Lipid Levels Are Regionally Associated with Cerebral Microbleeds in Patients with Intracerebral Hemorrhage Chunyan Lei, MD,* Sen Lin, MD, PhD,† Bo Wu, MD, PhD,* Hao Li, MD, PhD,† Ming Liu, MD, PhD,* and Chao You, MD, PhD†
Background: Lipid levels may be involved in intracerebral hemorrhage (ICH) etiology, which suggested that lipid levels may be associated with the etiology of microbleeds (MBs) in patients with ICH. To explore this, we examined the association between lipid levels and MBs in different cerebral regions in patients with ICH. Methods: Patients admitted to our hospital with ICH were consecutively and prospectively included. Demographic and clinical information were collected and analyzed according to the occurrence and location of MBs and levels of triglycerides (TGs). Results: Of the 77 patients included in our study, 63 (81.8%) were found to have MBs. Prevalence of MBs in the ‘‘deep or infratentorial’’ region and any region increased with increasing tercile in TG concentration; however, no such trend was observed for strictly lobar MBs. The odds ratio (OR) for occurrence of MBs in deep or infratentorial region was even higher for the third tercile relative to the first: 6.77 (95% confidence interval [CI] 1.31-34.96). The OR for occurrence of MB in any region was even higher for the third tercile relative to the first: 12.24 (95% CI 1.40-106.83). However, the OR for occurrence of deep or infratentorial region and any region in the second tercile relative to the first tercile did not reach significance. Moreover, TG levels did not appear to be associated with the occurrence of strictly lobar MBs. Conclusions: High TGs were associated with deep or infratentorial and any MBs but not with lobar MBs. This finding may shed light on the role of lipids in MB and ICH etiology. Key Words: Lipid levels—triglyceride—microbleeds— intracerebral hemorrhage. Ó 2014 by National Stroke Association
Introduction From the *Department of Neurology, West China Hospital, Sichuan University, Sichuan Province; and †Department of Neurosurgery, West China Hospital, Sichuan University, Sichuan Province, People’s Republic of China. Received September 28, 2013; revision received October 9, 2013; accepted October 16, 2013. C.L. and S.L. contributed equally to this study. Grant support: This research was supported by the Science and Technology Support Program of the Department of Science and Technology of Sichuan Province (2012FZ0006) and the National Key Technology R&D Program for the 12th Five-Year Plan of People’s Republic of China (2011BAI08B05). Conflict of interest: None. Address correspondence to Bo Wu, MD, PhD, Department of Neurology, West China Hospital, Sichuan University, No. 37 Guo Xue Xiang, Chengdu 610041, Sichuan Province, People’s Republic of China. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2014 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2013.10.016
Intracerebral hemorrhage (ICH) in a lobar location is associated with cerebral amyloid angiopathy, whereas ICH in the basal ganglia, cerebellum, or pons is attributed mainly to hypertension or atherosclerotic microangiopathy.1 Different factors contribute to ICH occurrence in different locations of the brain, suggesting different etiologies. Recent work suggests that lipid levels may be involved in ICH etiology,2 raising the question of whether they are involved in ICH at all locations or only at certain ones. Microbleeds (MBs) tend to occur in advance of ICH, suggesting that they are an early stage of the same disease.3 MBs are pathologic deposits of hemosiderin caused by minor blood extravasation from lipohyalinized small arterioles. If lipid levels are, indeed, involved in the etiology of ICH, they may also be involved in MBs. Therefore,
Journal of Stroke and Cerebrovascular Diseases, Vol. 23, No. 5 (May-June), 2014: pp 1195-1198
1195
C. LEI ET AL.
1196
analyzing whether lipid levels are associated with MBs, in particular brain locations, may reveal whether the same is true for ICH. In this prospective study, we investigated the prevalence and location of MBs in a population of patients with ICH. We examined whether lipid levels correlated with MB occurrence and location.
Subjects and Methods Patients and Evaluation Patients with spontaneous ICH were prospectively and consecutively enrolled after being admitted to West China Hospital, Sichuan University, Chengdu, China, between May 1, 2012, and March 31, 2013. Spontaneous ICH was defined as intraparenchymal hemorrhage or intraventricular hemorrhage not caused by trauma, brain tumor, or hemorrhagic transformation of arterial ischemic stroke or established pathogenesis, that is, arteriovenous malformation, head trauma, or cavernous angioma. Patients were also excluded if they did not undergo susceptibility-weighted imaging (SWI) or if lipoprotein analysis data were unavailable.
Baseline Data Collection and Categorization Baseline information was collected at admission on age, gender, and stroke risk factors (hypertension, diabetes mellitus, current smoking, alcohol intake). Cerebral MBs were defined as homogeneous, round focal areas observed throughout the brain, with a diameter less than 10 mm and very low signal intensity on SWI. Cerebral locations of MBs were categorized as either strictly lobar or as ‘‘deep or infratentorial.’’ Patients with 1 or more MBs restricted to a lobar location were defined as having strictly lobar MBs. Persons without lobar MBs and with MBs in a deep or infratentorial location were defined as having deep or infratentorial MBs. The presence and location of MBs on SWI was determined independently by 2 neurologists (inter-rater reliability k 5 .82) blinded to clinical data. In case of disagreement, a third neurologist was consulted, and a consensus decision was reached.
Statistical Analysis The c2 or Fisher exact tests were used to assess the significance of differences in categorical variables. The Mann–Whitney U test was used to assess the significance of differences in continuous variables. When appropriate, results were reported as an odds ratio (OR) and 95% confidence interval (CI). Two-sided values of P less than .05 were considered statistically significant. All statistical analyses were performed using SPSS version 16 (IBM, USA). Logistic regression modeling was used to identify variables associated with MB occurrence and location.
Results A total of 105 consecutive patients were assessed for inclusion in the study, but 21 (20.0%) were excluded because of established pathologies and 6 (5.7%) because of missing data on lipid levels. Of 77 patients included in the final analysis, 54 (70.1%) were men, and mean age at stroke onset was 61.06 6 12.96 years. Fifty-five patients had supratentorial ICH, 22 patients had infratentorial ICH, 9 patients had thalamic ICH, 33 patients had putaminal ICH, 2 patients had external capsule, and 11 patients had lobar ICH. A total of 63 patients (81.8%) had MBs. Baseline characteristics of patients with ICH are shown in Table 1 according to whether they had MBs. Patients with MBs were more likely to be older and to have higher triglyceride (TG) levels. Then, we examined that which risk factor influenced the MB lesion development in the ICH by univariate logistic regression analysis. Logistic regression identified increasing tercile in TG concentration as an independent marker of MBs lesions (P 5 .013, OR 5 2.86, 95% CI 1.25-6.58). The data were analyzed to determine whether TG levels were associated with particular demographic or clinical variables or with the occurrence and location of cerebral MBs. Triglyceride levels were not associated with demographic variables or risk factors or with the occurrence of strictly lobar MBs. There was a significant trend toward more occurrence of MBs in deep or infratentorial region (P 5 .04) and any region (P 5 .03) with increasing TG tercile (Table 2); however, no such trend was observed for strictly lobar MBs. Post hoc analysis showed occurrence of MBs in deep or infratentorial region rate to be significantly higher in the third tercile (37.0%) than in the first tercile (P 5 .014). In contrast, occurrence of MBs in deep or infratentorial region rate in the second tercile (28.0%) was not significantly different from those in the first (8.0%; P 5 .069). The OR for occurrence of MBs in deep or infratentorial region was even higher for the third tercile relative to the first: 6.77 (95% CI 1.31-34.96). The OR for occurrence of deep or infratentorial region in the second tercile relative to the first tercile was 4.47 (95% CI .83-24.19). Moreover, the rate of MBs in any region was significantly higher in the third tercile (96.3%) than in the first tercile (68.0%, P 5 .008). Rate of MBs in any region was not significant between the second tercile (76.0%) and the first (68.0%, P 5 .29). The OR for MBs in any region in the second tercile relative to the first tercile was 1.78 (95% CI .496.53). The OR was significantly higher for the third tercile relative to the first: 12.24 (95% CI 1.40-106.83).
Discussion In this prospective study of patients with ICH, we found that 82% had MBs, consistent with the idea that MBs are strongly associated with ICH and may represent an early stage of the disease. In addition, we
LIPID LEVELS AND CEREBRAL MICROBLEEDS IN ICH
1197
Table 1. Baseline characteristics of ICH patients with and without MBs
Male, n (%) Age, mean 6 SD Hypertension, n (%) Diabetes mellitus, n (%) Current smoking, n (%) Alcohol intake, n (%) Systolic BP (mm Hg), mean 6 SD Diastolic BP (mm Hg), mean 6 SD Glucose (mmol/L), mean 6 SD Total cholesterol (mmol/L), mean 6 SD TG (mmol/L), mean 6 SD HDL (mmol/L), mean 6 SD LDL (mmol/L), mean 6 SD
With MBs, n 5 63
Without MBs, n 5 14
P value
41 (65.1) 62.31 6 12.86 44 (69.8) 5 (7.9) 14 (22.0) 13 (20.6) 169.30 6 28.76 96.41 6 16.61 7.17 6 2.35 4.41 6 .84 1.51 6 .74 1.36 6 .44 2.55 6 .75
12 (85.7) 56.16 6 12.80 8 (57.1) 0 (.0) 4 (28.6) 4 (28.6) 166.14 6 33.04 98.07 6 16.57 6.37 6 1.47 4.51 6 .59 1.30 6 1.56 1.44 6 .43 2.72 6 .56
.20 .11 .36 .56 .73 .49 .72 .74 .22 .39 .04 .50 .44
Abbreviations: BP, blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MB, microbleed; TG, triglycerides.
found that high TG levels were associated with deep or infratentorial MBs but not with strictly lobar MBs. These findings help clarify the potential role of lipids in ICH. Our finding of the MB frequency of 82% in patients with ICH is higher than the frequencies of 20%-70% reported in other studies.4 These differences may be due in part to ethnic and clinical differences among the patient populations. Moreover, we used SWI, which is more sensitive at detecting MBs by combining a long echo time and a fully flow-compensated 3-dimensional gradient-echo sequence than conventional T2*-weighted gradient-recalled echo imaging used in previous studies.5-7 Our results suggested that high TG levels were associated with MBs, especially in the deep or infratentorial region. Hypercholesterolemia has been identified as an independent risk factor for atherosclerotic disease and
high TG as a risk factor for atherosclerosis and vascular inflammation.8,9 Hypercholesterolemia was equally prevalent among patients with arterial atherosclerosis and patients with smallvessel cerebral infarction.10 Small-artery atherosclerosis weakens small-artery endothelial cells, making the vessels more prone to leakage and rupture,11-13 when the risk of vessel leakage and rupture is also greater in patients with elevated blood pressure or blood glucose and in patients on antithrombotic therapy, which is a standard treatment for cerebral infarction. Although above evidences partly explained that high TG levels are associated with MBs, future work should further identify relation. Our finding that TG levels are associated with occurrence of deep or infratentorial MBs, but not with occurrence of lobar MBs, provides insight into the potential
Table 2. Distribution of demographic and clinical characteristics across different triglyceride levels
Male, n (%) Age, mean 6 SD Hypertension n (%) Diabetes mellitus n (%) Current smoking n (%) Alcohol intake n (%) Systolic BP (mm Hg) mean 6 SD Diastolic BP (mm Hg) mean 6 SD Glucose mean 6 SD Total cholesterol mean 6 SD HDL mean 6 SD LDL mean 6 SD Location of cerebral microbleeds Strictly lobar MBs (%) Deep or infratentorial MBs (%) Any MBs (%)
Tertile 1 (n 5 25) (, .97)
Tertile 2 (n 5 25) (.97-1.47)
Tertile 3 (n 5 27) (. 1.47)
P value
18 (72.0%) 62.4 16 (64.0%) 1 (4.0%) 9 (36.0%) 7 (25.0%) 170.56 6 28.99 99.08 6 18.40 6.75 6 2.12 4.25 6 .88 1.59 6 .49 2.45 6 .66
18 (72.0%) 61.4 16 (64.0%) 3 (13.6%) 4 (16.0%) 5 (20.0%) 169.76 6 30.96 96.68 6 16.54 7.60 6 2.57 4.27 6 .78 1.29 6 .41 2.62 6 .74
18 (66.7%) 59.5 20 (74.1%) 1 (3.7%) 5 (18.6%) 5 (18.6%) 164.96628.37 94.89 6 15.30 6.75 6 1.94 4.46 6 .76 1.24 6 .34 2.66 6 .76
.88 .72 .67 .40 .19 .68 .76 .67 .30 .57 .01 .55
3 (12.0%) 2 (8.0%) 17 (68.0%)
4 (16.0%) 7 (28.0%) 19 (76.0%)
4 (14.8%) 10 (37.0%) 26 (96.3%)
.92 .04 .03
Abbreviations: BP, blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MB, microbleed.
C. LEI ET AL.
1198
role of lipids in MBs. MBs in the deep or infratentorial area were associated with radiologic signs of small-vessel disease, white-matter hyperintensities, and lacunar infarcts and clinically related to hypertension or atherosclerotic microangiopathy.5,12,13 However, MBs restricted to a lobar location is related to cerebral amyloid angiopathy. Consistent with our TG findings, some studies showed that cardiovascular risk factors and classic markers of ischemic small-vessel disease have reported associations with deep or infratentorial MBs but not with lobar MBs.3 These results may suggest that high TG levels are associated with the occurrence of deep or infratentorial MBs but not with the occurrence of lobar MBs. MBs in a given brain area strongly associated with later occurrence of ICH in the same area,4,14 which may provide novel insights into TG in the etiology of ICH and MBs. Although our results showed that TG was associated with deep or infratentorial MBs, but not with lobar MBs, which provided novel insights into TG in the etiology of MBs and ICH, we do not provide an explicit mechanism for why lipid levels would be associated specifically with MBs in deep or infratentorial regions. Assuming that this relationship is confirmed in larger studies, future research should examine the basis for this region-specific lipid involvement.
3.
4.
5.
6.
7.
8.
9. 10.
11. 12.
References 1. Lee SH, Ryu WS, Roh JK. Cerebral microbleeds are a risk factor for warfarin-related intracerebral hemorrhage. Neurology 2009;72:171-176. 2. Wieberdink RG, Poels MM, Vernooij MW, et al. Serum lipid levels and the risk of intracerebral hemorrhage:
13. 14.
the Rotterdam Study. Arterioscler Thromb Vasc Biol 2011;31:2982-2989. Vernooij MW, van der Lugt A, Ikram MA, et al. Prevalence and risk factors of cerebral microbleeds: the Rotterdam Scan Study. Neurology 2008;70:1208-1214. Roob G, Lechner A, Schmidt R, et al. Frequency and location of microbleeds in patients with primary intracerebral hemorrhage. Stroke 2000;31:2665-2669. Goos JD, van der Flier WM, et al. Clinical relevance of improved microbleed detection by susceptibilityweighted magnetic resonance imaging. Stroke 2011; 42:1894-1900. Cheng AL, Batool S, McCreary CR, et al. Susceptibilityweighted imaging is more reliable than T2*-weighted gradient-recalled Echo MRI for detecting microbleeds. Stroke 2013;44:2782-2786. Nighoghossian N, Hermier M, Adeleine P, et al. Old microbleeds are a potential risk factor for cerebral bleeding after ischemic stroke: a gradient-echo T2*-weighted brain MRI study. Stroke 2002;33:735-742. Soo YO, Yang SR, Lam WW, et al. Risk vs benefit of antithrombotic therapy in ischaemic stroke patients with cerebral microbleeds. J Neurol 2008;255:1679-1686. Laloux P, Galanti L, Jamart J. Lipids in ischemic stroke subtypes. Acta Neurol Belg 2004;104:13-19. Schwartz EA, Reaven PD. Lipolysis of triglyceride-rich lipoproteins, vascular inflammation, and atherosclerosis. Biochim Biophys Acta 2012;1821:858-866. Talayero BG, Sacks FM. The role of triglycerides in atherosclerosis. Curr Cardiol Rep 2011;13:544-552. Shoamanesh A, Kwok CS, Benavente O. Cerebral microbleeds: histopathological correlation of neuroimaging. Cerebrovasc Dis 2011;32:528-534. Braun H, Schreiber S. Microbleeds in cerebral small vessel disease. Lancet Neurol 2013;12:735-736. Lee SH, Bae HJ, Kwon SJ, et al. Cerebral microbleeds are regionally associated with intracerebral hemorrhage. Neurology 2004;62:72-76.
Background: Lipid levels may be involved in intracerebral hemorrhage (ICH) etiology, which suggested that lipid levels may be associated with the etiology of microbleeds (MBs) in patients with ICH. To explore this, we examined the association between lipid levels and MBs in different cerebral regions in patients with ICH. Methods: Patients admitted to our hospital with ICH were consecutively and prospectively included. Demographic and clinical information were collected and analyzed according to the occurrence and location of MBs and levels of triglycerides (TGs). Results: Of the 77 patients included in our study, 63 (81.8%) were found to have MBs. Prevalence of MBs in the ‘‘deep or infratentorial’’ region and any region increased with increasing tercile in TG concentration; however, no such trend was observed for strictly lobar MBs. The odds ratio (OR) for occurrence of MBs in deep or infratentorial region was even higher for the third tercile relative to the first: 6.77 (95% confidence interval [CI] 1.31-34.96). The OR for occurrence of MB in any region was even higher for the third tercile relative to the first: 12.24 (95% CI 1.40-106.83). However, the OR for occurrence of deep or infratentorial region and any region in the second tercile relative to the first tercile did not reach significance. Moreover, TG levels did not appear to be associated with the occurrence of strictly lobar MBs. Conclusions: High TGs were associated with deep or infratentorial and any MBs but not with lobar MBs. This finding may shed light on the role of lipids in MB and ICH etiology. Key Words: Lipid levels—triglyceride—microbleeds— intracerebral hemorrhage. Ó 2014 by National Stroke Association
Introduction From the *Department of Neurology, West China Hospital, Sichuan University, Sichuan Province; and †Department of Neurosurgery, West China Hospital, Sichuan University, Sichuan Province, People’s Republic of China. Received September 28, 2013; revision received October 9, 2013; accepted October 16, 2013. C.L. and S.L. contributed equally to this study. Grant support: This research was supported by the Science and Technology Support Program of the Department of Science and Technology of Sichuan Province (2012FZ0006) and the National Key Technology R&D Program for the 12th Five-Year Plan of People’s Republic of China (2011BAI08B05). Conflict of interest: None. Address correspondence to Bo Wu, MD, PhD, Department of Neurology, West China Hospital, Sichuan University, No. 37 Guo Xue Xiang, Chengdu 610041, Sichuan Province, People’s Republic of China. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2014 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2013.10.016
Intracerebral hemorrhage (ICH) in a lobar location is associated with cerebral amyloid angiopathy, whereas ICH in the basal ganglia, cerebellum, or pons is attributed mainly to hypertension or atherosclerotic microangiopathy.1 Different factors contribute to ICH occurrence in different locations of the brain, suggesting different etiologies. Recent work suggests that lipid levels may be involved in ICH etiology,2 raising the question of whether they are involved in ICH at all locations or only at certain ones. Microbleeds (MBs) tend to occur in advance of ICH, suggesting that they are an early stage of the same disease.3 MBs are pathologic deposits of hemosiderin caused by minor blood extravasation from lipohyalinized small arterioles. If lipid levels are, indeed, involved in the etiology of ICH, they may also be involved in MBs. Therefore,
Journal of Stroke and Cerebrovascular Diseases, Vol. 23, No. 5 (May-June), 2014: pp 1195-1198
1195
C. LEI ET AL.
1196
analyzing whether lipid levels are associated with MBs, in particular brain locations, may reveal whether the same is true for ICH. In this prospective study, we investigated the prevalence and location of MBs in a population of patients with ICH. We examined whether lipid levels correlated with MB occurrence and location.
Subjects and Methods Patients and Evaluation Patients with spontaneous ICH were prospectively and consecutively enrolled after being admitted to West China Hospital, Sichuan University, Chengdu, China, between May 1, 2012, and March 31, 2013. Spontaneous ICH was defined as intraparenchymal hemorrhage or intraventricular hemorrhage not caused by trauma, brain tumor, or hemorrhagic transformation of arterial ischemic stroke or established pathogenesis, that is, arteriovenous malformation, head trauma, or cavernous angioma. Patients were also excluded if they did not undergo susceptibility-weighted imaging (SWI) or if lipoprotein analysis data were unavailable.
Baseline Data Collection and Categorization Baseline information was collected at admission on age, gender, and stroke risk factors (hypertension, diabetes mellitus, current smoking, alcohol intake). Cerebral MBs were defined as homogeneous, round focal areas observed throughout the brain, with a diameter less than 10 mm and very low signal intensity on SWI. Cerebral locations of MBs were categorized as either strictly lobar or as ‘‘deep or infratentorial.’’ Patients with 1 or more MBs restricted to a lobar location were defined as having strictly lobar MBs. Persons without lobar MBs and with MBs in a deep or infratentorial location were defined as having deep or infratentorial MBs. The presence and location of MBs on SWI was determined independently by 2 neurologists (inter-rater reliability k 5 .82) blinded to clinical data. In case of disagreement, a third neurologist was consulted, and a consensus decision was reached.
Statistical Analysis The c2 or Fisher exact tests were used to assess the significance of differences in categorical variables. The Mann–Whitney U test was used to assess the significance of differences in continuous variables. When appropriate, results were reported as an odds ratio (OR) and 95% confidence interval (CI). Two-sided values of P less than .05 were considered statistically significant. All statistical analyses were performed using SPSS version 16 (IBM, USA). Logistic regression modeling was used to identify variables associated with MB occurrence and location.
Results A total of 105 consecutive patients were assessed for inclusion in the study, but 21 (20.0%) were excluded because of established pathologies and 6 (5.7%) because of missing data on lipid levels. Of 77 patients included in the final analysis, 54 (70.1%) were men, and mean age at stroke onset was 61.06 6 12.96 years. Fifty-five patients had supratentorial ICH, 22 patients had infratentorial ICH, 9 patients had thalamic ICH, 33 patients had putaminal ICH, 2 patients had external capsule, and 11 patients had lobar ICH. A total of 63 patients (81.8%) had MBs. Baseline characteristics of patients with ICH are shown in Table 1 according to whether they had MBs. Patients with MBs were more likely to be older and to have higher triglyceride (TG) levels. Then, we examined that which risk factor influenced the MB lesion development in the ICH by univariate logistic regression analysis. Logistic regression identified increasing tercile in TG concentration as an independent marker of MBs lesions (P 5 .013, OR 5 2.86, 95% CI 1.25-6.58). The data were analyzed to determine whether TG levels were associated with particular demographic or clinical variables or with the occurrence and location of cerebral MBs. Triglyceride levels were not associated with demographic variables or risk factors or with the occurrence of strictly lobar MBs. There was a significant trend toward more occurrence of MBs in deep or infratentorial region (P 5 .04) and any region (P 5 .03) with increasing TG tercile (Table 2); however, no such trend was observed for strictly lobar MBs. Post hoc analysis showed occurrence of MBs in deep or infratentorial region rate to be significantly higher in the third tercile (37.0%) than in the first tercile (P 5 .014). In contrast, occurrence of MBs in deep or infratentorial region rate in the second tercile (28.0%) was not significantly different from those in the first (8.0%; P 5 .069). The OR for occurrence of MBs in deep or infratentorial region was even higher for the third tercile relative to the first: 6.77 (95% CI 1.31-34.96). The OR for occurrence of deep or infratentorial region in the second tercile relative to the first tercile was 4.47 (95% CI .83-24.19). Moreover, the rate of MBs in any region was significantly higher in the third tercile (96.3%) than in the first tercile (68.0%, P 5 .008). Rate of MBs in any region was not significant between the second tercile (76.0%) and the first (68.0%, P 5 .29). The OR for MBs in any region in the second tercile relative to the first tercile was 1.78 (95% CI .496.53). The OR was significantly higher for the third tercile relative to the first: 12.24 (95% CI 1.40-106.83).
Discussion In this prospective study of patients with ICH, we found that 82% had MBs, consistent with the idea that MBs are strongly associated with ICH and may represent an early stage of the disease. In addition, we
LIPID LEVELS AND CEREBRAL MICROBLEEDS IN ICH
1197
Table 1. Baseline characteristics of ICH patients with and without MBs
Male, n (%) Age, mean 6 SD Hypertension, n (%) Diabetes mellitus, n (%) Current smoking, n (%) Alcohol intake, n (%) Systolic BP (mm Hg), mean 6 SD Diastolic BP (mm Hg), mean 6 SD Glucose (mmol/L), mean 6 SD Total cholesterol (mmol/L), mean 6 SD TG (mmol/L), mean 6 SD HDL (mmol/L), mean 6 SD LDL (mmol/L), mean 6 SD
With MBs, n 5 63
Without MBs, n 5 14
P value
41 (65.1) 62.31 6 12.86 44 (69.8) 5 (7.9) 14 (22.0) 13 (20.6) 169.30 6 28.76 96.41 6 16.61 7.17 6 2.35 4.41 6 .84 1.51 6 .74 1.36 6 .44 2.55 6 .75
12 (85.7) 56.16 6 12.80 8 (57.1) 0 (.0) 4 (28.6) 4 (28.6) 166.14 6 33.04 98.07 6 16.57 6.37 6 1.47 4.51 6 .59 1.30 6 1.56 1.44 6 .43 2.72 6 .56
.20 .11 .36 .56 .73 .49 .72 .74 .22 .39 .04 .50 .44
Abbreviations: BP, blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MB, microbleed; TG, triglycerides.
found that high TG levels were associated with deep or infratentorial MBs but not with strictly lobar MBs. These findings help clarify the potential role of lipids in ICH. Our finding of the MB frequency of 82% in patients with ICH is higher than the frequencies of 20%-70% reported in other studies.4 These differences may be due in part to ethnic and clinical differences among the patient populations. Moreover, we used SWI, which is more sensitive at detecting MBs by combining a long echo time and a fully flow-compensated 3-dimensional gradient-echo sequence than conventional T2*-weighted gradient-recalled echo imaging used in previous studies.5-7 Our results suggested that high TG levels were associated with MBs, especially in the deep or infratentorial region. Hypercholesterolemia has been identified as an independent risk factor for atherosclerotic disease and
high TG as a risk factor for atherosclerosis and vascular inflammation.8,9 Hypercholesterolemia was equally prevalent among patients with arterial atherosclerosis and patients with smallvessel cerebral infarction.10 Small-artery atherosclerosis weakens small-artery endothelial cells, making the vessels more prone to leakage and rupture,11-13 when the risk of vessel leakage and rupture is also greater in patients with elevated blood pressure or blood glucose and in patients on antithrombotic therapy, which is a standard treatment for cerebral infarction. Although above evidences partly explained that high TG levels are associated with MBs, future work should further identify relation. Our finding that TG levels are associated with occurrence of deep or infratentorial MBs, but not with occurrence of lobar MBs, provides insight into the potential
Table 2. Distribution of demographic and clinical characteristics across different triglyceride levels
Male, n (%) Age, mean 6 SD Hypertension n (%) Diabetes mellitus n (%) Current smoking n (%) Alcohol intake n (%) Systolic BP (mm Hg) mean 6 SD Diastolic BP (mm Hg) mean 6 SD Glucose mean 6 SD Total cholesterol mean 6 SD HDL mean 6 SD LDL mean 6 SD Location of cerebral microbleeds Strictly lobar MBs (%) Deep or infratentorial MBs (%) Any MBs (%)
Tertile 1 (n 5 25) (, .97)
Tertile 2 (n 5 25) (.97-1.47)
Tertile 3 (n 5 27) (. 1.47)
P value
18 (72.0%) 62.4 16 (64.0%) 1 (4.0%) 9 (36.0%) 7 (25.0%) 170.56 6 28.99 99.08 6 18.40 6.75 6 2.12 4.25 6 .88 1.59 6 .49 2.45 6 .66
18 (72.0%) 61.4 16 (64.0%) 3 (13.6%) 4 (16.0%) 5 (20.0%) 169.76 6 30.96 96.68 6 16.54 7.60 6 2.57 4.27 6 .78 1.29 6 .41 2.62 6 .74
18 (66.7%) 59.5 20 (74.1%) 1 (3.7%) 5 (18.6%) 5 (18.6%) 164.96628.37 94.89 6 15.30 6.75 6 1.94 4.46 6 .76 1.24 6 .34 2.66 6 .76
.88 .72 .67 .40 .19 .68 .76 .67 .30 .57 .01 .55
3 (12.0%) 2 (8.0%) 17 (68.0%)
4 (16.0%) 7 (28.0%) 19 (76.0%)
4 (14.8%) 10 (37.0%) 26 (96.3%)
.92 .04 .03
Abbreviations: BP, blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MB, microbleed.
C. LEI ET AL.
1198
role of lipids in MBs. MBs in the deep or infratentorial area were associated with radiologic signs of small-vessel disease, white-matter hyperintensities, and lacunar infarcts and clinically related to hypertension or atherosclerotic microangiopathy.5,12,13 However, MBs restricted to a lobar location is related to cerebral amyloid angiopathy. Consistent with our TG findings, some studies showed that cardiovascular risk factors and classic markers of ischemic small-vessel disease have reported associations with deep or infratentorial MBs but not with lobar MBs.3 These results may suggest that high TG levels are associated with the occurrence of deep or infratentorial MBs but not with the occurrence of lobar MBs. MBs in a given brain area strongly associated with later occurrence of ICH in the same area,4,14 which may provide novel insights into TG in the etiology of ICH and MBs. Although our results showed that TG was associated with deep or infratentorial MBs, but not with lobar MBs, which provided novel insights into TG in the etiology of MBs and ICH, we do not provide an explicit mechanism for why lipid levels would be associated specifically with MBs in deep or infratentorial regions. Assuming that this relationship is confirmed in larger studies, future research should examine the basis for this region-specific lipid involvement.
3.
4.
5.
6.
7.
8.
9. 10.
11. 12.
References 1. Lee SH, Ryu WS, Roh JK. Cerebral microbleeds are a risk factor for warfarin-related intracerebral hemorrhage. Neurology 2009;72:171-176. 2. Wieberdink RG, Poels MM, Vernooij MW, et al. Serum lipid levels and the risk of intracerebral hemorrhage:
13. 14.
the Rotterdam Study. Arterioscler Thromb Vasc Biol 2011;31:2982-2989. Vernooij MW, van der Lugt A, Ikram MA, et al. Prevalence and risk factors of cerebral microbleeds: the Rotterdam Scan Study. Neurology 2008;70:1208-1214. Roob G, Lechner A, Schmidt R, et al. Frequency and location of microbleeds in patients with primary intracerebral hemorrhage. Stroke 2000;31:2665-2669. Goos JD, van der Flier WM, et al. Clinical relevance of improved microbleed detection by susceptibilityweighted magnetic resonance imaging. Stroke 2011; 42:1894-1900. Cheng AL, Batool S, McCreary CR, et al. Susceptibilityweighted imaging is more reliable than T2*-weighted gradient-recalled Echo MRI for detecting microbleeds. Stroke 2013;44:2782-2786. Nighoghossian N, Hermier M, Adeleine P, et al. Old microbleeds are a potential risk factor for cerebral bleeding after ischemic stroke: a gradient-echo T2*-weighted brain MRI study. Stroke 2002;33:735-742. Soo YO, Yang SR, Lam WW, et al. Risk vs benefit of antithrombotic therapy in ischaemic stroke patients with cerebral microbleeds. J Neurol 2008;255:1679-1686. Laloux P, Galanti L, Jamart J. Lipids in ischemic stroke subtypes. Acta Neurol Belg 2004;104:13-19. Schwartz EA, Reaven PD. Lipolysis of triglyceride-rich lipoproteins, vascular inflammation, and atherosclerosis. Biochim Biophys Acta 2012;1821:858-866. Talayero BG, Sacks FM. The role of triglycerides in atherosclerosis. Curr Cardiol Rep 2011;13:544-552. Shoamanesh A, Kwok CS, Benavente O. Cerebral microbleeds: histopathological correlation of neuroimaging. Cerebrovasc Dis 2011;32:528-534. Braun H, Schreiber S. Microbleeds in cerebral small vessel disease. Lancet Neurol 2013;12:735-736. Lee SH, Bae HJ, Kwon SJ, et al. Cerebral microbleeds are regionally associated with intracerebral hemorrhage. Neurology 2004;62:72-76.
