2.1 Outbreak identification and initial response

In April 2020, Italy recorded its first B. canis outbreak. Prior to this event, serological cases had been sporadically reported, but the bacterium had never been isolated. The outbreak occurred in a large breeding kennel in central Italy that housed mainly Chihuahuas along with other toy breeds such as Pomeranian Spitz, Maltese, and Toy Poodle [15].

Serological analyses were performed using a microplate serum agglutination test (mSAT), adapted from the tube agglutination method by Alton et al. [17] for use in 96-well U-shaped plates. B. canis strain RM66 was used to prepare the antigen. Serum samples were diluted 1:10 in Tris-maleate buffer (pH 9.0 ± 0.5), then serially diluted two-fold in the microplate with equal volumes of antigen and serum (50 μl each), resulting in final dilutions from 1:20–1:640. Plates were incubated at 37 ± 2 °C for 48 hours. Samples showing 100% agglutination at dilutions ≥1:20 were considered positive, with titers reported as the highest dilution showing complete agglutination [15].

Blood and urine samples were streaked onto selective Farrell’s Brucella Agar and inoculated into enrichment broth with equine serum. All cultures were incubated at 37 ± 1 °C. Broth cultures were subcultured weekly onto Farrell’s Brucella Agar for one month. Plates were checked daily for growth and considered negative if no colonies appeared in direct or subcultures. Suspected Brucella colonies were confirmed by species-specific PCR [15,18].

2.2 Implementation of the No-Kill strategy

Given the widespread public concern in Italy regarding pet welfare, euthanasia was not considered an option. In fact, it is prohibited by national law unless legally justified such as in cases of terminally illness or dangerous animals. Instead, a no-kill strategy was adopted, despite the significant logistical and financial challenges it posed, particularly due to the large number of animals involved and the substantial resources required for their care.

Following the confirmation of B. canis in the kennel, strict movement restrictions and a ban on sales were implemented to prevent further disease spread. These measures applied to both external transactions and internal transfers within the kennel. Additionally, male and female dogs were separated to prevent further breeding, reduce the risk of abortions, and minimize pathogen release.

2.3 Dog population control and testing protocol

To control the outbreak while managing the kennel population, a structured protocol combining neutering and castration and systematic testing was implemented. However, due to the frequent colonization of B. canis in the prostate of male dogs, which leads to bacterial shedding through urine, efforts to select negative animals eligible for rehoming were primarily focused on female dogs. Subsequently, the same protocol was applied to a restricted number of male castrated dogs that remained constantly negative since the beginning of the outbreak.

Female dogs underwent neutering, together with primary serological and bacteriological testing. Laboratory analyses were carried as previously described [15,19]. Concerning serological tests, seronegative status was determined using mSAT, with a cut-off titre of 1:20; dogs with titres below this threshold were classified as seronegative. The diagnostic performance of the mSAT, as validated by Perletta et al. [19], indicates an estimated sensitivity of 96.7% (88.8–98.7) and a specificity of 92.3% (86.7–95.1).

Simultaneously, all male dogs underwent castration and animals testing negative to serological and bacteriological screening carried out during outbreak investigations were considered for rehoming and followed the same protocol of female dogs with an additional bacteriological control on urine samples. Animals testing negative to serological and bacteriological investigations were classified as negative and eligible for adoption.

2.4 Seronegative animals selected for potential rehoming

Firstly, seronegative dogs were transferred to a separate facility under quarantine conditions. Then, sequential testing in quarantine has been done, as follows: both serological and bacteriological tests were performed at day 10, in a first round of quarantine; at day 28 only serological testing was conducted and an additional bacteriological testing was carried out on urine samples only for male dogs (second round); in the final evaluation round (week 8) dogs that remained negative throughout all testing phases were cleared for adoption. Furthermore, dogs testing positive at any stage were returned to the original outbreak kennel for further management.

The testing intervals (days 10, 28, and 8 weeks) were selected based on (i) the known kinetics of B. canis infection, including the typical window for seroconversion; (ii) the intermittent nature of bacteraemia and urinary shedding; and (iii) the need to balance diagnostic accuracy with feasible management of a large canine population. The day 10 assessment allowed early detection of dogs that might have been incubating the infection earlier; the day 28 testing corresponded to the expected timeframe for seroconversion in animals exposed shortly before the start of quarantine; and the final evaluation at 8 weeks provided a sufficient temporal buffer to identify delayed serological or bacteriological positives before rehoming for infections acquired around the 28^th day of testing (but still silent at the 28^th day’s sampling). These intervals were therefore chosen to minimize the risk of releasing dogs with early, subclinical, or intermittently detectable infections.

A small proportion of dogs tested positive during the first and second testing rounds and were consequently removed from the rehoming pathway and returned to the outbreak facility for continued management. No dogs that remained negative through all testing rounds converted at later timepoints.

The outbreak kennel remained under official movement restrictions. These dogs continued to be housed at the facility under veterinary supervision, following national guidelines for infectious disease containment. Due to the absence of a standardized treatment protocol and the high relapse rate associated with antibiotic therapy, no further rehoming of positive animals was permitted. Their long-term management remained under the control of the local veterinary authority.

Dogs were kept in a dedicated, physically separated facility under strict biosecurity measures. Housing adhered to national animal-welfare standards, ensuring adequate space, environmental enrichment, climate control, and daily monitoring by veterinary staff. Additionally, caretakers used individualized equipment and PPE to reduce cross-contamination risks. These details are now included to provide assurance that the quarantine procedures safeguarded both animal well-being and biosafety.

Fig 1 summarizes the flow of our rehoming testing strategy.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Rehoming testing strategy flow chart.

https://doi.org/10.1371/journal.pone.0351742.g001

2.5 Post-adoption monitoring and owner responsibilities

To ensure long-term health monitoring, all adopted dogs were subjected to post-adoption testing at defined intervals:

• Three months post-adoption
• Six months post-adoption
• Twelve months post-adoption
• Annually thereafter (recommended for life)

Adoptive owners were required to sign a consent and awareness form outlining the risks associated with adopting a dog from an outbreak setting. This document provided a comprehensive guidance on preventive measures, hygiene protocols, and importance of continued testing. Owners were also responsible for covering the costs of post-adoption testing, reinforcing their critical role in maintaining effective disease control.

2.6 Ethical, legal, and animal welfare considerations

All procedures involving animals were conducted in accordance with national and institutional guidelines for the care and use of laboratory animals. This study was carried out under the supervision of the official veterinary services and adhered to Italian legislation prohibiting euthanasia unless medically or legally justified. Informed consent was obtained from all adoptive owners prior to inclusion in the post-adoption monitoring program.

4.1 Validation of the No-Kill strategy

To the best of our knowledge, this is the first large-scale attempt at managing a B. canis outbreak in a breeding kennel through a no-kill strategy. The absence of infection in rehomed dogs supports the effectiveness of this approach. This aligns with findings from smaller-scale studies, which suggest that a combination of castration and neutering and rigorous screening can mitigate transmission risks [20]. In some cases, low seropositivity titers, close to test cut-off values (1:20 dilution for mSAT) were detected during initial post-adoption screenings. However, these cases were not confirmed in following testing, and all affected dogs consistently tested negative in bacterial isolation, suggesting that the observed low titers were likely nonspecific or transient responses rather than active infections.

4.2 Comparison with international outbreak responses

Unlike outbreaks in other countries where euthanasia has been a primary response, our strategy demonstrates that rehoming non-infected animals can be a viable alternative. [21]. The recommended UK approach for confirmed infected animals is euthanasia, however for test-positive clinically normal dogs antimicrobial therapy and neutering could be a solution, combined with a limitation in contact with other dogs and further hygiene measures for the rest of the dog’s life. [22]. In Netherlands euthanasia is the only approach for seropositive dogs and dogs they were in close contact with, while a periodically testing (6 months) is chosen for the seronegative remaining dogs [23]. The Canadian generally accepted recommendation is that truly infected dogs should be euthanized due to the risk to canine and human populations. However, when euthanasia is not possible, due to client opinion, then isolation can be considered after appropriate client education and medical record documentation [24]. In the U.S., delays in veterinary responses to multi-state outbreaks highlighted the need for standardized control measures [21]. European outbreaks similarly lacked coordinated response protocols, often relying on ad hoc interventions [25].

Our findings suggest that structured rehoming programs can serve as a humane and effective disease management tool, and demonstrate the effectiveness of the structured rehoming and monitoring protocol, ensuring that no cases of late-onset B. canis infection have emerged to date. The results support the viability of controlled adoption as a strategy for managing B. canis outbreaks while adhering to ethical and legal constraints.

While our approach demonstrated success, it remains contingent on rigorous follow-up testing and owner compliance. In scenarios where resources or owner adherence are insufficient, alternative containment strategies may need to be explored. For example, future studies should investigate the role of long-term antibody monitoring and the feasibility of a vaccination program. Additionally, genomic sequencing of B. canis strains involved in outbreaks could provide critical epidemiological insights and improve tracking of disease spread across borders [26].

4.3 Challenges in surveillance and control

The lack of a globally standardized protocol for B. canis control creates persistent challenges in disease management and prevention. In Europe, dogs are frequently imported without mandatory pre- or post-arrival testing, making surveillance difficult [10]. Even within Italy, fragmented canine registry system further complicates trace-forward investigations during outbreaks. Establishing mandatory identification and health certification protocols for imported dogs could significantly enhance disease tracking and prevention.

The reluctance of some pet owners to comply with follow-up testing requirements further complicates monitoring efforts. In some cases, the absence of legal enforcement mechanisms prevents veterinary authorities from compelling compliance. Future policies should address these gaps by integrating owner responsibility clauses and government-funded testing incentives for high-risk animals [25].

4.4 Policy and one health implications

Given the zoonotic potential of B. canis, a One Health approach is crucial for controlling its spread. Recommendations include:

• Mandatory pre-import screening for dogs entering non-endemic regions.
• Standardized testing protocols for breeders and veterinary clinics.
• Owner education initiatives to ensure proper monitoring and reporting.
• Further research into antimicrobial treatments and vaccine development [27].

The ethical considerations surrounding euthanasia versus treatment should also be addressed. While antimicrobial therapy remains inconsistent, euthanasia faces resistance from both the public and animal welfare organizations. Future guidelines must balance disease control with ethical concerns [26].

4.5 Study limitations

Despite the structured nature of the rehoming and monitoring protocol, the approach is not without limitations. Brucella canis, often referred to as “the great imposter,” poses a significant diagnostic challenge due to its potential for asymptomatic carriage and intermittent bacterial shedding. This, combined with imperfect test sensitivity and specificity, necessitates cautious interpretation of negative results.

A critical limitation encountered in our study was the approximately 20% loss to follow-up among rehomed dogs. Although adopters were informed of the need for continued testing and signed consent forms acknowledging their responsibilities, enforcement of long-term compliance remains a challenge in the absence of binding legal mechanisms. This gap introduces a residual risk of undetected transmission in the community.

Dogs that lost to follow-up may not be representative of those who completed all scheduled post-adoption tests. These animals could differ in owner compliance, access to veterinary care, or underlying health conditions, introducing the possibility that undetected late infections occurred outside the monitoring framework. Although all rehomed dogs underwent a rigorous multi-step testing protocol before adoption, this attrition still represents a source of uncertainty that may modestly affect the estimated reliability of the rehoming strategy.

Because B. canis is characterized by intermittent shedding and variable antibody responses, negative results, even when repeated, cannot absolutely exclude infection. We highlight that this limitation necessitates cautious interpretation of our findings and underscores the importance of continued long-term monitoring and the development of more sensitive diagnostic methods.

Et al.To mitigate such risks, only dogs that remained seronegative throughout multiple testing phases were deemed eligible for adoption. Furthermore, we recommend that future implementations of this strategy be accompanied by policy enhancements such as mandatory post-adoption testing requirements, government-funded diagnostic programs, and integration of health status monitoring via national microchip registries.