Hemogram Findings in Cats from an Area Endemic for Leishmania infantum and Feline Immunodeficiency Virus Infections

Simple Summary Some cats positive for Leishmania infantum (Li) show clinical signs and clinicopathological changes, including hemogram abnormalities. However, co-infections or co-morbidities are often detected in cats with leishmaniosis, and they can have a role in the clinical abnormalities observed. In feline Leishmania infantum infections and in clinical cases of feline leishmaniosis, a significant association with feline immunodeficiency virus (FIV) has been detected, but the role of this co-infection is controversial. The aims of this study were to retrospectively evaluate hemogram changes in cats from areas endemic for Leishmania infantum and feline immunodeficiency virus infections (Sicily and Calabria regions, Southern Italy) and to analyze the role of both pathogens in the abnormalities detected. A retrospective cross-sectional study was carried out on 496 cats. Hematological changes in cats positive for Leishmania infantum were found, particularly abnormalities compatible with chronic inflammation including moderate anemia, monocytosis, and morphologically activated monocytes. Moreover, some abnormalities, such as thrombocytosis, seem to arise when cats are co-infected with FIV. Feline leishmaniosis should be considered when some hemogram abnormalities compatible with chronic inflammation are detected. Abstract In feline Leishmania infantum (Li) infection and in clinical cases of feline leishmaniosis, co-infection with feline immunodeficiency virus (FIV) has been reported. However, the role of the retroviral co-infection in the impairment of feline clinical health is still controversial. The aim of this study was to evaluate hemogram changes in cats from regions endemic for both Li and FIV infection. Four hundred and ninety-six cats tested for Li (EDTA blood polymerase chain reaction and immunofluorescence antibody test) and for FIV infection (enzyme-linked immune assay) were retrospectively evaluated. Hemogram results including blood smear morphological evaluation were statistically compared considering four infection patterns: Li+FIV+, Li+FIV−, Li−FIV+, and Li−FIV−. Significantly lower values of erythrocytes (Li+FIV−: p = 0.0248; Li−FIV+: p = 0.0392) and hemoglobin (Li+FIV: p = 0.0086; Li−FIV+: p = 0.0249) were found in both infections when compared to Li−FIV− cats, and severity of anemia was more frequently moderate in Li-positive cats (p = 0.0206) and severe in FIV infection (p = 0.024). Li infection was associated with monocytosis (p = 0.0013) and morphologically activated monocytes (p = 0.0209). Moreover, FIV infection was associated with the presence of inflammatory leukogram (p = 0.023), and an association between thrombocytosis and the co-infection was found (p = 0.0347). Li infection in cats induces hematological changes compatible with chronic inflammation, some of which are due to co-infection with FIV.

Few studies compared hematological abnormalities between Li-positive and -negative cats, and they did not find significant differences [5,8]. Spada et al. (2020) evaluated co-infections, but they found only one case of FIV co-infection in that study [5]. In a recent study reporting clinical signs and hematological abnormalities associated with Li and FIV co-infection, no CBC abnormalities were significantly associated with this co-infection [14].
The aims of this study were to retrospectively evaluate CBC values in cats from areas endemic for Li and FIV infections, such as Sicily and Calabria regions (Southern Italy) [2,[25][26][27][28][29][30][31], and to analyze the role of both pathogens in the alterations detected.

Study Sites, Cat Enrollment, and Clinical Classification and Sampling Procedures
A retrospective cross-sectional study was carried out, analyzing results from three studies already published [28][29][30]. Cats were sampled between 2012 and 2019 at four veterinary clinics in Sicily (Ospedale Veterinario Universitario Didattico, Dipartimento di Scienze Veterinarie, Università di Messina, Messina; Ambulatorio Veterinario S. Lucia, Lipari, Messina) and Calabria (Clinica Veterinaria Camagna, Reggio Calabria; Ambulatorio Dr. Cardone, Gioia Tauro, Reggio Calabria) regions. Signalment, history, and physical examination findings were registered in a clinical form. According to physical examination, cats were classified as apparently healthy, affected by signs compatible with feline leishmaniosis [1], and affected by other signs. Clinical manifestations considered compatible with feline leishmaniosis were lymph node enlargement, skin and/or mucocutaneous lesions (mainly ulcerative or nodular), ocular lesions (mainly uveitis), oral lesions, weight loss, anorexia, lethargy, dehydration, pale mucous membranes, hepatomegaly, icterus, cachexia, fever, diarrhea, chronic nasal discharge, splenomegaly, polyuria/polydipsia, itching, dyspnea, wheezing, abortion, and hypothermia [1]. In most cases, cats included in this category showed multiple signs among those listed above. Signs and pathologies categorized as "other signs" were overweight, otitis, cystitis, and cough.
Cats were phlebotomized under physical restraint, and from each cat, about five milliliters of venous blood was collected. One milliliter was placed into a K 3 EDTA tube and used for CBC evaluation within two hours after collection or stored at +4 • C and used after a maximum of 24 h. The leftover K 3 EDTA blood was then stored at −20 • C until analyzed for detection of Li DNA or feline leukemia virus infection (FeLV) RNA. The remaining blood was used to make blood smears and to obtain serum after clotting in a dry tube and centrifugation. Each aliquot of serum was stored at −20 • C until analyzed for Li and FIV antibody detection, and FeLV p27 antigen.
Informed consent was obtained from owners, and approval of the ethics committee from the author's academic institution was obtained.

Complete Blood Count
Complete blood count was performed using a laser hematology analyzer ProCyte Dx (IDEXX laboratories, Westbrook, ME, USA). Reference intervals of CBC parameters are shown in Table 1. Blood smears were stained by May-Grünwald-Giemsa stain (Merck KgaA, Darmstadt, Germany) and evaluated microscopically at oil immersion ×1000 magnification [32]. Low platelet count as well as any "smart flag" message reported by the analyzer about leukocyte or platelet count (i.e., inability of the analyzer to make the count or inaccuracy of the analyzer count) were correspondingly confirmed or settled after microscopic examination of blood smears. Blood smears were also examined for cell morphological abnormalities and to exclude as thrombocytopenic samples where platelet clumps were observed.

Diagnosis of L. infantum Infection
Anti-Li immunoglobulin G (IgG) antibodies (Ab) were tested by immunofluorescence antibody test (IFAT) using Li (strain MHOM/IT/80/IPT1) antigen slides produced by CReNaL (Centro di Referenza Nazionale per la Leishmaniosi, Palermo, Italy) following Office International des Epizooties (OIE) Terrestrial Manual protocol [36].   [39]. Samples were amplified in a single 96-well plate, and in each one, a positive control containing genomic L. infantum DNA and a negative control without DNA were included. Each standard, sample, and negative control was analyzed in duplicate for each run. Cycle threshold (Ct) value was calculated for each sample by determining the point of the fluorescence value exceeding the threshold limit. The parasitic DNA load was defined in each examined sample by comparison of the data with a specific standard curve on the basis of the number of Leishmania per milliliter of extracted volume. Standard curves were prepared for both the Leishmania gene target and the internal PC (IPC Applied Biosystems). A stock solution of L. infantum DNA was obtained by extraction from 109 promastigotes/mL. Tenfold serial dilutions of the DNA stock solution were performed to obtain the six points of the curve spanning from 106 to 101 DNA equivalent cells. The standard curve, calculated by independent experiments, was linear over at least 6 log ranges of DNA concentration points with an average correlation coefficient of 0.988. The difference for each point of the curve was one log factor [39].

Diagnosis of Feline Retroviral Infections
Cats were tested for FIV and FeLV. Different diagnostic tests were used in the time interval considered for this retrospective cross-sectional study (2012-2019). In detail, anti-FIV antibodies were investigated with commercial kits (SNAP Combo Plus FeLV antigen and FIV antibody test, Idexx Laboratories, Westbrook, ME, USA (257 cats); Pet Check FIV anti-body test kit, IDEXX Laboratories, Westbrook, ME, USA (239 cats)). The FeLV positivity was assessed by a rapid ELISA test detecting p27 antigenemia (SNAP Combo Plus FeLV antigen and FIV antibody test, Idexx Laboratories, Westbrook, ME, USA (257 cats) or by blood real-time PCR (U3 region LTR-genesig ® Advanced kit, Rownhams, UK (239 cats)). All tests were performed according to the manufacturer's protocol.

Statistical Analysis
Statistical analysis was performed using GraphPad Prism version 7.0 for Windows (GraphPad Software, San Diego, CA, USA). Distribution of continuous variables was evaluated by D'Agostino-Pearson omnibus normality test, and descriptive statistics was performed for all the evaluated variables.
Cats Fisher's exact test was used to evaluate associations between the four infection patterns considered and signalment, history, clinical examination findings, CBC with blood smear morphological examination, degree of anemia, and its classification (regenerative/non regenerative, mycro/normo/macrocytic, hypo/normochromic). Parameters with 0 frequency were excluded from Fisher's exact test.
Kruskal-Wallis test and Dunn's multiple comparisons test were used to compare CBC values among the four infection patterns observed. This was performed for all cats enrolled as well as for those with CBC alterations.
The Spearman's rank correlation test was used to assess the correlation between antibody titer and hemogram values of cats with CBC abnormalities.

Cat Population Demographic and Clinical Data
Four hundred and ninety-six cats were enrolled: 319 from Calabria and 177 from Sicily regions. Signalment, history, and clinical data collected are shown in Table 2. Cats were aged between 5 and 228 months (median = 36 months, 25% percentile = 12 months, 75% percentile = 74 months). Table 2. Data from signalment, history, and clinical examination (n (%)) of total enrolled cats (total) and of the cats classified according to their infection pattern (Li+FIV+; Li+FIV−; Li−FIV+; Li−FIV−). Prevalence of males was significantly higher in Li−FIV+ than both Li+FIV− and Li−FIV− cats (Tables 2 and 3). Moreover, prevalence of males was higher in Li+FIV+ than Li−FIV− animals (Tables 2 and 3). Prevalence of adult cats was higher, when compared to young cats, in Li+FIV+ and Li−FIV+ than Li−FIV− (Tables 2 and 3). Finally, prevalence of senior cats was higher compared to young cats in Li+FIV+, Li+FIV−, and Li−FIV+ than Li−FIV− (Tables 2 and 3). Table 3. p-value, odds ratio (OR), and 95% confidence interval (CI) for the significant differences concerning signalment, history, and clinical examination.

Complete Blood Count
Lower values of erythrocytes were detected in Li+FIV− and in Li−FIV+ cats compared to Li−FIV− cats (Tables 4 and 5 Table 5).   Moderate anemia was more frequently observed when compared with mild anemia in Li+FIV− cats than Li−FIV− cats (Tables 6 and 7). Moreover, severe anemia was more frequently found when compared to mild anemia in Li−FIV+ cats than Li−FIV− cats (Tables 6 and 7). Finally, hypochromic anemia was more frequently found in Li−FIV+ (30.77%) than in Li−FIV− cats (7.53%), and macrocytic anemia was more frequently found in Li−FIV+ cats (36.36% and 50%, respectively) when compared to normocytic and microcytic cases than Li−FIV− cats (4.82% and 12.9%, respectively) ( Table 7).  Monocytosis and morphologically activated monocytes were more frequently observed in Li+FIV− cats compared to Li−FIV− cats (Tables 8 and 9). Inflammatory leukogram was more frequently detected in Li−FIV+ cats than Li−FIV− cats (Tables 8 and 9).   Frequency of thrombocytosis was significantly higher in Li+FIV+ cats than Li−FIV− cats (Tables 8 and 9).
No correlation was found between antibody titer against Li and CBC values. A statistical evaluation of the association between clinical signs and hematological alterations was not made because almost all the examined cats showed at least one clinical sign compatible with leishmaniosis (Table 10). Table 10. Description of hematological abnormalities observed in total enrolled cats with number of cats (n) presenting each of them. The prevalence of each abnormality is reported (n (%)) in cats classified according to their infection pattern (Li+FIV+; Li+FIV−; Li−FIV+; Li−FIV−) and clinical status (apparently healthy, signs compatible with FeL, different/other signs).

Discussion
In this retrospective cross-sectional study, a prevalence of 13.9% for Li antibody and 3.2% for Li DNA detection in blood samples was found, and 10.1% of cats had antibodies against FIV. Conversely, no cats were FeLV positive. These results confirm that Southern Italy, particularly Sicily and Calabria, is an endemic area for Li and FIV, while FeLV is sporadic.
According to the results of the present study, there was no significant association between anemia and infections with Li or FIV; however, significantly lower values of erythrocytes (Li+FIV− and Li−FIV+) and hemoglobin (Li−FIV+) were found in both infections when considering all results, and significantly lower values of hemoglobin were detected in Li infection (Li+FIV−) when only anemic cats were considered. Co-infection did not cause significantly lower values compared to the other patterns. However, the number of Li+FIV+ cats was small (16), and a higher number of co-infected cats should be analyzed. Severity of anemia was more frequently moderate in cats infected with Li and severe in the case of FIV infection. A considerable percentage of Li−FIV− cats had anemia (28.8%), and it was mostly a mild anemia (88.2%). However, only 21.3% of Li−FIV− cats were apparently healthy at physical examination and, unfortunately, we could not analyze data about other pathogens and pathologies potentially causing anemia in the studied cats. Additionally, FIV-infected cats had more frequently lower MCHC values and presented hypochromic and macrocytic anemia.
Mild-to-severe normocytic normochromic non-regenerative anemia is the most frequent hematological abnormality reported in clinical cases of feline leishmaniosis [1,6]. Anemia is also one of the most common laboratory findings in dogs with leishmaniosis with multiple and not fully known pathomechanisms [40][41][42][43][44][45][46]. In fact, blood loss, hemolysis, renal failure, chronic inflammation, and bone marrow alterations may be responsible for anemia in dogs with leishmaniosis. Erythroid hypoplasia [19,[47][48][49]; erythroid dysplasia [50]; myelophthisis with infiltration of lymphocytes, plasma cells, and macrophages [19]; and erythrophagocytosis of erythroblasts [50] are the bone marrow alterations found in canine leishmaniosis. Interleukins produced by Th1 cells, such as IFN-γ, would play a role in the pathogenesis of anemia through inhibition of the earliest stages of erythroid differentiation and proliferation [51], alteration of iron homeostasis, induction of iron retention in macrophages, and consequent limitation of iron availability during erythropoiesis [49,51,52]. Moreover, the increased number of activated macrophages typical of leishmaniosis can promote erythrophagocytosis [53]. Changes in erythrocyte membrane fluidity, which may lead to their sequestration in the spleen, was also related to anemia in canine leishmaniosis [54]. Currently, there is a lack of data on the possible pathogenesis of anemia in feline leishmaniosis.
Both regenerative and non-regenerative anemia are commonly observed in FIVinfected cats because of a direct pathogenic effect of the virus. Multifactorial mechanisms have been proposed: ineffective hematopoiesis due to FIV infection of the bone marrow stromal cells that alters normal hematopoietic functions [55] and immunomediated destruction of erythrocytes due to an overactive immune response [56,57]. However, concomitant diseases often play an important role. This is the case of hemolytic anemia caused by hemotropic mycoplasmas [58,59]. Moreover, suppression of the immune system induced by FIV is associated with chronic opportunistic infections responsible for various hematological changes included normocytic normochromic non-regenerative anemia [23].
Most of the morphological alterations of erythrocytes were very rare and, when it was possible to use Fisher's exact test, no significant differences were found according to the infection pattern.
We found that Li infection was associated with monocytosis and with the occurrence of morphologically activated monocytes. Monocytosis has rarely been reported in Liinfected cats [14]. Conversely, it is considered a finding consistent with leishmaniosis in dogs [46], and higher monocyte concentrations have been detected in dogs with positive splenic culture when compared with dogs with negative spleen culture [49], in dogs with leishmaniosis when compared to healthy dogs [60], and in oligosymptomatic Li-infected dogs when compared to non-infected dogs [47].
Monocytes play a crucial role in Leishmania infection, contributing to both innate immune defense and adaptative immune responses [47,61,62]. Moreover, monocytosis occurs in many conditions in cats, including acute and chronic inflammation and tissue destruction due to trauma-related injuries, suppuration, necrosis, pyogranulomatous inflammation, hemolysis, malignancy, and immune-mediated disorders [34].
In the present study, FIV infection was associated with the presence of inflammatory leukogram. FIV-induced immunosuppression could have facilitated a secondary or opportunistic infection, to which an appropriate inflammatory response was made [22]. Many FIV-infected cats, with concomitant or secondary infections, still have an appropriate hematological response and show leukocytosis, neutrophilia, and left shift in association with purulent or inflammatory processes [59]. Neutrophilia and monocytosis were common hematological abnormalities seen in cats with naturally occurring FIV infection in previous studies [14,21,22,63]. We could not evaluate the clinical stage of FIV-positive cats under study but, on the basis of the association of the retroviral infection with the occurrence of inflammatory leukograms, they were not in the end stage of disease.
We found an association between thrombocytosis and the co-infection by Li and FIV, although this infection pattern involved only 11 cats, and this abnormality affected a small number of cats (2/11). Thrombocytosis has never been described before in cats with Li or FIV infection and in the case of co-infection. Thrombocytosis has been found in dogs with Li infection, and it was assumed that it can be related to the chronic inflammatory process [44]. The so-called reactive thrombocytosis is the most common form of thrombocytosis in dogs and cats, and it is caused by a high level of inflammatory cytokines that stimulate thrombopoiesis [64]. Importantly, the two cases of thrombocytosis occurred in cats with an inflammatory leukogram. This supports the hypothesis of a reactive mechanism for thrombocytosis of cats coinfected with Li and FIV.
We evaluated for the first time the reticulocyte hemoglobin content in Liand FIVpositive cats. The rationale for investigating this index was that chronic inflammation is associated with functional iron deficiency and can cause a non-regenerative normocytic normochromic anemia. Reticulocyte hemoglobin content is one of the reticulocyte indices considered as markers of iron deficiency and has been recently included in the hemogram variables investigated by veterinary laser hematology analyzers. As it was introduced in the analyzer used in this study in 2019, we were able to include this variable in a limited number of cats (113). Most of them (81%) were negative to both the investigated infections, and a low value was found in only one negative cat. As far as we know, only two studies investigated reticulocyte hemoglobin content in cats, and Keiner et al. (2020) found that sensitivity for iron-limited erythropoiesis was low [65,66]. Further clinical studies about cat reticulocyte indices are indeed needed.
As previously reported, a significant association between Li antibody positivity and old age was detected [1,6,31]. This finding can be explained by a long-lasting course of antibody positivity after cat seroconversion and a good life expectancy for antibodypositive cats. When these two latter conditions are met, in Li-endemic areas where no preventative measures are implemented, the number of seropositive cats can increase year by year, and prevalence will be higher according to the number of transmission seasons the cat passed.
Moreover, male sex [22,24,67] and adulthood or old age were found to be risk factors for FIV infection [21,22,67], as largely reported in the literature.
This study has some limitations, and they are related to the retrospective approach. In fact, this has prevented the possibility to better assess relationships between hematological changes and the clinical status of cats. In particular, the stage of the disease was not considered for FIV-positive cats. This could have been a bias, as most cats do not exhibit immunosuppressive or myelosuppressive effects for a long time [56], and hematological sequelae of co-infection possibly depend on the cat FIV stage. Moreover, apart from FeLV infection, other pathogens were not considered, as well as co-morbidities affecting the hemogram. Additionally, methods used for testing retroviral infections were inhomogeneous. However, all the different tests we used have high sensitivity and specificity and are widely used in relevant studies regarding feline pathogens [68][69][70]. Finally, despite the high prevalence found for Li and FIV, the infection pattern that included co-infected cats (Li+FIV+) consisted of a small number of subjects. A more homogeneous number of samples is required in each category to obtain a robust analysis of results.

Conclusions
In conclusion, Li infection in cats is associated with hematological changes; is compatible with a chronic inflammation; and includes moderate anemia, monocytosis, and morphologically activated monocytes. Some of them, such as thrombocytosis, arise in the case of co-infection with FIV and should be further investigated. The hemogram obtained with a routine test such as CBC can therefore provide clinicians with data supporting a suspicion of feline leishmaniosis when alterations suggestive of chronic inflammation are found.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data set analyzed for the current study is available from the corresponding author upon reasonable request.