How many horses have EPM?
One thing we’d like to know: how many horses in the United States have EPM? Most authors start out telling us how important and serious EPM is and then they tell us the seroprevalence, how many horses have antibodies to S. neurona. And then they say that EPM occurs in less than 1% of horses. That’s rare.
Published studies determine the prevalence of EPM on seroprevalence, the presence of antibody determined by Western blot. That number is between 30% to (more than) 60% and is regional. Seroprevalence has little to do with actual cases of EPM. Horses with antibodies to S. neurona are horses that can succumb to EPM. An infection is a prerequisite for disease and antibodies result from infections. Seroprevalence can tell us how many horses are at risk to get protozoal myeloencephalitis due to S. neurona. The number of horses with EPM, based on antibody (tested by Western blot) is high.
Even if we knew how many horses had a definitive diagnosis of EPM, that won’t help get us to the true incidence of disease. A definitive diagnosis is defined as an animal from which the organism was cultured from the central nervous system. The number of horses seen at referral clinics can give us an idea of true disease. If our defined population is the group of horses diagnosed at a referral clinic for their ataxia (or suspect EPM)– we’d expect the number of horses that have EPM to predominate and therefore skew the data to an increase in the number of horses with the disease. Other lesser causes of neurological disease (viral infections, trauma, wobbler syndrome) are included. It is difficult to isolate the organism, thus some of these horses are diagnosed with EPM based on histological lesions that consist of inflammation.
The comprehensive NAHMS study conducted in 1998 and published in 2001–the only government document, tells us that this serious disease is rare. For this study the authors surveyed veterinarians and horse facilities.
An estimate of the EPM horses in the United States
We expect to approximate the number of horses that contract EPM annually and guess it will be between “less than 1%” and the number predicted by Western blot seroprevalence. This is the morbidity. The current United States horse population is considered to be 7, 200,000. That means that the number of horses with EPM is less than 72,000 if we believe that disease is less than 1%.
The NAHMS reports that a veterinary diagnosis (with or without the support of diagnostics) was made in 14 cases out of 10,000 horses in 1998. That is a rate of 0.0014 or 0.14%. This rate means that only 10,080 horses have EPM. Intuitively, that is low. The NAHMS number is less than 72,000 and is the low estimate of 10,080.
I spoke with Frank Andrews, current (and past) president of the EPM Society. Dr. Andrews thought 50,000 would be a high number. He presented the Proportional Morbidity Data at the EPM meeting recently. The rate “which is PMR, proportional morbidity rate, is 0.88% and roughly correlates to incidence (the number of new cases presented to Universities over 17 years).” He agrees that the actual incidence of disease is low, but exposure is high in some areas. He also says that the PMR has not changed in the 17 years included in the study but pointed out that there was a spike in cases from 1995-1998. The reasons for the spike seen in morbidity data for EPM were the development and use of the Western Blot and an increase in cases sent to universities due to drug efficacy trials, not an outbreak of disease. That means there are potentially 63,360 cases.
Narrowing it down
The increase due to drug study trials and hopes of a diagnostic test gave a falsely elevated rate for the years 1995-1998. We removed these years from the data set– 14 years for analysis. Andrews says the rate was unchanged over all the years and removing the spike should render the data to more accurately reflect the true rate over time. The adjusted rate is more realistic and is 0.65%. That would still yield the high value for morbidity because these cases are from referral centers. It’s high because this is the rate determined from cases of ataxic horses that were seen at referral university clinics, other causes of ataxia are not sent for a diagnosis. But are there any other factors we should consider?
Our data from the Peptide ELISA shows that only the animals with a titer against SnSAG 1 and SnSAG 5 should be considered for animals that are susceptible to EPM. Data based on the Ohio State model that showed that SnSAG 5 can’t experimentally induce disease (disease is defined as parasites isolated from the CNS of a horse).
The number of isolates from horses with SnSAG 5 strains and seroprevalence of SnSAG 5 antibodies in pure and mixed infections will yield how many animals can get EPM from this strain. Our phenotype data shows that 6.6% of the animals with a presumptive diagnosis of EPM by clinical examination are infected with the SAG 5 phenotype. The Ohio studies indicate this strain will produce signs in response to inflammation.
Observations in the field support our findings. In the published field EPM cases for which the organism was isolated, only 7% displayed the SAG 5 phenotype, and there was evidence of mixed infections based on some of the published reports. The SnSAG 5 infections often resolve without treatment, based on the OSU studies.
Potentially all the SnSAG 1 infected animals are at risk for EPM.
The PRM report, adjusted for the spike, estimates the rate of EPM at 43, 992. That is 61 horses in 10,000. We averaged the high and the low estimates numbers to approximate the number of horses that have EPM and it is 27,003.