The Specific Associations Between Inflammatory Markers And Risk Of Abnormal Glucose Metabolism

Objective 1.To compare baseline characteristics of the Tianjin Chronic Low-grade Systemic Inflammation and Health cohort(TCLSIHealth)and the Swedish Malm? Diet and Cancer cohort(MDC)as well as its cardiovascular sub-cohort(Malm? Diet and Cancer cardiovascular cohort,MDC-CV),and to explore whether factors associated with abnormal glucose metabolism(inflammatory factors,etc.)differ among ethnic groups.2.Based on the TCLSIHealth cohort and the MDC(-CV)cohort,to comprehensively explore the predictive value of multiple inflammatory markers for abnormal glucose metabolism,thus,to identify inflammatory factors that can independently predict abnormal glucose metabolism.3.For inflammatory factors that have predictive ability for abnormal glucose metabolism,to further compare whether their associations with hyperglycemia risk differ from their associations with hypertension,hyperlipidemia,nonalcoholic fatty liver(NAFLD),and cardiovascular disease(CVD),thus to identify inflammatory factors that are specific for abnormal glucose metabolism and to help distinguish risk for abnormal glucose metabolism from other common metabolic disorders among high-risk individuals,which is a prerequisite for inflammation-targeted hyperglycemia prevention and treatment.Methods 1.Analysis of covariance was used for continuous variables and multiple logistic regression analysis was used for categorical variables,to compare differences in baseline characteristics among individuals in the TCLSIHealth cohort and those in the MDC(-CV)cohort.Stepwise logistic regression analysis was used to identify factors associated with abnormal glucose metabolism.2.Cox regression with time-on-study as time-scale was used to estimate hazard ratios(HRs)and 95% confidential intervals(CIs)for incident hyperglycemia.Multiple potential confounders were taken into consideration,including baseline glucose,etc.Restricted cubic spline functions were incorporated into the Cox model to detect potential non-linearity.In sensitivity analyses,inflammatory factors without ignificant collinearity by the variance inflation factor(VIF)tests were adjusted for each other in the multivariate model.3.For inflammatory markers that showed predictive ability for abnormal glucose metabolism,their associations with incident hyperglycemia versus incident hypertension,hyperlipidemia,NAFLD,and CVD,respectively,were compared using the Lunn-Mc Neil competing risks approach.In sensitivity analyses,those who developed both hyperglycemia and hypertension or hyperlipidemia or NAFLD or CVD during the follow-up period were additionally excluded.Results 1.Analyses of baseline characteristics 1.1.Comparisons between the TCLSIHealth cohort and the MDC(-CV)cohort The prevalence of diabetes in the TCLSIHealth cohort based on the American Diabetes Association criteria was 10.3%,which was higher than that based on a history of diabetes and/or fasting blood glucose in the MDC-CV cohort(8.09%)(P<0.0001).After adjusting for age and sex,total leukocyte and its subtypes,complement 3(C3),C-reactive protein(CRP),and other baseline characteristics significantly differed between the two groups(P<0.0001).1.2.Factors associated with abnormal glucose metabolism Inflammatory factors and common metabolic risk factors,including age,sex,waist circumference,hypertension,etc.were associated with abnormal glucose metabolism,with no obvious ethnic differences.2.Prospective analyses of inflammatory factors and risk of abnormal glucose metabolism 2.1.Results from the TCLSIHealth cohort After multiple adjustments,total leukocytes,neutrophils,lymphocytes,C3,and CRP were linearly and significantly associated with prediabetes risk.All P values for effect were <0.01 and P values for nonlinear were >0.05.However,the neutrophil to lymphocyte ratio(NLR),C4,and fibrinogen were not significantly associated with prediabetes.Results were generally consistent when diabetes was used as the outcome instead of prediabetes.2.2.Results from the MDC cohort Total leukocytes,neutrophils,lymphocytes were linearly and significantly associated the risk of diabetes.All P values for effect were <0.0001 and P values for onlinear were >0.05.The association between NLR and diabetes was not statistically significant.2.3.Results from the MDC-CV cohort C3,CRP,and growth differentiation factor 15(GDF15)were linearly and significantly associated risk of diabetes.P values for effect were <0.01,0.03,and 0.04,respectively,while all P values for nonlinearity were >0.05.Ceruloplasmin,alpha1-antitrypsin,orosomucoid,and haptoglobin were not associated with diabetes.2.4.Inflammatory factors adjusted for each other Total leukocytes and NLR were separated from neutrophils and lymphocytes,as suggested by the VIF assessment.They were then put together with the other inflammatory markers in a multivariate Cox model,so as to be adjusted for each other.Based on the TCLSIHealth cohort,using the combination of prediabetes and diabetes as the outcome event,total leukocytes/neutrophils and lymphocytes,C3 remained significantly associated with hyperglycemia.The hazard ratios and 95% confidence intervals(CIs)per 1 standard deviation change of total leukocytes/neutrophils and lymphocytes,C3 were 1.12(1.07,1.18)/1.08(1.03,1.13)and 1.09(1.04,1.14),1.15(1.09,1.21),respectively.However,the association of CRP with hyperglycemia became insignificant.The associations of total leukocyte and its subtypes with diabetes in the MDC cohort were consistent with those obtained from the TCLSIHealth cohort.Data from the MDC-CV cohort showed that the associations of C3 and GDF15 with diabetes remained significant,with the corresponding HRs(95% CIs,P values)were 1.14(95% 1.04-1.25,P<0.01)and 1.12(95% CI 1.01-1.25,P=0.03).However,CRP was no longer associated with the outcome after adjusting for other inflammatory markers.3.Comparison of risk of abnormal glucose metabolism and other diseases in relation to inflammatory markers 3.1.Results from the TCLSIHealth cohort Both total leukocyte and lymphocyte counts had stronger associations with risk of hyperglycemia compared with hypertension(P<0.02 and P=0.02,respectively).Lymphocyte count had stronger association with hyperglycemia as compared with hyperlipidemia(P=0.03).The predictive value of C3 for hyperglycemia was weaker than for NAFLD(P<0.01).Consistent results were observed after excluding those who developed both hyperglycemia and hypertension or hyperlipidemia or NAFLD during he follow-up period.Using diabetes instead of the combination of prediabetes and diabetes as the outcome event generated relatively less significant results.3.2.Results from the MDC cohort Lymphocyte count had stronger association with diabetes compared with CVD(P=0.02).Additional analysis yielded consistent results(P<0.01).3.3.Results from the MDC-CV cohort Both primary and additional analyses showed that the predictive ability of C3 for hyperglycemia was considerately stronger than for CVD(P<0.01).Conclusion 1.Differences were observed in the prevalence of diabetes and baseline characteristics between the Chinese urban population and the Swedish urban population.Factors associated with abnormal glucose metabolism were similar in the two ethnic groups.2.Total leukocytes(mainly neutrophils and lymphocytes),C3,GDF15,and CRP were associated with the risk of hyperglycemia.Except for CRP,the other inflammatory factors remained significant after mutual adjustments,indicating that different inflammatory cytokines capture distinct aspects of inflammation,and thus leading to distinguished associations with abnormal glucose metabolism.3.Metabolic diseases had shared inflammatory profiles.It is suggested that there is an inflammatory “common soil” for metabolic disorders,but there are also important differences.While being a less efficient predictor for hyperglycemia as compared with NAFLD,C3 is a stronger predictor for hyperglycemia as compared with CVD.Meanwhile,lymphocyte count is a stronger predictor for hyperglycemia as compared with ether hypertension,hyperlipidemia,or CVD.These findings may help discriminate between individuals at elevated risk of hyperglycemia and those at elevated risk of other metabolic disorders,which is a prerequisite for targeted therapies.Among various inflammatory markers,the special roles of lymphocytes and C3 in the development and progression of hyperglycemia is worthy of further investigation.4.Despite of the significant differences in the baseline characteristics,substantial similarities exist in the associations of inflammatory markers with abnormal glucose metabolism between the two cohort studies.It is suggested that specifying an inflammation-targeted prevention and control strategy for chronic diseases may have global benefits.