■Background High sensitivity methods for measurement of CRP (hs-CRP) are necessary for cardiovascular risk stratification, which is based upon discrimination.

Presentation on theme: "■Background High sensitivity methods for measurement of CRP (hs-CRP) are necessary for cardiovascular risk stratification, which is based upon discrimination."— Presentation transcript:

1 ■Background High sensitivity methods for measurement of CRP (hs-CRP) are necessary for cardiovascular risk stratification, which is based upon discrimination of CRP levels extending below 0.3 mg/dl. For the determination of cardiovascular risk, low, intermediate, and high risk values were defined as <0.1, 0.1 to 0.3, and > 0.3 mg/dl (Ridker RM, et al. Circulation 2003; 107: 363 and Pearson TA, et al. 2003; 107: 499). It has been considered that hs-CRP is associated with conventional cardiovascular risk factors, however little is still known about its significance in primary prevention of cardiovascular events in Japanese population. 今回の検討を行った背景として、次のようなことが挙げられます。 まず、CRPは近年、従来の炎症マーカーとしての役割だけではなく、心血管イベント発生のリスクを表す指標としても認められており、特に、通常のCRPでは正常範囲にあたる、0.1 以上0.3 未満でも、中等度のリスク群に該当すると考えられています。 また、LDLとHDLに関しては、周知のとおりです。

2 ■Objective ■Study Subjects
To assess the relation among hs-CRP ( within normal standard CRP levels ), serum lipid profiles and other cardiovascular risk factors measured in primary medical examination for subclinical Japanese population. ■Study Subjects One thousand and five hundred two ( 95.7 % ) of consecutive 1,570 people, who underwent primary medical examination in Shin-Urayasu Toranomon Clinic from April to August in 2007, had serum hs-CRP levels < 0.3 mg / dl.

3 ■Study Subjects (cont’d)
One thousand ninety nine people ( 70 % ) were left after exclusion of 403 people with smoking habit, apparently inflammatory disorders such as leukocytosis and any medication for hypertension, hyperlipidemia and diabetes mellitus. They consisted of 494 males ( 31.5 %; averaged age 49.5 ( ) years old ) and 605 females ( 38.5 %; averaged age 47.0 ( ) years old ).

4 ■Methods We analyzed data of hs-CRP, serum lipid profiles and other cardiovascular risk factors in each gender and age subgroup ( < 50 and > 50 year-old ) of the subjects to evaluate the following issues: (1) Correlation between hs-CRP and LDL, HDL, TG or LDL / HDL ratio. (2) Comparison of LDL, LDL / HDL ratio, HDL and TG between low risk ( hs-CRP < 0.1 mg / dl ) and intermediate risk ( hs-CRP > 0.1 mg / dl ) group in each subgroup of genders and ages. 検討は、健診の際に得られた患者データから、3項目の検査値を抽出し、男女別にデータの解析を行いました。 統計ソフトには、JMP 5.0.1を使用しました。 4

5 ■Methods (cont’d) (3) Multivariate analysis of correlation among hs-CRP and other cardiovascular risk factors. We used JMP software (SAS Institute, Cary, NC, USA) in statistical analysis and p value < 0.05 was considered as significant.

6 ■ Results Hs-CRP was significantly correlated with LDL, HDL, LDL / HDL and TG levels, and most strongly correlated to LDL / HDL ratio (Figures 1 - 4). Threshold levels of LDL / HDL between low and intermediate risks were 2.4 in men and 2.0 in women calculated from regression lines (Figures 1, 2). By ROC analysis, the thresholds of LDL / HDL were 2.5 in men (sensitivity 65.3 %, specificity 67.0 %) and 2.1 in women (sensitivity 59.6 %, specificity 66.2 %). LDL was significantly less in low risk group than intermediate risk group only in a subgroup of > 50 year-old men (Figures 5, 6). まとめです。 まず、高感度CRPとの相関関係については、LDL値よりもLDL/HDL比の方が比較的強い傾向にあることがわかりました。 次に、高感度CRPのリスクが異なる2つのグループ間での平均値を比較した検討では、LDLの平均値は50歳以上の男性区分以外では、有意な差が認められませんでした。 一方で、LDL/HDL比の平均値は、男女ともに各年齢区分において有意な差が認められました。

7 ■ Results (cont’d) LDL / HDL was significantly less in low risk group than intermediate risk group in all gender and age subgroups (Figures 7, 8). HDL was significantly greater in low risk group than intermediate risk group in all gender and age subgroups (Figure 9). TG was significantly smaller in low risk group than intermediate risk group only in younger-age (< 50 year-old) subgroups of men and women (Figure 10). In terms of cardiovascular risk factors, the order of significance of correlation with hs-CRP was BMI, LDL / HDL, gender and age (Table).

8 A) Men Spearman’s correlation coefficient　 ρ =0.3313（p ＜0.0001） Spearman’s correlation coefficient　 ρ =0.2097（p ＜0.0001） Y = X Figure 1. Correlation of hs-CRP with LDL level and LDL/HDL ratio in men.

9 B) Women Spearman’s correlation coefficient　 ρ =0.3730（p ＜0.0001） Spearman’s correlation coefficient　 ρ =0.2906（p ＜0.0001） Y = X Figure 2. Correlation of hs-CRP with LDL level and LDL/HDL ratio in women.

10 Figure 3. Correlation between hs-CRP and HDL level in men and women.
A) Men B) Women Spearman’s correlation coefficient　 ρ = －0.2995（p ＜0.0001） Spearman’s correlation coefficient　 ρ = －0.2876（p ＜0.0001） Figure 3. Correlation between hs-CRP and HDL level in men and women.

11 Figure 4. Correlation between hs-CRP and TG level in men and women.
A) Men B) Women Spearman’s correlation coefficient　 ρ = （p ＜0.0001） Spearman’s correlation coefficient　 ρ = （p ＜0.0001） Figure 4. Correlation between hs-CRP and TG level in men and women.

12 * * : p=0.0286 　以上の傾向をふまえ、さらに男女をそれぞれ、高感度CRPが0.1未満のグループと0.1以上の2グループに分け、このグループの間でLDLとLDL/ HDL比の平均値にどのような違い認められるかを検討しました。 　この際、LDLやHDLは年齢による変動を受けるため50歳で区切って検討いたしました。また、サンプルが等分散を示さなかったものについてはt検定のウェルチの方法を用いました。 ではまず、男性のLDLの平均値から比較していきます。 このグラフは、縦軸がLDL値を示し、横軸向かって左側が高感度CRPが0.1未満のグループ、右側が0.1以上のグループとなっています。 比較してみると、オレンジ色、つまり50歳以上の男性では高感度CRPのリスクが上がるとLDLの平均値も上昇し、また両グループの間に有意差が認められました。一方で、水色、つまり50歳未満の男性では、高感度CRPのリスクが上昇してもLDLの平均値の上昇は認められませんでした。 また、年齢区分間での有意差は認められませんでした。 Figure 5. Difference of LDL level between low and intermediate risk groups.

13 *** * : p <0.0001 女性についても同様に比較してみます。 どちらの年齢区分でも、高感度CRPのリスク上昇に伴うLDL平均値の上昇は認められたものの、有意差は認められませんでした。 一方、年齢区分間では、有意な差が認められました。 Figure 6. Difference of LDL level between low and intermediate risk groups.

14 ** * * : p＝0.004 **: p＝0.0009 次に、LDL/HDL比の平均値について、まずは男性から比較します。 50歳未満、50歳以上のどちらの年齢区分でも、高感度CRPのリスクが上昇するとLDL/HDL比の平均値は上昇し、両グループの間で有意差が認められました。 一方、年齢区分間の有意差は認められませんでした。 Figure 7. Difference of LDL / HDL between low and intermediate risk groups.

15 *** * ** * : p =0.0129 ** : p < 0.001 次に、女性の比較です。 女性についても、男性と同様に、どちらの年齢区分でも、高感度CRPのリスクが上昇するとLDL/HDL比の平均値は上昇し、両グループの間で有意差が認められました。 一方、年齢区分間の有意差は、CRP0.1未満のグループでは認められ、0.1以上のグループでは認められませんでした。 *** : p < Figure 8. Difference of LDL / HDL between low and intermediate risk groups.

16 * * * : p=0.0124 **:p＝0.001 * : p＜0.001 Figure 9. Difference of HDL level between low and intermediate risk groups.

17 *** *** *** : p＜0.0001 * : p =0.0172 Figure 10. Difference of TG level between low and intermediate risk groups.

18 Table. Multivariate analysis of cardiovascular risk factors related to hs-CRP.

19 ■Discussion Some guidelines already showed threshold levels of LDL, HDL and TG in primary prevention of cardiovascular events. However, significance of LDL / HDL and hs-CRP in the primary prevention has not been well evaluated. In the present study, the threshold levels of LDL / HDL ratio between cardiovascular risk classes according to serum hs-CRP levels in both genders could be demonstrated in subclinical Japanese population. Prospective observational study is needed to evaluate whether the threshold is valuable for primary prevention of cardiovascular disease in Japanese population.

20 ■Conclusions LDL / HDL ratio is more strongly correlated to hs-CRP level than LDL and HDL. According to risk stratification by hs-CRP, LDL / HDL ratio of > 2.5 in men and > 2.1 in women may show intermediate risk for future cardiovascular events in subclinical Japanese population with normal standard CRP levels.