BPV associated with imaging features of SSI on MRI

Abstract Objectives A retrospective study was performed to investigate the relationship between blood pressure variability (BPV) and imaging features of single small infarction (SSI) on magnetic resonance imaging (MRI). Materials and Methods Two hundreds and five patients with SSI and 120 healthy subjects matched with age and sex as the control group were enrolled into this study. All subjects came from the Affiliated Hospital to Qingdao University and Qingdao Municipal Hospital from October 2011 to June 2016. Research subjects were classified into different groups. Blood pressure was measured once a day and recorded during the hospitalization period (7–10 days). The followed up data of patients after discharging from hospital was collected from the follow‐up records. Results Twenty‐four hours BPV (SBPMean, DSBPMax, DSBPSD, NDBPMax, NDBPSD, and DDBPCV), day‐to‐day, and visit‐to‐visit BPV (SBPMax, SBPSD, DBPMax, and DBPSD) in the SSI group were significantly higher than that in control group. Compared with the giant lacunar group, day‐to‐day BPV (SBPMean, SBPMax, SBPSD, SBPCV, DBPMean, DBPMax, DBPSD), and visit‐to‐visit BPV (SBPMean, SBPMax, SBPSD, DBPMean, DBPMax, DBPSD) were significantly higher in the small lacunar infarct group (p < .05). The 24 hr BPV (SBPMean, DDBPMax, DDBPMean), day‐to‐day BPV (SBPMax, SBPSD, SBPCV), and visit‐to‐visit SBPMax in nonround lesion group were significantly higher than that in round group (p < .05). Compared with nondeep lesion group, some parameters in day‐to‐day BPV and visit‐to‐visit BPV were significantly higher in the deep small lesion group (p < .05). Conclusion Increased BPV parameters such as day‐to‐day and visit‐to‐visit (SBPMax, SBPSD, DBPMax) were related to the SSI characterized by small lesion in deep brain region.

that stroke-related target organ damage is closely connected with BPV (Takase et al., 2011). Independent stroke prognosis information carried by BPV will change our current understanding of the importance of blood pressure (Veluw et al., 2017). Clinical studies have confirmed that improving BPV is another important goal in managing blood pressure level (Turtzo et al., 2009). There is a lot of evidence that the predictive effect of BPV on target organ damage is independent of the mean blood pressure. The visit-to-visit systolic BPV and maximal systolic blood pressure (SBP) are predictors of stroke independent of the mean systolic blood pressure. Rothwell et al. (2010a) found that the maximum of visit-to-visit systolic blood pressure and systolic BPV are independent risk factors for cardiovascular and cerebrovascular events (Rothwell et al. 2010a;Rothwell et al. 2010b). The predictive value of visit-to-visit BPV for cardiovascular events is higher than average blood pressure and 24 hr BPV. Lacunar infarction was always found both in well-controlled hypertensive patients and in nonhypertensive subjects (Musini & Wright, 2009).
Hypertensive factors are currently considered to play a dominant role in the development of small lacunar infarction.
Although the visit-to-visit BPV is linked to the incidence of stroke, the degree of association between visit-to-visit BPV and stroke may vary from different study populations. It has been noted that elevated SBP is a risk factor for lacunar infarction (Chen et al., 2008).
While few studies have involved in the relationship between BPV and imaging feature of SSI on MRI. Therefore, this study was performed to investigate the relationship between BPV and imaging feature of SSI on MRI.

| Patients
All subjects who were enrolled into this study came from the Affiliated Hospital to Qingdao University and Qingdao Municipal Hospital from October 2011 to June 2016. Two hundred and five patients with SSI were enrolled into this study according to inclusion criteria and exclusion criteria. ECG, echocardiography, intracranial and extracranial MRA (or CTA), TCD, and carotid color ultrasonography were used to search for etiologies. The etiological classification was based on the Chinese Ischemic Stroke Subtype (CISS) classification standard (Gao et al., 2011). Depending on the purpose of the study, different grouping methods were used: the patients were divided into round-shaped lesions and nonround lesions (wedge, triangle, and irregular) according to the morphology of the lesions; giant lacunar lesions group (40 mm > d ≥ 20 mm) and small lacunar lesions group (d < 20 mm) according to lesion size; deep infarction group (lesion located in basal ganglia, internal capsule, thalamus, brain stem); and nondeep infarction group (focus in cortex, radiography crown, and semi-ovoid center) according to lesion distribution. At the same time, in order to investigate the relationship between BPV and SSI, one hundred and twenty healthy subjects matched with age and gender were collected as the control group.
Inclusion criteria: (1) All cases were confirmed as new infarction (DWI positive and low apparent diffusion coefficient) by cranial MRI; (2) Hospitalized within 3 days after onset, the diameter of cerebral infarction was less than 40 mm in DWI. Exclusion criteria: (1) With obvious stroke sequelae; (2) Large areas of cerebral infarction (complete posterior circulation infarction and over 1/3 anterior circulation); (3) Transient ischemic attack (TIA); (4) Combined intracranial hemorrhage and hemorrhagic transformation after cerebral infarction, infection, and tumor; (5) With severe heart, liver, kidney, and other underlying diseases; (6) Age > 80 years, or with severe dementia or Parkinson's disease and Parkinson's syndrome; (7) Nervous system demyelinating diseases such as Guillain-Barre syndrome, multiple sclerosis, etc.; (8) Disturbance of consciousness. The hospital's institutional review committee on human research approved this study protocol.

| Clinical and imaging assessment
The severity of neurological impairments of the index stroke was measured using the National Institutes of Health Stroke Scale (NIHSS). NIHSS scores were recorded when patients were admitted to hospital. The risk factors including gender, smoking, alcoholism, hypertension, coronary heart disease, diabetes, fasting blood glucose, and high blood lipids were recorded. The main risk factors were determined as follow: (1) hypertension: antihypertensive drugs before admission or systolic blood pressure > 140 mmHg, or diastolic blood pressure ≥ 90 mmHg after hospital admission; (2) type 2 diabetes: a history of diabetes; fasting blood glucose >7.0 mmol/L or blood glucose 2 hr after meal > 11.1 mmol/L, while glycated hemoglobin >6.5%; (3) hyperlipidemia: past dyslipidemia or abnormalities after hospital admission, total cholesterol > 5.72 mmol/L, TG > 1.72 mmol/L, or LDL > 3.12 mmol/L; (4) Smoking: Currently smoking or quitting smoking (10 sticks/day for 5 years or more (5) Drinking: Daily alcohol consumption > 50ml, abstinence or not drinking).  Lesion with d < 20 mm is regarded as a small lacuna. For DWI being very sensitive to small lesions, and there is no lower limitation for the lesion diameter. Lesion with 40 mm > d ≥ 20 mm is regarded as giant lacuna. SSI was determined as follow: lesions occur in the cortex, cortex infarcts in the lower half of the oval center and in the coronal, basal ganglia, inner cystic striatum, thalamus, and brainstem regions, regardless of whether the main artery has any degree of stenosis, not considering the size of the lesion (not to exceed 1/3 anterior circulation, the maximum of transverse diameter of the lesion not exceed 40 mm). Neurologists and neuro-radiologists integrated the clinical manifestations and neuroimaging findings, respectively.

| The followed up data collecting about 24 hr ambulatory blood pressure and follow-up blood pressure
All the followed up data collected from the follow-up records were analyzed. Twenty-four-hour ambulatory blood pressure measurement: in order to minimize the adverse disturbance of hyperacute cerebral infarction on blood pressure, the MC-6800 ambulatory blood pressure monitoring (Shenzhen Meili Biomedical Electronics Co., Ltd.,) was performed after the onset of the disease once every 60 min, continuously monitor the blood pressure for 24 hr.08:00 and 20:00 were set as two time periods of blood pressure monitoring during the day and night. For some patients with acute cerebral infarction who were treated with intravenous thrombolysis after admission, ambulatory blood pressure monitoring was performed 24 hr after admission. The effective SBP measurement ranges from 70 mmHg to 260 mmHg, and DBP 50 mmHg to 150 mmHg. 24 hr effective blood pressure reading >90% is qualified. During hospitalization (general hospitalization 7-10 days), blood pressure is measured and recorded every morning. Patients were followed up after discharge. Patients were referred once a month in the outpatient clinic, and blood pressure measurements were taken at rest for 5 min and recorded. Parameters of blood pressure included as follow: daytime maximum systolic blood pressure (DSBP Max ), daytime maximum diastolic pressure (DDBP Max ), nighttime maximum systolic blood pressure (NSBP Max ), nighttime maximum diastolic blood pressure (NDBP Max ), 24 hr systolic blood pressure, 24 hr SBP Mean ), 24 hr mean diastolic blood pressure (24 hr DBP Mean ), daytime systolic blood pressure (DSBP Mean ), daytime diastolic blood pressure (DDBP Mean ), and mean nighttime systolic blood pressure (night systolic blood pressure, NSBP Mean ), nighttime diastolic blood pressure (NDBP Mean ), daytime systolic blood pressure coefficient of variation (DSBP CV ), nighttime systolic blood pressure (nighttime systolic blood pressure)-coefficient of variation (NSBP CV ), daytime diastolic blood pressure coefficient of variation (DDBP CV ), nighttime diastolic blood pressure coefficient of variation (NDBP CV ), standard deviation of daytime systolic blood pressure (daytime systolic blood pressure-standard deviation (DSBP SD ), nighttime systolic blood pressure-standard deviation (NSBP SD ), day diastolic blood pressure-standard deviation (DDBP SD ), and standard deviation of night diastolic blood pressure (nighttime diastolic blood pressurestandard deviation, NDBP SD ). The SBP Mean , SBP, DBP, SD, CV, and other parameters were collected. The CV calculation method is CV = 100 × SD/mean.
Measurement data and count data were expressed as mean ± standard deviation (Mean ± SD) and percentage. The t-test was used for data with normal distribution of data. The Mann-Whitney U test was used for nonparametric data that did not meet the normal distribution. The chi-square test was used for the count data. A logistic regression analysis was thus performed, before and after adjustment, for the effect of the potential confounding variables to evaluate the risk relationship between the radiographic phenotypes of SSI and BPV parameters. The following potential confounders were considered: demographic characteristics (sex, age), baseline MMSE score, vascular risk factors (hypertension, diabetes, smoking habits, and hyperlipidemia).SPSS 16.0 package for Windows was used for statistical analysis. p < .05 was considered statistically significant.

| Demographic and clinical characteristics of participants
A total of 205 patients with SSI were enrolled into this study. The average age was (69.4 ± 10.1) years, of which 113 (55.1%) were male.
The average age of the control group including 120 healthy subjects was (68.1 ± 10.8) years, of which 64 (53.3%) were male. The baseline characteristics of the SSI and noncerebral infarction groups are shown in Table 1. There were statistically significant differences in risk factors such as blood lipids, blood glucose, and hypertension (p < .05) between them. There were nonround cerebral infarction in 97 cases (36.5%), round-like cerebral infarction in 108 cases (63.5%).

cases of small lacuna lesions and 83 cases of giant lacunar le-
sions. According to the distribution of infarction lesions, patients were divided into deep infarction group (130 cases) (lesions located in the basal ganglia, internal capsule, thalamus, brainstem), and nondeep infarction group (75 cases) (lesions located in the cortex, radiant crown, and semi-oval center). Baseline characteristics of every subgroup showed in the Table 2.

| The composition of the different imaging feature categories of SSI
In the present study, MRI imaging features of SSI lesion were defined as x = round lesion (round), y = nonround lesion (nonround), m = small lacunar lesion (small), n = giant lacunar lesion (giant), Figure 1 shows the composition of the different imaging feature catego-

| The correlation between BPV and SSI
Compared with the noninfarction group, the SBP Mean , DSBP Max , DSBP SD , NDBP Max , NDBP SD, and DDBP CV in the SSI group were significantly higher in the aspect of 24 hr BPV. There was a statistically significant difference between these two groups (p < .05, Table 3). SBP Max , SBP SD , DBP Max , and DBP SD were significantly higher in the SSI group, and the difference between these two groups was statistically significant in the aspect of day-to-day and visit-to-visit BPV (p < .05). Visit-to-visit SBP CV was higher in the SSI group that in the noninfarction group. There was no statistically significant difference between these two groups (p = .06, p = .07), but the clinical significance should not be ignored (Table 3).

| Comparison of BPV parameters of 24 hr, dayto-day and visit-to-visit in different subgroups of SSI
In the aspect of 24 hr BPV parameters, SBP Mean , DDBP Max, and DDBP Mean were significantly different between the round group and the nonround group. SBP Mean , DDBP Max were higher in the round group, whereas DDBP Mean was higher in the nonround group.
There was no statistically significant trend in 24 hr BPV parameters between the small lacunar lesions group and giant lacunar lesions group. There was no significant difference in 24 hr BPV between deep infarction and nondeep infarction (p > .05).
In terms of day-to-day BPV parameters, SBP Max , SBP SD, and SBP CV were statistically significant in the round group and the nonround group (p < .05), and were higher in the nonround group. infarction group were significantly increased, and there was statistically significant between these two groups (p < .05). There were statistically significant differences in SBP Max and DBP Max among these four subgroups. The other parameters were not statistically different among these subgroups (p > .05).
In terms of visit-to-visit BPV parameters, SBP Max was statistically significant between the round group and the nonround group (p < .05), and was higher in the nonround group. SBP Mean , SBP Max , SBP SD , DBP Mean , DBP Max, and DBP SD were statistically significant between the small lacunar lesions group and giant lacunar lesions group (p < .05). Compared with nondeep infarction group, SBP Mean and SBP Max were significantly increased in the deep infarction group (p < .05). (Table 4).

| BPV related to imaging feature of SSI on MRI
In the present study, 64.8% of lacunar infarction appeared as round- showed that 61.4% of the SSI lesions were located in the deep brain.
Seventy-five percent of round lesions belonged to deep brain infarction, suggesting that different imaging feature of SSI had different propensity distribution. Small and round lacunar lesions are tending to locate in deep brain area such as basal ganglia and thalamus.
The proportion of hypertension in the small lacunar lesion group was higher than that in the giant lacunar lesion group, but there was not statistically significant in the 24-hr ambulatory blood pressure parameters between these two groups. However, there was a significant increase in visit-to-visit BPV in the small lacunar lesion group.
The maximum of SBP was significantly increased in the non-round lesion group and the small lacunar lesion group. Hypertension, SBP and DBP have similar epidemiological features in both regular and irregular lacunar infarction, whereas BPV is more pronounced in irregular lesions (Feng et al., 2013). The results of this study suggest that long-term BPV is closely related to nonround lesions. The imaging features of small focal in deep brain structures were closely related to BPV parameters of day-to-day and visit-to-visit including SBP Max , SBP SD, and DBP Max . This study found that systolic and diastolic blood pressure peaks and systolic BPV were independent high-risk factors for deep focal infarction. Hypertension mainly results in hyaline degeneration or fibrinoid necrosis of arterioles, consequently resulting in lacunar infarction or microinfarction. Clinical and animal experimental studies have demonstrated that long-term hypertension with significant BPV can lead to severe small-vascular endothelial dysfunction (Diaz et al., 2012), leading to rupture of the blood-brain barrier and expansion of the Virchow-Robin spaces (VRS) (Klarenbeek et al., 2013).

TA B L E 4 Comparison of BPV parameters of 24 hr, day-to-day and visit-to-visit in different subgroups of SSI
It is known that lacunar infarction is most closely related to hypertension (Altmann et al., 2015;Das et al., 2008), and lacunar infarction is mostly belong to SSI in the deep brain. Patients with SSI in deep brain had higher BPV parameters; especially visit-tovisit BPV was significantly higher in patients with SSI in deep brain than that with SSI patients in nondeep brain. Different pattern of BPV might be related to different distribution patterns of SSI.
The incidence of infarcted lesions was higher in the basal ganglia, brainstem, and centrum semiovale and corona radiata. BPV was associated with impaired endothelial function in small vessels (Diaz et al., 2012). The BPV factor has greater influence on the change of tensile stress that the blood vessel wall is underwent.
However, there was no significant difference in 24 hr ambulatory blood pressure parameters between these two groups.

| CON CLUS ION
This study finds that increased BPV parameters such as day-today and visit-to-visit (SBP Max , SBP SD , DBP Max ) were related to the SSI characterized by small lesion in deep brain region. Of course, this conclusion still needs to be confirmed by prospective studies. .186 CV 11.9 (5.3) 11.7 (4.7) .481 11.9 (5.1) 11.8 (4.9) .305 12 . .233 CV 9.6 (3.1)

TA B L E 4 (Continued)
In addition to 24 hr average blood pressure, long-term BPV, blood pressure peak, maximum control of rapid fluctuations should be focus on.

CO N FLI C T O F I NTE R E S T
None declared.

AUTH O R CO NTR I B UTI O N S
Putting forward this idea and project administration: Yong peng Yu.
Data curation, Methodology: Yong peng Yu, Lan Tan, Yali Zheng and Tingting Jiang. Software and Writing, review and editing: Yong peng Yu and Yali Zheng.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1002/brb3.2155.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.