Hantavirus antibodies in rodents and human cases with pulmonary syndrome, Río Negro, Argentina

Edmundo Larrieu^{1, 2}, Gustavo Cantoni², Eduardo Herrero², Alicia Pérez², Gabriel Talmon², Gabriela Vázquez², Odila Arellano², Paula Padula³

¹Facultad de Veterinaria de la Universidad Nacional de La Pampa;
²Ministerio de Salud de la Provincia de Río Negro;
³Instituto Nacional de Enfermedades Infecciosas INEI-ANLIS Dr. C. G. Malbrán, Buenos Aires

Postal address: Dr. Edmundo Larrieu, Laprida 240, 8500 Viedma, Río Negro, Argentina. Fax: (54-2920) 430007 e-mail: elarrieu@salud.rionegro.gov.ar

Abstract
In Río Negro Province, Argentina, human cases of hantavirus pulmonary syndrome (HPS) appeared in the region of subantartic forests. The Andes virus (ANDV) has been identified in the region both in Oligoryzomys longicaudatus rodents and in humans, with the main transmission being from rodents to humans but also showing the possibility of human to human transmission. Between 1996 and 2004, in 40 campaigns, 29.960 night-traps for capturing live rodents were set up. Blood samples were obtained from the rodents and processed using enzyme immunoassay with recombinant antigens made from ANDV. A total of 1767 rodents were captured, with a capture success of 5.9% and an antibody prevalence of 2.1%. Important differences were observed among the species captured from Andes and Steppe regions. Seropositive Oligoryzomys longicaudatus, Abrotrix olivaceus, Abrotrix xanhtothinus and Loxodontomus microtus were captured. During the 1993-2004 period, 40 HPS cases were registered.

Key words: Hantavirus; Epidemiology; Rodents; Pulmonary syndrome; Andes virus

Resumen
Anticuerpos contra hantavirus en roedores y casos humanos con síndrome pulmonar, Río Negro, Argentina. En la Provincia de Río Negro, Argentina, se presentaron casos humanos de síndrome pulmonar por hantavirus (SPH) en la región de los bosques subantárticos. El virus Andes (AND) fue identificado en la región, tanto en el roedor Oligoryzomys longicaudatus como en seres humanos, demostrándose la transmisión principalmente del roedor al hombre y la posibilidad de la transmisión de persona a persona. Para ello, se procedió a la colocación de 29.960 trampas para captura viva de roedores, tipo Sherman, en 40 operativos efectuados desde 1996 hasta 2004. Se obtuvieron muestras de sangre de los roedores, las que fueron procesadas mediante enzimoinmunoensayo con antígenos recombinantes elaborados a partir de virus AND. Fueron capturados 1767 roedores, con un éxito de trampeo del 5.9% y una prevalencia de anticuerpos contra hantavirus del 2.1%. Se observaron importantes diferencias en las especies capturadas en cada una de las regiones. Se capturaron O. longicaudatus, Abrothrix olivaceus y Abrothrix xanhtothinus y Loxodontomys microtus seropositivos. Se registraron 40 casos humanos en el período 1993-2004.

Palabras clave: Hantavirus; Epidemiología; Roedores; Síndrome pulmonar; Virus Andes

The prototype of the genus Hantavirus, Bunyanviridae family, was isolated in Korea in the 1970 decade and was called Hantaan virus. All related viruses are known as hantavirus and have been associated with two clinical syndromes: hemorrhagic fever with renal syndrome (FHSR), predominant in Asia and Europe and hantavirus pulmonary syndrome (HPS), described in 1993 and present exclusively in America¹.
The first identified reservoir for HPS was the wild rodent, Peromyscus maniculatus, in the USA in 1994. Later other reservoirs such as Sigmodon hispidus, Oryzomis palustris and Peromyscus leucopus, were identified. In Argentina initially Oligoryzomys longicaudatus, O. flavescens and O. chacoensis have been identified as reservoirs, while in the rest of America the list of identified reservoirs becomes longer each day (Calomys laucha in Paraguay and Bolivia, O. microtis in Peru, O. fulvescens in Panama, O. fornesi and Holochilus sciureus in Brasil, O. flavescens in Uruguay, etc.). All of them belong to the Muridae family, Sigmodontinae subfamily^2-7.
In the USA and in several countries of South America, especially in Brazil, Chile and Argentina, increasing numbers of cases have been observed.
During the period 1996-2001 the Argentine Health Ministry, reported 324 cases of human hantavirus infections. Of these, 138 came from provinces of the North  (Salta, Jujuy), 124 were from the Central region (Buenos Aires, Santa Fe) and 62 from the South of the country (Chubut, Neuquen, Rio Negro). Mortality rate reached 30% between 1996 and 2001. The cases were associated with any one of the six genotypes of Andes Hantavirus: AND Nort Oran, AND Nort Bermejo, AND Sout, AND Cent Lec, AND Cent Buenos Aires and AND Cent Plata⁵.
In the province of Río Negro, all cases occurred in the region of subantartic forests and were associated to AND virus (genotype AND Sout) transmitted by infected O. longicaudatus rodents⁸ or from person-to-person^{9, 10}. Successive studies have been carried out in this particular region to define reservoirs^{8, 11}, evaluate risk factors¹², evaluate clinical aspects¹³ and antibody prevalence for Hantavirus in rodents^{8, 14}.
The first objective of this study was to present new information on the species of rodents potentially harboring hantavirus in the subantartic forest region of the province of Río Negro, Argentina, and rodent seroprevalence as well as the relationship between human cases and both the size of rodent populations and rodent seroprevalence. The second objective was to identify differences in rodent population and in the prevalence of antibodies against hantavirus in this region, according to whether they were captured in areas related or not to HPS. Additionally,  results from studies on rodents carried out in the steppe region of the province of Río Negro are presented.

Materials and Methods

The first area under study was Subantartic forest: Bariloche Department, to the west of Rio Negro, Argentina (latitude/longitude S41.79/W71.31 to S41.04/ W71.10) and in neighbouring areas of La Angostura (North, Neuquen) and Lago Puelo (South, Chubut). The climate is cold and humid, with 800 to 1000 mm rain fall. Vegetation is lush, with dense forests coexisting with areas of shrub vegetation (Rose rubiginosa, Rubus idaneus, among others). Farming systems include mushrooms, strawberries, hops and fine fruits. The stable population reaches 142.000 inhabitants, a number which greatly increases during the winter and summer tourist periods (Fig. 1).

Fig. 1

The second area was Steppe region: Departments of Pilcaniyeu to the East of Río Negro, General Roca and 25 de Mayo in the center, and Valcheta and San Antonio to the West (latitude/longitude S41.08/W70.50 to S40.40/ W66.09). The climate is cold in winter and hot in summer, with 200-400 mm rainfall. Vegetation is scarce, with a predominance of shrub and xerophilous species and a minimum of herbaceous stratum.
Sherman type traps were used to capture live rodents, using oats as bait. Traps were set 2 m apart, and each line (consisting of 10-25 traps) were set out in such a way to cover the widest variety possible of shrub and herbaceous stratum, water courses and around rural buildings, favouring the presence of rodents. Standardized procedures were used for setting the traps-night and for caring for the operators' bio safety¹⁵.
In the period 1996-2004, a total of 29.960 traps-night were set as follows:
- Subantartic forests: A total of 4311 traps-night were set in nine campaigns, carried out in homes and sites of exposure to disease, where 24 cases had been detected, and other twenty campaigns unrelated to cases of disease, with a total of 18.625 traps-night set. It was not possible to carry  out this type of study during the years 2000 and 2003.
- Steppe: In the period 2001-2005 eight campaigns were carried out, setting a total of 7024 traps-night.
Captured rodents were anesthetized using either ether or methoxifluorane, and blood samples were obtained using cardiac puncture. After this, the animals were sacrificed by cervical dislocation and necropsy was carried out to obtain the liver, lungs, spleen and kidney. All samples were kept in liquid nitrogen until they were processed at the National Institute of Infectious Diseases INEI-ANLIS Dr. C. G. Malbrán laboratory.
The province of Río Negro maintains a vigilance system that includes a register of the cases occurred during the 1993-2004 period and epidemiological studies to identify sites of exposure.
In 2004, a serological screening was carried out in the rural human population over 18 years of age, in the area of El Manso (subantartic forests, (latitude/longitude S41.35/W71.44), where a 50% of the existing population was studied (490 people).
Blood samples from both rodents and humans were processed using enzyme immunoassay (ELISA) with recombinant antigens produced from AND virus¹⁶.
The relationships between rodent population (expressed as the success of trapping or the percentage of rodents captured every 100 traps set), antibody prevalence against han-tavirus in rodents and human cases occurrence, were analysed using Pearson's coefficient of correlation. Supposing the distribution of the variables was normal, BMDP (SPSS, Chicago, IL) software was used for the analysis.
Odds ratio (OR), Z-test, Chi square test of association and confidence intervals (CI) of 95% were estimated using Epidat 3.0 (Xunta de Galicia, Spain).

Results

In the subantartic forest region, campaigns associated with homes and probable sites of exposure to virus, captured 138 rodents (trapping success rate of 3.2%; CI 95% 2.6-3.7) with an AND seroprevalence of 2.9% (CI 95% 0.8-7.2) (Table 1).

Table 1.- Rodents' captures in sites of exposure to 24 hantavirus in different seasons of the year in the subantartic forests region, positivity for Andes virus (AND) antibodies (percentage), and Oligoryzomys longicaudatus specimens captured. Río Negro, Argentina, 1996-2004.

In campaigns unrelated to disease cases, a significantly higher number of rodents was found (1410  rodents; 7.6% trapping success rate; CI95% 7.2-7.9;  p: 0.0000; OR 0.4) with a lower, non significant difference in the prevalence of antibodies against AND virus (2.4% prevalence; CI 95% 1.5-3.1; p: 0.6398) (Table 2).

Table 2.- Rodents' captures in places unrelated to disease cases, in different seasons of the year in the subantartic forests region, Río Negro, Argentina, 1996-2004

The range of trapping success rate in studies unrelated to disease cases varied from 1.9% to 15.4% in summer, 2.9% to 88.4% in spring, 3.3% to 38.0% in fall and 2.7% to 16.0% in winter (Table 2).
The species most often captured were Oligoryzomys longicaudatus (39.1% related to disease cases, 48% unrelated to disease cases) with no significant differences being observed (p: 0.2204), compared with campaigns associated with homes and probable sites of exposure to virus, Abrothrix longipilis (34.1%, 40.1%) and Abrothrix olivaceus (18.1%, 5.8%) (Table 4).
Among rodents captured in places associated to disease cases, 7.4% of O. longicaudatus were found reactive for AND and from captures unrelated to disease cases, seropositive O. longicaudatus, Loxodontomys microtus and A. longipilis  were respectively 4.4%, 8.3% and 0.2% (Table 4). Differences between the presence of seropositive O. longicaudatus from places related (7.4%) and unrelated (4.4%) to disease cases were not significant (p: 0.2873). The majority of seropositive O. longicaudatus, 70% (23), were adult males (Table 4).
Among captures unrelated to disease cases, Pearson's correlation was negative for trapping success of O. longicaudatus and for seroprevalence (p: 0.1990).
Considering all captures carried out, seasonal success of trapping was 8.9% in spring, 4.8% in summer, 8.3% in autumn and 6.1% in winter. Significant differences were observed between spring and winter (p: 0.0000) and spring and summer (p: 0.0000) with no significant differences observed between spring and autumn (p: 0.2280). Reactive antibodies for Andes virus were 3.2%, 3.2%, 1.4% and 0.9% respectively, with no significant differences observed between spring and the rest of the seasons (p: 0.2428).
In the Steppe region a total of 219 rodents were captured (3.1% trapping success; CI95% 2.7-3.5) with a 0.5% prevalence of antibodies against Andes virus (Table 3). The most frequently captured species were: Abrothrix xanhtorhinus (37.9%), A. olivaceus (19.2%) and O. longicaudatus (11.9%). From the West of the region (latitude/longitude S41.08/W70.50), in the transition zone with subantartic forests, a seropositive A. olivaceus (2.4%) was captured (Table 4).

Table 3.- Rodents' captures in places unrelated to disease cases, in different seasons of the year in the steppe region, Río Negro, Argentina, 1996-2004

Table 4.- Different rodent species captured and the prevalence of antibodies against hantavirus in the subantartic forests and the steppe regions, Río Negro, Argentina, 1996-2004

Trapping success was significantly greater in the subantartic forests region (7.6%) than in the Steppe region (3.1%)  (p: 0.0000; OR 2.7).
During the 1993-1995 period, 7 human cases were recorded and during the 1996-1998 period, 22 cases. Of the latter, 16 corresponded to a single outbreak in 1996, which included one index case and 15 cases in which inter-human transmission was demonstrated by genetic evidence¹⁰, for the rest of the cases in that period the source of transmission was not identified. The 1999-2001 period registered 5 cases and the 2002-2004 period, 6 cases, giving rise to a total of 40 cases with  43.9% lethality rate (Table 5).

Table 5.- Cases and deaths by hantavirus pulmonary syndrome in Río Negro, Argentina, 1993-2004

Font: Ministry of Health, Río Negro, Argentina

Eleven out of 24 cases with rodent-transmission, (45.8%) occurred in spring, 2 (8.3%) in summer, 7 (29.2%) in autumn and 4 (16.7%) in winter, with significant differences being found between spring and the other seasons (Z: 65, p: 0.0000). Fifteen cases with inter-human transmission occurred in spring¹⁰.
Capture of rodents related to homes and sites of exposure to disease, allowed identification of infected rodents in only 4 of 24 cases (16.7%) (Table 1). The latter were the only samples where the AND seroprevalence in O. longicaudatus was equal or superior to 15% (Table 2). This prevalence in studies unrelated to cases of the disease was only observed in the spring and summer of 1996 (the year with a maximum number of cases).
Of the 243 blood samples obtained from adult inhabitants (range 18-83 years old) 3 were positive (1.2%, CI 95%  0.25-3.6),  26, 32 and 36 years old and without antecedent of SPH.

Discussion

This study reinforces the importance of O. longicaudatus as a possible reservoir for hantavirus in the region of subantartic forests^{8, 11, 12, 17}. Animals with antibodies against AND virus were captured on both the Northern and Southern borders of Río Negro, in the localities of El Bolsón and San Carlos de Bariloche, in different natural environments and related to homes and sites of exposure to human cases to hantavirus. The rates of rodent capture, 3.2% and 7.6% respectively, related or unrelated to disease cases showed extreme variability according to seasonality.
The greater number of rodents and human cases in spring and fall suggested the possible role of the increased rodent population as a risk factor for man, and spring and fall as the seasons with the highest risk of transmission. In previous studies in the neighbouring province of Neuquen¹⁷ and in North of Argentina¹⁸  also a greater number of rodents was found in spring than in other seasons.
Overall the proportion of infected rodents found in this study was low, and comparable to the reported in other regional studies^{5, 11, 17}. However prevalence close to or higher than 10% was observed during the four seasons of the year, when the captures of infected rodents were from places related to human cases. Nevertheless, no coincidence was found between seasonal prevalence in rodents and human outbreaks, and between traping success rate and seroprevalence in O. longicaudatus.
In only 16.7% of studies in homes and sites of exposure to disease cases positive rodents were captured. In the North of Argentina, positive rodents were captured in only 1 out of 9 (11%) sites of exposure to disease cases¹⁹. This is indicative of the existing difficulties and limitations to reveal and precisely identify sites of infection and risk activities for man.
Previous studies have shown increased infection rates in O. longicaudatus males⁸. Local environmental conditions could generate increased rodent prevalence and gender distribution. According to this study, the potential role of Abrothrix spp and of other species as hantavirus reservoirs in the subantartic forests region is extremely limited, either because of their low seroprevalence (A. longipilis, 0.2%) or their low number (L. microtus, 0.9%).
Seroprevalence for hantavirus found in the adult human rural population of the subantartic forests region was also low (1.2%) and closer with the low seroprevalence found in rodents (2.9% and 2.4% respectively related or unrelated to disease cases). Studies carried out in the North of Argentina identified higher human prevalence (6.3%) associated with an equally high rodent seroprevalence (5.8%)¹⁸.
In the Steppe region, successful capture of O. longicaudatus occurred only in transition zones with subantartic forests (Pilcaniyeu) and in small numbers. Nevertheless, in the neighbouring province of Neuquen, to the North of Río Negro, the presence of O. longicaudatus with antibodies against hantavirus has been notified¹⁷, which raises the need to maintain vigilance systems in this region.
Abrohtix xanthtorimus was the predominant species in this study (37.9%) but until now this species has not been reported as hantavirus carrier. Other studies identified Eligmodontia spp and A. iniscatus as the predominant rodent species in the north of the steppe region of Río Negro¹⁷. To the west of the region, one A. olivaceus specimen with antibodies against hantavirus was captured, being this a species predominant in the region (19.2%). Other studies will be required to define the possible role of Abrotrix spp. in hantavirus ecology in the Steppe region.

References

1. Nichol ST, Spiropoulou CF, Morzunov S, et al. Genetic identification of a hantavirus associated with an outbreak of acute respiratory illness. Science 1993; 5: 914-7.        [ Links ]

2. Child JE, Ksiazek TG, Spiropoulou CF, et al. Serologic and genetic identification of Peromiscus maniculatus as the primary reservoir for a new hantavirus in the southwestern United States. J Infect Dis 1994; 169: 1271-80.         [ Links ]

3. Martínez VP, Colavecchia S, García Alay M, et al, A. Síndrome pulmonar por hantavirus en la Provincia de Buenos Aires. Medicina (Buenos Aires)  2001; 61: 147-56.        [ Links ]

4. López N, Padula P, Rossi C, et al. Genetic identification of a new hantavirus causing severe pulmonary syndrome in Argentina. Virology 1996; 6: 220-3.        [ Links ]

5. Padula P, Colavecchia S, Martínez P, et al. Genetic diversity, distribution and serological features of hantavirus infection in five countries in South America. J Clin Microbiol 2000; 38: 3029-35.         [ Links ]

6. Riquelme R, Riquelme M, Torres A, et al. Hantavirus pulmonary syndrome, southern Chile. Emerg Infect Dis 2003; 9: 1438-43.        [ Links ]

7. Rosa ES, Mills JN, Padula PJ, et al. Newly recognized hantaviruses associated with hantavirus pulmonary syndrome in northern Brazil: partial genetic characterization of viruses and serologic implication of likely reservoirs. Vector Borne Zoonotic Dis 2005; 5: 11-9.        [ Links ]

8. Cantoni G, Lázaro M, Resa A, et al. Hantavirus pulmonary syndrome in the Province of Río Negro, Argentina, 1993-1996. Rev Med Trop São Paulo 1997; 39: 191-6.         [ Links ]

9. Enría D, Padula P, Segura E, et al. Hantavirus Pulmonary Syndrome in Argentina. Possibility of person to person transmission. Medicina (Buenos Aires) 1996; 56: 709-11.        [ Links ]

10. Padula P, Edelstein A, Miguel S, et al. Hantavirus pulmonary syndrome outbreak in Argentina: molecular evidence for person-to-person transmission of Andes virus. Virology 1998; 15: 323-30.        [ Links ]

11. Cantoni G, Padula P, Calderón G, et al. Seasonal variation in prevalence to hantaviruses in rodents from southern Argentina. Trop Med Int Health 2001; 6: 811-6.        [ Links ]

12. Levis S, Morzunov S, Rowe J, et al. Genetic diversity and epidemiology of hantavirusses in Argentina. J Infec Dis 1998; 177: 529-38.        [ Links ]

13. Yadón Z. Epidemiología del síndrome pulmonar por hantavirus en la Argentina (1991-1997). Medicina (Buenos Aires) 1998; 58: 25-26.        [ Links ]

14. Lázaro M, Resa A, Barclay C, et al. Síndrome pulmonar por hantavirus en el sur argentino. Medicina (Buenos Aires) 2000; 60: 289-301.        [ Links ]

15. Mills J, Childs J, Ksiazek G, et al. Methods for trapping and sampling small mammals for virologic testing. US Department of Health and Human Services, Center for Disease Control and Prevention, Atlanta; 1995:1-65        [ Links ]

16. Padula P, Rossi C, Della Valle M, et al. Development and evaluation of a solid phase enzyme immunoassay based on Andes hantavirus recombinant nucleoprotein. J Med Microb 2000; 49: 149-55.        [ Links ]

17. Piudo L, Monteverde M, González Capria S, et al. Distribution and abundance of sigmodontine rodents in relation to hantavirus in Neuquen, Argentina. J Vector Ecol 2005; 30: 119-25.        [ Links ]

18. Sosa Estani S, Martínez V, Gonzáles M, et al. Hantavirus en población humana y de roedores de un área endémica para el Síndrome Pulmonar por hantavirus en la Argentina. Medicina (Buenos Aires) 2002; 62: 1-8.        [ Links ]

19. Gonzáles Della Valle M, Edelstein A, et al. Andes virus associated with hantavirus pulmonary syndrome in northern Argentina and determination of the precise site of infection. Am J Trop Med Hyg 2002; 66: 713-20.        [ Links ]

Received: 20-11-2007
Accepted: 14-7-2008