In this study, we using fasting and non-fasting lipid profiles to assess 10-year risk of ASCVD in the same Chinese general patients through both China ASCVD risk estimator and Europe SCORE risk charts respectively. It is showed that China ASCVD risk estimator assessed half of the participants as low risk, while the European risk charts assessed half of the participants as moderate risk in the same participants. And the agreement between China ASCVD risk estimator and SCORE system was relatively poorly. Interestingly, there was substantial agreement on China ASCVD risk estimator between fasting and non-fasting lipid profiles, which indicated that non-fasting lipid profiles could also be considered to evaluate 10-year CVD risk via China ASCVD risk estimator in clinical practice.
Non-fasting changes in lipid profiles, especially LDL-C and TG, in Chinese participants in this study were relatively different from those in other studies with large population in Denmark, which showed that changes in lipid profiles were insignificant after a daily meal [6, 31–33]. The maximal mean changes between non-fasting versus fasting blood samples as measured in random are + 0.3 mmol/l for TG level and − 0.2 mmol/l for TC or LDL-C level [33]. However, TC and LDL-C levels decreased by 0.3 and 0.7 mmol/l, respectively, while TG level increased by 0.6 mmol/l at 4 h after a daily breakfast in this study. The potential causes for non-fasting reduction in LDL-C and increment in TG were complicated and controversial. We previously found that postprandial reduction in direct measured LDL-C level was more prominent than that in calculated LDL-C level by Friedewald formula at both 2 h and 4 h after a daily breakfast in Chinese subjects [34], and LDL-C level was detected by the direct method in the present study. Additionally, some scholars believed that mildly reduction in LDL-C level after a daily meal was due to fluid intake, but changes could be corrected by adjustment for albumin levels [6, 12]. Compared with the western breakfast, the traditional Chinese breakfast could have a higher carbohydrate content and fluid intake, such as porridge, noodles, vermicelli, soy milk and so on, but not breakfast rich in protein and/or fat and solid food such as cheese, sausage, ham and bacon [35, 36]. Unfortunately, changes in albumin levels were not included in this study. Moreover, the potential role of difference in commercial lipid test kits and race between different studies could also be considered.
China ASCVD risk estimator is primarily based on the results of the China-PAR (i.e. Prediction for ASCVD Risk in China) project which was the first study to develop and validate the 10-year ASCVD risk in four large, contemporary Chinese populations, which fasting lipid profiles but not non-fasting lipids were used in that project [25]. The equations directly included age, BP, TC level, LDL-C level, HDL-C level current smoking, and diabetes mellitus. It was demonstrated that the China-PAR project had excellent performance in ASCVD risk prediction with good internal consistency and external validation compared with western estimators [37, 38].
SCORE system, which estimates 10-year cumulative risk of first fatal atherosclerotic events, has been recommended for risk stratification in European countries, and the corresponding risk estimator has been produced as charts for low-and high-risk regions [27]. SCORE system is based on large, representative European cohort datasets, and includes several risk factors, such as age, gender, smoking, systolic blood pressure and total cholesterol level in risk charts. Agreement between fasting and non-fasting in risk stratification is high in both low-and high-risk charts, and it is TC level that used in the SCORE system.
Of course, we noticed that agreement between SCORE low-and high-risk charts was high, whereas relatively poorly with China ASCVD risk estimator as the former classified half of the participants as moderate risk, while China ASCVD risk estimator assessed half of the participants as low risk in the same participants. That is to say that SCORE system could be not appropriate for Chinese individuals to evaluate CVD risk. Substantial differences in prediction capability between China ASCVD risk estimator and SCORE risk charts may be due to ethnic heterogeneities, distinctive risk characteristics of CVD, as well as different treatment and control rates for risk factors (e.g. hyperlipidemia) [29, 39–42]. Since the two established risk assessment systems above were based on per se research results or databases of the respective populations, so it is more suitable for individuals to use specific and regional assessment tools when conducting cardiovascular risk assessments.
There were several limitations in this study. Firstly, the number of participants in this study was relatively small. Secondly, this is a cross-sectional study, and thus it was impossible to verify the occurrence of cardiovascular events. A prospective study with large sample size could be needed to further explore the application of non-fasting blood lipids in CVD risk assessments.