Abstract
Objectives:
Cerebral microbleeds (CMB) are associated with increased risk of hemorrhagic transformation (HT) of ischemic stroke with alteplase. Whether the presence of CMB influences the risk of HT and discharge outcomes of stroke patients not receiving alteplase is unclear. We evaluated the factors associated with the presence of CMB, and if the rates of HT and discharge outcomes were modified by the presence of CMB among stroke patients not treated with alteplase.
Methods:
Ischemic stroke patients who had MRI and did not receive alteplase were included in the study. CMB, HT and white matter hyperintensity (WMH) were evaluated using Microbleed Anatomical Rating Scale, Heidelberg bleeding classification, and Fazekas scales, respectively. Multivariate regression analysis was performed to evaluate factors associated with the presence of CMB.
Results:
Among 196 patients in the study, 58 (30%) patients had CMB. Nine patients had ≥10 CMBs. Median National Institutes of Health stroke scale score was 4. In multivariate analysis, age (OR=1.07;95%CI=1.01–1.12), history of stroke (OR=3.10;95%CI=1.08–8.92), congestive heart failure (OR=7.26;95%CI=1.58–33.42), admission diastolic blood pressure (OR=1.03;95%CI=1.003–1.06) and severe WMH defined as Fazekas score 4–6 (OR=4.69;95%CI=1.80–12.23) were significantly associated with the presence of CMB. There was no difference in HT (10% vs 12%, p=0.80) or discharge outcomes (modified Rankin Scale 0–2: 53% vs 57%, p=0.62) of patients with CMB compared to those without CMB.
Conclusion:
CMB are associated with severe WMH and higher diastolic blood pressure. CMB are not associated with the HT occurrence or discharge outcome of mild ischemic stroke in the absence of alteplase.
Keywords: Ischemic stroke, cerebral microbleeds, hemorrhagic transformation, white matter hyperintensity, alteplase
1. Introduction
Cerebral microbleeds (CMB) are chronic hemorrhages that have had growing interest in their role in cerebral amyloid angiopathy, dementia and risk for intracerebral hemorrhage [1, 2]. Patients with CMB and recent history of ischemic stroke or transient ischemic attack have higher risk of future ischemic stroke and intracranial hemorrhages [1]. CMB increases the risk of anticoagulation-related intracranial hemorrhage [3].
In patients with ischemic stroke, hemorrhagic transformation (HT) can be a natural course with several factors potentially increasing the risk including uncontrolled hypertension, diabetes mellitus, older age, increased severity of stroke as measured by the National Institutes of Health Stroke Scale (NIHSS) score, concurrent use of antithrombotics, use of alteplase, mechanical thrombectomy, hyperglycemia, lower low-density lipoprotein or total cholesterol, and renal impairment [4]. Presence of CMB and higher number of CMB are associated with increased risk of symptomatic HT following treatment with alteplase in acute ischemic stroke [5–7]. One previous study reported no increased risk of HT in patients with CMBs who did not receive alteplase.[8] Whether the discharge outcomes of stroke patients not receiving alteplase is modified by the presence of CMBs is unclear.
The objective of this study is 1) to evaluate the clinical and imaging factors associated with the presence of CMB, and 2) to evaluate the risk of HT and discharge outcomes stratified by presence of CMB among stroke patients not treated with alteplase or endovascular therapy.
2. Methods
2.1. Data source and study design
This is a retrospective observational study of consecutive ischemic stroke patients admitted to the stroke service from July 1, 2015 to June 30, 2016. The Institutional Review Board at University of Florida approved the study.
2.2. Study population
Patients who did not receive alteplase or endovascular therapy were included in the study if they had interpretable diffusion-weighted imaging (DWI), susceptibility weighted imaging (SWI) or gradient-recalled echo (GRE), and fluid-attenuated inversion recovery (FLAIR) sequences on magnetic resonance imaging (MRI) of the brain and if stroke lesion was present on DWI. Patients with transient ischemic attack, clinically diagnosed ischemic stroke without DWI lesion and uninterpretable MRI sequences were excluded. Patients included were chart reviewed to abstract the clinical, laboratory and discharge data elements needed for the study ― (1) demographics, (2) medical history, (3) social history, (4) clinical presentation, (5) vital signs, (6) admission NIHSS score, (7) laboratory readings for hemoglobin, platelet count, blood urea nitrogen, creatinine, coagulation profile, lipid profile and hemoglobin A1c, (8) modified Rankin scale (mRS) score at the time of discharge, (9) discharge disposition and (10) TOAST (Trial of Org 10172 in Acute Stroke Treatment) etiological classification [9]. Intracranial atherosclerosis was included in large artery disease when determining TOAST category.
2.3. Imaging protocol
MRI images were obtained from 1.5T Avanto or 3T Verio Siemens scanners. Stroke protocol MRI included DWI, FLAIR, and SWI sequences. Typical parameters for DWI were TR/TE=4100/102 or 7300/80 or 8200/89 with b=0 and 1000 s/mm2, 4–5mm thickness; for FLAIR TR/TE=9000/90 or 9000/126, 4mm thickness; and for SWI, TR/TE=28/20 or 49/40, and 3mm thickness. One patient had GRE.
2.4. Imaging assessment
Two investigators (NN, AF) evaluated all DWI, SWI and FLAIR images independently. Discrepancies in image reads were resolved by consensus. DWI lesions were categorized based on vascular territory and structures involved. Images were reviewed blinded to clinical history.
White matter hyperintensity (WMH) was evaluated on FLAIR sequence using the semi quantitative Fazekas rating scale [10]. FLAIR hyperintensity in deep white matter (DWM) was graded on a scale of 0–3; 0–no lesion, 1–punctate foci, 2–beginning confluence of foci, and 3–large confluent areas [10]. Periventricular white matter (PVWM) lesions were also graded on a 0–3 scale; 0–no lesion, 1–caps or pencil-thin lining, 2–smooth halo, and 3–irregular periventricular hyperintensities extending into deep white matter [10]. FLAIR changes corresponding to acute DWI lesions were excluded in this grading. The total score calculated from the sum of scores for DWM and PVWM hyperintensities was dichotomized as 0–3 (mild) versus 4–6 (severe) for analysis.
CMB and HT were evaluated on the same SWI sequence obtained during admission. Microbleed Anatomical Rating Scale (MARS) was used to classify CMB [11]. It is shown to have good interrater reliability for presence of definite microbleeds. Circular homogenously hypointense lesions measuring 2–10 mm in diameter on SWI were identified as CMB. Those hypointensities likely to be mimics as described in MARS criteria were excluded. Hypointensities in basal ganglia were compared to CT head obtained at admission to exclude mimics such as correlating calcium deposits. Additionally, hypointense lesions were followed through sequential slices to ensure they were not small vessels. CMB distribution was recorded and classified into lobar (frontal, parietal, temporal, occipital, insula), deep (basal ganglia, internal and external capsule, corpus callosum, thalamus, deep and periventricular white matte), and infratentorial (brainstem, cerebellum) per MARS scale. Definite and possible CMB were combined to obtain total count for analysis.
HT was graded using the Heidelberg bleeding classification system [12]. Broadly, it is differentiated into hemorrhagic infarct (HI) or parenchymal hematoma (PH). HI was classified into HI1 – scattered small petechiae, no mass effect, and HI2 – confluent petechiae, no mass effect. PH was classified as PH1– hematoma within infarcted tissue, occupying <30% and no substantive mass effect, PH2 – hematoma occupying ≥30% infarcted tissues, with obvious mass effect. Intraventricular hemorrhages, subdural hemorrhages, subarachnoid hemorrhages, and remote PH from the area of infarct were also noted. Patients who do not receive alteplase usually do not have a follow-up imaging unless there is clinical change. Therefore, the MRI brain performed during admission was used to evaluate both CMB and HT.
2.5. Outcomes
The primary outcome was radiologically defined HT assessed on SWI. Secondary outcomes were discharge outcome defined by mRS and discharge disposition. Good functional outcome was defined as mRS 0, 1 or 2 at the time of discharge. Good discharge outcome was defined as discharge to home, home with home health services, and acute inpatient rehabilitation. Discharge to a skilled nursing facility, long-term care facility, hospice and death were considered as poor discharge outcome.
2.6. Statistical analysis
The study dataset was divided into two groups ― patients with CMB and without CMB. Categorical variables were reported as percentages and continuous variables are reported as mean with standard deviation or median with interquartile range. Chi-squared test was performed for categorical variables and univariate regression analysis was performed on continuous variables to assess significance between two groups. Multivariate logistic regression analysis was performed to evaluate factors associated with the presence of CMB adjusting for covariates. The variables that were significant in the univariate analysis and the variables that were considered as potential confounders were included as covariates. It included demographics (age, sex, race, ethnicity), medical history (hypertension, diabetes mellitus, hyperlipidemia, stroke, transient ischemic attack, atrial fibrillation, coronary artery disease, and congestive heart failure), use of antithrombotics, blood pressure, hemoglobin, platelet count, renal function (blood urea nitrogen and creatinine), lipid profile (low density lipoprotein, high density lipoprotein, total cholesterol and triglycerides), and total Fazekas score. Model goodness-of-fit was expressed as c-statistic. The alpha level for statistical significance was 0.05. Odds ratio (OR) and 95% confidence intervals (CI) were reported for the results of regression models. Statistical analysis was performed using SAS version 9.4 SAS Institute Inc, Cary, NC.
3. Results
3.1. Baseline clinical, laboratory and imaging characteristics
A total of 196 patients met the study criteria. CMB were present in 58 (30%) patients – 43 patients had 1–5 CMBs, 6 patients had 6–10 CMBs, and 9 patients had >10 CMBs (one patient each with 11, 24, 33, 38, 49, 52 and 171 CMBs and two patients with 13 CMBs). Patients with CMB were older compared to those without CMB (72.0 ±13.0 vs 63.6±14.4 years, p=0.0002). In univariate analysis, compared to patients without CMB, patients with CMB were more likely to have a history of hypertension (90% vs 72%, p=0.009), stroke (45% vs 21%, p=0.0007), atrial fibrillation (26% vs 8%, p=0.0008) and congestive heart failure (17% vs 4%, p=0.007), use of antithrombotics (67% vs 38%, p=0.0002), higher systolic blood pressure (median 163 vs 152, p=0.003), diastolic blood pressure (median 89 vs 82, p=0.008), and high-density lipoproteins (median 48 vs 45, p=0.02). There was no difference in social history between the two groups (table 1 and 2). Strokes due to cardioembolism and small vessel disease were higher in patients with CMB than those without CMB but it did not reach statistical significance (p=0.08).
Table 1:
Clinical characteristics
| CMB Absent, n=138 | CMB Present, n=58 | p value | |
|---|---|---|---|
| Age, years, mean (SD) | 63.6 (14.4) | 72.0 (13.0) | 0.0002 |
| Sex, female, n (%) | 67 (49) | 33 (57) | 0.29 |
| Race, n (%) | |||
| 1. White | 106 (77) | 40 (69) | 0.47 |
| 2. Black | 31 (22) | 17 (29) | |
| 3. Other | 1 (1) | 1 (2) | |
| Ethnicity, Hispanic, n (%) | 5 (4) | 3 (5) | 0.70 |
| Past medical history, n (%) | |||
| 1. Hypertension | 100 (72) | 52 (90) | 0.009 |
| 2. Diabetes mellitus | 50 (37) | 20 (34) | 0.82 |
| 3. Hyperlipidemia | 44 (32) | 27 (46) | 0.05 |
| 4. Stroke | 29 (21) | 26 (45) | 0.0007 |
| 5. Transient ischemic attack | 7 (5) | 7 (12) | 0.12 |
| 6. Atrial fibrillation | 11 (8) | 15 (26) | 0.0008 |
| 7. Coronary artery disease | 29 (21) | 15 (26) | 0.46 |
| 8. Congestive heart failure | 6 (4) | 10 (17) | 0.007 |
| 9. Peripheral vascular disease | 4 (3) | 0 (0) | 0.32 |
| Social history, n (%) | |||
| 1. Tobacco use | 74 (54) | 27 (46) | 0.37 |
| 2. Alcohol use | 48 (35) | 19 (33) | 0.79 |
| 3. Illicit drug use | 10 (7) | 4 (7) | 1.00 |
| Medication history, n (%) | |||
| 1. Antiplatelet | 48 (35) | 36 (62) | 0.0004 |
| 2. Anticoagulants | 7 (5) | 6 (10) | 0.18 |
| 3. Antithrombotic (antiplatelet or anticoagulant) | 53 (38) | 39 (67) | 0.0002 |
| Vitals sings, median (IQR) | |||
| 1. Systolic blood pressure, mm Hg | 152 (137–167) | 163 (143–181) | 0.003 |
| 2. Diastolic blood pressure, mm Hg | 82 (71–93) | 89 (76–100) | 0.008 |
| 3. Body mass index, kg/m2 | 28 (24–33) | 27 (23–31) | 0.08 |
| NIH stroke scale score, median (IQR)† | 4 (2–8) | 3 (1–8) | 0.80 |
| TOAST classification, n (%) | |||
| 1. Large artery disease | 30 (22) | 9 (16) | 0.08 |
| 2. Cardioembolism | 23 (17) | 17 (29) | |
| 3. Small vessel occlusion | 36 (26) | 20 (34) | |
| 4. Other determined etiology | 13 (9) | 2 (3) | |
| 5. Undetermined etiology | 36 (26) | 10 (17) |
SD = standard deviation; IQR = interquartile range;
Missing NIH stroke scale for 12 and 3 patients in CMB absent and present group respectively;
Table 2:
Laboratory and imaging characteristics
| CMB Absent, n=138 | CMB Present, n=58 | p value | |
|---|---|---|---|
| Laboratory characteristics | |||
| Lipid profile, mg/dl, median (IQR)† | |||
| 1. Low density lipoprotein | 102 (68–131) | 93 (72–120) | 0.45 |
| 2. High density lipoprotein | 45 (33–55) | 48 (39–64) | 0.02 |
| 3. Total cholesterol | 174 (140–209) | 171 (149–204) | 0.96 |
| 4. Triglycerides | 112 (81–164) | 97 (73–131) | 0.11 |
| Complete blood count, median (IQR) | |||
| 1. Hemoglobin, g/dl | 14 (13–15) | 14 (12–15) | 0.14 |
| 2. Platelets, x103/mL | 217 (185–268) | 209 (169–237) | 0.10 |
| Renal function, mg/dl, median (IQR) | |||
| 1. Blood urea nitrogen | 15 (10–19) | 16 (12–21) | 0.35 |
| 2. Creatinine | 0.9 (0.7–1.1) | 0.9 (0.8–1.1) | 0.27 |
| Coagulation profile, median (IQR)‡ | |||
| 1. Prothrombin time, seconds | 13 (13–14) | 13 (13–14) | 0.55 |
| 2. INR | 1 (1–1.1) | 1 (1–1.1) | 0.50 |
| 3. Partial thromboplastin time, seconds | 29 (27–32) | 29 (27–32) | 0.63 |
| Hemoglobin A1c, %, median (IQR)† | 6.1 (5.6–7.4) | 5.9 (5.5–6.4) | 0.08 |
| Imaging characteristics | |||
| Last known well to MRI, days, median (IQR) | 1.07 (0.66 – 1.96) | 0.90 (0.55 – 1.41) | 0.03 |
| Vascular Territory, n (%) | |||
| 1. Anterior cerebral artery | 10 (7) | 4 (7) | 1.00 |
| 2. Middle cerebral artery | 89 (64) | 43 (74) | 0.19 |
| 3. Anterior choroidal artery | 6 (4) | 2 (3) | 1.00 |
| 4. Posterior cerebral artery | 27 (20) | 16 (28) | 0.22 |
| 5. Basilar artery | 27 (20) | 4 (7) | 0.03 |
| 6. Vertebral artery | 11 (8) | 7 (12) | 0.36 |
| Fazekas scale score, n (%) | |||
| 1. Periventricular hyperintensity | |||
| a) Score 0 | 21 (15) | 4 (7) | <0.0001 |
| b) Score 1 | 57 (41) | 10 (17) | |
| c) Score 2 | 44 (32) | 21 (36) | |
| d) Score 3 | 16 (12) | 23 (40) | |
| 2. Deep white matter hyperintensity | |||
| a) Score 0 | 38 (28) | 4 (7) | <0.0001 |
| b) Score 1 | 67 (49) | 18 (31) | |
| c) Score 2 | 27 (20) | 28 (48) | |
| d) Score 3 | 38 (4) | 8 (14) | |
| 3. Total Fazakas score | |||
| a) Score 0–3 | 104 (75) | 20 (34) | <0.0001 |
| b) Score 4–6 | 34 (25) | 38 (66) |
Missing labs for 2 and 1 patients in CMB absent and present group respectively;
Missing labs for 21 and 14 patients in CMB absent and present group respectively
IQR = interquartile range;
Patients with CMB had MRI performed earlier from stroke onset compared to patients without CMB (median 0.90 vs 1.07 days, p=0.03). They were also less likely to have strokes involving the basilar artery and its perforating branches. There was no difference between the two groups for stroke location (not shown). Patients with CMB had more severe PVWM (40% vs 12%, p<0.0001) and DWM (14% vs 4%, p<0.0001) hyperintensities compared to those without CMB (table 2). Table 3 shows the distribution of CMBs based on MARS scale. Severe WMH (total Fazekas score 4–6) was associated with deep CMB compared to no deep CMB [21/33 (64%) vs 51/163 (31%), p=0.0004]; and similarly, with infratentorial CMB [15/24 (63%) vs 57/172 (33%), p=0.005]; and lobar CMB [22/27 (81%) vs 50/169 (30%), p<0.0001].
Table 3:
Distribution of cerebral microbleed for each location based on Microbleed Anatomical Rating Scale (MARS)
| Location of microbleed | Definite | Possible |
|---|---|---|
| 1. Lobar, n (%) | 25 (13) | 9 (5) |
| a) Frontal | 18 (9) | 8 (4) |
| b) Parietal | 15 (8) | 4 (2) |
| c) Temporal | 15 (8) | 5 (3) |
| d) Occipital | 15 (8) | 2 (1) |
| e) Insula | 0 (0) | 0 (0) |
| 2. Deep, n (%) | 26 (13) | 13 (7) |
| a) Basal ganglia | 11 (6) | 9 (5) |
| b) Internal capsule | 2 (1) | 1 (1) |
| c) External capsule | 4 (2) | 0 (0) |
| d) Corpus callosum | 2 (1) | 0 (0) |
| e) Thalamus | 16 (8) | 3 (2) |
| f) Deep and periventricular white matter | 6 (3) | 0 (0) |
| 3. Infratentorial, n (%) | 19 (10) | 7 (4) |
| a) Brainstem | 7 (4) | 3 (2) |
| b) Cerebellum | 16 (8) | 6 (3) |
3.2. Factors associated with presence of CMBs in stroke
On multivariate analysis, higher age (OR=1.07 95%CI=1.01–1.12), history of stroke (OR=3.10 95%CI=1.08–8.92), history of congestive heart failure (OR=7.26 95%CI=1.58–33.42), admission diastolic blood pressure (OR=1.03 95%CI=1.003–1.06) and severe WMH (OR=4.69 95%CI=1.80–12.23) were significantly associated with the presence of CMB (table 4).
Table 4.
Multivariable logistic regression analysis for predictors of presence of cerebral microbleeds.
| Variable | OR (95% CI) | p value |
|---|---|---|
| Cerebral microbleed | ||
| 1. Age | 1.07 (1.01–1.12) | 0.0157 |
| 2. History of stroke | 3.10 (1.08–8.92) | 0.0356 |
| 3. History of congestive heart failure | 7.26 (1.58–33.42) | 0.0109 |
| 4. Diastolic blood pressure | 1.03 (1.003–1.06) | 0.0327 |
| 5. Total Fazekas score 4–6 vs 0–3 | 4.69 (1.80–12.23) | 0.0016 |
| Model c-statistic 0.876 |
3.2. Outcomes
HT was present in 22 (11%) patients – HI1 and 2 in 9 and 8 patients, respectively, and PH1 and 2 in 2 patients each. One patient with subarachnoid hemorrhage was included in PH2 for analysis. There was no difference in HT in patients with CMB compared to those without CMB [6/58 (10%) vs 16/138 (12%), p=0.80, table 5]. None of the nine patients with CMB ≥10 had HT. Good functional outcome and good discharge outcome was not significantly different between the two groups (table 5).
Table 5.
Outcomes of stroke stratified by cerebral microbleeds (CMB)
| Outcome | CMB Absent (n=138) | CMB Present (n=58) | p value |
|---|---|---|---|
| Hemorrhagic transformation, n (%) | 16 (12) | 6 (10) | 0.80 |
| 1. Hemorrhagic infarction 1 | 6 (4) | 3 (5) | |
| 2. Hemorrhagic infarction 2 | 6 (4) | 2 (3) | |
| 3. Parenchyma hematoma 1 | 2 (1) | 0 (0) | |
| 4. Parenchyma hematoma 2 | 2 (1) | 1 (2) | |
| Good functional outcome, mRS (0–2), n (%) | 79 (57) | 31 (53) | 0.62 |
| Discharge disposition, n (%) | |||
| 1. Good discharge outcome | 114 (83) | 46 (79) | 0.59 |
| 2. Poor discharge outcome | 24 (17) | 12 (21) |
4. Discussion
Our study shows that in patients presenting with ischemic stroke, CMBs are found among patients with higher age, history of stroke, congestive heart failure, higher admission diastolic blood pressure and severe WMH. Presence of CMB or the burden of CMB (≥10) was not associated with HT or discharge outcomes of stroke not treated with alteplase.
CMBs were present in 30% of patients with ischemic stroke in our study which is similar to that noted in other studies [7, 13]. Our finding of higher diastolic blood pressure being associated with CMB is consistent with Rotterdam scan study [14], and the Age, Gene/Environment Susceptibility (AGES)–Reykjavik Study [15]. Our study consisted of ischemic stroke patients compared to the Rotterdam scan study which included healthy adults >60 years. The Rotterdam scan study reported that higher diastolic blood pressure was associated with CMB in the lobar locations [14]. A follow-up to the Rotterdam scan study noted an association of severe hypertension and high systolic blood pressure with the deep or infratentorial CMB [16]. In the AGES study, while the severe hypertension was associated with CMB, mild hypertension was not [15]. In our study, hypertension was not categorized into mild or severe, and it is likely that this led to hypertension being non-significant in multivariable model. Multi-collinearity could be a potential reason for lack of hypertension association with CMBs in multivariable analysis [17]. This is usually inevitable as the pathophysiology of many diseases stem from the same risk factors. It is likely that the Fazekas’s score and age that were more strongly associated with CMB in univariate analysis overshadowed hypertension. When the Fazekas score was removed from the model, hypertension (OR=3.744 95%CI=1.032–13.591) was significantly associated with presence of CMB.
Our study showed that CMB in all locations – lobar, deep and infratentorial were associated with WMH as noted in Rotterdam scan study [16]. CMB and WMH represents various manifestations of small vessel disease. Lobar CMB are suggested to reflect CAA and deep CMB are suggested to reflect hypertensive vasculopathy although they may coexist [18]. Both CAA (a cerebral small vessel disease) and hypertensive vasculopathy are associated with CMB and WMH at baseline with increased progression in number of CMB and volume of WMH over time [19, 20].
Our study finding that CMB is not associated with HT or poor discharge outcomes in the absence of thrombolysis is reassuring. In one study, 44.6% (66/148) patients had CMB and it was not associated with increased risk of HT of stroke not treated with alteplase.[8] This study did not evaluate discharge outcomes stratified by presence of CMB.
Being aware of presence of CMB, its location and burden has important clinical considerations. A meta-analysis showed that compared to patients without CMBs, patients with CMB and recent stroke or transient ischemic attack have higher risk of future intracranial hemorrhage [adjusted hazard ratio 2.45 (1.82–3.29)] than ischemic stroke [adjusted hazard ratio 1.23 (1.08–1.40)] with a steeper increase especially with ≥5 CMB.[1] However, among patients with CMBs the absolute increase in ischemic stroke was higher compared to increase in intracranial hemorrhage regardless of the burden or distribution (lobar, deep or mixed) of CMBs [1]. A meta-analysis showed that presence of CMB was associated with 2.7 times increased risk of intracerebral hemorrhage at follow-up after 6 months of anticoagulation [3]. The risk increased to six times when ≥5 CMB were present, with annual incidence of 2.48% [3]. Patients with atrial fibrillation can make educated decisions when provided with these hemorrhagic risk probabilities when on anticoagulation and risk of ischemic stroke (based on CHADS2 criteria) when not on anticoagulation. Given these consequences of CMB, aggressive management of modifiable risk factors particularly hypertension is of utmost important.
Our study has limitations, a major one being the use of one-time SWI performed to evaluate both CMB and HT. MRI was performed approximately one day/24 hours from symptom onset. Patients treated with alteplase usually get 24-hour post-treatment follow-up imaging to evaluate for HT. The MRI performed in our patient population is closer to this time line. Unlike alteplase that may cause hemorrhages mimicking CMB, HT is unlikely to cause CMB. Further, we excluded CMBs within the area of acute stroke. Therefore, the one-time MRI performed may suffice to evaluate CMB and HT.
HT may occur few days after stroke and this is not captured in our study. Our study attempted to evaluate the risk of HT among patients with CMB within the usual stroke care protocol. Our study patients had mild strokes and therefore HT was noted in only 22 patients (11.2%). The risk of HT is higher with severe stroke and those with larger infarct volume. It could be conceived that patients with more severe strokes are more likely to seek medical attention early and be treated with alteplase or mechanical thrombectomy and these patients were excluded in our study. This is a single center study with small number of patients. Larger studies with mild to severe strokes may provide more insight on association of CMB with HT in absence of alteplase.
5. Conclusion
CMBs are associated with severe WMH and higher diastolic blood pressure. Aggressive blood pressure control could lower occurrence of CMB. Presence of CMB was not associated with the HT occurrence or discharge outcomes of mild ischemic stroke not treated with alteplase.
Funding:
Research reported in this publication was supported in part by the National Center For Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR001427. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Disclosures
Nandakumar Nagaraja: Consultant stroke adjudicator for Women’s Health Initiative study. Receives research funding from 1Florida Alzheimer’s Disease Research Center.
Amreen Farooqui, Abdullah Bin Zahid, and Supreet Kaur: Not applicable
References
- 1.Wilson D, Ambler G, Lee KJ, Lim JS, Shiozawa M, Koga M et al. Cerebral microbleeds and stroke risk after ischaemic stroke or transient ischaemic attack: a pooled analysis of individual patient data from cohort studies. Lancet Neurol. 2019;18(7):653–65. doi: 10.1016/S1474-4422(19)30197-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Akoudad S, Wolters FJ, Viswanathan A, de Bruijn RF, van der Lugt A, Hofman A et al. Association of Cerebral Microbleeds With Cognitive Decline and Dementia. JAMA Neurol. 2016;73(8):934–43. doi: 10.1001/jamaneurol.2016.1017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Charidimou A, Karayiannis C, Song TJ, Orken DN, Thijs V, Lemmens R et al. Brain microbleeds, anticoagulation, and hemorrhage risk: Meta-analysis in stroke patients with AF. Neurology. 2017;89(23):2317–26. doi: 10.1212/WNL.0000000000004704. [DOI] [PubMed] [Google Scholar]
- 4.Álvarez-Sabín J, Maisterra O, Santamarina E, Kase CS. Factors influencing haemorrhagic transformation in ischaemic stroke. The Lancet Neurology. 2013;12(7):689–705. doi: 10.1016/S1474-4422(13)70055-3. [DOI] [PubMed] [Google Scholar]
- 5.Charidimou A, Turc G, Oppenheim C, Yan S, Scheitz JF, Erdur H et al. Microbleeds, Cerebral Hemorrhage, and Functional Outcome After Stroke Thrombolysis: Individual Patient Data Meta-Analysis. Stroke. 2017. doi: 10.1161/STROKEAHA.116.012992. [DOI] [PubMed] [Google Scholar]
- 6.Tsivgoulis G, Zand R, Katsanos AH, Turc G, Nolte CH, Jung S et al. Risk of Symptomatic Intracerebral Hemorrhage After Intravenous Thrombolysis in Patients With Acute Ischemic Stroke and High Cerebral Microbleed Burden: A Meta-analysis. JAMA Neurol. 2016;73(6):675–83. doi: 10.1001/jamaneurol.2016.0292. [DOI] [PubMed] [Google Scholar]
- 7.Nagaraja N, Tasneem N, Shaban A, Dandapat S, Ahmed U, Policeni B et al. Cerebral Microbleeds are an Independent Predictor of Hemorrhagic Transformation Following Intravenous Alteplase Administration in Acute Ischemic Stroke. J Stroke Cerebrovasc Dis. 2018;27(5):1403–11. doi: 10.1016/j.jstrokecerebrovasdis.2017.12.044. [DOI] [PubMed] [Google Scholar]
- 8.Ozbek D, Ozturk Tan O, Ekinci G, Midi I. Risk of hemorrhage in ischemic stroke and its relationship with cerebral microbleeds. Clin Neurol Neurosurg. 2018;168:112–7. doi: 10.1016/j.clineuro.2018.03.007. [DOI] [PubMed] [Google Scholar]
- 9.Adams HP Jr., Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL et al. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment. Stroke. 1993;24(1):35–41. doi: 10.1161/01.str.24.1.35. [DOI] [PubMed] [Google Scholar]
- 10.Kapeller P, Barber R, Vermeulen RJ, Ader H, Scheltens P, Freidl W et al. Visual rating of age-related white matter changes on magnetic resonance imaging: scale comparison, interrater agreement, and correlations with quantitative measurements. Stroke. 2003;34(2):441–5. [DOI] [PubMed] [Google Scholar]
- 11.Gregoire SM, Chaudhary UJ, Brown MM, Yousry TA, Kallis C, Jager HR et al. The Microbleed Anatomical Rating Scale (MARS): reliability of a tool to map brain microbleeds. Neurology. 2009;73(21):1759–66. doi: 10.1212/WNL.0b013e3181c34a7d. [DOI] [PubMed] [Google Scholar]
- 12.von Kummer R, Broderick JP, Campbell BC, Demchuk A, Goyal M, Hill MD et al. The Heidelberg Bleeding Classification: Classification of Bleeding Events After Ischemic Stroke and Reperfusion Therapy. Stroke. 2015;46(10):2981–6. doi: 10.1161/STROKEAHA.115.010049. [DOI] [PubMed] [Google Scholar]
- 13.Cordonnier C, Al-Shahi Salman R, Wardlaw J. Spontaneous brain microbleeds: systematic review, subgroup analyses and standards for study design and reporting. Brain. 2007;130(Pt 8): 1988–2003. doi: 10.1093/brain/awl387. [DOI] [PubMed] [Google Scholar]
- 14.Poels MM, Vernooij MW, Ikram MA, Hofman A, Krestin GP, van der Lugt A et al. Prevalence and risk factors of cerebral microbleeds: an update of the Rotterdam scan study. Stroke. 2010;41(10 Suppl):S103–6. doi: 10.1161/STROKEAHA.110.595181. [DOI] [PubMed] [Google Scholar]
- 15.Ding J, Sigurdsson S, Garcia M, Phillips CL, Eiriksdottir G, Gudnason V et al. Risk Factors Associated With Incident Cerebral Microbleeds According to Location in Older People: The Age, Gene/Environment Susceptibility (AGES)-Reykjavik Study. JAMA Neurol. 2015;72(6):682–8. doi: 10.1001/jamaneurol.2015.0174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Poels MM, Ikram MA, van der Lugt A, Hofman A, Krestin GP, Breteler MM et al. Incidence of cerebral microbleeds in the general population: the Rotterdam Scan Study. Stroke. 2011;42(3):656–61. doi: 10.1161/STROKEAHA.110.607184. [DOI] [PubMed] [Google Scholar]
- 17.Vatcheva KP, Lee M, McCormick JB, Rahbar MH. Multicollinearity in Regression Analyses Conducted in Epidemiologic Studies. Epidemiology (Sunnyvale). 2016;6(2). doi: 10.4172/2161-1165.1000227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Vernooij MW, van der Lugt A, Ikram MA, Wielopolski PA, Niessen WJ, Hofman A et al. Prevalence and risk factors of cerebral microbleeds: the Rotterdam Scan Study. Neurology. 2008;70(14):1208–14. doi: 10.1212/01.wnl.0000307750.41970.d9. [DOI] [PubMed] [Google Scholar]
- 19.Gregoire SM, Brown MM, Kallis C, Jager HR, Yousry TA, Werring DJ. MRI detection of new microbleeds in patients with ischemic stroke: five-year cohort follow-up study. Stroke. 2010;41(1):184–6. doi: 10.1161/STROKEAHA.109.568469. [DOI] [PubMed] [Google Scholar]
- 20.Chen YW, Gurol ME, Rosand J, Viswanathan A, Rakich SM, Groover TR et al. Progression of white matter lesions and hemorrhages in cerebral amyloid angiopathy. Neurology. 2006;67(1):83–7. doi: 10.1212/01.wnl.0000223613.57229.24. [DOI] [PMC free article] [PubMed] [Google Scholar]