Canine leishmaniosis is pushing northwards out of its traditional endemic regions as the distribution of its vector expands and increasing numbers of dogs travel between countries. In non-endemic areas, imported disease represents a threat to animal and human health. This article provides an understanding of the epidemiology, life cycle and transmission of Leishmania infantum, reviews its diagnosis and treatment, and briefly discusses the current understanding of feline leishmaniosis.
Statistics from Altmetric.com
Paolo Silvestrini qualified from the University of Perugia, Italy, in 2002 and spent three years working in private practice in Rome. In 2007 he gained the certificate of Specialist in small animal clinic and pathology from the University of Pisa, Italy. He completed an 18-month internship at the Universitat Autónoma de Barcelona in Spain, followed by a three-year residency in internal medicine at the same university. He is now senior lecturer in small animal internal medicine at the University of Liverpool. He is a diplomate of the European College of Veterinary Internal Medicine and holds RCVS recognised Specialist status in small animal internal medicine. In 2016 he completed a PhD focusing on novel clinicopathological aspects of canine leishmaniosis.
Canine leishmaniosis (CanL) is caused by the protozoan parasite Leishmania infantum (syn Leishmania chagasi). It is endemic in more than 70 countries throughout the world (Solano-Gallego and others 2011), including those of the Mediterranean basin, Portugal, Latin America and southern Asia. However, CanL is also an important concern in non-endemic countries where imported disease constitutes a veterinary and public health problem. Dogs are the main animal reservoir for human visceral leishmaniosis and, in people, the disease is usually fatal if not treated. Susceptible dogs commonly develop both visceral and cutaneous forms and if left untreated they generally die or are euthanased due to progressive severe renal disease.
Cats have been considered as a relatively resistant host species but several studies in the past decade have confirmed that feline leishmaniosis may be relatively common in areas where CanL is endemic (Box 1).
Feline leishmaniosis (FeL) has often been underestimated since clinical illness due to Leishmania infantum in cats is rare and cats have been generally considered a relatively resistant host species. However, subclinical feline infections are common in areas endemic for canine leishmaniosis. The prevalence rates of feline infection in serological and/or molecular-based surveys range from 0 per cent to more than 60 per cent (Pennisi and others 2015). Sandflies are able to acquire the parasite from cats and, therefore, cats may act as a secondary reservoir.
Cutaneous lesions such as dermal nodules, ulceration and crusts are predominant in FeL and are often localised on the head, nose and lips or on the distal parts of the limbs. Currently, there is no strong evidence regarding the best treatment options for FeL, but the majority of cases have been treated with variable success with allopurinol alone. Prognosis varies from good to poor and dissemination of Leishmania amastigotes to the spleen, lymph nodes, liver and gastrointestinal tract is often associated with clinical deterioration and death. No vaccines against Leishmania are available for cats and the single preventive strategy is topical insecticides. Unfortunately, most pyrethroids cannot be used in cats due to their toxicity in this species.
Epidemiology, life cycle and transmission
CanL is endemic in the Mediterranean areas of Europe (Cyprus, Greece, Albania, Croatia, Italy, Malta, France, Spain) and Portugal, the Middle East, Latin America and southern Asia (Maia and Cardoso 2015). Canine infection rates approach 70 per cent to 90 per cent, as shown by PCR and serology, in highly endemic foci, such as the Balearic Islands of Spain (Solano-Gallego and others 2001), the Marseille area in France (Berrahal and others 1996), throughout Greece (Leontidas and others 2002) and the Naples area in Italy (Oliva and others 2006). At least 2.5 million dogs are infected in south-western Europe alone (Moreno and Alvar 2002). The infection is spreading to non-endemic areas, such as North America and northern Europe, probably due to wider distribution of its vector and especially to larger numbers of dogs travelling from/to endemic countries.
Leishmania species completes its life cycle in two hosts, a phlebotomine sandfly vector, which transmits the flagellated infective promastigote form (Fig 1a), and a mammal, where the amastigote form (Fig 1b) develops and replicates inside the host’s macrophages. Female sandflies of the genus Phlebotomus in the Old World and Lutzomyia in the New World are the main vectors of CanL but other modes of transmission are possible. In utero transmission from an infected dam to its offspring and venereal transmission from infected males to healthy bitches have been documented (da Silva and others 2009). Transmission by haematophagus arthropods other than sandflies has been suspected but is not yet proven to have an epidemiological significance. Rhipicephalus sanguineous ticks have been shown to acquire Leishmania species in their gut after feeding on infected dogs (Coutinho and others 2005). Blood transfusion is another possible route of transmission and canine blood donors living in endemic countries should be routinely tested for CanL (Owens and others 2001, Giger and others 2002, Tabar and others 2008). Direct dog-to-dog transmission has also been suspected, especially in those areas where sandflies are not present (Duprey and others 2006).
Pathogenesis and clinical signs
CanL is the classical example of a disease where the clinical signs and underlying pathology are intrinsically related to the interaction between the parasite (virulence), arthropod vector (repeated infectious bites) and host (genetic background, immune response, coexisting diseases). Leishmania species has three general pathogenic features:
The parasite’s target cell is the macrophage, which is the site for its replication;
Establishment of infection and evolution of the disease both depend on the host’s immunological responses;
Once established, the infection usually persists in tissues.
Leishmania species tend to localise in all tissues in which monocytic-macrophagic cells exist in high numbers. Not all dogs infected with L infantum will eventually develop clinical leishmaniosis (Killick-Kendrick and others 1994, Pinelli and others 1995). Dogs that respond to infection with a T cell-mediated immunity (T helper 1; Th1) will remain asymptomatic (resistant dogs) while those that respond predominantly with a humoral immunity (T helper 2; Th2) will develop clinical signs (susceptible dogs). However, subclinical infection is not necessarily permanent, and factors such as immunosuppressive conditions or concomitant disease can break the equilibrium and lead to the progression of clinical disease.
The usually prolonged incubation period, which can extend from three months to seven years after infection, may explain the inconsistent and variable transition from a resistant to a susceptible state (Baneth and Solano-Gallego 2012). There is a strong genetic basis influencing susceptibility and resistance to CanL. Boxers, cocker spaniels, rottweilers and German shepherd dogs are more susceptible to developing signs of leishmaniosis in contrast to Ibizan hounds, in which clinical disease is rather rare due to the predominant cell-mediated Th1-type immune response (Sideris and others 1999, Solano-Gallego and others 2000, Franca-Silva and others 2003).
Age seems to be another important risk factor and clinical disease generally has two peaks, one in young dogs (two to four years old) and another in older dogs (more than seven years old) (Miranda and others 2008).
CanL can potentially involve any organ, tissue and/or biological fluid, and can manifest with a wide range of clinical signs. The proliferation of B lymphocytes, plasma cells, histiocytes and macrophages results in generalised lymphadenomegaly, splenomegaly and hyperglobulinaemia. This last response is not protective but detrimental, with the generation of autoantibodies and circulating immune complexes, which can consequently deposit in various tissues and organs, resulting in glomerulonephritis, vasculitis, uveitis, myositis and polyarthritis. Renal disease is recognised as the main cause of death or euthanasia in dogs with leishmaniosis (Planellas and others 2009) and generally progresses from mild azotaemia and proteinuria to nephrotic syndrome and end-stage kidney disease.
History and physical examination may reveal reduced appetite or anorexia, lethargy, emaciation, cachexia, peripheral lymphadenomegaly, exercise intolerance, skin lesions, temporal muscle atrophy, splenomegaly, polyuria/polydipsia, epistaxis, ocular lesions, onychogryphosis, lameness, vomiting and diarrhoea, which appear alone or, more often, in various combinations.
Skin lesions occur in 80 per cent to 90 per cent of cases (Ciaramella and others 1997, Koutinas and others 1999). They are rarely pruritic and include exfoliative dermatitis with focal or multifocal alopecia, generally localised on the face, ears and limbs; ulcerative dermatitis over bony prominences and in mucocutaneous junctions, paws and the ear pinnae; and focal or multifocal nodular dermatitis (Fig 2) (Ordeix and others 2005).
Ocular lesions are also common and can include uveitis, conjunctivitis, keratoconjunctivitis sicca, blepharitis, glaucoma and a combination of these (Fig 3).
Dogs with leishmaniosis may show signs of a haemorrhagic diathesis, manifest primarily as epistaxis, and less commonly as haematuria and haemorrhagic diarrhoea. The leading causes of epistaxis, which may be acute or chronic/recurrent, unilateral or bilateral, and sometimes severe enough to cause anaemia due to uncontrollable blood loss, are consistent with thrombocytopathy, increased serum viscosity due to hyperglobulinaemia, and rhinitis, ulcerative or not (Mylonakis and others 2008, Petanides and others 2008). Vasculitis may also contribute in some cases, causing bleeding ulcerations of the nasal philtrum and/or nostrils. Anaemia usually develops as a sequel to the decreased erythropoiesis of chronic disease or chronic kidney disease but may be aggravated by blood loss.
Less commonly, CanL has been recognised as the cause of myositis (Paciello and others 2009, Vamvakidis and others 2000), erosive or non-erosive mono- or polyarthritis, tongue nodules and papules or multifocal to diffuse ulcerative glossitis and stomatitis, chronic hepatitis and chronic colitis, pericarditis, myocarditis and pneumonia (Font and others 1993, Torrent and others 2005), meningoencephalomyelitis (Vinuelas and others 2001, Font and others 2004, Jose-Lopez and others 2012), orchitis, epididymitis, chronic prostatitis, penile granulomatous disease and balanoposthitis (Diniz and others 2006, Manna and others 2012, Mir and others 2012).
The diagnosis of CanL is difficult and complex because of the spectrum of clinical signs and laboratory abnormalities, the high prevalence of subclinical infection and coinfections with other vectorborne diseases (Baneth and others 2008) and, more recently, the use of vaccines in endemic areas. The main purpose for which diagnosis of CanL is carried out is to confirm disease in affected dogs. However, it is also important to investigate possible subclinical infection in clinically healthy dogs living in endemic areas, including blood donors, breeding dogs and dogs before vaccination; in dogs heading towards disease progression; and in clinically healthy dogs in non-endemic areas (travelling dogs), to avoid importing infected dogs to non-endemic regions and to monitor response to treatment (Solano-Gallego and others 2009, 2011).
The diagnostic investigation should always be based on an integrated approach considering signalment, history, clinical findings and the results of basic blood and urine analysis and of more specific diagnostic tests. Table 1 summarises the laboratory findings generally associated with CanL (Paltrinieri and others 2010).
Numerous techniques have been developed to help in the diagnosis of CanL. The detection of L infantum infection in dogs includes parasitological (cytology, histology, immunochemistry and culture of the organism), molecular (conventional, nested and real-time PCR) and serological methods (qualitative and quantitative antibody tests). Cutaneous lesions, bone marrow, lymph nodes, and spleen, and, less commonly, other tissues or body fluids such as joint, cerebrospinal and abdominal fluids, are good choice samples to observe Leishmania species amastigotes in both cytological and histological specimens (Solano-Gallego and others 2009).
Definitive histopathological identification of parasites within tissue macrophages may be difficult and an immunohistochemical staining method can be used. PCR on bone marrow, lymph node, spleen or skin has a high sensitivity and specificity, and quantitative real-time PCR allows the quantification of Leishmania species loads in tissues of infected dogs, which is important for diagnosis as well as for follow-up during the treatment.
Various serological methods have been used to detect serum anti-Leishmania-specific IgG antibodies, including indirect fluorescent antibody tests (IFAT), ELISA, immunochromatography with rapid in-house devices, direct agglutination assays and Western blotting. Generally, IFAT, ELISA and the rapid in-house kits are the most commonly employed methods (Bourdeau and others 2014). The IFAT is considered the reference method for anti-Leishmania serology in dogs (Gradoni and Gramiccia 2008, EFSA AHAW Panel 2015) based on its high sensitivity and specificity (near 100 per cent for both) except in areas endemic for Trypanosoma cruzi where it may give false positive results. ELISA is also very sensitive and specific (near 100 per cent for both) when a combination of immunodominant recombinant proteins is used as antigen; it has slightly lower specificity when crude parasite lysates are employed instead (Maia and Campiono 2008, Rodríguez-Cortés and others 2010, Solano-Gallego and others 2014).
Immunochromatographic rapid in-house tests have a quite acceptable specificity, but their sensitivity is usually low (in the range of 30 to 70 per cent) and largely dependent on the stage of infection (Paltrinieri and others 2016). Lowest sensitivity is associated with infected dogs without clinical signs, while highest sensitivity is seen for dogs with overt disease. Various in-house serological tests are commercially available and are particularly attractive to practising veterinarians because they give immediate results (Bourdeau and others 2014). However, their major disadvantage is that they are qualitative and thus any positive result needs to be followed by a quantitative test.
High antibody concentrations in a non-vaccinated dog with compatible clinical signs and clinicopathological abnormalities are almost diagnostic of the disease (Solano-Gallego and others 2016). However, the presence of lower antibody levels is not necessarily indicative of latent disease and needs to be confirmed by other diagnostic methods such as PCR, cytology or histology. Moreover, the interpretation of serological titres should take into consideration the possibility of the presence of antibodies elicited by vaccines since these have been introduced as preventive strategies against CanL in endemic areas.
Serological assays based on the detection of antibodies reactive with recombinant proteins and rapid in-house tests are usually less sensitive for the recognition of antibodies elicited by vaccination than those based on whole-parasite antigens and quantitative serological techniques (Marcondes and others 2013, Moreno and others 2014, Starita and others 2016). The rapid serological test Speed Leish K (Virbac) detects circulating antibodies against kinesins of L infantum and might be used to distinguish between vaccinated and naturally infected dogs. However, its diagnostic performance has been variable in some studies (Montoya and others 2017). It therefore remains important to develop new serological techniques that are able to discriminate between naturally occurring antibodies and antibodies elicited by vaccination; currently, if there is a recent history of vaccination then other diagnostic techniques based on observation of lesions (cytology/histology) and detection of parasites or parasite DNA (PCR) should be employed.
Therapy and follow up
CanL is more resistant to therapy than human leishmaniosis, and only rarely are Leishmania organisms completely eliminated with available drugs (Baneth and Shaw 2002, Noli and Auxillia 2005). Relapses necessitating retreatment are common, although dogs frequently become cured of the clinical disease. In addition, in endemic areas, reinfections occur and contribute to apparent treatment failure.
In order to prevent a recurrence of CanL, parasitostatic drugs, such as allopurinol, are usually combined with leishmanicidal therapy and continued for several months or years beyond apparent clinical cure (Baneth and Shaw 2002, Noli and Auxillia 2005). Allopurinol can be stopped only when all the following conditions are met:
Complete clinical recovery;
The most commonly prescribed dosages for allopurinol range between 5 and 20 mg/kg every 12 hours. Use of allopurinol causes hyperxanthinuria, which may produce urolithiasis in about 12 per cent of treated dogs (Torres and others 2011).
N-methyl-glucamine (meglumine) antimoniate is the most used pentavalent antimony compound for treating leishmaniosis in dogs and people. The drug selectively inhibits leishmanial glycolysis and fatty acid oxidation, causing the subsequent death of the parasite. Treatment with meglumine antimoniate induces a generalised reduction of the parasite load, together with a temporary restoration of cell-mediated immunological response. Pain and swelling of the injection site are the most common adverse effects. Fever, diarrhoea and loss of appetite have been also reported (Denerolle and Bourdoiseau 1999). To date, there has been no evidence of renal damage induced by antimonials in dogs. The most commonly reported treatment regimen is 100 mg/kg subcutaneously once daily for four weeks.
Miltefosine at 2 mg/kg orally once daily for four weeks combined with allopurinol is an alternative first-line protocol. Miltefosine was originally developed as an antineoplastic agent and is able to kill parasites in vitro and in vivo by disturbing signalling pathways and cell membrane synthesis, thus leading to apoptosis (Verma and Dey 2004, Farca and others 2012). The side effects usually include vomiting, seen in about 11 to 23 per cent of treated dogs (Mateo and others 2009, Woerly and others 2009, Andrade and others 2011). The combination of miltefosine and allopurinol is clinically as effective as the standard protocol based on meglumine antimoniate and allopurinol. In one study (Miró and others 2009), a significant reduction in clinical scores and parasite load was observed in both groups with no significant differences.
Dogs should be re-evaluated after one, three and six months of therapy and then once every six months for life. Each evaluation should include complete physical examination, haematology, biochemistry profile, serum protein electrophoresis and urinalysis with urine protein:creatinine ratio to quantify the level of proteinuria. After one month, dogs responding to therapy and showing clinical improvement will have a significant reduction of total proteins and globulins and an increase of albumin on serum protein electrophoresis. In contrast, the serological titre will remain elevated and a significant decrease will be evident only around the third month after the start of therapy. It is therefore recommended that serology should be rechecked after three and six months of therapy (omitting serology at the one month check) to document a progressive reduction in the titre. Real-time PCR can help monitor the response to therapy and identify relapses. Table 2 summarises clinical staging of CanL based on serological status, clinical signs and laboratory findings, and the types of therapy and prognosis for each clinical stage (Solano-Gallego and others 2017).
There are two predominant methods to prevent CanL:
Vector control through the use of insect repellent to avoid infectious bites;
Chemotherapeutic and/or immunological control using leishmanicidal/leishmaniostatic drugs, immunomodulators and vaccines.
As discussed above, sandflies are the primary route of infection in endemic areas, so the first line of control is the blockade of transmission though the use of insect repellents. Multiple studies have evaluated the efficacy of various active ingredients and compounding, including collars, spot-ons and sprays.
Immunotherapy is an expanding area of research in which new and old molecules have been tested. Given that CanL dramatically and negatively modulates the host’s immune system, the main objective of immunotherapy is to help the host’s immune system to control the infection. Domperidone is a gastric prokinetic and antiemetic drug, acting as a dopamine D2 receptor antagonist, that is able to stimulate the production of serotonin, which in turn increases prolactin secretion. Prolactin, besides its well-known role in milk production, is an important pro-inflammatory lymphocyte-derived cytokine that is able to stimulate the cell-mediated Th1 lymphocyte-driven immune response. Domperidone has been licensed for the veterinary market with an indication for the treatment and prevention of CanL at the dose of 0.5 mg/kg orally once daily for one month, repeated every three months (Sabaté and others 2014).
Three commercial vaccines – Leishmune (Fort Dodge), Leish-Tec (Hertape Calier Saúde Animal) and CaniLeish (Virbac) – have been licensed for the control of CanL, the first two in Brazil, and the third in Europe. The European Medicines Agency has authorised a new vaccine product (Letifend; Laboratorios LETI) for the European market, while the Brazilian health authorities have withdrawn Leishmune. Table 3 summarises the main vaccine products for CanL in South America and Europe (Solano-Gallego and others 2017).
Past and currently marketed vaccines appear to elicit a long-lasting, parasite-specific, cellular immunity (Moreno and others 2014). However, the main controversy regarding CanL vaccines is that they do not block the establishment of infection. This could potentially keep an infected dog (without clinical disease) healthy, but possibly able to spread infection to people and dogs. Moreover, all vaccines are recommended solely for clinically healthy and seronegative dogs. This can sometimes be very difficult to ensure as prescreening tests may not be sensitive enough to detect all subclinically infected dogs. It is important to highlight that vaccinated infected dogs have been shown to be infectious to sandflies (Bongiorno and others 2013). Moreover, the consequences of vaccinating seropositive dogs are currently unknown, but there is a risk that these dogs would develop clinical leishmaniosis (Solano-Gallego and others 2017).
Cases of CanL are uncommon in the UK, occurring only in dogs that have travelled to endemic areas. The UK also lacks the sandfly vector for the parasite, removing the threat of onward transmission of the disease by natural means. However, with the increasing numbers of dogs that enter the UK each year from countries where the disease is endemic, practitioners need to be aware of the signs of the disease, how to diagnose it, and the action to take should they confirm a case.
In Practice has partnered with BMJ OnExamination to host the self-assessment quizzes provided with each clinical article. These can be completed on the online version of each article at www.inpractice.bmj.com
Supplementary file 1
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.