NAFLD and CKD share similar pathological mechanisms; therefore they are speculated to have some links. This cross-sectional study analyzed the association between ultrasound-defined NAFLD and CKD in two health examination datasets from the US and China. The results of this study confirmed the higher prevalence of CKD among patients with NAFLD. Slightly difference was found between Chinese and US population in this study: in the Chinese population, NAFLD was significantly associated with increased risk of CKD after adjustment for metabolic factors, but in the US population, no significant association was found after adjustment for the same factors. However, when we excluded patients with advanced reduced renal function (eGFR<60 ml/min/ 1.73 m2, CKD stages 3-5), NAFLD was significantly correlated with increased risk for early renal function decline in both populations.
The discrepancy between eastern and western population regarding the relationship between NAFLD and CKD has already been found by previous studies. Sirota et al.[12] found that NAFLD is not associated with the prevalence of CKD among US adults after adjusting for features of metabolic syndrome. While other studies from Asia reported a strong independent risk of CKD in ultrasonography-diagnosed NAFLD patients [13, 17, 18]. The meta-analysis also confirmed the difference between Asian and European populations[8]. The reason for this discrepancy has not been clarified since all the studies were not analyzed under same statistical condition. In this study, in order to compare the association of NAFLD and CKD in different populations, we creatively calculated the risks for CKD by adjustment for same confounders in different datasets. Consistent with previous studies, we confirmed the ethnic difference in the relationship between CKD and NAFLD in two population-based datasets.
The answer for the ethnic difference on the relationship between NAFLD and CKD might result from different proportion of severe renal dysfunction. The US population had a higher percentage of stages 3-5 CKD than the Chinese population. In fact, after we excluded those with late stages of CKD, NAFLD was strongly associated with early renal decline in both populations. Supporting our results, several prospective studies confirmed the influence of NALFD on the development of CKD, most of which were developed from the early stages[19-21]. But an important fact we should emphasize is that, compared with the early stage of CKD, the later stage of CKD is more complicated and severe. For example, increased synthesis and decreased clearance of triglycerides, extremely altered glucose homeostasis and uncontrollable blood pressure are more common among patients with late stages CKD[22-24]. Thus NAFLD might have an impact on renal function, yet it alone apparently not strong enough to contribute to the late stage function decline when severe renal decline is concerned. That explains why the association between late stages of CKD and NAFLD is not significant in both populations. When there are more late stages of CKD, such as the US population, the relationship of NALFD and CKD might attenuate during multivariate analysis.
To our best knowledge, this study is the first to compare the association between NAFLD and early renal function decline across different ethnicities. And we also answer the questions of conflicting conclusion from different countries from some aspects. The results of our study suggest that NAFLD plays a more important role in mild renal dysfunction, which providing more evidences for the hypothesis that pre-existing NAFLD is an independent risk factor for the development of renal injury.
There are several limitations in this study that deserve a mention. Firstly, the diagnosis of CKD usually requires the presence of an abnormality of kidney function or kidney structure for more than 3 months; however, it is difficult to perform a second examination in such a large population in a population-based study. We had to admit that the diagnostic criteria used in this study, which has also been used by several similar researches[7, 12, 25], might overestimate the prevalence of CKD. Given the cross-sectional design of this study, we are unable to draw conclusions about the causality of the relationship between NAFLD and early renal function decline. Secondly, we have not use cystatin C to define eGFR, which is less affected than creatinine by muscle mass and more accurate for different ethnicities. Thirdly, there is a huge difference in period of database between the US (1984-1994) and Chinese (2010) populations, therefore, the discrepancy in the diet and lifestyle might be a major concern. It is widely accepted that patients with CKD should limit the intake of certain foods to reduce the accumulation of unexcreted metabolic products and also to protect against hypertension, proteinuria and other health problems. In fact, the diet and lifestyle in China are closer to that in western country since 21th century. According to a previous report, China’s food consumption patterns and dietary behaviors changed dramatically between 1991 and 2011. The diet has shifted from macronutrient composition toward fats, and protein and sodium intakes[26]. Another study also demonstrated that the structure of the Chinese diet has been shifting away from the traditional diet toward high-fat, low-carbohydrate and low-fiber diets, and nutrients intakes in Chinese people have been changing even worse than those in American people[27]. Therefore, the heterogeneous database might partially influence the results, but the impact is not as larger as expected.
In conclusion, our findings demonstrated a significant positive association between the presence of NAFLD and early stage of CKD in both the US and Chinese adults. For the late stages of CKD, NAFLD might not be strong enough to be an independent contribut