Abstract
Objectives
The aims of this study were to investigate the prevalence of Leishmania species infection in cats in Northern Italy and to evaluate the associations between infection and signalment and clinicopathological data.
Methods
The study was carried out in a veterinary university hospital from June to November 2017. Blood, urine, conjunctival swabs and hair were collected from all randomly selected cats. Leishmania species infection was evaluated using the indirect fluorescent antibody test (IFAT), setting a cut-off value of 1:80, and using real-time PCR on blood, conjunctival and hair samples. A complete blood count, serum chemistry profile, serum electrophoresis and urinalysis were also carried out. The cats were grouped on the basis of the results of the diagnostic criteria adopted in positive, negative and unconfirmed Leishmania cases. Non-parametric variables and continuous data were compared among the study groups using the χ2 test and the Mann–Whitney U-test, respectively.
Results
One hundred and fifty-two cats were included. Nineteen of the 152 (12.5%) cats were positive (18/152 [11.8%] showed an IFAT titre of ⩾1:80 and 1/152 [0.7%] was real-time PCR-positive from a hair sample); 106/152 (69.7%) cats were negative; and 27/152 (17.8%) cats were unconfirmed for Leishmania species. Total proteins, beta2-globulin and gamma-globulin were significantly increased in the positive Leishmania group compared with the negative group.
Conclusions and relevance
The results of the present study demonstrated the spread of Leishmania infantum infection in cats in Northern Italy. Hyperproteinaemia and hypergammaglobulinaemia appeared to be significant clinicopathological abnormalities in this population of cats with L infantum infection.
Keywords
Introduction
To date, in Italy, canine leishmaniosis (CanL) is considered to be endemic throughout most of the national territory, with a variable prevalence. Recent studies have detected Leishmania species infection in domestic cats, as well in areas where CanL is endemic, 1 with seroprevalence ranging from 0% to 68.5% and molecular prevalence ranging from 0% to 60.7%. 2 Cats have a neglected and controversial role in the cycle of Leishmania infantum, 3 but whether they serve as primary, secondary or accidental hosts is currently unknown. 4 Furthermore, only a few studies have been carried out regarding clinical and clinicopathological abnormalities in cats with feline leishmaniosis (FeL). 2
The majority of the diagnostic techniques for Leishmania species infection available for dogs are also used for cats. The most common serological test used is the indirect fluorescent antibody test (IFAT).2,4–7 Of the molecular methods, real-time PCR (qPCR) allows quantitative and fast detection of Leishmania species DNA.8,9 Detection using DNA can be applied to a variety of biological matrices, including bone marrow, skin biopsy and lymph node aspirate. These samples are difficult to obtain and, in some cases, sedation of the animal is required.2,10 Still little is known regarding the use of non-invasive biological matrices for the molecular detection of Leishmania species DNA in cats, such as peripheral blood, conjunctival swabs and hair.6,11–13
The aims of this study were: (1) to investigate the presence and distribution of Leishmania species infection in cats in Northern Italy using IFAT and qPCR assays applied to different biological samples (serum, blood, conjunctival swabs and hair); and (2) to evaluate possible associations between infection by L infantum and signalment and clinicopathological data.
Materials and methods
Study design, inclusion criteria, sampling and groups
This was a prospective study carried out in a veterinary university hospital in Northern Italy (Department of Veterinary Medical Sciences, University of Bologna, Bologna, Italy). Client-owned cats were included in the study during the sandfly activity season (from June to November 2017) after random selection and owner consent. Cats were included if they: (1) were living in the Emilia-Romagna region; and (2) had undergone a blood sample after clinical examination. Signalment, lifestyle (indoor or outdoor) and clinical data were recorded for each cat sampled.
Blood sampling was carried out by venepuncture, and samples were collected using a vacuum system (Vacutest Kima). The K3EDTA samples were used for a complete blood count (CBC) and molecular analyses. Serum samples underwent a chemistry profile and serum protein electrophoresis. The urine samples were collected by spontaneous voiding, cystocentesis or urethral catheterisation, when required, by the attending clinicians, and underwent urinalysis. Conjunctival swabs from both eyes and hair samples (plucked) were also collected from each cat. Clinicopathological evaluation was carried out within 1 h from the sampling and the samples were stored at −80°C after examination. Leishmania species infection was investigated using IFAT on serum and qPCR on blood, conjunctival swabs and hair samples.
Cats with a positive IFAT titre (⩾1:80) 14 and/or positive qPCR on blood and/or conjunctival swabs and/or hair samples were included in the positive Leishmania (PL) group; cats with a negative IFAT titre (<1:40) and negative qPCR results were included in the negative Leishmania (NL) group; cats with an IFAT titre equal to 1:40 and negative qPCR results were included in the unconfirmed Leishmania (UL) group.
Furthermore, the medical records of the PL group were analysed to evaluate signalment, history, and clinical and clinicopathological data.
IFAT
Anti-Leishmania IgG antibodies were investigated in serum samples using IFAT according to the CanL diagnosis described in the World Organisation for Animal Health manual of diagnostic tests and vaccines for terrestrial animals. 15 Slides coated with 30 µl of an in-house antigen consisting of promastigotes of L infantum (strain MHOM/TN/80/IPT1) 16 were probed with sera serially diluted in phosphate-buffered saline (PBS) starting with a concentration of 1:40 until reaching a concentration of 1:5120, incubated for 30 mins at 37°C and washed three times with PBS. Internal feline positive (1:160) and negative sera controls were included on each slide. The slides were probed with 30 µl of fluorescein isothiocyanate (FITC) conjugated anti-cat IgG antibody diluted in PBS at a concentration of 1:32 (Anti-Cat IgG-FITC antibody; Sigma-Aldrich) for 30 mins at 37°C and were washed three times with PBS and examined under a fluorescent microscope. The highest dilution showing fluorescent promastigotes was the final antibody titre.
Molecular diagnosis
Genomic DNA was extracted from blood, conjunctival swabs and hair samples using a commercially available kit (NucleoSpin Tissue Kit; Macherey-Nagel) according to the manufacturer’s protocol. Before extraction, for each cat sampled, the two conjunctival swab samples (right and left eyes) were resuspended together in 250 µl of PBS. The hair samples were cut into three segments (proximal, central and distal) and lysed at 56°C overnight with 250 µl of lysis buffer (Buffer T1) and 50 µl of proteinase K supplied in the extraction kit.
Leishmania species DNA detection was carried out using SYBR Green qPCR as reported by Ceccarelli et al. 17 A reaction was carried out using a commercially available kit (PowerUp SYBR Green Master Mix Kit; Life Technologies) and the StepOnePlus Real-Time PCR System (Thermo Fisher Scientific, Life Technologies). Serial 10-fold dilutions of a plasmid (pCR4 plasmid; Invitrogen, Life Technologies) containing one copy of the target sequence was used as an external standard for the construction of the assay standard curve. Melting experiments were carried out after the last extension step with a continuous increment from 60°C to 95°C; the specific melting temperature was 80.7°C. The limit of detection (LOD) of the assay was assessed by testing serial 10-fold dilutions of the recombinant plasmid and was found to be one copy/µl. The DNA samples and standards were repeated within each run in duplicate. A no template control underwent analysis simultaneously. The specimens were considered positive if the fluorescence curve in the amplification plot showed an exponential increase, a specific melting peak was observed and the mean of the DNA copy number obtained from the replicates was greater than the LOD.
Clinicopathological investigations
A CBC was carried out using an automated haematology analyser (ADVIA 2120; Siemens Healthcare Diagnostics). The haematology was completed with a microscopic blood smear examination using May-Grünwald Giemsa staining. A serum chemistry profile, including creatinine, urea, phosphate, total protein, albumin, albumin to globulin ratio (A:G), alanine transaminase, aspartate transaminase, alkaline phosphatase, gamma (γ)-glutamyltransferase, total bilirubin, cholesterol, total calcium, sodium, potassium, chloride, glucose and serum amyloid A (SAA), as reported previously, 18 was determined. Urinalysis included urine specific gravity, dipstick (Combur 10 TestUX; Roche) and microscopic sediment examination, and a urine protein:creatinine ratio (UPC). Urine samples with a visible red colour and/or >250 red blood cells in a high-power field were excluded from the UPC analysis. Serum and urine chemical analysis were carried out using an automated analyser (AU480; Beckman Coulter-Olympus).
Serum protein electrophoresis was carried out on agarose gel using a semi-automated system (Hydragel protein kit, Hydrasys; Sebia).
Statistical analysis
The data were evaluated using standard descriptive statistics and reported as mean ± SD or median and range, based on their distribution. The cats included in the UL group were excluded from all analyses. The categorical data were analysed using the χ2 test. The upper limits of the reference interval used in the university laboratory for cats were used as cut-off values for hyperproteinaemia and hypergammaglobulinaemia and were 8.8 g/dl and 1.30 g/dl, respectively. The continuous data (age and clinicopathological results) were compared among the study groups using the Mann–Whitney U-test with compensated post-hoc analysis. A P value <0.05 was considered significant. An inter-rater agreement statistic (Cohen’s kappa [κ] coefficient) was calculated to compare the results obtained by IFAT and qPCR, and between the qPCR carried out on three different biological samples. Statistical analysis was carried out using commercially available software (MedCalc Statistical Software version 16.8.4).
Results
Study groups and diagnosis of L infantum infection
During the study period, 152 cats met the inclusion criteria. Nineteen of the 152 (12.5%) cats were included in the PL group: 18/152 (11.8%) showed an IFAT titre of ⩾1:80 and 1/152 (0.7%) tested qPCR-positive with the hair sample. The NL group included 106/152 (69.7%) cats and the UL group included 27/152 (17.8%) cats. Table 1 shows the descriptive statistics results and data regarding the prevalence of infection in the study population.
Descriptive statistics and prevalence of infection among the cats included in the study groups
The χ2 test was carried out on the positive Leishmania (PL) and negative Leishmania (NL) groups. Data are n (%)
P value referred to male/female
P value referred to male (intact or castrated)/female (intact or spayed)
UL = unconfirmed Leishmania; NA = not available; DSH = domestic shorthair
Thirteen of 19 (68.4%) cats included in the PL group were male, of which 12/13 (92.3%) were castrated, and 6/19 (31.6%) were female, of which 5/6 (83.3%) were spayed. Seventeen of 19 (89.5%) were >24 months of age and 2/19 (10.5%) were between 6 and 24 months of age, with a median age of 8 years (range 9 months to 15 years). Fourteen of 19 (73.7%) were domestic shorthair cats and 5/19 (26.3%) were from other various breeds (Maine Coon, n = 1; Bengal, n = 1; Birman, n = 1; Persian, n = 1; Chartreux, n = 1); 12/19 cats (63.2%) were indoor/outdoor, while 7/19 (36.8%) were exclusively indoor. Sixteen of 19 (84.2%) cats were from the province of Bologna. The highest frequencies of positivity were detected in July (n = 5/35; 14.3%) and August (n = 7/22; 31.8%), while the frequency decreased in the other months of sampling.
Seroreactivity (IFAT titre ⩾1:40) to L infantum was found in 45/152 (29.6%) cats tested. Of the 18 seropositive cats (IFAT titre ⩾1:80), 11 (61.1%) had an antibody titre equal to 1:80, five (27.8%) to 1:160 and two (11.1%) to 1:320 (Tables 2 and 3).
Results obtained using the indirect immunofluorescence antibody test (IFAT) and real-time PCR (qPCR) in blood, conjunctival swabs and hair samples
Data are n unless otherwise indicated
Prevalence of Leishmania species infection in cats detected by indirect fluorescent antibody test (IFAT)
Data are n (%)
NA = not available; DSH = domestic shorthair
The cat identified as positive from the hair sample using qPCR showed eight copies of Leishmania kinetoplast DNA/µl of DNA extract and was seronegative to L infantum (Table 2). This cat was a 10-year-old domestic shorthair castrated male that lived indoors and was sampled in August.
Forty-nine of 106 cats included in the NL group were male, of which 43/49 (87.8%) were castrated, while the remaining 57 were female, of which 45/57 (78.9%) were spayed. Ninety-two (86.8%) were >24 months old, with a median age of 7 years (range 4 months to 20 years); 92/106 (86.8%) were domestic shorthair cats and 93/106 (87.7%) were from the province of Bologna. Forty-nine of 106 (46.2%) were indoor cats and 47/106 (44.3%) were indoor/outdoor cats.
No statistical association was found regarding positivity to L infantum and sex, age, breed, geographical origin, access to outdoor environment and month of sampling. Although not significant, an increased frequency in positivity among cats sampled in July and August was observed (P = 0.071) (Table 1).
No agreement between the results obtained by IFAT and qPCR and between the qPCR carried out on three different biological samples was shown by the value of Cohen’s κ coefficient (P = 0.309).
Clinical and clinicopathological findings
The results of the CBC, serum chemistry, protein electrophoresis and urinalysis, including UPC, obtained in the PL and NL groups are shown in Table 4. Total proteins, beta (β)2-globulin and γ-globulin fractions were significantly increased in the PL group (Table 4). Hyperproteinaemia and hypergammaglobulinaemia were detected in 3/19 (15.8%) and 11/19 (57.9%) PL cats, respectively. The frequency of hyperproteinaemia and hypergammaglobulinaemia were significantly increased in the PL group when compared with the NL group (P = 0.028 and P = 0.038, respectively). No other significant differences between the groups selected were detected.
Clinicopathological results of the cats included in the positive Leishmania (PL) and negative Leishmania (NL) groups
Data are mean ± SD or median (range). Values in bold indicate statistical significance
RI = reference interval; RBCs = red blood cells; WBCs = white blood cells; Hb = haemoglobin; Hct = haematocrit; MCV = mean cell volume; MCH = mean cell haemoglobin; MCHC = mean cell haemoglobin concentration; MPV = mean platelet volume; RDW = red cell distribution width; ALT = alanine aminotransferase; AST = aspartate aminotransferase; ALP = alkaline phosphatase; GGT = gamma-glutamyl transferase; A:G = albumin:globulin ratio; SAA = serum amyloid A; USG = urine specific gravity; UPC = urine protein:creatinine ratio
No clinical signs associated with leishmaniosis were observed in any cat included in this study, and leishmaniosis was never suspected by the attending clinicians. In the PL group, cats with an antibody titre equal to 1:80 (n = 11/18) had gastroenteritis (n = 6), neurological disorders (n = 1) and chronic kidney disease (CKD; n = 2). Moreover, 2/11 in the PL group tested positive for feline calicivirus (FCV) and 1/11 for feline immunodeficiency virus. Cats with an antibody titre >1:80 (n = 7/18) had more severe diseases, including autoimmune thrombocytopenia (n = 1), hypertrophic cardiomyopathy (n = 1), hyperthyroidism (n = 2), neoplasia (n = 2), CKD (n = 3) and Addison’s disease (n = 1). One of the seven presented with coinfection with feline herpesvirus and FCV, and was treated with glucocorticoids. The only cat identified as positive by qPCR had CKD.
Six months after serological and molecular positivity, 8/19 (42.1%) cats included in the PL group died from causes not apparently attributable to leishmaniosis. Two of them had an IFAT titre of 1:80 upon admission, four had 1:160 and one had 1:320. The qPCR-positive cat was euthanased within 6 months of inclusion in the study owing to CKD progression.
Discussion
The primary aim of this study was to investigate the prevalence of Leishmania species infection in cats in Northern Italy, an area that is becoming endemic for CanL. The secondary aim was to determine whether there was an association between L infantum positivity and signalment and clinicopathological data.
Two different diagnostic methods, IFAT and qPCR, were used to evaluate the presence of Leishmania species infection in cats, and the overall prevalence detected was 12.5%. Recent studies carried out in Italy have reported a similar (12.2%) or a greater (25.8%) prevalence of infection.5,6 In the present survey, the use of a molecular test allowed the identification of an infected subject that had not developed a detectable antibody response, suggesting that serology may underestimate the real number of infected animals. 19
No standardised IFAT assay for the detection of antibodies against Leishmania species is currently available for cats, and an antibody titre value universally accepted as indicative of active infection has not been identified. 20 In the present study, seroreactivity (⩾1:40) to L infantum was found in 29.6% of the cats tested. Other studies have reported seroreactivity to range from 16.3% to 30% in Italy.5,21 However, positivity with a 1:40 titre is not indicative of infection; therefore, the seroprevalence could be overestimated using lower cut-offs.4,14 Based on the cut-off value of ⩾1:80,2,4,14 the seroprevalence detected decreased to 11.8%. Similar results have been reported in Northern Italy by Spada et al, 5 suggesting that feline infection has spread in recent years. In the present study, L infantum DNA was only identified in the hair sample of one cat. L infantum DNA was not detected in any of the blood and conjunctival swab samples, even in seropositive cats. The possible reasons for the inconsistency of DNA detection in blood is the absence of parasitaemia in cats except in the cases of severe disease or intermittent parasitaemia as described in dogs.22,23 Nevertheless, higher and variable values of molecular prevalence of L infantum infection assessed by testing feline blood and conjunctival samples have been reported in Italy,5,6,11,13,24 and in other European countries.12,19,25–28 These results suggest that future molecular studies should be carried out to evaluate the diagnostic accuracy of non-invasive samples in cats, as has already been done in dogs.
In the present study, L infantum DNA was detected in the hair of one cat, confirming the possible utility of this biological sample for the diagnosis of FeL, as has been reported for CanL. 29 The negative result obtained in the IFAT carried out on this cat could reflect an ineffective immune system response or could be explained by the absence of antibody production during an early stage of the infection. 25 Discordant findings between the serological testing and molecular methods have also been reported previously.20,30–33
No significant association was found between L infantum infection and signalment data, as has been reported in previous studies.5,25,33,34 An interesting finding of the present study, although not sustained by statistical significance, is the increased prevalence of infection in the cats sampled in July and August, probably owing to the increased Phlebotomus species activity in Southern Europe during these months. 35
Positivity for L infantum in cats was associated with a significant increase in total proteins, β2 and γ globulins, and with a decrease in A:G, although not significant when compared with the negative cases. Interestingly, 15.8% and 57.9% of the PL group had hyperproteinaemia and hypergammaglobulinaemia, respectively, and the frequency of these abnormalities was significantly increased when compared with the NL group. Hyperproteinaemia associated with hyperglobulinaemia has been commonly found in FeL,4,36–38 and, in a study carried out on four cats, hyperglobulinaemia was classified as hypergammaglobulinaemia. 36 However, the presence of hyperproteinaemia, decreased A:G, and polyclonal β and γ hyperglobulinaemia have long been reported in the course of CanL. 39 Therefore, it can be hypothesised that in cats, as in dogs, the humoral immune response could also be activated by the production of immunoglobulins after infection. When FeL is suspected in a feline patient, performing serum electrophoresis is recommended to support the diagnosis by means of evidence of β2 and γ polyclonal hyperglobulinaemia.
No other significant difference was found in the haematological and serum chemistry variables between the two groups studied, although mild-to-severe non-regenerative normocytic anaemia, thrombocytopenia, leukopenia, pancytopenia and hypoalbuminemia increased liver enzyme activity, and serum creatinine and urea concentrations have been the most frequent abnormalities previously reported in clinical cases of FeL.2,4,36–38,40 Proteinuria is often the only abnormal clinicopathological finding in dogs with CanL, 39 whereas it has been described only occasionally in cats with FeL.4,36,41 In this study, an increase in the UPC in the PL group was not observed as compared with the NL group. These results could be explained by the composition of the NL group. In fact, the NL group was not composed of healthy cats, but also included patients suffering from other diseases not related to L infantum infection, including numerous cats suffering from renal disease, which can be accompanied by several laboratory abnormalities.
The clinical cases of feline leishmaniosis reported have frequently been associated with skin lesions, as well as other less specific clinical signs.30,34 None of the positive cats included in this study showed clinical signs of leishmaniosis, and the disease had never been suspected by the attending clinicians. This suggests that subclinical feline infections could be common in areas that are becoming endemic for CanL and that clinical illness owing to L infantum in cats is currently rare. 2
Some limitations should be acknowledged when interpreting the present results. Sampling should have lasted for more than a year, and the number of cats sampled within the inclusion period should have been more homogeneous in each month. A second staggered sampling would have been necessary for the cats included in the UL group to assess the changes in the antibody response. Moreover, the study population consisted mainly of cats referred to a veterinary hospital and, consequently, the animals included in the study could have presented with clinicopathological abnormalities not related to L infantum infection, affecting the comparison between the PL and NL groups.
Conclusions
The results of the present study suggest the spread of L infantum infection in cats in Northern Italy, although in the majority of cases, the infection did not appear to lead to active disease. The use of a molecular technique for the detection of Leishmania species does not replace serology but may support the final diagnosis. Hyperproteinaemia and hypergammaglobulinaemia appeared to be significant clinicopathological abnormalities in cats with Leishmania species infection. Additional studies are needed to better assess the role of cats in the epidemiological cycle of Leishmania species and the pathogenic role of L infantum in this host.
Footnotes
Conflict of interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Ethical approval
This work involved the use of non-experimental animals only (owned or unowned) and followed established internationally recognised high standards (‘best practice’) of individual veterinary clinical patient care. Ethical approval from a committee was therefore not necessarily required.
Informed consent
Informed consent (either verbal or written) was obtained from the owner or legal custodian of all animals described in this work for the procedures undertaken. No animals or humans are identifiable within this publication, and therefore additional informed consent for publication was not required.
