Emergence of new leptospiral serovars in American Samoa - ascertainment or ecological change?

Leptospirosis is the most common bacterial zoonosis in the world, cause by bacteria belonging to the phylum Spirochaetes and genus Leptospira [1, 2]. Mammals serve as reservoir hosts for leptospires and maintain enzootic transmission cycles within and between animal species. Leptospires have been isolated from almost all mammal species including rodents, livestock, domestic pets, and wildlife. There are over 200 known pathogenic serovars which have host specificity for particular species of animals [2], and host-pathogen adaptation can develop where leptospires colonise the renal tubules of animal hosts without causing any apparent illness in chronically infected animals. When infected animals urinate, the bacteria are excreted into the environment, potentially infecting other animals, and continuing the transmission cycle. Animal species can also be incidental hosts if they are infected with serovars where host-pathogen adaptation has not developed, potentially leading to severe illness or even death. Animal species can be reservoir hosts for some serovars, and incidental hosts for others [3, 4]. The presence of particular serovars and their geographic distribution therefore depend on the variety of local animal species, and host-pathogen adaptation of serovars. Emergence of serovars can result from their adaptation to new species of animal hosts, the introduction of infected animal hosts to new areas, or evolution in animal populations and transmission dynamics driven by ecological change.

Humans are incidental hosts for leptospires, and therefore do not contribute to the transmission cycle. Infection can occur through direct contact with infected animals or through exposure to an environment that has been contaminated by animal urine. The usual route of infection is through cuts or abrasions in the skin, but can also occur through intact (especially waterlogged) skin, the conjunctiva, and ingestion or inhalation of water or aerosols. Human infection can result in a wide spectrum of disease ranging from subclinical infections to renal failure, liver failure, pulmonary haemorrhage, and death [3, 4].

Leptospirosis is found throughout the world, but is particularly common in tropical and subtropical regions where environmental conditions favour the survival and transmission of leptospires. An estimated 500,000 severe cases occur each year (accounting for only 5 to 15% of all clinical infections), and case-fatality rates of over 30% have been reported in some areas [3, 5]. The World Health Organization (WHO) has identified leptospirosis as a neglected tropical disease, and estimates the median global incidence of leptospirosis to be at least 5.1 cases per 100, 000 per year in endemic areas, and 14 cases per 100,000 per year during epidemics. However, incidences vary significantly between regions, with estimated annual incidences per 100,000 per year ranging from 95.5 in Africa, 66.4 in the Western Pacific, 12.5 in the Americas, 4.8 in South-East Asia, to 0.5 in Europe [6].

Most emerging infectious diseases are zoonotic in origin, and ecological changes are major drivers of their emergence [7, 8]. Leptospirosis has recently been discussed as an emerging disease in many different contexts around the world [2, 9], and different environmental and ecological drivers of disease transmission have been identified [10‐12]. Disease outbreaks have been associated with flooding in many areas including India, Philippines, Thailand, New Caledonia, Hawaii, Guyana and Nicaragua [13‐19]. Recently, outdoor recreation and ecotourism have emerged as increasingly important risk factors in developed countries [20]. There have also been recent reports on the changing epidemiology of leptospiral serovars and emergence of serovars in the Pacific region [21‐24]. In Australia, serovar Arborea was first detected in 1998, and has emerged to become the most common serovar in Queensland, responsible for 35% of cases in 2009 [25]. A recent study in Hawaii also reported the emergence of a new serovar [23].

In American Samoa (AS), the first laboratory confirmed case of leptospirosis was detected in 2003, followed by three additional cases and one death over the next 12 months. In response to this outbreak, the Centers for Disease Control and Prevention (CDC) in Atlanta conducted a seroprevalence study in AS to investigate the epidemiology of leptospirosis and the risk factors for infection. Adults were randomly selected from 13 villages on the main island of Tutuila, and 17% of the 341 participants had serological evidence of previous leptospirosis infection using the Microscopic Agglutination Test (MAT) [3]. On multivariable modelling, male gender and contact with dogs were significant risk factors for being seropositive; and high income and bathing in treated municipal water were associated with being seronegative. The study also made an observational assessment that the risk of human leptospirosis was associated with contamination of streams by rodents, dogs, and particularly pig waste [26].

To further investigate the environmental drivers of leptospirosis in American Samoa, we conducted a more extensive seroprevalence study in 2010, and the methods and results have been described in detail in a recent report [27]. In summary, the study included 807 adult participants from 55 villages on five islands of AS: the main island of Tutuila, the adjacent island of Aunu'u, and the remote Manu'a Islands (Ta'u, Ofu, and Olosega). An overall seroprevalence of 15.5% was detected, with three predominant serovars accounting for over 90% of seropositive tests. Questionnaires were used to assess individual-level risk factors, and geo-referenced environmental data were used to investigate environmental risk factors around the home. Significant risk factors for overall seropositivity included male gender, outdoor occupation, low income, lack of knowledge about leptospirosis, living below the median altitude of one's village, and the density and location of piggeries around the home. Risk factors for infection varied between the three most common serovars: serovar Hebdomadis was associated with outdoor occupations and having piggeries near the home; serovar LT 751 with contact with rain puddles and living further from streams; and serovar LT 1163 with flooding at home, flooding at work, living at lower elevation, swimming, and having piggeries near the home [27].

The present paper adds to this picture by comparing the epidemiology of leptospiral serovars in our study in 2010 with results of the CDC study in 2004; and by examining the association between serovar distribution and ecological zones by using geo-referenced environmental data, including newly available vegetation mapping data [28]. We discuss the serological evidence of serovar emergence in the six years between the two studies, and explore the potential epidemiological and ecological implications of our findings.

This study found that three predominant serovars were responsible for human leptospirosis infection in American Samoa in 2010. To the best of our knowledge, the three serovars were not previously known to occur in the territory. The epidemiology of serovars differed significantly from the CDC study in 2004 with changes in the predominant serovars as well as serogroups. Serovar distribution also varied between islands and ecological zones.

Our findings might indicate true serovar emergence, or an arefactual emergence where the serovars were long established but previously undetected because of limitations in laboratory detection methods. The three most common serovars found in our study were not included in the panel of serovars used for MAT in 2004, and therefore would not have been identified even if they were present in AS at the time. It is difficult to make direct comparisons of predominant serovars between the studies because of differences in the serovars used for MAT, but a comparison of serogroups showed marked differences between the two studies, and indicates that changes in serovar epidemiology were not merely the result of cross-reactions between serovars within a serogroup. Further, the studies also used different cut-off titres for MAT, which makes it difficult to compare seroprevalence, but the predominant serovars remain unchanged if a cut-off titre of 1:100 was used in our study [27]. Sampling distribution also differed between the two studies, with the 2004 study focusing on 13 villages on Tutuila, while our study collected data across Tutuila and four other islands. However, the predominant serovars on Tutuila in 2010 were the same three serovars discussed above (Hebdomadis, LT 751, LT1163), and therefore also differed from those found in the 2004 study. We believe the results provide supportive evidence of true serovar emergence in the 6 years between the studies.

Ecological changes are major drivers of the emergence of zoonotic diseases around the world. Changes in interactions between humans and the environment can result in exposure to different animal hosts; and ecological change can alter pathogen transmission dynamics and consequent disease emergence. For example, leptospiral serovar epidemiology can be affected by change in distribution and/or abundance of animal reservoirs (e.g. pigs); introduction of new serovars through transport of animals (e.g. rats on cargo ships) or natural migration (e.g. bats); or environmental changes (e.g. climate, land use) that differentially favor the survival and/or transmission of specific serovars. Recent reports of changing epidemiology of leptospiral serovars in the Pacific region [21‐24] support the hypothesis that ecological factors are likely to be responsible for driving serovar emergence.

Differences in serovar distribution seen in our study could be caused by recent emergence where the pathogens have not yet been introduced to some areas, or the filling of particular ecological niches by the animal hosts of particular serovars. The variation in serovar distribution between islands and ecological zones found in our study might provide insights into possible reasons for the emergence of serovars in AS.

Disparities in serovar distribution between the five islands are likely to be related to variations in environmental conditions. Tutuila is the most urbanized and densely populated island, and home to 95% of the territory's population. The only international port in AS is located on Tutuila, and introduction of animals (e.g. rodents from ships) is most likely to occur here. The Manu'a Islands are very small, remote, isolated, and sparsely populated, with limited air and sea connections to Tutuila. The islands therefore vary in many factors that can affect the transmission dynamics of leptospirosis including animal populations; biodiversity; probability of importation of animals and pathogens (including leptospiral serovars); environmental degradation, pollution, and contamination of streams; flooding risk; and contact between humans and infected animals and/or a contaminated environment. For example, serovar Hebdomadis was only found on Tutuila, and could potentially be a newly introduced serovar from rodents through international shipping.

Differences in geographic distribution of serovars between ecological zones on Tutuila suggest that the three main serovars have different animal hosts that live in distinct environments. Different risk factors for each serovar also support this explanation [27]. For example, serovar LT 1163 is completely absent from high population density areas, and is therefore unlikely to be carried by common peridomestic animals in AS such as pigs, dogs, house mice (Mus musculus), Norway rats (Rattus norvegicus), or roof or black rats (Rattus rattus). More likely animal hosts for serovar LT 1163 include the Polynesian rat (Rattus exulans) and the three species of endemic bats, all of which are more abundant in forested areas [31]. In contrast, serovar Hebdomadis was widely distributed across Tutuila and more prevalent in agricultural areas, and is therefore more likely to be carried by one or more of the peridomestic animal species.

Variation in the geographic distribution of serovars in different ecological zones support the hypotheses that environmental factors play an important role in the transmission dynamics of serovars, and ecological change will therefore be a major driver of serovar emergence.

There are limitations to the use of serological methods for leptospirosis, and future studies involving bacterial cultures will be required to confirm the presence of serovars detected on MAT. It is also possible that yet undiscovered serovars in American Samoa might have produced cross-reactions with serovars in our MAT panel. Veterinary studies would provide valuable information on local animal species responsible for the transmission of each serovar, and improve our understanding of pathogen transmission dynamics. Although our study provided supportive evidence of serovar emergence and showed different spatial patterns in serovar distribution, many parts of AS are sparsely populated and a larger follow-up survey would be required to confirm true serovar emergence, determine whether serovar epidemiology is continuing to evolve, and further explore the relationship between environmental factors and serovar distribution.

Despite limitations in serological diagnosis and possible effects of ascertainment, this paper provides supportive evidence for the emergence of previously unknown leptospiral serovars in AS. Future research should focus on specifically testing the hypothesis that ecological change has direct impact on serovar distribution and disease risk in humans. Such a finding would support environmental monitoring as a valuable tool for identifying high-risk areas for infection and directing public health interventions [32]; and contribute to the growing evidence that ecosystem conservation is beneficial to human health [8, 33].

CLL - MBBS, MPH&TM, FRACGP. PhD Candidate, School of Population Health, The University of Queensland, Brisbane, Australia.

CS - MSc, PhD. School of Population Health, The University of Queensland, Brisbane, Australia.

LDS - BSc. Director, WHO/OIE/FAO Collaborating Centre for Reference and Research on Leptospirosis, Brisbane, Australia.

SBC - MSc, PhD. Supervising scientist, WHO/OIE/FAO Collaborating Centre for Reference and Research on Leptospirosis, Brisbane, Australia. Associate Professor, University of the Sunshine Coast, Sippy Downs, Queensland, Australia.

PW - MBBS, PhD, MAppEpi, MA, FAFPHM. Professor of Ecosystem Health and Dean of Graduate Studies, University of South Australia, Adelaide, Australia.