• Open Access

Review of Equine Piroplasmosis

Authors

  • L.N. Wise,

    Corresponding author
    1. Department of Veterinary Microbiology and Pathology, Washington State University, Pullman, WA
    2. Animal Disease Research Unit, U.S. Department of Agriculture, Agricultural Research Service, Pullman, WA
    • Corresponding author: L.N. Wise, DVM, DACVIM, Department of Veterinary Microbiology and Pathology, Washington State University, PO Box 646630, Pullman, WA 99164; e-mail: nwise@vetmed.wsu.edu.

    Search for more papers by this author
  • L.S. Kappmeyer,

    1. Animal Disease Research Unit, U.S. Department of Agriculture, Agricultural Research Service, Pullman, WA
    Search for more papers by this author
  • R.H. Mealey,

    1. Department of Veterinary Microbiology and Pathology, Washington State University, Pullman, WA
    Search for more papers by this author
  • D.P. Knowles

    1. Department of Veterinary Microbiology and Pathology, Washington State University, Pullman, WA
    2. Animal Disease Research Unit, U.S. Department of Agriculture, Agricultural Research Service, Pullman, WA
    3. OIE Reference Laboratory for Equine Piroplasmosis, Pullman, WA
    Search for more papers by this author

Abstract

Equine piroplasmosis is caused by one of 2 erythrocytic parasites Babesia caballi or Theileria equi. Although the genus of the latter remains controversial, the most recent designation, Theileria, is utilized in this review. Shared pathogenesis includes tick-borne transmission and erythrolysis leading to anemia as the primary clinical outcome. Although both parasites are able to persist indefinitely in their equid hosts, thus far, only B. caballi transmits across tick generations. Pathogenesis further diverges after transmission to equids in that B. caballi immediately infects erythrocytes, whereas T. equi infects peripheral blood mononuclear cells. The recent re-emergence of T. equi in the United States has increased awareness of these tick-borne pathogens, especially in terms of diagnosis and control. This review focuses in part on factors leading to the re-emergence of infection and disease of these globally important pathogens.

Equine piroplasmosis is an infectious, tick-borne disease caused by the hemoprotozoan parasites Theileria equi and Babesia caballi. Piroplasmosis, which is also known in the literature as equine babesiosis, theileriosis (concerning T. equi), and biliary fever, affects all equid species, including horses, donkeys, mules, and zebras.[1, 2] Infection with either or both of these obligate, intraerythrocytic organisms can cause varying degrees of hemolytic anemia and associated systemic illness. As T. equi infection “silently” re-emerged in the United States, questions are being raised concerning the tick-vector-parasite-host requirements necessary for the development of clinical disease. The parasites and their natural tick vectors are endemic to most countries with tropical and subtropical climates.[3-5] The goals of control and disease eradication vary tremendously between endemic and non-endemic nations. Recent outbreaks of infection with limited disease expression in the United States and the Netherlands coupled with the identification of novel vectors within the United States have prompted a renewed interest in this historically important disease.[6-9]

Taxonomy of the causative agents of piroplasmosis has been in question since their discovery and remains controversial for T. equi.[10, 11] Currently, the parasites are classified within the phylum Apicomplexa, which contains other hemoprotozoan such as Plasmodium and Theileria. The parasite, termed Piroplasma equi (reclassified later as B. equi), was first recognized in South Africa as the causative agent of disease in 1901.[12] A few years later, it was discovered that 2 separate and distinct parasites could infect equid erythrocytes, one significantly larger than the other.[13] The larger of the parasites was termed Piroplasma caballi, only to be later reclassified as B. caballi.[14] Whereas B. caballi is considered a classic “babesia” species, the taxonomy of B. equi remains controversial. Based in part on finding an extra-erythrocytic stage within equine PBMC, B. equi was reclassified as T. equi in 1998.[11] Molecular phylogenetic investigations indicated that the organism possesses characteristics of both babesia and theileria lineages, possibly placing it between the two.[15-18] Recent genomic analysis supports the concept of a new genus for T. (B.) equi.[10] Additional data are needed to determine the final placement of this parasite and therefore this review will use the most recent designation, T. equi.

Epidemiology/Ecology

Piroplasmosis occurs in most countries worldwide and infection is maintained within equine populations as long as competent vectors are present.[19] For T. equi, the reservoir is the persistently infected equid; however, for B. caballi, both infected horses and the primary tick vector are reservoirs.[20-24] Although the precise tick-vector-parasite-host requirements for infection or clinical disease are not known, the outcome of increasing densities of infected horses and ticks in an area is an increase in infection and potentially disease.[25] Clinically silent transmission appears common.[7] The risk of life-threatening clinical disease increases with the presence of factors such as immunological naivety and increased density of infected ticks and horses.[26]

Although numerous studies have been published regarding the epidemiology and distribution of infection within specific countries and regions, these publications should be interpreted with caution given the profound variation in experimental design, sample population, and diagnostic testing. Particularly important is the previous use of the complement fixation test (CFT) for detection of antibody against B. caballi or T. equi. Abundant data have shown that the CFT lacks sensitivity in detecting persistent infections.[27]

Ixodid tick vectors occur in tropical, subtropical, and some temperate climates.[28] T. equi is more prevalent than B. caballi, yet B. caballi has been identified in more northern regions of the Northern Hemisphere than T. equi.[1] Given the currently available information on geographic distribution of infected horses according to the Office International des Epizooties/The World Organisation for Animal Health (OIE), Central and South America, Cuba, Africa, Asia, the Middle East and Southern Europe are considered endemic regions.[8] Within South America, the disease is readily identified in all regions with the exception of the southernmost areas of Chile and Argentina. The prevalence and distribution of the infection within the Caribbean nations are questionable, but the disease has been reported on most islands, including Trinidad and Cuba.[8, 29, 30] Disease is widespread in Africa and Asia with the highest prevalence reported in South Africa.[31, 32] Not all countries report identified cases to the OIE, making an accurate understanding of the current parasite distribution difficult. Countries such as Mexico and China are not considered endemic because the OIE routinely receives no information regarding distribution of piroplasmosis cases in those countries. Yet, articles from both Mexico and China have been published citing cases within those countries.[33-35] Compiling a list of currently non-endemic regions is equally challenging given the difference in surveillance, import/export restrictions, and disease reporting that occurs. Current disease status for those countries that report to the OIE can be found on their website.[36]

Detection of infection within the United States in recent years has placed the “piroplasmosis free” status of this country under scrutiny.[7, 37] The 1st case of B. caballi was identified in a horse in 1961 in southern Florida and while the source of infection was never definitively determined, it was speculated that it was attributable to importation of horses from Cuba.[38, 39] During the following decade during extensive surveillance programs, several hundred total cases of both B. caballi and T. equi were diagnosed in 7 states. Dermacentor (Anocenter) nitens, a tick capable of transmitting B. caballi, was identified in affected Florida counties, yet this documented vector's role in disease transmission was never confirmed. All infected horses were deported, quarantined, or treated with various drugs until a negative serologic result was obtained. Intense surveillance of horses and ticks continued in Florida until 1988 when the United States was deemed free of disease. To maintain this status, USDA APHIS improved restrictions on importation of horses from endemic areas.

In 2008, 20 T. equi-infected horses were identified on 7 separate premises in Florida.[37] All affected horses were associated with horses that had been imported from Mexico and all were engaged in illegal horse racing. Given the history and distribution of infected horses, inappropriate management practices including needle sharing and “blood doping” were assumed to be the mode of transmission. No tick vectors were identified despite aggressive surveillance. More recently in 2009, an outbreak of T. equi was identified on a ranch in southern Texas involving approximately 400 horses.[7] All infected horses resided on the premises or had been at some time associated with the ranch. An investigation of this property allowed identification of 2 previously unrecognized competent vectors of T. equi within the United States (Amblyomma cajennense and Dermacentor variabilis). Ticks collected from the infected horses on the premises were capable of biologically transmitting disease to naïve horses.[9]

Competent ticks and iatrogenic blood transfers are efficient modes of transmission.[19-21, 25, 40] With over 850 tick species worldwide and approximately 85 within the United States, the potential for transmission is high.[28] However, the presence of a competent tick vector and infected horses within the same area does not always lead to further infection or disease. Many factors must be considered including season, climate, host-specificity, and the particulars of a competent tick's life cycle.[25]

The life cycle of a tick involves 4 life stages: egg, larva, nymph, and adult. After hatching from an egg, the larva feeds on its host and molts into a nymph. The nymph then feeds and molts into an adult. Females and males proceed through these life stages, but the female dies after laying her eggs. Adult male ticks seeking females have the potential to feed on multiple hosts. Ticks are classified as hard ticks (Ixodidae) or soft ticks (Argasidae).[28] Both are vectors for pathogen transmission, yet only hard ticks are natural vectors for B. caballi and T. equi.

Tick transmission can occur via 3 forms: intrastadial, transtadial, or transovarial. Intrastadial transmission occurs when acquisition and transmission of the parasite occurs within 1 life stage (no stage transition before transmission). Transtadial describes acquisition of infection in 1 stage and the ability for the same tick to transmit the infection during subsequent life stages. The parasite is maintained within the tick as it develops. Transovarial transmission occurs when the female acquires parasites, which enter ovaries and are transmitted to offspring, allowing maintenance of the parasites across tick generations.[20]

Multiple ixodid tick species have been identified as either natural or experimental vectors of piroplasmosis. B caballi is transmitted by 15 separate species (7 Dermacentor [Anocenter] sp., 6 Hyalomma sp., and 2 Rhiphicephalus sp.) and T. equi by 14 species (4 Dermacentor sp., 4 Hyalomma sp., 5 Rhiphicephalus (Boophilus) sp., and Acajennense). B. caballi is transmitted transtadially and transovarially by its vectors.[41] T. equi is generally transmitted through transstadial and intrastadial transmission.[20] Transovarial transmission of T. equi occurs, but the precise role in epidemiology has not been detailed.[42, 43]

Discovery of the vector D. reticulatus in the Netherlands in 2010 combined with recognition of a subclinically B. caballi-infected horse led to a surveillance of that area that resulted in identification of several of T. equi- and B. caballi-infected horses.[6] Before 2009, only 2 tick species known to transmit T. equi naturally had been identified within the southernmost parts of the United States: D. nitens and R. microplus. Within the Texas outbreak, infected horses were found to be infested with 4 different tick species: A. cajennense, A. maculatum, Dermacentor (Anocenter) nitens, and D. variabilis. A. cajennense was the most abundant, being identified on approximately 79% of the infected horses, followed by A. maculatum (19%), Dvariabilis (16%), and D. nitens (3%). Before this outbreak, A. cajennense had not been identified as a vector for T. equi. The geographic distribution of these 3 host ticks within the United States appears to be limited to Texas and Florida.[44] The role of A. maculatum in the outbreak currently remains unclear and although D. nitens is a known vector of B. caballi, evidence supporting transmission of T. equi is lacking. Adult Dvariabilis ticks collected from horses on the ranch were able to experimentally transmit disease to naïve horses, but their role as a natural vector in this outbreak remains questionable.[9]

Pathogenesis (Transmission/Life Cycle of Parasites)

The life cycle of both B. caballi and T. equi involves distinct stages that occur in the host and tick[45, 46] (Fig 1, 2). Both parasites progress through 3 life stages: the sporozoite (asexual transmission stage), the merozoite (asexual blood stage), and the gametocyte (sexual blood stage). The development of these parasites within the tick can vary depending on species of tick involved. Regardless of species variation, for both Bcaballi and T. equi, infectious sporozoites are transmitted through the tick-saliva to the equid host. The duration of the extrinsic incubation period (acquisition feed; parasite replication time in the tick vector and transmission feed) is not defined for Bcaballi or T. equi. Once within the host, Bcaballi sporozoites directly invade erythrocytes where they multiply and develop first into trophozoites and then into merozoites. After erythrocyte rupture, merozoites are released and invade other erythrocytes. Tequi's initial invasion is different in that it first enters PBMCs.[11] This life cycle event is part of the justification for the most recent taxonomic classification as Theileria.[10] Inside PBMCs, Tequi sporozoites develop into large schizonts and after approximately 9 days, merozoites are released and invade erythrocytes. For both parasites, asexual replication results in an expanding population of merozoites and parasitized erythrocytes. Some merozoites develop into gametocyte forms within equine peripheral blood. Upon ingestion of merozoites (and/or gametocytes) by a competent tick, the parasites undergo sexual reproduction, with gametocytes developing into gametes, which combine to form zygotes within the tick midgut. The zygotes develop differently depending on the tick species and the parasite. After a period of 6–24 days, continued development results in the presence of sporozoites within the salivary gland of the tick.[43, 47, 48]

Figure 1.

Life cycle of Babesia caballi. Illustration by Massaro Ueti.

Figure 2.

Life cycle of Theileria equi. Illustration by Massaro Ueti.

Transmission can occur iatrogenically through inappropriate mixing of the infected and uninfected blood.[37] This occurs most frequently during the practice of needle sharing between positive and naïve horses, but use of any blood-contaminated equipment could result in transmission.[49] Infection can also be caused when chronically infected horses serve as blood donors to naïve horses. The illegal practice of “blood doping” (prerace blood transfusions) was implicated in the 2008 Florida outbreak.[37] Experimental infection can be achieved through intravenous and subcutaneous routes as well as tick transmission.[20, 50-52]

Although some details of pathogenesis remain unknown, infection with either T. equi or B. caballi causes erythrocyte lysis resulting in varying degrees of hemolytic anemia. Physical rupture of erythrocytes during release of merozoite stages causes intravascular hemolytic anemia. Infected red blood cells are removed from the circulation by splenic macrophages further contributing to hemolytic anemia. Nonparasitized erythrocytes are also removed from circulation, but the reason for this phenomenon is unknown. Data from experimentally infected splenectomized donkeys indicate that the biochemical structure of erythrocyte membranes changes dramatically during infection with T. equi.[53] It was suggested that this conformational change causes decreased deformability of the red cells, which could lead to reduced microvascular blood flow. Plasma levels of malondialdehyde (marker of lipid peroxidation) are significantly increased, suggesting that an accumulation of oxidative ions also contributes to erythrocyte lysis.[53] These parasites also alter coagulation in infected horses through unknown mechanisms. B. caballi-infected erythrocytes cause microthrombi by clumping within small vessels, leading to venous stasis and vasculitis.[52, 54] Varying degrees of thrombocytopenia and prolonged clotting times have been reported during infection with T. equi and B. caballi.[55] Hypotheses regarding the pathogenesis of decreased platelet counts include immune-mediated destruction, splenic sequestration and/or excess consumption as is observed in disseminated intravascular coagulation. Severe piroplasmosis can result in hypercoagulability, systemic inflammatory response syndrome, and subsequent multiorgan system dysfunction.[56]

Placental transmission from infected carrier mares to their fetuses has been documented.[15, 57-59] This transmission can result in abortion (most commonly in late gestation), stillbirth, or neonatal infection and can occur across placentas that are histologically normal. The prevalence of this type of transmission is unknown. The natural outbreak documented in Texas of 2009 resulted in infection of pregnant mares, none of which transmitted infection to their foals based on serial negative polymerase chain reaction (PCR) results. Conversely, T. equi has been reported to be responsible for 11% of all abortions in South Africa.[60] Based on the variation in reported occurrence, it is likely that individual horse genetics or geographic isolate/strain differences could influence the prevalence of placental transmission. Exposure to semen from an infected stallion is not considered to be a means of transmission, yet blood contamination during breeding practices could present a transmission risk.[61]

In most cases, the horses become persistently infected and become inapparent carriers. The inapparent carrier state is life-long with T. equi and possibly for B. caballi. A number of accounts indicate that horses infected with B. caballi can undergo self-clearance of the parasite without treatment.[1, 3, 5] It is assumed that persistent subclinical infection is due in part to sequestration of the organism and immune evasion strategies. Various theories for location of parasite sequestration in asymptomatic horses have been reported, including capillaries, central nervous system vasculature, and bone marrow.[4, 62, 63] The mechanism(s) of persistent infection remain unknown.

After transmission, depending on factors including parasite dose and immunity, clinical signs develop within 10–30 days for B. caballi and 12–19 days for T. equi.[3] The fatality rate of naïve horses in endemic nations has been estimated at 5–10% dependent on parasite, transmission dose, overall health of the horse, and administration of treatment.[4] Uniformly, infection with T. equi typically results in more severe clinical disease than B. caballi.[4] Disease signs and severity can vary significantly from 1 region to another.[7, 31]

Clinical Disease

Clinical disease can manifest in different forms. With acute T. equi infection, clinical signs are usually related to marked hemolysis and resulting anemia. Although Bcaballi-infected horses do become anemic as well, the rare cases of acute death from B. caballi have occurred reportedly as a result of multiple organ dysfunction related to systemic formation of microthrombi and development of disseminated intravascular coagulation.[56] In these cases, the exact clinical signs vary depending on the organ system affected.

Horses with acute infection initially develop nonspecific signs such as high fevers, sometimes in excess of 104°F, lethargy, anorexia, weight loss, and peripheral edema.[4] Petechiations caused by thrombocytopenia are often observed on mucous membranes, including the nictitating membrane.[3] Signs of hemolytic anemia follow and include icteric or pale mucous membranes, tachycardia, tachypnea, weakness, and pigmenturia (because of either hemoglobinuria or bilirubinuria).[53, 64] Some horses show signs of gastrointestinal complications including colic or impactions followed by diarrhea. Other less common clinical presentations include secondary development of pneumonia, pulmonary edema, cardiac arrhythmias, catarrhal enteritis, laminitis, and central nervous system disease characterized by ataxia, myalgia, and seizures.[32, 39, 64, 65] Temporary or permanent infertility has been reported in stallions.[56] Acute renal failure occurs as a result of hemoglobin-induced pigment nephropathy and systemic responses to severe inflammation (hypotension) can worsen the kidney disease.[3] Severe infections can also culminate in liver failure or disseminated intravascular coagulation.[56]

Fulminant, abrupt onset of signs of disease, termed peracute disease, has been documented. Collapse and sudden death from overwhelming T. equi can occur and introduction of naïve horses into an endemic region can lead to rapid onset of severe disease. In the 1930s, relocation of a group of naïve horses into an endemic area of southern France resulted in a 69% fatality rate.[26] A report from Jordan documented 5 presumed inapparent carriers undergoing strenuous exercise that immediately after completion of the exercise, developed profound weakness with two dying suddenly.[66] Recrudescence of marked T. equi parasitemia was assumed. Neonatal foals infected in utero with T. equi can present with acute, severe signs.[57, 58, 67-71] These foals can exhibit clinical signs at birth or can become ill at 2–3 days of age. Clinical signs are often nonspecific, such as weakness and decreased suckling, but progress to resemble those of an infected adult, including icterus, fever, and anemia (with or without petechiations and hemoglobuinuria). Cases of B. caballi fetal and neonatal infection have been reported but are rare.[70, 72]

Chronic T. equi or B. caballi infection can result only in nonspecific signs, including lethargy, partial anorexia, weight loss, and poor performance. Mild anemia might be present and the spleen might be enlarged upon rectal palpation. It has been suggested that splenic enlargement is caused by the increased rate of extravascular hemolysis that occurs within the spleen in less severely affected horses.[3, 5, 52]

Importantly, horses infected with either T. equi or B. caballi in both endemic and non-endemic regions are most commonly inapparent carriers with no appreciable signs of disease. Pregnancy in carrier mares can result in abortion or neonatal infection.[15, 32] Because inapparent carriers can serve as reservoirs for transmission via ticks, placentally or iatrogenically, these horses represent the largest challenge to non-endemic nations attempting to prevent apparently healthy carriers from crossing their borders.[20, 73]

An appropriate list of differential diagnoses should be determined based on whether the horse resides in or has visited an endemic region. In general, acute onset of the aforementioned clinical signs could also be caused by equine infectious anemia virus, African horse sickness virus, equine viral arteritis virus, equine ehrlichiosis, purpura hemorrhagica, immune-mediated hemolytic anemia, and red maple leaf toxicity.[8, 74]

Results of laboratory analyses may aid in diagnosis. Most horses regardless of clinical syndrome exhibit some degree of anemia characterized by decreased packed cell volume, hemoglobin, and erythrocyte count. Although acutely infected horses can have profound anemia with packed cell volumes (PCV) as low as 10%, the PCV rarely falls below 20%.[52, 54, 64] Red blood cell indices, MCV, MCH, and MCHC are variable.[53] Thrombocytopenia is commonly identified.[3, 52, 54, 55, 64] One report noted a decrease in platelet count in 39% of T. equi infections, 80% of B. caballi infections, and 100% of dual infection.[54, 64, 75] Clotting times can be prolonged or normal. The leukogram can vary depending on infection stage and severity.[76] Fibrinogen concentration can be elevated and albumin concentration can vary depending on hydration status, chronicity of the disease, and associated conditions that result in protein loss.[55] Hyperbilirubinemia is often observed and the liver enzyme activities, ALP, AST, and GGT can be elevated.[64] These elevations are attributed to reduced blood flow to the liver, which can in severe cases result in centrilobular necrosis. Hypophosphatemia and hypoferremia are common, attributed to altered erythrocytic metabolism.[77] Infected erythrocytes can be identified in sternal bone marrow aspirates of asymptomatic horses, but utility of this test as a diagnostic tool is limited.[62]

Gross and histopathologic findings at necropsy vary depending on the severity of disease and associated complications. Gross examination might demonstrate evidence of anemia as well as varying degrees of icterus, edema, and splenomegaly. Other findings can include pulmonary edema and congestion, cardiac hemorrhages, hydropericardium, hydrothorax, hepatomegaly, ascites, enlarged discolored kidneys, and lymphadenopathy.[3] Histopathologic findings can include centrilobular necrosis of the liver, renal tubular necrosis with hemoglobin casts, and microthrombi within the liver and lungs.[32] Pulmonary tissue examination can demonstrate congestion, edema, and hemosiderin-laden macrophages within the pulmonary alveolar walls. Parasites can be observed within red blood cells within blood vessels and within macrophages in the lymph nodes.[3, 39, 78]

The apparent global variation in clinical disease might be caused by a variety of factors contributing to emergence of infection and disease. Infection with T. equi in South Africa often results in severe disease, requiring treatment.[31, 32] In contrast, in the outbreak identified in the United States, in which 475 horses were affected, only 1 horse was reported to exhibit mild clinical signs.[7] It is difficult to compare these instances because the differences between these T. equi strains are unknown, but it at minimum provides a single comparison between non-endemic and endemic countries.

Immunity

The response of the equine immune system to infection with T. equi or B. caballi is not completely defined, but is undoubtedly complex and multifaceted. It is well accepted that infection with either parasite results in carrier status, which confers protection against disease. There is no documented cross-protection between T. equi and B. caballi, as horses can be infected with both parasites simultaneously.[4]

The spleen plays a necessary role in control of most hemoprotozoan parasites. A horse with a spleen is typically able to overcome acute T. equi-induced disease, whereas splenectomized horses invariably succumb to disease with parasitemias that can reach 80%.[51, 53, 79] Inapparent carriers of T. equi will also develop terminal disease upon splenectomy.[80] Splenectomized horses inoculated experimentally with B. caballi may or may not develop obvious clinical disease, but death caused by infection has been documented.[81] This discrepancy may be attributable to differences in parasite strain, infective dose, the overall health of the horse, or a combination of factors.

Although essential in other hemoprotozoan infections like Babesia bovis, the function of cell-mediated immunity in piroplasmosis has yet to be fully determined.[82, 83] Production of nitric oxide by macrophages might be an essential effector mechanism of immune control against experimental B. caballi infection.[84]

Importantly, innate immune responses and the presence of a spleen are not sufficient to control T. equi infection, because spleen-intact foals with severe combined immunodeficiency (SCID) are unable to control T. equi parasitemia[85] (Fig 3). Although innate immunity is unaffected, SCID foals lack functional T and B lymphocytes and are incapable of mounting antigen-specific antibody and cellular immune responses.[85-91] Inoculation of these foals with T. equi resulted in fulminant, severe infection within 7 days, and subsequent death. Terminal parasitemias ranged between 29 and 41% and PCV decreased by 50% of the normal value. Thus, adaptive immune responses are required to control T. equi parasitemia, but are not required for lysis of erythrocytes.

Figure 3.

Electron micrograph of a splenic macrophage from a severe combined immunodeficiency foal showing phagocytized Theileria equi-infected erythrocytes. Arrow denotes the organism inside an erythrocyte and the asterisk illustrates the nucleus of the splenic macrophage. Image courtesy of Lowell Kappmeyer.

Antibody responses correlate with control of parasitemia.[92] T. equi-infected horses produce antibodies against immunodominant merozoite proteins termed equi merozoite antigens (EMAs), which are surface expressed on merozoites.[85] The exact role of this antibody response in immunity and persistence remains unclear. The nomenclature of equine immunoglobulin G has recently been adjusted to accommodate the 7 unique IgG heavy chain genes discovered in the equine genome.[93] In the acute stages of T. equi infection, high levels of IgGa (now IgG1) and IgGb (now IgG4&7) correlate with control, whereas IgG(T) levels (now predominately IgG5 and to a lesser extent IgG3) increase after resolution of parasitemia during the chronic phase of infection.[92] Antibodies are first detected within 7–11 days after natural infection and peak at 30–45 days. All examined subclasses remain detectable into the chronic/inapparent stage of disease.

The correlates of adaptive immune responses to T. equi infection are currently unknown. Donkeys vaccinated with T. equi immunogen were able to mount a protective response characterized by high antimerozoite antibody titers and merozoite-specific lymphocyte proliferation.[83] Recently, it was documented that SCID foals that received repeated infusions of T. equi hyperimmune plasma before inoculation with the parasite were able to delay the time to peak parasitemia.[94] All foals developed disease, but partial protection was associated predominantly with transfused IgG3-specific antimerozoite antibodies. Overall, additional research is needed to define the protective roles of antibody and cellular adaptive immune responses against T. equi.

Even less is known about protective immune responses against B. caballi infection. Infected horses produce antibodies to rhoptry associated protein-1 (RAP-1), which is utilized on serologic detection of infection. This conserved apical merozoite protein remains partially uncharacterized in B. caballi, but in B. bovis plays a pivotal role in induction of humoral immunity.[95]

In most endemic areas, foals that ingest colostrum from a carrier mare are protected from infection and clinical disease for the first 1–5 months and can be protected up to 9 months of age.[3, 56] As maternal antibodies decline, the foal becomes susceptible and most young horses in endemic nations are infected by age 2. It has also been suggested that foals can be born as healthy inapparent carriers of T. equi, which would also infer some level of protection.[56]

Rarely, inapparent T. equi carriers can exhibit relapses of clinical disease associated with stress, strenuous exercise, immunosuppression, and steroid administration.[66, 96, 97] Experimental treatment with beclamethasone at a dose of 0.1 mg/kg once daily for 5 days before and 5 days after inoculation with T. equi resulted in a 50% increase in parasitemia as compared with controls.[96] These relapses have not been reported for B. caballi.

Diagnosis

Various diagnostic modalities can be used alone or in combination to diagnose infection. During management of an outbreak within a non-endemic nation, involvement of the state and national regulatory agencies is required and often, multiple diagnostic methods will be utilized in an effort to obtain the most accurate information. Only a few laboratories in the world are authorized to perform certain tests, so proper handling of samples is crucial.

Light microscopy can be used to identify the organisms within the erythrocytes. A thin blood smear, stained with Giemsa, Wright's, or Diff-Quik®, may reveal organisms during the acute stage of infection. The smears must be thoroughly examined since even during severe infection, the percent parasitemia remains so low that false-negative results are not uncommon.[5, 98] The piroplasms of T. equi and B. caballi can be easily distinguished from one another. Within the erythrocyte, B. caballi typically appears as 2 large pyriform (pear-shaped) merozoites that measure approximately 2–5 μm in length (Fig 4). During clinical infection with B. caballi, the percentage of erythrocytes parasitized is typically less than 1% and may be less than 0.1%. T. equi merozoites occur within erythrocytes as polymorphic, small piroplasms occasionally in a distinct Maltese cross-formation (Fig 5). The T. equi merozoites are smaller and typically measure 2–3 μm in length.[3] The percent of infected erythrocytes during clinical disease caused by T. equi is usually between 1 and 5%, but in severe cases can exceed 20%.[56] In cases of chronic or inapparent infection, parasite numbers remain too low for reliable detection on blood smear.[99]

Figure 4.

Equine erythrocyte containing Babesia caballi merozoites. Diff-Quik®, ×100 oil magnification. Image courtesy of Peter Awinda.

Figure 5.

Equine erythrocyte containing Theileria equi merozoites in characteristic Maltese cross-formation. Diff-Quik®, ×100 oil magnification.

Several serologic tests were developed to increase diagnostic sensitivity, especially in those carrier horses exhibiting no clinical signs. These tests include the CFT, indirect immunofluorescence assay, western blot, and competitive enzyme-linked immunosorbent assay.

The CFT relies on activation of complement upon specific interaction of antibody and antigen.[99] A positive result is defined as a positive reaction at a dilution of 1 : 5. Infected horses seroconvert on CFT approximately 8–11 days after infection with titers beginning to decline at 2–3 months.[99] CFT is a very specific test, yet lacks sensitivity, especially in chronic infection or after treatment.[27] Horses can transiently become negative 3–15 months after treatment for B. caballi and 24 months for T. equi.[99, 100] IgG(T), now classified as IgG5 and to a lesser extent IgG3, remains elevated in chronic T. equi infections. IgG(T) does not fix complement and recently, it was demonstrated that IgG3 fixes complement, whereas IgG5 does not.[93, 101] Thus, it is not surprising that the CFT lacks sensitivity for diagnosis of chronic or inapparent T. equi infection. Cross-reactivity between antibodies against T. equi and B. caballi when using the CFT has been reported.[99, 102] Regardless of the fact that the CFT was previously the official regulatory test for establishing piroplasmosis status before travel to a non-endemic country, the CFT is not considered the diagnostic test of choice for chronic infection.[98]

Indirect immunofluorescent antibody tests (IFAT) demonstrates high specificity, but lacks sensitivity. It is, however, considered more sensitive than the CFT.[99] In this test, fluorescently labeled antibodies react with antigen bound to a glass slide. A sample is considered positive if strong fluorescence is noted at a dilution of 1 : 80 or higher. Experimental intravenous infection with T. equi or B. caballi resulted in positive IFAT results at days 3–20 post infection.[100] Titers were more consistently detected and remained elevated longer with the IFAT versus the CFT.[103] Typically, the IFAT is used as an adjunct test to aid in analysis of CFT results, but it remains one of the prescribed tests for equine piroplasmosis recommended by the OIE.[98]

Until recently, Western blot (or immunoblot) has been utilized primarily in a research setting for diagnosis of T. equi and B. caballi. The National Veterinary Services Laboratory in Ames, Iowa, is now offering an immunoblot as an adjunct diagnostic tool for detection of B. caballi infection.[104] Research is under way to validate these tests for use in routine diagnosis.

Since 2004, the competitive inhibition enzyme-linked immunosorbent assay (cELISA) has been one of the regulatory tests prescribed by the OIE for international horse transport.[98] The test is considered to be the most sensitive means of detection of chronic T. equi infection.[27] The cELISA for T. equi utilizes recombinant EMA-1 and specific monoclonal antibodies.[105] EMA-1 is an immunodominant, highly conserved surface antigen specific to T. equi. Horses infected with T. equi are detectable with the cELISA as early as 21 days after experimental infection and approximately 5 weeks after tick transmission.[105] The EMA-1 cELISA is validated for use against multiple different strains of T. equi found around the world.[27] Generation of a recombinant form of this epitope and associated monoclonal antibodies allowed standardization of this test and markedly increased sensitivity as compared with the other serologic tests.[27, 106, 107]

A recombinant form of RAP-1 was also developed for the B. caballi cELISA.[108] This test, upon comparison with CFT using 300 equine serum samples from around the world, was able to diagnose infection in 25% more cases than with the CFT. However, the currently available RAP-1 cELISA relies on the recognition of epitopes that are not conserved across all B. caballi strains. Because of sequence heterogeneity between the recombinant RAP-1 used in the test and South African isolates of B. caballi, the test is unable to detect infected horses in South Africa.[109] Although similar differences exist between the recombinant EMA-1 used in the T. equi cELISA and South African T. equi isolates, the cELISA continues to detect infected horses in South Africa.[110] Both cELISAs are marketed by VMRD (Pullman, WA) and are not available to general practitioners. Lastly, a field validated ELISA using whole T. equi merozoite antigen has been described as an easy, economical, and reliable test.[111]

Recently, the previously reported data on the specificity and sensitivity of the CFT and cELISA were statistically analyzed.[25, 106, 108, 112] Overall, the sensitivity of the CFT to detect T. equi is 47% and the cELISA is 96%. The specificities of the 2 tests are 94% and 95%, respectively. For testing of B. caballi, the sensitivity of the CFT is 88% and the cELISA is 91%. The specificities of the 2 tests for B. caballi are 98% and 70%, respectively.

Polymerase chain reaction tests for the presence of the organism of interest by amplifying and detecting specific fractions of the DNA. This test is exquisitely sensitive and thus far has only been utilized in research settings for the detection of T. equi. Three variations of PCR include real-time PCR, nested PCR, and nested PCR with hybridization.[113-117] Comparison of these tests and the results from these tests is difficult, given that no standardization exists between laboratories. Nested PCR for T. equi using the EMA-1 gene sequence has been shown to detect a positive result equivalent to a percent parasitemia of 0.000006%.[115] One report also indicated that nested PCR detects 3.6× more infections that microscopy and 2.2× more than traditional PCR.[30] The validity of nested PCR routinely in the United States has been questioned as a diagnostic tool for horses in South Africa.[110, 118] Upon examination of the genetic composition of EMA-1, it was recognized that the strains from around the world were not 100% homologous. These discoveries further emphasize the issues involved in standardization of PCR as a diagnostic test. Thus far, the use of nested PCR for detection of B. caballi DNA in chronically infected horses has proven unreliable. As it is currently performed, it is unlikely that nested PCR will ever be standardized and commercially marketed for detection of T. equi or B. caballi infection.[119]

With increasing use of PCR in laboratory settings, the validity of a positive cELISA result has come into question. Despite treatment and apparent clearance of T. equi, as demonstrated by negative PCR and transmission studies, cELISA results remain positive, sometimes for up to 24 months after clearance.[7, 120] The transmission risk that these PCR-negative, seropositive horses pose must be determined.

Treatment

In endemic regions, treatment of piroplasmosis is used only as a means of decreasing clinical signs and reducing fatalities. Clearance of the organism serves no purpose in these countries as life-long immunity (premunity) is assumed to be conferred with chronic, inapparent infection. In non-endemic regions attempting to remain free of piroplasmosis, treatment of infected horses with the intent of clearance (chemosterilization) is desired. T. equi infections are more typically difficult to treat than B. caballi infections. Numerous drugs have been reported to have variable efficacy in inhibiting T. equi and B. caballi both in cell culture and in vivo, making the literature difficult to interpret.[117, 121-128] Historically, it was reported that B. caballi infection was self-limiting with clearance noted after several years, yet this is not always the case.[63] Chemotherapeutic clearance of T. equi in the horse has been previously reported, yet the research was conducted before the development of tests with increased sensitivity for parasite persistence.[122, 123] Until recently, it was widely accepted that chemosterilization of a T. equi-infected horse was unachievable. Data collected during the outbreak in Texas indicate that T. equi can be eliminated from an infected horse with appropriate dosing of imidocarb diproprionate.[7]

For alleviation of clinical signs, several drugs have been used with success, yet imidocarb, in its diproprionate salt form (ID), is considered to be the most effective. The alternate form of this drug, composed of dihydrochloride salt, will cause more severe muscle damage at the site of injection.[122, 129] ID, a carbanilide derivative, is typically administered to horses intramuscularly. Although its mechanism of action remains unclear, proposed mechanisms include inhibition of inositol entry into infected erythrocytes or alteration in metabolism of polyamines.[130, 131] After intramuscular (IM) injection, ID is rapidly eliminated from the plasma, yet remains sequestered in certain tissues of the body.[132] Reported dosages for alleviation of clinical signs vary, yet most sources indicate that 2.2–4.4 mg/kg given IM once is effective. If necessary, lower dosages can be repeated at 24–72 hour intervals for 2–3 treatments. In non-endemic nations where chemotherapeutic clearance of the organism is desired, animals infected with B. caballi can be cleared with a dose of 4.4 mg/kg IM every 72 hours for 4 treatments.[117] For clearance of T. equi, data from both naturally and experimentally infected horses indicate that the same dose is effective.[7, 120] Of the 25 naturally infected treated horses, one remained positive for T. equi after the initial treatment and had to undergo a 2nd treatment regimen to obtain chemosterilization.[7] In the study using ID to attempt clearance in experimentally T. equi-infected horses, 1 horse remained resistant to parasite elimination.[120] Currently, clearance is determined by negative PCR results and inability to transmit the parasite to a naïve splenectomized horse via blood transfusion, but studies are under way to more clearly define parasite elimination.[7] In the United States, if a horse is diagnosed with T. equi, the owner and veterinarian can enroll the horse in the USDA controlled treatment program as to ensure appropriate quarantine, treatment, and subsequent release of cleared horses.[120]

ID has anticholinesterase activity, so reactions to the drug often present as sweating, signs of agitation, colic, and diarrhea.[129, 133] Typically, these signs are transient and rarely life-threatening. Effects can be prevented with an intravenous dose of glycopyrolate at 0.0025 mg/kg once or reversed with a single intravenous dose of atropine at 0.2 mg/kg. Both of these anticholinergic drugs cause adverse effects as well. Administration of the anticholinergic n-butylscopolamine1 can lessen clinical signs without addition of adverse effects.[7] As ID undergoes hepatic and renal clearance, periportal hepatic necrosis and renal tubular necrosis can occur with toxicity.[129, 133] Horses undergoing treatment with this drug should be monitored carefully for development of complications. Transient azotemia, elevations in urinary GGT/creatinine ratios, or both as well as transient elevations in liver enzyme activities (AST, ALT, ALP, and SDH) can be observed during treatment, but typically resolve with discontinuation of the drug. Donkeys and mules are exquisitely sensitive to ID, therefore its use in these species is not recommended.[122] Information regarding treatment of T. equi-infected neonatal foals with ID is limited with only one case report.[71] When a nursing mare is given a single dose of at 2.4 mg/kg IM, ID is detectable in the milk 2 hours after administration.[132] It remains unclear if this could lead to toxicity in the suckling foal. One report of administration of ID to pregnant mares followed by induced abortion resulted in circulating levels of ID in each fetus comparable to the dam's serum concentrations.[60]

Diminazene aceturate and diaminazene diaceturate have been used with success against T. equi and B. caballi at a dose of 3.5 mg/kg IM every 48 hours for 2 treatments.[134] Diminazene aceturate is more effective than diaminazene diaceturate, but both drugs have been reported to cause significant injection site muscle damage. Efficacy of both drugs increases with the 2nd dosage, yet no chemosterilization has been reported. Signs of toxicity include respiratory distress and lethargy.

The antibiotic oxytetracycline when administered IV at a dose of 5–6 mg/kg once daily for 7 days is effective against T. equi, but not against B. caballi.[64] Other drugs reported to have efficacy in treatment of babesiosis include amicarbilade isothionate, euflavine, artesunate and arteether (arteminsin derivatives), buparvaquone, and atovaquone, yet these drugs are no longer commonly utilized in practice.[56, 124, 125] Ponazuril inhibits T. equi organisms in vitro.[135] No in vivo research has been conducted to date, yet it is possible that this class of drugs could offer additional treatment options in the horse.

Aside from antiprotozoal drugs, acutely infected horses often require supportive care including but not limited to intravenous fluids, nonsteroidal anti-inflammatory drugs, pain management, and blood transfusions. Adequate hydration is essential upon initiation of and during treatment with imidocarb.

Prevention/Vaccination

Prevention of infection in endemic nations is virtually impossible, and it is assumed that the premunity conferred with initial infection acts to protect the horse from recurrent disease upon subsequent exposures. In non-endemic nations, the cornerstone of protection is regulation of equine movement between endemic nations. Depending on the non-endemic country in question, horses must test negative for T. equi and B. caballi on the serologic test designated specifically by the country of import, typically the cELISA or the IFAT. If positive, horses are generally denied entrance unless for tightly regulated athletic events. All imported horses from endemic nations undergo strict quarantine and are examined thoroughly for ticks. Application of acaricides before removal from the endemic nation is used to ensure that ticks are not introduced with the horses. The regulatory system put in place by the OIE has been successful, yet isolated cases continue to occur in non-endemic nations. These isolated cases are rarely caused by tick transmission and are most often linked to the use of blood contaminated equipment and practices involving needle sharing or blood transfusions from untested donors.[37, 39, 78] In the United States, a horse that is identified as positive on cELISA must be immediately quarantined and the state and federal authorities must be notified. Until a reliable means of sterilization is identified, these horses must remain quarantined, be exported, donated to research facilities, or euthanized. The most appropriate action is determined by the state and federal regulatory officials on a case-by-case basis.[25, 120]

Non-endemic nations that border endemic nations cannot completely prevent introduction of ticks, so diligent measures must be taken to reduce horses' contact with ticks. This includes routine application of acaricides, surveillance for the presence of ticks, and reduction in vegetation.[73] A variety of chemical are available to reduce tick exposure, but not all effective acaricides conform to EPA regulations, so careful selection and use of these products is necessary.[25] Although no direct scientific evidence exists in horses, systemic administration of the dewormer ivermectin likely aids in control of blood-feeding ticks. Being aware of the habitat and the seasonality of ticks in your area is important. Guidelines for examining horses for ticks and appropriate removal of ticks have been published by the USDA.[73]

Several studies have assessed the potential use of vaccination to induce immunity to B. caballi and T. equi infection, yet no vaccine is commercially available. Immunization with whole merozoite induced protection in 4 donkeys.[83] The correlates of protective immunity in horses are unknown, and until these are elucidated, developing an effective vaccine will be difficult.

Vaccination and treatment strategies are dependent on the infection status (endemic versus non-endemic) of the region. A transmission blocking vaccine in the absence of full compliance in at-risk horses would not be appropriate in an endemic region. A vaccine preventing transmission or clinical disease and death is needed for naïve horses moved into endemic regions. Chemotherapeutics, which aid in control of acute parasitemia and associated clinical signs, but do not eliminate infection, are important in endemic regions. In non-endemic regions, the goal is remaining infection free; therefore, when infected horses are detected, a safe chemotherapeutic with efficacy for eliminating persistent infection is needed. To remain infection free, necessary tools include accurate screening diagnostic tests and thorough knowledge of tick populations and their ability to transmit Bcaballi, T. equi, both.

Increasing globalization of the equine industry coupled with constantly changing climates provides major challenges for controlling persistent infections as those caused by T. equi and B. caballi. As with all vector-borne infections, measures must be taken to enhance control through surveillance of equine populations and detailed knowledge of vector competence and habitat. Detailed information regarding life cycles, transmission, immune responses, accurate diagnostics, and treatment of these parasites is paramount to control these insidious diseases, while promoting growth in the movement of horses internationally.

Acknowledgments

Conflict of Interest Declaration: The authors disclose no conflict of interest.

Footnote

  1. 1

    Buscopan; Boehringer Ingelheim, Rheinland-Pfalz, Germany

Ancillary