Evaluation of changes in the epidemiology of leptospirosis in dogs after introduction of a quadrivalent antileptospiral vaccine in a highly endemic area

Abstract Background Since 2003, a marked increase in leptospirosis serogroup Australis has been observed in dogs in Switzerland. In 2013, a new quadrivalent antileptospiral vaccine (L4) was introduced, adding serogroups Australis and Grippotyphosa to Canicola and Icterohaemorrhagiae of the previous bivalent vaccines (L2). Objective To examine whether introduction of L4 was associated with decreased incidence of leptospirosis and decreased odds for dogs with acute kidney injury (AKI) to be diagnosed with leptospirosis. Animals Four hundred and sixty‐nine dogs with AKI presented to a referral hospital, including 269 dogs with leptospirosis and 200 controls with other causes. Methods Descriptive section: disease incidence was evaluated for 3 consecutive periods: before (PRE, 2011‐2012), transition (TRANS, 2013‐2014), and after introduction of L4 (POST, 2015‐2017). Analytical section: variables associated with a diagnosis of leptospirosis were investigated in a case‐control study using multivariable logistic regression, and focusing on vaccination. Results The number of dogs diagnosed with leptospirosis (AKI‐L) decreased from 56.5 (PRE) to 15.7 (POST) cases/year while controls increased from 16.5 to 38.0 cases/year. Control dogs (AKI‐nL) showed a decrease in L2 vaccination (100% to 26%) and an increase in L4 vaccination (0% to 70%). The odds ratio for vaccinated dogs to be diagnosed with leptospirosis was 0.11 (95% confidence interval [CI], 0.06‐0.22; P < .001) for L4 and 2.08 (0.58‐7.42; P = .26) for L2. Conclusions and Clinical Importance The introduction of L4 was associated with a marked decrease in dogs with leptospirosis and AKI in Switzerland. Use of the L4 vaccine was associated with significantly decreased odds of disease.

leptospirosis in dogs diagnosed at a veterinary teaching hospital, exceeding the incidence in other European countries. 4 The main clinical manifestations of this re-emerging epidemic of leptospirosis in dogs in Switzerland often were severe and included acute kidney injury (AKI, 99%), acute liver injury (26%), hemorrhagic tendencies (18%), and leptospiral pulmonary hemorrhage syndrome (77%).
Approximately 70% of dogs in this cohort showed serologic evidence of infection with serovars Bratislava and Australis, both belonging to serogroup Australis. 4 In 2013, a new quadrivalent antileptospiral vaccine (Nobivac Lepto 6, MSD Animal Health, also known as Nobivac L4 in other countries; L4) was introduced onto the Swiss market. 12 This killed whole cell vaccine includes serovars of serogroups Australis and Grippotyphosa in addition to serogroups Canicola and Icterohaemorrhagiae present in the previously available bivalent vaccines (L2). The vaccine had been shown to provide excellent protection against experimental challenge of dogs with heterologous strains from the same serogroups. 12 Because of effective marketing and continuous education efforts by specialists working in academia, the uptake of this vaccine in Switzerland appeared to be quick, providing the unique opportunity to study the change in epidemiology of leptospirosis in dogs associated with the change in prevention.
Therefore, the aims of our retrospective case-control study were to describe the changes in numbers of cases of AKI in dogs with causal evidence of leptospirosis (AKI-L) over a 7-year period (2011-2017) compared to a control group of similarly affected dogs with AKI not caused by leptospirosis (AKI-nL), and to investigate vaccination-related variables associated with a diagnosis of leptospirosis.

| Study design
The study was divided into 2 sections. The descriptive section was designed to report the incidences of AKI-L and AKI-nL in dogs presented to a veterinary teaching hospital. The incidences were evalu- ). An analytical section was designed as a retrospective case-control study to evaluate variables associated with a diagnosis of leptospirosis, focusing on antileptospiral vaccination status. 13 For the case-control study, a case was defined as a dog diagnosed with AKI caused by leptospirosis (AKI-L). A control was defined as a dog diagnosed with AKI of other causes (AKI-nL). This control population was chosen because it consists of dogs with similar clinical presentation and disease severity diagnosed during the same time. Acute kidney injury was defined by the combination of historical, clinical, laboratory, and imaging evidence, with at least 2 of the following criteria 14 : (a) presence of renal azotemia with a serum creatinine concentration ≥ 1.7 mg/dL persisting at least 24 hours after correction of prerenal factors; (b) increase in serum creatinine concentration ≥0.3 mg/dL during a 48-hour interval in the absence of prerenal factors; (c) persistent pathological oligo-anuria (<1 mL/kg/h over 6 hours) after volume repletion; (d) and evidence of tubular injury with renal glucosuria or granular casts on urinalysis. Dogs with evidence of underlying chronic kidney disease were not excluded from the study, as long as they fulfilled the criteria for AKI.

| Acute kidney injury not due to leptospirosis (AKI-nL)
Control dogs were diagnosed and confirmed with AKI-nL based on a known alternative diagnosis (2a) or a suspicion of a cause other than leptospirosis with negative paired MAT serology (2b). A tentative diagnosis of AKI-nL was considered for dogs with a suspicion of AKI-nL and 1 negative MAT serology (2c), and for dogs without evidence of leptospirosis in which no cause was positively identified (2d).   To minimize potential misclassification bias, both logistic regression analyses were repeated using a restricted case definition after excluding cases and controls diagnosed based on clinical evidence only (1e, 1f, 2d).

| Confirmatory testing
T A B L E 1 Number of cases, demographics, main disease characteristics, and basis for diagnosis of 469 dogs with AKI due to leptospirosis (AKI-L) or other causes (AKI-nL) AKI-L AKI-nL  The basis for the diagnosis of dogs as AKI-L and AKI-nL is described in Table 1

| Vaccine use and vaccination status
Data on vaccine use and vaccination status are summarized in Table 2, and changes over time are represented in Figures 2 and 3.
The proportion of the control AKI-nL dogs never vaccinated remained low throughout the study (0/18 PRE, 0/38 TRANS, and 4/89 POST).
The partial immunization status for L4 observed in 32 dogs was because of incorrect primary vaccination (n = 25), incorrect yearly booster schedule (n = 7), or a last vaccine administration >365 d before presentation for AKI (n = 14); >1 condition was not fulfilled in 10 dogs.

| Logistic regression analysis
Based on sample size analysis, the study population of 368 dogs with complete vaccination information was considered sufficient to address the study aims, including the evaluation of confounder variables. The following variables were evaluated as potential  (Tables 3 and 4). Sex and breed group were eliminated because they were not associated with both the independent and dependent variables. Neuter status was associated with the dependent variable and age with both the independent and dependent variables. They both were included in the multivariable models.
In the L4 multivariable analysis, vaccination status and age were strongly associated with diagnosis (

CONFLICT OF INTEREST DECLARATION
Financial support from MSD Animal Health was limited to facilitate collection of data and vaccination history by hiring support staff. Data analysis, interpretation and presentation was performed independently by the investigators. The final manuscript was submitted to MSD before submission for preapproval.

OFF-LABEL ANTIMICROBIAL DECLARATION
Authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION
Authors declare no IACUC or other approval was needed.

HUMAN ETHICS APPROVAL DECLARATION
Authors declare human ethics approval was not needed for this study.