H-type Hypertension is a risk factor for Cerebral Small Vessel Disease: a retrospective study

Background: The correlation between H-type hypertension and cerebral small-vessel diseases (CSVD) remains uncertain. The present study was designed to explore the possible relationship between H-type hypertension and CSVD spectrum and total burden. Methods: We included 329 patients in the present study and divided them into four groups: H-type hypertension group, the isolated hypertension group, the isolated hyperhomocysteinemia (HHcy) group and the control group. Clinical variables of interest and the MR examination sequences were obtained. We count the presence of each CSVD feature and rated the total burden of CSVD on an ordinal scale from 0 to 4 according to a recent described score rule. Results: The results showed that H-type hypertension was associated with the presence of CMB and the severity of WMH and PVS. CSVD total burden was signicantly related to age (OR: 1.059, 95% CI: 1.037-1.082), systolic pressure (OR: 1.122, 95% CI: 1.007-1.136), triglycerides (OR: 1.386, 95% CI: 1.037-1.854), isolated HHcy (OR: 4.154, 95% CI 1.836–9.401) and H-type hypertension (OR: 5.028, 95% CI 2.323-10.883). And we further observed hypertension and HHcy had a synergistic effect on CSVD total burden (OR: 2.776, 95% CI 1.564-4.927). Conclusions: H-type hypertension may act as an independent risk factor of CSVD and cause increased susceptibility to CSVD total burden and CSVD spectrum, which deserved further prevention measures.

Background Cerebral small-vessel diseases (CSVD) is an intrinsic disorder of small arteries, arterioles, venules and capillaries of the brain (1). From a clinical point of view, CSVD contributes to a risk of cognitive decline, dementia and stroke, and causes considerable worsening of cognitive function, gait, and balance (2). In recent years, the development of neuroimaging has improved the diagnostic rate of CSVD. The recognized neuroimaging spectrum ascribable to CSVD has been expanded including leukoaraiosis, cerebral microbleeds, lacunar infarcts, perivascular spaces and brain atrophy currently (1). Magnetic resonance (MR) is the gold standard imaging for CSVD, and four closely correlated features are markers on brain MR: white hyperintensities (WHM), lacunes, cerebral microbleeds (CMBs), and perivascular spaces (PVS).
Recently, a total SVD burden score was proposed (3)(4)(5), which capture the global effect of cerebral SVD and quanti ed the global burden with a combined score. In this score, one point is allocated to each of the following: presence of lacunes, presence of microbleeds, moderate to severe WHM, PVS grated. The score has been tested partly for association with vascular risk factors or stroke subtype in few studies(6, 7 fasting serum glucose > 7.0 mmol/L or postprandial 2h plasma glucose > 11.1 mmol/L or having previous history of diabetes. Hyperlipidemia was de ned as having elevated level of one of triglyceride, total cholesterol or low density lipoprotein. Venous blood samples from the participants were collected on an empty stomach the second day after admission. The serum Hcy level were measured within 24h of hospitalization using enzymatic cycling method. HHcy was de ned as Hcy concentration ≥12.0 umol/L.
Full neurological examination, brain CT or MRI scan and carotid ultrasonography were also recorded.
2.3 Brain MRI acquisition and Analysis: The MR examination was performed within 48 hours after admission and sequences included T1-weighted, T2-weighted, uid-attenuated inversion recovery (FLAIR), axial diffusion-weighted imaging (DWI), and TOF-MRA sequences. MR was rated for the presence of lacunes, white matter hyperintensities, cerebral microbleeds, and perivascular spaces independently.
Lacunes was de ned as rounded or ovoid shaped lesions, >3-and <20-mm diameter on T1, T2 or FLAIR sequences in basal ganglia, the white matter or brainstem. We de ned microbleeds as small (<5 mm), homogeneous, round foci of low signal intensity on gradient echo images in basal ganglia, white matter, cerebellum, brainstem or cortico-subcortical junction (10). EPVS was de ned as small (<3 mm) round or linear hyperintensities in the basal ganglia or centrum semiovale on T2 images, and they were rated using a ve-point ordinal scale12 as follows: 0=no EPVS, 1=1-10 EPVS, 2=11-20, EPVS, 3=21-40 EPVS, and 4=>40 EPVS. Three trained neurologists and 2 neuroradiologists, each of whom was blinded to clinical information rated all the available scans for the presence and severity of SVD features. Deep and periventricular WMH were both coded according to the Fazekas scale from 0 to 3 (11).
Based on the recent described score(4), we count the presence of each SVD feature and rated the total burden of SVD on an ordinal scale from 0 to 4. The MR manifestation of WMHs graded 2-3 according to Fazekas grading was recorded as 1 point, presence of CMBs or lacunes was recorded 1 point respectively, PVS graded 2-4 (≧11) was counted 1 point ( Table 1). The severity of total SVD burden score were divided into three categories: mild (0 or 1 point), moderate (2 points), severe (3 or 4 points)(3, 12).
Normally distributed variables were presented as the mean ± Standard Deviation (SD) and categorical data were presented as frequency or ratio. Kolmogorov-Smirnov was used for the test of normality of quantitative data. Levene's test was used to test homogeneity of variance. T test or one-way ANOVA was performed to compare the distribution of quantitative variables. χ 2 test was used to compare the distribution of classi cation index. To determine the independent factors related to CSVD, we performed one variable analysis and multiple logistic regression analysis by using a backward elimination method and set the probability at 0.10 for removal. The statistical signi cance level was set at p <0.05 in the present study.

Results
3.1 Baseline characteristics. A total of 329 (220 males and 109 females) participants were nally included and they were divided into four groups. There were 76 participants (23.10%) in the control group, 112 participants (34.04%) in the isolated HBP group, 53 participants (16.11%) in the isolated HHcy group and 88 participants (26.75%) in the H-type hypertension group at baseline.  Figure 1).

3.3
The association of risk factors with CSVD total burden. A number of predictors of CSVD were shown in the logistic regression model (Table 4). Single factor analysis indicated that age, systolic pressure and diastolic pressure were contributed to the high CSVD total burden. However, after the multivariate adjustment, CSVD total burden was signi cantly related to age

Discussion
A direct relationship between Hcy and CSVD spectrum has been observed in several studies. One study reported that patients suffered from CSVD with con uent leukoaraiosis had the highest serum Hcy level compared with other TOAST subtypes of stroke (13). Another study found that in Japanese patients elevated Hcy level is independently associated with leukoaraiosis, but not with the incidence of microbleeds (14). Lin et. al reported both smoking and total HCY levels were shown to be risk factors for LA occurrence(15). Wang et al observed an independent association between Hcy level and severity of the CMBs(16). Miwa et al reported that serum Hcy was associated with lacunas, CMBs, and strictly deep CMBs (17). Elevated Hcy level increases hypercogulability(18) and oxidative stress (3), induces endothelial dysfunction and smooth muscle cell proliferation (19), increases hypercoagulability, and thus contributed to the damage of blood brain barrier.
Our study is the rst to focus on the predictive impact of H-type hypertension on CSVD. In our research population, we studied the relationship between H-type hypertension and CSVD spectrum, and observed that H-type hypertension was associated with the presence of CMB and the severity of WMH and PVS.
However, CSVD imaging features usually occur together, a relative study showed that signs of two or more severe CSVD features may appear in around one-third of the patients suffered acute ischemic stroke. A number of researches indicated that quanti cation of global burden of CSVD is feasible, meaningful and has clinical relevance (20,21). The total CSVD burden may provide a more comprehensive view of the global impact of CSVD than do the single feature separately. Thus, in the present study, we aimed to study the effect of H-type hypertension on the CSVD total burden. We observed that isolated homocysteine and H-type hypertension were associated with CSVD total burden. Like other traditional risk factors, such as age, systolic pressure and triglycerides, H-type hypertension is an independent risk factor for CSVD total burden. Our study is the rst to focus on the predictive impact of H-type hypertension on CSVD spectrum and CSVD total burden.
In the present study, CSVD total burden was signi cantly related to age, systolic pressure, triglycerides, isolated HHcy and H-type hypertension. And we further observed hypertension and HHcy had a synergistic effect on CSVD total burden (OR: 2.776, 95% CI 1.564-4.927).
The certain synergistic effect of hypertension and HHcy can be observed in some recent studies (22,23).
A large population-based study showed that H-type hypertension contributed to a remarkable increase of stroke incidence compared with isolated HBP (24). Another study that enrolled 750 subjects of cardiac, cerebral, and peripheralreported disease reported that the incidence of atherosclerotic vascular diseases was about ve times higher than that of the patients who suffered from the isolated hypertension (25).
The primary mechanisms may be explained by the fact that HHcy activates the angiotensin-converting enzyme by inhibiting the production of endogenous hydrogen sul de to aggravate hypertension(26-28).
Another study discovered that reduction of Treg cells percentage might be an important cause of immune disorders in H-type hypertension patients. According to their results, HHcy oxidized to peroxide causing Tcell subsets imbalance and vascular injury aggravation(29). Thus, when hypertension and HHcy are combined, the adverse effects may be increased.
The potential limitations in the study need to be acknowledged. First, vitamin B12, pyridoxal-5-phosphate and some medicines are related to Hcy metabolism and may affect the results. However, in our study, we only tested the plasma homocysteine level and did not record patients' medications at baseline. In future follow-up studies, we will add these parameters and expect to further interpreted the relationship between H-type hypertension and CSVD. Subsequently, we only detected blood pressure and the level of plasm Hcy at one time-point and have no data on possible changes in the long-term. Finally, the sample size in our study was relatively small that might have an impact on the overall assessment of the results.

Conclusions
In conclusion, our study indicated that H-type hypertension may cause increased susceptibility to CSVD total burden and CSVD spectrum, which deserved further prevention measures.

Conclusions
In conclusion, our study indicated that H-type hypertension may cause increased susceptibility to CSVD total burden and CSVD spectrum, which deserved further prevention measures Availability of data and materials: The datasets used and/or analyzed during the current study are available from the corresponding author upon request.
Competing interests: The authors declare that they have no competing interests. Funding: This study was funded by Jiangsu provincial science and technology department key diseases standardized diagnosis and treatment project BE2016670 . The funding body made contributions to the design and data collection of the study, Authors' contributions: CXY and FQ devised the study. LT and LXY collected and analyzed data, and drafted the manuscript. DSS and KY participated in the clinical evaluation of the patients, and performed MRI data analysis and interpretation. DXY, YS and LSJ conceived of the study, and helped to draft and revise the manuscript. All authors provided intellectual input to the study and approved the nal version of the manuscript.   Figure 1 The comparison of CSVD total burden among the four groups. Mild CSVD total burden: 0 or 1 point, moderate CSVD total burden: 2 points, severe CSVD total burden :3 or 4 points.