Risk Factors for Chronic Kidney Disease in Older Adults with Hyperlipidemia and/or Cardiovascular Diseases in Taipei City, Taiwan: A Community-Based Cross-Sectional Analysis

This cross-sectional study aimed to compare risk factors for chronic kidney disease (CKD) in older adults with or without dyslipidemia and/or cardiovascular diseases (CVD) in Taipei City, Taiwan. The data on 2912 participants with hyperlipidemia and/or CVD and 14,002 healthy control participants derived from the Taipei City Elderly Health Examination Database (2010 to 2011) were analyzed. The associations between conventional CKD risk factors and CKD were comparable between participants with and without hyperlipidemia. Participants with high uric acid and BUN had a higher risk of CKD if they also had hyperlipidemia and CVD [odds ratio (OR) in uric acid = 1.572, 95% CI 1.186–2.120, p < 0.05; OR in BUN = 1.271, 95% CI 1.181–1.379, p < 0.05]. The effect was smaller in participants with hyperlipidemia only (OR in uric acid = 1.291, 95% CI 1.110–1.507, p < 0.05; OR in BUN = 1.169, 95% CI 1.122–1.221, p < 0.05). The association between uric acid/BUN and CKD was also observed in the healthy population and participants with CVD only. In conclusion, older adults with hyperlipidemia and CVD are at high of CKD. Physicians should be alert to the potential for CKD in older patients with hyperlipidemia and CVD.


Introduction
Chronic kidney disease (CKD) is a worldwide health problem with a steady annual increase in occurrence of approximately 6%, and significant differences in prevalence between populations [1,2]. CKD represents an important public health issue because such patients have an increased risk of end-stage renal disease (ESRD). Taiwan has a high prevalence of both CKD [3] and ESRD [4]. Significantly elevated cardiovascular morbidity and mortality have been observed in the course of CKD. In patients with CKD, the cardiovascular mortality rate is 10 to 20 times higher than in the general population, and in the ESRD population, it is 20-30 times higher [5,6]. The spectrum of cardiovascular diseases (CVD) in the CKD population includes arterial vascular disease such as atherosclerosis and arteriosclerosis, concentric left ventricular hypertrophy, heart failure, and non-atherosclerotic CVD, which becomes dominant in more advanced stages of CKD [7,8]. However, unravelling the exact mechanisms and causal pathways linking CKD and CVD remains a challenge.

Population and Definition
The present study utilized data derived from the Taipei City Elderly Health Examination Database (2010 to 2011), which collected health examination data from community-dwelling Taipei citizens aged 65 years or older. Taipei City, located in northern Taiwan, is the capital of Taiwan. The Taipei City Elderly Health Examination Database sponsored by the Department of Health, Taipei City Government, has been used for research as previously described [20][21][22][23]. The protocol of this study was reviewed and approved by the Institutional Review Board (IRB) of Taipei City Hospital (TCHIRB-10514118-W). In this cross-sectional study, participants who had missing values for age, sex, or serum creatinine level were excluded because all these variables were required to calculate the estimated Glomerular Filtration Rate (eGFR) based on the Chronic Kidney Disease Epidemiology Collaboration formula [24]. In addition, 91 participants were excluded because of cancer, nephrectomy, kidney transplantation, or ESRD and waiting for a renal transplant. Finally, a total of 16,914 (7533 male and 9381 female) older participants (mean age 74.9 years) were included in the study.

Data Extraction
The physical examination data and physical and mental questionnaire data included sex, age, body mass index (BMI), clinical laboratory data, and other parameters, which were defined as descripted in previous studies [20][21][22][23]. The lifestyle behaviors include smoking, exercise habit, alcohol drinking, and betel nut chewing. The behavioral data was based on a standardized self-administered questionnaire developed by the Health Promotion Administration.
In addition, proteinuria was determined by dipstick test as previously described [27], and the presence (positive: 1+ or more) or absence of proteinuria was recorded and analyzed in this study. The presence of hyperlipidemia, CVD, and other comorbidities was indicated based on the self-reported medical condition and medication use.

Statistical Analysis
Characteristics between individuals with low and high eGFR were compared; continuous variables were presented as mean and standard deviation, tested using the t test; categorical variables were presented as count and percentage, tested using the chi-square test. The association between risk factors and the eGFR level (low eGFR: eGFR < 60 mL/min/1.73 m 2 and high eGFR: eGFR ≥ 60 mL/min/1.73 m 2 ) was assessed by logistic regression. Multivariate logistic regression was performed to adjust for variables that were significant in the univariate regression model. Stratified analyses were performed based on dyslipidemia and CVD to explore the association between risk factors and the eGFR level. The significance level was set as two-sided p < 0.05. All statistical analyses were performed using the statistical software R, version 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria).

Clinical Characteristics of the Study Population
A total of 16,914 older participants were included in this study, with a mean age of 74.86 years. The patients' characteristics are presented in Table 1. This study cohort was slightly female-predominant (55.46%) with the prevalence of comorbidities as follows: hypertension, 8.14%; diabetes mellitus, 2.93%; hyperuricemia/gout, 7.17%; proteinuria, 16.66%; and CVD, 9.31%. Moreover, the majority of participants were married/cohabiting (74.07%) and 48.73% had normal BMI. In terms of socioeconomic characteristics, 56.8% had a high school diploma or higher and 97.66% were financially better than poor. Most participants had a regular exercise habit (38.28%), did not drink (81.61%), had no current smoking habit (95.24%), and reported no betel nut chewing (99.35%).

The Association of Laboratory Values, Comorbidity, and Future Risk of CKD
Participants were divided into four groups based on dyslipidemia and CVD status: Group 1: participants without dyslipidemia and CVD; Group 2: participants with dyslipidemia only; Group 3: participants with CVD only; and Group 4: participants with dyslipidemia and CVD ( Figure 1). married/cohabiting (74.07%) and 48.73% had normal BMI. In terms of socioeconomic characteristics, 56.8% had a high school diploma or higher and 97.66% were financially better than poor. Most participants had a regular exercise habit (38.28%), did not drink (81.61%), had no current smoking habit (95.24%), and reported no betel nut chewing (99.35%).

The Association of Laboratory Values, Comorbidity, and Future Risk of CKD
Participants were divided into four groups based on dyslipidemia and CVD status: Group 1: participants without dyslipidemia and CVD; Group 2: participants with dyslipidemia only; Group 3: participants with CVD only; and Group 4: participants with dyslipidemia and CVD ( Figure 1). Demographic, socioeconomic, lifestyle factors, laboratory data, and comorbidities of the four groups, stratified by eGFR level (eGFR < 60: CKD; eGFR ≥60: non-CKD), are shown in Table 2.
Aging, high uric acid, high blood urea nitrogen (BUN), proteinuria, and a history of hyperuricemia/gout were significantly associated with CKD in all four groups (all p < 0.05, Table 2). Male sex, obesity, high triglyceride, HDL-C, and prior hypertension were significantly associated with CKD in Group 1, Group 2, and Group 3; however, low income, low education, and high exercise habit were significantly associated with CKD in Group 1 (all p < 0.05, Table  2).
Univariate logistic regression analysis found a similar relationship between these factors (having a significant difference between CKD and non-CKD individuals in each Group in Table  2) and the risk of CKD in these subgroups (all p < 0.05, Table 3). In multivariate logistic regression analysis adjusted for the risk factors associated with CKD as shown in Table 3, we found similar relationships between these factors (having a significant difference between CKD and non-CKD individuals in each Group in Table 2) and the risk of CKD in these subgroups (all p < 0.05, Table 4). However, the effect of low income on the risk of CKD did not differ significantly by subgroup after adjustment for conventional risk factors (p > 0.05, Table 4). Demographic, socioeconomic, lifestyle factors, laboratory data, and comorbidities of the four groups, stratified by eGFR level (eGFR < 60: CKD; eGFR ≥ 60: non-CKD), are shown in Table 2.
Aging, high uric acid, high blood urea nitrogen (BUN), proteinuria, and a history of hyperuricemia/gout were significantly associated with CKD in all four groups (all p < 0.05, Table 2). Male sex, obesity, high triglyceride, HDL-C, and prior hypertension were significantly associated with CKD in Group 1, Group 2, and Group 3; however, low income, low education, and high exercise habit were significantly associated with CKD in Group 1 (all p < 0.05, Table 2).
Univariate logistic regression analysis found a similar relationship between these factors (having a significant difference between CKD and non-CKD individuals in each Group in Table 2) and the risk of CKD in these subgroups (all p < 0.05, Table 3). In multivariate logistic regression analysis adjusted for the risk factors associated with CKD as shown in Table 3, we found similar relationships between these factors (having a significant difference between CKD and non-CKD individuals in each Group in Table 2) and the risk of CKD in these subgroups (all p < 0.05, Table 4). However, the effect of low income on the risk of CKD did not differ significantly by subgroup after adjustment for conventional risk factors (p > 0.05, Table 4).

Discussion
In the present study, male sex, greater age, high triglyceride, high uric acid, high BUN, high proteinuria, and prior hyperuricemia/gout were risk factors for CKD in participants without hyperlipidemia or CVD. We also found that BMI, marital status, income level, smoking, alcohol intake, betel nut chewing, fasting glucose, triglyceride, and history of diabetes or cancer were not associated with CKD in participants without hyperlipidemia or CVD. Consistently, several studies indicated that male sex and physical inactivity are two conventional risk factors for CKD in adults [28][29][30]. In our study, participants were divided into four groups, depending on the presence of dyslipidemia and CVD, and significant associations between these conventional risk factors and CKD were observed in certain groups, but not in all four groups.
Low income is a critical predictive factor for CKD. Consistently, our data also indicated that low income was significantly associated with a high risk of CKD in participants without hyperlipidemia or CVD. On the other hand, people with lower socioeconomic status are often associated with many common health risk factors including smoking [31], second-hand tobacco smoke exposure [32], alcohol consumption [33], unhealthy diets [34], and betel nut chewing [35]. Cigarette smoking and betel nut chewing have been reported to be associated with CKD [36,37]; however, no significate associations between these common health risk factors and CKD in all four groups were observed in the present study.
The discrepancy between the current findings and the results of two abovementioned studies might be partially due to the differences in age range, inclusion criterion, participant grouping, and/or adjustment for socioeconomic status [36,37]. Hence, further large-scale investigation is warranted to clarify the association between individual socioeconomic status-related factors and CKD in older adults with hyperlipidemia and/or CVD.
Older participants with high uric acid and BUN had a higher risk of CKD than those without it in all four groups, and the association was very high after adjustment for conventional risk factors in the present study. The prevalence of hyperuricemia appears higher in older Taiwanese and the total population [38,39], which may, in part, contribute to the high incidence of ESRD observed in Taiwan [40]. Similarly, an elevated prevalence of hyperuricemia in patients with CKD and nonalcoholic fatty liver disease (NAFLD) in Iran was recently reported [41]. Several studies revealed that hyperuricemia is significantly associated with CKD in the general population [29,39]; however, the present study further found that older adults with hyperuricemia are at a much higher risk of developing CKD compared to the general population. Taken together, we suggest that hyperuricemia and BUN, compared to other conventional risk factors, might be associated with a greater risk of developing CKD and ESRD.
It has been documented that awareness of CKD in people who displayed markers and/or clinical manifestations associated with CKD, such as increasing age, obesity, CVD, and hyperlipidemia, is extremely low [42,43]. Consequently, it is a high priority for health professionals to promote awareness of CKD in high-risk populations, thereby facilitating early detection of CKD [43]. The present study found that uric acid and BUN levels were strongly associated with CKD in older patients with or without hyperlipidemia and CVD, suggesting the potential of uric acid and BUN levels as biomarkers for CKD in older adults aged 65 and over. Notably, hyperlipidemia is a risk factor for hyperuricemia in patients with CKD and NAFLD [41]. Furthermore, dyslipidemia and CVD have been demonstrated to be independently associated with CKD in the general population [44,45]. Taken together, older patients with hyperlipidemia and/or CVD are at very high risk of developing CKD, and their awareness of CKD should be enhanced by health professionals.
In the present study, the harmful effects of high lipid levels (triglyceride, total cholesterol, and HDL) on developing CKD were only observed in the general population and participants with hyperlipidemia only, but not in those with CVD only. Male participants were at a higher risk of CKD than their female counterparts; however, the negative impact of high lipid levels on CKD was less severe in males compared to that of females [29]. Obesity was a risk factor related to high lipid levels. In the present study, obesity and high lipid levels produced comparable harmful effects on CKD, suggesting the importance of hyperlipidemia and obesity in the prediction of CKD. Renal lipid accumulation was demonstrated to be nephrotoxic and may be involved in the progression of CKD [46]. However, further research is warranted to disclose the underlying mechanism responsible for all associations with CKD observed in older adults with hyperlipidemia and/or CVD.
The strength of this study is the utilization of the Taipei City Elderly Health Examination Database, which allows for comprehensive investigation of health issues in older adults aged 65 and over. However, several limitations of this study have to be discussed. First, no cause-effect relationships could be explored and established in this cross-sectional study. Second, self-reported questionnaires were used to collect data on lifestyle factors, so the possibility of recall bias cannot be ruled out. Finally, quantitative data on proteinuria level was not available in this study. Furthermore, the current findings collected from northern Taiwan should be confirmed by additional large-scale cross-sectional and longitudinal studies conducted in different geographic areas.

Conclusions
Sex, age, income, lipid levels, proteinuria, and prior hyperuricemia/gout represent potential candidates for developing an effective screening/prevention strategy for CKD in older adults with or without hyperlipidemia. However, smoking, alcohol intake, and betel nut chewing were not good candidates for CKD prevention in these populations. In addition, high uric acid or BUN levels may also be considered in the screening strategy for CKD in older patients with hyperlipidemia and CVD. While implementing a specific CKD screening/prevention strategy for older patients with hyperlipidemia and/or CVD, the awareness of CKD in such high risk populations should be enhanced by health professionals. Funding: The study did not receive any funding.