Chronic kidney disease (CKD) is a major global public health problem [1
]. Diabetes mellitus and hypertension are the most common causes of CKD in developed countries. However, a severe form of CKD that has been reported in individuals without any association of these known risk factors is defined as CKD of unknown etiology (CKDu) [2
]. CKDu has emerged as a significant public health crisis in rural agricultural communities in Sri Lanka, among sugarcane farmers in Central American countries, and in some areas of India and Egypt [3
]. In Sri Lanka, the number of CKDu patients has been increasing rapidly since the 1990s, mainly in the dry zones of North-Central, Uva and North Western provinces. CKDu prevalence ranges from 15.1% to 22.9% in highly prevalent areas [4
Hantaviruses belong to the order Bunyavirales, family Hantaviridae, and genus Orthohantavirus and they are naturally maintained in rodent and other small mammal reservoirs [5
]. Several species of hantaviruses cause two important zoonoses: Hemorrhagic fever with renal syndrome (HFRS) in the Old World and hantavirus pulmonary syndrome (HPS) in the New World [6
]. In addition to pathogenic viruses, many hantaviruses are known to have low pathogenicity or no pathogenicity in humans. We recently reported high seroprevalence of hantaviruses among CKDu patients in Girandurukotte, one of the CKDu hotspots in Sri Lanka [7
]. Preliminary findings suggested that Thailand orthohantavirus (THAIV) or a THAIV-related virus was circulating in this area [8
]. THAIV was originally isolated in Thailand in 1985 [9
]. However, THAIV is not associated with classic HFRS in Asia, and its pathogenicity in humans has not been determined yet [10
]. Moreover, the relationship between hantavirus infection and CKDu in Sri Lanka has not been evaluated, as there were no comparative analyses between a CKDu endemic and a non-endemic area in previous studies.
In this study, we conducted a cross-sectional study followed by an unmatched case-control comparison to identify risk factors of exposure to THAIV among renal disease patients and healthy controls in two geographically distinct areas: endemic area (Girandurukotte) and a non-endemic area (Kandy).
This is the first epidemiological study addressing hantavirus infection in renal disease patients in a CKDu endemic and a non-endemic area in Sri Lanka. A higher seroprevalence of hantavirus was found in kidney disease patients than in healthy controls in a CKDu endemic area, as reported in recent studies [7
Girandurukotte, one of the CKDu hotspots, had a greater proportion of exposed individuals than Kandy. The study results provide interesting insights for the hypothesis of exposure to hantavirus as a possible risk factor for the development of CKDu in Sri Lanka [25
]. Clinical records showed that 90 of the 104 Girandurukotte renal disease patients had diagnosed CKDu cases according to the 2009 guideline [22
]. Significant associations were found between renal disease and hantavirus exposure in both Girandurukotte and Kandy. The majority of seropositive renal disease patients in Girandurukotte were CKDu patients, but none of the renal disease patients in Kandy were diagnosed with CKDu. Furthermore, the nine seropositive renal disease patients in Kandy included seven CKD patients, one end-stage renal disease patient, and one acute kidney injury (AKI) patient. The diagnosis of CKDu in Sri Lanka is made according to guidelines that are solely based on clinical parameters (GFR levels). These results confirm the existence of sporadic hantavirus infection even in Kandy after the first report by Gamage et al. in 2011 [12
In Girandurukotte, the majority of residents are engaged in agriculture such as paddy farming. In contrast, fewer people are engaged in paddy farming in Kandy than in Girandurukotte. In this study, we showed that age of 40 years or older, male gender, paddy farming, storing crops at home, and witnessing rodents in or surroundings of the house were high-risk factors for hantavirus infection in CKDu endemic areas. Moreover, the presence of anti-hantavirus antibodies was the most significant risk factor among renal disease patients. In Kandy, individuals in the exposed group were older than those in the non-exposure group and were more likely to be male, paddy farmers, and to have sighted rats. However, storing crops at home was not found to be a risk factor for exposure to hantavirus in Kandy. This finding suggested that hantavirus infection in this area might not cause by rodents invading the house and suggests the rationale of lower seroprevalence in females than in males.
Renal disease patients in both areas tended to be older than the controls and there was a larger percentage of males than that in the control groups. However, our results were likely to have been affected by biased information provided by the participants with regard to interactions with rodents and their excreta during the survey-based interview. Despite these constraints, this study provides empirical information of exposure to hantavirus as a risk factor for renal disease in a CKDu endemic area.
In 2016, we reported high seropositivities in the CKDu patients living in Girandurukotte [7
]. By using their sera, we established a reliable serodiagnostic method for this study. These sera were also applied to ICG by using HS103 antigen as mentioned in a previous report [26
]. They showed varied reactivities. Some showed relatively strong bands, whereas others showed faint or no bands (data not shown). Although ICG tests are considered to be the best method for field surveys, they require high affinity and avidity of antigen-antibody binding. In order to develop ICG method specific to Sri Lanka, it is necessary to use an antigen derived from an orthohantavirus specific to Sri Lanka.
In Kandy, which is considered to be a non-endemic area, seroprevalences in renal disease patients and controls (18.0% and 7.0%, respectively) were lower than those in Girandurukotte. Furthermore, in this study, the rate of seroprevalence for hantavirus in the apparently healthy population was higher than that reported elsewhere. The seroprevalence of THAIV among healthy controls in Thailand has not been reported yet. However, seroprevalence of 0.4% (1/260) for THAIV was found in febrile patients in Surin Province of Thailand who were suspected of having leptospirosis and were serologically negative for Leptospira antigens [10
]. In Chile, the seroprevalences for Andes orthohantavirus in apparently healthy individuals were reported to be 1.3% and 1.5% in rural and urban poor communities, respectively [27
]. Similarly, a cross-sectional survey was carried out in an endemic area of the state of Minas Gerais, a peri-urban and rural area in Brazil, to assess the proportion of persons exposed to hantaviruses. Antibody positivity was determined to be 3% by ELISA using a recombinant antigen of Araraquara orthohantavirus associated with HPS [28
], whereas the results of the current study showed a seropositive rate of 17.4% in controls living in rural agricultural communities of Girandurukotte and a seropositive rate of 7% in controls living in peri-urban Kandy.
The recent study showed that the innate immune response was activated in CKDu patients, including interferon, inflammasome and triggering receptor expressed in myeloid cell-1 (TREM1) signaling. Transcriptome analysis suggests that viral infections and fluoride or other toxic environmental factors appear to be contributing to the molecular mechanisms underlying the development of CKDu in Sri Lanka [29
]. However, the causes of CKDu have remained unknown for more than 20 years. The results of this study showed that exposure to hantavirus infection is a possible risk factor underlying renal disease in both endemic and non-endemic areas. Antibodies detected in these individuals were associated with a lack of past clinical symptoms or a medical history of being exposed to hantaviruses, suggesting that infection with THAIV-related hantavirus might be an asymptomatic or a mild infection with nonspecific clinical symptoms at the onset of the disease.
Several studies carried out in Europe has revealed pathogenesis related to mild to moderate forms of kidney damage during several hantavirus infections. In Northern Germany, Dobrava-Belgrade virus (DOBV) infected patients showed a diverse range of mild to moderate clinical manifestations associated with kidney injury [32
]. In another study, patients with nephropathia epidemica (NE) often referred as a mild form HFRS caused by Puumala virus exhibited upregulation of biomarkers of inflammation suggested damage in kidney proximal tubule and leukocyte chemotaxis at the onset of NE in a cohort of patients in a Russian hospital [33
]. On the other hand, CKDu in Sri Lanka is known to develop from tubular damage [34
]. Several Old World hantaviruses have been reported to cause AKI but not HFRS [36
]. Furthermore, it has been reported that chronic proteinuria continues even several years after recovery from hantavirus infection [32
]. Thus, there are similarities between CKDu in Sri Lanka and hantavirus infection. In addition to the epidemiological data obtained in this study, it is necessary to clarify the pathological association between CKDu and hantavirus infection.
Thus, there is a necessity for island-wide sero-epidemiological studies on exposure to hantavirus using both retrospective (renal disease patients) and prospective (nonspecific febrile patients) approaches. Future directions of this study are focused on identifying the reservoir host carrying the THAIV-related hantavirus in these CKDu endemic areas and understanding the dynamics of its transmission to humans. The pathological and molecular biological mechanisms underlying the progression of renal damage due to circulating THAIV-related hantaviruses should also be investigated to understand the disease dynamics.