable assessment of the research. Personally I doubt that the figures are genuinely this high. Let's all hope so.

I think the KC may be seeing that as in the North Carolina study affected dogs are actually in the recruited sample so a high positive outcome is inevitable.

frecklesmom
6th November 2008, 12:26 AM
Mark, peer review can be so interesting. What is your specialty? :smile:

Mark
7th November 2008, 10:53 PM
I'm not sure that interesting is the right word:cffee: peer review really refers to the process of publishing research where independent experts in the field review the study methodology and data analysis to ensure that the conclusions are robust. It is, at least a system of sorts but it is flawed in so many ways. That said, the key is in understanding scientific research methodology and being able to judge the validity of the conclusions that the researchers are proposing. That means examining the entire research approach that they have taken and the statistical analysis i.e. were the stats tests appropriate and applied correctly.

It takes a good deal of experience and training to be able to see past the figures quoted in many papers and put them into context of the research methodology which can obviously be difficult for non-scientists/non-medics/vets.

The important thing is to question what is in the literature and identify good quality, well planned and sound research.

Ok chaps...wake up:rolleyes:

frecklesmom
8th November 2008, 12:00 AM
Many types of peer review-what you stated is one type examining form but it actually cannot examine the substance (only how the research was written up). Only an expert in the very specific field can lay question to substance.

frecklesmom
8th November 2008, 12:02 AM
Forgot to ask, Mark, what's your specialty ?

Mark
8th November 2008, 12:16 AM
What do you mean by substance? The write up should contain every aspect of the research in a manner that it could be repeated.

In terms of publishing research 'Peer review' only has one meaning which is an independent scrutiny of the research process and validity of the study. A reviewer needs knowledge of the research process moreso than to be an expert in the exact field in question. In most cases review by someone in the same field can cause the reviewer to be blinkered to the faults in the methodology or to be unfairly harsh in their review in the competitive field of research in which they are competing for recognition, I feel.

Mark
8th November 2008, 12:19 AM
Forgot to ask, Mark, what's your specialty ?

I review for clinical laboratory medical journals, my speciality is clinical and laboratory pathology and basic biomedicine (basic as in fundamental, not 'low level')

Karlin
8th November 2008, 12:20 AM
It takes a good deal of experience and training to be able to see past the figures quoted in many papers and put them into context of the research methodology which can obviously be difficult for non-scientists/non-medics/vets.


I agree -- but the people that are doing the research in this particular area are considered some of the best internationally in their field. Guy Rouleau, who heads the genome research, is one of the finest researchers in this particular area. Clare Rusbridge is held in very high esteem by the leading human SM researchers and physicians. Dominic Marino is one of the cutting edge neurosurgeons in vet medicine internationally, and so on.

Any of these folks would greatly welcome a massive study on incidence -- but who would fund it? So far, Dr Rusbridge has been almost entirely self financing and working with small grants and donations from people like many of us on the board -- yet is the preeminent expert in this particular area, working closely with experts at the Animal Health Trust, Cambridge University, etc. Club sponsored projects have generally been modest -- the largest grant to date is to Dr Blott, yet her results are largely dependent on the success of the Montreal genome research for any truly meaningful accuracy for her EBVs. The Montreal research has been given grants late in the game by the clubs in the US, but the original research again was funded by many individuals who gave blood and scans and this innovative project still needs financial and breeder/pet owner support to succeed (I am sending blood from three dogs next week as well as scans). The grants were spurred along by many individuals who wrote to the clubs to ask about the status of the grants -- and who donated to the specified grant funds themselves in order to show private interest. I would urge anyone who would like a huge study of MRId dogs to write and ask the clubs and KCs internationally to fund such a study as it isn't going to emerge out of the blue. For that matter the same sort of study would be useful for MVD -- but honestly, there's almost nobody questioning the serious rate of incidence in the breed and I think almost anyone involved in breeding or ownership of these dogs cares more about finding ways of addressing the problems rather than caring remotely whether the incidence is 1% 5% 30% or 70%. Every breeder I know who has scanned has had more than a 'low' incidence in their own dogs. I only know of one single line where more than one dog has scanned entirely clear. And sadly there are only a couple of dogs in that particular line. If incidence is low, then breeders would scan and find those clear dogs. They are not finding totally clear dogs. The people most insistent on the falsity of the suggested levels of incidence are those who do not scan. Half of my four have SM -- and the affected dogs were both diagnosed at under 5. That seems pretty typical amongst those willing to discuss their scan results.

Peer review is of course open to flaws, like any area of human overview of research, but what is the alternative? It is the best method of independently assessing research, and Clare Rusbridge's work amongst others has been peer reviewed in some of the leading international vet journals.

Mark
8th November 2008, 12:49 AM
I think we should be caring about the incidence, particularly when so many owners are worried sick by the figures they hear. Their worry affects how they interact with their puppies and dogs and how fulfilling their relationship is. Pet owners do care about the figures as forums such as this are constantly seeing these queries from worried owners. So I disagree that it isn't important.

I don't dispute the reputation of any of the researchers mentioned but there is no way the research of anyone should ever go unquestioned just because they are renowned in the field. It is not a reflection on the researchers, any research degree will actively encourage researchers to sit and brainstorm and criticise each others work. The researchers mentioned aren't offering these figures on incidence as part of published work anyway since they know can't provide them at this stage.

I'm not sure that there being only one line with more than one dog scanned clear is correct. Then again that information isn't really available for anyone to make a judgement. Neither really can anyones individual experience with SM in their pets be a marker of incidence. Many people on this board have multiple cavs who are asymptomatic for SM, does this mean that the incidence is zero? No, but put like that someone could argue it.

Karlin
8th November 2008, 04:26 PM
Mark, your comments have suggested a questioning of the skills of the researchers on my reading and that of others, and the reason I point out the reputation of the individuals not just in their own field but across into human medicine as well is to emphasise that some of the leading experts in other areas have also read this work, spoken with these researchers, and trust the figures they are getting. Thousands of illnesses have levels of incidence measured not from vast studies but from smaller studies so this is hardly unusual to do this especially in the case of a condition that is expensive to actually assess. Nonetheless I'd say the overall sample now is running close to 2000 MRId cavaliers if you add in all the studies plus the work of Geoff Skerritt who has scanned over a thousand dogs, many hundreds of those for breeders, not for people bringing in sick dogs. Geoff confirms at least the minimal published levels of incidence amongst those that arrive in with no clinical symptoms, presumed clear. MVD is measured the same way and I can definitely confirm those estimates are accurate in the dozens and dozens of cavaliers I take into rescue and have heart checked. Almost every cavalier age 5 or older has a murmur.

Did you attend any of the SM conferences so far? Have you read the existing studies? Are you following with current progress on the genome work? Spoken to Sarah Blott? I'd respectively suggest that actually, it isn't all that hard to get information as extensive notes, abstracts, publications and CDs of the Rugby talks, for example, are all easily available. I recommend Rod Russell's extensive summary of all current research and publications at his cavalierhealth.org site.

I do know for a fact that there is only ONE line with any significant degree of fully clear dogs -- which still I believe is only three -- with the key scanning dog now elderly and beyond reproducing. Unless there are breeders keeping their clear/clear lines secret from everybody, both other breeders and researchers, believe me, there are no fully clear lines. It would be so exciting if there were even one, but so far, there are none.

For going on a decade now many of us with affected dogs have listened to some breeders denying incidence; some of these being people that I know for a fact have threatened pet owners with legal action if they reveal their affected dog of that line. Meanwhile every neurologist can attest to huge numbers of cavaliers presenting with this disease; I have talked directly to several of them who see this condition constantly in cavaliers. I know of breeders that have had to completely end their lines and start over because SM was so widespread amongst nearly every dog scanned; I know of people with more typically, about a fourth to half of dogs showing syrinxes. These are dogs all related to all the significant lines, as almost every cavalier is closely related anyway. Unless all the scans on all those dogs are catching some serious anomoly then I think it is quite safe to say that existing studies are accurately reflecting incidence in wider populations. We all hope these are somehow flukes and it will not be the case.

I know you would definitely be happier over on the CKCS-SM list discussing these issues though, so I am going to suggest you move your discussions on this issue over there -- this forum is really addressed to giving support to those with questions about SM and with dogs suffering from it, not to have arguments on level of incidence (that is the type of discussion that the CKCS-SM list was set up for). Even the KC and CKCS Club have said it is pointless to keep returning to such areas of disagreement when all accept that too many cavaliers are suffering and the level of incidence is by any means of calculating it, now alarming -- and on a forum like this, with primarily a support focus and dozens of owners of affected dogs living with pain, your insistence on returning to this here is increasingly insensitive. Thank you for your contributions to the board, but your particular interests in the subject would be more appropriate in other places.

Karlin
9th November 2008, 06:52 PM
I've had a predictably childish and silly private reply from 'Mark' which pretty well revealed their own agenda and also indicated to me who this actually was. :rolleyes:

Why some people are obsessed with being on this board, I don't know -- :rotfl: -- but 'Mark', you need to get out a bit more I think!

Maybe 'Mark' is a bit jealous of well regarded researchers? icon_nwunsure