br Methods br Results A total of
Results A total of 2136 cases, including 614 controls, and 707 LAA, 375 SVO, and 440 HICH cases, were enrolled in this study. Clinical parameters between the control individuals and patients with different stroke subtypes are listed in Table 1. The average age and serum HDL-C level were significantly lower in all the stroke subtypes than those of the control group (p<0.001). Female gender, hypertension, smoking, alcohol consumption, lower cholesterol level, and no family history of CAD were high risk factors of stroke development. LAA (p<0.001) and SVO (p<0.001) frequencies, but not HICH frequency (p=0.956), were significantly higher in patients with diabetes than in the control individuals. Low serum LDL-C and no family history of stroke were associated with HICH, but not with LAA and SVO. Serum triglyceride level was associated with SVO, but not with LAA or HICH. Table 2 presents the rs5070 genotypes in the control individuals and patients with different stroke subtypes. The frequency of ApoA1 rs5070 polymorphism in all the stroke subtypes was not significantly different from that in the control group. After age, gender, hypertension, diabetes, triglyceride level, cholesterol level, smoking, alcohol consumption, family history of stroke, family history of CAD, and LDL-C and HDL-C levels were adjusted, the difference of frequency in the polymorphisms between the two groups was not significant. The association of rs5070 pacap genotype (AA/AG+GG) with the clinical parameters is summarized in Table 3. Patients with AG+GG genotypes showed a more significant family history of CAD than those with the AA genotype. Age, gender, hypertension, diabetes, serum cholesterol, smoking, alcohol, triglyceride, family history of stroke, LDL-C, and HDL-C were not associated with ApoA1 rs5070 polymorphism. The HDL-C levels were 49.9±13.2, 41.6±12.7, 44.8±12.6, and 44.2±16.3 in the control individuals with the GG genotype, LAA, SVO, and HICH, respectively. Compared with that of the patients with the AA allele, the OR of the diabetic patients with the AG+GG allele was 1.58 (95% CI=1.00–2.42; p=0.046; Table 4), associated with the development of LAA but not with the development of SVO (OR=1.21; 95% CI=0.71–2.07; p=0.428) or HICH (OR=0.62; 95% CI=0.91–2.87; p=0.100). Furthermore, the OR of male diabetic patients with the AG+GG allele associated with the development of LAA was increased to 1.90 (95% CI=1.04–3.46; p=0.035). The OR of female diabetic patients with the AG+GG allele associated with the development of LAA was not significant (OR=1.23; 95% CI=0.59–2.59; p=0.577; Table 5).
Discussion Stroke is the final event in a complex interplay between the environment and genes. In this study, several conventional risk factors, such as age, gender, hypertension, diabetes, and cholesterol levels, were associated with stroke. Epidemiological studies have shown that the prevalence of stroke is lower in premenopausal females than in age-matched males. In this study, the incidence of different stroke subtypes was significantly lower in women than in men. Hypertension is a well-known critical risk factor of stroke development. In Japan, the hazard ratios of stroke in normal (<120/80 mmHg), normal-to-high [(120–139) mmHg/(80–89) mmHg], hypertension stage I [(140–159) mmHg/(90–99) mmHg], and hypertension stage II (>160/100 mmHg) are 2.12, 2.43, 2.62, and 4.38 in men and 1.05, 1.29, 1.21, and 2.02 in women, respectively. The Asia Pacific Cohort Studies Collaboration also indicated that an increase in systolic and/or diastolic blood pressure can increase the hazard ratio of stroke development. Diabetes is also an independent risk factor of stroke development. The risk of stroke is higher in diabetic patients than in normal individuals. However, intensive glucose control in diabetic patients is not associated with a decreased incidence of CAD. Hypercholesterolemia is associated with an increased risk of stroke. Ebrahim et al demonstrated that a high serum cholesterol level is associated with ischemic stroke, and a low cholesterol level is correlated with hemorrhagic stroke. By contrast, our results indicated that low cholesterol levels were associated with both ischemic and hemorrhagic strokes. We could not exclude the possibility that low lipid levels in stroke patients were attributed to lipid-lowering medications or other treatments because drug information was not available in our study. Further research should be conducted to elucidate the relationship between low cholesterol levels and ischemic stroke. In Korea and Japan, smoking of a large number of cigarettes per day increases the risk of ischemic stroke and subarachnoid hemorrhage, but not that of intracerebral hemorrhage. An increased HDL-C level reduces not only the risk of CAD and stroke, but also the recurrence of stroke. In our study, the HDL-C level of the stroke patients was significantly lower than that of the control group. Therefore, conventional risk factors, such as gender, age, hypertension, diabetes, serum cholesterol level, smoking, and low HDL-C, were associated with different stroke subtypes.