Article Text

Reptile-associated salmonellosis in children aged under 5 years in South West England
  1. Dan Murphy1,
  2. Femi Oshin2
  1. 1Royal Cornwall Hospital, Truro, Cornwall, UK
  2. 2Cornwall & Isles of Scilly Health Protection Team, Sedgemoor Centre, St Austell, Cornwall, UK
  1. Correspondence to Dr Dan Murphy, Royal Cornwall Hospital, Penventinue Lane, Truro, Cornwall TR1 3LJ, UK; Daniel.murphy{at}


Objective To determine the proportion of Salmonella cases in children aged <5 years that were reptile-associated salmonellosis (RAS) and to compare the severity of illness.

Design To analyse all cases of salmonellosis reported to public health authorities in children aged under 5 years in the South West of the UK from January 2010 to December 2013 for reptile exposure, age, serotype, hospitalisation and invasive disease.

Results 48 of 175 (27%) Salmonella cases had exposure to reptiles. The median age of RAS cases was significantly lower than non-RAS cases (0.5 vs 1.0 year). RAS cases were 2.5 times more likely to be hospitalised (23/48) compared with non-RAS cases (25/127; p=0.0002). This trend continued in cases aged under 12 months, with significantly more RAS cases hospitalised (19/38) than non-RAS cases (8/42; p=0.003). Significantly more RAS cases had invasive disease (8/48: 5 bacteraemia, 2 meningitis, 1 colitis) than non-RAS cases (4/127: 3 bacteraemia, 1 meningitis).

Conclusions Reptile exposure was found in over a quarter of all reported Salmonella cases in children under 5 years of age. RAS is associated with young age, hospitalisation and invasive disease.

  • Infectious Diseases
  • salmonellosis
  • reptile

Statistics from

What is already known on this topic

  • Ownership of exotic reptiles is reported to have increased in recent years, as have cases of reptile-associated salmonellosis (RAS), particularly among children.

  • RAS from indoor reptiles is associated with more severe disease at a younger age.

What this study adds

  • Over a quarter of Salmonella cases in children under 5 years of age had exposure to reptiles.

  • Children with reptile-associated salmonellosis were younger and were more likely to have invasive disease and be hospitalised.


Salmonellae are motile gram-negative bacilli that infect or colonise hosts. In humans, characteristic clinical features of infection include gastroenteritis, enteric fever, bacteraemia and focal infections such as osteomyelitis, abscess formation and meningitis. Zoonotic transmission occurs especially from reptiles that can excrete Salmonella from their gut. Reptile-associated salmonellosis (RAS) can occur from pet reptiles and amphibians, particularly in infants and preschool children.1

Irish researchers identified 14 cases of RAS from 2005 to May 2008. In total, 8 of the 14 cases were in infants <1 year of age.2 Researchers retrospectively analysing 177 cases in the literature since 1965 identified that RAS is significantly associated with invasive disease (including meningitis, bacteraemia, bone and joint infection) at younger ages, particularly below 6 months. Exposure to reptiles that are kept indoors, such as iguanas, bearded dragons, snakes, chameleons and geckos, is associated with RAS at a younger age and more invasive disease.3

A case–control study conducted in the USA from 1996 to 1997 estimated that RAS accounted for 21% of all laboratory-confirmed sporadic Salmonella cases in humans under age 21 and suggested that RAS is associated with approximately 74 000 Salmonella infections annually.4


To determine the proportion of Salmonella cases in children aged <5 years that were RAS and to compare the severity of illness.


All cases of salmonellosis in children aged under 5 years in the South West area of the UK (covering Dorset, Somerset, Devon, Cornwall and the Isles of Scilly), reported to and recorded by the local health protection team, were identified from January 2010 to December 2013.

Individual case records were reviewed to identify whether there had been any exposure to reptiles or whether the case had been hospitalised as a result of the infection.

Of 250 identified cases, 75 were excluded on the basis of incomplete information to enable the primary research questions to be answered.

Proportions of hospitalised children and invasive disease in reptile-exposed and non-reptile-exposed salmonellosis groups were analysed for evidence of a statistically significant difference using the Z-test of two proportions.

The distributions of ages of children in each group were compared using a Mann–Whitney U test to account for dissimilar sample sizes and non-parametric distributions.


Of 175 cases with complete data sets, 48 had exposure to reptiles (27.4%).

The median age of RAS cases was 0.5 years (IQR 0.16–0.66), which was significantly different to that of non-RAS cases whose median age was 1.0 years (p=0.001).

Significantly more of the cases exposed to reptiles were admitted to hospital (n=23/48; 48%) compared with 25/127 (19%) of non-RAS cases (p=0.0002) (see figure 1).

Figure 1

Hospitalisation rates for reptile-associated salmonellosis (RAS) and non-RAS in children aged under 5 years.

Significantly more of the reptile-associated cases under 12 months of age were hospitalised (n=19/38; 50%) compared with 8/42 (19%) of non-RAS cases under 12 months of age (p=0.003).

Invasive disease occurred in 8/48 (17%) RAS cases. Five had bacteraemia, two meningitis and one colitis requiring surgery. This was significantly more than the 4/127 (3%) of non-RAS cases, three of which were bacteraemia, one meningitis (p=0.0016).


This study found reptile exposure in over a quarter of all reported Salmonella cases in children under 5 years of age in South West England. This proportion of cases is higher than previously reported in the literature and represents a substantial modifiable disease burden.

Reptiles asymptomatically excrete Salmonella from their gut. Children, particularly those under 12 months of age, may become infected with Salmonella via environmental exposure through sharing living space with an indoor reptile. While crawling and undergoing developmental stages that include oral exploration, the younger child would be particularly at risk of contracting Salmonella shed from a reptile sharing the same space. Further studies may wish to examine the extent to which reptiles roam free in spaces shared with cases and contaminate those spaces.

In terms of age distribution, RAS cases appear to occur at a younger age than non-RAS cases (p=0.001). While this may reflect the clinician's inclination to investigate earlier in younger age groups, this finding concurs with previous studies, which have shown that exposure to indoor reptiles is associated with more severe disease at a younger age.3

Of all cases studied in children under 5 years and under 12 months of age, significantly more RAS cases were admitted to hospital than non-RAS, suggesting that reptile-associated infection carries a higher likelihood of more severe symptoms. Of the eight RAS cases with invasive disease, two bacteraemic infections were due to serotype Muenchen, but no other dominant serotypes were evident. Of the four non-RAS cases with invasive disease, two cases of bacteraemia were due to serotype Agama, with one due to ser. Newport and one case of ser. Typhimurium meningitis.

Salmonella carried by reptiles tend to be different serotypes from those acquired by food poisoning. While no significant differences in serotype distribution could be attributed to invasive disease in RAS, there were some interesting descriptive differences between RAS and non-RAS serotype patterns. In total, 46/127 non-RAS cases were ser. Typhimurium and 13 were ser. Enteritidis, equally distributed between PT1, PT4 and PT8 subtypes. Conversely, no Typhimurium serotypes were isolated in the RAS cases. Four ser. Enteritidis RAS isolates were all PT8 subtypes. Also, 8/48 RAS cases were ser. Arizonae, 3/48 were ser. Muenchen. Other RAS serotypes seen included Urbana, Carmel, Wassenaar, Newport, Monschaui and Oranieburg. Further work linking specific serotypes to invasive disease in children is needed.

Cases were excluded on the basis of incomplete data collection. Further work may be to look at the serotypes of these excluded cases, examining the serotype and potential of recent travel. A Salmonella Typhimurium subtype and a history of recent foreign travel may not be followed up and investigated in the same way, leading to potential bias away from RAS cases in those excluded.


RAS constitutes a substantial proportion of Salmonella infections in young children. RAS is associated with a severe outcome—hospitalisation and invasive disease. Coupled with evidence of increasing ownership of indoor reptile pets,6 the incidence of RAS hospitalisation is likely to increase. Health professionals such as general practitioners and paediatricians need to be aware of this risk. Further work is required to inform the public of this risk and allow them to make informed choices around pet selection and potential health risks to young children.


Supplementary materials


  • Contributors DM: conception and design, analysis and interpretation of data; drafting the article; revising it critically for important intellectual content. FO: acquisition of data, critical appraisal and final approval of the version to be published.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.