Table 2.

Relation between hs-CRP and cardiovascular events and mortality in tertiles of hs-CRP

Tertiles of log(hs-CRP)	Tertile 1	Tertile 2	Tertile 3
n = 1679	n = 558	n = 561	n = 560
Range hs-CRP, mgl/L	0.13–1.35	1.35–3.22	3.22–9.97
Myocardial infarction
n	41	53	62
Model I	1.00 (ref)	1.22 (0.81–1.83)	1.46 (0.98–2.18)
Model II	1.00 (ref)	0.91 (0.59–1.39)	1.07 (0.70–1.63)
Model III	1.00 (ref)	0.87 (0.57–1.32)	1.23 (0.81–1.85)
Stroke
n	24	33	28
Model I	1.00 (ref)	1.28 (0.76–1.17)	1.12 (0.76–2.17)
Model II	1.00 (ref)	0.96 (0.55–1.68)	0.81 (0.45–1.47)
Model III	1.00 (ref)	0.71 (0.40–1.22)	0.67 (0.37–1.20)
Cardiovascular events
n	80	102	125
Model I	1.00 (ref)	1.19 (0.88–1.59)	1.52 (1.14–2.02)
Model II	1.00 (ref)	0.93 (0.68–1.27)	1.16 (0.85–1.57)
Model III	1.00 (ref)	0.88 (0.65–1.21)	1.24 (0.91–1.68)
Vascular mortality
n	38	58	88
Model I	1.00 (ref)	1.47 (0.98–2.22)	2.34 (1.59–3.43)*
Model II	1.00 (ref)	1.26 (0.81–1.95)	1.87 (1.23–2.84)*
Model III	1.00 (ref)	1.37 (0.89–2.10)	1.87 (1.22–2.85)*
All-cause mortality
n	74	113	156
Model I	1.00 (ref)	1.47 (1.09–1.97)*	2.11 (1.59–2.79)*
Model II	1.00 (ref)	1.26 (0.92–1.72)	1.72 (1.27–2.33)*
Model III	1.00 (ref)	1.34 (0.98–1.84)	1.89 (1.39–2.57)*

hsCRP per tertiles, T1 = ref, hazard ratio (95% confidence interval); *statistically significant p < 0.05

Model I: age + sex

Model II: Model I + BMI, smoking, alcohol, non-HDL-c, micro-albuminuria

Model III: Model II + lipid-lowering therapy, aspirin