Equine Granulocytic Anaplasmosis: A Systematic Review and Meta-Analysis on Clinico-Pathological Findings, Diagnosis, and Therapeutic Management

Simple Summary Equine granulocytic anaplasmosis (EGA) is a tick-borne disease affecting horses worldwide, caused by Anaplasma phagocytophilum. The disease ranges from non-specific clinical signs to fatal outcomes. A large number of horses has been tested for serological and molecular surveys concerning EGA, but the number of clinical EGA reports is smaller and mostly as single or case series reports. Therefore, the aim of this paper was to analyze reported cases regarding clinico-pathology, diagnosis, and therapeutic management in equids that developed clinical signs of EGA and were confirmed to be infected with A. phagocytophilum. The variety of clinical and pathological findings and the challenging therapeutic approaches registered strongly suggest that EGA should be included in the differential diagnosis when fever occurs. In addition, the findings emphasize the importance of monitoring EGA to initiate proper medication in order to avoid complications. Abstract Equine granulocytic anaplasmosis (EGA) is a tick-borne disease affecting horses worldwide, caused by Anaplasma phagocytophilum. The disease ranges from non-specific clinical signs to fatal outcomes. This paper aimed to analyze EGA cases reported in peer-reviewed journals, particularly on clinico-pathological findings, diagnosis, and therapeutic management. Overall, 189 clinical cases from 31 publications were included in the study. Extensive symptomatology for the EGA cases was reported, of which mostly was fever (90.30%), followed by limb edema (48.51%), anorexia (41.79%), depression (32.84%), icterus (22.39%), ataxia (17.91), tachycardia (16.42%), and lethargy (15.67%). Laboratory tests revealed thrombocytopenia (90.32%), anemia (75%), decreased hematocrit (70.59%), leukopenia (55.88%), lymphopenia (58.14%), and neutropenia (41.67%) as the most common hematological abnormalities. For a subset of tested animals, hyperbilirubinemia (20/29), hyperfibrinogenemia (13/15), and hyponatremia (10/10) were also reported. The diagnosis was established by microscopic identification of morulae (in 153 cases), and/or PCR (120 cases), isolation (1 case), or serology (56 cases). For treatment, oxytetracycline was used in the majority (52.24%) of EGA cases, but recovery without antibiotherapy (10.34%) was also noted. In conclusion, the variety of clinical and pathological findings and the challenging therapeutic approaches reported suggest that EGA should be included in the differential diagnosis when fever occurs.


Introduction
Equine granulocytic anaplasmosis (EGA) is a tick-borne infection caused by Anaplasma phagocytophilum, an obligate intracellular Gram-negative bacterium (order Rickettsiales; family Anaplasmataceae) [1][2][3].The causative species was previously known as Ehrlichia phagocytophila, Ehrlichia equi, and human granulocytic ehrlichiosis agent, which have been classified into a single species based on the similarities observed between them [4].Supporting this theory, a human pathogenic strain of A. phagocytophilum has been identified in the blood of a horse [5].
The infections caused by A. phagocytophilum, namely human granulocytic anaplasmosis (HGA) and animal granulocytic anaplasmosis (equine, canine, and feline granulocytic anaplasmosis, or tick-borne fever in ruminants), are among the most widespread tickborne infections in animals in Europe.Moreover, A. phagocytophilum is also regarded as an emerging human pathogen with growing importance in the Northern Hemisphere [4,6].
EGA was reported for the first time in 1969 in California, USA.In North America, it is transmitted by Ixodes pacificus (in California and other western states) and Ixodes scapularis (in the eastern and northern Midwest of the United States) ticks [7][8][9].
The reservoir hosts for this bacterium are considered to be small mammals, wild boar, cervids, and sheep, while humans, dogs, and horses are considered accidental hosts [9, 16,17].Recently, it has been noted that Marmota himalayana (Himalayan marmot) could serve as a reservoir for A. phagocytophilum [10].
A. phagocytophilum has been reported worldwide in a variety of hosts.In Romania, the pathogen has been detected in ticks, red foxes, small mammals, and wildlife hosts (mammal hosts, birds, and hedgehogs) [18][19][20][21][22][23].Antibodies against Anaplasma spp.have been recently reported in cattle [24].Also, a recent serological survey conducted by the authors reported, for the first time in Romania, anti-A.phagocytophilum antibodies in horses [25].
A large number of horses has been tested using serological and molecular surveys concerning EGA, but the number of clinical EGA reports is smaller and mostly as single or case series reports.Therefore, the aim of the present systematic review and meta-analysis was to analyze clinical cases reported in peer-reviewed journals regarding clinico-pathology, diagnosis, and therapeutic management in horses that developed clinical signs of EGA and were confirmed infected with A. phagocytophilum.

Search Methodology, Inclusion and Exclusion Criteria
Publications selected for this systematic review and meta-analysis were found in the PubMed and Google Scholar databases.PRISMA guidelines were used to select eligible articles.For this, two authors (A.M.B. and M.I.) searched independently, using keywords and expressions like Anaplasma phagocytophilum, anaplasmosis, equine anaplasmosis, equine granulocytic anaplasmosis, Anaplasma phagocytophilum in horses, anaplasmosis in horses, Ehrlichia equi, ehrlichiosis, equine ehrlichiosis, and ehrlichiosis in horses.'My ncbi filters' were used to select the results by year and the type of articles needed (i.e., case reports).Once a title has been selected, the abstract was checked to select articles eligible for inclusion.Available full texts of the selected publications were reviewed, including their references, in order to include additional articles that might have been selected in the initial literature search.
The following inclusion criteria were established: Publications that did not fulfill the above criteria were excluded.

Information Extracted
In a Microsoft ® Excel ® 2019 MSO (16.0.15831.20098)spreadsheet, the following information was extracted from the selected articles: country where the cases were reported, continent, number of clinical cases, description, anamnesis, symptomatology, pathological findings, diagnostic methods, therapy, outcome, and reference.

Results
The bibliographic research using the above-mentioned criteria resulted in a total of 460 accessed full-text articles; of them, 257 described the pathogen's occurrence in animals (other than equids) and/or in vectors (ticks, fleas, and lice), 81 were studies regarding A. phagocytophilum, only 21 reported A. phagocytophilum in equids, other species and/or ticks, 4 emphasized the importance of climate change and ecology in its appearance, 4 were publications in which the bacterium had not been detected in equids or other species, and 93 described its occurrence in equids only.Of the last 93 publications, 31 reported clinical signs of EGA in 189 equids.The PRISMA flow diagram is presented in Figure 1.Therefore, in the present study, a total number of 189 clinical EGA cases described in 31 publications were used to synthesize and analyze findings on EGA with regard to (i) clinical signs and laboratory abnormalities; (ii) diagnosis; (iii) therapy and the outcome.

•
Clinical cases of EGA with a description of symptomatology; • Clinical cases of EGA with confirmatory diagnosis by microscopic and/or molecular methods (polymerase chain reaction).Publications that did not fulfill the above criteria were excluded.

Information Extracted
In a Microsoft ® Excel ® 2019 MSO (16.0.15831.20098)spreadsheet, the following information was extracted from the selected articles: country where the cases were reported, continent, number of clinical cases, description, anamnesis, symptomatology, pathological findings, diagnostic methods, therapy, outcome, and reference.

Results
The bibliographic research using the above-mentioned criteria resulted in a total of 460 accessed full-text articles; of them, 257 described the pathogen's occurrence in animals (other than equids) and/or in vectors (ticks, fleas, and lice), 81 were studies regarding A. phagocytophilum, only 21 reported A. phagocytophilum in equids, other species and/or ticks, 4 emphasized the importance of climate change and ecology in its appearance, 4 were publications in which the bacterium had not been detected in equids or other species, and 93 described its occurrence in equids only.Of the last 93 publications, 31 reported clinical signs of EGA in 189 equids.The PRISMA flow diagram is presented in Figure 1.Therefore, in the present study, a total number of 189 clinical EGA cases described in 31 publications were used to synthesize and analyze findings on EGA with regard to (i) clinical signs and laboratory abnormalities; (ii) diagnosis; (iii) therapy and the outcome.Overall, EGA clinical cases were reported from 1994 to 2022 (Figure 2a) on four continents: Europe (127/189; 67.20%), Asia (29/189; 15.34%), America (24/189; 12.70%), and Africa (9/189; 4.76%) (Table 1, Figure 2b).

Diagnostic Assays
In the eligible analyzed studies, EGA was diagnosed using one or more of the four following diagnostic methods: microscopy, PCR, isolation, and/or serology.The summarized methods used for diagnosis of the 189 clinical cases of EGA are presented in Table 3.
In addition, in one study, 18 of 24 analyzed samples were seropositive by IFA after 8 months [28].Also, one horse showed ataxia and conscious proprioceptive deficits five years after the initial EGA diagnosis; microscopic examination and PCR were negative, while the serological test was positive at presentation and after 36 days [51].
The pathogen was isolated in only one case (1/189; 0.53%), using specific cellular lines and infected blood cells (confirmed by microscopic examination of buffy coat smear) preserved at −80 • C [31].
Oxytetracycline was administered intravenously (i.v.) at doses from 3 mg/kg to 10 mg/kg for 4 to 14 days; in some cases (4/67; 5.97%), it was followed by or switched to oral doxycycline at a dose of 10 mg/kg for 5 to 14 days [46,47,50].In one horse, oxytetracycline was administered after initiation of oral minocycline therapy, which was continued at a dose of 4.4 mg/kg for 14 days [51].Three horses received oxytetracycline intramuscularly (i.m.) at doses of 5 mg/kg, 8.4 mg/kg, and 11.6 mg/kg (continued with 5.8 mg/kg) for 5, 10, and 7 days, respectively [30,51].One horse was treated only with oral doxycycline on the farm, respecting the owners' decision.Despite the fact that the treatment for this horse was discontinued after 5 days because of diarrhea, clinical signs resolved without recurrence [49].
One horse was treated with the combination trimethoprim sulphonamide and showed full recovery in 4 days [28].Also, in four horses, treatment included drugs that were not specific for anaplasmosis until the diagnosis of EGA could be established.In three of these cases, horses were treated with penicillin before admission to the clinic and recovered spontaneously without any other treatment after 3-14 days of clinical signs [37], while one horse received a dose of imidocarb dipropionate before EGA diagnosis and recovered within five days without any other treatment [36].
Data about recovery after therapy were available in 44 out of 67 cases (44/67; 65.67%); complete recovery was reported for 24 (54.55%)treated equids, at 12 h up to 8 months p.t.For the remaining 20 cases, the period was not mentioned.It can be noted that recovery for some untreated horses and for several horses treated with antibiotics other than tetracyclines was reported within the same period as in those treated with tetracycline [28].
Vet. Sci.2024, 11, 269 Table 4. Meta-analysis results of therapy and outcome of the equine granulocytic anaplasmosis clinical cases reported in peer-reviewed published articles .

Discussion
Analyzing the clinical cases resulting from the selection shows that the largest number of clinical cases of EGA was registered in Europe (127/189), and the majority of studies (20/31) were reported after 2005.A high number of EGA cases, more than 10 cases each, were from Sweden, Spain, Germany, Poland, and the Czech Republic.EGA was diagnosed in equids aged between 4 months and 30 years old, mostly males, of different pure-or crossed-breeds.
EGA develops as an acute self-limiting or subclinical disease [7,54].Clinical signs are usually non-specific, except for limb edema and ataxia, so the fever of unknown origin should be an impulse to meticulously investigate the diagnosis for EGA [36].In the selected articles, the most commonly observed clinical signs were fever (90.30%), limb edema (48.51%), and anorexia or inappetence (41.79%).
Experimental studies in horses reported a percentage from 0.5 to 16.0% of neutrophils with inclusion bodies [55,56].These were first observed in neutrophils by day 12-14 postexposure to experimentally infected ticks and from 27 days post-exposure to naturally infected ticks [55,57].In contrast, the selected articles reported a percentage of 4% to nearly 40% of neutrophils [5,27,35,37,42,48,49], 3-4 days after tick bite [38].After 48-72 h from the initiation of antibiotherapy in naturally infected horses, morulae are no longer present in peripheral blood.Hence, inclusions can be observed for a short period of time, which decreases the sensitivity of this diagnosis method [47].However, morulae were still observed after 2 weeks after their first detection in a horse that did not receive any treatment, but it was considered clinically recovered [28].
In experimental studies in horses, the PCR detection period after exposure to naturally infected ticks was 3-17 days [55,57,58].After intravenous inoculation of A. phagocytophilum Swedish strain, PCR was positive for a longer period, until 18-21 days [56].In a naturally infected horse, PCR remained positive at 2 days but was negative after 28 days [51].In the case of chronic anaplasmosis, the pathogen was detected 4 months after tick contact, so the detection period was almost equal to the one described in another experimental study, in which the pathogen persisted for at least 129 days, without any clinical or pathological abnormalities observed [41,54].
DNA of A. phagocytophilum can be isolated from whole blood, leukocytes, bone marrow, or fragments of the spleen [7].In the selected publications, besides blood samples, cerebrospinal fluid, muscle tissue, pericardial fluid, and pleural effusion showed positive PCR results for A. phagocytophilum [41,46,50].
Anti-A.phagocytophilum antibodies are detected by serological methods, such as IFAT and ELISA, to determine a present or past infection.These methods are very useful in epidemiological studies, but the information obtained is not enough to establish a diagnosis in acute clinical cases [59,60].One selected article reported seropositive horses 8 months after the acute stage of the disease [28].Another article reported a seroconversion in a horse 2 years after it was first diagnosed with EGA, without any evidence of infection in those years [51].
A. phagocytophilum can be cultured from infected blood and kept for up to 18 days under refrigerated conditions [2].In vitro culturing offers a sensitivity equivalent to that of PCR and microscopy; however, positive results are reported in approximately one week and can remain negative for more than 2 weeks [17].Only one selected article reported the isolation of the pathogen using specific cellular lines, performed by use of infected cells preserved at −80 • C [31].
A differential diagnosis should comprise purpura hemorrhagica, liver diseases, equine infectious anemia, equine viral arteritis, encephalitis, and borreliosis [7,61].It has also been suggested to consider EGA as a differential diagnosis for acute severe rhabdomyolysis and myalgia [46].
Co-infection with Borrelia burgdorferi sensu lato was also reported in six EGA cases in two studies [5,28], but all were considered most probably old infections.Of them, five horses were seropositive for Borrelia afzelii, but most antibody titers declined during repeated checking over a period of 12-15 months [28]), while for one Lyme borreliosis seropositive horse, discrimination between a previous and recent exposure was not pursued [5]).However, given the potential of co-infection with B. burgdorferi s.l., and since clinical signs of borreliosis may resemble those of EGA, this is a diagnosis to be considered.A recent serological and molecular study in Brazil (Rio de Janeiro state) reported a moderate frequency of seropositive horses for A. phagocytophylum (17.3%) but also simultaneous exposure to A. phagocytophilum and T. equi (18.2%).However, co-infection of A. phagocytophilum with T. equi was molecularly confirmed in only one horse, which was also the only one displaying clinical signs such as pale mucosa, cachexia, and fever [62].In Brazil, EGA is considered an emerging disease.High seroprevalence rates have been reported in horses from the central-west region [63] and southeastern Brazil [64], of 65% (by ELISA) and 76% (by IFA test), respectively.However, characteristic inclusions of A. phagocytophilum were observed in 12.8% of buffy coat smears [64].
Also, serological evidence of exposure to A. phagocytophilum has been reported also in horses in Portugal, with a prevalence from 3% to 13% [65,66].However, the human A. phagocytophilum HZ strain was found in one seropositive horse from mainland Portugal, suggesting the potential for HGA in Portugal [65].
Horses can fully recover from EGA without any treatment, but antibiotics shorten the course of the disease, and clinical signs abate after intravenous administration of oxytetracycline for 3-7 days [36,47,54].In an experimental study, all horses diagnosed with acute EGA recovered without antibiotherapy by 21 days after inoculation of A. phagocytophilum strain, with negative results in PCR tests by day 22 [54].Nine horses (10.34%) from the selected articles recovered without antibiotherapy, and some untreated horses and several horses treated with antibiotics other than tetracyclines recovered within the same period as those treated with tetracycline [5,28,34,36].

Conclusions
Clinical cases of EGA selected for meta-analysis allow for a better understanding of the clinico-pathological features.Therefore, the EGA cases were mainly characterized clinically by fever (90.30%), limb edema (48.51%), and anorexia (41.79%), while among the frequent pathological findings were thrombocytopenia (90.32%), lymphopenia (58.14%), and hyperbilirubinemia (68.97%).Diagnosis was mainly based on the detection of the pathogen by microscopy (88.44%) and/or molecular biological methods (PCR-based) (98.36%).Besides blood samples, A. phagocytophilum could also be detected in a cytocentrifuge preparation of the abdominal fluid (microscopic examen-M.E.), in pericardial fluid (M.E., PCR), pleural effusion (M.E., PCR), mucous sample of the endotracheal tube (M.E.), cerebrospinal fluid (PCR), and from muscle tissue (PCR).In terms of therapy and outcome, equids recovered after specific treatment with oxytetracycline, which was mostly used (52.24%), but clinical signs resolved also in one horse treated only with doxycycline.The small percentage of horses that recovered without antibiotherapy (11.25%) emphasizes the important role of monitoring EGA to initiate proper medication in order to avoid complications.

3. 1 .
Data about the Period and Originating Area of Reported EGA Clinical Cases

Figure 2 .
Figure 2. (a) Number of published articles per year reporting equine granulocytic anapla cases.(b) Map with the countries where the equine granulocytic anaplasmosis (EGA) reporte (n = 189) originated from.

Figure 2 .
Figure 2. (a) Number of published articles per year reporting equine granulocytic anaplasmosis cases.(b) Map with the countries where the equine granulocytic anaplasmosis (EGA) reported cases (n = 189) originated from.

Table 1 .
General data of the equine granulocytic anaplasmosis clinical cases (n = 189) reported in peer-reviewed published articles.

Table 2 .
Meta-analysis results of clinical and laboratory findings of the animals diagnosed with equine granulocytic anaplasmosis.
* Anemia included decreased red blood cell count, decreased hemoglobin, and/or decreased PCV/hematocrit); ** cases for which only decreased PCV/hematocrit was reported.

Table 3 .
Diagnostic methods used for diagnosis of 189 clinical cases of equine granulocytic anaplasmosis reported in peer-reviewed articles.