Multidrug-Resistant Campylobacter jejuni Outbreak Linked to Puppy Exposure — United States, 2016–2018

Epidemiologic Investigation

Campylobacteriosis became a nationally notifiable condition in 2015, and many states routinely interview patients with campylobacteriosis.* For this investigation, a standardized, supplemental questionnaire was used by state and local health departments to collect dog exposure information from persons with Campylobacter infection who reported recent dog or pet store exposure during routine interview. A case definition relevant to this outbreak (Box) was developed to aid in case finding and characterization.

By February 28, 2018, a total of 118 persons meeting the case definition for Campylobacter infection, including 29 pet store employees, were reported from 18 states.^† Age was available for 115 persons and ranged from <1 year to 85 years (median = 26 years); 74 of 115 (63%) infected persons were female. Among 107 persons for whom hospitalization information was available, 26 (24%) were hospitalized; no deaths occurred. In total, 105 of 106 (99%) infected persons reported dog exposure, including 101 (95%) who had contact with a pet store puppy (Table). Eight patients reported buying or having contact with puppies from five pet store companies other than company A (companies B–F), indicating that puppies became infected with Campylobacter before reaching pet stores.

State and local health and agriculture departments in four states (Kentucky, Ohio, Pennsylvania, and Wisconsin) visited 20 pet stores and collected antibiotic administration records for 154 puppies. Among 149 puppies with available information, 142 (95%) received one or more antibiotic courses before arriving or while at the store. Among 142 puppies that received antibiotics, treatment indication was available for 134 (94%); 78 (55%) treated puppies received antibiotics for prophylaxis only, 54 (38%) for prophylaxis and treatment, and two (1%) for treatment only. Median antibiotic treatment duration was 15 days (range = 2–67 days). Four antibiotics (metronidazole, sulfadimethoxine, doxycycline, and azithromycin) accounted for 81% of all antibiotics administered (Figure). Use of broad-spectrum antibiotics also was noted, including tetracyclines, quinolones, aminoglycosides, and chloramphenicol.

Laboratory Investigation

Stool specimens from infected persons or Campylobacter isolates were submitted to state public health laboratories. Health and agriculture officials from six states (Florida, Kansas, Kentucky, Ohio, Pennsylvania, and Wisconsin) visited 29 pet stores (27 company A and two company B) to collect puppy fecal samples. All company A and B stores in Ohio were visited, and a convenience sample of stores in other states was selected. Some states also requested fecal samples from patient households that had purchased puppies. Human stool specimens and puppy fecal samples underwent Campylobacter culture, and whole-genome multilocus sequence typing (wgMLST) was performed to compare genetic relatedness. Antibiotic susceptibility testing for nine antibiotics was performed by broth microdilution (Sensititer, Thermo Fisher Scientific) on selected isolates and interpreted using epidemiologic cutoff values established by the European Committee on Antimicrobial Susceptibility Testing. In this report, “resistant” refers to isolates with non–wild-type results (2). To explore pet food as a possible source of Campylobacter infection in puppies, dog food samples from company A and one person’s home were collected for culture.

Campylobacter jejuni isolates were obtained for 51 persons and 23 puppies. Outbreak isolates from 45 persons and 11 puppies grouped into three distinct clades by wgMLST. Six persons whose illnesses did not meet the case definition because their isolates were unrelated by wgMLST were excluded. Twelve puppy isolates were also unrelated to the outbreak by wgMLST. Two clades contained isolates from persons and puppies that were genetically related (≤32 alleles difference within each clade). The third clade contained six patient isolates that were related (≤30 alleles difference). Eighteen outbreak isolates (10 human and eight puppy) representing all three clades were selected for antibiotic susceptibility testing, and all were resistant to azithromycin, ciprofloxacin, clindamycin, erythromycin, nalidixic acid, telithromycin, and tetracycline. In addition, 16 of 18 isolates were resistant to gentamicin, and four of 18 were resistant to florfenicol. None of the cultures of 12 dog food samples yielded Campylobacter.

Traceback Investigations

Records, including microchip identification numbers of puppies when available, were collected for puppies owned by infected persons and those sampled in stores. Microchips are implanted subcutaneously, usually before the puppy arrives at the store, and their corresponding identification numbers allowed investigators to trace puppies back to their breeders and distributors. Distributors are companies that purchase puppies wholesale from breeders and sell them to pet stores and other locations. Additional transport information was collected from stores when available. Practices identified during records review indicated that pet store puppies travel from breeders to distributors to stores by third-party transport companies. Information collected for eight puppies owned by infected persons and 20 puppies with fecal samples that were positive for Campylobacter jejuni traced back to 25 breeders and eight distributors. No single breeder, distributor, or transporter was identified as the infection source. However, potential for Campylobacter transmission among puppies exists because puppies from different breeders were commingled at distributors, during transport, and in stores.

Public Health Response

CDC developed educational materials on campylobacteriosis prevention. CDC and states shared these with pet industry partners, including retail pet stores. Educational messages focused on handwashing, separating human eating areas from animal areas, and using personal protective equipment correctly, such as wearing gloves when cleaning cages in pet stores. CDC posted an outbreak advisory online, which included information for clinicians and veterinarians recommending culture and antibiotic susceptibility testing to guide antibiotic treatment decisions (3).

Discussion

Epidemiologic, laboratory, and traceback evidence indicates that puppies sold through the commercial dog industry, an uncommon source of Campylobacter outbreaks, were the source of a multistate outbreak of multidrug-resistant Campylobacter infections. This evidence, combined with the prolonged nature of the outbreak and the potential for puppy commingling, indicates a potential for continued transmission of multidrug-resistant Campylobacter industrywide, including at breeders, distributors, transporters, and stores, and ultimately in customers’ homes. Although the investigation is completed, the risk for multidrug-resistant Campylobacter transmission to employees and consumers continues.

Dog-associated Campylobacter outbreaks have been reported previously, but those outbreaks involved fewer illnesses, and the isolates were not multidrug-resistant (4–6). The investigation of this outbreak revealed widespread administration of multiple antibiotic classes, including all classes to which the outbreak Campylobacter strains were resistant. Hygiene and animal husbandry practices can reduce the need for antibiotics and decrease transmission of Campylobacter between animals and from animals to humans (7). Adherence to antibiotic stewardship practices in these settings might reduce the selection of highly drug-resistant Campylobacter. Implementation of antibiotic stewardship principles and practices in the commercial dog industry is needed.

Clinicians should consider that persons can acquire Campylobacter infections, including multidrug-resistant infections, from puppies. If antibiotics are indicated, consider stool culture and antibiotic susceptibility testing. Pet stores, commercial distributors, transporters, and breeders should ensure that existing biosecurity measures are sufficient to reduce ongoing risk for Campylobacter transmission between puppies and humans. Pet stores should provide employee and customer education and training on handwashing and provide employees with personal protective equipment when cleaning animal areas (8). Educational information^§ that veterinarians and pet stores provide to pet owners could include information on reducing the risk for pathogen transmission. Finally, antibiotics should only be administered under veterinary supervision with a valid veterinary-client-patient relationship, consistent with existing stewardship principles.^¶