Lacunar strokes (LS), caused by the occlusion of individual penetrating arteries, are small subcortical infarcts in the deep hemisphere or in the brainstem.¹ Although LS accounts for one-quarter of all ischemic strokes and generally has a favorable outcome, 20-30% of patients with LS show neurological deterioration (ND), especially motor deficits, within a few days after stroke onset.^2-4 ND in LS has been shown to be associated with cerebral edema, low collateral blood flow, thrombus propagation, or recurrent embolism.^2,5 Because ND in LS leads to a high rate of functional disability, early identification of patients who are at risk of progression may improve their clinical and therapeutic management.

Blood viscosity (BV) is defined as the measurement of intrinsic resistance applied to blood flow, and describes the blood stickiness and thickness.⁶ BV is related to elevated concentrations of blood cells and plasma components, and high aggregability and low deformability of erythrocytes.^6,7 Prior studies have demonstrated that high BV increases the major thromboembolic risk, and is associated with cerebro-cardiovascular diseases.^6-8 Regarding stroke, several studies have suggested that BV appears significantly elevated more commonly in LS among stroke subtypes.^9,10 It has been postulated that when the blood passes the cerebral small arteries, increased BV can aggravate the flow disturbance and cause luminal occlusion.^9,11

Considering the underlying pathomechanisms of ND in LS, we hypothesized that increased BV could influence cerebral vessel viscosity, which is related to thrombus propagation and collateral blood flow circulation. Hence, this study aimed to determine the relationship between ND and BV in patients with LS.

In total, 215 patients (26% of the total stroke population of the institute during the study period) who had experienced LS were screened for enrolment, of whom 33 (15%) were excluded from the study (12 due to admission after 5 days of stroke onset, nine due to Hb level <10 or >18 mg/dL, eight with IV thrombolysis, two with recurrent stroke during admission, and two with prior anticoagulant use). Consequently, a total of 182 patients were included in the final analysis.

The baseline characteristics of the enrolled patients are shown in Table 1. Of these, 40 patients (22%) showed clinical progression of stroke symptoms, and 78% remained stable or improved. For males, the study group consisted of 116 patients (22% with progression of stroke symptoms), and the mean age was 68.20±12.01 years. Of these, 69% had a history of hypertension, 36% had a history of diabetes, 49% had a history of dyslipidemia, and 42% were current smokers. Thirty-four patients (29%) regularly used antiplatelets (aspirin alone, 47%; clopidogrel alone, 29%; aspirin plus clopidogrel, 12%; and others, 12%) at admission. The median time from symptom onset to hospital arrival was 16 hours, and 67% of patients visited the hospital within 24 hours. The median NIHSS score at admission was 3, and 32% of patients presented with NIHSS ≥4. All baseline characteristics were well balanced across the two groups except time to admission (p=0.002), onset to admission ≤24 hours (p=0.043), and NIHSS score ≥4 at admission (p=0.015), suggesting that the progressive stroke group visited the hospital earlier with more severe symptoms than the non-progressive stroke group. For females, 66 patients were enrolled in the study (23% with progression of stroke symptoms), and the mean age was 72.70±11.35 years. There were no differences in the baseline characteristics between the two groups, except for time to admission (p=0.006), onset to admission ≤24 hours (p<0.0001), and lesion localization (p=0.009). The progressive stroke group visited the hospital earlier and had more anterior circulation lesions than that in the nonprogressive stroke group.

Table 2 shows the laboratory findings of the study population. In men, no significant differences were observed in laboratory findings between the groups, except for SBV and DBV. The SBV and DBV were higher in the progressive stroke group (p=0.007 and p=0.006, respectively), showing that the progressive stroke group had increased BV at admission compared to the non-progressive stroke group. Multivariable logistic regression analysis revealed that onset to admission ≤24 hours (odds ratio [OR], 4.03; 95% CI, 1.10-14.71; p=0.035), NIHSS ≥4 at admission (OR, 2.63; 95% CI, 1.03-6.71; p=0.043), SBV (OR, 3.57; 95% CI, 1.39-9.16; p=0.008), and DBV (OR, 1.09; 95% CI, 1.02-1.16; p=0.012) were significantly associated with progressive stroke. In men, the cutoff value of the SBV and DBV in progressive stroke were 4.77 cP (sensitivity, 70.8%; specificity, 31.7%) and 32.35 cP (sensitivity, 66.7%; specificity, 65.9%), respectively. The AUC was 0.68 (95% CI, 0.56-0.80; p=0.007) and 0.67 (95% CI, 0.55-0.80; p=0.011) (Fig. 1). Analysis of the female groups showed that Hb, Hct, SBV, and DBV were significantly higher in the progressive stroke group. Multivariable logistic regression analysis revealed that onset to admission ≤24 hours (OR, 17.92; 95% CI, 2.16-148.62; p=0.008) was related to progressive stroke, suggesting that most ND occurred within days after stroke onset. There was also a trend towards increased BV being associated with progressive stroke (SBV OR, 3.25; 95% CI, 0.97-10.97; p=0.057; and DBV, OR, 1.07; 95% CI, 0.99-1.18; p=0.054).

In this study, we evaluated the relationship between ND and BV in LS patients. Our study showed that admission within 24 hours, higher NIHSS score, and higher BV at admission were associated with an increased risk of ND in male patients with LS. In women, the onset to admission within 24 hours was associated with progressive stroke, confirming that most ND occurred within days after stroke onset. These data were consistent with previously reported data demonstrating that in 20-30% of LS patients, neurological deficits worsen in hours or days after LS onset.^5,15 In our study, the median time from symptom onset to hospital arrival was 11 hours in the progressive stroke group and 21 hours in the nonprogressive stroke group. Regarding NIHSS score, recent studies have suggested that a higher NIHSS score at admission is associated with an increased risk of ND.^13,16 Eighteen patients (45%) with progressive stroke presented with NIHSS ≥4 in our study. From a therapeutic point of view, patients with NIHSS ≥4 at admission were excluded from the POINT trial.¹⁷ However, one study demonstrated that 5 days of DAPT are associated with improved functional outcomes in progressive LS.¹⁴ Although DAPT efficacy and the mechanisms underlying ND are not yet fully understood, DAPT might improve functional outcomes and reduce clinical fluctuation by reducing thrombus propagation or arterial reocclusion, which is closely related to the mechanism of progressive LS.

There is still no generally accepted definition of ND for patients with LS. Several studies have used diverse neurological scales and definitions when classifying patients into a progressive stroke group.^5,13-15 One study reported that worsening more than 1 point in NIHSS score is reasonable since the major ND occurred in motor function.⁵ Considering the criteria for NIHSS score and clinical fluctuations in LS, we defined progressive stroke as worsening by ≥2 points on the NIHSS score for motor function or ≥3 points in all NIHSS scores.^13,14 We believe that an advanced evidence-based definition of ND in LS is essential to perform good clinical trials and provide the best treatments.

Failure of collateral circulation, thrombus propagation, arterial reocclusion, hemorrhagic transformation, stroke recurrence, seizure, brain edema, neuronal excitotoxicity, inflammation, and peri-infarct depolarizations may be associated with ND in LS.^15,18,19 The most plausible explanation tis that local thrombosis superimposes to the ostial atheroma, and thrombus propagation proximal to distal, or distal to proximal segments of a perforating artery.^15,19 Hemodynamic insufficiency of perforators near the one responsible for the ischemic insult could also contribute to transformation of the penumbra into infarction.¹⁵ It is well known that increased BV reflects an acute phase response to stroke and presents an independent stroke risk factor.^20,21 There are several potential explanations for the relationship between BV and progressive LS.^9,20 Firstly, a reduction in the collateral circulation could be one of the potential mechanisms that lead to progressive LS. Further, cerebral perfusion is inversely related to BV and thus,²⁰ an increased BV can lead to decreased cerebral perfusion in patients with progressive LS. BV is elevated in the acute phase of ischemic stroke and gradually improves in the chronic phass,⁷ and appears to be significantly higher in LS than in other stroke subtypes.^9,11 When the blood passes through the stenotic cerebral perforating arteries, increased BV can aggravate the flow disturbance and cause vascular endothelial remodeling and luminal occlusion. These hemorheological changes can be related to thrombus propagation or arterial reocclusion in progressive LS.²² Taken together, it can be assumed that increased BV may play an important role in the mechanism of progressive LS.

Our study has several limitations. First, the number of patients with progressive LS may have been underestimated. The frequency of progressive LS is described to be dependent on the time of admission.⁵ The ND in LS usually occurred within the first days following stroke onset, but only 63% of patients visited the hospital within 24 hours after stroke onset in our study. Second, BV is elevated in the acute phase of ischemic stroke and gradually decreases over time.⁷ The earlier hospital visit in the progressive stroke group and the time differences of measuring BV between the groups may explain the discrepancy in our results. However, Hct, the major determinant of BV, was similar between the groups in men, suggesting that these results were not related to the BV measurement time. Third, several clinical and neuroimaging data have been studied as potential predictors of LS progression. Lesion location (internal capsule, pons, and corona radiata), motor function involvement, and the presence of a few risk factors linked to atherosclerosis are the most consistently reported. Regarding imaging risk factors, previous brain MRI studies have shown that corona radiata or pontine lesions are associated with ND in patients with LS. Although a trend towards concordance with previous studies was observed (p=0.113), our study failed to demonstrate significant imaging data. Finally, the statistical power of the results supporting our conclusion was fairly weak, mainly because of the small sample size and lack of age-matched controls. These limitations should be considered when interpreting our results, and the present findings need to be confirmed in larger trials.

In conclusion, this study showed that admission within 24 hours, higher NIHSS score, and higher BV at admission were all associated with an increased risk of ND in male patients with LS. Increased BV may play an important role in the reduction of cerebral collateral circulation, thrombus propagation or arterial reocclusion in progressive LS. Further studies focusing on a larger sample size with acute onset time are essential to confirm our results.