Abstract
OBJECTIVE To identify important subspecies and serovars of Salmonella enterica in a captive reptile population and clinically relevant risk factors for and signs of illness in Salmonella-positive reptiles.
DESIGN Retrospective cross-sectional study.
ANIMALS 11 crocodilians (4 samples), 78 snakes (91 samples), 59 lizards (57 samples), and 34 chelonians (23 samples) at the Bronx Zoo from 2000 through 2012.
PROCEDURES Data pertaining to various types of biological samples obtained from reptiles with positive Salmonella culture results and the reptiles themselves were analyzed to determine period prevalence of and risk factors for various Salmonella-related outcomes.
RESULTS Serovar distribution differences were identified for sample type, reptile phylogenetic family, and reptile origin and health. Salmonella enterica subsp enterica was the most common subspecies in Salmonella cultures (78/175 [45%]), identified across all reptilian taxa. Salmonella enterica subsp diarizonae was also common (42/175 [24%]) and was recovered almost exclusively from snakes (n = 33), many of which had been clinically ill (17). Clinically ill reptiles provided 37% (64) of Salmonella cultures. Factors associated with an increased risk of illness in reptiles with a positive culture result were carnivorous diet and prior confiscation. Snakes had a higher risk of illness than other reptile groups, whereas lizards had a lower risk. Bony changes, dermatitis, and anorexia were the most common clinical signs.
CONCLUSIONS AND CLINICAL RELEVANCE This study provided new information on Salmonella infection or carriage and associated clinical disease in reptiles. Associations identified between serovars or subspecies and reptile groups or clinical disease can guide management of Salmonella-positive captive reptiles.
Bacteria of the genus Salmonella are important causes of gastrointestinal and extraintestinal disease in humans and other animals.1–3 Salmonellosis is a major public health problem, with an estimated 1 million people affected annually in the United States.4 Although most cases of disease are attributable to food contamination, salmonellosis acquired by contact with or exposure to infected animals contributes to the disease burden.5–13 Salmonellosis in humans following contact or exposure to reptiles is associated with outcomes ranging from gastrointestinal illness to sepsis, shock, and death, which has led to recommendations and legislation to reduce the risk of exposure or infection.14,15
Much research has already been conducted, but the role of salmonellae in disease of reptiles remains unclear. When a reptile is exposed to the organism, a range of clinical outcomes is possible. Subclinical carriage, with intermittent shedding of salmonellae through feces, is common.16–18 Constant reinfection can occur through this shedding, making clearance of the pathogen challenging. This intermittent shedding and risk of reinfection complicate the understanding of epidemiology and pathophysiology of salmonellosis in reptiles.3,16,19
Historically, serotyping was performed on Salmonella isolates in an attempt to elucidate the course of disease in individuals and the epidemiology of infection in populations. This has allowed for tracking and tracing of outbreaks of salmonellosis by serovar. In a population of animals, serotyping can permit monitoring and longitudinal analyses, although molecular diagnostic methods, such as pulsed-field gel electrophoresis, now allow for more exact epidemiological tracing.20
Nomenclature for the genus Salmonella has been constantly evolving, leading to some inconsistencies in the literature, particularly for subspecies of veterinary importance. The genus includes 2 extant species: Salmonella enterica and Salmonella bongori. The 6 subspecies of Salmonella enterica (enterica [I], salamae [II], arizonae [IIIa], diarizonae [IIIb], houtenae [IV], and indica [VI]) are further categorized into serovars, with 2,637 extant serovars recognized.21,22 Serovars are determined by phenotyping of the O (somatic) and H (flagellar) antigens. Certain serovars are associated with specific clinical syndromes, and although the inherent virulence of a serovar can increase a bacterium's propensity to cause disease, host immunity plays a large role in its pathogenicity.21,23,24 This is of particular importance in ectothermic animals, in which ambient temperature influences metabolism and immune response.19
Multiple serovars from multiple subspecies of S enterica have been associated with reptile-associated salmonellosis in humans and with subclinical carriage and clinical disease in reptiles.1,5,13,16,25–42 Historically associated with reptiles, serovars from subspecies arizonae (IIIa) and diarizonae (IIIb) were once listed in their own genus (Arizona),29,31,42,43 and some laboratories fail to differentiate between these 2 subspecies during initial biochemical testing.23
Much published information on salmonellae in reptiles focuses on prevalence in routinely collected biological samples, point prevalence in opportunistically collected samples, or clinical illness associated with specific strains. Reports of associations between clinical disease and Salmonella carriage are uncommon, but understanding the importance of positive Salmonella culture results in reptiles is important to practicing best medicine.
At the Wildlife Conservation Society Bronx Zoo, reptiles have been kept in an ever-changing population for more than a century. Salmonella spp have been cultured from various surveillance and diagnostic biological samples from both apparently healthy and clinically ill animals. Given the diversity of this reptile population and the ability to track demographic information and health status, the Bronx Zoo provides a unique opportunity to evaluate the epidemiology of Salmonella carriage and infection in a captive reptile population over several years. The purpose of the study reported here was to retrospectively characterize S enterica isolates in this population, subspecies and serovar distribution, squamate population prevalence, individual factors associated with clinical disease, and common clinical signs and outcome.
Materials and Methods
Animals and samples
The study was designed to focus on all positive Salmonella culture results obtained from reptiles housed at the Bronx Zoo from January 1, 2000, to December 31, 2012. Biological samples for microbial culture were collected by multiple methods during routine or diagnostic examinations of the reptiles, most commonly via collection of fresh fecal samples. All samples were shipped from the Bronx Zoo in bacterial transport mediaa to the Cornell University Animal Health Diagnostic Center, which is the state veterinary diagnostic laboratory of New York. Microbial culture of fecal samples was performed by standard methods, often including testing for multiple enteric pathogens.44 Colonies with biochemical properties consistent with Salmonella spp were confirmed as salmonellae by use of an automated microbiological identification systemb and serogrouped via slide agglutination testing. Results of these laboratory tests included subspecies of S enterica and often serogroup (O antigen grouping). In the event of a positive culture result, samples were also sent to the National Veterinary Services Laboratory for full serotyping.
Sample data sets
The study was designed in 4 portions. For the first portion involving overall characterization of all Salmonella isolates recovered from reptiles, the study population consisted of all samples from any reptile housed at the Bronx Zoo for which positive results of Salmonella testing were obtained from January 1, 2000, to December 31, 2012. These data were identified through a laboratory record search at the Cornell University Animal Health Diagnostic Center. Information from the Cornell University Animal Health Diagnostic Center and Bronx Zoo medical recordsc was catalogued to include reptile group, family, genus, and species; sample source (feces, swab specimen of solid organ, aspirate, or other source); and Salmonella serogroup and serovar. If full serotyping had not been performed or was not possible, the sample was included in the dataset with notation of lack of full typing. These data were analyzed for serogroup and serovar distribution. Reptiles from which multiple isolates were recovered were included. Gathered data were used to calculate the prevalence of positive Salmonella culture results in squamate populations (lizards and snakes) during the study period.
The second portion of the study included characterization of all individual reptiles from which salmonellae had been isolated during the study period, with the intention of identifying individual risk factors for and population prevalence of positive Salmonella culture results. This dataset was developed from the laboratory records from the Cornell University Animal Health Diagnostic Center and evaluation of individual reptile records, including review of medical and zoological collection data in Bronx Zoo medical records.c Any reptile with an active clinical problem recorded in the medical record, abnormal behavior, or clinical signs at point of sample collection was defined as ill. All medical records for the 60 days prior to and following sample submission were reviewed by the same investigator for consistency of clinical illness designation, and reptiles without medical records or husbandry notes were excluded from analysis. Reptile order or suborder (chelonian, crocodilian, lizard, and snake), family, genus, species, sex, age, and origin were cataloged as well as sample source, collection point (during the routine 30- day quarantine period, after quarantine, or after death), and type (diagnostic or routine). Diet type of the various reptile species was established on the basis of zoological collection information and the natural history of the species. When an individual reptile could not be identified via sample submission data or medical record review, the sample was excluded from this portion of the study. Individual animals that had missing data points (eg, sample source, collection point, or type) were excluded from analysis of that putative risk factor but were included in serotype distribution, prevalence, and incidence calculations.
The third portion of the study was designed to characterize antimicrobial susceptibility in recovered Salmonella isolates. Information was collected regarding signalment, grouping, and taxa (family, genus, and species) of the reptile from which each isolate had been recovered; sample source, type, and collection point; and isolate antimicrobial susceptibility patterns. Because of changes in the panel of antimicrobials used in susceptibility testing during the study period, proportions of susceptible samples rather than absolute numbers were used for analysis. Total susceptibility patterns for each sample were also analyzed to compare with reported patterns for multidrug-resistant Salmonella isolates.
The final portion of the study included evaluation of fecal shedding of salmonellae and prevalence of positive Salmonella culture results over a 5-year subset of the study period (2008 to 2012). All fecal samples from reptiles during this period were evaluated to identify prevalence within reptile group and family.
Statistical analysis
To examine true population prevalence, squamate (lizard and snake) population totals were established per annum from 2000 through 2012, along with total population numbers during the study period. Squamates were selected only because they contributed most of the Salmonella-positive samples, and the authors believed this would reduce possible inaccuracies in chelonian population totals caused by colony accessions with incomplete population data and low numbers of positive crocodilian samples. The numerator consisted of the total number of individuals with positive culture results, and the denominator consisted of the total number of individuals in the relevant squamate population. Individuals from which multiple positive samples had been collected were counted only once in this calculation.
Bronx Zoo sample submission records allowed calculation of the prevalence of positive Salmonella culture results for fecal samples during the last 5 years of the study period (January 1, 2008, to December 31, 2012). Calculations were made by identifying the total number of fecal samples collected from reptiles housed at the Bronx Zoo that were submitted for microbial culture over that period and the total number of fecal samples with positive Salmonella culture results over the study period. Individual reptiles from which multiple positive samples had been obtained were counted only once, and the denominator for prevalence calculations was the total number of fecal samples submitted rather than the number of individual reptiles.
In addition to prevalence calculations, binary logistic regression was performed to identify risk factors for positive Salmonella culture results by maximum likelihood, and RR and 95% CIs were computed. Binary logistic regression was also performed to evaluate serovar and subspecies differences. A similar approach was used to calculate the RR that Salmonella spp would be isolated from a clinically ill (vs non–clinically ill) reptile. Values of P < 0.05 were considered significant. Statistical softwared was used for all calculations.
Results
Overall characterization of Salmonella isolates
Between January 1, 2000, and December 31, 2012, positive results of Salmonella culture were obtained for 175 biological samples (4 crocodilian, 23 chelonian, 57 lizard, and 91 snake) collected from 182 reptiles. Full serotyping was not performed or not possible for 27 (15%) of these samples. Multiple isolates were recovered from 30 reptiles (22 samples), yielding a total of 35 isolates, with samples collected from groups of reptiles (rather than individuals) contributing to the total number of samples (175) and isolates (182). Biological sample types from which salmonellae were isolated included fecal samples (117/175 [67%]) as well as swab or wash specimens from the cloaca (20/175 [11%]), coelom (9/175 [5%]), gastrointestinal tract and liver (11/175 [6%]), and other solid organs or masses (18/175 [10%]).
Serovars identified via serotyping of isolates from samples with a positive culture result were unevenly distributed among the subspecies of S enterica: 39 enterica (I), 3 salamae (II), 6 arizonae (IIIa), 25 diarizonae (IIIb), and 5 houtenae (IV). Many serovars were identified repeatedly in the study, with only 78 unique serovars detected. These serovars represented only a small percentage of extant serovars (3%), but the representation of each subspecies differed, with higher prevalences of subspecies diarizonae (IIIb; 7%), houtenae (IV; 7%), and arizonae (IIIa; 6%). Serovar distribution among subspecies of S enterica was significantly (P < 0.001) associated with reptile order or suborder and family insofar as certain subspecies appeared to have limited host range in this population. Salmonella enterica subsp enterica (I) was the most common (n = 78 [45%]), and S enterica subsp diarizonae (IIIb) was the second most common (42 [24%]). Isolates of S enterica subsp diarizonae (IIIb; 42) were mostly recovered from snakes (34 [81%]), and all isolates of S enterica subsp arizonae (IIIa; 10) were also recovered from snakes. The single most common serovar was Salmonella Newport (n = 10). Salmonella serovar IV 43:z4, z32:– was recovered multiple (8) times, including 5 times from a snake described in a previous case report.67 Other common serovars include Salmonella serovars Braenderup (n = 6) and Montevideo, Oranienburg, and Thompson (4 each).
Characterization of all reptiles with positive Salmonella culture results
Samples with positive Salmonella culture results included in the study dataset were obtained from individual reptiles (n = 182) representing 67 species (Table 1). The fact that some samples had been collected from groups of reptiles rather than individual reptiles contributed to a discrepancy between individual and sample counts. Identities could not be discerned for the reptiles from which 8 samples with positive results were obtained; therefore, these samples were excluded from the second portion of the study, in which individual risk factors were identified.
Relative risk of positive Salmonella culture results for biological samples obtained from various groups of reptiles with versus without clinical illness at the Bronx Zoo from 2000 through 2012.
| Reptile group | Total No. of individuals contributing samples (total ill) | Total No. of individuals with positive samples (total ill) | Total No. of samples (total pertaining to ill reptiles) | RR (95% CI) that sample was from ill reptile | P value |
|---|---|---|---|---|---|
| Crocodilian | 11 (0) | 2 (0) | 4 (0) | 0.37 (0.01–5.31) | 0.46 |
|   Alligatoridae | 9 (0) | 1 (0) | 3 (0) | — | — |
|   Crocodylidae | 1 (0) | 1 (0) | 1 (0) | — | — |
| Chelonian | 34 (15) | 11 (5) | 23 (10) | 1.27 (0.12–6.54) | 0.36 |
|   Emydidae | 3 (1) | 3 (1) | 3 (1) | — | — |
|   Geomydidae | 1 (1) | 1 (1) | 1 (1) | — | — |
|   Testudinidae | 30 (13) | 8 (3) | 19 (8) | — | — |
| Lizard | 59 (16) | 22 (9) | 57 (6) | 0.55 (0.32–0.98) | 0.03 |
|   Agamidae | 10 (2) | 3 (2) | 6 (2) | — | — |
|   Anguidae | 1 (0) | 1 (0) | 1 (0) | — | — |
|   Crotaphytidae | 1 (1) | 1 (1) | 1 (1) | — | — |
|   Gekkonidae | 14 (0) | 4 (0) | 15 | — | — |
|   Gerrhosauridae | 1 (0) | 1 (0) | 1 (0) | — | — |
|   Helodermatidae | 1 (1) | 1 (1) | 1 (1) | — | — |
|   Iguanidae | 3 (0) | 3 (0) | 3 (0) | — | — |
|   Opluridae | 2 (0) | 1 (0) | 1 (0) | — | — |
|   Pygopodidae | 8 (2) | 1 (1) | 8 (2) | — | — |
|   Scincidae | 1 (0) | 1 (0) | 1 (0) | — | — |
|   Varanidae | 17 (6) | 6 (4) | 20 (7) | — | — |
| Snake | 78 (33) | 31 (20) | 91 (39) | 1.57 (1.03–2.39) | 0.04 |
|   Boidae | 25 (12) | 5 (5) | 28 (11) | — | — |
|   Colubridae | 17 (7) | 8 (4) | 19 (9) | — | — |
|   Elapidae | 4 (3) | 3 (2) | 4 (3) | — | — |
|   Pythonidae | 11 (6) | 6 (3) | 12 (7) | — | — |
|   Viperidae | 21 (5) | 9 (3) | 28 (9) | — | — |
— = Not calculated. Values of P < 0.05 were considered significant.
Most samples (148/175 [85%]) originated from squamates, with uneven distribution among reptile families and genera (Table 2). Total prevalence of positive Salmonella culture results over the study period for lizards was 16% (95% CI, 15% to 18%), with a high prevalence identified for lizards in the genera Uroplatus (42%) and Lialis (80%). Snakes had a similar prevalence of 18% (95% CI, 16% to 19%). Certain genera in each snake family had a high prevalence of positive Salmonella culture results, including Corallus (40%), Epicrates (29%), and Eunectes (32%) in the family Boidae; Morelia (57%) in the family Pythoidae; and Agkistrodon (34%) and Bitis (71%) in the family Viperidae. No positive culture results were obtained for 2 genera of Boidae (Candoia and Sanzinia) and 3 genera of Pythonidae (Aspidites, Calabaria, and Malayopython).
Distribution of reptiles of squamate genera (lizards [n = 59] and snakes [78]) per population year with positive results of Salmonella culture at the Bronx Zoo from 2000 through 2012.
| Reptile family | Genus | Proportion | Percentage (95% |
|---|---|---|---|
| Lizard |  | 56/1,672 | 16 (15–18) |
|   Agamidae | Chlamydosaurus | 20/115 | 18 (11–25) |
|  | Pogona | 11/117 | 9 (4–13) |
|  | Uromastyx | 13/93 | 14 (7–21) |
|   Anguidae | Ophisaurus | 9/45 | 20 (8–32) |
|   Crotaphytidae | Crotaphytus | 63/253 | 25 (20–30) |
|   Gekkonidae | Phelsuma | 39/326 | 12 (8–16) |
|  | Uroplatus | 137/325 | 42 (34–50) |
|   Gerrhosauridae | Zonosaurus | 27/159 | 17 (11–23) |
|   Helodermatidae | Heloderma | 10/115 | 9 (4–14) |
|   Iguanidae | Cyclura | 18/92 | 20 (12–28) |
|  | Sauromalus | 8/190 | 4 (1–7) |
|   Opluridae | Oplurus | 12/146 | 8 (4–12) |
|   Pygopodidae | Lialis | 155/194 | 80 (74–86) |
|   Scinicidae | Corucia | 22/731 | 3 (2–4) |
|   Varanidae | Varanus | 427/2,247 | 19 (17–21) |
| Snake | |||
|   Boidae | Boa | 21/192 | 11 (7–15) |
|  | Corallus | 17/43 | 40 (25–55) |
|  | Epicrates | 32/112 | 29 (21–38) |
|  | Eunectes | 45/140 | 32 (25–40) |
|   Colubridae | Drymarchon | 6/67 | 9 (2–16) |
|  | Elaphe | 28/87 | 32 (22–42) |
|  | Erpeton | 7/133 | 5 (1–9) |
|  | Gonyosoma | 15/167 | 9 (5–13) |
|  | Lampropeltis | 7/43 | 17 (6–28) |
|  | Leioheterodon | 4/16 | 25 (4–46) |
|  | Pituophis | 7/48 | 14 (5–23) |
|   Elapidae | Naja | 7/145 | 5 (1–9) |
|  | Ophiophagus | 2/48 | 3 (0–8) |
|   Pythonidae | Morelia | 60/106 | 57 (50–64) |
|  | Python | 24/174 | 14 (9–19) |
|   Viperidae | Agkistrodon | 48/141 | 34 (26–42) |
|  | Bitis | 26/36 | 71 (57–86) |
|  | Crotalus | 31/123 | 25 (17–33) |
|  | Protobothrops | 3/34 | 9 (0–19) |
|  | Sistrurus | 2/6 | 33 (0–70) |
With respect to fecal samples collected from 2008 to 2012, snakes had the highest prevalence of positive Salmonella culture results (33%; 95% CI, 21% to 44%), followed by lizards (22%; 95% CI, 10% to 34%). Prevalence for crocodilians (8%; 95% CI, 0% to 21%) and chelonians (4%; 95% CI, 0% to 10%) during this period was considerably lower.
Characteristics of clinically ill reptiles with positive Salmonella culture results
Sixty-four individual reptiles had clinical illness at the point a Salmonella-positive specimen was collected. Common clinical signs in these reptiles were bony changes (n = 15), dermatitis (15), anorexia (14), and lethargy or weakness (10). Other clinical signs included failure to thrive (n = 8), neurologic abnormalities (7), gastrointestinal parasitism (7), loss of body weight or condition (7), coelomic swelling or mass (5), dehydration (5), and gasping or other respiratory abnormalities (5).
Clinically ill reptiles provided 37% of all samples with positive culture results (61 samples from 64 individuals). Although most biological samples pertaining to the overall population were fecal samples collected during screening tests, the distribution of sample sources in the clinically ill subgroup was significantly (P = 0.006) differerent, with most samples having been obtained from other sources, such as bone biopsy specimens (3/61 [5%]) or abscess aspirate samples (6/61 [10%]), for diagnostic purposes.
Distributions of subspecies of S enterica isolated from clinically ill reptiles differed from those isolated from the overall study population. Proportions of isolates belonging to each subspecies for clinically ill reptiles varied (Table 3). Individual species distribution was generally similar to that in the overall Salmonella-positive population.
Numbers of isolates of particular Salmonella subspecies recovered from all reptiles and from clinically ill reptiles alone (parentheses) at the Bronx Zoo from 2000 through 2012.
| Reptile group | enterica | salamae | Untypable III | arizonae | diarizonae | houtenae | Other |
|---|---|---|---|---|---|---|---|
| Crocodilian | 4 (0) | — | — | — | — | — | 1 (0) |
|   Alligatoridae |  | — | — | — | — | — |  |
| Crocodylidae | — | — | — | — | — | — | — |
|   Chelonian | 16 (8) | — | — | — | 2 (0) | — | 3 (2) |
|   Emydidae | 1 (0) | — | — | — | 1 (0) | — | 1 (1) |
|   Geomydidae | — | — | — | — | — | — | 1 (1) |
|   Testudinidae | 15 (8) | — | — | — | 1 (0) | — | — |
| Lizard | 28 (9) | 3 (1) | 6 (0) | — | 7 (2) | 4 (0) | 9 (1) |
|   Agamidae | 4 (2) | 1 (0) | — | — | — | 1 (0) | — |
|   Anguidae | — | — | — | — | — | 1 (0) | — |
|   Crotaphytidae | 1 (1) | — | — | — | — | — | — |
|   Gekkonidae | 1 (0) | — | 3 (0) | — | 3 (0) | 2 (0) | 5 (0) |
|   Gerrhosauridae | — | — | — | — | — | — | 1 (0) |
|   Helodermatidae | — | 1 (1) | — | — | — | — |  |
|   Iguanidae | 2 (0) | 1 (0) | — | — | — | — | — |
|   Opluridae | — | — | 1 (0) | — | 1 (0) | — | — |
|   Pygopodidae | 7 (1) | — | — | — | 1 (1) | — | — |
|   Scincidae | — | — | — | — | — | 1 (0) | — |
|   Varanidae | 13 (5) | — | 2 (0) | — | 2 (1) |  | 3 (1) |
| Snake | 30 (8) | — | 11 (9) | 10 (1) | 33 (17) | 9 (5) | 5 (3) |
|   Boidae | 13 (3) | — | — | 3 (0) | 11 (9) | 1 (0) | 1 (0) |
|   Colubridae | 5 (2) | — | 5 (4) | 3 (0) | 4 (2) | — | 2 (1) |
|   Elapidae | 2 (1) | — | — | — | 3 (2) | — | — |
|   Pythonidae | 2 (0) | — | 5 (4) | — | 3 (2) | 1 (0) | 1 (1) |
|   Viperidae | 8 (1) | — | 1 (1) | 4 (1) | 12 (2) | 7 (5) | 1 (1) |
— = No isolates of this subspecies recovered.
The 2 strongest risk factors for obtaining positive Salmonella culture result from a clinically ill (vs non–clinically ill) reptile were diet type and origin (Table 4). Carnivores with a positive result had a significantly (P < 0.001) greater risk of clinical illness (RR, 4.61; 95% CI, 2.10 to 10.10) than did noncarnivores. Confiscation was also associated with a significantly (P < 0.001) greater risk of clinical illness than were other sources of origin (RR, 2.03; 95% CI, 1.42 to 2.90). Snakes with a positive culture result had a significantly (P = 0.04) greater risk of clinical illness than did other reptile groups with a positive culture result (RR, 1.57; 95% CI, 1.03 to 2.39); lizards had a significantly (P = 0.03) lower risk of clinical illness (RR, 0.55; 95% CI, 0.32 to 0.98).
Associations of various factors with clinical illness (vs no clinical illness) in reptiles with positive Salmonella culture results at the Bronx Zoo from 2000 through 2012.
| Factor | No. (%) of all Individuals | No. (%) of ill individuals | RR (95% CI) | P value |
|---|---|---|---|---|
| Origin | ||||
|   Dealer | 49 (27) | 15 (31) | 0.80 (0.08–4.6) | 0.45 |
|   Confiscation | 38 (21) | 23 (63) | 2.03 (1.5–3.1) | < 0.001 |
|   Captivity | 35 (19) | 12 (34) | 1.10 (0.6–1.4) | 0.72 |
|   Private donation | 38 (21) | 10 (29) | 0.70 (0.4–1.2) | 0.23 |
|   Wild | 4 (2) | 1 (25) | — | — |
|   Unknown | 18 (10) | 3 (17) | — | — |
| Point of sample collection | ||||
|   Quarantine | 88 (48) | 24 (28) | 0.64 (0.3–0.8) | 0.04 |
|   After quarantine | 54 (30) | 28 (54) | 1.66 (1.2–2.9) | 0.006 |
|   Unknown | 40 (22) | 12 (30) | — | — |
| Diet type | ||||
|   Carnivore | 130 (71) | 58 (46) | 4.61 (1.8–8.4) | < 0.001 |
|   Omnivore | 7 (4) | 3 (43) | — | — |
|   Insectivore | 15 (8) | 3 (20) | — | — |
|   Herbivore | 30 (17) | 0 (0) | 0.030 (0.0–0.5) | 0.01 |
| Sample type | ||||
|   Fecal | 136 (75) | 31 (48) | 0.32 (0.2–0.5) | < 0.001 |
|   Nonfecal | 46 (25) | 33 (52) | 3.57 (2.2–4.5) | < 0.001 |
For calculations of RR, the referent group was the group of reptiles lacking the indicated factor.
See Table 2 for remainder of key.
Characterization of antimicrobial susceptibility patterns of Salmonella isolates
Results of full antimicrobial susceptibility testing were available for 116 (66%) Salmonella isolates pertaining to 121 reptiles. Antimicrobial resistance was uncommon, with a mean of > 90% of isolates susceptible to most antimicrobials (Table 5). Only 7 isolates were multidrug resistant, with resistance to penicillins and cephalosporins; partial resistance to tetracyclines, ticarcillin, gentamicin, and trimethoprim-sulfamethoxazole; and intermediate resistance to enrofloxacin and orbifloxacin. These isolates had similar antimicrobial susceptibility patterns, with 5 isolates pertaining to specimens collected within the same 5-month period, all of which were from confiscated chelonians in the family Testudinidae. Four of the 5 isolates were identified as Salmonella Potsdam, constituting the only representatives of this serovar in the study. Each of the 4 isolates originated from different members of a group of Chelonians of genus Pyxis that had been confiscated at the same time. The fifth isolate in this cluster was identified as Salmonella Newport and had been recovered from a group sample from Burmese star tortoises (Geochelone platynota).
Percentage of all Salmonella isolates (n = 116) recovered from reptiles at the Bronx Zoo from 2000 through 2012 with susceptibility to various antimicrobials.
| Antimicrobial | Percentage susceptible* |
|---|---|
| ß-Lactams | |
|   Penicillins | 93.1–94.8 |
|   Cephalosporins | 28.4–98.6 |
|   Potentiated ß-lactams | 94.0–94.8 |
| Aminoglycosides | 96.6–100 |
| Fluoroquinolones | 83.5–100 |
| Tetracyclines | 91.3–100 |
| Sulfonamides | 76.2–96.6 |
| Macrolides | 0.0 |
| Lincomycin | 0.0 |
| Chloramphenicol | 97.1 |
| Carboxypenem | 99.1 |
| Rifamycin | 0.0 |
Reported ranges pertain to multiple antimicrobials within a given class.
Two additional isolates of multidrug-resistant Salmonella spp had similar but not identical susceptibility patterns. Both had been cultured from fecal samples obtained from apparently healthy reptiles, and both were recovered > 1 year after the original cluster of 5 multidrug-resistant isolates had been identified. One isolate was from an emerald monitor (Varanus prasnius), and serotyping was not possible. The second was from a Western diamondback rattlesnake (Crotalus atrox) and was typed as Salmonella Muenchen.
Discussion
Salmonella carriage or infection was prevalent in both apparently healthy animals and clinically ill captive reptiles at the Bronx Zoo in the present study. The percentage of cultures from reptiles with clinical signs of disease that yielded salmonellae (37%) was higher than expected. Although S enterica serovars were associated with clinical illness in this particular reptile population, its role as a primary pathogen remains unclear.
The ability of salmonellae to serve as primary pathogens in snakes has been established.3,19,29,34,35,45–47 For reptiles with positive Salmonella culture results in the present study, some of the clinical signs most commonly observed were consistent with those of other studies17,19 and of the established clinical signs of salmonellosis in reptiles, such as bony changes, weight loss, and lethargy. Other clinical signs, such as dermatitis and gastrointestinal parasitism, may have been related to debilitation, which is common in ill reptiles. Although nearly two-thirds of all biological samples with a positive culture result had been obtained from apparently healthy reptiles, the number of ill individuals with positive Salmonella culture results was considerable.
Because the present study focused on reptiles with positive Salmonella culture results rather than on all reptiles at the Bronx Zoo, analyses to identify risk factors for Salmonella carriage could not be performed, but risk factors associated with (vs without) clinical illness at the point of a positive culture result could be and were identified. Reptile group (with an increased risk for snakes and decreased risk for lizards), diet type, and origin and source and collection point of biological samples were all important. Many of these risk factors were clinically relevant.
A confiscated reptile with a positive Salmonella culture result was more likely to be ill than a nonconfiscated reptile. Confiscated reptiles housed at the Bronx Zoo often come from illegal trade and arrive dehydrated and otherwise compromised. Dehydration can reinitiate shedding of salmonellae in chelonian species.43 Whether this shedding results from alterations in immune function or directly from changes in the gastrointestinal tract that result from dehydration is undetermined but may be important in other reptile groups and in the Bronx Zoo reptile population. Many confiscated reptiles in the present study developed disease associated with a positive Salmonella culture result years after the related sample had been collected, although reptiles in the quarantine period, regardless of origin, had a decreased risk of illness. These findings suggested that the factors underlying the association between confiscation and increased risk of illness at point of a positive culture result were more complex than simply immune alterations caused by transport and stress or that transport and its associated stressors may have more lasting effects on reptilian immune function than suspected.
Other environmental factors, including diet and housing, have been identified as potential risk factors for Salmonella shedding by reptiles.26,34,37,48 Carnivorous diet had a close association with illness at the point of a positive culture result in the reptiles of the present study. Snakes, a reptile group with an increased risk for illness at time of a positive Salmonella culture result relative to other groups, are obligate carnivores. Other carnivorous reptiles, such as varanid lizards, had a relatively high proportion of ill individuals with a positive culture result, whereas most herbivorous lizard families had a relatively low proportion. Previous studies37,48 in which mice and other dietary sources were investigated as a potential origin for salmonellae in reptiles revealed a lack of a direct association.
Enclosure size and usage are other husbandry variables that have not been well explored but could contribute to the risk of reinfection in captive settings. Enclosure characteristics were not investigated for associations with outcome in the present study, and the risk for reinfection because of environmental contamination or housing with a Salmonella carrier could be expected to increase the risk of Salmonella exposure. Enclosure characteristics were inconsistently recorded to allow assessment, which represents a study limitation. Future studies should include comparison of captive and natural environments and evaluation of enclosure and other husbandry variables.
In other studies29–31,34,36,37,45,49–54 of captive reptile populations, prevalence estimates for Salmonella carriage have differed widely (2% to 55% in chelonians, < 5% in crocodilians, 36% to 84% in lizards, 28% to 100% in snakes, and 22% to 76% in entire reptile populations). In the present study, the prevalence of positive Salmonella culture results was lower in lizards (16.5%) and snakes (17.5%) than in the aforementioned studies. Numerous possible explanations exist for these discrepancies, including true differences in prevalence among captive populations and differences in testing methods and sensitivities.55 In the present study, the same testing methods were used consistently throughout the sample collection period.
Husbandry factors, such as whether a natural or captive climate is provided, have been hypothesized to be risk factors for Salmonella carriage and reinfection. One study48 revealed that reptiles from an arid natural environment may be more susceptible to reinfection when housed in a less arid captive enclosure. A similar pattern was not identified in the present study. Snake genera with the highest prevalence of positive Salmonella culture results were largely semiaquatic, most notably in the families Boidae (Corallus, Epicrates, and Eunectes) and Viperidae (Agkistrodon and Bitis). Not all semiaquatic or tropical species had a high prevalence; the aquatic genus Erpeton and tropical genus Gonyosoma both had a relatively low prevalence. Multiple genera of lizards naturally found in arid climates (Heloderma, Oplurus, Pogona, and Sauromalus) had the lowest prevalence of positive Salmonella culture results. No similar environmental or natural history factor was identified between the 2 lizard genera with the highest prevalence (the arboreal tropical Uroplatus and the ground-dwelling temperate Lialis).
Serovar distribution among the subspecies of S enterica in the present study was uneven, with a higher prevalence of serovars from subspecies diarizonae (IlIb), houtenae (IV), and arizonae (IIIa) than from other subspecies. Although subspecies arizonae (IIIa) has already been identified in reptile species, particularly snakes,3,29,45,56 and numerous serovars of S enterica subsp enterica (I) have been isolated from all reptile groups,25,26,29,33,51,52,54,57–62 particularly turtles,15,43,53,63,64 the high prevalence of serovars from subspecies diarizonae (IIIb) and houtenae (IV) in the present study was unique. These 2 serovars were important in this study population, particularly when considering the likelihood of serovars of these subspecies to be associated with illness.24,28,30,37,38,50,62,65 Serovars from S enterica subsp enterica (I), the most common subspecies associated with human disease,4,21,66 were identified in samples from all reptile taxa in the present study. Many of the Salmonella enterica serovars linked to reptile-associated salmonellosis in humans were also identified in the reptiles, including serovars Pomona, Poona, Typhimurium, I 4,[5],12:i:–, and IV 44:z24, z23:–.8,18,25,61 Distribution of serovars among O antigen serogroups was similar to that in another study16 in which certain O antigen serogroups, particularly serogroups other than B and D, were evaluated for an association with reptiles.
Information regarding serovar contributes only partially to evaluations of the epidemiology of Salmonella infection or carriage. Serotyping remains common in veterinary medicine, although alternative techniques that rely on genetic rather than phenotypic characterization20 may provide more valuable epidemiological data in a collection such as the reptiles in the study reported here. Such evaluation was not possible given the available study data, but use of these techniques in a prospective study may further contribute to our understanding.
The relatively low period prevalence of positive Salmonella culture results in the present study, compared with point prevalence estimates from other studies involving similar reptiles, could be attributed to opportunistic rather than routine testing. Frequency of sample collection changed substantially in the last 5 years of the study as the frequency of quarantine testing decreased at the Bronx Zoo. Prevalence estimates reported here may also have been imprecise because of the retrospective population counts, which may have introduced a minimal degree of inaccuracy, and group sample collection, whereby 2 or more individuals could have contributed feces to a single sample. Pooled fecal samples are often used in reptile prevalence studies to reduce the frequency of sample collection and testing, but such collection methods can reduce the accuracy of calculations in that only 1 individual from the group could have been shedding salmonellae but the positive result would represent the whole group. Selection bias may have influenced the lower risk of a fecal sample from a quarantined, clinically ill (vs non-clinically ill) reptile yielding a positive Salmonella culture result, given that samples were more likely to be collected from a source or particular individual if illness was observed.
The veterinary literature14,64 currently recommends against treatment of Salmonella infection or carriage in reptiles unless a reptile is clinically ill. At the Bronx Zoo, antimicrobial treatment has been directed by antimicrobial susceptibility results, but empirical treatment of a clinically ill reptile before results are received often included ceftazidime (a first-line antimicrobial in reptile medicine not included in the antimicrobial susceptibility panel), amikacin, or enrofloxacin. Antimicrobial treatment has not been highly successful for most ill reptiles at the Bronx Zoo or in the literature.3,46 A Taylor's cantil (Agkistrodon bilineatus taylori) with osteomyelitis that was treated with amikacin delivered via osmotic pump reportedly had cessation but not reversal of lesions for the 10-month treatment period, and the disease recurred when treatment stopped.67 Antimicrobial resistance occurs in salmonellae, although multidrug resistance was uncommon in the present study beyond the expected pattern for gram-negative bacteria of resistance to clindamycin, erythromycin, penicillin, and rifampin. Overinterpretation of resistance patterns is possible when interpreters fail to consider the characteristics of the gram-negative bacteria and family Enterobacteriaceae.54,56,59,68
Findings of the present study underscored the importance of salmonellae as potential pathogens in captive reptile collections. Salmonella spp were detected in a third of all snake fecal samples and in biological samples from 15% to 20% of squamates. Clinical illness was evident in a third of reptiles with positive Salmonella culture results. Even though these data supported the notion that salmonellae are a typical component of reptilian intestinal flora, the detected period prevalence was higher than predicted for this captive population. Certain risk factors for illness associated with positive Salmonella culture results were identified, including snake suborder, carnivorous diet, and confiscation, but these factors did not account for all reptiles at risk for developing clinical illness. The difficulties in identifying high-risk reptiles, coupled with the insidious nature of this pathogen, prevent a complete understanding of the epidemiology of Salmonella infection or carriage in reptile collections. Careful assessment of husbandry factors should be considered in future investigations of risk factors for Salmonella carriage and disease, which would be best performed in a prospective study with a similar setting. Treatment of salmonellosis in reptiles is often unsuccessful but should be guided by results of antimicrobial susceptibility testing.
Acknowledgments
This manuscript represents a portion of a capstone project submitted by Dr. Clancy to the Johns Hopkins Bloomberg School of Public Health as partial fulfillment of the requirements for a Master of Public Health degree.
Presented in abstract form at the 45th Annual Meeting of the American Association of Zoo Veterinarians, Salt Lake City, October 2013.
The authors thank Jean Lay and Dr. Kimberly Rainwater for data collection and collation, Dr. Shelley Rankin for subject-matter expertise, and Dr. Pat McDonough for sample testing and initial data compilation.
ABBREVIATIONS
| CI | Confidence interval |
| RR | Relative risk |
Footnotes
BBL Port-A-Cul specimen collection and transport products, Becton Dickinson, Franklin Lakes, NJ.
Sensititre automated microbiology system A80 panel, TREK Diagnostic Systems, Cleveland, Ohio.
MedARKS, International Species Information System, Bloomington, Minn.
Stata, version 12, StataCorp, College Station, Tex.
References
1. Kolker S, Itsekzon T, Yinnon AM, et al. Osteomyelitis due to Salmonella enterica subsp. arizonae: the price of exotic pets. Clin Microbiol Infect 2012; 18:167–170.
2. Mahajan RK, Khan SA, Chandel DS, et al. Fatal case of Salmonella enterica subsp arizonae gastroenteritis in an infant with microcephaly. J Clin Microbiol 2003; 41:5830–5832.
3. Ramsay EC, Daniel GB, Tryon BW et al. Osteomyelitis associated with Salmonella enterica SS arizonae in a colony of ridgenose rattlesnakes (Crotalus willardi). J Zoo Wildl Med 2002; 33:301–310.
4. Scallan E, Hoekstra RM, Angulo FJ, et al. Foodborne illness acquired in the United States—major pathogens. Emerg Infect Dis 2011; 17:7–15.
5. CDC. Turtle-associated salmonellosis in humans—United States, 2006–2007. MMWR Morb Mortal Wkly Rep 2007; 56:649–652.
6. CDC. Multistate outbreak of human Salmonella infections associated with exposure to turtles—United States, 2007–2008. MMWR Morb Mortal Wkly Rep 2008; 57:69–72.
7. CDC. Multistate outbreak of human Salmonella Typhimurium infections associated with pet turtle exposure—United States, 2008. MMWR Morb Mortal Wkly Rep 2010; 59:191–196.
8. CDC. Notes from the field: outbreak of salmonellosis associated with pet turtle exposures—United States, 2011. MMWR Morb Mortal Wkly Rep 2012; 61:79.
9. CDC. Notes from the field: Infections with Salmonella I 4,[5],12:i:- linked to exposure to feeder rodents—United States, August 2011- February 2012. MMWR Morb Mortal Wkly Rep 2012; 61:277.
10. Keen JE, Durso LM, Meehan TP. Isolation of Salmonella enterica and Shiga-toxigenic Escherichia coli O157 from feces of animals in public contact areas of United States zoological parks. Appl Environ Microbiol 2007; 73:362–365.
11. LeJeune JT, Davis MA. Outbreaks of zoonotic enteric disease associated with animal exhibits. J Am Vet Med Assoc 2004; 224:1440–1445.
12. Olsen SJ, Bishop R, Brenner FW, et al. The changing epidemiology of Salmonella: trends in serotypes isolated from humans in the United States, 1987–1997. J Infect Dis 2001; 183:753–761.
13. Vora NM, Smith KM, Machalaba CC, et al. Reptile- and amphibian-associated salmonellosis in childcare centers, United States. Emerg Infect Dis 2012; 18:2092–2094.
14. Bradley T, Angulo FJ. Salmonella and reptiles: veterinary guidelines. J Herp Med Surg 2009; 19:36–37.
15. Cohen ML, Potter M, Pollard R, et al. Turtle-associated salmonellosis in the United States. Effect of public health action, 1970 to 1976. JAMA 1980; 243:1247–1249.
16. Mermin J, Hutwagner L, Vugia D, et al. Reptiles, amphibians, and human Salmonella infection: a population-based, case-control study. Clin Infect Dis 2004; 38(suppl 3):S253–S261.
17. Mitchell MA. Salmonella: diagnostic methods for reptiles. In: Mader D, ed. Reptile medicine and surgery. 2nd ed. St Louis: Elsevier Saunders, 2006; 900–905.
18. CDC. Reptile-associated salmonellosis—selected states, 1996–1998. MMWR Morb Mortal Wkly Rep 1999; 48:1009–1013.
19. Origgi FC. Reptile immunology. In: Jacobson E, ed. Infectious diseases and pathology of reptiles. Boca Raton, Fla: CRC Press, 2007;131–166.
20. Wattiau P, Boland C, Bertrand S. Methodologies for Salmonella enterica subsp enterica subtyping: gold standards and alternatives. Appl Environ Microbiol 2011; 77:7877–7885.
21. Grimont PAD, Weill F-X. Antigenic formulae of the Salmonella serovars. Geneva: WHO Collaborating Centre for Reference and Research on Salmonella, 2007.
22. Issenhuth-Jeanjean S, Roggentin P, Mikoleit M, et al. Supplement 2008–2010 (no. 48) to the White-Kauffmann-Le Minor scheme. Res Microbiol 2014; 165:526–530.
23. Ackman DM, Drabkin P, Birkhead G, et al. Reptile-associated salmonellosis in New York State. Pediatr Infect Dis J 1995; 14:955–959.
24. Bauwens L, Vercammen F, Bertrand S, et al. Isolation of Salmonella from environmental samples collected in the reptile department of Antwerp Zoo using different selective methods. J Appl Microbiol 2006; 101:284–289.
25. Bemis DA, Grupka LM, Liamthong S, et al. Clonal relatedness of Salmonella isolates associated with invasive infections in captive and wild-caught rattlesnakes. Vet Microbiol 2007; 120:300–307.
26. Blaylock RS. Normal oral bacterial flora from some southern African snakes. Onderstepoort J Vet Res 2001; 68:175–182.
27. Cambre RC, Green DE, Smith EE, et al. Salmonellosis and arizonosis in the reptile collection at the National Zoological Park. J Am Vet Med Assoc 1980; 177:800–803.
28. Grupka LM, Ramsay EC, Bemis DA. Salmonella surveillance in a collection of rattlesnakes (Crotalus spp.). J Zoo Wildl Med 2006; 37:306–312.
29. Jackson CG Jr, Jackson MM. The frequency of Salmonella and Arizona microorganisms in zoo turtles. J Wildl Dis 1971; 7: 130–132.
30. Mermin J, Hoar B, Angulo FJ. Iguanas and Salmonella marina infection in children: a reflection of the increasing incidence of reptile-associated salmonellosis in the United States. Pediatrics 1997; 99:399–402.
31. Nakadai A, Kuroki T, Kato Y, et al. Prevalence of Salmonella spp. in pet reptiles in Japan. J Vet Med Sci 2005; 67:97–101.
32. Onderka DK, Finlayson MC. Salmonellae and salmonellosis in captive reptiles. Can J Comp Med 1985; 49:268–270.
33. Orós J, Rodriguez JL, Herraez P, et al. Respiratory and digestive lesions caused by Salmonella arizonae in two snakes. J Comp Pathol 1996; 115:185–189.
34. Pasmans F, De Herdt P, Haesebrouck F. Presence of Salmonella infections in freshwater turtles. Vet Rec 2002; 150:692–693.
35. Scheelings TF, Lightfoot D, Holz P. Prevalence of Salmonella in Australian reptiles. J Wildl Dis 2011; 47:1–11.
36. Schröter M, Roggentin P, Hofmann J, et al. Pet snakes as a reservoir for Salmonella enterica subsp. diarizonae (serogroup IIIb): a prospective study. Appl Environ Microbiol 2004; 70:613–615.
37. Waterman SH, Juarez G, Carr SJ, et al. Salmonella arizona infections in Latinos associated with rattlesnake folk medicine. Am J Public Health 1990; 80:286–289.
38. Weiss SH, Blaser MJ, Paleologo FP, et al. Occurrence and distribution of serotypes of the Arizona subgroup of Salmonella strains in the United States from 1967 to 1976. J Clin Microbiol 1986; 23:1056–1064.
39. Woodward DL, Khakhria R, Johnson WM. Human salmonellosis associated with exotic pets. J Clin Microbiol 1997; 35:2786–2790.
40. Zwart P, Poelma FG, Strik WJ. The distribution of various types of salmonellae and arizonas in reptiles. Zentralbl Bakteriol [Orig] 1970; 213:201–212.
41. DuPonte MW, Nakamura RM, Chang EM. Activation of latent Salmonella and Arizona organisms by dehydration of red-eared turtles, Pseudemys scripta-elegans. Am J Vet Res 1978; 39:529–530.
42. Brenner FW, Villar RG, Angulo FJ, et al. Salmonella nomenclature. J Clin Microbiol 2000; 38:2465–2467.
43. NYS Veterinary Diagnostic Laboratory Bacteriology/Mycology Section work area and standard operating procedures. Ithaca, NY: Cornell University, 2013.
44. Bemis DA, Owston MA, Lickey AL, et al. Comparison of phenotypic traits and genetic relatedness of Salmonella enterica subspecies arizonae isolates from a colony of ridgenose rattlesnakes with osteomyelitis. J Vet Diagn Invest 2003; 15:382–387.
45. Isaza R, Garner M, Jacobson E. Proliferative osteoarthritis and osteoarthrosis in 15 snakes. J Zoo Wildl Med 2000; 31:20–27.
46. Orós J, Rodriguez JL, Espinosa de los Monteros A, et al. Tracheal malformation in a bicephalic Honduran milk snake (Lampropeltis hondurensis) and subsequent fatal Salmonella arizonae infection. J Zoo Wildl Med 1997; 28:331–335.
47. Pfleger S, Benyr G, Sommer R, et al. Pattern of Salmonella excretion in amphibians and reptiles in a vivarium. Int J Hyg Environ Health 2003; 206:53–59.
48. Burnham BR, Atchley DH, DeFusco RP, et al. Prevalence of fecal shedding of Salmonella organisms among captive green iguanas and potential public health implications. J Am Vet Med Assoc 1998; 213:48–50.
49. Geue L, Loschner U. Salmonella enterica in reptiles of German and Austrian origin. Vet Microbiol 2002; 84:79–91.
50. Jang YH, Lee SJ, Lim JG, et al. The rate of Salmonella spp. infection in zoo animals at Seoul Grand Park, Korea. J Vet Sci 2008; 9:177–181.
51. MacNeill AC, Dorward WJ. Salmonella prevalence in a captive population of herptiles. J Zoo Anim Med 1986; 17:110–114.
52. Strohl P, Tilly B, Fremy S, et al. Prevalence of Salmonella shedding in faeces by captive chelonians. Vet Rec 2004; 154:56–58.
53. Pasmans F, Martel A, Boyen F, et al. Characterization of Salmonella isolates from captive lizards. Vet Microbiol 2005; 110:285–291.
54. Kruse T, Sebro M, Clancy M, et al. Comparison of fresh and frozen fecal samples for detection of enteric Salmonella from captive Indian start tortoises (Geochelone elegans). J Zoo Wildl Med 2015; 46:187–190.
55. Corrente M, Madio A, Friedrich KG, et al. Isolation of Salmonella strains from reptile faeces and comparison of different culture media. J Appl Microbiol 2004; 96:709–715.
56. Chambers DL, Hulse AC. Salmonella serovars in the herpetofauna of Indiana County, Pennsylvania. Appl Environ Microbiol 2006; 72:3771–3773.
57. Friedman CR, Torigian C, Shillam PJ, et al. An outbreak of salmonellosis among children attending a reptile exhibit at a zoo. J Pediatr 1998; 132:802–807.
58. Gopee NV, Adesiyun AA, Caesar K. Retrospective and longitudinal study of salmonellosis in captive wildlife in Trinidad. J Wildl Dis 2000; 36:284–293.
59. Manolis SC, Webb GJ, Pinch D, et al. Salmonella in captive crocoiles (Crocodylus johnstoni and C. porosus). Aust Vet J 1991; 58:102–105.
60. Mitchell MA, Shane SM. Preliminary findings of Salmonella spp. in captive green iguanas (Iguana iguana) and their environment. Prev Vet Med 2000; 45:297–304.
61. Pedersen K, Lassen-Nielsen AM, Nordentoft S, et al. Serovars of Salmonella from captive reptiles. Zoonoses Public Health 2009; 56:238–242.
62. Chiodini RJ, Sundberg JP. Salmonellosis in reptiles: a review. Am J Epidemiol 1981; 113:494–499.
63. Koopman JP, Kennis HM. Treatment of reptiles contaminated with salmonellas. Z Versuchstierkd 1976; 18:141–145.
64. van der Walt ML, Huchzermeyer FW, Steyn HC. Salmonella isolated from crocodiles and other reptiles during the period 1985–1994 in South Africa. Onderstepoort J Vet Res 1997; 64:277–283.
65. Jones TF, Ingram LA, Cieslak PR. Salmonellosis outcomes differ substantially by serotype. J Infect Dis 2008; 198:109–114.
66. Nowinski RJ, Albert MC. Salmonella osteomyelitis secondary to iguana exposure. Clin Orthop Relat Res 2000; 250–253.
67. Clancy MM, Newton A, Sykes JM. Management of osteomyelitis caused by Salmonella enterica subsp. houtenae in a Taylor's cantil (Agkistrodon bilineatus taylori) using amikacin delivered via osmotic pump. J Zoo Wildl Med 2016. In press.
68. Ebani VV, Cerri D, Fratini F, et al. Salmonella enterica isolates from faeces of domestic reptiles and a study of their antimicrobial in vitro sensitivity. Res Vet Sci 2005; 78:117–121.
