gribbon wrote:775 examination on 412 dobermanns
it is a very small number!
RESULTS: DCM prevalence in various age groups was as follows: age group 1 (1 to <2 years) 3.3%, age group 2 (2 to <4 years) 9.9%, age group 3 (4 to <6 years) 12.5%, age group 4 (6 to <8 years) 43.6%, and age group 5 (>8 years) 44.1%. The cumulative prevalence of Doberman Pinscher cardiomyopathy was 58.2%.
Cumulative prevalence is NOT average.In another words that DOESN'T mean what you claimed at another post : " Some studies I have read are sure that DCM are the cause of 57% of deaths in the Dobermann. "
I told you that the dcm is in the genes of the breed.It was also in the past.Please don't say to me about ages of death of dogs from the past in comparison to the modern dogs.There is no sense because there is a variety of reasons which lead a dog to death and mostly because for every dog which you will mention,i can mention to you another one from the modern era who lived long.But this kind of discussion makes no sense.
.....to improve the breeding.We need a genetic test and not fears and rumors.
Epidemiological measures of the frequency of a disease in a population can be evaluated from studies of Prevalence or Incidence. “Prevalence” is a measurement of all individuals affected by the disease at a point of time in all population, whereas “Incidence” is a measurement of the number of new individuals who contract a disease during a particular period of time in the risk population. “Accumulate Prevalence” is a measurement of the proportion of all individuals affected by the disease during a particular period of time. Risk population only includes healthy dogs at final time of the study (dogs died are not at risk).
DCM in Dobermann is a disease with variable Prevalence and variable Incidence over time: risk period for developing the disease can be between juvenile ages to most aged dogs (8 old). Disease is mortal in all cases and average time of survival is short. With these characteristics of the Dobermann DCM, most adequate method to evaluate the frequency of the disease in the population is the study of Accumulated Prevalence over a long trial period (5 years in this study).
Most accurate evaluation of risk of the DCM disease in the Dobermann population is given when we calculate the Accumulated Prevalence dividing the” total number of all affected dogs” of all age classes during the trial period, not by the total number of dogs but by the “total number of healthy dogs”, not including those with the disease, since they are not at risk of developing it.
Is in this case Accumulated Prevalence representative of the average risk of DCM for the subpopulation of European Dobermann?
Yes! If the sample is representative (and it is), accumulated prevalence (58.2%) is the average risk. A prevalence of 58.2% for DCM means that now in a random group of 1000 healthy European Dobermanns, they are a most probable distribution of 582 of dogs that will be affected by DCM at some point in their lives, and 418 dogs not affected. It's that simple!
Is it enough and representative sample to obtain an accuracy scientific conclusion?
I guess so. A level of statistical significance P<0.05 does not seem a lax level of significance. It is the most used significance level in veterinary epidemiological studies. Otherwise 5 years is a long enough time to obtain an accurate measure of DCM risk. Geographical distribution of the sample including dogs from Germany, the Netherlands, Austria, Switzerland, Italy, and some eastern European countries seems representative also.
But you are right, 412 Dobermanns can be a small number; a sample of 4000 is better and a sample of 40 is worse. Obviously, the main way to reduce the random error is increase the size of the sample. The appropriate size of a sample in an epidemiological study is not chosen at random
but are calculated based on statistical formulas which relate the size of the sample with trial variables as level of statistical significance requested, probability of not detecting a real error, relative size of groups compared, etc.
Your argument is very clever but biased and confuses.
If I deny the representativeness of the sample size shredding the validity… I remember you that scientific protocols are strict, and expert scientific referees evaluate the validity of the results, all methods (included the statistical methods and representativeness of the sample size), consulted bibliography and adequate relation of the discussion and conclusions topics with experimental data and cited bibliography before it publication in a scientific journal. The Journal of Veterinary Internal Medicine is not a “gossip magazine” it is the official publication of the American College of Veterinary Internal Medicine, the European College of Veterinary Internal Medicine, the European College of Veterinary Neurology, and the European College of Equine Internal Medicine, with a vey prestigious Editorial Board and an impact factor of 2.168, one most higher between Veterinary Sciences Journals.
Is the data of Accumulate Prevalence for DCM of European Dobermanns 58.2% all the truth?
All statistical data ere subject to a degree of uncertainty but surely it is more near the exact value that our erratic speculations. In any case there appears a Dobermann forum the most appropriate place to discuss the validity of a scientific paper between neophytes. We would have to be grateful to the authors for their contribution, although their findings are an “inconvenient truth” for many people. When numbers appear fade fables.
Data of prevalence reported in the Discussion by the same authors for dogs from the United States or Canada vary between 45 and 63% (average between extreme values =54%). Is it a coincidence? Surely not: same founders + same bottleneck + same selection method on popular sires (although different popular sires) = same result. A catastrophe! Probably causative gene(s) are ancestral genes concentrated during selection (not recent mutations) but this matter little now. I think that the appropriate response to these data is to question our methods and strategies, not the validity and meaning of a scientific data
What about interbreed risk of DCM?
Interbreed quantitative relative risk of DCM is not documented in the bibliography. There are some quantitative data not corrected for the risk population of each breed (the number of live individuals of every breed, variable according to the popularity of each). In the paper of Petric et al. (Dilated Cardiomyopathy in Doberman Pinschers: Survival, causes of death and a pedigree review in a related line. J. Vet. Cardiol. 2002; 4:17–24
) there are some data restrained to a territory (Slovenia). These data can give us some perspective of how things. Selecting data in this paper only for breeds ordinarily considered as “breed at risk” for DCM in the literature (Dobermann, Boxer, Great Dane, Newfoundland and English Cocker Spaniel). Dobermann accumulate more risk (almost double) than all others breeds at risk. Newfounland is not a popular breed but all others can be considered with a similar popularity. Approximate conclusion but it indicates a disturbing tendency.
What about selection of the progenitors?
The DCM is an ugly disease: autosomal dominant inheritance with reduced penetrance (Meurs et al. A prospective genetic evaluation of familial dilated cardiomyopathy in the Doberman pinscher. J Vet Intern Med. 2007; 21(5): 1016-20
). Incomplete penetrance means that not all genotypes content the causative gene(s) exhibit DCM phenotype. More is reduced the level of penetrance more important is the percent of “silent carriers” in the not affected population.
Incomplete penetrance diseases is most likely a result of the way of a gene interacts with other genes as well as environmental factors of endogenous origin (for example sex hormones, immune system, etc) or exogenous origin (for example stress, food, etc). For simplicity and to understand the prospective risk for DCM in the selection of the progenitors, we can think in an inheritance model of a “major dominant gene D” responsible of DCM in Doberman and analyze it as a monogenic Mendelian trait. Just some examples:
First scenario: Penetrance 1 for DD homozygote and 0.5 for Dd heterozygote genotypes. In this scenario only rest an approximate 17.6 % of free dd genotype in the population; p(D)=0.58; q(d)=0.42.
Second scenario: Penetrance 0.75 for DD homozygote and 0.5 for Dd heterozygote genotypes. In this scenario only rest an approximate 8.8 % of free dd genotype in the population; p(D)=0.70; q(d)=0.30.
Another time approximate but disturbing. Even when a test is available we will certainly have to use heterozygous silent carriers Dd*, otherwise breeding population (dd) would be very small.
Paper of Petric et al. (Dilated cardiomyopathy in Doberman Pinschers: Survival, causes of death and a pedigree review in a related line. J Vet Cardiol 2002; 4:17–24)
show that occult DCM cases (sudden death) are half that previously diagnosed DCM cases (congestive heart failure). Also diagnosed DCM cases (congestive heart failure) are more frequent in the more aged dogs (average =7.1 years old) and sudden death DCM cases are more frequent in dogs that have died younger. In this case, sample is restricted to a group of 21 affected Slovenian Dobermanns, but data may be indicative: half of all cases are occult to the ECHO-EKG Holter scanning diagnostic.
A most detailed analysis of data of accumulate prevalence from West et al.
show two age peaks for the Incidence (new cases in age groups) at 2-4 years (minor) and 6-8 years (major). This distribution is worrying for the progenitor’s selection because age class 6-8 shows an incidence of 31.1%, almost 1/3 of new cases of DCM in this age group, too late.
Risk perception is subjective
and everyone can see the glass half full or half empty as it seems. It is discussed in this section about Fedor. Can be him an affected DD or Dd? Probably no, because he has almost surpassed the age at risk (8 years). Can be him a silent carrier DD* or Dd*? Maybe yes, maybe not (his mother is probably affected DD or Dd). The time and the incidence of DCM in their progeny (the sample is very large) will reveal its likely condition. Otherwise if he is a silent carrier, most probably Dd*, no problem. In fact I have proposed before the use of heterozygous silent carriers Dd* when a genetic test can be available.
However, is a safe practice that a male as Fedor breed 150 or 200 bitches?
No. If he is a free genotype dd for DCM this practice is bad (popular sire effect), if he is a silent carrier Dd* worse because he would contribute to spread the risk allele D in the population and increasing the DCM prevalence.
We could continue discussing on many other suspect dogs and the discussion would be endless with the same results among the sceptics and the cautious. I do not think starting a “witch hunt” as convenient and would not lead anywhere. Only objective and scientific data is that prevalence of the disease in Europe is 58.2%. A shame!
This number has a reasonable explication only when a strong popular sire effect has disseminated the disease in the population. Retrospective origin of DCM in the North American Dobermanns is documented by the DCM status of famous Seven Sires (for example see Rod Humphries in Doberman Pinchers Magazine: The Cardiomyopathy Chronicles 1
), in Europe data are scattered on different dogs but not in a synthetic article.
In absence of a genetic diagnostic, only reasonable way to prevent the increasing of the DCM prevalence in the population it is risk diversification.
Risk diversification only works when you reduce the genetic quote of each individual in the Dobermann population. Limitation of litters by sire is necessary
. Is limitation possible? It seems that no with the actual state of opinion of influential people and hierarchies. DCM is not Von Willebrand disease (rarely mortal and recessive), DCM is dominant although with incomplete penetrance but mortal in all affected dogs. What do you ameliorate without selection on health if a 58% of born dogs will die of DCM before 8 old?
Is it coherent that the average longevity of the breed (7-8 years) is closely approximates with the limit risk age (8 old) for DCM and with data of pravalence (58%) in this study? Another coincidence? Unless we act now the percentage of affected dogs will continue to grow. What priority in this catastrophic situation, multi Ch & ZTP criteria of selection? If you add a criterion of clean DCM bloodline, how many dogs rest in this group? Sufficient number to make an amelioration job without going into a population bottleneck?
Your proposal that everything will remain as always, it is a proposal for risk
. It's like if you recommended to a lung cancer patient continue smoking. Well, there is a subtle difference: the lung cancer patient can choose but the dog does not, the choice is dictated by the breeder. No, unless we act now, things that go wrong likely will get worse.
I repeat, only reasonable way it is risk diversification and to limit sire utilisation
I am aware that the reality, although inconvenient, it is stronger than any scientific or ethical argument. Perhaps the happiness of the innocent is least annoying, except for the affected dogs and for the breed survival. Even the Devil has advocates, usually much because he is very powerful, but not dogs.