Heliyon 5 (2019) e02373 Contents lists available at ScienceDirect Heliyon journal homepage: www.heliyon.com Leptospira seroprevalence in owned dogs from Spain pez a, *, A. Vila a, J. Rodo M.C. Lo n b, X. Roura a a Hospital Clínic Veterinari, Autonomous University of Barcelona, Campus UAB, Carrer de l'Hospital, s/n, 08193, Bellaterra, Barcelona, Spain b Idexx Laboratories, Carrer del Plom, 2, 08038, Barcelona, Spain A R T I C L E I N F O A B S T R A C T Keywords: Leptospira is a widespread zoonosis that has been linked to transmission between dogs and humans. The main Microbiology purposes were to determine the seroprevalence of anti-Leptospira serum antibody and to identify the most com- Systems biology mon serovars in dogs in Spain. This is a cross-sectional study with 1,310 records of canine Leptospira testing data Zoology from Spain since 2015 to 2017. Inclusion criteria were individual cases with MAT test results for 8 serovars Infectious disease Veterinary medicine (Bratislava, Icterohaemorrhagiae, Australis, Pomona, Grippotyphosa, Autumnalis, Canicola and Saxkoebing) and Public health to have the zip code data. Three hundred and thirty-eight samples (25.8%; 95%CI 23.6–28.4) were seropositive Epidemiology (1:100). According to geographic areas, North had the highest seroprevalence (38.0%; 95%CI 28.9–47.1) fol- Infectious disease lowed by South (29.4%; 95%CI 20.1–38.8), Center (28.6%; 95%CI 24.3–33.0), Mediterranean (22.3%; 95%CI Internal medicine 19.1–25.6) and Northwest (22.2%; 95%CI 7.9–36.4). Seropositivity (MAT 1:100) was most common to serovars Leptospirosis Icterohaemorrhagiae (19.4%; 95%CI 17.2–21.5) and Bratislava (8.5%; 95%CI 7.0–10.0), followed by Grippoty- MAT phosa (7.2%; 95%CI 5.8–8.6), Australis (6.4%; 95%CI 5.0–7.7), Autumnalis (5.0%; 95%CI 3.8–6.2), Pomona Canine leptospirosis (4.5%; 95%CI 3.3–5.6), Canicola (3.4%; 95%CI 2.4–4.4) and Saxkoebing (0.8%; 95%CI 0.3–1.3). An association Europe was found between positivity (MAT 1:100) and males (P ¼ 0.003) and dogs that were 6 years old or older were at higher risk of exposure (P ¼ 0.001; OR 4.61; 95%CI 1.86–11.43). This study has shown that dogs in Spain are commonly exposed to Leptospira infection and points out the necessity to control the prevalence of this severe widespread zoonosis in dogs and humans. 1. Introduction Grippotyphosa is common in mainland Europe but appears to be rare in the UK and Ireland, perhaps associated with the distribution of relevant Leptospirosis is a zoonotic disease caused by pathogenic spirochetes reservoir hosts (Ellis, 2010). In Germany, the most common serovars in with worldwide distribution affecting most mammalian species (Bharti dogs are Grippotyphosa and Sejroe (Geisen et al., 2007), while the et al., 2003). serovar Australis appears to be common in Italy (Mastrorilli et al. 2007; Several studies have reported variable prevalence in dogs ranging Tagliabue et al., 2016). Infection with serovar Javanica has been from 1.8% in Australia to 71.1% in India (Davis et al., 2008; Zwijnenberg described in dogs and humans in Europe, however, no reports have been et al., 2008; Shi et al., 2012; Ambily et al., 2013; Bier et al., 2013; Ojha found correlating infection with this serovar and clinical signs in dogs et al., 2018). In Europe, the prevalence ranges from 6% in Ireland to 49% (VandenBroek et al., 1991; Cacciapuoti et al., 1994). As currently in Italy (Scanziani et al., 2002; Burriel et al., 2003; Schuller et al., 2015b; available anti-leptospiral vaccines are primarily serovar specific, the Llewellyn et al., 2016; Habus et al., 2017). recognition of leptospires commonly involved in canine exposure in Leptospirosis in dogs has been associated with serovars Canicola and specific areas is crucial in order to keep vaccination strategies up to date. Icterohaemorraghiae but it is now clear that dogs are susceptible to Canine vaccines used in Spain contain L. interrogans serovars Canicola infection with multiple serovars (Schuller et al., 2015a). In Europe, and Icterohaemorragiae, although since 2012 vaccines also include Grippotyphosa and Bratislava have emerged as major causes of canine L. interrogans serovar Bratislava and L. kirshneri serovar Grippotyphosa leptospirosis (Ellis, 2010; Mayer-Scholl et al., 2013; Renaud et al. 2013). (Klaasen et al., 2013). The rate of use of these vaccines has not been Based on the available antibody prevalence data in Europe, most prev- described in Spain, however multiple positive titers in MAT to vaccine alent serovars are Icterohaemorrhagiae, Grippotyphosa, Australis, serovars could be an indication of recent vaccination. Hardjo, and Canicola (Ellis, 2010). Seropositivity to the serogroup In contrast to other European countries, there have been limited * Corresponding author. E-mail address: mariavirgolopez@hotmail.com (M.C. L opez). https://doi.org/10.1016/j.heliyon.2019.e02373 Received 19 March 2019; Received in revised form 15 May 2019; Accepted 22 August 2019 2405-8440/© 2019 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). M.C. L opez et al. Heliyon 5 (2019) e02373 canine seroprevalence studies in Spain. In dogs, prevalence data is although no consistent agreement on age groupings was found among available for two regions of Spain; Andalusia with 35.7% (Millan et al., studies. For association and risk factor analysis, age groups were 2009) and Comunidad Valenciana with 19.8% (Benito et al., 2005). regrouped in <6 years and 6 years. Furthermore, serovars were only determined in Andalusia by indirect Geographic mapping was used to characterize the distribution of the microscopic agglutination test (MAT), considered 1:100 the cut-point to disease in dogs throughout Spain by zip code. Spain is geographically positive sera, detecting Canicola, Icterohaemorrhagiae, and Australis divided into five domains (Estrada-Pe~ na et al. 2017). These are: 1–2) (Mill an et al., 2009). North and Northwest, which are humid, with mild winters and summers, A recent meta-analysis study identified being male sex and an urban and the rough features of natural vegetation are similar in both North and dogs as major risks factors for leptospirosis (Az ocar-Aedo and Monti, Northwest, but weather is warmer and humid in the latter; 3) Center, 2016). Dog ownership has also been identified as a risk factor for human which is a large area at high altitude (average above 800 m above the sea leptospirosis in Nicaragua (Trevejo et al., 1998), Barbados (Douglin et al., level and therefore has a continental type climate); 4) Mediterranean, 1997) and Germany (Jansen et al., 2005) suggesting transmission of which is also wet because of the influence of the Mediterranean Sea, with Leptospira spp. from dogs to humans. Dogs could serve as an important mild winters and hot summers; and 5) South, which is the warmest and sentinel specie for human infection, as well as indicator of the presence of driest region. leptospires in specific environments (Ghneim et al., 2007; Major et al., Descriptive statistics, relative prevalence calculations, univariate and 2014; Schuller et al., 2015a). multivariate analyses were performed using IBM SPSS Statistic version The aims of this study were: 1) to determine the seroprevalence of 25 for Mac. All statistical analyses were performed on data from 1,310 anti-Leptospira serum antibody in owned dogs from Spain; 2) to know the dogs and to the serovars they were exposed without exclusion of vaccinal most common canine serovars and if there is any relation between the serogrups (Icterohaemorrhagiae, Canicola, Grippotyphosa and Aus- region and serovars; and 3) to determine if geographic area, season, age tralis). Prevalence of different serovars in seropositive dogs was and sex could be risk factors for exposure to pathogenic leptospires in computed by geographic region. Season prevalence was determined for dogs. complete year data samples. In order to identify risk factors, the inde- pendent variables as sex (male, female), season (winter, spring, summer 2. Material and methods and autumn), age (<6 years and 6 years) and region (Mediterranean, Northwest, North, Center and South) with MAT seropositivity at recip- A cross-sectional study design was used to determine the prevalence rocal titers of 1:100 (ALL 100) or 1:400 (ALL 400) were assessed. of antibodies against various Leptospira serovars in owned dogs from Moreover, independent variables were also assessed with positive titers Spain. Sampling was done by convenience with serological canine (100) to each serovar tested in MAT. leptospirosis MAT results in Spain obtained through an agreement with Univariate analysis was performed using χ2 or Fisher's exact test for IDEXX Laboratories (Ludwisburg). The results of 1,310 individual tests categorical variables. A value of P < 0.05 was considered as the critical were obtained from the proprietary database. The clinical history, level of significance. For multivariate analysis a logistical regression was vaccination status and whether samples were submitted for paired MATs performed. Risk factor was statistically significant at a p-value <0.05. were unknown. All MAT results for tests conducted from January 2015 to Geographic mapping was used to characterize the distribution of the July 2017 were included. The laboratory is accredited according to ISO disease in dogs throughout Spain by zip code using QGIS 3.4 (QGIS 17025 and regularly participates in the International Leptospirosis So- Development Team, 2018. QGIS Geographic Information System. Open ciety MAT Proficiency Testing Scheme. Source of cultures was the na- Source Geospatial Foundation Project). tional reference laboratory (NRL) for leptospirosis (Bundesinstitut für Risikobewertung, BfR). MAT was performed according to OIE standards 3. Results (Office International des Epizooties OIE, 2008). Samples were tested for the presence of antibodies to a panel of 8 serovars (Table 1). Dogs tested A total of 1,310 samples met the inclusion criteria. All dogs included against the eight serovars with zip code of origin and dates of the test in this study were privately owned pet dogs. Three hundred thirty-eight done, were eligible for inclusion. MAT positivity was defined as positive out of 1,310 dogs (25.8%; 95%CI 23.6–28.4) were positive titer (100) reaction to at least one serovar included in the 8 serovars panel at to at least one serovar tested. One hundred and twenty-two (9.3%; 95%CI reciprocal titers of 1:100 (ALL 100) or 1:400 (ALL 400). Two instead 7.8–10.9) were positive for 400. One hundred and fifty-one dogs out of of one cut-off titers were chosen in the absence of a consensus to what 338 (44.6%; 95%CI 39.3–50.0) had more than one positive titer (100) represents an ideal cut-off titer to document exposure in a population of with a total of 723 different positive titers. The geographical distribution dogs. A dog could be classified as having more than one positive serovar of positive sera across Spain is shown in Fig. 1. titer. As described in Table 2, seropositivity for MAT titer 100 (n ¼ 338) Samples analyzed were grouped by month of diagnosis (spring - was most common to serovar Icterohaemorrhagiae followed by Bra- March 1st, summer - June 1st, autumn - September 1st, and winter - tislava, Grippotyphosa, Australis, Autumnalis, Pomona and Canicola with December 1st), by sex (female, male), and by age (categorized as 0 to <2 similar prevalence, and Saxkoebing with the lowest percentage. A MAT years, 2 to <6 years, 6 to <10 years, and 10–15 years). Cut-offs for age titer 400 (n ¼ 122) was quite similar among Icterohaemorrhagiae, categories were chosen to create 4 age groups approximately corre- Bratislava and Grippotyphosa, and between Pomona, Australis, Autum- sponding to those in previous studies (Grayzel and DeBess, 2016) nalis and Canicola, with lowest prevalence of Saxkoebing. For a higher MAT titer 1600 (n ¼ 44) distribution kept same patter where Icter- Table 1 ohaemorrhagie was the most common serovar, followed by Grippoty- Panel of 8 Leptospira spp. used as live antigens for microscopic agglutination phosa and Bratislava (Table 2). testing. Serogroup analyses was also performed for all samples, at lower titers Icterohaemorrhagiae was the most common serogroup found followed by Genomospecies Serogroup Serovar Australis and Grippothyposa but at higher titers, Grippothyposa was L. interrogans Australis Bratislava more common than Australis (Table 3). Australis Australis Autumnalis Autumnalis Total co-reactivity prevalence of positive dogs with 100 and 400 Icterohaemorrhagiae Icterohaemorrhagiae cut-off was analyzed (Table 4). The most common serovars with double Pomona Pomona exposure (400) were Bratislava and Icterohaemorrhagiae (7/25, L. borgpetersenii Sejroe Saxkoebing 28.0%), Icterohaemorrhagie and Canicola (4/25, 16.0%), and Bratislava L. kischeneri Grippothyphosa Grippothyposa and Australis (3/25, 12.0%). Co-reactivity for serogroups in multivalent 2 M.C. L opez et al. Heliyon 5 (2019) e02373 Fig. 1. The geographical distribution of positive dogs tested for serum antibodies Leptospira spp. with MAT (n ¼ 338). Geocoding was performed based on the zip code. Geographic areas: Mediterranean (M), Northwest (NW), North (N), Center (C), and South (S). Table 2 Number of individuals (n), relative prevalence and 95% confidence interval (CI) per Leptospira serovar for all dogs (n ¼ 1310). Tested serovars: Bratislava (BRA), Icterohaemorrhagie (ICT), Australis (AUS), Pomona (POM), Grippotyphosa (GRI), Autumnalis (AUT), Canicola (CAN) and Saxkoebing (SAX). MAT Titer All Serovars BRA ICT AUS POM GRI AUT CAN SAX 1:100 n 338 111 254 84 59 94 66 44 11 % 25.8 8.5 19.4 6.4 4.5 7.2 5.0 3.4 0.8 CI 23.6–28.4 7.0–10.0 17.2–21.5 5.0–7.7 3.3–5.6 5.8–8.6 3.8–6.2 2.4–4.4 0.3–1.3 1:200 n 223 84 155 56 43 67 41 21 9 % 17.0 6.4 11.8 4.3 3.3 5.1 3.1 1.6 0.7 CI 15.0–19.1 5.1–7.7 10.1–13.6 3.2–5.4 2.3–4.2 3.9–6.3 2.2–4.1 0.9–2.3 0.2–1.1 1:400 n 122 54 72 29 30 35 25 13 8 % 9.3 4.1 5.5 2.2 2.3 2.7 1.9 1.0 0.6 CI 7.8–10.9 3.0–5.2 4.3–6.7 1.4–3.0 1.5–3.1 1.8–3.5 1.2–2.6 0.4–1.5 0.2–1.0 1:800 n 77 35 39 14 16 23 15 3 4 % 5.9 2.7 3.0 1.1 1.2 1.8 1.1 0.2 0.3 CI 4.6–7.2 1.8–3.5 2.1–3.9 0.5–1.6 0.6–1.8 1.0–2.5 0.5–1.7 0.0–0.5 0.0–0.6 1:1600 n 44 11 21 7 6 16 5 1 2 % 3.4 0.8 1.6 0.5 0.5 1.2 0.4 0.1 0.2 CI 2.4–4.4 0.3–1.3 0.9–2.3 0.1–0.9 0.1–0.8 1.6–1.8 0.1–0.7 0.0–0.2 0.0–0.4 1:3200 n 19 2 8 2 1 11 - 1 - % 1.5 0.2 0.6 0.2 0.1 0.8 - 0.1 - CI 0.8–2.1 0.0–0.4 0.2–1.0 0.0–0.4 0.0–0.2 0.3–1.3 - 0.0–0.2 - 1:6400 n 11 2 4 - 1 5 - 1 - % 0.8 0.2 0.3 - 0.1 0.4 - 0.1 - CI 0.3–1.3 0.0–0.4 0.0–0.6 - 0.0–0.2 0.1–0.7 - 0.0–0.2 - vaccine (Australis, Icterohaemorrhagiae, Canicola, Grippothyphosa) for for Bratislava, Icterohaemorrhagiae and Autumnalis serovars (Table 5). titer 100 and 400 was 4/183 and 3/67, respectively. Seasonality was analyzed only for 2015 and 2016 samples. Autumn Age was known in 209/1310 (15.9%) of dogs and ranged from 0 to 15 had the major prevalence with 82 positive dogs (37 and 45, respectively); years old (median 5.1 IQR 2–8). Distribution of age groups was 49/209 followed by summer 60 (29 and 31), winter 57 (29 and 28) and spring 54 (23.4%) from 0 to <2 years, 64/209 (30.6%) from 2 to <6, 71/209 (30 and 24). Statistically significant relation was found between Grip- (34.0%) from 6 to <10, and 25/209 (12%) for 10. Prevalence for titers pothyposa positive titers and season (P ¼ 0.039), higher prevalence was 100 was 8.2% (4/49) in <2 years, 21.8% (14/64) from 2 to <6, 32.3% found in autumn and winter (9,0% and 9.6%, respectively) compared to (23/71) from 6 to <10, and 32.0% (8/25) in 10. Dogs with 6 years in spring (5.4%) and summer (4.0%). 100 cut-off were in major risk of being seropositive (OR 4.6; 95% CI According to geographic areas with a 100 cut-off, association is 1.9–11.4, P ¼ 0.001) as well as for Icterohaemorrhagiae (OR 3.7; 95% CI described in Table 5, statistically significant association was found with 1.4–9.7, P ¼ 0.007) (Table 6). Sex was available for 309/1310 (23.6%) higher prevalence in North area (P ¼ 0.001). For a 400 cut-off this dogs and was equally distributed with 180/309 (58.3%) male and 129/ relation was also found statistically significant (P ¼ 0.042) with preva- 309 (41.7%) female. A total of 51/180 (28.3%) males and 18/129 lence of 15.9% in North, 13.7% in South, 9.0% in Center, 8.3% in (13.9%) females were positive at least to one serovar. Evidence of posi- Northwest and 7.8% in Mediterranean region. In multivariate analyses tive serovars were more frequently found in males (87.5%) than females no statistically significant risk factor was found between region and (12.5%, P ¼ 0.015) for MAT titers 400. This association was also found positivity (Table 6). 3 M.C. L opez et al. Heliyon 5 (2019) e02373 Table 3 Number of individuals (n), relative prevalence and 95% confidence interval (CI) per Leptospira serogroup for all dogs (n ¼ 1310). Tested serogroups: Australis (AUS), Icterohaemorrhagie (ICT), Pomona (POM), Grippotyphosa (GRI), Autumnalis (AUT), Canicola (CAN) and Sejroe (SAX). MAT Titer All Serogroups AUS ICT POM GRI AUT CAN SEJ 1:100 n 338 117 254 59 94 66 44 11 % 25.8 8.9 19.4 4.5 7.2 5.0 3.4 0.8 CI 23.6–28.4 7.4–10.5 17.2–21.5 3.3–5.6 5.8–8.6 3.8–6.2 2.4–4.4 0.3–1.3 1:200 n 223 89 155 43 67 41 21 9 % 17.0 6.8 11.8 3.3 5.1 3.1 1.6 0.7 CI 15.0–19.1 5.4–8.2 10.1–13.6 2.3–4.2 3.9–6.3 2.2–4.1 0.9–2.3 0.2–1.1 1:400 n 122 57 72 30 35 25 13 8 % 9.3 4.4 5.5 2.3 2.7 1.9 1.0 0.6 CI 7.8–10.9 3.3–5.5 4.3–6.7 1.5–3.1 1.8–3.5 1.2–2.6 0.4–1.5 0.2–1.0 1:800 n 77 37 39 16 23 15 3 4 % 5.9 2.8 3.0 1.2 1.8 1.1 0.2 0.3 CI 4.6–7.2 1.9–3.7 2.1–3.9 0.6–1.8 1.0–2.5 0.5–1.7 0.0–0.5 0.0–0.6 1:1600 n 44 15 21 6 16 5 1 2 % 3.4 1.1 1.6 0.5 1.2 0.4 0.1 0.2 CI 2.4–4.4 0.6–1.7 0.9–2.3 0.1–0.8 1.6–1.8 0.1–0.7 0.0–0.2 0.0–0.4 1:3200 n 19 3 8 1 11 - 1 - % 1.5 0.2 0.6 0.1 0.8 - 0.1 - CI 0.8–2.1 0.0–0.5 0.2–1.0 0.0–0.2 0.3–1.3 - 0.0–0.2 - 1:6400 n 11 2 4 1 5 - 1 - % 0.8 0.2 0.3 0.1 0.4 - 0.1 - CI 0.3–1.3 0.0–0.4 0.0–0.6 0.0–0.2 0.1–0.7 - 0.0–0.2 - et al., 2009) and in a mix (owned and un-owned) population of dogs Table 4 (167/864, 16.7%) in Comunidad Valenciana (Benito et al., 2005). The Number of individuals positive against one, two or more serovars at a MAT Titer prevalence in Andalusia was higher than the mean of the present study, of 100 and 400. even if it is compared with our South region's prevalence (29.4%). A Nº seropositive serovars MAT 1:100 n (%) MAT 1:400 n (%) possible explanation for this difference is that Millan's study (2009) 1 183 (54.2) 55 (45.0) evaluated rural and un-owned dog's population. However, in Comunidad 2 54 (16.0) 25 (20.5) Valenciana the described prevalence was lower than in the present study. 3 32 (9.5) 20 (16.4) A plausible explanation to this would be that dogs were randomly taken 4 27 (8.0) 11 (9.0) 5 24 (7.1) 9 (7.4) for analyses in the previous study (Benito et al., 2005). It should be noted 6 14 (4.1) 2 (1.6) that in our study a possible bias could raise the prevalence because 7 4 (1.2) 0 (0) samples were collected from a laboratory and we hypothesized that 8 0 (0) 0 (0) probably received most of them from veterinary clinicians in order to confirm or reject leptospirosis as a potential diagnosis of sick dogs. Dogs with MAT titer 100 were divided by serovar and region as described Although there was some variability between the different in Table 7, only one statistically significant association was found between geographic areas, the most frequent serovars found in this study were Grippothyposa serovar and region (P ¼ 0.003), with higher prevalence in Icterohaemorrhagiae, Bratislava, Grippotyphosa and Australis. Serovar Center (47/422, 11.1%) followed by South (7/95, 7.4%), North (10/113, Icterohaemorrhagiae was the most prevalent especially when 100 cut- 8.8%), Northwest (2/36, 5.4%) and Mediterranean (28/644, 4.3%). off was used. This could be explained as a result of vaccination bias; however, it is also remarkable that this high prevalence was kept in 4. Discussion higher cut-offs (400) when vaccination bias could be more unlikely. This serovar has been previously described as the most frequent in wild The analyses conducted in this study evidenced that Spain's dogs are and domestic animals in southern Spain (Millan et al., 2009). exposed to eight pathogenic Leptospira serovars. Our prevalence results Conversely, our study showed a low rate of positivity for Canicola (25.8% at 100 and 9.3% at 400 cut-off) were similar to previous data serovar consistent with the wide use of vaccination could be responsible of seroprevalence of anti-Leptospira antibodies reported in Europe of the decrease of positivity in Italy (Scanziani et al. 1994, 2002; (Scanziani et al., 2002; Burriel et al., 2003; Krawczyk, 2005; Schuller Tagliabue et al., 2016) and Europe (Andre-Fontaine, 2006). Although et al., 2015b; Llewellyn et al., 2016; Delaude et al., 2017; Habus et al., both vaccine serogroups (Icterohaemorrhagiae and Canicola) are ex- 2017). pected to induce agglutinating antibodies, in our study the seropreva- Leptospira prevalence in dogs from Spain has only been previously lence of Icterohaemorrhagiae antibodies was higher than those of described in free-roaming dogs (10/28, 35.7%) in Andalusia (Millan Canicola similar to a previous study with canine serum samples in France Table 5 Analysis of association (χ2) between season, sex, region or age with positive microscopic agglutination test against any of the tested serovars at MAT cut-offs 1:100 (ALL 100) and 1:400 (ALL 400); and against each serovar at MAT cut-off 1:100 for serovars Bratislava (BRA), Icterohamorrhagiae (ICT), Australis (AUS), Gripo- thypposa (GRI), Autumnalis (AUT), Canicola (CAN) and Saxkoebing (SAX). Variable ALL100 ALL400 BRA100 ICT100 AUS100 POM100 GRI100 AUT100 CAN100 SAX100 SEASON 0.187 0.060 0.232 0.254 0.512 0.149 0.039* 0.09 0.373 0.426 SEX 0.003* 0.015* 0.015* 0.023* 0.072 0.407# 1.000# 0.023# * 0.085# 1.000# REGION 0.001* 0.042* 0.373 0.301 0.402 0.208 0.003* 0.257 0.241 - AGE 0.001* 0.735# 0.729# 0.000* 1.000# 0.689# 0.456# 1.000# 0.663# - * Statistically significant (p < 0.05). # Fisher's test used. 4 M.C. L opez et al. Heliyon 5 (2019) e02373 Table 6 increase with titers to Grippotyphosa and Pomona in dogs (Brown et al., Multivariable analysis of the association between sex, age, season and region 1996; Greenlee et al., 2004) making the analysis of the infecting serovar with positive against any serovar at MAT titers 100, 400 and Icterohaemor- even more challenging. Co-reactivity results for titers 400 also showed rhagiae (ICT) serovar for titer 100. a high prevalence of Bratislava and Icterohaemorrhagiae (28.0%) as well Covariate Levels Cut-Off Odds ratio 95%CI p-value as Icterohaemorrhagiae and Canicola (16.0%), all these serovars are Sex Female Ref. found in multivalent vaccines and that could be an indicator of recent Male 1:100 2.16 0.90–5.19 0.086 vaccination. Positive titers for only the four vaccinal serogroups was 1:400 4.03 0.44–37.14 0.218 found in four dogs, while multiple co-reactivities with no vaccinal ICT 1.53 0.60–3.87 0.367 serogroups were found, this fact could be explained by cross-reaction in Age <6y Ref 6y 1:100 4.61 1.86–11.43 0.001* MAT test which is a challenge for MAT interpretation for vaccination 1:400 2.73 0.41–18.17 0.300 without previous history of vaccination status. ICT 3.73 1.42–9.77 0.007* In order to assess areas of increased risk, the prevalence of Leptospira Season W Ref. serovars amongst 5 different geographic regions was compared. This SP 1:100 0.72 0.21–2.44 0.602 geographic classification was based on weather variations found in 1:400 1.37 0.09–20.15 0.818 ICT 0.62 0.18–2.22 0.467 different areas of Spain but does not consider the potentially significant SU 1:100 0.50 0.14–1.83 0.295 regional variations as a flora and fauna or the presence of stagnant water 1:400 1.21 0.09–16.31 0.890 and the wildlife populations. The prevalence was markedly higher in ICT 0.73 0.20–2.67 0.634 North region (38.9%) compared with quite similar prevalence found in A 1:100 0.52 0.15–1.87 0.319 the rest of areas of Spain. Furthermore, a significant relation between 1:400 0.87 0.04–37.14 0.926 ICT 0.36 0.09–1.48 0.160 region and seropositive samples was found but no significant risk factor Region S Ref. with any region was found. All this data could be explained due to the M 1:100 0.33 0.08–1.37 0.126 humid weather and more water sources available in North area. How- 1:400 0.15 0.01–2.70 0.195 ever, this high prevalence was not found in Northwest with similar ICT 0.40 0.08–1.86 0.242 NW 1:100 - - - climatological features, but the sample size from this region was too 1:400 - - - small to draw any conclusions and also the different sample size from ICT - - - different regions could be a bias for our study. N 1:100 1.27 0.18–8.93 0.813 Higher prevalence could be biologically plausible in regions with 1:400 2.04 0.09–46.21 0.654 lower altitudes due to less extreme climatic conditions throughout the ICT 0.90 0.10–7.83 0.242 C 1:100 0.55 1.86–11.43 0.423 year (Adler, 2015). Moreover, a recent study about Leptospira infection in 1:400 0.68 0.06–8.11 0.760 Mediterranean periurban micro-mammals reported a prevalence of 12% ICT 0.67 0.14–3.21 0.618 that could be a high risk factor for infection in dogs and humans in this *Statistically significant (p < 0.05). area (Millan et al., 2018). In our study, no risk association with Medi- terranean area was found. (Andre-Fontaine, 2006). Perhaps, this could be explained by the higher Leptospirosis in the United States has been characterized as having a infection pressure in the environment of the former strain, compared to seasonal distribution that favors late summer and early fall (Ward, 2002). the latter. A similar pattern can be found across Europe (Jansen et al., 2005; Bar- The second most prevalent serovar reported in this study was Bra- anton and Postic, 2006; Habus et al., 2017). In this study, only a signif- tislava. This was consistent with previous studies in dogs in Italy (Scan- icant association between Grippotyphosa and season was found (Table 5) ziani et al., 2002) and Scotland (VandenBroek et al., 1991). In Spain, with higher prevalence in autumn. Bratislava was considered the most prevalent serovar in cattle (Alon- The high prevalence in male dogs observed in our study agreed with so-Andicoberry et al., 2001; Guitian et al., 2001; Atxaerandio et al., previous results described for canine leptospirosis (Az ocar-Aedo and 2005). Monti, 2016). Similar to previous studies in dogs, our study also revealed Our study showed that the third most frequent co-reactivity was to correlation between age and MAT seropositivity, where being younger Australis and Bratislava serovars (12.0%), both belong to the serogroup than 1-year-old was a protective factor and, in contrast, being 6 years old Australis. That co-reactivity could be due to a cross-reactivity in MAT or older increased the risk of seropositivity (Harland et al., 2013; Grayzel because this test is good detecting serogroups, being a bias in the anal- and DeBess, 2016; Delaude et al., 2017). This relationship could be ysis. Moreover, it was described that Bratislava antibody titers often plausible as antibodies related to infection can persist for years and the Table 7 Number of titers (n) and relative prevalence for MAT cut-off 100 divided by geographic areas and serovar. Serovars tested: Bratislava (BRA), Icterohaemorrhagie (ICT), Australis (AUS), Pomona (POM), Grippotyphosa (GRI), Autumnalis (AUT), Canicola (CAN), and Saxkoebing (SAX). Region All Serovars BRA ICT AUS POM GRI AUT CAN SAX Mediterranean n 147 48 115 40 20 28 27 21 3 % 22.3 7.4 17.9 6.2 3.1 4.3 4.2 3.2 0.5 CI 19.1–25.6 5.4–9.5 14.9–20.8 4.3–8.1 1.7–4.4 2.7–5.9 2.6–5.7 1.9–4.6 0.0–1.0 Northwest n 8 3 4 2 2 2 2 1 0 % 22.2 8.3 11.1 5.6 5.6 5.6 5.6 2.8 - CI 7.9–36.4 0.0–17.8 0.3–21.9 0.0–13.4 0.0–13.4 0.0–13.4 0.0–13.4 0.0–8.4 - North n 44 13 25 12 7 10 8 1 6 % 38.0 11.5 22.1 10.6 6.2 8.8 7.1 0.8 5.3 CI 28.9–47.1 5.5–17.5 14.3–29.9 4.8–16.4 1.7–10.7 3.5–14.2 2.3–11.9 0.0–2.6 1.1–9.5 Center n 125 36 88 23 25 47 23 18 1 % 28.6 8.5 20.8 5.4 5.9 11.1 5.4 4.2 0.2 CI 24.3–33.0 5.8–11.2 17.0–24.7 3.3–7.6 3.3–7.6 8.1–14.1 3.3–7.6 2.3–6.2 0.0–0.7 South n 27 11 22 7 5 7 6 3 1 % 29.4 11.5 23.2 7.4 7.4 7.4 6.3 3.2 1.0 CI 20.1–38.8 5.0–18.1 14.5–31.8 2.0–12.7 2.0–12.7 2.0–12.7 1.3–11.3 0.0–6.7 0.0–3.1 5 M.C. L opez et al. Heliyon 5 (2019) e02373 probability of having contacted with Leptospira increases over time Competing interest statement (Delaude et al., 2017). The most important limitation of this study was the fact that vacci- The authors declare no conflict of interest. nation status of the dogs was unknown, limiting the interpretation of the results of MAT. While the MAT is far from being a perfect method, it is Additional information currently the only tool available to document exposure in dogs (Adler, 2015). Non-infected dogs vaccinated with bivalent or quadrivalent whole No additional information is available for this paper. cell anti leptospiral vaccines can have post-vaccinal titers of 1:6400 or higher to both vaccinal and non-vaccinal serovars (Barr et al., 2005; References Midence et al., 2012; Martin et al., 2014). Although the majority of vaccinated dogs have been shown to become antibody negative by week Adler, B., 2015. Leptospira and Leptospirosis. Springer-Verlag Berlin Heidelberg. 15 post-vaccination, vaccinal titers can persist for 12 months in a small Alonso-Andicoberry, C., García-Pe~ na, F.J., Ortega-Mora, L.M., 2001. Epidemiología, percentage of dogs (Martin et al., 2014). Despite its limitations the MAT diagnostico y control de la leptospirosis bovina (Revisi on). Investig. Agrar. Prod. Sanid. Anim. 16, 205–226. is well accepted and extensively used to derive information about the Ambily, R., Mini, M., Joseph, S., Krishna, S.V., Abhinay, G., 2013. Canine leptospirosis - a presence of exposition to Leptospira in sick (Boutilier et al., 2003; Ward seroprevalence study from Kerala, India. Vet. World 6, 42–44. et al., 2004; Ghneim et al., 2007; Hennebelle et al., 2013; Mayer-Scholl Andre-Fontaine, G., 2006. Canine leptospirosis-Do we have a problem? Vet. Microbiol. 117, 19–24. et al., 2013; Major et al., 2014) and healthy dogs (Davis et al., 2008; Arent, Z.J., Andrews, S., Adamama-Moraitou, K., Gilmore, C., Pardali, D., Ellis, W.A., Arent et al., 2013; Schuller et al., 2015a; Llewellyn et al., 2016). In the 2013. Emergence of novel Leptospira serovars: a need for adjusting vaccination absence of a consensus of what represents an ideal cut off titer for MAT policies for dogs? Epidemiol. Infect. 141, 1148–1153. Atxaerandio, R., Aduriz, G., Ziluaga, I., Esteban, J.I., Maranda, L., Mainar-Jaime, R.C., positivity, our study used two different cut offs (100 and 400) to 2005. Serological evidence of Leptospira interrogans serovar Bratislava infection and calculate seroprevalence in order to compare the results with previous its association with abortions in cattle in northern Spain. Vet. Rec. 156, 376–380. publications worldwide. Furthermore, MAT does not discriminate be- Azocar-Aedo, L., Monti, G., 2016. Meta-analyses of factors associated with leptospirosis in domestic dogs. Zoonoses Public Health 63, 328–336. tween vaccination titers and titers due to exposure, thus adding further Baranton, G., Postic, D., 2006. Trends in leptospirosis epidemiology in France. Sixty-six difficulty to the interpretation of canine tests, as a large proportion of years of passive serological surveillance from 1920 to 2003. Int. J. Infect. Dis. 10, dogs are vaccinated with either bivalent or multivalent Leptospira vac- 162–170. Barr, S., McDonough, P., Scipioni-Ball, R., Starr, J., 2005. Serologic responses of dogs cines (Sykes et al., 2011). Moreover, multivalent vaccines were available given a commercial vaccine against Leptospira interrogans serovar pomona and in Spain since the study started which made the interpretation of the Leptospira kirschneri serovar grippotyphosa. Am. J. Vet. Res. 66, 1780–1784. results even more challenging. In dogs with confirmed leptospirosis the Benito, M., Perez, J., Vega, S., Bernabeu, G., García, F., 2005. Comparison of the sensitivity of MAT has been estimated to be 50–67% at 1:400 dilution prevalence of the infection by leptospira spp, Leishmania infantum and Ehrlichia canis in dogs in the Comunidad Valenciana (Spain). Epidemio Sante Anim 45, 83–86. and 22–67% at 1:800 dilution, while the specificity ranges from 69–93% Bharti, A.R., Nally, J.E., Ricaldi, J.E., Matthias, M.A., Diaz, M.M., Lovett, M.A., at 1:400 dilution and 69–100% for 1:800 dilution (Miller et al., 2011). Levett, P.N., Gilman, R.H., Willig, M.R., Gotuzzo, E., Vinetz, J.M., 2003. However, the sensitivity and specificity of the MAT in detecting chronic Leptospirosis: a zoonotic disease of global importance. Lancet Infect. Dis. 3, 757–771. Bier, D., Shimakura, S.E., Morikawa, V.M., Ullmann, L.S., Kikuti, M., Langoni, H., subclinical infection or post exposure titers has not been reported Biondo, A.W., Molento, M.B., 2013. Analise espacial do risco de Leptospirose canina (Delaude et al., 2017). In the present study all statistical analyses were na Vila Pantanal, Curitiba, Parana. Pesqui. Vet. Bras. 33, 74–79. performed using both cut-off 100 and 400 in order to minimize the Boutilier, P., Carr, A., Schulman, R.L., 2003. Leptospirosis in dogs: a serologic survey and case series 1996 to 2001. Vet. Ther. 4, 387–396. important limitation of vaccine interference. Finally, multiple serocon- Brown, C., Roberts, A., Miller, M., Davis, D., Brown, S., Bolin, C., Jarecki-Black, J., versions to two or more serovars contained in the current vaccines were Greene, C., Miller-Liebl, D., 1996. Leptospira interrogans serovar grippotyphosa detected in many dogs in this study, probably also related to vaccination. infection in dogs. J. Am. Vet. Med. Assoc. 209, 1265–1267. Burriel, A.R., Dalley, C., Woodward, M.J., 2003. Prevalence of Leptospira species among In this case the MAT was also not able to identify the real etiological farmed and domestic animals in Greece. Vet. Rec. 153, 146–148. agent as has been previously described (Cerri et al., 2003; Tagliabue Cacciapuoti, B., Ciceroni, L., Pinto, A., Apollini, M., Rondinella, V., Bonomi, U., et al., 2016). Benedetti, E., Cinco, M., Dessì, S., Dettori, G., Grillo, R., Falomo, R., Mansueto, S., Miceli, D., Marcuccio, L., Marcuccio, C., Pizzocaro, P., Schivo, M.L., Varaldo, E., Lupidi, R., Ioli, A., Marzolini, A., Rosmini, F., 1994. Survey on the prevalence of 5. Conclusions leptospira infections in the Italian population. Eur. J. Epidemiol. 10, 173–180. Cerri, D., Ebani, V.V., Fratini, F., Pinzauti, P., Andreani, E., 2003. Epidemiology of This study shows that owned dogs in Spain are commonly exposed to leptospirosis: observations on serological data obtained by a ‘diagnostic laboratory for leptospirosis’ from 1995 to 2001. New Microbiol. 26, 383–389. Leptospira spp., with Icterohaemorrhagiae, Bratislava, Grippotyphosa and Davis, M.A., Evermann, J.F., Petersen, C.R., VanderSchalie, J., Besser, T.E., Huckabee, J., Australis being the most common serovars. Sex is associated with sero- Daniels, J.B., Hancock, D.D., Leslie, M., Baer, R., 2008. Serological survey for positivity and being older than 6 years old was a risk factor for sero- antibodies to Leptospira in dogs and raccoons in Washington State. Zoonoses Public Health 55, 436–442. positivity. Multiple positive serovars occurrence is very frequent and this Delaude, A., Rodriguez-Campos, S., Dreyfus, A., Counotte, M.J., Francey, T., produces difficulties to identify the real circulation of Leptospira in dogs. Schweighauser, A., Lettry, S., Schuller, S., 2017. Canine leptospirosis in Since dogs are considered as sentinel of human zoonosis, it makes sense Switzerland—a prospective cross-sectional study examining seroprevalence, risk factors and urinary shedding of pathogenic leptospires. Prev. Vet. Med. 141, 48–60. to consider a regular monitoring of them. Douglin, C., Jordan, C., Rock, R., Hurley, A., Levett, P.N., 1997. Risk factors for severe leptospirosis in the parish of st. Andrew, Barbados. Emerg. Infect. Dis. 3, 78–80. Declarations Ellis, W.A., 2010. Control of canine leptospirosis in Europe: time for a change? Vet. Rec. 167, 602–605. Estrada-Pe~ na, A., Roura, X., Sainz, A., Mir o, G., Solano-Gallego, L., 2017. Species of ticks Author contribution statement and carried pathogens in owned dogs in Spain: results of a one-year national survey. Ticks Tick. Borne. Dis. 8, 443–452. Geisen, V., Stengel, C., Brem, S., Müller, W., Greene, C., Hartmann, K., 2007. Canine M.C. Lopez: Analyzed and interpreted the data; Wrote the paper. leptospirosis infections - clinical signs and outcome with different suspected A. Vila, X. Roura: Conceived and designed the experiments; Analyzed Leptospira serogroups (42 cases). J. Small Anim. Pract. 48, 324–328. and interpreted the data; Wrote the paper. Ghneim, G.S., Viers, J.H., Chomel, B.B., Kass, P.H., Descollonges, D.A., Johnson, M.L., J. Rodon: Contributed reagents, materials, analysis tools or data. 2007. Use of a case-control study and geographic information systems to determine environmental and demographic risk factors for canine leptospirosis. Vet. Res. 38, 37–50. Funding statement Grayzel, S.E., DeBess, E.E., 2016. Characterization of leptospirosis among dogs in Oregon, 2007–2011. J. Am. Vet. Med. Assoc. 248, 908–915. Greenlee, J.J., Bolin, C.A., Alt, D.P., Cheville, N.F., Andreasen, C.B., 2004. Clinical and This research did not receive any specific grant from funding agencies pathologic comparison of acute leptospirosis in dogs caused by two strains of in the public, commercial, or not-for-profit sectors. Leptospira kirschneri serovar grippotyphosa. Am. J. Vet. Res. 65, 1100–1107. 6 M.C. L opez et al. Heliyon 5 (2019) e02373 Guiti an, F.J., García-Pe~na, F.J., Oliveira, J., Sanjuana, M.L., Yusa, E., 2001. Serological Ojha, K.C., Singh, D.K., Kaphle, K., Shah, Y., Pant, D.K., 2018. Sero-prevalence of study of the frequency of leptospiral infections among dairy cows in farms with leptospirosis and differentiation in blood parameters between positive and negative suboptimal reproductive efficiency in Galicia, Spain. Vet. Microbiol. 80, 275–284. cases in dogs of Kathmandu Valley. Trans. R. Soc. Trop. Med. Hyg. 00, 1–5. Habus, J., Persic, Z., Spicic, S., Vince, S., Stritof, Z., Milas, Z., Cvetnic, Z., Perharic, M., Renaud, C., Andrews, S., Djelouadji, Z., Lecheval, S., Corrao-Revol, N., Buff, S., Turk, N., 2017. New trends in human and animal leptospirosis in Croatia, Demont, P., Kodjo, A., 2013. Prevalence of the Leptospira serovars bratislava, 2009–2014. Acta Trop. 168, 1–8. grippotyphosa, mozdok and pomona in French dogs. Vet. J. 196, 126–127. Harland, A.L., Cave, N.J., Jones, B.R., Benschop, J., Donald, J.J., Midwinter, A.C., Scanziani, E., Calcaterra, S., Tagliabue, S., Luini, M., Giusti, A.M., Tomba, M., 1994. Squires, R.A., Collins-Emerson, J.M., 2013. A serological survey of leptospiral Serological findings in cases of acute leptospirosis in the dog. J. Small Anim. Pract. antibodies in dogs in New Zealand. N. Z. Vet. J. 61, 98–106. 35, 257–260. Hennebelle, J.H., Sykes, J.E., Carpenter, T.E., Foley, J., 2013. Spatial and temporal Scanziani, E., Origgi, F., Giusti, A.M., Iacchia, G., Vasino, A., Pirovano, G., Scarpas, P., patterns of Leptospira infection in dogs from northern California: 67 cases Tagliabue, S., 2002. Serological survey of leptospiral infection in kennelled dogs in (2001–2010). J. Am. Vet. Med. Assoc. 242, 941–947. Italy. J. Small Anim. Pract. 43, 154–157. Jansen, A., Sch€ oneberg, I., Frank, C., Alpers, K., Schneider, T., Stark, K., 2005. Schuller, S., Arent, Z.J., Gilmore, C., Nally, J., 2015a. Prevalence of antileptospiral serum Leptospirosis in Germany, 1962-2003. Emerg. Infect. Dis. 11, 1048–1054. antibodies in dogs in Ireland. Vet. Rec. 177, 126–128. Klaasen, H.L.B.M., Van Der Veen, M., Molkenboer, M.J.C.H., Sutton, D., 2013. A novel Schuller, S., Francey, T., Hartmann, K., Hugonnard, M., Kohn, B., Nally, J.E., Sykes, J., tetravalent Leptospira bacterin protects against infection and shedding following 2015b. European consensus statement on leptospirosis in dogs and cats. J. Small challenge in dogs. Vet. Rec. 172, 172–181. Anim. Pract. 56, 159–179. Krawczyk, M., 2005. Serological evidence of leptospirosis in animals in northern Poland. Shi, D., Liu, M., Guo, S., Liao, S., Sun, M., Liu, J., Wang, L., Wang, Z., Wang, S., Yang, D., Vet. Rec. 156, 88–89. Chai, T., 2012. Serological survey of canine leptospirosis in southern China. Pak. Vet. Llewellyn, J.R., Krupka-Dyachenko, I., Rettinger, A.L. en., Dyachenko, V., Stamm, I., J. 32, 280–282. Kopp, P.A., Straubinger, R.K., Hartmann, K., 2016. Urinary shedding of leptospires Sykes, J.E., Hartmann, K., Lunn, K.F., Moore, G.E., Stoddard, R.A., Goldstein, R.E., 2011. and presence of Leptospira antibodies in healthy dogs from Upper Bavaria. Berl. 2010 ACVIM small animal consensus statement on leptospirosis: diagnosis, Munch. Tierarztl. Wochenschr. epidemiology, treatment, and prevention. J. Vet. Intern. Med. 25, 1–13. Major, A., Schweighauser, A., Francey, T., 2014. Increasing incidence of canine Tagliabue, S., Figarolli, B.M., D’Incau, M., Foschi, G., Gennero, M.S., Giordani, R., leptospirosis in Switzerland. Int. J. Environ. Res. Public Health 11, 7242–7260. Natale, A., Papa, P., Ponti, N., Scaltrito, D., Spadari, L., Vesco, G., Ruocco, L., 2016. Martin, L.E.R., Wiggans, K.T., Wennogle, S.A., Curtis, K., Chandrashekar, R., Lappin, M.R., Serological surveillance of Leptospirosis in Italy: two-year national data (2010-2011). 2014. Vaccine-associated leptospira antibodies in client-owned dogs. J. Vet. Intern. Vet. Ital. 52, 129–138. Med. 28, 789–792. Trevejo, R.T., Rigau-Perez, J.G., Ashford, D.A., McClure, E.M., Jarquin-Gonzalez, C., Mastrorilli, C., Dondi, F., Agnoli, C., Turba, M.E., Vezzali, E., Gentilini, F., 2007. Amador, J.J., de los Reyes, J.O., Gonzalez, A., Zaki, S.R., Shieh, W.J., McLean, R.G., Clinicopathologic features and outcome predictors of Leptospira interrogans Australis Nasci, R.S., Weyant, R.S., Bolin, C.A., Bragg, S.L., Perkins, B.A., Spiegel, R.A., 1998. serogroup infection in dogs: a retrospective study of 20 cases (2001-2004). J. Vet. Epidemic leptospirosis associated with pulmonary hemorrhage-Nicaragua, 1995. Intern. Med. 21, 3–10. J. Infect. Dis. 178, 1457–1463. Mayer-Scholl, A., Luge, E., Draeger, A., N€ockler, K., Kohn, B., 2013. Distribution of leptospira VandenBroek, A.H., Thrusfield, M.V., Dobbie, G.R., Ellis, W.A., 1991. A serological and serogroups in dogs from Berlin, Germany. Vector Borne Zoonotic Dis. 13, 200–202. bacteriological survey of leptospiral infection in dogs in Edinburgh and Glasgow. Midence, J.N., Chandler, A.M., Goldstein, R.E., 2012. Assessing the effect of recent J. Small Anim. Pract. 32, 118–124. leptospira vaccination on whole blood quantitative Pcr testing in dogs. J. Vet. Intern. Ward, M.P., 2002. Seasonality of canine leptospirosis in the United States and Canada and Med. 26, 149–152. its association with rainfall. Prev. Vet. Med. 56, 203–213. Mill an, J., Candela, M.G., L opez-Bao, J.V., Pereira, M., Jimenez, M.A., Le on-Vizcaíno, L., Ward, M.P., Guptill, L.F., Prahl, A., Wu, C.C., 2004. Serovar-specific prevalence and risk 2009. Leptospirosis in wild and domestic carnivores in natural areas in Andalusia, factors for leptospirosis among dogs: 90 cases (1997-2002). J. Am. Vet. Med. Assoc. Spain. Vector Borne Zoonotic Dis. 9, 549–554. 224, 1958–1963. Mill an, J., Cevidanes, A., Chirife, A.D., Candela, M.G., Le on-Vizcaíno, L., 2018. Risk Zwijnenberg, R.J.G., Smythe, L.D., Symonds, M.L., Dohnt, M.F., Toribio, J.A., 2008. factors of Leptospira infection in Mediterranean periurban micromammals. Zoonoses Cross-sectional study of canine leptospirosis in animal shelter populations in Public Health 65, 1–7. mainland Australia. Aust. Vet. J. 86, 317–323. Miller, M.D., Annis, K.M., Lappin, M.R., Lunn, K.F., 2011. Variability in results of the microscopic agglutination test in dogs with clinical leptospirosis and dogs vaccinated against leptospirosis. J. Am. Vet. Med. Assoc. 25, 426–432. 7