Temporal Relationships Between Blood Lipids And HOMA-IR In Perimenopausal Women

Backgrounds:In recent decades,the incidence and prevalence of dyslipidemia and diabetes have been increasing rapidly.As the main risk factors for the morbidity and mortality of cardiovascular diseases,the prevention and control of them has become nonnegligible public health challenge.In reality,dyslipidemia and diabetes always occur concomitantly and the causal relationships between blood lipids and insulin resistance are still controversial.As women get older,the protection of estrogen fades away,and womens risk of developing diabetes and dyslipidemia increases significantly.Perimenopause is a transition period before menopause,and it is also a window period for disease prevention.It is valuable to explore the temporal relationships between blood lipids and insulin resistance.And it has great public health significance to prevent post-menopausal dyslipidemia,diabetes and its complications,and improve the quality of life of middle-aged and elderly women.Methods:The Study of Womens Health Across the Nation(SWAN)is an ongoing,multicenter,and multi-ethnic prospective cohort Study.The baseline and the first 10 follow-ups in the cohort have been available to the public.It included a large amount of information about the health of women before,during and after menopause,providing a good database for this study.The cross-lagged path analysis model is a method based on the cross-lagged panel design,using two or more panels to study to dissecting a causal relationship between intercorrelated variables.Its path coefficient indicates a clear temporal sequence,which satisfies the temporality that cause precedes outcome in causal inference and has the ability to prompt the direction of causality.It is an effective method to explore the temporal relationships between blood lipids and insulin resistance in perimenopausal women.Our study is based on the Study of Women's Health Across the Nation(SWAN).First,we used the perimenopausal womens baseline and last follow-up data to build a 2 waves cross-lagged path analysis model and explore the temporal relationships between blood lipids,including total cholesterol(TC),triglyceride(TG),high density lipoprotein cholesterol(HDL-C),low density lipoprotein cholesterol(LDL-C),lipoprotein A-I(LpA-I),apolipoprotein A-I(ApoA-I)and apolipoprotein B(ApoB),and homeostasis model assessment-IR(HOMA-IR),fasting plasma glucose(FPG),or insulin.Then we built appropriate latent variables by the structural equation model and explored the temporal relationships between lipid factors and HOMA-IR.Finally,a 3 waves latent variable cross-lagged path analysis model was built to explore the influence of follow-up interval on the temporal relationships between lipid factors and HOMA-IR for the sensitive analysis.The aim of this study was to find the temporal relationships between blood lipids and insulin resistance in perimenopausal women,and provide a scientific basis for the prevention and control of dyslipidemia and diabetes in women.RESULTS:1.In the two waves cross-lagged path analysis models,1386 perimenopausal women were included,with an average baseline age of 46.35±2.63years and an average follow-up time of 3.53±1.83 years.In the three waves cross-lagged path analysis models,722 perimenopausal women were included,with an average baseline age of 45.71±2.37 years.Their average follow-up time at the second follow-up(T2)was 2.51± 1.34 years,and at the average follow-up time at the third follow-up(T3)was 4.76± 1.40 years.2.The results of the two waves cross-lagged path analysis model are as follows:(1)Temporal relationships between blood lipids and HOMA-IR:? There is no temporal relationship between TC and HOMA-IR,TC is independent of HOMA-IR.?There is a two-way temporal relationship between TG and HOMA-IR:The increase of TG or HOMA-IR can cause an increase of each other.? There is a two-way temporal relationship between HDL-C/LpA-I/ApoA-I and HOMA-IR:The increase of HDL-C/LpA-I/ApoA-I or HOMA-IR can cause a decrease of each other.? There is no temporal relationship between LDL-C and HOMA-IR,LDL-C is independent of HOMA-IR.?There is a unidirectional temporal relationship between ApoB and HOMA-IR:Increased ApoB might predict the increase of HOMA-IR.(2)Temporal relationships between blood lipids and FPG or Insulin:?There is no temporal relationship between TC and FPG,TC is independent of FPG.There is a unidirectional temporal relationship between TC and Insulin:Increased Insulin might predict the decrease of TC.? There is a two-way temporal relationship between TG and FPG/Insulin:The increase of TG or FPG/Insulin can cause an increase of each other.?There is a unidirectional temporal relationship between HDL-C and FPG:Increased FPG might predict the decrease of HDL-C.There is no temporal relationship between LpA-I/ApoA-I and FPG,LpA-I/ApoA-I is independent of FPG.? There is a two-way temporal relationship between HDL-C/LpA-I/ApoA-I and Insulin:The increase of HDL-C/LpA-I/ApoA-I or Insulin can cause a decrease of each other.?There is no temporal relationship between LDL-C/ApoB and FPG/Insulin,LDL-C/ApoB is independent of FPG/Insulin.3.In this study,the two waves latent variable cross-lagged path analysis model was built to explore the temporal relationships between lipid factors and HOMA-IR.The lipid factors are the cholesterol transport factor(TC,LDL-C,ApoB),the reverse cholesterol transport factor(HDL-C,LpA-I,ApoA-I),and TG.The results are as follows:? There is no temporal relationship between the cholesterol transport factor and HOMA-IR.? There is a two-way temporal relationship between the reverse cholesterol transport factor and HOMA-IR.The increase of the reverse cholesterol transport factor or HOMA-IR can cause a decrease of each other.4.The results of the three waves latent variable cross-lagged path analysis model are as follows:? There is no temporal relationship between the cholesterol transport factor and HOMA-IR fromT1 to T2 and T2 to T3.? There is a two-way temporal relationship between the reverse cholesterol transport factor and HOMA-IR from T1 to T2 and T2 to T3.? There is a two-way temporal relationship between TG and HOMA-IR from T1 to T2,but there is a unidirectional temporal relationship between TG and HOMA-IR from T2 toT3,TG?HOMA-IR.CONCLUSION:1.There is no temporal relationship between cholesterol transport-related lipids and HOMA-IR.2.There is a two-way temporal relationship between reverse cholesterol transport-related lipids and HOMA-IR:The increase of reverse cholesterol transport-related lipids or HOMA-IR can cause a decrease of each other.3.There is a two-way temporal relationship between TG and HOMA-IR:The increase of TG or HOMA-IR can cause an increase of each other.4.The relationships between insulin resistance with the process of reverse cholesterol transport and TG are stronger than the process of cholesterol transport.