What is the Impact of Sex on Cardiovascular Disease Risk Factors in Patients with Chronic Kidney Diseases in China: A Cross-sectional Study

Background: It is controversial about the sex differences in the association of chronic kidney diseases (CKD) and cardiovascular disease (CVD) risk. Thus, we examined CVD risk makers of CKD and non-CKD men and women in China, especially some “non-traditional” ones. Methods: This cross-sectional study used 7999 participants from the China Health and Nutrition Survey in 2009. This study examined the “traditional” risk factors and of CVD, such as lipoprotein cholesterol (LDL-C), total cholesterol (TC) and non-high density lipoprotein cholesterol(non-HDL-C).Also, the “non-traditional” risk factors of CVD were calculated, such as lipoprotein (a) (Lp(a)), white blood cell (WBC) count, visceral adiposity index (VAI) and lipid accumulation product (LAP). Results: Compared with men with CKD, higher levels of TC and LDL-C were observed in women with CKD. Furthermore, compared with men with CKD, the relative difference of WBC count was greater between women with CKD and their non-CKD ones. Meanwhile, the level of LAP and VAI of women with CKD were higher than men with CKD, which indicate the visceral obese. We also observed that the sex by CKD status interactions were statistically signicant for TC, LDL-C, non-HDL-C, LAP, VAI and Lp(a) (all p <0.05). After adjusted the covariates, the sex differences effect on CVD risk factors among CKD patients couldn’t be eliminated as well. Conclusions: In CKD situation, women had greater lipid proles and put on more visceral adiposity than men, which may indicate a higher CVD risk of women with CKD.

geography. The Lots of institutions supported this survey as mentioned in previous study (17) and all participant were provided a written informed consent (18).
Since CHNS began to collect the fasting blood in 2009, data from CHNS 2009 were used in this study. At the 2009exam, 10243 adult respondents were surveyed with 1120 did not give blood. Meanwhile, there 64 participants were pregnant and 805 participants' age was under 18, there 8254 participants with fasting blood samples nally. There also some different exclusion criteria including participants without the information of body mass index (BMI), age or waist circumference (WC) and with extreme triglycerides (TG) (> 500 mg/dl), BMI (≥ 40 kg/m 2 ) or high-density lipoprotein cholesterol (HDL-C) (> 100 mg/dl) values. At the same time, in order to eliminate the different effects caused by the curing of CVD risk factors on the result of this study between different genders, participants were using anti-diabetic agent, lipid-lowering agent and anti-hypertension agent were excluded as well. This analysis included 7999 participants in the end.

Measurements
The stuff of CHNS used standard protocols from the World Health Organization (WHO) to measure participants' height, weight, systolic/diastolic BP and WC. Also, they used a calibrated beam scale to measure the participants' weight with a light clothing and portable SECA stadiometer to measure height. Using an inelastic tape to measure WC at middle between the top of the exhalation ilium and the bottom of chest. BMI was equal to the value using square of height (in meters) to divide weight (in kilograms).
Meanwhile, trained technicians used mercury manometers to measure BP after a rest of ten minutes three times and averaging the three readings of BP in the end.

Measurements of biochemical
All tested samples were collected after a more than 8-hour overnight fast. We analyzed total samples with strict quality control in a lab of Beijing. Picric acid method was used to measure serum creatinine (Scr). GODPAP method was used to measure fasting plasm glucose (FPG). Hitachi 7600 automated analyzer was used to measure all lipids parameters, including, TG, low density lipoprotein cholesterol [LDL-C] and HDL-C. The value of non-HDL-C equaled to TC minus HDL-C. Measured Lp (a) with an immunoturbidimetric method. Also, apolipoprotein B (apo B), apolipoprotein a1 (apo a1), alanine aminotransferase (ALT), uric acid (UA), fasting insulin concentration and hypersensitive C-reactive protein (hs-CRP) were measured accurately with proper methods. Details on these procedures have been described previously (17). With drawing a conclusion of data from Chinese chronic kidney disease, estimated glomerular ltration rate (eGFR) was calculated with an equation developed by adaptation of the

Statistical analysis
We used SPSS software (version 25.0 for windows; SPSS, Chicago, IL, USA) to conducted the statistical analysis. And control persons were those participants who had eGFR≥60ml/min/1.73 m 2 . All continuous variables were calculated as mean ± standard deviation (SD). We used two-way analysis of covariance to computed CKD status and gender as the two main effects in ve tables in this study. We also conducted the statistical analysis of the association of sex × CKD interactions with CVD risk factors. A logtransformed (natural logarithm) was used in insulin to approximate normal distributions before conducting analyses. A two-sided Pvalue < 0.05 was regarded as statistically signi cant.  Table 1 showed the non-adjusted mean values of CVD risk factors of CKD status in different genders. We can found an increase of CVD risk pro le, such as TG, TC, LDL-C, TG/ HDL-C, Non-HDL-C, HOMA-IR, TyG, VAI, LAP and so on, in men and women with CKD in which compared with their non-CKD participants. Meanwhile, in CKD patients, we found that the TC, LDL-C, non-HDL-C and Apo B of women were signi cantly higher than men. Notably, between CKD patients and non-CKD people, women were more diverse than men in terms of WBC values. In addition, women with CKD have higher level of Lp(a), VAI and LAP compared with men with CKD.

Results
There was no difference of BMI between patients with CKD compared with their non-CKD ones. Besides, compared with non-CKD men, men with CKD had a lower level of ferritin, which was opposite to women.  Since CKD is related with age and CVD is related with BMI, then we adjusted age and BMI in Table 2. As the table showed, the mean levels of TC, LDL-C, Non-HDL-C and Apo B of women with CKD were greater than men with CKD under the adjusted of age and BMI. Besides, the differences of WBC between women with CKD and non-CKD ones were more striking than men. Moreover, compared with men with CKD, women with CKD have higher level of Lp(a), VAI and LAP. However, there was little statistical evidence about sex heterogeneity in the association of CKD with TG, TG/HDL, Apo B/apo a1, FPG, HbA1c, HOMA-IR, TyG, ALT and hs-CRP. Also, we observed some interactions between sex with CKD status with statistically signi cant in some CVD risk factors, such as TC, HDL-C, LDL-C, non-HDL-C, Lp (a), LAP and VAI(all p values < 0.05). After further adjustment for HOMA-IR, there was a little change between women with CKD and men with CKD about the difference of these risk factors (Table 3).     The visceral adiposity has a closely relationship with metabolic disorders. Thus, Table 4 adjusted for VAI instead of BMI as a represent of visceral adiposity. Women with CKD had signi cantly higher levels of TC, LDL-C, non-HDL-C and Apo B than their men counterparts with CKD. The differences of WBC between women with CKD and non-CKD ones was much more signi cantly than men. Also, women with CKD have higher level of Lp(a) and LAP compared with men with CKD. Besides, the magnitude of differences in HbA1c, ApoB/Apo a1, FPG, Hs-CRP and WBC between different genders with CKD were less marked. The Hb of women with CKD was lower compared with men with CKD. The interactions of sex × CKD mentioned in Table 2 remained signi cant still. After adjusting the HOMA-IR (Table 5), the difference of these CVD factors between women with CKD and men with CKD remained essentially unchanged.

Discussion
This was a study of the differences between men and women in the comparison of traditional and "non-traditional" CVD markers in CKD patients. Compared with men, women were more likely to have an atherogenic lipid pattern and visceral obese in patients with CKD. Besides, in comparison with the non-CKD counterparts, women with CKD showed greater differences in the value of WBC than men with CKD. The sex differences effect on CVD risk factors of CKD patients and non-CKD ones can't be eliminated considering with BMI, VAI or HOMA-IR or both.
The level of traditional lipid markers, such as TC, were found to be positively associated with an increased risk of CVD, coronary heart disease (CHD) and cardiac death (CHD plus heart failure) (24). In this research, the value of TC in women with CKD was higher than men. Previous research showed that higher level of LDL-C, non-HDL-C and Apo B indicated a higher incidence of future cardiovascular events (25). In this study, the level of LDL-C, non-HDL-C and Apo B of women with CKD were higher than men with CKD as well, which indicate a worse situation of risk factors of CVD of women with CKD. Thus, the traditional lipid markers of women may need more attention.
In this study, we observed that Lp (a) of women with CKD was signi cantly higher than that of men with CKD. Lp (a) is one of the cardiovascular factors that have been discovered and showed signi cantly positively correlated with CVD in previous studies (7).
Besides, apolipoprotein(a) bound to apolipoprotein B of an LDL-like particle covalently composed Lp(a) (26) and it mediates atherogenicity by using its LDL moiety. Moreover, Lp(a) can induce the proin ammatory responses as one of its normal function (27). In previous studies, androgens had a greater affection of decreasing the level of Lp(a) compared with estrogens. (28).Besides, most of the women with CKD patients are old, and mostly are in menopause (14). Thus, there is a possibility of women with CKD have a worse situation of Lp(a) compared with men with CKD and we need to promote the realization of Lp (a) in women with CKD.
WBC is also one of the newly discovered cardiovascular risk factors, which participates various stages of cardiovascular disease progression and complication (29). Oxidative stress and in ammation were the important mechanisms for WBC to be considered as a risk factor for CVD in patients with CKD (30). In our study, the value of WBC increased in both women and men with CKD. Moreover, the difference between CKD and non-CKD women was signi cantly higher than men. It indicated that women may get more serious processes of in ammation from non-CKD to CKD compared with men.
VAI and LAP are important markers of visceral obesity (31) and we found that the level of VAI and LAP in women with CKD were higher than men with CKD in this study. Lots of studies showed that VAI and LAP are associated with a high risk of CVD (32). The INTERHEART study had showed that VAI contributes a lot to CVD risk, evaluating the impact of obesity on CVD (33). Also, woman was found as a risk factor of abdominal obesity in the previous study (34). The speci c sex-based mechanisms are still unclear (35).
Simultaneously, the LAP is considered as an better index compared with BMI to recognize CVD risk as an index which describes lipid over-accumulation based on WC and fasting triglycerides (36). In this study, compared with BMI, VAI and LAP were more sensitive to re ect the change of lipid pattern in CKD patients. Thus, paying more attention to these two factors instead of BMI only in CKD patients may help doctors to get awareness of CVD risk.
The reasons of the severe condition of the risk factors of CVD of women with CKD may be the following points. Frist, previous studies have shown that autoimmune diseases, pregnancy, dialysis and transplantation lead speci c challenges to women with CKD (37,38). Second, it is not equal among men and women to access the medical care of CKD in lots of places around the world (39). Third, the lost protection of estrogen in women is also a dangerous factor for the risk of CVD in women with CKD (40). Finally, the progress of the times has made women no longer constrained by family life, but also faces more pressure from work and society, especially in China (41).This stress may increase the in ammation of women. It is important to keep the health of families, communities and populations to advocate for improving the access to medical for women (42). Therefore, more attention needed to pay to the risk factors of CVD of women with CKD to aware of the unsatisfactory condition of women with CKD.
Our study had some limitations at the same time. Firstly, it was di cult to explain whether the CKD in women will experience worse changes in visceral adiposity, some lipid pro le and TyG during the process of CKD because of the cross-sectional study design. Second, since we only have the only one blood sample, the de nition of CKD may not exactly as it showed. Finally, the causality between the greater relative CVD risk and these great change in the CVD risk factors among women with CKD was hard to nd.
However, there were also some strengths of our study, such as these data used in this study were population-based. Also, covariates of interest in this research were assessed in detail. Some non-traditional risk factors such as in ammation markers and visceral adiposity were included and they were calculated as well.

Conclusions
In summary, our study documented that CKD women had a tendency of having worse CVD risk factors than men. Increased CVD risk factors were observed in this study in both genders with CKD. CKD women had higher value of CVD risk factors than men. Moreover, the greater changes in some CVD risk factors of women with CKD compared with man were independent of important factors, such as BMI, age, visceral adiposity or HOMA-IR. These important ndings may help to suggest the increased relative CVD risk in CKD women compared with men.