Comparison Among Inflammatory Markers About Their Ability To Identify Hypertension

Aims: Inflammation is closely related to the development of hypertension. Research has shown that many inflammatory factors associated with an increased risk of high blood pressure, such as high-sensitivity C-reactive protein, and so on. However, most previous studies were only focus on the association of one single inflammatory factor with hypertension. Very few study has compare different of inflammatory markers(Hs-CRP, WBC, NLR, CC3 and CC4) for their ability to identify hypertension.Methods: A cross-sectional(n=1,857) study was conducted on participants recruited from the Health Management Center of Tianjin Medical University General Hospital in Tianjin, China. Multiple logistics regression analysis was used to assess the relationships between several important inflammatory markers quintiles and hypertension. ROC curve analysis was used to obtain the area under ROC curve and optimal cutoff points of inflammatory markers for diagnosing hypertension.Multivariable linear regression models were calculated between inflammatory markers and blood pressure(systolic blood pressure and diastolic blood pressure).Results: The prevalence of hypertension were 33.28%. In cross-sectional analysis,after adjusted for potential confounders, for all genders, the odds ratios(95%confidence interval) of hypertension for increasing quintiles of Hs-CRP were 1.00(reference), 1.31(0.94, 1.82), 1.69(1.22, 2.33), 1.75(1.27, 2.41)(P for trend < 0.01),the odds ratios(95% confidence interval) of hypertension for increasing quintiles of CC3 were 1.00(reference), 1.12(0.80, 1.55), 1.50(1.08, 2.09), 1.56(1.13, 2.17)(P for trend <0.01). For men, the odds ratios(95% confidence interval) of hypertension for increasing quintiles of Hs-CRP were 1.00(reference), 1.33(0.91, 1.94), 1.68(1.15,2.45), 1.58(1.08, 2.30)(P for trend< 0.05). The odds ratios(95% confidence interval)of hypertension for increasing quintiles of CC3 were 1.00(reference), 1.26(0.87,1.85), 1.47(1.00, 2.17), 1.55(1.05, 2.28)(P for trend < 0.05). For women, the odds ratios(95% confidence interval) of hypertension for increasing quintiles of Hs-CRP were 1.00(reference), 1.23(0.63, 2.43), 1.70(0.90, 3.26), 2.25(1.19, 4.35)(P for trend< 0.05). In other situation, there are no correlation. The area under ROC curve(AUC) of Hs-CRP was 0.63 for all genders, which was 0.59 for men and 0.69 for women. The AUC of WBC was 0.56 for all genders, which was 0.54 for men and0.56 for women. The AUC of NLR was 0.53 for all genders, which was 0.50 for men and 0.59 for women. The AUC of CC3 was 0.62 for all genders, which was 0.59 for men and 0.66 for women. The AUC of CC4 was 0.56 for all genders, which was 0.54 for men and 0.61 for women. The optimal cutoff point(sensitivity; specificity) of Hs-CRP was 0.77 mg/l(0.67, 0.55) for all genders, which was 0.72 mg/l(0.00, 0.98)for men and 0.73 mg/l(0.71, 0.60) for women. The optimal cutoff point(sensitivity;specificity) of WBC was 5.55×10-9/L(0.55, 0.57) for all genders, which was7.12×10-9/L(0.60, 0.48) for men and 5.65×10-9/L(0.39, 0.73) for women. The optimal cutoff point(sensitivity; specificity) of NLR was 1.57(0.55, 0.57) for all genders, which was 2.08(0.73, 0.30) for men and 1.59(0.52, 0.65) for women. The optimal cutoff point(sensitivity; specificity) of CC3 was 98.5 mg/dl(0.68, 0.50)for all genders, which was 96.5 mg/dl(0.73, 0.30) for men and 99.8 mg/dl(0.68, 0.60)for women. The optimal cutoff point(sensitivity; specificity) of CC4 was 21.2 mg/dl(0.52, 0.58) for all genders, which was 24.35 mg/dl(0.29, 0.78) for men and 18.15mg/dl(0.81, 0.36) for women. In multiple linear regression analysis, the independent variables included in Hs-CRP, WBC, NLR, CC3, CC4, age, BMI, smoking, alcohol consumption, family history and so on, For all genders, the standard regression coefficient of CC3 with diastolic blood pressure and systolic blood pressure were0.055(P < 0.05) and 0.072(P < 0.01). For men, the standard regression coefficient of CC3 with diastolic blood pressure and systolic blood pressure were 0.049(P = 0.12)and 0.080(P < 0.05). For women, the standard regression coefficient of WBC with diastolic blood pressure and systolic blood pressure were 0.162(P < 0.0001) and0.084(P < 0.05). Which is the largest.Conclusions: In conclusion, CC3 and Hs-CRP are superior to other inflammatory markers(WBC, NLR and CC4) as an indicator to identify hypertension. Considering the gender difference, CC3 has a stronger association with blood pressure for men.While for women, WBC has a stronger association.