The present large screening programmes of general populations showed a high prevalence of eGFR<60 ml/min/1.73 m2
and microalbuminuria in Nepal and Bolivia, similar to that found in the US population by NHANES. Hypertension, DM and obesity were relatively common among these populations and clustered in subjects with renal dysfunction. As expected, prevention programmes in Bangladesh and Georgia focusing on subjects with already known CVD risk factors such as DM, hypertension or previous CV events found higher prevalence of microalbuminuria. In accordance with previously published studies, male gender was associated with an increased prevalence of hypertension, reduced renal function and proteinuria.10–12
The prevalence of renal dysfunction and microalbuminuria found in Nepal and Bolivia is in line with those reported by other ISN-sponsored screening programmes in indigent residents of Guadalajara, Mexico13
or of Kinshasa, in the Democratic Republic of Congo,14
highlighting the generalisability of this approach for case detection in diverse settings around the world. Notably, the availability of the NHANES dataset allowed us to infer that the burden of CKD is similar across different populations worldwide, including both high-income and low-income countries. Importantly, data on the prevalence of albuminuria retrieved from the NHANES cohort are similar to those reported by the Prevention of Renal and Vascular ENd-stage Disease study, a large European screening programme.15
This is consistent with other evidence that chronic non-communicable diseases are now the major cause of morbidity and mortality in the entire world.16–19
Intriguingly, prevalence of CKD and of subjects with high CV risk was consistent across different countries, independent of per capita incomes (available at http://www.heritage.org/index/default
). Of note, prevalence of CKD in low-income countries and in the USA was confirmed when data were corrected for different age distributions across nations. In low-income countries, levels of physical activity were higher than in the US population, whereas smoking habit and alcohol consumption tended to be lower. However, in light of progressive spread of Western lifestyles,20
incidence of DM and hypertension has been forecast to increase in these nations during the next few years, which is also expected to translate into increased CKD and CVD. This clearly highlights the importance of prevention programmes especially in low-income countries, where renal replacement therapies and coronary revascularisation procedures are available only for a minority of people.
About 15% of subjects had an estimated 10% or higher risk of developing a CV risk in the following 10 years. In consideration of the high prevalence of renal dysfunction and microalbuminuria, this risk was probably underestimated, although the net additional contribution of eGFR and albuminuria on CVD risk is still unclear. Long term outcome analysis of the present screened cohorts will allow assessing whether inclusion of albuminuria and/or reduced GFR among the variables considered in algorithms for prediction of individual CVD risk will improve the performance of current WHO prediction algorithms.
Evidence is emerging that CKD and CVD have a major impact on macroeconomic development due to diminished labour supply related to premature death and disability in people of working age. According to WHO, these conditions decrease the potential annual growth rate in gross domestic product by 1–5% in developing countries experiencing rapid economic growth.21
Importantly, data from large trials have consistently shown that off-patent drugs such as ACE inhibitors can reduce albuminuria and prevent GFR decline and CV events.5
Thus, prevention programmes should identify and treat renal abnormalities early, with the primary goal to reduce CV mortality and morbidity, which, on its turn, may translate into an economical benefit.
There is little debate that screening for albuminuria should be performed in patients with DM and/or hypertension, where early intervention can slow down deterioration of renal function. However, due to the difficulty of identifying subjects at risk in low-income countries, a prescreening phase including clinical history, BP and anthropometric values might be instrumental in identifying patients in whom screening with serum and urine testing could be most cost-effective. On the contrary, data from Nepal and Bolivia showed that more than 5% of people younger than 60 years without previous history of diabetes and hypertension had microalbuminuria/proteinuria. Consistently, screening also low-risk groups, or even the general population, has been advocated to identify and treat those at risk for progressive renal disease, arguing that most persons with albuminuria and/or reduced eGFR (<60 ml/min/1.73 m2
) are asymptomatic. However, concerns towards general population screenings regard not only the cost of screening itself, but, more importantly, the risk and the cost of treating false-positive subjects with no other modifiable risk factors. Thus, patients with a first positive test should be always asked to repeat the analyses and only those with confirmed positivity should be treated.22
However, when debating how to address the issue of screening for non-communicable diseases and, more in general, of health in developing nations, we should look at the problem from the perspective of the low-income and middle-income countries, not from that of industrialised nations. Indeed, there is not a unique blueprint of screening strategy even among developing countries, so that the approaches should be adapted on single-nation conditions and socioeconomic status.23
Strengths and limitations of the study
Our study has several limitations. At variance with NHANES data, showing characteristics of a cohort representative of US population, results from Bolivia cannot be taken to infer the absolute or relative prevalence of renal abnormalities in this country, since all subjects were referred to a single centre. However, in consideration of the limited availability of local resources, current data seem reasonably reliable and representative of a large fraction of people, at least those living at the altitude regions of this country. Reliability of data is supported by the fact that, in line with available evidence, prevalence of microalbuminuria was higher in older subjects and in those with hypertension or diabetes. Also data from Nepal cannot be formally considered representative of the whole Nepalese population. However, the multiple sites used for the screening, along with the large sample size and consistency with data retrieved from a previous smaller cohort of Nepalese subjects24
make the present dataset insightful of the prevalence of CKD and CVD in this nation. Importantly, data from Bolivia were similar to those found in Nepal and this suggests that these data may reflect patterns throughout the developing world. Of note, although formal comparison across different populations is prevented by different inclusion criteria and sampling strategies, use of the same cut-off points for clinical and laboratory parameters and the same web-based database for data entry and centralised data monitoring make the present analysis a unique opportunity to study prevalence of CKD and CVD in countries with different incomes.
Owing to limited resources, laboratories in developing countries were not calibrated according to the National Kidney Disease Education Program. However, each site made calibrations for creatinine measurement according to guidelines suggested by the manufacture, which should have prevented major bias in our findings. Some authors have advocated to repeat renal function measurements in screening programmes to prevent overestimation of disease prevalence.22
However, in low-income countries, repeated measurements on a standard basis would unnecessarily increase costs and could be burdened by high rate of non-compliance, which, on its turn, could reduce the number of subjects identified with renal abnormalities. In addition, the primary goal of any screening programme is to avoid false negative results that might affect the identification of subjects at risk, whereas the risk of false-positive results is of relatively limited importance since subjects erroneously identified with albuminuria can be correctly characterised at follow-up evaluation. Therefore, different assays to measure albuminuria should be titrated to single resources. Consistently, a pilot phase of National Kidney Foundation Kidney Early Evaluation Program in México City and Jalisco State25
reporting prevalence of eGFR<60 ml/min/1.73 m2
and albuminuria close to the ones we found in high-risk subjects from Georgia and Bangladesh, evaluated albuminuria levels once and used the same cut-off points considered in our analyses.
Finally, an issue of the present study was the use of GFR-predicting equations that may not fully fit to different ethnicities, thus some participants from these countries might have been misclassified with respect to the presence of eGFR<60 ml/min/1.73 m2. This could account for the variability across different countries. On the contrary, performance of formulas is poor for high levels of GFR, whereas they tend to improve for lower values. Thus, risk of misclassification of subjects below the threshold of 60 ml/min/1.73 m2 is reasonably low. Notably, since no ad hoc formulas are available to estimate GFR more precisely in considered populations, such ISN screening programmes might offer the unique opportunity to implement formulas to fit best different ethnicities.