Prevalence of High Non-high-density Lipoprotein Cholesterol and Associated Risk Factors in Patients with Diabetes Mellitus in Jilin Province， China： A Cross-sectional Study*

HE Huan， ZHEN Qing， LI Yong， KOU Chang Gui， TAO Yu Chun， WANG Chang， Joseph Sam Kanu，LU Yu Ping， YU Ming Xi， ZHANG Hui Ping， YU Ya Qin， LI Bo，#， and LIU Ya Wen，#



Prevalence of High Non-high-density Lipoprotein Cholesterol and Associated Risk Factors in Patients with Diabetes Mellitus in Jilin Province， China： A Cross-sectional Study*

HE Huan1，¶， ZHEN Qing1，¶， LI Yong1， KOU Chang Gui1， TAO Yu Chun1， WANG Chang1， Joseph Sam Kanu1，LU Yu Ping1， YU Ming Xi1， ZHANG Hui Ping2， YU Ya Qin1， LI Bo1，#， and LIU Ya Wen1，#

Dyslipidemia is a risk factor for cardiovascular diseases （CVDs） in patients with diabetes， and non-high-density lipoprotein cholesterol （non-HDL-C） is a better predictor of CVDs than low-density lipoprotein cholesterol （LDL-C） in patients with diabetes. Therefore， we aimed to investigate the distribution of non-HDL-C and the prevalence of high non-HDL-C level in Chinese patients with diabetes mellitus and identify the associated risk factors. Non-HDL-C concentration positively correlated with total cholesterol，triglycerides， and LDL-C concentrations. Although both non-HDL-C and LDL-C concentration both related positively with TC concentration， the magnitude of correlation was relatively higher for non-HDL-C. The prevalence of high non-HDL-C（≥4.14 mmol/L） was higher in two age groups （55-64 years： 46.7%； 65-79 years： 47.3%） than other age groups （18-24 years： 4.2%； 25-34 years： 43.6%；35-44 years： 38.1%； 45-54 years： 41.0%）. It was also higher among overweight （45.1%）， generally obese （50.9%）， or abdominally obese （47.3%） subjects，compared with normal weight subjects （34.5%）. The risk of high non-HDL-C increased with advancing age. Both general obesity ［odds ratio （OR）=1.488，95% confidence interval （CI）： 1.003-2.209］ and abdominal obesity （OR=1.561， 95% CI： 1.101-2.214）were significantly associated with high non-HDL-C levels.

Diabetes is associated with a greater risk of morbidity and mortality from cardiovascular diseases （CVDs）［1］. The coexistence of diabetes and other risk factors， in particular dyslipidemia， further increases the risk of CVDs［2］， and ‘lipotoxic' dyslipidemia can accelerate the manifestation and persistence of complications of diabetes［3］. The United Kingdom Prospective Diabetes Study （UKPDS） also showed that dyslipidemia was a risk factor for CVDs in patients with diabetes［4］. Previous studies demonstrated that an elevated level of low-density lipoprotein cholesterol （LDL-C） was a powerful coronary heart disease risk factor and recommended using LDL-C as the primary marker to guide therapy［5］. The Third Adult Treatment Panel （ATP III） of the National Cholesterol Education Program （NCEP）suggested that non-high-density lipoprotein cholesterol （non-HDL-C） be used as a secondary target for lipid control［5］. A recent study also reported that non-HDL-C was a better predictor of CVDs than LDL-C in patients with diabetes［6］. Estimation of non-HDL-C levels， which is calculated by subtracting HDL-C from total cholesterol （TC），does not require measurement of triglyceride （TG） or LDL-C levels， and it is not influenced by plasma TG concentration［5］. Non-HDL-C reflects the concentrations of both cholesterol-rich and TG-rich atherogenic apolipoprotein B-containing lipoproteins，including very-low-density lipoprotein cholesterol （VLDL-C）， intermediate-density lipoprotein cholesterol （IDL-C）， LDL， and lipoprotein （a）［5］. Most studies on non-HDL-C are performed for normal populations. Very few cross-sectional studies have been performed to investigate the prevalence of high non-HDL-C and associated factors among patients with diabetes in China. In this study， we aimed to describe non-HDL-C levels among patients with diabetes mellitus， examine the prevalence of high non-HDL-C levels and the risk factors influencing high non-HDL-C levels in patients with diabetes mellitus in Jilin Province， China.

Data for this study were collected from the Survey of Chronic Diseases and Associated Risk Factors among Adults in Jilin Province， China in 2012. The survey was a cross-sectional study that used a multistage cluster random sampling design to select a representative sample of permanent residents aged 18 to 79 years in nine different cities in Jilin Province. The survey is described in detail elsewhere［7］. A total of 16，975 participants completed the survey. Participants with incomplete information on blood glucose or lipid levels were excluded from the present study. Our study sample consisted of 1825 patients with diabetes mellitus，which was defined as fasting plasma glucose （FPG）level ≥7.0 mmol/L， 2-h post prandial plasma glucose （2hPG） concentration ≥11.1 mmol/L， patient administering anti-diabetic medications， or having a history of diabetes mellitus in the past one year.

To account for the complex sampling design，prevalence rates were weighted to produce representative estimates of the total population of Jilin Province. The weights were obtained from the 2010 Chinese population census data and calculated based on four factors： geographical region，residential type （urban/rural）， sex， and age.

Of 1825 patients with diabetes mellitus （representing 2，074，217 subjects）， 54.4% resided in urban areas while 45.6% resided in rural areas. There was a 56.4% male predominance. The mean age of subjects was 53.20±0.55 years， with 29.0% of them aged 55-64 years， 28.1% aged 45-54 years， and 1.7% aged 18-24 years. Other characteristics of the study sample population are shown in Table S1 （See the BES website： www. besjournal.com）.

In our study， the weighted mean levels of TC， TG，LDL-C， HDL-C， and non-HDL-C were 5.31±0.04 mmol/L， 2.98±0.09 mmol/L， 3.11±0.03 mmol/L，2.13±0.10 mmol/L， and 4.07±0.04 mmol/L，respectively （Table 1）. These values were higher than those reported for the general and normal population［8］， and these levels are similar to the levels reported by Wagner et al.［9］for patients with type 2 diabetes.

A positive correlation between non-HDL-C and TG concentrations was observed （r=0.531， P＜0.001）（Figure 1A）， but the correlation between LDL-C and TG concentrations was not significant （r=-0.040，P=0.087） （Table S2， see the BES website：www.besjournal.com）. Although both non-HDL-C （r=0.947， P＜0.001） （Figure 1B， see the BES website：www.besjournal.com） and LDL-C （r=0.717， P＜0.001；Table S2） concentration positively correlated with TC， the magnitude of correlation was relatively higher for the former. Non-HDL-C concentration significantly correlated with LDL-C concentration （r=0.657， P＜0.001） （Figure 1C）， and non-HDL-C exhibited a significant inverse relationship to HDL cholesterol （r=-0.101， P＜0.001） （Figure 1D）.

Table 1. Blood Glucose Concentration and Serum Lipid Levels of Study Subjects in Jilin Province， China，2012

The present data show that non-HDL-C concentration， unlike LDL-C， better correlated with TC concentration. Because non-HDL-C reflects the concentrations of both cholesterol-rich and triglyceride-rich atherogenic apolipoprotein B-containing lipoproteins， including VLDL-C， IDL-C，LDL， and lipoprotein （a）， compared with LDL-C，non-HDL-C could be a better atherogenic index for patients with diabetes and hypercholesterolemia and/or hyper triglyceridemia.

According to the NCEP-ATPIII criteria， non-HDL-C levels are routinely calculated as TC levels minus HDL-C level. The reasonable non-HDL-C concentration was 0.78 mmol/L （30 mg/dL higher than LDL-C concentration［5］. Based on the subdivided LDL-C cut-off point， we categorized non-HDL-C concentrations into four levels： ＜3.37 mmol/L （130 mg/dL）， 3.37-4.13 mmol/L （130-159 mg/dL），4.14-4.90 mmol/L （160-189 mg/dL）， and ≥4.92 mmol/L （190 mg/dL）. We defined high non-HDL-C concentrations as those ＞4.14 mmol/L （160 mg/dL）. As shown in Table S3 （See the BES website： www. besjournal.com）， the non-HDL-concentration for 28.9% of subjects was in the range of 3.37-4.13 mmol/L， for23.6% in the range of 4.14-4.91 mmol/L， and for 19.4% it was ≥4.92 mmol/L. The overall prevalence of high non-HDL-C was 43.0% ［95 confidence interval （CI）%： 33.9%-46.3%］ （Table S4， See the BES website：www.besjournal.com）. The odds ratios （ORs） for high non-HDL-C increased with age. Compared to the normal weight group， the OR for high non-HDL-C in the overweight and obesity groups were 1.560 （95% CI： 1.119-2.175） and 1.973 （95% CI： 1.348-2.887），respectively. In the abdominal obesity group， the OR for non-HDL-C was 2.064 （95% CI： 1.436， 2.966）. However， the ORs for high non-HDL-C based on variables such as having a family history of dyslipidemia， administering lipid-modifying medication， smoking， drinking， and physical exercise，were not significantly different.

Measurements TG-rich lipoproteins， which are included in non-HDL-C measurements， are of a great significance in clinic because these lipoproteins tend to increase as people advance in age and become more obese， insulin resistant， and hyperglycemic［10］. Moreover， with the increasing prevalence of obesity， metabolic syndrome， and diabetes among adults， using non-HDL-C levels to predict CVDs may become more important［6］. A study by Banu et al.［11］showed that the ages of pre-diabetes and diabetes patients were associated with non-HDL-C levels. Our study also showed that the risk of high non-HDL-C increased with advancing age， but subjects aged 25-34 years had a greater risk of high non-HDL-C （OR=16.628， 95% CI：1.503-183.994） than two other proceeding age groups： 35-44 years （OR=11.973， 95% CI：1.351-106.130） and 45-54 years （OR=12.981， 95% CI：1.486-113.370） （Table 2）. This is possibly because the proportion of obese participants in the 25-34 years age group （33.0%） was greater than that in other age groups （less than 29.0%）.in the multivariate logistic regression model.

Table 2. Multivariate Logistic Regression Analysis of Factors Influencing High non-HDL-C Levels in Study Subjects in Jilin Province， China， 2012

Multivariate logistic regression analysis results from the present study suggested that obesity （OR=1.488， 95% CI： 1.003-2.209） and abdominal obesity （OR=1.561， 95% CI： 1.101-2.214） were significantly associated with the prevalence of high non-HDL-C （Table 2）. The finding is similar to that from a study conducted by Ram et al.［12］， which showed that with increased age and obesity， the degree of insulin resistance becomes more serious in patients with diabetes， leading to a higher prevalence of lipid abnormalities， and most diabetes patients have insulin resistance and lipid metabolism disorders.

In conclusion， our study showed that non-HDL-C concentration positively correlated with TC， TG， and LDL-C concentration. Although both non-HDL-C and LDL-C concentration correlated positively with TC level， the magnitude of correlation was relatively higher for non-HDL-C. The prevalence of high non-HDL-C varied according to age， BMI， and waist circumference in Chinese patients with diabetes mellitus. Our results suggest that we should pay more attention to the level of non-HDL-C in diabetics，especially for older subjects and those who are more obese or abdominally obese. These subjects should monitor and control their blood non-HDL-C levels to prevent complications and deterioration due to diabetes as well as the occurrence of CVDs.

ACKNOWLEDGEMENTS

We thank WU Yan Hua， TANG Yuan， YAN Zhi Hui，LYU Le and LIU Fan Yu as investigators participated in this study.

¶These authors contributed equally to this work.

#Correspondence should be addressed to LI Bo， Tel：86-433-85619451， E-mail： li_bo@jlu.edu.cn； LIU Ya Wen，Tel： 86-433-85619419， E-mail： ywliu@jlu.edu.cn

Biographical notes of the first authors： HE Huan，female， born in 1988， Master of medicine， majoring in chronic epidemiology； ZHEN Qing， female， born in 1988，Associate Professor， majoring in infect edpideminology.

1. Davidson MH. Cardiovascular risk factors in a patient with diabetes mellitus and coronary artery disease： therapeutic approaches to improve outcomes： perspectives of a preventive cardiologist. Am J Cardiol， 2012； 110， 43B-9B.

2. Jaiswal M， Schinske A， Pop-Busui R. Lipids and lipid management in diabetes. Best Pract Res Clin Endocrinol Metab，2014； 28， 325-38.

3. Unger RH. Lipotoxic diseases. Annu Rev Med， 2002； 53， 319-36.

4. Turner RC， Millns H， Neil HA， et al. Risk factors for coronary artery disease in non-insulin dependent diabetes mellitus：United Kingdom Prospective Diabetes Study （UKPDS： 23）. BMJ（Clinical research ed）， 1998； 316， 823-8.

5. National Cholesterol Education Program （NCEP） Expert Panel on Detection， Evaluation， and Treatment of High Blood Cholesterol in Adults （Adult TreatmentPanel III）. Third Report of the National Cholesterol Education Program （NCEP） Expert Panel on Detection， Evaluation， and Treatment of High Blood Cholesterol in Adults （Adult Treatment Panel III） final report. Circulation， 2002； 106， 3143-421.

6. Kilgore M， Muntner P， Woolley JM， et al. Discordance between high non-HDL cholesterol and high LDL-cholesterol among US adults. J Clin Lipidol， 2014； 8， 86-93.

7. Shi-bin W， Chang-gui K， Ya-wen L， et al. Analysis on prevalence rate and disease spectrum of chronic disease in adult residents in Jilin province. Journal of Jilin University （Med Ed）， 2013； 39，1045-50. （In Chinese）

8. Wang Y， Chen X， Yuan Q. Analysis on Levels of Non-high Density Lipoproterin Cholesterol and atherogenic Index of Plasma in Normal Population. Zhejiang Prev Med， 2008； 20，10-4. （In Chinese）

9. Wagner AM， Perez A， Zapico E， et al. Non-HDL cholesterol and apolipoprotein B in the dyslipidemic classification of type 2 diabetic patients. Diabetes care， 2003； 26， 2048-51.

10.Blaha MJ， Blumenthal RS， Brinton EA， et al. The importance of non-HDL cholesterol reporting in lipid management. J Clin Lipidol， 2008； 2， 267-73.

11.Banu S， Jabir NR， Manjunath NC， et al. Comparative study of non-high density lipoproteins cholesterol level and lipid profile in pre-diabetic and diabetic patients. CNS Neurol Disord Drug Targets， 2014； 13， 402-7.

12.Ram N， Ahmed B， Hashmi F， et al. Importance of measuring non-HDL cholesterol in type 2 diabetes patients. J Pak Med Assoc， 2014； 64， 124-8.

10.3967/bes2016.071

December 28， 2015；

*This work was funded by the Scientific Research Foundation of the Health Bureau of Jilin Province， China （grant number： 2011Z116） and the National Natural Science Foundation of China with grant （grant number： 81573230）.

1. Department of Epidemiology and Biostatistics， School of Public Health， Jilin University， Changchun 130021， Jinlin，China； 2. Department of Psychiatry， Yale University School of Medicine， VA Medical Center， West Haven， CT 06516， USA

Accepted： May 20， 2016