A 4-year-old male, mixed breed, disabled dog who had spent 21 months in a dog shelter in Thailand was adopted into a shelter in the United States. The dog arrived at John F. Kennedy International Airport in New York State near midnight on 19 April 2016, on a commercial airline flight, travelling in a crate in the cargo hold. He then travelled by car, confined to a crate, approximately 200 miles, to his new home at a shelter/sanctuary for chronically ill or disabled animals, arriving approximately 0300 on 20 April 2016. The shelter is in Broome County, NY, about 50 miles from Cornell University Hospital for Animals (CUHA) in Tompkins County.

He presented to CUHA on 21 April 2016 for evaluation of spinal trauma and inability to walk of approximately 2 years duration. He had been unable to ambulate since his arrival at the animal shelter in Thailand but had reportedly regained some function while there and was able to walk with assistance. He was incontinent of urine. He was up to date on his rabies vaccination. He was eating and drinking normally. Further medical history was unavailable.

On examination, the patient was bright, alert and responsive. He was briefly able to bear his own weight and ambulate once assisted into a standing position. There was marked muscle atrophy of both hindlimbs. The skin of the inguinal area was abnormally thickened, damp and malodorous, consistent with infection caused by urine scald related to incontinence.

Radiographs of the spine revealed damage to the cranial end-plate of the fourth lumbar vertebra. Absent lumbar puncture or magnetic resonance imaging (MRI), spinal infection could not be ruled out, but the radiographic changes appeared chronic and were more suggestive of trauma than of infection. Urinalysis was benign except for 5–20 red blood cells per high-power field, likely related to trauma during the cystocentesis by which the sample was acquired. The ptient was discharged to the care of his owners with appropriate nursing care and rehabilitation instructions. For the inguinal pyoderma, he was placed on oral cephalexin and a topical chlorhexidine spray. He returned a few days later for fitting of a hindlimb cart; urine cultures were pending at that time.

The bacteriology section at the Animal Health Diagnostic Center at Cornell University performed an aerobic culture on the urine sample following its standard operating procedures and using their automated identification system, Matrix-Assisted Laser Desorption Ionization Time-of-Flight (MALDI-TOF) mass spectrometry (Singhal, Kumar, Kanaujia, & Virdi, 2015) for identification of the bacterium. The fingerprint output revealed a close match with Burkholderia, precipitating additional biosafety precautions and reference to the New York State Biodefense Laboratory (NYSBDL) for definitive identification. All disposables and wastes generated from the case were secured pending results. After confirmation of Burkholderia pseudomallei at NYSBDL, all disposables and wastes associated with the case were destroyed by autoclaving with appropriate documentation. Cleaning and disinfection of hospital areas routinely used either accelerated hydrogen peroxide or an ammonium chloride-based neutral disinfectant; all areas the dog contacted were disinfected again using a 10% bleach solution.

Burkholderia pseudomallei, a saprophyte and the causative organism of melioidosis, is found most commonly in soil and surface water in the tropics, especially South-East Asia and northern Australia. With increasing awareness and diagnostic capacity over the past two decades, melioidosis is being recognized more frequently elsewhere, including Taiwan, China, Brazil, Central America, India and the Caribbean (Benoit et al., 2015; Currie, Dance, & Cheng, 2008; Doker et al., 2015; Vandana et al., 2016). One model estimates 68,000–412,000 human cases annually worldwide, arising from an estimated 79 different countries (Limmathurotsakul et al., 2016).

In the United States, most human cases occur in travellers who have visited endemic areas or in immigrants from endemic countries; however, two cases have been reported in individuals without a history of travel outside of the United States (Doker et al., 2014; Stewart et al., 2011).

Melioidosis is acquired by inhalation, ingestion or transdermal inoculation from environmental sources such as soil or standing water. There are limited case reports of person to person and perinatal transmission (Abbink & Orendi, 2001; Fang, Chen, Zhu, & Mao, 2016; McCormick et al., 1975). Individuals with diabetes, thalassaemia, alcoholism and renal impairment are more susceptible to melioidosis (Cheng & Currie, 2005). In addition to environmental sources, there have been two outbreaks associated with contaminated potable water supplies (Currie et al., 2001; Inglis et al., 1999).

In humans, melioidosis can present in many forms (Wiersinga, Currie, & Peacock, 2012), including:

Bacteremia and fulminant sepsis

Some forms of melioidosis, particularly the cutaneous form, can become chronic and relapsing.

Burkholderia pseudomallei can also cause infection and disease in a number of animals (Sprague & Neubauer, 2004). Goats and sheep are particularly susceptible, with several epizootics reported in the literature (Choy, Mayo, Janmaat, & Currie, 2000; Sutmoller, Kraneveld, & Van Der Schaaf, 1957). Cattle, swine, dogs and cats are considered to have low susceptibility (Sprague & Neubauer, 2004). Cases of melioidosis have been seen in iguanas and non-human primates (Johnson et al., 2013; Zehnder et al., 2014). Imported canine culture-positive melioidosis has not been described in the United States. Several such cases, some fatal, were reported among military working dogs in the Republic of Vietnam (Moe, Stedham, & Jennings, 1972; Stedham, 1971), but it appears that the animals remained in country and thus did not present an importation risk to a non-endemic area. Some military working dogs returning to the United States demonstrated serologic evidence of subclinical infection (Alexander et al., 1972).

Personnel from the Broome County Health Department visited the animal shelter on 4 May 2016 and interviewed the owner/operator, who lives on the premises.

The facility is essentially a small farm. At the time of the site visit, it housed approximately 135 animals, which included dogs, cats, horses, livestock and fowl. The grounds were organized and well-kept, and the house and outbuildings were well-maintained.

It was customary at the facility for newly arriving dogs to be housed for a period of days in a private downstairs room in the house, which is where he remained at the time of the visit to the facility. The dog looked healthy and behaved normally, except for paralysis of his hindlimbs. He was wearing a diaper. His antibiotic had been changed from cephalexin to enrofloxacin after the preliminary urine culture report from CUHA indicated Gram-negative rods resistant to the former (MIC = 256 μg/ml) but sensitive to the latter (MIC = 0.25 μg/ml).

Except for the two trips to the hospital detailed above, he had remained in the private room. He had voided in the crate during the car ride from the airport, and this had been cleaned and disinfected with bleach upon arrival home. Another partially paralysed dog from the shelter had accompanied him on the initial visit to CUHA. Each had been in their own crate in the back seat of the car for the trip.

The owner/operator had already taken several actions upon receipt of the urine culture results: continuing confinement to the single room, limiting access to a single caregiver, the use of latex gloves for direct contact, double diapering, an environmental cleaning regimen with a bleach solution and placement of a bleach shuffle pit at the entrance to the dog’s room.

In its endemic range, B. pseudomallei is acquired, by humans and animals, from soil and surface water, often in an agricultural setting. Although several cases of suspected animal-to-human transmission have been reported in the literature, none have been confirmed through genotyping, and evidence for zoonotic transmission is purely circumstantial (Fhogartaigh & Dance, 2015). Guidelines for the management of human exposures exist only within the context of laboratory settings (Peacock et al., 2008). The present case, however, included circumstances that complicated our interpretation and application of existing knowledge and exposure guidelines: B. pseudomallei culture-positive urine in an incontinent and paraplegic companion animal needing a high level of direct involvement by his human caregivers, all in the setting of a non-endemic area.

In the absence of any treatment regimen with demonstrated efficacy in eradicating B. pseudomallei from the canine urinary tract, and thus with an ongoing source of potential human exposure to a rare and serious infection, all clinicians and public health agencies involved recommended euthanasia, and the dog’s owner agreed. This was accomplished at CUHA on 6 May 2016. In the light of stringent federal regulations regarding select agents, and the infrastructure necessary to do so safely, an autopsy was not performed. Disposal of the body was by alkaline hydrolysis.

All six monitored contacts from the veterinary hospital remained well throughout their monitoring periods. No reports of illness among the other 10 contacts were received. The individual undergoing antibiotic post-exposure prophylaxis completed the course without incident. One individual’s sequential serology result doubled at week 4, rising from 1:20 to 1:40; this was considered indeterminate and non-actionable. All other serological results were negative.

By the end of their monitoring period, all three contacts at the animal shelter had remained well. Baseline serologic tests all returned negative.

Infection with B. pseudomallei is very rare in humans and animals in the United States, but it is endemic in many parts of South-East Asia, where an often-brutal dog meat trade motivates many dog rescue organizations to facilitate adoptions into the United States. This may result in importation of this rare select agent into almost any community, with implications for medical, veterinary and public health personnel.

It may be challenging for clinicians outside of endemic areas to recognize this unfamiliar illness; astute enquiry about travel history, for humans and for animals, can facilitate diagnosis in both. Vigilance in the microbiology laboratory, with timely referral of suspect isolates to reference facilities, is also advised.

In the present case, urinary B. pseudomallei infection combined with incontinence, neither of which was likely to resolve, presented some unique concerns over minimal but ongoing exposure of the dog’s human caretakers. Prospects for a good quality of life for the dog were also poor if he was faced with perpetual indoor confinement and separation from the many other animals on the premises. In the absence of a promising and practical treatment regimen, humane euthanasia was recommended. When such a recommendation is necessary, the animal and its owner deserve empathy and emotional support from clinicians and public health personnel. An empathetic approach also helps maintain important working relationships between local health departments and animal owners, shelters and boarding facilities.

Published guidelines regarding inadvertent exposure to B. pseudomallei pertain specifically to laboratory workers (Peacock et al., 2008). Clinical judgement is necessary when attempting to apply that guidance in non-laboratory settings (in assessment of the clinic and the household in this case). Expert consultation may be needed.

The question of environmental contamination also arises. Winters in upstate New York are probably not conducive to survival of the organism, but persistent soil contamination and associated cases have occurred in temperate climates, as described by Mollaret (Mollaret, 1988). Environmental conditions could con-ceivably be more favourable to B. pseudomallei in the southeastern United States (Limmathurotsakul et al., 2016; Portacci, Rooney, & Dobos, 2017).

Currently, animals imported to the United States are not screened for melioidosis. However, this dog had no signs referable to B. pseudomallei and would likely have escaped notice in any practicable import inspection regime. Thus, vigilance among veterinarians caring for animals from South-East Asia would be wise. Further research into the volume of international dog adoptions, and into the prevalence of B. pseudomallei infection in these animals, would be valuable.