Relationship Of Dyslipidemia To Stroke In Dialysis Patients And Chronic Kidney Disease

BackgroundChronic kidney disease (CKD) has been a global challenge for health care systems. CKD not only represents a progressive irreversible process that ultimately leads to end-stage renal disease (ESRD) for renal replacement therapy, but also increases the risk of cardiovascular diseases (CVD) and cardiovascular morbidity and mortality. The global incidence of ESRD has also been increasing steadily. Compared with the general population, long-term dialysis patients have a 10-to 20-times greater incidence of cardiovascular mortality.According to a position statement from Kidney Disease Improving Global Outcomes (KDIGO) in 2007,9.6% of non-institutionalized adults have CKD in the United States. Studies from Europe, Australia, and Asia confirm the high prevalence of CKD. The US Renal Data System (USRDS) 2014 Annual Data Report declared, CKD prevalence in US in 2012 (not including ESRD) was estimated at 13.6%, and has been relatively stable over the last decade. There were 114813 new cases of ESRD reported, and 636,905 individuals were treated for ESRD in 2012. The overall prevalence of CKD in China was established to be 10.8% in 2012. At the end of 2013 in Beijing, China, there were 12,249 patients on maintenance hemodialysis (MHD), which corresponded to 579 per million population. The prevalence rate showed an increasing trend in China.Stroke is the main cause of death in these patients with ESRD, and as the second and third leading causes of death in the United States and United Kingdom, respectively. Stroke was also the second main cause (20.3%) of death in Chinese patients with ERSD. Hemodialysis (HD) or peritoneal dialysis (PD) has been widely accepted for treatment in patients with ERSD. Previous studies in the US and Japan reported 2-to 10-fold increased risks of stroke in dialysis patients compared to the general population. This propensity for stroke has been attributed to their higher prevalence of factors recognized as risk factors for stroke in the general population, such as age, hypertension, diabetes mellitus and dyslipidemia, plus the presence of factors specific to patients receiving dialysis for ESRD, such as accelerated calcific arteriosclerosis, effects of uremic toxins, dialysis techniques, vascular access, and use of anticoagulants to maintain flow in the extracorporeal circuit.Dyslipidemia is an established risk factor for atherosclerotic disease, such as stroke and ischemic heart disease. A large number of epidemiologic studies have suggested the independent role of dyslipidemia on cardiovascular morbidity and mortality in the general population. However, the role of dyslipidemia as an independent risk marker for stroke in dialysis patients still remains uncertain. We designed the present prospective observational cohort study to estimate the incidence rates for different subtypes of stroke in patients undergoing HD or PD, and to identify if dyslipidemia is a prognostic risk factor of stroke events associated with each dialysis modality.Furthermore, the role of dyslipidemia in CKD progression has not been well understood. Experimental and clinical studies have suggested a correlation between dyslipidemia and the progression of CKD. High cholesterol and triglyceride could be independent risk factors for progression of CKD. To the best of our knowledge, the epidemiological studies of the relationship between CKD and dyslipidemia are inconsistent, even conflicting. Thus, current study aimed to explore the association between CKD and dyslipidemia in a southern Chinese population.Objective1. estimate current status of dyslipidemia in dialysis patients of China.2. estimate the incidence rates for different subtypes of stroke in patients undergoing HD or PD.3. identify prognostic risk factors of stroke events associated with each dialysis modality.4. identify if dyslipidemia is a prognostic risk factor of stroke in dialysis patients.5. explore the association between CKD and dyslipidemia in a southern Chinese population.1. the Association between Dyslipidemia and Stroke in Dialysis Patients1.1 Participants and Methods1.1.1 ParticipantsIn this prospective cohort study,590 patients undergoing hemodialysis (HD; n= 285) or peritoneal dialysis (PD; n= 305) were recruited from two dialysis centers (Guangzhou First People’s Hospital and General Hospital of Guangzhou Military Command of PLA) between January 1,2008 and December 30,2012. The follow-up period extended to December 31,2013. The inclusion criteria for this study were all patients≥ 18 years old who received maintenance HD or PD therapy for more than 3 months, except those who had ever received a kidney transplant prior to dialysis, had malignant disease, or refused to give written consent. Patients in whom the dialysis modality changed during the study period were classified as HD or PD according to their initial treatment modality. The study protocol was approved by the Clinical Research Ethics Committee of Guangzhou First People’s Hospital and the Clinical Research Ethics Committee of General Hospital of Guangzhou Military Command of PLA. All patients provided written informed consent before study entry.1.1.2 Study ProtocolThis was a prospective observational cohort study. Baseline demographic data included age, gender,24-h urine output, dry weight, primary cause of ESRD, history of diabetes, hypertension, or CVD, and biochemical parameters, including hemoglobin, serum albumin, albumin-corrected calcium, serum phosphorus, total cholesterol (TC), triglycerides (TG), low density lipoprotein cholesterol(LDL-C), high density lipoprotein cholesterol(HDL-C), serum high sensitive C-reactive protein (hs-CRP), serum uric acid, serum urea nitrogen, serum creatinine, total Kt/V, and residual kidney function (RKF). These data were obtained during the third month of dialysis. All parameters were measured in the clinical laboratory of the two hospitals. Medicine usage data were also collected from the patients’files.The major outcome of the study was stroke, which was defined as a focal neurological deficit of cerebrovascular persisting for>24 h that was diagnosed as an ischemic or hemorrhagic stroke by computed tomography (CT) or magnetic resonance imaging (MRI). They included the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes of 433.xx,434.xx, and 436.xx for infarction and 430.xx,431.xx, and 432.xx for hemorrhage. The quality of diagnosis data fulfilled the criteria of the ideal stroke incidence study suggested by Sudlow and Warlow. In addition, the diagnosis of dyslipidemia at the initiation of dialysis was determined if any one of the following indexes was met:TC level≥ 6.22 mmol/L, TG level≥ 2.26 mmol/L, LDL-C level≥ 4.14 mmol/L, HDL-C level<1.04 mmol/L and/or having received lipid-lowering therapy during the previous two weeks. Diabetes was based on diagnostic criteria from the American Diabetes Association. Fasting glucose≥7.0 mmol/1 or under treatment for treatment for previously diagnosed diabetes diagnosed for diabetes. Hypertension was recorded if the patient was taking any antihypertensive drug with hypertension history or had three separate blood pressure measurements≥ 140/90 mmHg without any antihypertensive drug. CVD was defined that includes myocardial infarction, atherosclerotic heart disease, cardiomyopathy, cardiac arrhythmia, cardiac arrest, congestive heart failure, cerebrovascular accident, ischemic brain damage, anoxic encephalopathy, and peripheral vascular disease. Corrected serum calcium (mmol/L)= total calcium (mmol/L)+0.2* [4-serum albumin concentration (g/L)/10]. Body mass index (BMI) was calculated as weight (kg) divided by height (m) squared. Residual kidney function (RKF) and total Kt/V were calculated using PD adequest software 2.0 (Baxter Healthcare Ltd). The RKF, in milliliters per minute per 1.73 m2, was estimated from mean values of creatinine clearance and urea clearance and adjusted for body surface area.1.1.3 Statistical AnalysisSummary statistics were presented as percentages for categorical data, mean± standard deviation for approximately normally distributed continuous variables, and median (inter quartile range) for skewed continuous variables. According to the primary dialysis modality, patients were divided into two groups:HD and PD. According to stroke events, patients were also divided into two groups:stroke and non-stroke. According to lipid parameters (TC, TG, LDL-C and HDL-C), patients were also divided into two groups:dyslipidemia and normal lipid subgroups. Characteristic differences between two groups were tested by using the chi-square test for categorical variables, Student’s t-test for approximately normally distributed continuous variables, and the non-parametric Mann-Whitney test for skewed continuous variables. Timeline incidence data were analyzed using a Poisson model. Time-to-event analysis of stroke events was performed using Kaplan-Meier survival curves, the Log-Rank test and the Cox proportional hazards model for the HD group compared to the PD group. Multivariate models were constructed sequentially by initially using only the group, then adding demographic characteristics (age at enrollment and gender), followed by adding comorbidities (diabetes, hypertension and cardiovascular disease), and finally by adding laboratory values (albumin-corrected calcium, total triglycerides, hs-CRP,24-h urine output, and RKF). Cox regression analysis was used to evaluate the risk factors for the presence of stroke in the total dialysis population, HD patients, or PD patients at baseline. Two different approaches were applied:(1) univariate Cox regression analysis of the occurrence of stroke using the following variables:age, diabetes, hypertension, cardiovascular disease, albumin-corrected calcium, total triglycerides,24-h urine output, RKF, and aspirin and clopidogrel use, and (2) multivariate Cox regression analysis modeling of covariates using a backward stepwise selection procedure (entry:p≤0.05; removal:p>0.1; the selection criterion was derived from acquiesce in SPSS software system, as well as the estimated clinical importance). Furthermore, Cox regression models were used to evaluate the relationship between total triglycerides tertiles with stroke, initially without adjustment, and subsequently adjusting for several groups of covariates. The multivariate Cox regression model were constructed using eligible covariates that demonstrated significant or near significant association with stroke (p<0.2) on bivariable analysis or for importance of clinical concern. Statistical significance was defined asp<0.05 using two-tailed tests. Statistical analyses were performed using SPSS 17.0 for Windows (SPSS, Chicago, IL, USA).1.2 Results1.2.1 Baseline Cohort CharacteristicsA total of 590 eligible patients (mean age 54.46±15.65 years,61.8% male) undergoing HD (n=285) or PD (n=305) were enrolled in this study. The median follow-up duration was 32.5 months (range:3-71.8). The primary cause of ESRD was glomerular disease (261 patients; 44.2%), followed by diabetic nephropathy (146 patients; 24.7%) and hypertension (96 patients; 16.2%). Compared to PD patients, HD patients more frequently had diabetes and CVD and presented with an older age and a higher hs-CRP, but lower total triglycerides, Kt/V,24-h urine output, and RKF. During the study period,91 (15.4%) patients underwent renal transplantation,30 (5.1%) patients were transferred to other centers,17 (3.0%) patients were lost to follow-up,4 (0.6%) patients declined further treatment, and the remaining 449 (75.9%) patients continued to be followed up.1.2.2 Characteristics of Stroke PatientsA total of 62 stroke events (10.5%) occurred in all patients, including 39 (62.9%) ischemic strokes and 23 (37.1%) hemorrhagic strokes. There were 38 (61.3%) events in the HD group and 24 (38.7%) events in the PD group. The ischemic stroke incidence rate was higher in the HD group than in the PD group (8.7%vs.4.6%; p= 0.039). The overall incidence rate of stroke was 49.2/1,000 patient-years for all patients,74.0/1,000 patient-years in the HD group, and 31.8/1,000 patient-years in the PD group (p<0.001).Compared to non-stroke patients, stroke patients had an older age (65.09±11.27 vs.50.53±16.30;p<0.001), total triglycerides (2.12±1.09 vs.1.62±0.87;p=0.004), and hs-CRP (2.11 [0.68-7.37] vs.1.68 [0.63-6.40];p<0.001). They also had a lower albumin-corrected calcium (2.29±0.39 vs.2.41±0.35;p=0.030),24-h urine output (536±295 vs.680±342; p=0.043), and RKF (2.05±1.26 vs.3.57±2.52;p=0.002). In addition, stroke patients more frequently had the comorbidities of diabetes, hypertension, and cardiovascular disease, and they were more commonly being treated with aspirin and clopidogrel. Furthermore, the cumulative hazard of developing stroke was significantly higher in the HD group than in the PD group (hazard ratio [HR],2.43; 95% confidence interval [CI],1.45-4.06;p<0.001). After adjusting for various potential confounders, there was still a significant association between HD and the risk of stroke (HR,1.75; 95%CI,1.15-3.62;p=0.046). Moreover, compared to PD patients, HD patients had a higher risk of ischemic stroke based on Kaplan-Meier curves and Cox regression analysis (HR,2.62; 95% CI, 1.56-4.58; p=0.002). However, the risk of hemorrhagic stroke did not differ between the two groups.1.2.3 Characteristics of Dyslipidemia PatientsThere were 131 patients with dyslipidemia in all long-term dialysis patients, accounting for 22.2%. Compared to normal lipid patients, dyslipidemia patients were more famale, more likely older (55.03±14.29 vs.51.16±16.93, P=0.009), with a higher hemoglobin (102.26±18.91 vs.94.05±17.93, P<0.001), albumin-corrected calcium (2.25±0.22 vs.2.17±0.25, P=0.001), hs-CRP (2.12[0.75-7.46] vs.1.72[0.62-6.13], P=0.001). In addition, dyslipidemia patients had a higher triglycerides (2.44±1.27 vs.1.35±0.41,p<0.001), total cholesterol (5.54±1.49 vs.4.39±0.84, P<0.001), low density lipoprotein cholesterol (3.64±0.87 vs.1.48±0.56, P<0.001), but high density lipoprotein cholesterol level did not differ between the two groups. Moreover, compared to normal lipid patients, dyslipidemia patients more frequently had the comorbidities of diabetes and cardiovascular disease. However, treatment with aspirin and clopidogrel did not differ between the two groups.1.2.4 Risk Factors for StrokeClinical and laboratory variables that were statistically different between stroke group and non-stroke group, were included in the Cox regression analysis. The multivariate analysis model identified these factors as being independently associated with an increased risk of stroke:older age (HR,1.05; 95%CI,1.02-1.09; p=0.003), diabetes (HR,1.98; 95%CI,1.31-3.46;p=0.001), CVD (HR,2.06; 95%CI, 1.62-3.05;p<0.001), higher triglycerides (HR,1.20; 95%CI,1.08-1.58;p= 0.034), and HD (with PD as the reference; HR,2.03; 95%CI,1.46-3.89; p= 0.005). In addition to the common stroke risk factors of older age, diabetes, and CVD that were predictors for both groups, lower albumin-corrected calcium (HR,0.92; 95%CI, 0.74-0.96; p= 0.035) was an independent prognostic predictor for stroke in HD patients, whereas higher triglycerides (HR,1.28; 95%CI,1.07-1.62;p=0.016) was an independent risk factor for stroke in PD patients.1.2.5 Relationship between Dyslipidemia and StrokeThe multivariate analysis model identified higher triglycerides as an independent risk of stroke in all dialysis patients. Furthermore, the Cox regression analysis of triglycerides for stroke in all dialysis patients, we found a J-shaped effect of total triglycerides with stroke. Adjusted for age, sex, BMI, comorbidities, hemoglobin, serum albumin, albumin-corrected calcium, total cholesterol, hs-CRP, 24-h urine output, residual kidney function, compared with the Tertile Ⅱ (≥ 1.09-≤ 1.48 mmol/L) as a reference, hazard ratios of stroke were 1.12 (95%　confidence interval [CI],0.88-1.41) for Tertile Ⅰ (<1.09 mmol/L),1.04 (95%　CI,0.90-1.28) for Tertile Ⅲ (>1.48-≤1.90 mmol/L), and 1.53 (95% CI,1.15-2.18) for Tertile IV (> 1.90 mmol/L).1.3 ConclusionsIn conclusion, we found that patients undergoing HD had a significantly higher risk of stroke, with an adjusted hazard ratio of 1.75 (1.15-3.62) compared to patients undergoing PD. This increased risk was especially apparent for the subtype of ischemic stroke. The risk factors for stroke for all dialysis patients were older age, diabetes, and CVD, whereas lower albumin-corrected calcium was a predictor in HD patients and higher triglyceride levels were associated with an increased risk of stroke in PD patients. Although different dialysis modalities involve different mechanisms of pathophysiology, comprehensive control of diabetes, CVD, calcium-phosphorus metabolism, and triglyceride levels may be an effective preventive strategy for stroke in dialysis patients. Dyslipidemia was not uncommon in dialysis patients. A J-shaped effect of total triglycerides was found with stroke in all dialysis patients.2. the Association between CKD and Dyslipidemia2.1 Participants and Methods2.1.1 study populationThe study data was extracted from a population-based, cross-sectional survey conducted between June,2012, and October,2012 in Wanzhai Town, Zhuhai City, a coast city in Southern China. A total of 1-834 residents (aged 18 years or older) participated in the survey were qualified. This study was approved by the Ethics Committee of the Third Affiliated Hospital of Southern Medical University, Guangzhou. All participants signed written informed consent forms. The present study was conducted based on the principles of the Declaration of Helsinki.2.1.2 Data collectionData regarding participants’ demographic characteristics and details regarding lifestyle (current or past cigarette smoking, alcohol intake, diet habits, educational status and physical activity), medical histories (coronary artery disease, hypertension, and diabetes), were obtained through a standardized questionnaire.2.1.3 Physical measuresAnthropometric indices such as height, weight, waist circumference were measured in the community clinics by physicians, medical students and nurses, who had received intensive training. Body weight, height, waist circumference and blood pressure were measured in the morning between 08:00 and 11:00 am. Blood pressure (BP) was measured in the sitting position after resting for at least 5 minutes. The average value was calculated after three different consecutive times’measurement of BP. Waist circumference was taken midway between the last rib and iliac crest with the participants standing with light garments and breathing out gently. The waist circumference were taken twice and the average value was recorded to the nearest 0.1 cm. Body mass index (BMI) was calculated as weight (kg) divided by height squared (m2).2.1.4 Laboratory assaysBlood specimens were collected after an overnight fast. First morning urine samples were collected from all participants. Women who were actively menstruating and individuals having urinary tract infection symptoms were excluded from the urine test. All blood samples and urine samples were stored at 4℃ and transported to the central laboratory in the Third Affiliated Hospital of Southern Medical University for later laboratory analysis in 3 hours from collection sites. Urinary albumin and creatinine were measured from a fresh morning spot urine sample. The urinary albumin to creatinine ratio (ACR, mg/g) was calculated.2.1.5 Determination of dyslipidemiaDyslipidemia is defined by the presence of at least one of the following:serum total cholesterol (TC) level≥ 6.22mmol/L, triglyceride (TG) level≥ 2.26mmol/L, low density lipoprotein cholesterol (LDL-C) level≥ 4.14 mmol/L, and high density lipoprotein cholesterol (HDL-C) level< 1.04 mmol/L, and/or having received lipid-lowering therapy during the previous 2 weeks. Non-HDL=TC-HDL.2.1.6 Determination of CKDThe estimated glomerular filtration rate (eGFR), an indicator of kidney function, was estimated using a formula from the Chinese-Modification of Diet Renal Disease (C-MDRD) study:GFR (ml/min/1.73m2)= 175x (Scr)-1234 × (Age)-0.179 × (if female, x0.79). Reduced renal function was defined as an eGFR of less than 60 ml/min per 1.73m2. For practical purposes, albuminuria was defined as a spot urinary albumin-to-creatinine ratio higher than 30 mg/g. CKD was defined as an eGFR of less than 60 ml/min per 1.73 m2 or albuminuria.2.1.7 Determination of Hypertension and Diabetessystolic BP≥140 mmHg or diastolic BP≥90 mmHg or under treatment for hypertension diagnosed for hypertension. Fasting glucose≥7.0 mmol/l or under treatment for treatment for previously diagnosed diabetes.2.1.8 Statistical AnalysisData were analyzed using SPSS (version 19.0). Summary statistics were presented as percentages for categorical data, mean ± standard deviation for approximately normally distributed continuous variables, and median (inter quartile range) for skewed continuous variables. The continuous variables were analyzed by t-test or Wilcoxon rank-sum test and the categorical variables were analyzed by the chi-squared test or Fisher’s exact test. All participants were divided into two subgroups:CKD subgroup and Non-CKD subgroup. Baseline characteristics of CKD subgroup and Non-CKD subgroup were examined. Logistic regression models were used to examine whether dyslipidemia is associated CKD. Model one was unadjusted. Model two was adjusted for age and sex. Model three was adjusted for age, sex, history of coronary heart disease, hypertension, diabetes, smoking, alcohol use, education status, physical activity and BMI. The odds ratios (OR) and the corresponding 95%　confidence intervals (95%CI) were calculated. All statistical tests were 2-sided, and p< 0.05 was considered statistically significant.2.2 Result2.2.1 Baseline characteristics of CKD and non-CKD subgroupsAmong the total participants,230 (12.5%) participants had CKD. Compared to participants without chronic kidney disease, participants with chronic kidney disease were more likely older, with a higher prevalence of history hypertension, higher prevalence of history coronary heart disease, higher prevalence of diabetes mellitus, lower education status, higher BMI, higher waist circumference, higher systolic blood pressure and diastolic blood pressure, higher fasting blood glucose, higher serum total cholesterol, higher serum triglycerides, higher serum low density lipoprotein cholesterol, higher C-reactive protein, higher serum creatinine, higher serum uric acid, higher blood urea nitrogen, higher ACR and lower eGFR. Dyslipidemia were more prevalent among patients with CKD than among those without CKD (45.8 vs 30.7, P<0.001). While, gender, smoking status, drinking status, physical exercise, and serum high density lipoprotein cholesterol did not differ between the two subgroups.2.2.2 Baseline characteristics of dyslipidemia and normal lipid subgroupsAmong the total participants,624 (30.4%) participants had dyslipidemia. Compared to participants with normal lipid, participants with dyslipidemia be were more likely older, more male, with a higher prevalence of hypertension, higher prevalence of diabetes mellitus, lower education status, more smokers, more drinkers, higher BMI, higher waist circumference, higher systolic blood pressure and diastolic blood pressure, higher fasting blood glucose, higher serum total cholesterol, higher serum triglycerides, higher low density lipoprotein cholesterol, higher C-reactive protein, higher serum creatinine, higher serum uric acid, higher blood urea nitrogen, higher ACR and lower eGFR. CKD were more prevalent among patients with dyslipidemia than among those with normal lipid (20.9% vs 12.1%, P<0.001). While, history of coronary heart disease, physical exercise and high density lipoprotein cholesterol did not differ between the two subgroups.2.2.3 The relationship between dyslipidemia and chronic kidney diseaseLogistic regression analysis was applied to examine whether dyslipidemia was associated with CKD. Dyslipidemia was associated with CKD in the unadjusted analyses. The odd ratio was 2.07 (95%CI 1.56-2.73, P<0.001). Further adjusted for age and gender, dyslipidemia was also associated with CKD. The odd ratio was 1.72 (95%CI 1.29-2.29, P<0.001). When further adjusted for history of coronary heart disease, history of hypertension, history of diabetes, smoking status, drinking status, education status (≥ high school), physical inactivity and BMI, the correlation still significant (OR 1.59,95%　CI 1.14-2.21, P=0.006).According to "Guidelines on Prevention and Treatment of Blood Lipid Abnormality in Chinese Adults" in 2007, we further divided triglyceride (the first group TG<1.70mmol/L, the second group 2.26>TG≥1.70mmol/L, the three group TG≥2.26mmol/L), total cholesterol (the first group TC<5.18mmol/L, the second group 6.22>TC≥5.18mmol/L, the three group TC≥6.22mmol/L), and low density lipoprotein cholesterol (the first group LDL-C<3.37mmol/L, the second group 4.14> LDL-C≥3.37mmol/L, the three group LDL-C≥4.14mmol/L) into three groups, respectively. Multivariate logistic regression was used to explore its relationship with chronic kidney disease.2.2.4 The relationship between triglyceride and chronic kidney diseaseThe OR for CKD increased with increasing of triglyceride, and a statistically significant trend was observed in subjects of the second group and the third group compared to the first group (TG<1.70mmol/L). In the unadjusted model, triglyceride was significantly associated with CKD (OR 1.83,95%CI 1.23-2.73, P=0.003, comparing the second group to the first group, and OR 2.14,95%CI 1.53-3.00, P< 0.001, comparing the third group to the first group). Further adjusted for age and gender, the relationship between triglyceride and CKD still has significance (OR 1.65, 95%CI 1.09-2.48, P=0.017, comparing the second group to the first group, and OR 1.91,95%CI 1.35-2.69, P<0.001, comparing the third group to the first group). Further adjusted for history of coronary heart disease, hypertension, diabetes, smoking status, drinking, drinking status, education status, physical inactivity and BMI, the relationship between triglyceride and CKD still has significance in the third group (OR 1.55,95%CI 1.04-2.30, P= 0.032, compared to the first group).2.2.5 The relationship between total cholesterol and chronic kidney diseaseThe OR for CKD increased with increasing of total cholesterol, but a statistically significant trend was observed only in subjects of the third group compared to the first group (TC<5.18 mmol/L). In the unadjusted model, total cholesterol was significantly associated with CKD (OR 1.54,95%CI 1.11-2.14, P=0.01, comparing the second group to the first group, and OR 2.18,95%CI 1.53-3.10, P<0.001, comparing the third group to the first group). Further adjusted for age and gender, the relationship between total cholesterol and CKD still has significance only in the third group (OR 1.63,95%CI 1.13-2.35, P=0.009, compared to the first group). Further adjusted for history of coronary heart disease, hypertension, diabetes, smoking status, drinking, drinking status, education status (≥ high school), physical inactivity and BMI, the relationship between total cholesterol and CKD still has significance in the third group (OR 1.68,95%CI 1.11-2.55, P=0.014, compared to the first group).2.2.6 The relationship between low density lipoprotein cholesterol and chronic kidney diseaseIn the unadjusted model, the relationship between low density lipoprotein cholesterol and CKD was significance only in the third group (OR 1.63,95%CI 1.12-2.38,.P=0.010, compared to the first group). However, the association was abolished when adding age and gender to the model (OR 1.22,95%CI 0.83-1.80, P=0.313, compared to the first group).2.2.7 The association between Non-HDL/HDL and chronic kidney diseaseNon-HDL/HDL was divided into three group, according to its three quantile (Q1 ≤2.21,2.96≥Q2>2.21, Q3>2.96). Logistic regression analysis was applied to examine whether Non-HDL/HDL was associated with CKD. In the unadjusted model, Non-HDL/HDL was associated with CKD only in the third group (OR 2.08,95%CI 1.46-2.96, P<0.001). Further adjusted for age and gender, the relationship still has significance (OR 1.56,95%CI 1.09-2.25, P=0.016). However, when adding history of coronary heart disease, hypertension, diabetes, smoking status, drinking, drinking status, education status (≥ high school), physical inactivity and BMI to the model, the association was abolished (OR 1.15,95%CI 0.75-1.76, P＝0.533).2.3 ConclusionIn conclusion, dyslipidemia was associated with chronic kidney disease, the association independent to age, sex, hypertension, diabetes, coronary heart disease, smoking, drinking, cultural degree, lack of physical exercise and BMI. Multiple logistic regression model was used to explore the relationship among total cholesterol, triglyceride, low density lipoprotein cholesterol (LDL-C) and chronic kidney disease. We found that only elevated triglyceride and total cholesterol were associated with chronic kidney disease. However, neither LDL-C nor Non-HDL/HDL was an independent risk factor for chronic kidney disease.