HEALTH SURVEY OF FREE-RANGING RACCOONS (PROCYON LOTOR) IN CENTRAL PARK, NEW YORK, NEW YORK, USA: IMPLICATIONS FOR HUMAN AND DOMESTIC ANIMAL HEALTH

Raccoons (Procyon lotor) occur in both rural and urban environments and have been used as sentinels to monitor environmental health, such as heavy metal contamination (Burger et al. 2002), and disease transmission risks to humans and domestic animals (Burger et al. 2002; Bischof and Rogers 2005; Rosatte et al. 2010). Raccoons have the potential to contaminate the environment with parasites (e.g., Baylisascaris procyonis; Yeitz et al. 2009) and serve as reservoirs for infectious disease agents such as Salmonella (Jardine et al. 2011), other enteric bacterial pathogens (Bigler et al. 1975), Toxoplasma gondii (Dubey et al. 2007), canine distemper virus (CDV), canine adenovirus-1 (CAV-1), parvoviruses (Junge et al. 2007), and the rabies virus (Jones et al. 2003). Raccoons are an ideal indicator species because they are ubiquitous habitat generalists, are relatively high in the food chain, and have relatively small ranges (Burger et al. 2002).

Central Park, in New York City (NYC), US, consists of 341 ha of fields, wooded areas, brush, and ponds and offers ideal habitat for raccoons in the center of a densely populated urban environment. Although raccoons have been reported to travel up to 45 km, most travel less than 7 km (Rosatte et al. 2010), and urban raccoons have significantly smaller home ranges than rural raccoons (Prange et al. 2004; Bozek et al. 2007). Therefore, the raccoon population in Central Park (40°46′52′′N, 73°58′2′′W) is likely isolated because Central Park is 4×0.8 km and the island of Manhattan is 21.6 km×3.7 km. Raccoons have been seen throughout Central Park, including on the grounds of Central Park Zoo. A recent survey estimated 8–9 million people visit Central Park annually, with 4.3 million visits by people walking dogs (Central Park Conservancy 2011). With an estimated park population of 500 raccoons, there is significant opportunity for interactions among raccoons, humans, and domestic animals (Slavinski et al. 2012), as well as zoologic or domestic species in Central Park's 2.6-ha zoo.

A raccoon rabies epizootic occurred in Central Park beginning in late August 2009, with 133 raccoons confirmed with rabies between December 2009 and December 2011 (Slavinski et al. 2012). Raccoon rabies was first detected in NYC in 1992, with rabies surveillance beginning that year. Before the 2009 epizootic, only one rabid raccoon had been reported from Central Park (in 1999). A multiagency task force, including the US Department of Agriculture Wildlife Services (USDAWS) and the NYC Department of Health and Mental Hygiene, was assembled during this epizootic to implement a raccoon trap-vaccinate-release (TVR) program starting in February 2010 to reduce exposure of humans and domestic animals to rabies (Slavinski et al. 2012). During this effort, we conducted health assessment and disease screening on a subset of raccoons in the TVR program to evaluate the general health of this population, to determine whether these raccoons harbor infectious disease agents to which humans and domestic animals visiting Central Park might be exposed, and to identify any clinical or subclinical diseases associated with environmental contaminants (e.g., lead exposure). Our findings could be used to assess the potential risk for zoonotic or interspecies disease transmission and for human and domestic animal exposure to environmental contaminants in Central Park.

Between 17–19 February 2010 (Feb2010); 20–21 September 2010 (Sep2010); and 28–29 November 2011 (Nov2011), Tomahawk model 108 wire box traps (Tomahawk Live Trap, Hazelhurst, Wisconsin, USA) baited with marshmallows and anise oil were set nightly by USDAWS (Slavinski et al. 2012). Latitude and longitude coordinates were recorded for each trap. The USDAWS technicians transported trapped raccoons to a nonpublic location in Central Park (median coordinates 40°47′10′′N, 73°57′43′′W) each morning for evaluation. The USDAWS technicians anesthetized each raccoon in the trap by intramuscular injection of a 5:1 premixed combination of ketamine (Ketaset 100 mg/mL, Fort Dodge Animal Health, Fort Dodge, Iowa, USA) and xylazine (AnaSed 100 mg/mL, Lloyd Laboratories, Shenandoah, Iowa, USA). The volume of the premixed combination given to cause adequate sedation was 0.5–2 mL, yielding a median ketamine dose of 12.17 mg/kg (range 5.67–34.72 mg/kg) and a median xylazine dose of 2.43 mg/kg (range 1.13–6.94 mg/kg). Once anesthetized, we performed a physical examination with recording of abnormal findings; relative age (juvenile vs. adult) based on tooth eruption and wear; sex; weight; rectal temperature; heart rate; and respiratory rate. The USDAWS technicians administered the rabies vaccine Imrab^® (Merial Limited, Duluth, Georgia, USA) to all raccoons via intramuscular injection. We placed Monel ear tags (National Band and Tag, Newport, Kentucky, USA) and subcutaneous transponders (Avid Identification Systems, Norco, California, USA) for permanent identification.

We euthanized anesthetized raccoons with significant wounds and those that scratched people during handling with 2 mL of a sodium pentobarbital solution by intracardiac injection. Pathologists performed gross necropsy and histologic examination on a subset of raccoons. Necropsies were not performed on all euthanized raccoons due to the zoonotic disease risk of performing a necropsy on a potentially rabid animal. Tissues collected for histologic review included heart, trachea, lung, esophagus, stomach, small intestine, large intestine, pancreas, liver, gall bladder, kidney, urinary bladder, spleen lymph node, bone marrow, thyroid gland, parathyroid gland, adrenal gland, testis, ovary, uterus, mammary gland, skin, diaphragm, skeletal muscle, spinal cord, and peripheral nerve. Pathologists fixed tissues in 10% neutral buffered formalin, and then they were processed using routine methods, sectioned (5 μm), stained with H&E, coverslipped, and reviewed by board-certified veterinary pathologists. We submitted brain tissue from all euthanized raccoons for rabies antigen detection via direct fluorescent antibody (DFA) testing (Brown et al. 2011) using standard methodologies and from necropsied raccoons for CDV antigen detection via DFA testing with same methods as for rabies antigen testing using a commercially available conjugate (Canine Distemper Virus Direct FA Conjugate, Veterinary Medical Research and Development, Pullman, Washington, USA) using standard methodologies at the Rabies Laboratory at the Wadsworth Center (RLWC, Albany, New York, USA).

We collected a representative sample of ectoparasites, placed them in 70% isopropyl alcohol, and submitted them to Cornell University Animal Health Diagnostic Center (CUAHDC, Ithaca, New York, USA) for identification. We collected blood samples via jugular venipuncture and placed the blood in serum separator tubes and/or tubes containing ethylenediaminetetraacetic acid or heparin. Veterinary technicians made smears from whole blood immediately after sample collection. We stored whole blood and feces on ice until processing 4–6 h later. Rectal swabs collected for enteric culture were placed in ParaPak C&S nonnutritive stool transport media (Meridian Bioscience, Cincinnati, Ohio, USA) and held no more than 3 d until shipment to a reference laboratory.

Veterinary technicians processed samples at the Bronx Zoo's Wildlife Health Center (Bronx, New York, USA) the day of collection. Veterinary technicians performed complete blood counts (CBCs), packed cell volume, total solids, and fecal parasite examinations on samples from each animal by using standard techniques. All CBCs were performed using the Hemavet hematology analyzer (Drew Scientific, Waterbury, Connecticut, USA) with manual differential cell counts. Whole blood lead analysis was performed with the Leadcare^® Blood Lead Test Kit (ESA, Chelmsford, Massachusetts, USA).

Veterinary technicians evaluated feces via direct microscopy of wet preparation, flotation in a sodium nitrate solution (Fecasol^®, Vétoquinol USA, Fort Worth, Texas, USA), and sedimentation with a 2% glycerin solution in water. Parasites were classified as coccidia or nematodes. Nematodes were further classified as ascarid-type ova, strongyle-type ova, ova with bipolar plugs, and larvae. We submitted feces from one raccoon to CUAHDC for parasite identification. We submitted rectal swabs to CUAHDC for aerobic enteric, Campylobacter jejuni, and Salmonella cultures. All Salmonella isolates were submitted to the National Veterinary Services Laboratory (Ames, Iowa, USA) for serotyping. We submitted serum for biochemical analysis to a commercial laboratory (Antech Diagnostics, New Hyde Park, New York, USA).

Rabies serology was conducted at RLWC by using an in vitro virus neutralization (VN) test (Trimarchi et al. 1996). The CDV serologies using VN (Epstein et al. 2013) were done at CUAHDC. We considered rabies virus neutralizing antibody (VNA) titers <0.125 IU/mL as negative, 0.125–0.25 IU/mL as low positive, and ≥0.5 IU/mL as positive. We defined individual CDV titers <4 as negative, 4–12 as inconclusive, and >12 as positive.

The remaining sera and heparinized plasma samples were frozen for 6–27 mo at −80 C until all samples from the three sampling times were collected. Serologic testing for additional diseases was then conducted on subsets of randomly chosen adult raccoons. Due to financial and logistical considerations, it was not possible to perform all serologic tests on all raccoons. However, we selected adults over juveniles due to the expectation that results from adults would be more representative of the population exposure status to these disease agents. Serum samples were evaluated for antibodies against CAV-1 by using VN and canine parvovirus-2 (CPV-2) by using hemagglutination inhibition (HI) at CUAHDC (Carmichael et al. 1980) as well as T. gondii by using an immunoglobulin G (IgG)–specific Modified Agglutination Test (Dubey and Desmonts 1987) at the University of Tennessee Veterinary Diagnostic Laboratory (Knoxville, Tennessee, USA). We defined T. gondii IgG titers ≥32 as positive and <32 as negative; CAV-1 titers <4 as negative, 4–12 as inconclusive, and >12 as positive; and CPV-2 titers ≥20 as positive and <20 as negative.

Serum samples were also evaluated for antibodies against Borrelia burgdorferi and Rickettsia rickettsii at the Michigan State University Diagnostic Center for Population and Animal Health (Lansing, Michigan, USA) by using standard methods for indirect antibody staining. Serial twofold dilutions of test serum were applied to antigen-spotted wells on a glass slide and were treated with fluorescein conjugated anti-IgG. We defined Borrelia burgdorferi titers <160 as negative, 160–640 as inconclusive, and >640 as positive and R. rickettsii titers <40 as negative, 40 as inconclusive, and ≥80 as positive. Serum samples from these raccoons were also submitted for another study of raccoon polyomavirus (RacPyV; Church et al. 2016).

We compiled and analyzed data in a computerized spreadsheet (Microsoft Office Excel 2010, Redmond, Washington, USA). All data were evaluated for normality using the Shapiro-Wilk W test (Royston 1982), with significance set at P<0.05. The median latitude and longitude of trap sites were determined because these data were not normally distributed. Medians, minimums, and maximums were calculated for body weights, ketamine and xylazine doses, rectal temperatures, heart rates, respiratory rates, CBC data, blood lead levels, and biochemistry data. Means and SDs were also calculated for packed cell volume, mean corpuscular volumes, hemoglobin, total solids, total protein, albumin, globulin, albumin:globulin, phosphorus, calcium, potassium, sodium:potassium, and cholesterol because these data were distributed normally. We determined the percentage of live raccoons with coccidia, nematodes, and subcategories of nematodes. We calculated the percentage of raccoons with intestinal parasites detected at necropsy. We calculated the percentages of raccoons that were enteric culture-positive for C. jejuni or Salmonella sp. as well as the percentage of each Salmonella serotype. We calculated percentages with antibodies for rabies, CDV, CAV-1, CPV-2, T. gondii, Borrelia burgdorferi, and R. rickettsii.

We performed physical examinations and collected biological samples on 113 raccoons trapped in Central Park during Feb2010, Sep2010, and Nov2011 (Table 1). Five raccoons were recaptured and sampled at a second time point, for 118 raccoon examinations in total. Vital parameters (mean; range) included rectal temperature (38.4 C; 33.1–40.2 C), heart rate (112 beats/min; 70–190 beats/min), and respiratory rate (32 breaths/min; 12–108 breaths/min). Eleven raccoons trapped in Sep2010 and four raccoons trapped in Nov2011 previously received rabies vaccination in Feb2010, but biologic samples were not collected at the initial vaccination and health assessments were not performed on them at that time.

We noted no abnormalities on physical examination of 52 of 118 (44.1%) raccoons. Tooth abnormalities, including devitalized and fractured crowns, dental calculus, and missing teeth, were detected in 28 of 118 (23.7%) raccoons. Wounds significant enough to necessitate euthanasia were detected in 10 of 118 (8.5%) raccoons and 21 of 118 (17.8%) raccoons had minor wounds. All significant wounds and 8 of 21 (38%) minor wounds were likely sustained before trapping. Minor abrasions and ulcerations on the front feet in 13 of 21 (62%) raccoons may have been sustained in the traps. Minor corneal ulcers were found in 2 of 118 (1.7%) raccoons. We found ectoparasites on 16 of 118 (13.6%) raccoons, including ticks (15/118, 12.7%) identified as Ixodes texanus and lice (1/118, 0.8%) identified as Trichodectes octomaculatus. Nonwound skin abnormalities, such as dermatitis and alopecia, were found in 4 of 118 (3.4%) raccoons.

We euthanized 13 raccoons due to puncture wounds (7/13, 54%), wounds from dog attacks (1/13, 8%), abnormal behavior and puncture wounds (1/13, 8%), large wounds (1/13, 8%), and scratching a person while being handled during the TVR program (3/13, 23%). Rabies and CDV DFA testing was completed on brain tissue from 13 of 13 (100%) and 8 of 13 (61%) euthanized raccoons, respectively; all were negative for detection of viral antigens. Gross and histologic examination, excluding examination of the tissues of the head submitted for rabies and CDV testing, were performed on 8 of 13 (61%) raccoons. External trauma and mild gastrointestinal nematodiasis with associated mild eosinophilic and lymphoplasmacytic gastroenteritis and/or colitis were the most common gross and histologic postmortem findings. The B. procyonis parasite was definitively identified in six of eight (75%) postmortem examinations.

The CBC values were determined for all three sampling times, and biochemistry analytes were determined for raccoons sampled in Nov11 (Table 2). Compared to International Species Information System reference intervals for this species (Teare 2013), we detected leukocytosis in 8 of 32 (25%) raccoons in Feb2010, 6 of 38 (16%) raccoons in Sep10, and 2 of 36 (6%) raccoons in Nov11. In Nov11, one individual had a normocytic, normochromic anemia. We obtained blood lead values in 99 individuals and 5 of those individuals were sampled at two time points (Table 2). Of the five raccoons sampled twice, blood lead ranged from 4.8 to 15.6 μg/dL, and lead concentration increased less than twofold from the first sampling time to the second in two individuals (Table 3). The exact blood lead value was not known on one raccoon sampled in Feb2010 because it was above the detection limit of the test and was reported as >65 μg/dL.

We performed parasitology on 98 fecal samples collected on anesthetized raccoons, including 11 before euthanasia. We found no parasites in the feces of 11 of 98 (11%) raccoons, whereas nematodes were found in 80 of 98 (82%) raccoons and included ova with bipolar plugs (62/80, 78%), strongyle-type ova (47/80, 59%), ascarid-type ova (25/80, 31%), and larvae (7/80, 9%). Nematodes in one sample were definitively identified as B. procyonis, Capillaria procyonis, Capillaria putorii, Crenosoma sp., and Placoconus lotoris. Coccidia were seen in 68 of 98 (69%) raccoons and were definitively identified in one sample as Eimeria nutalli and Eimeria procyonis.

No significant organisms were detected on routine aerobic enteric culture (n=111). Campylobacter jejuni was detected in in 5 of 79 (6%) samples. Salmonella enterica was detected in 70 of 111 (63.1%) fecal samples of which 37 of 70 (53%) were serogroup B and 34 of 70 (49%) were serogroup C. One fecal sample was positive for both serogroups B and C. Eleven Salmonella serotypes were detected (Table 4), with the most common being Salmonella Newport (20/70, 29%) and Salmonella Oranienburg (20/70, 29%).

Overall, rabies VNA ≥0.5 IU/mL was found in 22 of 108 (20.4%) samples of which 9 of 88 (10%) were from raccoons that had not previously received rabies vaccination (Table 5), including three juveniles and six adults. No raccoons that previously received rabies vaccination had a rabies VNA titer <0.125 IU/mL. No raccoons were seropositive for CDV (n=108), CAV-1 (n=60), or Borrelia burgdorferi (n=30). A minority were seropositive for R. rickettsii (3/30, 10%), whereas most were seropositive for CPV-2 (54/59, 92%) and T. gondii (39/60, 65%) (Table 6).

We collected biologic samples from five individual raccoons at two time points and completed serial serologic testing for rabies and CDV on all five raccoons (Table 3). Raccoon 2724 was vaccinated in Feb2010, but biologic samples were not collected then. A positive rabies VNA titer in raccoon 2724 dropped to a low positive value from Sep2010 to Nov2011; the animal was vaccinated against rabies at both time points. The rabies VNA titer of raccoon 2962 did not change from Feb2010 to Nov2011. Rabies VNA titers in the other three raccoons changed from negative at the first sample time to positive at the second sample time. In addition, most had S. enterica isolates at sequential sampling, but none of these raccoons had the same serotype at both time points (Table 3).

Significant potential exists for direct and indirect contact between raccoons and humans and/or domestic animals in Central Park. Many of the infectious agents, parasites, and indications of environmental lead presence determined in this study are capable of causing adverse health effects in animals and in humans, with rabies being the most significant threat. The nine raccoons positive for rabies VNA titer (median titer 1 IU/mL, range 0.5–4 IU/mL), but naïve to rabies vaccination, may have been harboring rabies virus asymptomatically at the time of sample collection (Slavinski et al. 2012); or more likely, they may have been exposed to rabies virus previously (Hill et al. 1992; Brown et al. 2012).

Most of the raccoons in this study were apparently healthy on physical examination and blood analysis. No raccoon euthanized in this TVR program was positive for rabies or CDV antigens. The majority of raccoons previously vaccinated against rabies had a positive rabies VNA titer (13/20, 65%). This TVR program was successful in ending the raccoon rabies epizootic in Central Park (Slavinski et al. 2012). It is possible that the rabies epizootic would have ended regardless of the TVR program because this is likely an isolated raccoon population, but the length of time required for a spontaneous end of the epizootic is unknown. Therefore, it was better from a public health perspective to proceed with the TVR program in an effort to prevent human exposure to rabid raccoons.

At least one raccoon (raccoon 2724 in Table 3) did not maintain an elevated rabies VNA titer from Sep2010 to Nov2011, despite receiving rabies vaccinations in Feb2010 and Sep2010. This lack of rabies titer maintenance could have been due to a variety of factors including age, other subclinical illness, stress, or underlying immunodeficiencies (Brown et al. 2012). It was likely not related to lead-related immunosuppression (Hamir et al. 1994) because the blood lead in this individual decreased with time and was below the median blood lead level (7.3 μg/dL) in this study. The median blood lead level in this study was higher than that previously reported for raccoons in urban New Jersey (4.4 μg/dL) and rural Pennsylvania (2.6 μg/dL; Hamir et al. 1994). However, raccoons have been found to be relatively resistant to the toxic effects of lead compared to other mammal species (Hamir et al. 1999).

Raccoons can become infected with CDV (Bischof and Rogers 2005; Rosatte et al. 2007), a virus that can cause neurologic disease and death in domestic dogs and in a wide variety of noncanid carnivore species (Williams 2001; Terio and Craft 2013). The CDV-associated neurologic signs in domestic and wildlife species can closely resemble clinical signs of rabies virus infection (Rosatte et al. 2007). Another disease that has been found to cause neurologic signs that may resemble rabies in raccoons on the west coast of the US is RacPyV-associated neuroglial neoplasia (Giannitti et al. 2014). Although no raccoons in Central Park have been found with RacPyV brain tumors, 57% tested were seropositive for RacPyV (Church et al. 2016).

Our study indicates that raccoons in Central Park have not been exposed to CDV, possibly due to a well-vaccinated dog population in this urban environment. In contrast, CDV seroprevalence in nonurban raccoons has been reported to be between 23% and 33% (Mitchell et al. 1999; Bischof and Rogers 2005) and was reported to be 55% during a CDV outbreak in urban raccoons and gray foxes in Florida (Hoff et al. 1974). The naïve CDV status of the Central Park raccoon population may make these raccoons more susceptible to a CDV outbreak should the disease be introduced, similar to what was observed with the rabies epizootic in this population.

Carnivore protoparvoviruses such as CPV and feline panleukopenia virus (FPV) are a common cause of severe enteric disease in multiple domestic and nondomestic species (Mann et al. 1980; Nettles et al. 1980). The FPV also causes abortions in domestic cats and was documented to cause fatal disease in zoo-housed felids (Duarte et al. 2009; Sassa et al. 2011). These viruses are extremely stable in the environment, and raccoons may play a role in maintaining carnivore protoparvoviruses in the environment (Stuetzer and Hartmann 2014). Phylogenetic analysis of protoparvoviruses from raccoons (2010–12) from across the US revealed that the majority (22/23) were CPV like instead of FPV like, and it has been speculated that raccoons may have been the intermediate host between the feline-restricted parvoviruses and the emergent canine parvoviruses (Allison et al. 2013). Because we did not have a raccoon parvovirus isolate from Central Park, we chose a CPV-like antigen for the HI assay to assess exposure of raccoons to a carnivore protoparvovirus.

The high percentage of CPV-like seropositive raccoons in the study (92%) strongly suggested that a “raccoon” parvovirus is enzootic in the raccoons of Central Park. The HI test was not sufficiently discriminatory to determine which carnivore protoparvovirus had initiated an infection in a raccoon. Raccoons can become infected with FPV and CPV, but variants may be selected that are more compatible for the transferrin type 1 receptor found on raccoon cells, thus the preponderance of the CPV-like isolates from raccoons in the US (Lee et al. 2016). The CPV-like viruses found in raccoons are not found circulating in the domestic dog population, thereby supporting the concept of an enzootic “raccoon” parvovirus. The exact nature of the carnivore protoparvovirus in the Central Park in raccoons and transmissibility to domestic dogs must await its isolation and sequence analysis.

The CAV-1, which causes infectious canine hepatitis, is capable of causing hepatitis and encephalitis in many carnivore species including raccoons (Woods 2001). However, the absence of CAV-1 antibodies in raccoons in this study indicates that raccoons in Central Park have not been exposed to this virus. Dogs frequenting the park may be well vaccinated against CAV-1 in addition to CDV. Similarly, no raccoons were seropositive for Borrelia burgdorferi, the causative agent of Lyme disease, and sera from only three raccoons were seropositive to R. rickettsii, the causative agent of Rocky Mountain spotted fever (RMSF). Lyme disease and RMSF infections have been diagnosed in people residing in NYC, with 10,763 and 100 cases, respectively, between 2006 and 2013 (Greene et al. 2015). It is unknown exactly where these infections were contracted, but many people diagnosed with vectorborne diseases also had a history of travel outside of NYC (Greene et al. 2015).

No Ixodes scapularis, the vector of Lyme disease, were found on raccoons in this study. Lyme disease is the most common tickborne disease affecting NYC residents, but it is primarily associated with travel outside of the city (Slavinski and Abdool 2015). For these reasons and because no raccoons in this study were seropositive for Borrelia burgdorferi, Lyme disease is not a disease of concern in Central Park. The raccoon-adapted tick found on raccoons in this study, I. texanus, and Dermacentor variabilis are both common ectoparasites of raccoons (Fish and Dowler 1989; Kollars 1993). The latter is the vector of R. rickettsii, but I. texanus has also been found to harbor R. rickettsii (Anderson et al. 1981). The role of I. texanus in transmitting R. rickettsii to Central Park raccoons is unknown. Local transmission of RMSF has been reported in humans in NYC, but only in association with D. variabilis (Slavinski and Abdool 2015). Because a low percentage of raccoons were seropositive for R. rickettsii and none were found with D. variabilis, raccoons likely do not play a role in RMSF transmission in Central Park.

The raccoon is an intermediate host of T. gondii (Dubey and Jones 2008) and the prevalence of serum antibodies to T. gondii may reflect the presence of oocysts in the environment and tissue cysts in food items of these animals (Mitchell et al. 2006). Raccoons do not transmit T. gondii directly to humans because they do not shed oocysts in the feces, but they may serve as an indicator of the presence of T. gondii in a given area. Infection of raccoons with T. gondii may occur by several mechanisms: 1) consumption of oocysts in the environment shed in feces of feral cats entering Central Park, 2) consumption of water contaminated by infected cat feces, or 3) consumption of rodents with infective tissue cysts. Similarly, animals eating raccoons may ingest infective tissue cysts. Variable prevalence (15–100%) of T. gondii antibodies has been reported in raccoons, with 100% seroprevalence in raccoons in a Connecticut study (Mitchell et al. 2006; Hwang et al. 2007; Dubey and Jones 2008). Our results suggest that raccoons are indicators of environmental presence of T. gondii in Central Park, which demonstrates a potential risk to domestic animals and humans in the park.

The primary parasite of concern in this study was B. procyonis, a nematode parasite of raccoons. This parasite is a prolific egg producer and can cause visceral, ocular, and neural larval migrans in humans, domestic animals, and wildlife (Hanley et al. 2006; Thompson et al. 2008; Bauer 2013). Eggs of B. procyonis are extremely stable in the environment across a range of temperatures (Bauer 2013). Therefore, areas of Central Park that are or have been frequented by raccoons, especially areas containing raccoon latrines (Bauer 2013), are likely to be contaminated with B. procyonis eggs and present a potential health risk to humans and domestic animals.

Raccoon feces may also contaminate the environment with enteric bacterial pathogens (Bigler et al. 1975). Because C. jejuni was detected in a low percentage of raccoons in this study, it is possible that raccoons may play a role in the epidemiology of this bacterium in Central Park (Lee et al. 2011). Enteric culture for Salmonella revealed that more than half the raccoons tested in this study shed this organism in feces. As previously reported, raccoons in Central Park are likely not persistent carriers of a particular serotype, but rather are reinfected with Salmonella serotypes from food items or the environment (Jardine et al. 2011) and can therefore play a role in the cycling of this organism in the environment. Salmonellosis is a common cause of gastroenteritis in humans, causing more than 1.2 million cases of illness in the US (CDC 2013). Many of the serotypes found in this study, including the most common serotypes Salmonella Newport and Salmonella Oranienburg, can cause disease in humans (CDC 2013).

Serology was used to detect exposure of Central Park raccoons to several infectious agents significant in human, domestic animal, and wildlife health. Positive results do not necessarily reflect an active carrier state or infection. Because antibodies to these agents were detected in raccoon sera, and this raccoon population on the island of Manhattan is most likely isolated from other raccoon populations, these raccoons have likely been exposed to these agents in or around Central Park, indicating the presence of these agents in the local environment. Some cross-reactions with antigenically similar nontarget pathogens may occur with these serologic tests, and further research is needed to determine, for example, whether these raccoons have been exposed to CPV-2, FPV, or some related parvovirus. Further research involving DNA sequencing may also be helpful in confirming that this raccoon population is isolated from surrounding populations.

Only a subset of raccoons was tested for CAV-1, CPV-2, T. gondii, Borrelia burgdorferi, and R. rickettsii. Therefore, we were not able to determine the seroprevalence of these agents in this population. We found that raccoons in Central Park are likely to be exposed to CPV-2 or similar parvoviruses and T. gondii, whereas they are less likely to be exposed to CAV-1 and R. rickettsii and not likely to be exposed to Borrelia burgdorferi.

Given the high raccoon, domestic animal, and human density in Central Park, there is great potential for interspecies transmission of disease. Our results indicate the presence of several significant pathogens in the environment to which not only raccoons but also humans and domestic animals are susceptible, and for which raccoons may play an ecologic role. These pathogens include the rabies virus, B. procyonis, T. gondii, Salmonella, C. jejuni, and possibly parvoviruses. The most realistic public health recommendation is to increase public awareness of these diseases in such a way as to not cause public overreaction or panic. Other interventions such as preventing raccoons from entering certain areas of the park and increased surveillance would be difficult and costly. Further work is needed to fully characterize the potential impact raccoons in Central Park have on human and domestic animal health.

We thank Maren Connolly, Marc Valitutto, Meredith Clancy, Ihsaan Sebro, Karen Ingerman, and Berni Leahy for collecting and processing samples; Alisa Newton, Elizabeth Dobson, and Carlos Rodriguez for performing necropsies; Alfred Ngbokoli and Daniel Friedman for histologic processing and slide preparation; staff at the New York State Health Department's Wadsworth Rabies Diagnostic Laboratory for rabies and canine distemper virus testing; and USDAWS technicians for trapping and vaccinating the raccoons.