Risk factors of leptospirosis in Liping County, Guizhou Province, China—an expanded survey in general population and rodents

HUANG He, LIU Ying, LI Shi-jun, YAO Guang-hai , LIANG Wen-qin, ZOU Zhi-ting

(Department of Prevention and Control of Communicable Disease, Guizhou Center for Disease Control and Prevention, Guiyang 550004, China)

Leptospirosis is a widely distributed zoonoses, which mainly affects low-income rural farmers and people living in urban slums in temperate and tropical regions[1]. Transmission may occur when people are submitted to direct contact with urine, fresh tissue of infected animals, or indirectly exposed to contaminated water or soil[1-2]. Human leptospirosis features multiple, complex modes of transmission, numerous hosts, a multitude of pathogenic serovars, protean clinical manifestations, non-specific symptoms, and difficulties in early lab confirmation[3], which imposes great challenges to prevention and diagnosis. If improperly treated, there may be a risk of high mortality in humans (as high as 70%, reported by E.L.Gouveia[4]). Additionally, previous reports suggested that climate change (in particular floods and hurricanes) and population drift from rural to urban areas have impacted on the nature of the disease and may influence the magnitude and severity of outbreaks[5-7]. Therefore, long-term surveillance on host animals and comprehensive extended on environment and animals is recommended as basic measurements for leptospirosis prevention.

Qian Dongnan Prefecture, located in southeast of Guizhou Province, has been an epidemic area for leptospirosis, particularly in Liping County, where the mortality rate was as high as 22% from 2001 to 2008. All death cases had serious pulmonary haemorrhage with respiratory failure[8]. Previous research indicated the importance of rodents in transmission circle of leptospirosis, and suggested the probable epidemic serovar was serovar Lai. According to the China National Information System of Communicable Diseases Report, eight leptospirosis cases were reported in 2014, and two acute cases were deceased due to delayed proper treatment, which alerted local public health system that leptospirosis may be a potential threat to rural residents if the awareness of the public and clinicians remains low.

Our study sought to unravel the distribution of risk factors in three different villages in Liping County, in order to emphasis the importance of surveillance on host animals. The outcomes may serve to improve our understanding of leptospirosis, and propose key steps for disease prevention.

Materials and methodsInvestigation site selection

Three villages were involved in the present study. Cenghu and Xuejiaping were the two villages where the two fatality cases were reported in October 2014. We chose another village nearby (Luoli) where no leptospirosis case has been documented in past three years as the extended site, in order to figure out whether areas without reported cases was free fromLeptospirainfection in the general population and rodents. The local demographic data and environment information of the three target sites were collected by investigating village chiefs and doctors using the same questionnaire.

Materials and subjects

The inclusion criterion for present study were the following: general local residents who were aged 10 to 65, had not been diagnosed with leptospirosis clinically or by laboratory, and were able to understand the purpose of the study and signed informed consent at the same time. However, people who suffered health problems (moderate/severe diseases) or in special physiological condition (such as being pregnant), or had ever been diagnosed with leptospirosis were excluded from the study. Age-stratified sampling was applied during the planning stage; the subjects were divided into five age-groups from 10 to 50. When selecting the subjects, we tried to balance the distribution of age, and we chose about 10 participants in each group at every site. After informed consent, participants answered a few questions about their occupation, vaccination history, frequency of doing farm work and self evaluation of their health condition. Blood specimens of subjects were collected using 5 mL vacuum blood tubes.

Serum was isolated and frozen for identification ofLeptospira. Microscopic agglutination test (MAT) was conducted using the anti-serum againstLeptospiraserogroupIcterohaemorrhagiaeSerovarLaiStrain56601,CanicolaSerovarCanicolaStrain56603,AustralisSerovarAustralisStrain56607 andHebdomandisSerovarHebdomandisStrain56610 provided by the Chinese Center of Disease Control and Prevention. The selection of serovars from 15 serogroups 15 serovars was based on time saving and cost efficient considerations as well as previous studies[2,9-10]which proved these four strains were dominate in Liping. The samples were two-fold dilution from 1∶50 to 1∶800. Ultimate titer of MAT was defined as the reciprocal of the highest serum dilution with 50% ofLeptospiraagglutinated. The sample was recognized as positive if its titer ≥100. The positive case was assumed to have previous infection or current subclinical infection. The highest titer of the tests with four strains was recognized as the titer of the sample.

Night trapping method was used to collect rodents. The traps with peanut baits were set in rice-fields before sunset, and collected in the next morning. The species and gender of captured rodents were identified and recorded. The bodies of rodents were sterilized in 84 disinfectants for 10 min. Dissection was then performed, and kidney samples of rodents were collected and stored in 8 mL liquid Ellinghausen-McCullough-Johnson-Harris (EMJH) medium for isolation and cultivation ofLeptospira. The kidney specimens were incubated at 28 ℃, and observed weekly by dark field microscopy for approximately two months.

Laboratory work

The identification ofLeptospirain the rodent kidney specimens was performed by G1/G2-PCR method. The DNA Extraction Kit (SBS Genetech, Beijing, China) was used to extract DNA fromLeptospirastrains which cultured for five to seven days. The primer sequences of G1/G2 were listed as below. G1: 5′-CTGAATCGCTGTATAAAAGT-3′, G2: 5′-GGAAAACAAATGGTCGGAAG-3′. The 25 μL total volumes of PCR reaction system contained 12.5 μL of Pre Mix Taq, 2 μL of forward and reverse primers, respectively, 1 μL of DVA and 7.5 μL of deionized water. The PCR amplification parameters included an initial denaturation at 95 ℃for 5 min; 35 cycles of 94 ℃ for 60 seconds, 55 ℃ for 60 seconds, 72 ℃ for 90 seconds; followed by an extension at 72 ℃ for 10 min. The PCR production was detected by electrophoresis on 1.5% of agarose gel. The positive amplification product was identified by the 285 bp DNA band. The procedures follows the requirements of National Criteria for Leptospirosis Diagnosis[11].

Data analysis

The original information of humans and rodents was inputted using Epidata Version 3.1. The data of Xuejiaping Village and Cenhu Village were combined into epidemic area, while Luoli Village acted as non-epidemic area (the extended area). IBM SPSS Statistics 20 was utilized to analyze the data. Chi-square test and Fisher’s exact possibilities test were used to analyze the differences of variables in case and extended area. The level of significance was set at 0.05 (α = 0.05). Statistically significant factors or factors of clinical or epidemiological importance were included in Logistic regression to analyze the risk factors ofLeptospirainfection.

Results

Xuejiaping Village and Cenhu Village are located in the north and the south of Liping County with a population of 2 326 and 1 478, respectively. The coverage of paddy fields in these two places is approximately 40% and 60%. From 2009 to 2014, three leptospirosis cases had been reported, and two of them had died of acute pneumorrhagia caused byLeptospirainfection. However, in Luoli, a larger village in the west with 3 284 residents, 90% coverage of paddy fields had no cases reported during the same period.

In the three villages, people live separately from their barns. However, there were no closed sewerage systems in the barns. Since 2002, the vaccination for leptospirosis has not been available for general population. With the exception of a few villagers in Xuejiaping, most of the residents in the three areas had the habit of wearing shoes or waterproof socks when they work in paddy fields. In Luoli village and Cenghu village, feces of humans and livestock were discharged into the water body or fertilized without treatments, while in Xuejiaping, feces were composted before fertilization (Table 1).

Tab.1 Demographic information of two areas

ItemsExtendedareaEpidemicareaP-valueAgeMean±Std45.2±14.936.4±15.10.001*Gender0.046** Male1055 Female3588Occupation0.017# Famer3588 Student025 Theunemployed17 NonagriculturalWorker923Vaccinationhistory0.000# Vaccinated38142 Neverreceived71

Note:*The difference was examined by t-test. ** The outcome of Chi-square test. # The outcome of Pearson’s Exact possibilities method. α=0.05.

Both of the fatality cases had been exposed to a contaminated environment. Mr. Shi (the family name of the patient), a 47-year-old man, living in Cenghu, had worked in paddy fields and caught fish in a river several days before fell ill with fever. Mr Long, the other man who was 24, had been repairing his house in Xuejiaping before he felt onset of fatigue. Similarly, their initial symptoms were slight fever, myalgia and fatigue, which were treated as flu in local clinic at first. However, dyspnea and hemopstysis emerged within two days, although they were admitted to hospital immediately, it ultimately caused deaths after a few hours. The test of their blood samples showed Long was infected byIcterohaemorrhagiaeLeptospira, while the sample of Mr. Shi was severely hemolyzed, so the result cannot be detected.

The investigation among general population

In total 188 villagers were involved in our study, the gender ratio was 1∶1.89 (65/123, male to female). Among them, there were 123 famers (65.4%), 25 primary school students (13.3%), 13 self-employed (6.9%), 11 doctors (5.9%), 8 unemployed, five teachers and three manual workers. The subjects were grouped into famers, students, non-agricultural workers or the unemployed consequently. However, no one reported having receivingLeptospiralvaccine. The comparison of demographic valuables (which were significantly different between two areas) was listed in Table 1, which presented some discrepancy. However, self-evaluations of health condition in two areas were comparable.

The sero-epidemiological survey

In total 44 cases were tested positive by MAT and the positive rate was 23.4%. It showed that 37 samples were infected with one kind ofLeptospirastrains, while the other 7 cases were co-infected. The distribution of antibody against differentLeptospirastrains among positive cases are listed in Table 2. The infection spectrum in the two areas is illustrated in Figure 1.

Tab.2 Result of MAT Test

Typeofinfection*SingleinfectionCo-infectionS1S3S7S10S1+S7S1+S10S3+S10S7+S10+S1No.ofpositivecases200983211Percentages(%)45.5020.518.26.84.52.32.3

Note: *S1, S3, S7, S10 abbreviations forIcterohaemorrhagiaeSerovarLaiStrain56601,CanicolaSerovarCanicolaStrain56603,AustralisSerovarAustralisStrain56607 andHebdomandisSerovarHebdomandisStrain56610

S1,S7,S10 abbreviations for Icterohaemorrhagiae Serovar Lai Strain 56601, Australis Serovar Australis Strain 56607 and Hebdomandis Serovar Hebdomandis Strain 56610, co-infection includes infections with two or three strains.Fig.1 Spectrum of Leptospira infection in general population, by area

Positive rates of male and female were 26.2% and 22.0%, respectively, which showed no significant difference. It suggested age was a risk factor of infection, and there was a positive linear trend between age and positive rate (linear-by-linear associationP<0.05). The age distribution and age stratified MAT positive rate of participants were presented in Table 3.

The factors, such as area, vaccination history, occupation and self-evaluation of health condition were analyzed. Except for area, no significant difference was found within subgroups of other valuables. The positive rate of extended area was 53.3% (24/45) which is significantly higher than that of epidemic area (14.0%, 20/143). A similar trend was observed in the proportion of high titer (any strains ≥800 , Table 4). Self-report information showed 63.8% (113/177) of people having exposed to paddy field environment, while the MAT positive rate is not significantly different from those had not been exposed (P=0.360 by Pearson Chi-Square test).

Tab.3 Age-group and MAT result, by area

Age-groupEpidemicareaExtendedareaFrequencyNo.ofpositivecasesPositiverate(%)FrequencyNo.ofpositivecasesPositiverate(%)10-2528.0000.020-24416.711436.430-2015.09333.340-45613.34375.050-23417.48787.560-6350.013753.8Total1432014.0452453.3

Tab.4 Distribution of titers in different areas

Area(%)TitersofMAT-1∶501∶1001∶2001∶4001∶800Epidemicareapercentage9429329665.720.32.11.46.34.2Extendedareapercentage129845726.720.017.88.911.115.6

Result of rodent specimens

In total 71 rodents were captured and sampled. In total, four strains ofLeptospirawere isolated from 67 samples (three strains were isolated fromApodemusagrarius, one fromRattusrattussladeni), all belonged to serogroupIcterohaemorrhagiae. However, there was no difference in total germ-carrying rates of rodents in different areas (Fisher’s exactP=1.000) or between genders (Fisher’s exactP=1.000). The proportion ofApodemusagrariusin the captured rodents showed no significant difference compared with extended areas (P>0.05). Their germ-carrying rates share no difference in two areas, either (Table 5).

Tab.5 Distribution of Apodemus agrarius and germ-carrying rate of rodents in different areas

AreaNo.ofApodemusagrariusTotalno.ofrodentsProportionofApodemusagrarius(%)TestNegativePositiveGerm-carryingrateofApodemusagrarius(%)EpidemicArea113630.61019.1ExtendedArea163545.714212.5

Risk factors analysis

Area, age, sex, vaccination history and occupation were considered risk factors on statistical and clinical basis. The logistic regression analysis showed age and area were two significant factors. The risk of infection withLeptospiraincreased with age, and there was a higher positive rate (95% IC of OR 3.706-23.085) in general population in extended area than that of epidemic area.

Tab.6 Multivariate model of independent factors associate with leptospiral infection P value

PvalueOR95%CIforORLowerUpperAge0.0021.0651.0241.107Vaccinationhistory0.0981.1610.17527.307Area0.0009.2503.70623.085Gender0.0890.4830.2091.116Occupation0.197 Famer0.0380.3100.1020.940 Student0.8771.1610.1757.697 Unemployed0.8460.8280.1235.565

Discussion

Our study combined the geographic data of three sites, demographic information of subjects and their laboratory test results to describe distribution and risk factors in Liping County. Although on the stage of planning, we supposed the frequency of exposure to paddy fields, occupation and self-evaluated health condition were risk factors of leptospiral infection, which were identified in other researches[12-14], it turned out that there was no significant relation between these variables and population positive rate. Additionally, the positive rate was not statistically different between male and female participants, while the rate increased with age, and differed in the epidemic and the extended area. Based on these facts, it can be deduced that infectious factors were relevant to accumulation of time and location. Although only serogroupIcterohaemorrhagiaewas cultured from the rodent kidney samples, exposures of different strains of Leptospira commonly existed in the three villages to varying extents based on the findings of MAT among the general population. Preventive measures such as vaccination and health education campaigns should be taken among young people, so as to prevent them from the first infection withLeptospira. The sero-positive rate of Luoli villagers was obviously higher than that of other two villages, which indicated lepstospirosis infection was more common in Luoli although no cases were reported. The higher sero-positive rate and higher proportion of high titers may attribute to the hygienic habit (feces treatment, habits in farm work, etc.), more chance of exposure to environment contaminated by rodents (especiallyApodemusagrarius), high paddy coverage in Luoli village and its location between those two epidemic villages; people were prone to contraction of the antibody from previous unapparent or mild infections during frequent exposures[15], the titer of antibody would last for months or even years[16]. Immunologic memory mechanism will stimulate immune cells to produce plenty of antibody once organism was invaded by the pathogen again, which presents a high titer of antibody in serum. The antibody may prevent them from re-infection or severe acute symptoms. Building on these facts, it should be recognized that the fact that no cases have been reported does not mean the risk is low or relevant surveillance can be ignored. Conversely, the high positive rate among the general population stands for high risk of infection and implies high potential incidence, to which should be attached importance by hygiene system, and one of effective measures is surveillance in all susceptible areas. In epidemic areas, vaccination may prevent the onset of the disease or relieve the symptoms of severe cases, while adverse reactions, storage and target population are challenges of vaccine selection in the future[17]. Further research may be conducted to explore the target population of vaccination and epidemic strain ofLeptospira, and evaluate the cost-efficiency of population vaccination. The localized vaccination plan was recommended for various infection patterns in different places.

The present research in rodents showed that the local main germ-carrier wasApodemusagrarius. Other species likeRattusrattussladenimay occasionally carryLeptospira. SerogroupIcterohaemorrhagiaeStrainLaiwas isolated from rodents, but the results of MAT showed antibodies of four strains were found in human serum of all sites. This explained to an extent the source of theLeptospirainfection, and other strains (CanicolaStrain,AustralisStrainandHebdomandisStrain) ofLeptospiramay be related to other carriers, such as pigs, dogs, cattle, and so on[12-13,18], or due to limited sample size of captured rodents. According to Table 5 and Table 6,StrainLai(S1) was isolated from the samples ofApodemusagrariusin both areas, and the densities ofApodemusagrariuswere the same within Luoli Village and other two villages. Therefore, the role ofApodemusagrariusplays in leptospirosis epidemic process was critical. A previous study in the Pacific Islands[19]concluded the biomass of specific rat species influence the risk of humans being infected withLeptospira, which was not necessarily related to the total rodent density. Based on the thesis, it is reasonable to conclude that villagers of Luoli may have higher exposure risk ofLeptospirathough the germ-carrying rates showed no difference in two areas. Because Apodemus agrarius was likely to be found in Luoli, higher exposure probability was created. It was proved by higher proportion of human infection of S1 in Luoli (Figure 1). The aforementioned analysis makes it necessary to emphasize the importance of surveillance both in human and animal carriers for leptospiosis prediction, prevention and control.

Inevitably, the present study has three main limitations. Firstly, the information of factors was not completely explored. Social-economic variables such as income, education level and exposure chances of contaminated water body were not included in our study, which proved significant in the studies in Nicaragua[20], Brazil[21]and Asia-Pacific region[14]. Secondly, the sample sizes of both humans and rodents can support the facts in three sites, but they were not sufficient to represent the local distribution of leptospiral infection of Liping County. In regards to demographic valuables (gender, age, vaccination history and occupation), the target population of two areas showed not comparable; as a result, Logistic regression was applied to remove the bias attributed to these differences. Larger sample sizes and wider target areas were supposed to balance the distribution on these factors in further studies. Lastly, we chose to test four dominant strains ofLeptospiraamong the general population of Liping considering time-saving and cost-efficient factors, but it could not represent the whole infection spectrum of Leptospirosis. We assumed other strains were marginal amongLeptospirainfection. In further studies, more serotypes should be included to shed light on infection spectrum among the general population.

Long-term rodent carrier surveillance should be conducted in Liping County to provide valuable information and locate high risk areas, facilitating the preparation of preventive measures. Leptospiral infection background in general population suggested the infectious environment commonly exists and the youth should be protected and well informed with knowledge of leptospirosis prevention. Vaccination is recommended if economic feasibility and demand were proved.

Acknowledgement

We acknowledge the contribution of Liping County CDC for rodents trapping and townships public health centers of Luoli, Xuejiaping and Cenhu for population investigation.

Reference

[1] News O. Leptospirosis: an emerging public health problem[Z]. Geneva: WHO. 2011, 86(6): 45-50

[2] Li S, Wang D, Zhang C, et al. Source tracking of human leptospirosis: serotyping and genotyping ofLeptospiraisolated from rodents in the epidemic area of Guizhou province, China[J]. BMC Microbiol, 2013, 13: 75. DOI: 10.1186/1471-2180-13-75

[3] Durski KN, Jancloes M, Chowdhary T, et al. A global, multi-disciplinary, multi-sectorial initiative to combat leptospirosis[J]. GLEAN, 2014, 11(6): 6000-6008. DOI: 10.3390/ijerph110606000

[4] Gouveia EL, Metcalfe J, De Carvalho ALF, et al. Leptospirosis-associated severe pulmonary hemorrhagic syndrome, Salvador, Brazil[J]. Emerg Infect Dis, 2008, 14(3): 505-508. DOI: 10.3201/eid1403.071064

[5] Brinkerhoff RJ, Collinge SK, Ray C, et al. Typhoon-related leptospirosis and melioidosis, Taiwan, 2009[J]. Vectorborne Zoonotic Dis, 2011, 10(1): 47-52. DOI: 10.1089/vbz.2009.0044

[6] Bandara M, Ananda M, Wickramage K, et al. Globalization of leptospirosis through travel and migration[J]. Globalizat Hlth, 2014, 10: 61. DOI: 10.1186/s12992-014-0061-0

[7] Agampodi SB, Dahanayaka NJ, Bandaranayaka AK, et al. Regional differences of leptospirosis in Sri Lanka: Observations from a flood-associated outbreak in 2011 [J]. PLoS Negl Trop Dis, 2014, 8(1): e2626. DOI: 10.1371/journal.pntd.0002626

[8] Ke Y, Yongquan J, Yongping L. Epidemiology of leptospirosis in Liping county, Guizhou, 2001-2008[J]. Dis Surveill, 2009, 24(10): 768-769.

[9] Lu DQ, Zhou ZW. Sero-epidemiological survey of leptospirosis in Guizhou Province from 1991 to 1993[J]. Chin J Epidemiol, 1995, 16(1): 55.

[10] Shi MH, Tu QL. A survey of geographic distribution of leptospira in China[J]. Chin J Epidemiol, 1995, 16(5): 259-262.

[11] Diagnostic Criteris for Leptospirosis (WS290-2008)[Z]. 2008

[12] Johnson MAS, Smith H, Joeph P, et al. Environmental exposure and leptospirosis, Peru[J]. Emerg Infect Dis, 2004, 10(6): 1016-1022. DOI: 10.3201/eid1006.030660

[13] Keenan J, Ervin G, Aung M, et al. Risk factors for clinical leptospirosis from Western Jamaica[J]. Am J Trop Med Hyg, 2010, 83(3): 633-636. DOI: 10.4269/ajtmh.2010.09-0609

[14] Sakundarno M, Bertolatti D, Maycock B, et al. Risk factors for leptospirosis infection in humans and implications for public health intervention in Indonesia and the Asia-Pacific region[J]. Asia-Pacific J Public Hlth, 2014, 26(1): 15-32. DOI: 10.1177/1010539513498768

[15] Gancheva GI. Leptospirosis in elderly patients[J]. Brazilian J Infect Dis, 2013, 17(5): 592-595. DOI: 10.1016/j.bjid.2013.01.012

[16] Levett PN. Leptospirosis[J]. Clin Microbiol, 2001, 14(2): 296-326. DOI: 10.1128/CMR.14.2.296

[17] Xu FX. Current status and challenges ofLeptospiravaccine development in China[J]. Chin Pharmaceut Affairs, 2015, 29(7): 730-733.

[18] Khalili M, Sakhaee E, Aflatoonian MR, et al. Seroprevalence of bovine leptospiral antibodies by microscopic agglutination test in Southeast of Iran[J]. Asian Pacific J Trop Biomed, 2014, 4(5): 354-357. DOI: 10.12980/APJTB.4.2014C1206

[19] Theuerkauf J, Perez J, Taugamoa A, et al. Leptospirosis risk increases with changes in species composition of rat populations[J]. Die Naturwissenschaften, 2013, 100(4): 385-388. DOI: 10.1007/s00114-013-1033-6

[20] Bacallao J, Schneider MC, Najera P, et al. Socioeconomic factors and vulnerability to outbreaks of leptospirosis in Nicaragua[J]. Intl J Envir Res Public Hlth, 2014, 11(8): 8301-8318. DOI: 10.3390/ijerph110808301

[21] Felzemburgh RDM, Ribeiro GS, Costa F, et al. Prospective study of leptospirosis transmission in an urban slum community: role of poor environment in repeated exposures to theLeptospiraagent[J]. PLoS Neglected Trop Dis, 2014, 8(5): e2927. DOI: 10.1371/journal.pntd.0002927