Prevalence of chronic kidney disease among severely gas-exposed survivors in Bhopal, India

INTRODUCTION

Chronic kidney disease (CKD) has emerged as one of the leading causes of global morbidity and mortality with its global prevalence ranging from 8% to 10%.^1–6 The global burden of disease study estimated a 30% increase in CKD prevalence in 2020 compared to 1990.⁴ Despite being a global concern, CKD disproportionately affects people from lower-middle-income countries.⁵ In India, the prevalence of CKD has increased to epidemic proportions and population-based studies have reported a 4%–20% prevalence of CKD in India.^2,5–10

CKD in its early stages is considered as one of the major risk factors for fatal and non-fatal cardiovascular events. When it reaches its last stage, also known as end-stage renal disease (ESRD), the financial burden of treatment through dialysis and renal replacement therapy is enormous.^4,5,7,10 In resource-scarce countries such as India, <10% of patients with ESRD have access to any kind of renal replacement therapy.^2,10 Hence, it is appropriate that efforts are focused on prevention rather than treatment.

The Bhopal gas tragedy in December 1984, considered among the worst industrial disasters in the history of humankind, resulted in mortality of 2500–6000 and debilitating over 200 000 people, causing major morbidity and many premature deaths.^11,12 Various clinical and epidemiological studies undertaken subsequently showed a higher prevalence of chronic illnesses such as pulmonary fibrosis, bronchial asthma, chronic obstructive pulmonary disease, keratopathy and corneal opacities in the exposed population.^11,13,14 The Indian Council of Medical Research (ICMR) launched a long-term, population-based epidemiological study in January 1985 to assess the long-term health effects of toxic gases on a cohort of exposed people assembled according to surrogate exposure intensity, i.e. severely, moderately and mildly exposed and this cohort is still being followed up. Although animal studies show acute histopathological changes in renal epithelial cells upon exposure to methyl isocyanate,^11,15 there is a dearth of population-based studies to show the extent of renal disease in the gas-exposed survivors.

Concern has been repeatedly expressed by the gas-exposed survivors and several civil society groups that the prevalence of kidney-related ailments is too high among this vulnerable group as a consequence of the ill-effects of gas exposure in 1984. To address this concern, we did a cross-sectional study to estimate the prevalence of CKD among individuals belonging to the severely gas-exposed cohort of the ongoing long-term, population-based epidemiological study in Bhopal.

METHODS

RESULTS

A total of 215 gas-exposed survivors belonging to the severely exposed cohort of the long-term population-based epidemiological study were assessed for the presence of CKD. The mean (SD) age of the participants was 57.5 (12.07) years with higher participation (58.6%) of females. Nearly half the participants (50.2%) belonged to the 50–65 years age group. About 54.9% of participants did not have any formal education, whereas 30.2% of participants had completed primary level education. About 18% of participants were unemployed, 50.2% were homemakers and 71. 2% were non-vegetarian. Tobacco chewing was the most common (37.2%) substance use among the study participants (Table I).

The prevalence of CKD was found to be 16.7% (95% CI 12.2– 22.2). The mean (SD) e-GFR of the participants was 75.4 (17.9) with 14.4% of participants belonging to the G3a category of eGFR (Fig. 1). Proteinuria was present in 2.7% (6/215) and diabetes and hypertension were present in 20% (43/215) and 56.2% (121/215), respectively.

Close

FIG 1.

Distribution of the study population according to their estimated glomerular filtration rate category given by the National Kidney Foundation Kidney Disease Outcomes Quality Initiative (n=215)

Export to PPT

The mean (SD) age of the participants diagnosed as having CKD (n=36) was 64.2 (13.9) years and their mean e-GFR was 53.4 (10.9). Three-fourths (75%, 27/36) of the participants with CKD had hypertension, 30.6% (11/36) had diabetes, 27.8% (10/36) were overweight and 8.3% (3/36) were obese. Of the 36 participants with CKD, 32 did not have any history of renal disease and were newly diagnosed with CKD.

Univariate analysis revealed that factors such as gender, age, educational status, occupation, BMI and hypertension were significantly associated with CKD (Table II). To find out the individual effect of these factors on CKD after adjusting for other confounders, multivariate analysis was done. Factors with p<0.1 were included for multivariate analysis. This showed BMI and level of education to be independent predictors of CKD in our study population (Table III). Being literate reduced the risk of development of CKD by 65% (adjusted OR=0.347, 95% CI=0.13–0.93) when compared with illiterate. A high BMI increased the likelihood of development of CKD as there was a 3.6 times higher risk among overweight participants and 9.3 times higher risk among obese compared to participants with normal BMI. The R² (Nagelkerke) value of the final model was 0.212, thereby explaining about 21.2% variation of CKD in the study population.

DISCUSSION

Studies conducted after the Bhopal gas disaster documented a high prevalence of several chronic illnesses related to the respiratory, gastrointestinal, neurological, psychiatric and ophthalmic systems.^11,13,14 However, the burden of renal disease among this community remained undocumented. In the past few years, a concern has been voiced that the prevalence of kidney-related ailments is higher in the gas-exposed community compared to the general population as a consequence of the ill-effects of gas exposure. We did not include moderately and mildly exposed cohort survivors in the study primarily due to operational reasons and second, it was assumed that any long-term adverse effect of gas exposure on the renal system resulting in the development of CKD will be maximal in the severely exposed cohort and thus they will have the highest prevalence of CKD among the three cohorts.

We found a 16.7% prevalence of CKD in the severely exposed cohort of gas survivors. An earlier cross-sectional study conducted using similar case definitions and equations as used in our study, which covered 6120 subjects across various Indian cities including Bhopal reported in 2013, found a nearly similar prevalence (17.2%) of CKD among Indian adults.⁷ Further, Ene-Iordache et al.⁵ in their cross-sectional study carried out in 12 lower- and middle-income countries from six regions of the globe to assess the prevalence and awareness of CKD and its risk factors reported a 16.8% prevalence of CKD in India, which matches with the prevalence reported in our study. A current systematic review based on eight Indian studies reported a 10.2% pooled prevalence of CKD in India.² Among the studies included in this systematic review, the highest prevalence in India was 17.2% among participants of the Screening and Early Evaluation of Kidney Disease study, which screened 6120 subjects from 13 academic and private medical centres across India⁷ and the lowest prevalence of 4.2% was found among <20 years old adult residents from Delhi.¹⁹ Thus, the prevalence of CKD estimated in our study was comparable with the national prevalence.

The minor variations in the reported prevalence among different studies from India could be because different definitions of CKD were used, or the difference in the methods/equations adopted in estimating GFR, the difference in the study population, as well as differences in the geographical area studied, etc.^10,19–21 For instance, Agarwal and Srivastava²² reported a prevalence of CKD of 0.79% in India. The prevalence might have been underestimated due to the use of serum creatinine >1.8 mg/dl as the cut-off. On the other hand, Anupama and Uma¹⁰ reported the prevalence of CKD as 6.3% using the MDRD equation in a rural population of southern India. The difference in the CKD prevalence estimated in our study with that of other studies^2,10,22 could partly be explained by the high prevalence of diabetes (20%) and hypertension (56.2%) among the subjects in our study. In 2005, Modi and Jha²³ conducted a large ESRD incidence study among the gas-exposed victims of Bhopal visiting the dedicated tertiary care hospital, and reported that average crude and age-adjusted incidence rates were 151 and 232 per million population, respectively. In our community-based, cross-sectional study conducted among the severely gas-exposed cohort survivors, we found that the prevalence of later stage of CKD (G3b) was 2.3%. There is a need for future community-based prospective studies to ascertain the change in incidence rates in this population. Further, Modi and Jha in their study found that diabetic nephropathy was one of the leading causes of ESRD,²³ which was reflected in our study as well.

In our study, 89% of the subjects diagnosed with CKD were unaware of the status of their renal condition and hence were new cases of CKD identified during the study. This indicates the lack of screening for renal damage among those suffering from other chronic non-communicable diseases (NCDs). Hence, there is a need for designing and implementing country-specific standard guidelines for screening of patients with NCDs for their renal function. A notable finding in our study was the lower prevalence of proteinuria (2.7%), which is at variance to earlier population-based studies on CKD.^10,24 This could partly explain why a higher proportion of patients with CKD were unaware of their renal condition in our study because, at the primary care level, screening of renal function is based on urine protein analysis. This should be considered while planning any screening programme in this population.

In our study population, BMI and level of education were found to be significant independent predictors of CKD. Having some formal education significantly reduced the likelihood of development of CKD compared to illiterate individuals. Consistent with our findings, several studies in western countries^24–28 and India^19,29,30 reported an increased risk of CKD and its outcomes in individuals with a lower level of education. In an observational cohort of 61 457 participants of the Kidney Early Evaluation Program study, it was found that higher educational level was associated independently with a lower prevalence of CKD and lower mortality in those in the cohort who had chronic diseases including CKD.²⁶ The educational status of an individual may impact development of CKD and its diagnosis through several factors such as health literacy and knowledge of the impact of comorbid illnesses on renal function, health-related behaviour including healthcare seeking and utilization and access to healthcare delivery systems.²⁷ Many studies have reported the association of unhealthy behaviours such as consumption of unbalanced diet, smoking and alcohol intake with a lower educational level.^26–28 Similarly, an association has also been observed between lower educational level and diseases such as diabetes and hypertension.³ A study exploring the socioeconomic disparities in prevalence of CKD revealed that education was associated more closely with the prevalence of CKD and its clinical outcomes as compared to income.³¹ Factors such as health behaviour, comorbid illness and health system access that are influenced by lower educational status may lead to higher risk of CKD and thus need to be studied in detail in Indian settings.

We found a dose–response relationship with BMI and CKD. Compared to individuals with normal BMI, the likelihood of developing CKD among overweight individuals was three times higher and for obese individuals, the likelihood increased to nine times. This finding is in agreement with previous studies conducted across a diverse population.^32–34 Overweight and obesity result in a wide range of metabolic abnormalities, which may affect renal function.^35,36 Presumably, some of the harmful effects of obesity on kidneys are mediated through comorbid illnesses such as diabetes and hypertension. Evidence exists about the independent and direct effect of adiposity on kidneys induced by the endocrine activity of adipose tissue.^36,37 However, whether BMI is an appropriate indicator of adiposity is a debatable issue.³⁷