Background Some researches demonstrate that high-sensitivity C-reactive protein may be a risk factor to cause carotid atherosclerosis in patients with cerebral infarction. Inflammatory reaction may participate in formation of carotid atherosclerosis in patients with acute cerebral infarction. Objective To investigate the correlation between levels of serum high-sensitivity C-reactive protein and carotid atherosclerosis in patients with acute cerebral infarction accompanied with carotid atherosclerosis. Design Contrast observation between two groups. Setting Department of Neurology, Zhenzhou Hospital, Shenyang Medical College. Participants A total of 102 patients with acute cerebral infarction regarded as cerebral infarction group were selected from Department of Neurology, Shenzhou Hospital Affiliated to Shenyang Medical College from February 2005 to September 2006. There were 55 males and 47 females and their ages ranged from 55 to 86 years. All patients met the variously diagnostic points of cerebral infarction established by the Fourth National Cerebrovascular Disease Academic Meeting and were finally diagnosed with CT or MRI examination. Illness course was in an acute phase. A total of 96 healthy subjects were regarded as control group, including 51 males and 45 females aged from 48 to 78 years. All accepted subjects provided the confirmed consent. Methods ▪ Patients in the cerebral infarction group received carotid ultrasound Doppler examination and serum high-sensitivity C-reactive protein detection within 72 hours after onset. IMMAGE immune biochemical system and latex reinforcement particle-enhanced nephelometric immunoassay (PENIA) were used for quantitative detection of serum high-sensitivity C-reactive protein. ▪ Healthy subjects in the control group received the same detection. SEQUOIA512 color Doppler ultrasound (Siemens Company, USA) was used to detect carotid artery of all subjects so as to observe intima media thickness of artery and formation of artery atherosclerostic plaques. If artery atherosclerostic plaques were formed, their properties and amounts were determined based on the characteristics of light-echo signals. Evaluating criteria: Intima media thickness of artery was the vertical dimension from crossed face between lumen and tunica intima to crossed face between tunica media and tunica adventitia. Intima media thickness ≤ 0.9 mm was regarded as normal; 0.9 mm < intima media thickness ≤ 1.2 mm was regarded as thickening; when local eminence thickening was processed towards to lumen, the intima media thickness was more than 1.2 mm and plaque of tunica intima was formed at the same time. Properties of plaque were classified into 4 types: steady low-echo lipid malacoplakia, equal-echo fiber plaque, strong-echo or sound-imaging calcification hard plaque and unsteady-echo ulcer mixed plaque. Fiber plaque and calcification hard plaque were steady but malacoplakia and mixed plaque were unsteady. Main outcome measures Thickness of tunica media, characteristics of plaque and level of serum high-sensitivity C-reactive protein in carotid artery in two groups. Results All 102 patients with cerebral infarction and 96 healthy subjects were involved in the final analysis. ▪ Comparisons of level of high-sensitivity C-reactive protein: Level of high-sensitivity C-reactive protein in normal tunica media was higher in the cerebral infarction group [(4.66±1.55) mg/L] than the control group [(3.49±1.24) mg/L, t =2.541, P < 0.05]. In addition, level of high-sensitivity C-reactive protein in patients with thickening tunica media and plaque was not significantly different between the cerebral infarction group and the control group ( P > 0.05). ▪ Correlation between various degrees of vascular lesion and level of high-sensitivity C-reactive protein in the cerebral infarction group: Level of high-sensitivity C-reactive protein was statistically significantly higher in patients with thickening tunica media [(8.16±2.42) mg/L] than patients with normal tunica media [(4.66±1.55) mg/L, t =4.132, P < 0.01]. In addition, level of high-sensitivity C-reactive protein was statistically significantly higher in patients with carotid plaque [(12.08±3.85) mg/L] than patients with normal tunica media ( t =5.994, P < 0.01) and thickening tunica media ( t =4.197, P < 0.01). ▪ Levels of high-sensitivity C-reactive protein in patients with various kinds of carotid plaque: Level of high-sensitivity C-reactive protein was statistically significantly higher in patients with unsteady carotid plaque [(13.54±2.62) mg/L] than patients with steady carotid plaque [(8.61±3.71) mg/L, t =2.002, P < 0.05]. That was to say level of serum high-sensitivity C-reactive protein in patients who suffered acute cerebral infarction combined with carotid atherosclerosis especially carotid plaque was higher than that in those patients who did not have carotid lesions. This suggested that serum high-sensitivity C-reactive protein had a certain correlation with onset of carotid atherosclerosis in patients with acute cerebral infarction. Conclusion Serum high-sensitivity C-reactive protein certainly correlates with onset of carotid atherosclerosis in patients with acute cerebral infarction, while inflammatory reaction may participate in formation of carotid atherosclerosis in patients with acute cerebral infarction.