Global dimensions of chronic kidney disease of unknown etiology (CKDu): a modern era environmental and/or occupational nephropathy?

The historical outbreak that is most similar to CKDu in terms of mortality occurred in Queensland, Australia residents who survived lead poisoning as children but went on to die of ESRD as adults [14]. Lead paint was used in Queensland from 1890 until it was linked to lead poisoning in children and phased out, starting with a ban in 1922 [14]. Children were exposed through play on painted verandas and railings of raised houses; a type of housing unique to Queensland [14]. As shown in the Fig. 1 below, an epidemiological investigation of mortality from chronic nephritis revealed a dramatic spike in deaths starting in 1905 and peaking in the early 1930s [15].

In autopsy data, the content of lead in skull bone was higher in those who were born in Queensland and died in their third to fifth decade from ESRD of unknown cause with granular contracted kidneys compared to those who died of ESRD from known causes, such as chronic glomerulonephritis, and those who died from non-renal causes [16, 17]. Furthermore, levels of chelatable lead in urine were also higher in patients with CKD consistent with lead nephropathy [18].

However, in studies of adults who were lead poisoned as children in other geographic areas, increased mortality from ESRD and/or disease severe enough to require dialysis was uncommon. Only persistent partial Fanconi syndrome [19] and a few cases of kidney disease consistent with lead nephropathy have been reported [20, 21]. Differences in the extent of exposure may be involved since the Queensland children were not treated with chelation therapy. Researchers in the Queensland outbreak also implicated the additional impact of concentrated urine in children in this warm climate [16].

Higher ESRD mortality or incidence rates have also been reported in lead workers, most of whom were likely very highly exposed [22, 23]. However, the number of workers with a nonmalignant renal cause of death was still small [24] compared to residents in Queensland and in areas with CKDu. Furthermore, blood lead is easy to measure and elevated levels have not been detected in Sri Lanka [3] or in Central America [25]. Thus, lead exposure is not a likely explanation for CKDu.

Excessive cadmium exposure has occurred in several areas in Japan but is best known as the cause of Itai-itai (“ouch-ouch”) disease in the Jinzu River basin of Toyama prefecture. Ingestion of rice irrigated with industrially polluted water resulted in high level exposure to cadmium. One study reported mean urine cadmium levels of 25.6 and 36.7 μg/g creatinine in 34 males and 38 females, respectively [26]. High levels were also observed in other studies [27]. To put these levels in context, the geometric mean urine cadmium in the U.S. National Health and Nutrition Survey was 0.25 μg/g creatinine in the 2007–2008 survey [28]. The most obvious adverse health effects from the Japanese exposure, reflected in the disease name, were osteomalacia and osteoporosis with secondary fractures and severe pain. Older women were most affected; this is attributed to their poor nutritional status, especially following World War II, since iron deficiency increases cadmium absorption. Proximal tubule damage was also a common feature of this disease and decreased creatinine clearance was reported in affected residents in the Jinzu river basin and other cadmium polluted areas in Japan [26, 29, 30]. However, despite high levels of exposure based on cadmium measurement in urine and rice, kidney disease severe enough to result in early death was not commonly reported. Four clinically identified deaths from uremia in Itai-itai patients have been described in the literature [30]. Studies among inhabitants living in the cadmium polluted area of the Kakehashi river basin observed increased mortality from nephritis and nephrosis, also based on small numbers [31].

In a more recent environmental exposure in the Mae Sot district of Thailand, evidence of proximal tubular damage is present but no outbreak of ESRD requiring dialysis or resulting in death has been reported in the affected population with mean urine cadmium levels of 5 μg/g creatinine [32]. A general population study in Belgium reported an association between higher blood lead and lower creatinine clearance; no associations with this outcome were observed with urine or blood cadmium [33]. A similar situation was observed in occupational exposure. Even in exposure conditions considered extremely high by current standards, tubular damage and CKD occurred but were rarely severe enough to progress to ESRD [34, 35].

The recently reported values of 1.04 and 0.65 μg/g creatinine in participants with CKDu and controls in the Sri Lanka endemic area, respectively, were much lower than in any of the studies described above [3]. Thus, it is unlikely that cadmium is the sole cause of CKDu in Sri Lanka.

Aristolochic acid, a naturally occurring nephrotoxic compound found in plants of the genus Aristolochiaceae, has been implicated as the cause of interstitial nephritis from ingestion of Chinese herbal medications used for weight loss that were contaminated with this compound [13]. Aristolochic acid has also been implicated in Balkan endemic nephropathy via plants growing around wheat fields in affected areas [13]. CKD from aristolochic acid ingestion does progress to ESRD, within 2–3 years in the case of herb ingestion. However, both aristolochic acid nephropathy and Balkan endemic nephropathy are associated with increased risk for urothelial carcinoma which has not been reported in CKDu to date. Further, data to date, although limited, do not indicate greater exposure to this chemical in the affected areas compared to non-endemic surrounding areas [36, 37].

Elevated standardized mortality rates for kidney disease have been reported in ecologic studies of communities with moderate to high levels of arsenic in their water sources [38]. Reduction in mortality after cessation of exposure in a previously endemic area in Taiwan was also observed [39]. A recent longitudinal study reported an increased risk for incident CKD with baseline urine arsenic concentrations [40]. However, similar to the situation with cadmium, reported arsenic levels in water [3] were much lower in Sri Lanka than in the communities where increased risk was reported in the ecologic studies. Furthermore, mortality in CKDu is so high that it does not require a mortality study to detect it.

Other exposure-related outbreaks have involved acute kidney injury. Ingestion of formula adulterated with melamine resulted in an outbreak in Chinese infants due to urinary tract obstruction from kidney stones. The long term implications of this exposure remain a research focus [41]. In a much smaller outbreak, diethylene glycol contaminated acetaminophen syrup caused acute kidney injury in Haitian children. The case fatality rate was extremely high (88 of 109 exposed children) [42]. However, the acute nature of these outbreaks is quite different than the pattern described to date in CKDu. Certain medications can also cause or contribute to CKD. Analgesic nephropathy is one of the best known examples of this; also requiring substantial research to make the connection.

Considering the CKDu characteristics reported globally and the environmental causes of CKD outbreaks identified to date, none of the previously established causes alone appears to explain CKDu. Thus, this disease entity is either multifactorial and/or due to a previously unrecognized cause of kidney damage. Two reviews have discussed the range of risk factors under consideration, in Central America [37] and globally [6]. As noted above, different risk factors may be involved in the various locations in which CKDu has been reported. Ethnic diversity may also be a factor. For example, South Asians have an increased risk for coronary artery disease that is not explained by traditional cardiac risk factors [43]. Given the similarities between cardiac and renal risk factors, this risk may be relevant for CKDu in India and Sri Lanka but not necessarily related to CKDu in Central America. An additional challenge for etiologic determination is that the majority of countries report at least some baseline rate of CKD that appears unrelated to traditional risk factors and thus is considered to be of unknown etiology [2]. As noted by Rajapurkar and colleagues, disease in which the cause could not be determined was the second most common CKD presentation in India and was observed throughout the country [44]. They note that delayed presentation in areas with limited access to healthcare makes establishing a primary diagnosis very challenging. Thus, there is likely some proportion of disease that is due to lack of diagnostic capability within the category of disease not due to any common risk factor that actually represents a novel type of CKDu.

Extreme physical exertion, heat stress, water quality and exposure to agrochemicals are among the potential causes currently being considered for CKDu. The combination of heat stress and physically demanding occupations has received attention in the Central American outbreak based on studies reporting lower rates of CKDu at higher altitudes in the same agricultural processes [45]. A recent pilot study in Brazilian sugarcane harvesters observed evidence of acute kidney injury over the course of a workday [46] The impact of repeated episodes of heat-induced dehydration has been examined in an animal model to specifically address this cause [47]. Pathology consistent with CKDu, including elevated serum creatinine, proximal tubular injury, and renal inflammation and fibrosis, was observed. Interestingly, this pathology was not observed in fructokinase deficient mice. Fructose containing drinks in combination with heat-induced dehydration was implicated and intervention trials based on the United States’ Occupational Safety and Health Administration (OSHA) Heat Illness Prevention Campaign were proposed [48]. In considering this potential etiologic factor, one must consider whether agricultural work has changed within the time frame of the CKDu epidemic. The industrialization of animal husbandry in the form of concentrated animal feeding operations has dramatically changed that field. Has the same magnitude of change occurred for agricultural workers? Work in high levels of ambient heat was common in a recent study of working conditions in Central America [49]. However, in a recent spatial distribution analysis, high temperature did not explain CKDu occurrence after adjustment for area under cultivation for specific crops [50]. Given the increased reliance on agrochemicals in modern agriculture, direct exposure and water contamination remain important considerations as well.

A form of CKD that appears unrelated to traditional risk factors, such as diabetes and hypertension, is responsible for widespread morbidity and mortality in specific geographic locations across three continents. The cause(s) remains unknown; however none of the occupational or environmental exposures that have been implicated in previous outbreaks of nephropathy have been established as sole risk factors in the current outbreaks. Thus, multifactorial or novel risks must be considered. Addressing similarities and differences in CKDu reported in different geographic areas globally may be of value in unraveling the cause(s) of this severe form of kidney disease.

Future needs include funding for clinical care of affected populations and for continued etiologic research. CKDu mortality is extremely high, reflecting health care limitations in the low income countries impacted by this disease. Therefore, funding for patient care is urgently needed. Funding for preventive medical care to identify kidney disease at earlier, more treatable stages is also necessary. In addition, funding is needed for basic public health measures that are known to be cost effective and also address potential CKDu risk factors. Examples include water sources that are free of biological and chemical contaminants; work practices that prevent dehydration in workers; and use of accepted safety procedures for agrochemicals. Implementation of these measures will benefit the population overall and may reduce the risk of CKDu since, based on experience with Balkan nephropathy, establishing the etiology may be a prolonged process.

In terms of research funding needs, in addition to continued etiologic studies, communication between research groups in the different geographic areas where CKDu has been reported would be useful to consider similarities and differences in the disease in each location as well as consolidate research approaches when possible. This would allow development of an internationally accepted definition of CKDu and kidney biopsy pathology criteria. Conference funding would be very valuable in this regard. Given the extremely high mortality and morbidity reported with this disease, efforts to identify the cause(s), prevent future cases and provide care for those affected must be a global priority.