Small vessel disease burden and response to induced hypertension therapy in small vessel occlusion stroke with early neurological deterioration

Abstract

Background:

Induced hypertension therapy (iHT) is used to rescue early neurological deterioration (END) in small vessel occlusion (SVO) stroke. However, underlying perforator dysfunction reflected by small vessel disease (SVD) burden may attenuate its therapeutic effect.

Aim:

To investigate whether white matter hyperintensity (WMH) volume and total SVD score are associated with neurological response to iHT.

Methods:

Consecutive patients with acute SVO stroke who developed END and received iHT were identified from a prospective registry (January 2017–July 2025). Regional WMH volumes were quantified on magnetic resonance imaging fluid-attenuated inversion recovery images. Outcomes included early neurological improvement (ENI) following iHT and 3-month modified Rankin Scale (mRS). Associations were evaluated using binary and ordinal logistic regression.

Results:

Among 178 patients (median age, 68 years; 55.6% men), 93 (52.2%) achieved ENI. Higher WMH volumes were independently associated with a lower likelihood of ENI (odds ratio (OR) per twofold increase [95% confidence interval]: total, 0.57 [0.43–0.76]; periventricular, 0.57 [0.43–0.76]; deep, 0.68 [0.54–0.86]). Periventricular WMH volume was also associated with an unfavorable shift in 3-month mRS (common OR, 0.80 [0.65–0.99]). Increasing total SVD score was associated with worse 3-month mRS shift (common OR, 0.61 [0.43–0.85]) and a non-significant trend against ENI (OR, 0.70 [0.48–1.01]; p = 0.056).

Conclusion:

Greater SVD burden was associated with a poorer neurological response to iHT in SVO stroke with END. Imaging markers of SVD may help identify patients less likely to achieve neurological improvement following iHT, suggesting a need for alternative individualized rescue strategies.

Graphical abstract

Keywords

Cerebral infarction cerebral small vessel disease clinical deterioration therapeutics white matter diseases

Introduction

Small vessel occlusion (SVO) stroke is generally considered a mild subtype with favorable outcomes; however, a substantial proportion of patients experience early neurological deterioration (END) during the acute phase.^1,2 Because the END in SVO stroke is associated with worse functional outcomes,³ identifying effective acute-phase treatment strategies and the patients most likely to benefit from these interventions remains a major clinical challenge.

Although the mechanisms underlying END in SVO stroke are heterogeneous, hemodynamic insufficiency appears to play a central role. Impaired perfusion in perforator-dependent territories may lead to progressive ischemic injury and neurological worsening.^4,5 To augment perfusion in this setting, induced hypertension therapy (iHT) using vasopressors has been used as one of the rescue strategies in clinical practice, particularly in Korea.⁶ Although previous studies suggest that iHT may promote neurological recovery with acceptable safety, treatment responses vary widely, and reliable predictors of therapeutic response have not been established.⁷

This heterogeneity in response to iHT may be attributable to underlying small vessel dysfunction. When the perforator function is severely impaired, increases in systemic blood pressure may fail to translate into effective augmentation of microvascular perfusion, thereby limiting the therapeutic benefit of iHT. Because perforating arteries cannot be reliably evaluated using standard imaging modalities in routine clinical practice,⁸ assessment of small vessel dysfunction primarily relies on indirect imaging markers. White matter hyperintensity (WMH) volume and the total small vessel disease (SVD) score are readily available magnetic resonance imaging (MRI) markers of small vessel dysfunction relevant to perfusion augmentation strategies. Prior studies have demonstrated that greater WMH burden is associated with a higher risk of END in SVO stroke,^9,10 supporting WMH as a marker of vulnerability to acute hemodynamic failure. While WMH reflects an important aspect of small vessel dysfunction, the total SVD score integrates WMH with other manifestations of SVD, including lacunes, cerebral microbleeds, and enlarged perivascular spaces (ePVS), and has been validated as a robust marker of overall SVD burden and stroke-related outcomes.^11–14

Accordingly, this study aimed to determine whether regional WMH volumes and total SVD score, as imaging markers of small vessel dysfunction, are associated with neurological response to induced hypertension therapy in patients with acute SVO stroke who develop END.

Methods

Study population

We retrospectively analyzed consecutive patients with acute ischemic stroke (AIS) due to SVO from a prospective stroke registry at Asan Medical Center. This study included patients admitted within 7 days of symptom onset between January 2017 and July 2025. Patients were eligible if they (1) underwent brain MRI, including fluid-attenuated inversion recovery (FLAIR) sequences required for automated WMH segmentation and quantification, (2) developed END within 7 days of onset, and (3) subsequently received iHT for END. END was defined as an increase of ⩾2 points in the total National Institutes of Health Stroke Scale (NIHSS) score or ⩾1 point in the motor item after admission. Patients with pre-existing brain lesions on FLAIR images (e.g. chronic non-lacunar infarction sequelae, intracranial hemorrhage, brain tumors, perilesional edema, or postoperative changes) were excluded if visual review of the segmentation results confirmed that these non-WMH lesions were falsely segmented as WMH by the automated software. This study was approved by the Institutional Review Board of Asan Medical Center (No. 2025-0015), which waived the requirement for informed consent owing to the retrospective study design.

iHT protocol

At Asan Medical Center, iHT is used as a first-line rescue therapy for patients with SVO stroke who develop END, as previously described.¹⁵ Briefly, in the absence of contraindications—significant cardiac diseases (e.g. congestive heart failure, clinically relevant arrhythmias, or coronary artery disease with >50% stenosis), intracranial or other major sources of hemorrhage, or persistently elevated systolic blood pressure (SBP ⩾200 mm Hg) at the time of END—iHT was administered as a continuous intravenous infusion of phenylephrine, a selective α1-adrenergic agonist. Phenylephrine infusion was initiated at 0.3 μg/kg/min and titrated to achieve an approximately 20% increase in SBP from the baseline measured at the time of END, with serial neurological assessments using the NIHSS. If neurological improvement was not observed at the initial target, a further 10–20% increase in SBP was considered, while maintaining SBP below 200 mm Hg. Blood pressure targets were individualized according to neurological response. After achieving neurological stabilization, iHT was maintained for 1–3 days and gradually tapered. Patients were continuously monitored for blood pressure variability and treatment-related adverse events. Antithrombotic therapy was continued according to standard clinical practice and was not modified in relation to iHT administration.

Data collection

For eligible patients, data were collected on demographic characteristics, including age and sex, and stroke risk factors such as hypertension, diabetes, dyslipidemia, previous stroke, atrial fibrillation, smoking status, and prestroke use of antihypertensive medications. Index stroke-related variables included prestroke modified Rankin Scale (mRS) score, initial NIHSS score, lesion location (supratentorial or infratentorial), maximum axial diameter of the lesion on diffusion-weighted imaging (DWI), presumed stroke mechanism (lipohyalinotic degeneration (LD) or branch atheromatous disease (BAD)), receipt of intravenous thrombolysis, antithrombotic therapy regimen at the time of END, and onset-to-arrival time. The presumed stroke mechanism was considered BAD if the lesion had a maximum axial diameter of ⩾15 mm or extended to the basal pontine surface, and LD otherwise.¹⁶ Based on the documented time points of END onset, iHT initiation, and iHT termination, the following intervals were calculated: onset-to-END time, END-to-iHT time, and iHT duration. Time in therapeutic range (TTR) was calculated as the percentage of time during which SBP was maintained within the individualized target range, based on SBP values sampled at 4-h intervals during iHT, up to 72 h or until iHT discontinuation, whichever occurred first. NIHSS scores were recorded at the time of END and at completion of iHT. MRI field strength (1.5 or 3.0 T) was collected. Follow-up DWI obtained during hospitalization, when available, was reviewed for index lesion enlargement and new infarcts distinct from the index lesion.

Segmentation and quantification of regional WMH

WMH volume was quantified separately for periventricular and deep regions, with total WMH volume defined as the sum of these two components. Regional WMH volumes were assessed on FLAIR images acquired at the earliest available time point after presentation for the index stroke. Brain MRI was performed using 1.5 T (Magnetom Avanto, Siemens, Erlangen, Germany (n = 133); Signa HDxt, GE, Milwaukee, WI, USA (n = 1); Magnetom Essenza, Siemens (n = 1)) or 3.0 T (Magnetom Vida, Siemens (n = 34); Ingenia, Philips, Best, The Netherlands (n = 4); Signa Architect, GE (n = 3); Vantage Elan, Toshiba, Otawara, Japan (n = 2)) MR scanners. FLAIR images were acquired with the following parameters: slice numbers, 20–42; slice thickness, 3–5 mm; interslice gap, 3.6–7 mm; field of view, 192–230 × 219–230 mm; matrix, 192–325 × 168–255; TR, 9000–11,000 ms; TE, 90–140 ms; and flip angle, 90°–160°. Automated WMH segmentation and volumetric analysis were performed using a fully automated, deep learning-based software package (AQUA AD version 3.0; Neurophet Inc., Seoul, Korea), which has been validated for WMH quantification across neurological conditions, including stroke and dementia.^17–19 Periventricular WMH was defined as lesions with any portion located within 10 mm of the lateral ventricular margin, whereas lesions entirely beyond this distance were classified as deep WMH. All automated segmentation outputs were visually reviewed, and cases in which pre-existing brain lesions were erroneously identified as WMH were excluded, as described above. A representative illustration of automated regional WMH segmentation is shown in Supplemental Figure S1.

Total SVD score

The total SVD score was calculated for each patient according to a previously described method.¹¹ The score ranged from 0 to 4 and was derived by assigning one point for the presence of each of the following MRI markers of SVD: lacunes (⩾1), cerebral microbleeds (⩾1), ePVS (⩾11) in the basal ganglia of the most affected hemisphere,²⁰ and WMH, defined as periventricular WMH extending into the deep white matter and/or confluent or early confluent deep WMH.

Study outcomes

The primary outcome was early neurological improvement (ENI) after iHT, defined as neurological improvement within 48 h after iHT initiation, meeting either of the following criteria: a reduction of ⩾2 points in the total NIHSS score or an improvement of ⩾1 point in the motor component.¹⁵ The secondary outcome was the 3-month mRS score.

Statistical analysis

Baseline characteristics were compared using the Mann–Whitney U test for continuous variables and the chi-square test or Fisher’s exact test for categorical variables, as appropriate. Total, periventricular, and deep WMH volumes were treated as continuous variables and logarithmically transformed to correct for right-skewed distributions. For descriptive trend analyses across total SVD score categories, the Cochran–Armitage test was used for ENI rates, and the Jonckheere–Terpstra test was used for 3-month mRS scores. Binary logistic regression was used to assess associations between regional WMH volumes and the total SVD score with ENI. Associations with 3-month mRS scores were evaluated using shift analysis with proportional odds ordinal logistic regression. Owing to sparse observations, mRS categories 5 and 6 were combined (mRS 5, n = 2; mRS 6, n = 1). The proportional odds assumption was assessed using the Brant test of the parallel regression assumption. Covariates for multivariable adjustment were selected based on their clinical relevance as recognized determinants of SVD burden or neurological outcome after stroke. Covariate adjustment followed a hierarchical modeling strategy: unadjusted, Model 1, adjusted for age and sex; and Model 2, further adjusted for hypertension, diabetes, previous stroke, smoking, prestroke mRS score, initial NIHSS score, onset-to-END time, and END-to-iHT time. Model 2 thus included 10 covariates, corresponding to approximately 9 events per variable, a level at which the traditional 10 events-per-variable rule can reasonably be relaxed.²¹ Onset-to-END time and END-to-iHT time were log-transformed owing to right-skewness. When closely related variables reflected overlapping clinical constructs (e.g. initial NIHSS and NIHSS at END), only one representative variable was retained to minimize collinearity, with preference given to measures reflecting baseline status. Restricted cubic spline curves were constructed to visualize the shape of the association between regional WMH volumes and adjusted odds ratios for ENI across the range of WMH burden. Subgroup analyses were performed to examine the consistency of associations between total WMH volume and ENI according to key clinical factors, including age, sex, vascular risk factors, prestroke mRS score, initial NIHSS score, lesion location, presumed stroke mechanism, onset-to-END time, and END-to-iHT time. Several sensitivity analyses were performed to assess the robustness of the primary findings. First, the analysis was restricted to patients with END occurring within 48 h of stroke onset to ensure a more homogeneous population with END. Second, TTR was additionally adjusted to evaluate whether the observed associations were independent of the intensity of delivered therapy. Third, MRI field strength was additionally adjusted to account for potential scanner-related variability in WMH quantification. Fourth, the potential influence of the END-to-iHT time, previously identified as a determinant of ENI in this clinical setting,¹⁵ on the primary findings was further examined. The correlation between total WMH volume and END-to-iHT time was assessed using the Spearman test, and median END-to-iHT times were compared across quartiles of total WMH volume using the Kruskal–Wallis test. Whether the association between WMH burden and ENI was modified by this variable was also examined using a multivariable logistic regression model with a multiplicative interaction between regional WMH volumes and a restricted cubic spline of log-transformed END-to-iHT time; the p value for interaction was assessed by likelihood ratio test. All statistical analyses were performed using R (version 4.2.1; R Foundation for Statistical Computing, Vienna, Austria). A two-sided p value < 0.05 was considered statistically significant.

Results

Of 1652 patients admitted with acute SVO stroke within 7 days of onset, 190 patients (11.5%) developed END and received iHT. Among these, 12 patients were excluded owing to falsely segmented non-WMH lesions on automated analysis (chronic non-lacunar infarction sequelae, n = 6; intracranial hemorrhage, n = 2; brain tumor, n = 2; traumatic encephalomalacia, n = 1; and radiation necrosis–related edema, n = 1). Accordingly, 178 patients were included in the final analysis (Figure 1). The median [interquartile range] age of the study population was 68 [58–77] years, and 99 patients (55.6%) were men. Prestroke mRS and initial NIHSS scores were 0 [0–0] and 5 [3–6], respectively. The median onset-to-END and END-to-iHT times were 31.3 [16.0–48.0] and 3.0 [1.2–8.8] h. iHT was maintained for 68.0 [50.0–99.3] h, with a median TTR of 61.1% [47.6–77.8%]. ENI after iHT was achieved in 93 patients (52.2%). Patients who achieved ENI received iHT earlier after END onset than those who did not. Baseline characteristics stratified by ENI status are summarized in Table 1. Follow-up DWI was performed in 112 patients (62.9%) during hospitalization. Enlargement of the index lesion was observed in 42 patients (37.5%) and did not differ significantly between patients with and without ENI (32.7% vs. 42.1%, p = 0.41). No patient showed a new infarct distinct from the index lesion.

Figure 1.

Study flowchart. END, early neurological deterioration; iHT, induced hypertension therapy; SVO, small vessel occlusion; WMH, white matter hyperintensity.

Table 1.

Baseline characteristics according to early neurological improvement.

	Total (n = 178)	ENI (n = 93)	No ENI (n = 85)	p
Age, years	68 [58–77]	66 [57–74]	68 [61–78]	0.086
Male sex	99 (55.6%)	58 (62.4%)	41 (48.2%)	0.081
Hypertension	119 (66.9%)	63 (67.7%)	56 (65.9%)	0.92
Prestroke antihypertensive use	84 (47.2%)	46 (49.5%)	38 (44.7%)	0.63
Renin–angiotensin system inhibitors	61 (72.6%)	32 (69.6%)	29 (76.3%)
Calcium channel blockers	51 (60.7%)	24 (52.2%)	27 (71.1%)
Diuretics	21 (25.0%)	16 (34.8%)	5 (13.2%)
Beta blockers	14 (16.7%)	10 (21.7%)	4 (10.5%)
Diabetes	67 (37.6%)	32 (34.4%)	35 (41.2%)	0.44
Dyslipidemia	73 (41.0%)	37 (39.8%)	36 (42.4%)	0.85
Previous stroke	30 (17.0%)	12 (13.0%)	18 (21.4%)	0.20
Atrial fibrillation	9 (5.1%)	3 (3.2%)	6 (7.1%)	0.41
Smoking	68 (38.2%)	39 (41.9%)	29 (34.1%)	0.36
Prestroke mRS score	0 [0–0]	0 [0–0]	0 [0–0]	0.27
Initial NIHSS score	5 [3–6]	5 [3–6]	5 [3–7]	0.78
Lesion location				0.90
Supratentorial	115 (64.6%)	61 (65.6%)	54 (63.5%)
Infratentorial	63 (35.4%)	32 (34.4%)	31 (36.5%)
Maximal axial diameter, mm	13 [10–17]	12 [9–16]	13 [10–17]	0.18
Presumed stroke mechanism				0.96
Lipohyalinotic degeneration	104 (58.4%)	55 (59.1%)	49 (57.6%)
Branch atheromatous disease	74 (41.6%)	38 (40.9%)	36 (42.4%)
Intravenous thrombolysis	7 (3.9%)	4 (4.3%)	3 (3.5%)	>0.99
Antithrombotic therapy at END				0.90
Single antiplatelet	2 (1.1%)	1 (1.1%)	1 (1.2%)
Dual antiplatelet	171 (96.1%)	90 (96.8%)	81 (95.3%)
Triple antiplatelet	2 (1.1%)	1 (1.1%)	1 (1.2%)
Oral anticoagulant + antiplatelet	3 (1.7%)	1 (1.1%)	2 (2.4%)
Onset-to-arrival time, hours	15.6 [6.4–32.5]	14.2 [5.9–30.6]	16.3 [6.6–35.6]	0.41
Onset-to-END time, hours	31.3 [16.0–48.0]	28.3 [18.1–45.8]	32.4 [13.7–54.0]	0.69
END-to-iHT time, hours	3.0 [1.2–8.8]	2.5 [1.0–5.2]	5.0 [1.3–11.9]	0.009
iHT duration, hours	68.0 [50.0–99.3]	69.8 [50.0–100.2]	67.1 [53.0–91.3]	0.62
Time in therapeutic range, %	61.1 [47.6–77.8]	61.0 [46.4–72.2]	66.7 [50.8–80.3]	0.060
NIHSS at END occurrence	7 [5–8]	7 [5–9]	7 [5–8]	0.48
NIHSS at iHT end	6 [5–8]	6 [5–7]	7 [6–9]	<0.001
MRI field strength				0.74
1.5 T	135 (75.8%)	72 (77.4%)	63 (74.1%)
3.0 T	43 (24.2%)	21 (22.6%)	22 (25.9%)

Data are presented as numbers (%) or median [interquartile range], as appropriate. END, early neurological deterioration; ENI, early neurological improvement; iHT, induced hypertension therapy; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale.

Patients who achieved ENI had significantly lower total (ENI: median [interquartile range], 6.1 [2.1–12.6] mL vs. no ENI: 13.0 [5.6–25.7] mL; p < 0.001), periventricular (4.4 [1.3–8.8] mL vs. 8.8 [3.9–16.4] mL; p < 0.001), and deep WMH volumes (1.5 [0.7–4.6] mL vs. 3.6 [1.6–7.3] mL; p < 0.001) than those who did not achieve ENI. In regression analyses, greater total, periventricular, and deep WMH volumes were consistently associated with a lower likelihood of ENI, and these associations remained statistically significant after multivariable adjustment. Data on 3-month mRS were available for 170 of 178 patients (95.5%); functional outcome analyses were performed in this subset. The proportional odds assumption was satisfied for all models. Higher regional WMH volumes were associated with an unfavorable shift in the 3-month mRS distribution in unadjusted analyses; however, after adjustment, only periventricular WMH volume remained independently associated with the 3-month mRS (Table 2).

Table 2.

Associations between regional white matter hyperintensity volumes and total small vessel disease score with early neurological improvement and 3-month modified Rankin Scale score.

	ENI		3–month mRS
	OR (95% CI)	p	cOR (95% CI)	p
Total WMH volume
Unadjusted	0.66 (0.54–0.81)	<0.001	0.62 (0.52–0.74)	<0.001
Model 1^a	0.62 (0.48–0.79)	<0.001	0.72 (0.58–0.88)	0.002
Model 2^b	0.57 (0.43–0.76)	<0.001	0.80 (0.64–1.00)	0.052
Periventricular WMH volume
Unadjusted	0.68 (0.57–0.82)	<0.001	0.64 (0.55–0.75)	<0.001
Model 1^a	0.62 (0.49–0.79)	<0.001	0.73 (0.60–0.88)	0.001
Model 2^b	0.57 (0.43–0.76)	<0.001	0.80 (0.65–0.99)	0.042
Deep WMH volume
Unadjusted	0.70 (0.58–0.85)	<0.001	0.71 (0.60–0.84)	<0.001
Model 1^a	0.70 (0.57–0.87)	0.001	0.82 (0.68–0.98)	0.029
Model 2^b	0.68 (0.54–0.86)	0.001	0.89 (0.74–1.08)	0.25
Total SVD score
Unadjusted	0.68 (0.51–0.91)	0.009	0.43 (0.32–0.57)	<0.001
Model 1^a	0.70 (0.51–0.95)	0.024	0.50 (0.36–0.68)	<0.001
Model 2^b	0.70 (0.48–1.01)	0.056	0.61 (0.43–0.85)	0.004

ORs and cORs represent the effect per doubling of WMH volume or per 1-point increase in total SVD score. Analyses of 3-month mRS were based on 170 of 178 patients (95.5%). CI, confidence interval; cOR, common odds ratio; END, early neurological deterioration; ENI, early neurological improvement; iHT, induced hypertension therapy; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio; SVD, small vessel disease; WMH, white matter hyperintensity.

Adjusted for age and sex.

Adjusted for age, sex, hypertension, diabetes, previous stroke, smoking, prestroke mRS score, initial NIHSS score, onset-to-END time, and END-to-iHT time.

The total SVD score was modestly but significantly lower in patients with ENI (median [interquartile range], 1 [0–1] vs. 1 [0–2]; p = 0.021). Among the individual components of the total SVD score, fulfillment of the scoring criteria for WMH and ePVS was more prevalent in patients who did not achieve ENI (Supplemental Table S1). Increasing total SVD score was associated with a significant adverse trend toward lower ENI rates and a worse distribution of 3-month mRS scores (Figure 2); in regression analyses, the association with 3-month mRS remained significant, whereas that with ENI was attenuated (Table 2).

Figure 2.

Early neurological improvement and 3-month functional outcomes according to the total small vessel disease score. (a) Proportion of patients achieving ENI and (b) distribution of 3-month modified Rankin Scale scores stratified by total SVD score. ENI was defined as neurological improvement occurring within 48 h after iHT, indicated by either a reduction of ⩾2 points in the total NIHSS score or an improvement of ⩾1 point in the motor component of the scale.

Restricted cubic spline analyses demonstrated a decreasing likelihood of ENI after iHT with increasing total, periventricular, and deep WMH volumes (Figure 3). Subgroup analyses further demonstrated that the inverse association between total WMH volume and ENI was consistent across key clinical subgroups, with no significant interactions observed (Figure 4). Sensitivity analyses showed consistent findings across alternative models, including restriction to END within 48 h and additional adjustment for TTR and MRI field strength (Supplemental Table S2). No significant correlation was observed between total WMH volume and END-to-iHT time (Spearman ρ = 0.03, p = 0.70), and median END-to-iHT times did not differ significantly across quartiles of total WMH volume (Q1, 2.5 [1.3–8.7]; Q2, 3.1 [1.3–7.9]; Q3, 3.5 [1.2–11.8]; Q4, 3.0 [1.0–6.9] h; p = 0.96). Inverse associations between regional WMH volumes and ENI were consistent across END-to-iHT times, with no significant interactions (Supplemental Figure S2).

Figure 3.

Associations between regional white matter hyperintensity volumes and early neurological improvement. Restricted cubic spline curves derived from multivariable logistic regression models illustrate the associations between (a) total, (b) periventricular, and (c) deep WMH volumes and adjusted ORs for ENI. ORs were adjusted for covariates included in Model 2, as detailed in Table 2. Solid lines indicate adjusted ORs, shaded areas represent 95% confidence intervals, and the horizontal dashed line denotes an OR of 1. ORs are plotted on a logarithmic scale. ORs represent the effect per doubling of WMH volume.

Figure 4.

Odds ratios for early neurological improvement per total white matter hyperintensity volume across key clinical factors. The forest plot illustrates adjusted ORs and 95% confidence intervals for early neurological improvement across clinical subgroups, per doubling of total WMH volume. ORs were estimated using multivariable logistic regression models with multiplicative interaction terms and adjusted for the covariate set specified in Model 2 (Table 2). ORs are displayed on a logarithmic scale. END, early neurological deterioration; iHT, induced hypertension therapy; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio.

Discussion

In this study, greater total, periventricular, and deep WMH volumes were consistently associated with a lower likelihood of ENI following iHT in patients with acute SVO stroke who developed END. In addition, a higher total SVD score was associated with worse functional outcomes at 3 months and showed a trend toward reduced ENI rates. Notably, periventricular WMH volume showed the strongest overall association among regional WMH measures.

The inverse relationship between imaging markers of SVD and ENI suggests that the likelihood of ENI following systemic blood pressure augmentation decreases with increasing severity of small vessel pathology. Although direct assessment of perforating artery function is not routinely available in clinical practice, recent advances in 7 T MRI have enabled in vivo evaluation of perforating artery flow characteristics. While structural abnormalities remain difficult to detect, functional indices, including blood flow velocity and pulsatility index, have emerged as surrogate markers of perforating artery function.^22,23 Previous 7 T MRI studies have demonstrated significant associations between impaired perforating artery flow metrics and SVD markers, supporting a link between SVD severity and compromised microvascular perfusion.^24,25 Collectively, these findings suggest that greater SVD burden may reflect impaired perforating artery function and reduced hemodynamic effectiveness of iHT, potentially limiting delivery of augmented blood flow to at-risk tissue.

The association between WMH burden and 3-month mRS was less pronounced than that observed for ENI, possibly reflecting characteristics of the study population. All patients included in this study had already experienced END, a condition strongly associated with poor functional outcome,³ which may have reduced the ability to detect additional between-group differences in long-term outcomes. Nevertheless, periventricular WMH volume remained significantly associated with 3-month mRS. Although the pathophysiology of regional WMH is not fully understood and appears heterogeneous, the periventricular white matter is supplied by long terminal medullary arteries with limited anastomotic capacity, possibly contributing to structural susceptibility to ischemia in this region.²⁶ In acute SVO stroke, this may translate into vulnerability to perforator-related hypoperfusion. The closer association of periventricular WMH with both acute neurological response and long-term functional outcomes in our cohort may support this interpretation. Therefore, patients with extensive periventricular WMH may be less likely to achieve neurological improvement following blood pressure-based hemodynamic augmentation, and alternative rescue strategies for END, such as optimization of antithrombotic therapy, warrant consideration.

Several limitations should be acknowledged. First, although WMH volume was quantitatively assessed, other components of the total SVD score were not. In particular, the binary classification of lacunes and cerebral microbleeds within the total SVD score may have attenuated potential dose–response relationships, which may explain the lack of associations with ENI in our study. Second, several methodological aspects of WMH assessment warrant consideration. Because of software limitations, regional WMH analyses were not lateralized to the hemisphere of the index infarct, which may have provided additional insight into spatially specific treatment responses. In addition, because three-dimensional T1-weighted imaging is not routinely acquired in acute stroke protocols at our institution, intracranial volume could not be measured; consequently, WMH volumes were analyzed as absolute values without normalization to intracranial volume, potentially introducing measurement bias related to age, sex, and head size. Third, as all patients received iHT without an untreated comparator, spontaneous neurological recovery, possible in SVO stroke, cannot be excluded. To mitigate this, we limited ENI to 48 h of post-iHT initiation, when spontaneous improvement alone is less likely to fully explain the neurological response. Furthermore, all patients had established END, in whom spontaneous recovery is less likely to fully account for the observed improvement. Fourth, distinguishing a reduced iHT response from a general recovery deficit due to high SVD burden is challenging. Greater WMH burden may indicate broader SVD-related tissue vulnerability, including neuroinflammation, blood–brain barrier dysfunction, and endothelial impairment,^27,28 all potentially limiting ENI. Fifth, iHT targets only the hemodynamic component among the mechanisms underlying END development. Although subgroup analyses by presumed LD and BAD did not show a predominant effect of a subtype, conventional imaging cannot reliably distinguish among finer mechanistic processes, including hemodynamic compromise, edema, thrombus extension, and recurrent infarction in an adjacent area, each responding differently to iHT. Finally, although we carefully adjusted for major determinants of WMH burden and performed sensitivity analyses addressing potential sources of bias, the retrospective single-center design limits causal inference, and residual confounding cannot be excluded. Prospective studies are needed to validate these findings and clarify causal relationships.

Our findings challenge a one-size-fits-all approach to rescue treatment for END in SVO stroke and emphasize the importance of underlying microvascular heterogeneity in neurological responsiveness to iHT. Incorporating markers of small vessel pathology may help identify patients less likely to achieve ENI after iHT, for whom alternative rescue strategies warrant further investigation.

Supplemental Material

sj-docx-1-wso-10.1177_17474930261461834 – Supplemental material for Small vessel disease burden and response to induced hypertension therapy in small vessel occlusion stroke with early neurological deterioration

Supplemental material, sj-docx-1-wso-10.1177_17474930261461834 for Small vessel disease burden and response to induced hypertension therapy in small vessel occlusion stroke with early neurological deterioration by Wookjin Yang, Keon Yeup Kim, Jeong Kyu Lee, Hyun Sun Oh, Jeong Yun Song, Bum Joon Kim, Dong-Wha Kang, Sun U Kwon and Jun Young Chang in International Journal of Stroke

Footnotes

Author contributions

W.Y. and J.Y.C. contributed to the conception and the design of the study; W.Y., K.Y.K., J.K.L., H.S.O., J.Y.S., B.J.K., D.-W.K., S.U.K., and J.Y.C. contributed to the acquisition and analysis of data; W.Y. and J.Y.C. contributed to drafting the text and preparing the figures.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by a grant (No. 2025-00590) from the Asan Institute for Life Sciences, Asan Medical Center, Seoul, Korea.

Ethical considerations

This study was approved by the Institutional Review Board of Asan Medical Center (No. 2025-0015).

Informed consent

The Institutional Review Board waived the requirement for informed consent owing to the retrospective study design.

ORCID iDs

Wookjin Yang

Keon Yeup Kim

Jeong Kyu Lee

Bum Joon Kim

Jun Young Chang

Data availability statement

The data supporting the findings of this study are available from the corresponding author upon reasonable request.

Supplemental material

Supplemental material for this article is available online.

References

Steinke

Ley

SC.

Lacunar stroke is the major cause of progressive motor deficits. Stroke 2002; 33: 1510–1516.

Vynckier

Maamari

Grunder

, et al. Early neurologic deterioration in lacunar stroke: clinical and imaging predictors and association with long-term outcome. Neurology 2021; 97: e1437–e1446.

Yaghi

Raz

Yang

, et al. Lacunar stroke: mechanisms and therapeutic implications. J Neurol Neurosurg Psychiatry 2021; 92: 823–830.

Werring

Ozkan

Doubal

, et al. Early neurological deterioration in acute lacunar ischemic stroke: systematic review of incidence, mechanisms, and prospects for treatment. Int J Stroke 2025; 20: 7–20.

Kang

Yang

Lee

, et al. The role of phenylephrine in patients with small deep subcortical infarct and progressive weakness. J Neurol Sci 2017; 377: 107–111.

Bang

Chung

Kim

, et al. Therapeutic-induced hypertension in patients with noncardioembolic acute stroke. Neurology 2019; 93: e1955–e1963.

Lee

Kim

Jeon

Kang

Kwon

Kim

JS.

Pharmacologically induced hypertension therapy for acute stroke patients. J Stroke 2019; 21: 228–230.

Harteveld

De Cocker

Dieleman

, et al. High-resolution postcontrast time-of-flight MR angiography of intracranial perforators at 7.0 Tesla. PLoS ONE 2015; 10: e0121051.

Feng

Tan

, et al. Leukoaraiosis correlates with the neurologic deterioration after small subcortical infarction. J Stroke Cerebrovasc Dis 2014; 23: 1513–1518.

10.

Nannoni

Del Bene

Palumbo

, et al. Predictors of progression in patients presenting with minor subcortical stroke. Acta Neurol Scand 2015; 132: 304–309.

11.

Staals

Makin

Doubal

, et al. Stroke subtype, vascular risk factors, and total MRI brain small-vessel disease burden. Neurology 2014; 83: 1228–1234.

12.

Lau

Schulz

, et al. Total small vessel disease score and risk of recurrent stroke: validation in 2 large cohorts. Neurology 2017; 88: 2260–2267.

13.

Song

Kim

Song

, et al. Total cerebral small-vessel disease score is associated with mortality during follow-up after acute ischemic stroke. J Clin Neurol 2017; 13: 187–195.

14.

Ryu

Jeong

Kim

DE.

Total small vessel disease burden and functional outcome in patients with ischemic stroke. PLoS ONE 2020; 15: e0242319.

15.

Jung

Ryu

Joon Kim

, et al. Time window for induced hypertension in acute small vessel occlusive stroke with early neurological deterioration. Stroke 2024; 55: 14–21.

16.

Nakase

Yoshioka

Sasaki

Suzuki

Clinical evaluation of lacunar infarction and branch atheromatous disease. J Stroke Cerebrovasc Dis 2013; 22: 406–412.

17.

Lee

Kim

Lee

, et al. Evaluating brain volume segmentation accuracy and reliability of FreeSurfer and Neurophet AQUA at variations in MRI magnetic field strengths. Sci Rep 2024; 14: 24513.

18.

Lee

Park

Kim

, et al. Impact of systolic blood viscosity on deep white matter hyperintensities in patients with acute ischemic stroke. J Am Heart Assoc 2024; 13: e034162.

19.

Kim

Wang

Kang

, et al. Development of a prediction model for cognitive impairment of sarcopenia using multimodal neuroimaging in non-demented older adults. Alzheimers Dement 2024; 20: 4868–4878.

20.

Maclullich

Wardlaw

Ferguson

Starr

Seckl

Deary

IJ.

Enlarged perivascular spaces are associated with cognitive function in healthy elderly men. J Neurol Neurosurg Psychiatry 2004; 75: 1519–1523.

21.

Vittinghoff

McCulloch

CE.

Relaxing the rule of ten events per variable in logistic and cox regression. Am J Epidemiol 2007; 165: 710–718.

22.

Bouvy

Geurts

Kuijf

, et al. Assessment of blood flow velocity and pulsatility in cerebral perforating arteries with 7-T quantitative flow MRI. NMR Biomed 2016; 29: 1295–1304.

23.

Geurts

Biessels

Luijten

Zwanenburg

Better and faster velocity pulsatility assessment in cerebral white matter perforating arteries with 7T quantitative flow MRI through improved slice profile, acquisition scheme, and postprocessing. Magn Reson Med 2018; 79: 1473–1482.

24.

Van Den Brink

Pham

Siero

, et al. Assessment of small vessel function using 7T MRI in patients with sporadic cerebral small vessel disease: the ZOOM@SVDs study. Neurology 2024; 102: e209136.

25.

van den Brink

Kopczak

Arts

, et al. CADASIL affects multiple aspects of cerebral small vessel function on 7T-MRI. Ann Neurol 2023; 93: 29–39.

26.

Smirnov

Destrieux

Maldonado

IL.

Cerebral white matter vasculature: still uncharted?

Brain 2021; 144: 3561–3575.

27.

Wardlaw

Smith

Dichgans

Small vessel disease: mechanisms and clinical implications. Lancet Neurol 2019; 18: 684–696.

28.

Hannawi

Cerebral small vessel disease: a review of the pathophysiological mechanisms. Transl Stroke Res 2024; 15: 1050–1069.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.58 MB