Short Communication
Chronic kidney disease of unknown etiology (CKDu): Using a system dynamics model to conceptualize the multiple environmental causative pathways of the epidemic

https://doi.org/10.1016/j.scitotenv.2019.135766Get rights and content

Highlights

  • There is a global of epidemic chronic kidney disease of unknown etiology (CKDu)

  • High prevalence rates are seen in agricultural communities of the tropics

  • Its causation is controversial, partly due to the reductionist scientific paradigm

  • Multiple etiological pathways appear to contribute to the epidemic

  • A system dynamics model enables better understanding of these pathways

Abstract

A chronic kidney disease of unknown etiology is devastating agricultural communities of Sri Lanka, Central America, areas of India, and Egypt. Researchers are yet to agree on its etio-pathogenesis despite many cross-sectional, case-control and cohort studies done in these countries. These approaches are broadly based on a reductionist approach. We propose a complementary paradigm based on complexity science to deepen our understanding of the disease. Complexity science views a population as system that has several dynamically interacting and inter-dependent sub-systems and is ‘open’ to the ‘outer’ environment. Health outcomes or epidemics are viewed as ‘emergent’ properties of the population. Using available literature from Sri Lanka, the paper presents a system dynamics model incorporating exposures from pesticides and heavy metals, drinking hard water with high levels of fluoride, poverty, low birth weight, micronutrient deficiencies and heat stress. This approach can be used to model the epidemic, understand the impacts of different factors, predict potential populations at risk, and formulate multi-pronged prevention strategies that target leverage points of the system.

Introduction

Communities living in rural areas of several countries are being devastated by an epidemic of chronic kidney disease of unknown etiology (CKDu) (Gifford et al., 2017; Glaser et al., 2016). The pathology is characterized by a non-specific chronic interstitial nephritis with several synonyms based on geographic locations: Salvadoran Agricultural Nephropathy, Mesoamerican Epidemic Nephropathy, Udhanam Endemic Nephropathy, Chronic Tubular-Interstitial Disease from Central America. Terms such as Chronic Interstitial Nephritis in Agricultural Communities (CINAC), heat stress nephropathy and chronic agrochemical nephropathy were coined due to its high prevalence in agricultural communities which use agrochemicals extensively and proximity of the epidemic to equatorial regions affected by heat stress (Glaser et al., 2016; Jayasinghe, 2014; Oxford Textbook of Clinical Nephrology, 2018).

However, there is considerable degree of uncertainty on its etiology as evidenced by this proliferation of terms. This in turn is likely to be result of the non-specific histology which merely indicates an environmental toxic nephropathy (Ordunez et al., 2014). The uncertainty is probably encouraged by the presence of an uneven mosaic pattern in the distribution of cases in the highly endemic areas. There appears to be clusters of high prevalence, yet within the same neighborhood there is relative sparing of certain households (Oxford Textbook of Clinical Nephrology, 2018).

Scientists disagree on the results and interpretation of studies done in the same country (Ordunez et al., 2014; Lunyera et al., 2016). An example is from Sri Lanka, where cross-sectional, case-control and cohort studies have implicated agrochemicals (Gifford et al., 2017; Wanigasuriya et al., 2007) excess fluoride in water (Chandrajith et al., 2011), and food contaminated by cadmium, arsenic, lead, and chromium, individually or in combination (Kulathunga et al., 2019; Jayatilake et al., 2013). One detailed study analyzed for heavy metals in drinking water, rice and hair, and investigated renal tissues for electron microscopic evidence of trace elements, and almost ruled out cadmium and arsenic (Nanayakkara et al., 2019a). In contrast, a more recent study found that the diet used in the region has a high intake of lead and cadmium (Ananda Jayalal et al., 2019). Other hypotheses include exposure to glyphosate (Jayasumana et al., 2015), increased ionic strength of ground water (Dharma-Wardana et al., 2015) and a combination of fluoride, cadmium and hardness of water (Wasana et al., 2017). In Central America, suspects include pesticides, and a combination of heat stress from climate change and intake of fructose rich drinks (Glaser et al., 2016). More recently poverty, malnutrition and population-wide exposures to contaminated water have been implicated, particularly in southern Asia (Orantes-navarro et al., 2017; Abraham et al., 2019).

In India organochlorines are proposed as an etiology and a recent report hypothesizes a potential role for persisting DDT in the environment in Sri Lanka (Singh et al., 2017; Jayasinghe et al., 2018). Some reports have seriously contested individual hypotheses such as the heat stress as an etiology of the disease (Herath et al., 2018).

The reductionist research methodologies used to investigate the epidemic find it problematic to explain two observations: the unusual mosaic distribution pattern and the continuing contradictory results on causative factors. However, there is increasing interest to apply systems science approaches to conceptualize chronic disorders in populations so as to complement to this conventional paradigm (Jayasinghe, 2011; Jayasinghe and Jayasinghe, 2011). The paper develops a model on the causation of CKDu that draws on our knowledge in population-wide exposures and systems theories, using Sri Lanka as an example.

Highly endemic populations to CKDu in Sri Lanka demonstrate population-wide exposures to several environmental factors, contaminants in food and biological predispositions. Systems science views such populations as complex adaptive systems and patterns of diseases as an emergent property (Jayasinghe, 2011). The etiological factors or determinants are subsystems that form networks with each other and create feedback loops. Their interactions synergize or retard each other's impacts and affect population health or human organs in a non-linear manner.

System dynamics, one of the modeling techniques used in systems science, is used to conceptualize the complex environmental causative pathways in the CKDu epidemic (see figure). In this approach model, the environmental factors affecting the kidneys are shown in the diagrammatic abstract.

We begin with the central horizontal pathway (from ‘low renal mass’ to ‘rapid decline in GFR’) that illustrates the susceptibility of kidneys with low renal mass to several factors (depicted using arrows in the vertical axis). The low renal mass in turn is due to malnutrition from poverty and other determinants of fetal growth retardation, such as sub-clinical effects of the maternal CKDu (i.e. a trans-generational effect) or unknown environmental contaminants. The regions with a high prevalence of CKDu had high rates of malnutrition and low birth weight (LBW) during the period the present cohort of adult patients were born (i.e. 1990s). LBW was 14.8% to 19.9% (Sri Lanka Demographic and Health Survey 1993, 1995). Evidence for malnutrition includes a high proportion of stunting seen in 3 to 36-month age group (ranging from 21.7% to 23.7%). A population with LBW will grow to adulthood with lesser numbers of glomeruli and lower renal mass that increases their susceptibility to damage from external stressors (Luyckx and Brenner, 2010).

The external stressors or factors, shown as vertical arrows of influence, includes of drinking hard water with high fluoride concentrations and excess iconicity especially from tube wells used in this region has hard water (Chandrajith et al., 2011; Nanayakkara et al., 2019a). Recent studies of ground water have confirmed these findings with a variation in water quality depending on season and geographic locations (Cooray et al., 2019). There is a higher risk of dehydration due to the high ambient temperature and lack of easy access to drinking water (Glaser et al., 2016; Nanayakkara et al., 2019b).

The intense use of pesticide and fertilizer in agriculture in the country will lead to mass exposures from pesticides and heavy metals (Gifford et al., 2017; Marasinghe et al., 2011). Sri Lanka is one of the highest per-capita users of pesticide in the world and had used DDT extensively in its campaign against malaria (Herath et al., 2018). Recent experience from Europe suggest the ubiquitous nature of pesticide exposures in wide geographic areas, despite much stricter controls over its use (Silva et al., 2019). Two other factors that may have an impact on the renal functions of this population are the high rates of snake bite and the use of nephrotoxic drugs herbal preparations or non-steroidal anti-inflammatory drugs (NSAIDs).

The exposures described above are known to interact with other factors to heighten their individual adverse impacts. An example is the high prevalence of vitamin D and zinc deficiency in these areas (Abeywickrama et al., 2018). These deficiencies facilitate intestinal absorption of heavy metals (e.g. cadmium) and promote their toxicity (Abeywickrama et al., 2018). Furthermore, exposure from multiple heavy metals (e.g. cadmium and lead) has a synergistic effect on nephrotoxicity that is larger than a simple additive effect (Mauderly and Jonathan, 2009). Similar synergies are also reported from cadmium, fluoride content and hardness of water in animal models of CKDu (Jayasumana et al., 2015). Heat stress could directly affect the kidneys having a low reserve and increase its susceptibility to a cocktail of toxins (Glaser et al., 2016).

The distribution of these adverse factors will be uneven across geographic areas, communities and individuals (e.g. spatial differences in fluoride concentrations of drinking water, individual differences in pesticide exposures, variations in intensities of heat stress exposure). They interact with each other as described above, and these interactions are non-linear with feedback loops, enabling adaptation and self-organization of the system. One example of a feedback loop is the early CKDu affecting the mother which in turn leads to fetal growth retardation and LBW. Another feedback loop is the adverse socio-economic impact of CKDu and its contribution towards worsening of poverty: A chronic illness such as CKDu, affects households from expenditures due illness and decreased productivity from ill-health.

These webs of causations lead to the emergence of a CKDu epidemic that has a geographic pattern of prevalence much like a mosaic (Singh et al., 2017; Jayasinghe, 2015). Such patterns are poorly recognized and detected or described by statistical techniques such as Chi-square test and odds ratio, which assume the presence of independent factors influencing outcomes in a linear fashion, i.e. a mechanistic-reductionist paradigm of causality (Singh et al., 2017; Jayasinghe and Jayasinghe, 2011). They are less effective in describing and analyzing epidemiology of chronic diseases that arise from multiple interacting factors with feedback loops. Furthermore, cross-section and case-control studies assume the presence of discrete exposure vs. non-exposure groups, an unlikely situation in the case of CKDu. As Geoffrey Rose aptly stated “If every member of society smoked 20 cigarettes a day and the incidence of lung cancer was investigated through cohort studies and case-control studies and standard statistical methods, the conclusion would be that incidence of lung cancer is determined by genetic predispositions, rather than smoking (Rose, 1985).

Quantifying the interactions between the different factors or subsystems require methods such as systems dynamics modeling, rather than traditional statistical analyses. These modeling approaches will help to unravel etiologies and understand the epidemic better (Carey et al., 2015). They will enable policymakers to model different scenarios based on a range of interventions and predict potential populations at risk. This helps to formulate multi-stakeholder prevention strategies targeting leverage points of the entire system, rather than fragmented approaches that focus on a single intervention, as recommended in many studies. We believe such a systems approach and modeling can be extended to other countries to understand the multiple environmental causative pathways underling the global epidemic of CKDu. The world of CKDu research should look forward to this new paradigm of thinking.

Section snippets

Conclusion

The paper proposes a complexity science-based approach to deepen our understanding of CKDu. Using data from Sri Lanka, a system dynamics model is described. Exploring this approach may help to model the epidemic, understand the impacts of different factors, predict hot spot, and formulate multi-pronged prevention strategies that target leverage points of the system.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

None.

References (33)

  • H.M. Abeywickrama et al.

    Micronutrient status in Sri Lanka: a review

    Nutrients

    (2018)
  • T.B. Ananda Jayalal et al.

    A quantitative analysis of chronic exposure of selected heavy metals in a model diet in a CKD hotspot in Sri Lanka

    BMC Nephrol.

    (2019)
  • G. Carey et al.

    Systems science and systems thinking for public health: a systematic review of the field

    BMJ Open

    (2015)
  • T. Cooray et al.

    Assessment of groundwater quality in CKDu affected areas of Sri Lanka: implications for drinking water treatment

    Int. J. Environ. Res. Public Health

    (2019)
  • M.W.C. Dharma-Wardana et al.

    Chronic kidney disease of unknown etiology and ground-water ionicity: study based on Sri Lanka

    Environ. Geochem. Health

    (2015)
  • J. Glaser et al.

    Climate change and the emergent epidemic of CKD from heat stress in rural communities: the case for heat stress nephropathy

    Clin. J. Am. Soc. Nephrol.

    (2016)
  • Cited by (13)

    • A review of molecular mechanisms linked to potential renal injury agents in tropical rural farming communities

      2022, Environmental Toxicology and Pharmacology
      Citation Excerpt :

      Revealing these mechanisms may also become useful in identifying potential therapeutic targets. Currently, to the best of our knowledge, there are no detailed examination of networks of biological pathways involved in etiology of CKDu (Jayasinghe and Zhu, 2020). This review addresses this knowledge gap by examining the molecular mechanisms associated with renal injury following exposure to chemical contaminants and heat stress associated with dehydration.

    • Numerical and experimental investigation of a DC-powered RO system for Sri-Lankan villages

      2022, Renewable Energy
      Citation Excerpt :

      Chronic kidney disease of unknown etiology (CKDu) is a global health concern that impacts many rural, economically developing regions around the world [1], including the north central region of Sri Lanka, coastal regions of Central America, and Andra Pradesh in India [2,3].

    • System dynamics simulation for park-wide environmental pollution liability insurance

      2021, Resources, Conservation and Recycling
      Citation Excerpt :

      Tsai et al. (2019) developed a SD model to explore the potential for biofuel commercialization in Taiwan considering limited biofuel resources. SD can also be used to model epidemics (Saroj et al. 2020) and pandemics (Sy et al., 2020). John et al. (2019) discussed the potential of SD modeling to facilitate understanding of cumulative health risks.

    • Profiles of antibiotic resistome and microbial community in groundwater of CKDu prevalence zones in Sri Lanka

      2021, Journal of Hazardous Materials
      Citation Excerpt :

      In 2017, antibiotic resistance was listed as an emerging issue of the frontier environmental concern by United Nations Environment Programme (UNEP, Frontiers, 2017), and it was also estimated that annual ARGs related global deaths would increase from about 700 000 to 10 million by 2050 where effective antibiotics no longer existed (Roope et al., 2019). Meanwhile, an increasingly serious global health concern was emerging as the chronic kidney disease of unknown etiology (CKDu) in agricultural communities of Sri Lanka, Central America, areas of India, and Egypt (Jayasinghe and Zhu, 2020), which presented as a kidney disease in patients who did not exhibit common causative factors, such as diabetes or hypertension (Cooray et al., 2019a). The CKDu was generally prevalent among agricultural workers, and an increasing number of Sri Lankans suffered from CKDu since the 1990s, e.g., more than 50,000 patients diagnosed with late-stage kidney disease (Makehelwala et al., 2020).

    • Letter to the Editor re: Jayasinghe and Zhu (2020)

      2020, Science of the Total Environment
    View all citing articles on Scopus
    View full text