Among randomly sampled adults living in a middle-income neighborhood of Bangladesh, approximately one in five were found to have CKD. Nearly four of five participants with CKD had features of insulin resistance, which was the predominant risk factor linked with CKD in our study. In contrast to data from the U.S. and other countries where stage three CKD predominates [14–16], we found that stage one or two CKD (albuminuria only) was most common. The high prevalence of insulin resistance and associated albuminuria among our participants with CKD implies that they are at high risk for progression of CKD and for experiencing a cardiovascular event.
No prior data exist on prevalence of CKD in Bangladesh. Two studies from New Delhi, India have used variable definitions of CKD. A study of 4,712 adults living in South Delhi found that 0.8% had serum creatinine above 159 umol/L (1.8 mg/dL) ; the comparable percentage in our study was 3%. Another study of 5,252 adults also based in Delhi reported a prevalence of reduced eGFR (below 60 ml/min/1.73 m2) equal to 4% ; the comparable percentage in our study was 10%. The Thai-Global Screening and Early Evaluation of Kidney Disease (SEEK) study sampled 3,459 from 10 provinces. Both albuminuria and serum creatinine were evaluated and the standard NKF/KDOQI definition was applied; 17% of surveyed participants met criteria for CKD . As reference, the most recent estimates of CKD prevalence from Norway , Spain , U.S. , and Japan  are between 10 and 13% of the adult population.
We hypothesize three reasons for the higher prevalence of CKD noted in our study. First, we surveyed adults over 30 years old, whereas Thai-SEEK and NHANES enrolled adults over 18 and 20 years old, respectively. Second, by surveying a well-established neighborhood of a major urban hub, we are sampling from a population at higher risk, not representative of the risk experienced by people living in rural areas or poorer persons. Migrating to an urban area was associated with an approximately 1.5-fold increase in odds of a sedentary lifestyle, three-fold increase in odds of obesity, and two-fold increase in odds of T2DM among men working in factories in India, compared with their rural counterparts . In low-income countries, the prevalence of obesity increases with increasing wealth, although this trend may soon arrest or reverse as prices of fruits and vegetables climb while processed foods become cheaper and more easily accessible .
A majority of participants who were identified as having CKD had stage one or two CKD (i.e., albuminuria only). As we did not repeat assessment of albuminuria, we do not have population-specific data on persistent albuminuria. However extrapolating data from NHANES on spot versus persistent albuminuria, 51% of participants with microalbuminuria and stage one CKD and 75% of participants with microalbuminuria and stage two CKD would be estimated to have persistent microalbuminuria, yielding an adjusted (approximate) CKD prevalence equal to 19%, rather than 26%. Thus, the prevalence of CKD in the urban Bangladeshi population would still be higher than that in the U.S., Europe, and Japan. Strikingly, we identified four persons (1%) with stage five CKD (eGFR below 15 ml/min/1.73 m2). If we extrapolate only to the >30 year old population of Mohammedpur, a stage five CKD prevalence of 1% would imply that roughly 1,800 persons would be at risk for imminent kidney failure in a district that represents 0.3% of the Bangladeshi population. Even these 1,800 persons represent an immense burden in a country where kidney transplants are performed in only one and dialysis in just four of 13 government hospitals .
Aside from the need for transplantation or dialysis, persons with CKD (especially those with albuminuria) experience an increased risk for cardiovascular disease and mortality. Microalbuminuria, found in 70% of our participants with CKD, has been linked to a 30 to 80% increase in the risk for all-cause and cardiovascular mortality; overt proteinuria, found in 15% of our participants with CKD, can double or triple these risks . The majority of our participants with CKD also had T2DM, which increases risk for all-cause and cardiovascular mortality 20 to 90%, relative to persons with CKD who do not have T2DM .
In Bangladesh, care for chronic illnesses is not yet in line with the care for infectious diseases; much of the public sector health budget is devoted to treating episodic illnesses such as diarrhea or respiratory illness. No national guidelines exist for primary health care screening for chronic illnesses such as CKD. Even among patients identified to have a chronic illness, cost of treatment is a major burden: while the government subsidizes some of the care (e.g., the doctor’s visit) in the public sector, patients may still pay for other aspects (e.g., medications). Some subsidies are available from social welfare organizations. Those who can afford it do seek private care, and a negligible percent goes abroad for better treatment.
We also found that moderate to advanced CKD was associated with a decrement in self-reported physical health. While few studies have explored HRQoL in patients with non-dialysis requiring CKD, the associated decrement in self-reported physical health seen among participants with moderate to advanced CKD in this study is considered to be clinically meaningful . Thus, if the burden of CKD is not lessened in the coming decades, we can anticipate higher rates of cardiovascular disease, frailty, and debility in the urban Bangladeshi population, even in late middle and early older age.
Our study has several strengths. We adopted a random sampling procedure in this community-based survey. Trained and experienced nurses and research assistants used standardized protocols for physical measurements. Anticipating low rates of awareness of chronic diseases, we employed laboratory measurement to diagnose T2DM, metabolic syndrome, and CKD. Ours is one of few studies in the region to include an assessment of albuminuria and HRQoL.
However, given our modest sample size and that our survey was limited to a particular urban neighborhood, we cannot extrapolate our findings to all of Bangladesh. Our modest sample size also limited our assessment of the correlates of CKD. Ideally, repeat measures of serum creatinine and urinary albumin excretion might allow us to be more precise in our prevalence estimates.