UMI INFORMATION TO USERS

Transcription

1 INFORMATION TO USERS This manuscript has been reproduced from the microfilm master. UM I films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter fiice, while others may be from any type o f computer printer. The quality o f this reproduction is dependent upon the quality o f th e copy subm itted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely afreet reproduction. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand com er and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back o f the book. Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6 x 9 black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Infonnation Company 300 North Zeeb Road, Ann Arbor MI USA 313/ /

2

3 NOTE TO USERS The original manuscript received by UMI contains pages with Indistinct and slanted print. Pages were microfilmed as received. This reproduction is the best copy available UMI

4

5 THE EPOEMIOLOGY OF EQUINE PROTOZOAL MYELOENCEPHALITIS (EPM) A Dissertation Presented in Partial Fulfillment of the Requirements for the degree Doctor of Philosophy in the Graduate School o f The Ohio State University By William James Allan Saville, DVM, Diplomate ACVIM * * * * * The Ohio State University 1998 Dissertation Committee: Professor Paul S. Morley, Adviser Professor Stephen M. Reed Professor David E. Granstrom Professor Catherine W. Kohn Approved by Adviser Department of Veterinary Preventive Medicine Professor Thomas E. Wittum

6 UMI Number; Copyright 1998 by Saville, W illicua James Allcui All rights reserved. UMI Microform Copyright 1999, by UMI Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code. UMI 300 North Zeeh Road Ann Arbor, MI 48103

7 C opyright by W illiam Jam es A llan Saville 1998

8 ABSTRACT Equine Protozoal Myeloencephalitis (EPM) is the most common neurologic disease o f horses presented to the OSU Veterinary Hospital. The purpose of these studies was to investigate the epidemiology o f EPM. We examined the prevalence of serum antibody to Sarco cystis neurona using a random sample o f horse serum. The rate of exposure increased with age. Breed and gender were not associated with seroprevalence, but there were geographic differences. We followed horses (n=6) that were neurologically normal and negative for serum antibody to Sarcocystis neurona for approximately one year to investigate natural exposure to this parasite. Horses naturally exposed to S. neurona may not develop clinical signs of EPM for long periods and some may never develop clinical signs. A retrospective case-control study was performed to examine risk factors for EPM and factors associated with improvement and survival in EPM cases. Risk factors that appear to Increase the risk o f EPM include age, season, horses located on premises which had previously diagnosed cases of EPM, premises where opossums are commonly seen, the occurrence of stressful events prior to admission, and race or show horses. Factors that reduced the risk of developing clinical signs of EPM included protection o f feed from wildlife access and presence of a creek or river on the premises. The likelihood o f improvement in clinical signs was lower in u

9 breeding and pleasure horses, but higher in horses treated with antimicrobials. Two factors were associated with the likelihood of survival in EPM cases, severity of clinical signs at presentation and improvement in clinical signs. Neurologic and non-neurologic horses were examined prospectively to compare agreement between board certified internists. Th^ '** was good overall agreement when examining for presence or absence of neurologic signs, but poor agreement in diagnosing specific clinical signs. Prospective examination of ante-mortem tests used for diagnosis of EPM found that Western blot appears to have reasonable sensitivity but poor specificity in CSF. PCR testing appears to be insensitive. CSF indices also may not be helpful in diagnosing neurologic disease. There were no differences between clinically normal and horses with neurologic disease in parameters evaluated during routine CSF analysis. m

10 Dedicated to my wife Margo and my late parents, Max and Lil Saville IV

11 ACKNOWLEDGMENT I wish to thank my advisors for their support over the last 5 years. My first advisor, Steve Reed, for without him none o f this would have been possible. His continued efforts to obtain fiinding for research, and support for my stipend for the last three years, will not be forgotten. My second advisor, Paul Morley, receives no less recognition for his attempts to make me a better scientist. Paul s hours of helping me with the computer and aiding me in the understanding of epidemiology and statistics will keep me indebted for life. To Dave Granstrom, who has been an inspiration since our first meeting in May of 1993, thank you for the wonderful opportunity to work with you. Your knowledge and understanding of this disease we have been glued to for the last several years has provided a wonderful challenge for me and, as I m sure, others. To Catherine Kohn, my thanks for recognizing a worthy candidate for the residency in 1993 and her continued support throughout the last 5 years. Catherine s attempts to make me an English major will be forever etched in my memory. And to the last member of my committee, but certainly not the least important, I would sincerely like to thank Tom Wittum. His countless hours o f answering my endless tirade of questions about epidemiology and statistics, even when I m sure they were mundane questions to him.

12 To Ken Hinchcliflf, who was on my committee for the first 3 years, I would like to express my sincerest gratitude for his support in the residency program as well as countless hours fielding my questions about research and writing. I owe a great deal of gratitude to my wife Margo for tolerating my mid-life crisis and moving to a completely strange environment to start fi-esh. I m sure it has been a long 5 years for her with all my late hours and constant work habits. To Maggie Lauderdale I am grateful for the help she has provided over the last several months. She is well organized and has done a tremendous job trying to keep the samples collected, making phone calls, and numerous other tasks. There are 6 veterinary students who took a month out o f their busy exam schedule this spring to help us with the retrospective study and my sincerest gratitude goes to: Chrissy Schilling, Mindy Cavender, Joe M organ, Sean Morrison, Kim Brooks and Laura Porter. I also wish to thank others who have helped to enhance my academic and intellectual development. There is not enough paper nor is my memory for names good enough to remember all of their names, but thank you for your help. To David Daniels and Duncan Alexander, I am eternally grateful for your support over the last 5 years. American Live Stock Insurance Company is a credit to the horse industry for the support they have given veterinary medicine and in particular Ohio State. VI

13 VITA August 10, Bora - Wainwright, Alberta, Canada DVM, University of Saskatchewan Veterinary Practitioner, Western Canada present... Graduate Teaching and Research Associate, The Ohio State University Research Publication PUBLICATIONS 1. Saville WJ, HinchclififKW, Moore BR, Reed SM, Kohn CW, Mitten LA, Rivas LJ. Necrotizing Enterocolitis: A retrospective study. J Vet Intern Med 1996; Vol 10(4): Saville WJ, Reed SM, Granstrom DE, ffinchclifif KW, Kohn CW, Wittum TE, Stamper S. Prevalence of serum antibodies to Sarcocystis neurona in horses in Ohio. J Am Vet Med Assoc 1997; 210(4): vu

14 3. Granstrom DE, Saville WJ. Equine protozoal myeloencephalitis. In: Equine Internal Medicine, SM Reed and WM Bayly, Eds., W.B. Saunders Co., Philadelphia, PA, Saville WJ. Polyneuritis equi. In: Equine Internal Medicine, SM Reed and WM Bayly, Eds., W.B. Saunders Co., Philadelphia, PA, Major Field: Veterinary Preventive Medicine FIELDS OF STUDY Studies in Epidemiology, Veterinary Internal Medicine, and Biostatistics viu

15 TABLE OF CONTENTS A bstract...ü Dedication... iv Acknowledgment...v V ita... vii List o f Tables... xiii List o f F igures... xvi CHAPTER 1 Introduction... 1 R eferences... 4 CHAPTER 2 Literature R eview The S ix tie s Clinical Signs Discovery o f the Causative A gent Culture o f the Organism Other Causative A g e n ts Life Cycle o f S a rco cystis n eu ro n a Aberrant Intermediate H osts Transport (Mechanical) V ecto rs Pathogenesis Transmission E x p o su re Immunosuppression D iagnosis Post-mortem Examination Ancillary T e s ts Western Blot Analysis...19 IX

16 2.6.4 Polymerase Chain Reaction (P C R ) CSF In d ic es Controversy over Testing New Tests Treatm ent Standard Therapy Anti-inflammatory Therapy Immune Stimulants Supplemental T h erap y Duration of Therapy New Anti-protozoal T herapy P rognosis Relapse after Cessation of Therapy Epidem iology Prevalence o f Infection Prevalence of Disease Geographic Distribution Breed Predilection Age Distribution Sporadic versus Outbreak R eferences...32 CHAPTER 3 Prevalence o f Serum Antibodies to Sarcocystis N eurona in Horses In O h io A b stract Introduction Materials & M ethods Study D e sig n Samples and sample analyses Data analysis Results Descriptive information Bivariate analysis Multivariable analysis D iscussion References...61 CHAPTER 4 Natural Exposure o f 6 Horses to Sarcocystis Neurona'. Biweekly Monitoring for 12 Months A bstract Introduction Materials & M ethods Study D esign...64

17 4.3.2 Management of H o rse s Laboratory Analysis Statistical Analysis Results Discussion: References...80 CHAPTER 5 A Retrospective Case-Control Study to Examine the Risk Factors for Equine Protozoal Myeloencephalitis (E P M ) Abstract Introduction Methods Study D esign Statistical Analysis Results Discussion References CHAPTER 6 Inter-Observer Variation in the Diagnosis o f Neurologic Abnormalities in the H o rs e A b stract Introduction Methods Study D e sig n Statistical A nalysis Results Discussion References CHAPTER 7 A Prospective Investigation o f Antemortem CSF Analysis as Used in the Diagnosis of Equine Protozoal Myeloencephalitis (E P M ) A bstract Introduction Methods Study D e sig n Clinical Inform ation Sampling P rotocol Statistical A nalysis Results Discussion References CHAPTERS C onclusions XI

18 Bibliography Appendix Data Collection M aterials Retrospective Survey Q u estio n s Retrospective History F o r m XU

19 LIST OF TABLES Table 3.1 Table 3.2 Table 3.3 Table 4.1 Seroprevalence of antibodies to S. neurona in horses by categorical variables Final multiple logistic regression model for factors associated with the presence of antibody to S. neurona in serum from 1056 horses in Ohio Multiple logistic regression model for effect o f climate (by region) on seroprevalence of S. neurona antibodies among 1056 horses in O h io 60 Horse number with age, breed, gender, time on farm prior to the study and week o f seroconversion for each subject in the study, a = Thoroughbred horse; b = Quarter Horse; c = Standardbred; d = A rabian Table 4.2 CSF samples from horses that were positive (P) or suspect positive (S) for S. neurona antibody using the Western blot assay. Also included are respective CSF indices for each positive or suspect positive sample, the week involved and the horse n u m b e r Table 4.3 Table 5.1 Table 5.2 Table 5.3 Table 5.4 Table 5.5 Table 5.6 Cerebrospinal fluid indices (mean ± SD; ranges) for all horses for the entire study period along with published reference ran g es Table o f variables created from review of records and information collected from telephone follow-up. Data are organized by case and control series. Variables are listed according to subset grouping Types o f injuries, illness or other stressful events reported to occur prior to presentation Table o f independent variables that survived the initial screening (P<0.25) for risk factors for EPM compared to non-neurologic co n tro ls...i l l Table o f variables that survived the subset screening (P^O.IO) for risk factors for disease, comparing EPM cases and non-neurologic controls, and will be included in the final model building process Table o f the multivariable logistic regression final model with risk factors for EPM cases versus non-neurologic controls and outcome disease Table o f independent variables that survived the initial screening (P<0.25) for risk factors for EPM compared to neurologic controls xm

20 Table 5.7 Table 5.8 Table 5.9 Table o f risk factors that survived the subset screening process (PsO. 10) and will be used for the final model building process for EPM cases and neurologic controls with the outcome d isease Table o f the final multivariable logistic regression model o f risk factors for EPM cases versus neurologic controls with outcome disease Table of continuous variables describing the CSF analysis included in the model building process for EPM cases and the outcomes not improved and not survived. Comparison with continuous variables for the neurologic controls Table 5.10 Table of risk factors for EPM cases with the outcome not improved including the independent variables that survived the initial screening (P<0.25) Table 5.11 Table of the risk factors that survived the subset screening (P^O.IO) and will be included in the final model building process for EPM cases and outcome not improved Table 5.12 Table o f final multivariable logistic regression model with risk factors for EPM cases and outcome not improved. PTA=prior to ad m issio n 133 Table 5.13 Table o f risk factors for EPM cases and outcome not survived, including variables that survived the initial screening (P<0.25) Table 5.14 Table o f risk factors that survived subset screening (P^O. 10) and will be included in the final model building process for the EPM cases and outcome not survived Table 5.15 Table o f final multivariable logistic regression model for risk factors for EPM cases and outcome not survived Table 6.1 Table 6.2 Table 6.3 Summary o f agreement for all Examiners in assessing overall status (neurologic/not neurologic), each aspect of gait (truncal sway, asymmetry, lameness, ataxia, weakness, spasticity), presence or absence of muscle atrophy, and overall severity of the neurologic signs. Overall Kappa statistics (K ± SD) and 95% confidence intervals for inter-observer agreement between the 4 internists for each category o f clinical sign observed during a neurologic work-up Agreement (K) among five pairs o f boarded internists when observing clinical signs o f horses. K = k ap p a Summary of agreement between Examiners 1 and 2 in assessing overall status (neurologic/not neurologic), each aspect o f gait (truncal sway, asymmetry, lameness, ataxia, weakness, spasticity), presence or absence of muscle atrophy, and overall severity of the neurologic signs XIV

21 Table 6.4 Table 6.5 Table 6.6 Table 6.7 Table 7.1 Table 7.2 Table 7.3 Table 7.4 Table 7.5 Summary of agreement between Examiners 1 and 3 in assessing overall status (neurologic/not neurologic), each aspect of gait (truncal sway, asymmetry, lameness, ataxia, weakness, spasticity), presence or absence of muscle atrophy, and overall severity of the neurologic signs Summary of agreement between Examiners 1 and 4 in assessing overall status (neurologic/not neurologic), each aspect o f gait (truncal sway, asymmetry, lameness, ataxia, weakness, spasticity), presence or absence of muscle atrophy, and overall severity of the neurologic signs Summary of agreement between Examiners 2 and 3 in assessing overall status (neurologic/not neurologic), each aspect o f gait (truncal sway, asymmetry, lameness, ataxia, weakness, spasticity), presence or absence of muscle atrophy, and overall severity of the neurologic signs Summary of agreement between Examiners 2 and 4 in assessing overall status (neurologic/not neurologic), each aspect o f gait (truncal sway, asymmetry, lameness, ataxia, weakness, spasticity), presence or absence of muscle atrophy, and overall severity of the neurologic signs Descriptive statistics data for continuous variables Results of the Western blot and polymerase chain reaction (PCR) assays in all horses. a=differences between the neurologic and non-neurologic horses was not significant (P=0.69). b=dififerences between the neurologic and nonneurologic horses was not significant (P=0.30) Results of Western blot assays for horses in which the albumin quotient was <2.0. a=dififerences between the neurologic and non-neurologic horses were not statistically significant (P=0.22). b=dififerences between the neurologic and non-neurologic horses were not statistically significant (P=0.17) Serum and CSF Western blot results with albumin quotient and IgG index calculations for individual horses that did not have detectable neurologic deficits. n = Serum and CSF Western blot results with albumin quotient calculations for Individual horses that presented with detectable neurologic deficits. n = XV

22 LIST OF FIGURES Figure 3.1 Map depicting statistical districts established for the state o f Ohio by the Ohio Department of A g ricu ltu re Figure 3.2 Map depicting extension districts established for the state o f Ohio by the Ohio State University Extension S erv ice...55 Figure 3.3 Prevalence o f serum antibodies to S neurona versus age o f horse. Results are plotted for each statistical district...56 Figure 4.1 Cerebrospinal fluid (CSF) Albumin Concentration o f horses repeatedly sampled over 93 weeks (n=6)...76 Figure 4.2 Albumin Quotient (AQ) o f horses repeatedly sampled over 93 weeks (n=6)...77 Figure 4.3 Cerebrospinal fluid (CSF) Immunoglobulin (IgG) concentration o f horses repeatedly sampled over 93 weeks (n=6) Figure 4.4 Immunoglobulin (IgG) Index o f horses repeatedly sampled over 93 weeks (n=6)...79 Figure 5.1 Age categories by series for all horses in the study Figure 7.1 Box plot o f the CSF total protein concentration by neurologic status. Neurologic =exhibits neurologic signs; nonneurologic=does not exhibit neurologic signs Figure 7.2 Box plots of CSF white blood cell count by neurologic status. Neurologic =exhibits neurologic signs; nonneurologic=does not exhibit neurologic signs Figure 7.3 Box plots of CSF red blood cell count by neurologic status. Neurologic =exhibits neurologic signs; nonneurologic=does not exhibit neurologic signs Figure 7.4 Box plots o f serum albumin by neurologic status. Neurologic =exhibits neurologic signs; nonneurologic=does not exhibit neurologic signs 186 XVI

23 Figure 7.5 B ox plots o f CSF albumin concentration by neurologic status. Neurologic = exhibits neurologic signs; nonneurologic=does not exhibit neurologic signs Figure 7.6 Box plots o f albumin quotient (AQ) by neurologic status. Neurologic = exhibits neurologic signs; nonneurologic=does not exhibit neurologic signs Figure 7.7 Box plots of serum immunoglobulin G concentration by neurologic status. Neurologic =exhibits neurologic signs; nonneurologic=does not exhibit neurologic signs Figure 7.8 Box plots o f CSF immunoglobulin G concentration by neurologic status. Neurologic =exhibits neurologic signs; nonneurologic=does not exhibit neurologic signs Figure 7.9 B ox plots o f immunogloblin G index by neurologic status. Neurologic =exhibits neurologic signs; nonneurologic=does not exhibit neurologic signs xvu

24 CHAPTER I INTRODUCTION Equine Protozoal Myeloencephalitis (EPM) is a serious and often fatal neurologic disease of horses.the causative agent o f EPM is Sarcocystis neurona, a protozoan parasite which appears to have a predilection for the central nervous system o f horses. Equine Protozoal Myeloencephalitis is a common disease and has significant economic impact on the U.S. horse industry. There are approximately 170,000 horses in the state of Ohio according to the 1985 American Horse Council (AHC) census. The prevalence of EPM in the general horse population has been estimated at between 0.5% and 1%. Using these findings, the total direct costs of treating affected horses in Ohio have been estimated at 1.4 to 2.8 million dollars during Considering the magnitude of estimated losses for Ohio, nation-wide impact of this disease is likely tremendous. Recent figures provided by the USD A indicate that the equine industry may be valued at $20 billion annually.^ In 1995, carcasses were exported at a value of $67.5 million, and export of live horses was estimated at $286 million.^ This is in contrast to the export of live cattle in 1996 valued at $86 million. A recent survey was conducted to identify needs of the U.S. horse industry and to determine the priorities for the NAHMS Equine 1998 study.* There were 2,599 respondents to this survey, 76% of which were horse owners, the remainder being veterinarians and other representatives o f the horse industry *

25 O f the infectious diseases listed, equine protozoal myeloencephalitis (EPM; 24.9%) and equine infectious anemia (EIA; 22.4%) were most commonly identified as top priorities for research.* EPM is a high profile disease in the equine industry. Numerous lay journals publish articles on this disease with a different theme in every issue. There is little consistency in the information that is relayed which is a reflection of the state of knowledge about this devastating disease. In addition, lists are filled with often heated discussions between veterinarians and experts in the field of equine medicine regarding diagnosis and therapy of EPM. Equine Protozoal Myeloencephalitis (EPM) appears to create panic among horse owners. In the spring of 1997, a world wide web site was introduced at Ohio State University to aid in educating veterinarians as well as horse owners. Since that time, there have been approximately 10,000 visits to the site and several hundred letters were received with enquiries about the disease. There is a definite hysteria present among the horse owners when their horse is diagnosed with this disease. The most important and frequently asked questions concern diagnosis of EPM, therapy used, prognosis, and methods of prevention. Unfortunately, there is a lack of confidence in the diagnostic tests presently available, and many times the tests are not properly applied. It is currently thought that therapies used to treat horses with EPM help to improve neurologic condition in 60 to 70% o f the cases, but many horses do not return to their original fimction. Most importantly, effective preventive strategies have not been identified. It is

26 not surprising that horse owners are frustrated when veterinarians who work with this disease on a daily basis are equally frustrated. Controlled studies o f the epidemiology of this disease are lacking. Little information is available and published reports are currently limited to two case series. The purpose of the present studies was to examine the epidemiology o f Equine Protozoal Myeloencephalitis in a controlled fashion. We examined the seroprevalence of antibodies to Sarcocystis neurona in the state of Ohio to provide an indication o f the infection rate in the state and to gain information about the life cycle of the organism. We followed naturally infected horses prospectively to learn about the incubation period and the clinical course of the disease. In addition, we examined the risk factors for disease in a retrospective case-control study. The goal of these studies was to provide information about methods that could be used to manipulate management of horses in an effort to prevent disease. In a prospective comparison of horses with neurologic disease to those without neurologic disease, results o f diagnostic testing were compared, including Western blot, polymerase chain reaction, CSF cytology and CSF indices. We also examined the inter-observer variation between board certified internal medicine specialists when performing neurologic examinations in the horse. EPM is currently the most common neurologic disease among horses presented to The OSU Veterinary Teaching Hospital. However, it is difibcult to diagnose, treat and prognosticate based on the current understanding of the disease. The overall goal of this work is to obtain a better understanding of EPM with regard to diagnosis as well as to objectively develop strategies for disease prevention based on risk factor analysis.

27 References 1. Granstrom DE, Dubey JP, Davis SW, et al. Equine protozoal myeloencephalitisantigen analysis of cultured Sarcocystis neurona merozoites. J Vet D iagn Invest 1993;5: Granstrom DE, Dubey JP, Giles RC, et al. Equine protozoal myeloencephalitis: Biology and epidemiology. Proc VU Intern Coof Equine Infect Dis 1994; Granstrom DE, Saville WJ. Equine Protozoal Myeloencephalitis In: S. M. Reed and W. M. Bailey, eds. E quine In tern a l M edicine. Philadelphia, Pa.: WB Saunders Company, 1998; Reed SM, Granstrom DE. Equine protozoal encephalomyelitis. Proc Am Coll Vet Intern Med Forum 1993 ; Reed SM, Granstrom D, Rivas LJ, et al. Results of cerebrospinal fluid analysis in 119 horses testing positive to the western blot test on both serum and CSF to equine protozoal encephalomyelitis. Proc Am Assoc Equine Pract 1994; Granstrom DE. Equine Protozoal Myeloencephalitis: Parasite biology, experimental disease, and laboratory diagnosis. International Equine Neurology Conference 1997; Anonymous. The USD A s role in equine health monitoring. In: 1996, ed: USD A: APHIS VS, Anonymous. NAHMS Equine '98: Needs assessment survey results In: 1997, ed: USDA:APHIS:VS, 1997.

28 CHAPTER! LITERATURE REVIEW 2.1 THE SIXTIES In the early I960's, there were reports of a progressive, idiopathic neurologic disease with an apparent increased predilection among Standardbreds. This unusual disease was most likely equine protozoal myeloencephalitis (EPM). One author noted that muscle atrophy was seen in association with these neurologic signs and that affected horses progressed to terminal prostration. In the late 1960's, researchers from the University of Kentucky reported on 44 unusual cases o f neurologic disease characterized by either sudden or gradual onset o f clinical signs.^ The disease affected horses o f a wide range of ages, and appeared to primarily affect Standardbreds.^ Both focal or multifocal lesions of the spinal cord were observed.^ Unfortunately, these same signs are currently seen all too frequently in horses diagnosed with EPM. 2.2 CLINICAL SIGNS Descriptions of clinical signs of horses diagnosed with EPM can vary greatly because the organism that causes this disease can affect any tissue within the central

29 nervous system (CNS). Therefore, any neurological abnormality exhibited by a horse could be caused by EPM. Clinical signs recognized in the earliest studies o f this disease still characterize neurologic abnormalities in horses with EPM. Early workers described horses with EPM as having an asymmetric ataxia and associated muscle atrophy. Horses may have a sudden onset o f clinical signs or disease may progress slowly over several months.^ Vague, intermittent lameness that is non-responsive to therapy may be caused by EPM and encephalitic signs typified by asymmetric cranial nerve deficits may also be seen in affected horses.* Although EPM is typified by the presence o f asymmetrical, multifocal neurological abnormalities, horses with EPM may have focal or symmetric signs. Cerebral signs are rarely seen in horses with EPM. However, 3 horses with EPM presented to The Ohio State University Veterinary Teaching Hospital displayed seizure activity and evidence of cortical electrical activity abnormalities on EEG examination.* Horses with cerebral neurologic signs often have a poor prognosis. However, seizure activity in horses with EPM may be treatable. Visual deficits and behavioral abnormalities have been reported in horses with EPM.* Head shaking was also reported in a recent case series describing 3 horses with EPM.^ Head shaking resolved in these horses after treatment for EPM.

30 2.3 DISCOVERY OF THE CAUSATIVE AGENT Results o f early investigations suggested that this disease was infectious or inflammatory in nature, but an etiology was not immediately identified despite considerable efforts by researchers. Initially, researchers were suspicious that Toxoplasma gondii caused this disease.* However, cerebral signs are more prominent than the spinal cord signs in other species with Toxoplasmosis, whereas, the spinal cord signs predominated in this new disease of horses.* * A landmark paper published in the early 1970's demonstrated that the organism that caused EPM was likely a Sarcocystis spp. Dubey et al. reported that although organisms found in the CNS o f affected horses resembled T. gondii, there were some distinct differences. The presence o f large numbers of rosettes within neurons, the presence o f associated mononuclear inflammation, the lack o f periodic acid Schifif (PAS) staining of parasites and the absence of cyst wall staining with Wilder's silver were inconsistent with T. gondii^ These findings were confirmed by another research group that described PAS-negative organisms associated with a nonsuppurative inflammation and mononuclear pleocytosis in nervous tissue o f 8 horses suspected o f having EPM. Toxoplasma gondii antibody was absent in both serum and cerebrospinal fluid (CSF) in affected horses in another study.^ Further confirmation that the causative agent o f EPM was likely a Sarcocystis spp was based on the characteristics o f the organism s asexual reproduction. *^ Asexual reproduction of this agent occurs before nuclear division with the formation o f more than 2 daughter cells (endopolygeny) and differs from asexual reproduction o f Toxoplasma gondii which divides to form 2 daughter cells

31 (endodyogeny). *^ In other Swcocystis spp, schizogonous stages lack a parasitophorus vacuole, and there is an absence of rhoptries. There was also an absence of rhoptries in this organism found in horses with EPM/^ Schizonts were found in the cells o f the white and gray matter as well as in a circulating neutrophil o f an EPM-afifected horse, which indicated there may be a hematogenous phase to the infection/^ Hematogenous dissemination o f infection is typical in Sarcocystis spp infections in other host species.'^ C ulture of the O rganism It was not until several years after the identification of Sarcocystis as the likely causative agent of EPM that a Sarcocystis-\jk& organism was cultured from the spinal cord of a h o rse.b e c a u se of the association of this organism with neurons in the central nervous system, the organism was named Sarcocystis neurona}^ This isolate showed cross reactivity with S. cruzi-antiserum. " The S. cn/zz-antiserum also was tested against several schizonts and sporocysts o f other Sarcocystis spp {S. cruzi, S. tenella, S. capricanis)}^ The intensity of the reaction was similar for the S. neurona isolate and the controls." The organism isolated by culture was structurally similar to organisms found in fixed tissues o f other affected horses." Subsequently, reports document that this organism has been cultured from California (3 times), Panama and Kentucky (3 times)." There have been anecdotal reports of isolation of the organism from other states as well including Michigan (Linda Mansfield, personal communication) and Florida (Robert MacKay, personal communication). Recovery o f this organism from nervous system

32 tissue of affected horses was therefore repeatable. In addition, transport of the organism at 4 C for 5 days and subsequent culture demonstrated the hardiness o f the parasite.** Further investigation of S. neurona showed that it is different from other Sarcocystis spp in that the schizonts are generally not seen in the vascular endothelium, but have been found exclusively in the brain and spinal cord." Schizonts contained a prominent residual body.*^ The schizonts were typically seen in the cytoplasm of macrophages in the CNS, but rarely in neutrophils.*^ Schizonts were also found in neurons in various stages o f development, occasionally in multi-nucleated giant cells, and also extracellularly in necrotic lesions o f the CNS.*^ DNA analysis has been very important in characterizing and classifying S. neurona. Using a random primed polymorphic DNA assay (RAPD), a unique sequence of base pairs was identified that distinguished S. neurona from 8 related coccidia, specifically 2 Sarcocystis spp, 1 Toxoplasma spp and 5 Eim eria spp.* This research demonstrated that unique DNA sequences could be successfully utilized as a species-specific probe for S. neurorui, and that these probes permitted differentiation of S. neurona from other coccidia o f equines.* Further studies performed to determine the phylogenetic relationship of S. neurona to other members of the family Sarcocystidae were based on the sequence of the small ribosomal subunit gene o f S. neurona using polymerase chain reaction (PCR).* Phylogenetic analysis confirmed that the agent causing EPM belonged in the genus Sarcocystis. Interestingly, Toxoplasma gondii, Sarcocystis muris, zcad Sarcocystis gigantea have a close phylogenetic relationship to S. neurona.^^

33 2.3.2 Other Causative Agents Other research suggests that Neospora spp may rarely cause disease that is similar to However N eospora organisms form typical tissue cysts in the horse that have not been found in horses with EPM. *' Neospora is a worldwide cause o f abortion, particularly in cattle. However, previous research indicates that EPM is a disease principally found in North, Central and South America. Therefore, it seems unlikely that Neospora is a common cause o f disease that is similar to EPM. 2.4 LIFE CYCLE OF Sarcocystis neurona Sarcocystis species typically have a two-host predator-prey life cycle. The asexual stage (schizont) o f the Sarcocystis organism is usually found in the intermediate host (prey).^^ ^ The sexual stage (sporocyst) is found in the definitive host (predator). Most specific details regarding the life cycle o f S. neurona are currently unknown, although recent research has demonstrated that the opossum is likely the definitive host. Early work suggested that the definitive host of Sarcocystis neurona must be indigenous to North, Central and South America because EPM had been reported only in horses fi^om these regions, or in horses that have been exported fi"om these regions to the other continents. Although many species o f wildlife have been suggested to be the true definitive host for S. neurona, the skunk, raccoon and opossum were considered to be the most likely candidates because these animals are unique to the Western hemisphere.^ Subsequent studies of these species found that only the skunk had serum antibodies to S. neurona?^ 10

34 Feces and intestinal digest from 4 raccoons, 2 opossums, 7 skunks, 6 cats, I hawk and I coyote were screened using the small ribosomal subunit gene (SSURNA) primers to identify S. neurona}^ Amplified PCR product obtained from 2 different opossums, was sequenced and compared to the SSURNA gene from the SN5 strain of S. neurona^^ There was a difference of only 2 nucleotides over the entire SSURNA gene (1,806 bp) which represented a 99.89% homology between the two sequences.^ These studies suggested that the opossum might harbor S. neurona in its gastrointestinal tract. The geographic distribution o f opossums is similar to the geographic distribution of EPM and areas with lower seroprevalence of 5. neurona appear to coincide with regions outside the natural range of opossums.^ The opossum is known to be the definitive host for two species o f Sarcocystis (S. rileyi and S. falcatula)}^ However, the SSURNA gene sequence from S. rileyi obtained from sarcocysts in ducks differs greatly from that of S. neurona?^ Therefore, the organism harbored by opossums that was most closely related if not identical to S. neurona was S. falcatulcu The intermediate hosts for S. falcatula include passeriform, psittaciform and columbiform birds. Sarcocystis falcatula is the only apicomplexan protozoan reported to parasitize the muscle of cowbirds and grackles.^* The SSURNA gene sequences of sarcocysts obtained from skeletal muscle of a brown-headed cowbird and schizonts from lung of a Moluccan cockatoo were compared to 2 isolates of S. neurona (SN 5 and UCD-1). The S. falcatula sequence from the 2 birds was identical to the UCD-l isolate of S. neurona and differed in only three positions from the SN 5 isolate of S. neurona}* The sequence of the schizonts from the cockatoo lung was identical to the sequence from II

35 the sarcocysts in the muscle of the cowbird.'* Based on sequence variation between isolates o f other species o f apicomplexan parasites, the minor differences found between these 2 isolates o f S. neurona were well within the expected range o f sequence variation for a single species.^ Other known intermediate hosts for S. falcatula include grackles and grosbeaks. Experimental infections have been established in other passerine, psittacine and columbiform birds including finches, canaries, budgerigtu:;, sparrows, and pigeons.^ Like the opossum, brown-headed cowbirds and common grackles have geographic ranges which includes most o f the U.S. and southern Canada.^* While these studies strongly suggest that the opossum is the definitive host of Sarcocystis neurona, studies comparing gene sequences of sporocysts fi om the opossum to sequences o f S. neurona isolates are required to confirm this hypothesis. Further evidence that the opossum is the definitive host for S. neurona was obtained by experimental induction o f EPM. ^ When sporocysts firom feral opossums were fed to horses, neurologic disease developed.^ However, induction of clinical EPM by feeding laboratory derived S. fa lca tu la sporocysts was not successful (David E. Granstrom, Personal Communication). This suggests that the opossum is the definitive host for more than one species of Sarcocystis. While some studies suggest that S. neurona and S. falcatula are actually the same organism,another study suggests that they may in fact be different, closely related species.^' Budgerigars were given merozoites of.?, falcatula and subsequently died fi"om acute sarcocystosis, whereas, budgies given merozoites o f S. neurona did not develop 1 2

36 lesions or clinical disease/' In another investigation, nude mice were inoculated with S. neurona merozoites cultured from an S. neurona-mîqct&é horse. These mice developed S. newrona-associated encephalitis, further demonstrating the unique characteristics o f this organism Aberrant Intermediate Hosts Unlike most Sarcocystis spp, S. neurona may aberrantly infect a large number of intermediate hosts. While the full range of intermediate hosts for S. neurona has not yet been identified, several species o f animals and birds have been reported to exhibit symptoms similar to those seen in horses with EPM. Several reports indicate that an S. n e u ron a -^e organism infects and causes neurologic disease in dogs, sheep, cats, mink, raccoons, a striped skunk, a golden hawk, rhesus monkeys and chickens. This wide host range is atypical for Sarcocystis spp. This host range behavior is similar to that of Toxoplasma gondii which is phylogenetically close to S. neuronap '^ Transport (Mechanical) Vectors Based on the estimated numbers of opossums in North America, the poor survival rate of these animals, and the small areas in which they travel, there is speculation that this organism may be transmitted via methods other than direct contact with opossum feces. Experiments performed by researchers in the 1980's indicate there may be some transmission by birds. In experiments attempting to characterize the life cycle of S. falcatula, birds w ere apparently infected by aerosol spread. Vector transmission was 13

37 also demonstrated by the recovery of sporocysts after budgerigars, canaries, white mice and chickens, were fed opossum feces/^ The recovered sporocysts were then fed to budgerigars to assess the viability of the sporocysts/^ Four of 6 budgerigars died, demonstrating that the sporocysts were viable/^ These experiments suggest that sporocysts might be transmissible between intermediate hosts/^ Considering the apparent wide range of natural and aberrant intermediate hosts for S. neurona and S. falcatula, transmission o f infectious organisms between intermediate hosts implies that control of disease caused by these organisms may be extremely difficult. Insects such as flies and cockroaches may also be transport vectors for S. neurona. Early work demonstrated that flies may act as transport vectors for Toxoplasma gondii.** Subsequently, the same group found that cockroaches may also act as transport vectors for T. gondii.*^ In addition, fatal pulmonary disease developed in psittacine birds that were fed cockroaches after the cockroaches had been fed opossum feces.^ While this suggests that insects may play a role in transmission o f S. neurona, further investigation is necessary to determine which insects are actually involved in the life cycle o f this organism. 2.5 PATHOGENESIS Transmission Little is known about the pathogenesis of EPM. It is assumed that the horses ingest Sarcocystis neurona and that the course o f infection and disease is then similar to that observed in other host species infected with Sarcocystis spp. 14

38 Because the sporocysts of S. neurona are passed in the feces of the opossum, it is likely that infective oocysts are introduced into the feed and water supply o f intermediate hosts." Once ingested, the sporocysts excyst and release sporozoites which penetrate the gut and enter arterial endothelial cells of various organs." Meronts develop and rupture the host cell releasing merozoites into the blood stream." This is probably followed by a second round of merogony throughout the body." In most sarcocystis-like diseases, this process results in the formation o f sarcocysts in the muscle." Subsequent ingestion o f the infected muscle tissue by the predator or definitive host completes the life cycle." Sarcocysts of S. neurona have not been found in affected horses, indicating that the horse is an aberrant, dead-end host.^^ Sarcocystis neurona has been recovered firom CNS lesions in several horses and subsequently propagated in culture in the laboratory. When administered to horses parenterally or introduced via the epidural space, cultured merozoites have not induced clinical disease in the horse.* The merozoite stage of Sarcocystis spp is not known to be transmissible to other animals.* However, nude mice have been inoculated intraperitoneally with cultured merozoites and subsequently developed evidence of S. newrona-associated encephalitis.^^ These mice were immunosuppressed strains and intraperitoneal injection would not likely be the normal route of infection with S. neurona in horses. The mechanism by which the merozoites enter the CNS is currently unknown. The organism likely enters the CNS via infected leukocytes or through the cytoplasm of endothelial cells.* Transplacental infection of the fetus with S. neurona has not been reported.* 15

39 2.5.2 Exposure High prevalence o f exposure to S. neurona has been demonstrated in horses. Recent studies indicate that about 50% o f horses in some populations have been exposed to S. neurona.^'**'*^ Immunosuppression It has been postulated that stress may play a role in the development of EPM, but limited evidence is available to support this hypothesis. The severity of EPM may be related to the size o f the infective dose, immune competency o f the host, and the environmental stresses to which the horse is exposed. A similar association between immunosuppression and disease has been documented in other species with EPM-like symptoms. Raccoons have been identified that were concurrently infected with a Sarcocystis spp-like protozoan and canine distemper virus.^* This is interesting as canine distemper virus is known to be immunosuppressive and has often been associated with cerebral toxoplasmosis in dogs, foxes, and raccoons. Sarcocystis spp infection of the CNS was also identified with a concurrent SIV infection in a monkey that developed asymmetric neurologic signs similar to those seen in EPM.^ Immune compromised people are often infected with Toxoplasma gondii and it has been demonstrated that stress plays a major role in the recrudescence o f the clinical signs of ToxopUtsma gondii-^ssoci^x.qd encephalitis.** Infections with either Neospora canirtum or Toxoplasm a gondii can cause T-cell hyporesponsiveness to the parasite antigen. It has also been demonstrated that an intact 16

40 T-cell response, specifically appropriate EL-12 and IFNy production, is necessary for resistance against either Neospora canirtum or Toxoplasma gondii. The parasite may therefore facilitate further infection by compromising host immune responses. Recent evidence suggests that there are neuropeptides (neuroimmune proteins NIP) released from the CNS when an animal is stressed, which may lead to suppression o f lymphocyte production and function. It has also been demonstrated that stress leads to high circulating glucocorticoid concentrations which are also immunosuppressive. The combination o f high resting concentrations of glucocorticoids and an increase in NIP release may result in immunosuppression and facilitate development o f clinical disease in horses infected with S. neurona. Further controlled investigations are needed to examine the role o f stress in the development of clinical signs of EPM in horses. 2.6 DIAGNOSIS Post-mortem Examination The diagnosis o f EPM has been difficult because o f a lack o f understanding of the pathogenesis of the disease, and the variety of clinical signs. Post-mortem examination was the first method used to definitively diagnose EPM and is still considered by many to be the gold standard for diagnosis. Grossly, the CNS lesions identified at post-mortem are described as multifocal areas o f hemorrhage to light discoloration of the brain or spinal cord.*"* Histology often reveals a marked mononuclear perivascular cuffing with necrosis and loss o f neurons, with infiltration o f monocytes, lymphocytes, some eosinophils and rarely, neutrophils.^ 17

41 Protozoan organisms can be seen in some o f the lesions, but are often difibcult to detect.^ DifiBculty in detecting the organisms increases if the animal has been treated with antiprotozoal medications. Immunohistochemical staining techniques can be used to definitively identify parasites in A significant problem with this method o f diagnosis is that by definition it cannot be applied in horses ante-mortem, and therefore cannot be applied to the largest majority o f clinical cases. A reliable diagnostic test that could be used for ante-mortem diagnosis is needed to better understand EPM and appropriately manage horses with this disease A ncillary Tests Cerebrospinal fluid analysis has been used to aid in determining the etiology of neurologic diseases in the horse. Early studies suggested that horses with EPM had mildly elevated CSF protein concentrations, increases in numbers o f mononuclear cells in addition to mild elevations in CSF enzyme activity (creatine kinase, CK; aspartate aminotransferase, AST). Two early studies reported marked elevations in the CSF CK activity in horses diagnosed with EPM.*** However, more recent studies suggest that neurologic disease o f horses cannot be reliably dififerentiated based on CSF leukocyte counts, CK activity, AST activity, or protein concentration. 18

42 2.6.3 Western Blot Analysis The ability o f S. neurona schizonts to survive multiple sub-passages allowed cultivation o f S. neurona using both bovine monocyte cell cultures and bovine pulmonary artery endothelial cells. These culture techniques were employed to develop an immunoassay for S. neurona^ Eight S. «ewrowa-specific proteins have been identified using cultured merozoite antigens in an immunoblot assay. While eight S. neuronaspecific proteins were identified, only 3 specific proteins (22.5, 13, and 10.5 Kd) were found in all horses with EPM and an S. /let/ro/ior-inoculated horse. The results of this study precipitated the development o f an ante-mortem diagnostic test for detection of S. neurona infection in horses. Finding S. neurona antibody in serum using this Western blot (WB) assay indicates that a horse has been infected, while finding antibody in CSF is thought to indicate that the parasite has penetrated the blood-brain barrier and stimulated a local immune response.a negative WB test result for CSF may theoretically occur in horses with EPM if the disease is so acute that the horse has not had time to mount an immune response, if affected horses fail to mount an immune response, or when infection is chronic. The latter is thought to rarely occur. False positive test results may occur if the blood brain barrier has been compromised by some other disease process, or when the CSF is contaminated during sampling. The immunoblot assay has not been routinely used to detect specific IgM. The sensitivity and specificity of the immunoblot have both been reported to be 89% based on comparison of WB results to results o f post-mortem in 19