Association of blood viscosity with first-pass reperfusion in mechanical thrombectomy for acute ischemic stroke

Abstract

BACKGROUND:

Elevated blood viscosity has been reported as a risk factor for cerebrovascular disease.

OBJECTIVE:

The relationship between blood viscosity and outcomes of mechanical thrombectomy (MT) for large artery occlusion (LAO) were investigated in the present study.

METHODS:

A total of 238 patients were enrolled and systolic blood viscosity (SBV) and diastolic blood viscosity (DBV) were measured using the scanning capillary tube viscometer. Receiver operating characteristic (ROC) analysis was performed to specify the association of viscosity with the first-pass reperfusion (FPR). Multivariable and regression analyses were performed to evaluate the relationship of viscosity with FPR and various variables.

RESULTS:

Based on ROC analysis, the best DBV cutoff value was 10.55 (cP). In multivariable analysis, high DBV was associated with FPR failure (odds ratio 2.82, 95% confidence interval 1.64–4.22; p = 0.001). Increased DVB could be associated with elevated SBV, hematocrit level, and blood urea nitrogen/creatinine ratio (p = <0.001, 0.004, and 0.002, respectively).

CONCLUSIONS:

Elevated DBV was associated with FPR failure. Patients with high DBV had longer thrombus length and required more stent passages than patients with low DBV.

Keywords

Blood viscosity dehydration ischemic stroke reperfusion risk factors thrombectomy

1 Introduction

Blood viscosity plays a major role in atherosclerosis and has been demonstrated as a risk factor or predictor of cardiovascular and neurovascular disease [1 –10]. Blood viscosity is determined by various hemorheological components, such as fibrinogen, white blood cells, red blood cells, platelets, and may be affected by plasma concentration, erythrocyte permeability and aggregation, and dehydration [6 , 11–14]. Elevated blood viscosity affects the deterioration of tissue perfusion with an increase of shear stress and inflammatory reaction to blood vessels [3]. Shear rate is an important factor regulating blood viscosity as a non-Newtonian fluid, and blood viscosity measured at a high shear rate ≥300 s^–1 is defined as systolic blood viscosity (SBV) [3 , 15–17]. In addition, the role of diastolic blood viscosity (DBV), which is the blood viscosity measured at a low shear rate ≤5 s^–1, is emphasized more than SBV as a risk factor for cardiovascular disease [18 –21].

In previous studies, the association of blood viscosity with stroke was reported [4–7 , 22–26]. Similar to coronary artery disease, elevated DBV may play an important role in the development of neurovascular disease as well as acute and chronic lacunar manifestations of cerebral small vessel disease [24]. Blood viscosity level in patients with small artery occlusion was higher than in other stroke subtypes with large artery atherosclerosis or cardioembolism [6, 7]. Flow resistance may be proportional to blood viscosity and vessel length and inversely proportional to vessel diameter, thus, flow resistance is markedly increased in narrow vessels such as stenotic lesions or small perforating arteries [6, 27]. This is only a hypothesis and whether blood viscosity affects large arteries other than small vessels has not been proven. In addition, there are few papers analyzed the association of blood viscosity and results of mechanical thrombectomy (MT) which is a standard treatment for large vessel occlusion (LAO) with acute ischemic stroke (AIS) [28].

Based on previous studies, the authors hypothesized that blood viscosity may affect the results of MT for LAO. Thus, in the present study, the association of blood viscosity with outcomes of MT for LAO was analyzed and factors that affect the blood viscosity were identified.

2 Materials and methods

2.1 Patients and mechanical thrombectomy

This study was approved by the local Institutional Review Board (IRB No. VC17RESI0048) of each participating center. Data were retrospectively reviewed from each institution’s database of consecutive patients who underwent MT for LAO between January 2016 and December 2019, and a total of 238 patients were enrolled. In all patients, AIS and LAO were confirmed using diffusion-weighted magnetic resonance imaging and computed tomography angiography, respectively. Before the MT, intravenous thrombolysis (IVT) with tissue plasminogen activator (alteplase) was administered within 4.5 hours after stroke onset at a maximum dose of 0.9 mg/kg in accordance with the European Cooperative Acute Stroke Study (ECASS) III trial [29, 30]. MT with stent retriever or combined technique were performed in patients with LAO, including occlusion of the intracranial carotid artery, middle cerebral artery, anterior cerebral artery, or posterior circulation (vertebral artery or basilar artery) [31].

2.2 Blood viscosity measurements

The scanning capillary tube viscometer (Hemovister, Pharmode Inc., Seoul, Korea) was used to measure whole blood viscosity including SBV and DBV. Whole blood viscosity was analyzed at high (≥300 s^–1) and low (≤5 s^–1) shear rates for SBV and DBV, respectively [15, 24]. Whole blood (3 mL) was collected in an ethylenediaminetetraacetic acid (EDTA) tube separately by peripheral venous puncture within 3 hours after admission and stored in a refrigerator at 4°C until measurement. All venous samples were obtained during the initial blood sampling procedure in the emergency room before any medical or interventional treatment was given to reduce the diluting effect of blood viscosity due to hydration [6, 24], and all measurements were performed within 12 hours of blood sampling.

2.3 Baseline assessments and laboratory findings

Multivariable factors were reviewed, including age, sex, risk factors, stroke etiology based on TOAST criteria [32], arterial occlusion site, application of IVT, Alberta Stroke Program Early CT Score (ASPECTS), and time intervals affecting viscosity, such as symptom onset to groin puncture time and procedure time (time from groin puncture to reperfusion) [33]. Risk factors included history of hypertension, diabetes mellitus, atrial fibrillation, coronary artery disease, prior stroke or transient ischemic attack (TIA), current smoking status, dyslipidemia, and body mass index (BMI: weight divided by height) ≥25 kg/m². Laboratory findings which could affect blood viscosity and prognosis of patients were included, such as white blood cells, red blood cells, hemoglobin, hematocrit, platelets, total protein, blood urea nitrogen (BUN), creatinine (Cr), BUN/Cr ratio, fibrinogen, D-dimer, fibrin degradation products, high-sensitivity C-reactive protein, and lipoproteins [23 , 34–39].

2.4 Clinical and radiological outcomes

All patients underwent a clinical investigation including National Institutes of Health Stroke Scale (NIHSS; range, 0–42, with higher score indicating more severe neurological deficit), the modified Rankin Scale (mRS) score (favorable clinical outcome was defined as mRS score of 0–2 at 3 months), symptomatic intracerebral hemorrhage (any hemorrhage with an increase of ≥4 points on the total NIHSS score after MT) [29], and mortality at 3 months. Regarding radiological outcomes, a successful recanalization was defined as a Thrombolysis in Cerebral Infarction (TICI) grade of 2b or 3, the first-pass reperfusion (FPR) indicates achievement of modified TICI (mTICI 2b or 3) with the first pass of thrombectomy devices [40, 41], and total number of stent passages for successful recanalization were analyzed. The length of the thrombus was measured in the patient’s angiography images and was defined as the length of the lesion artery not contrasted on angiography.

2.5 Statistical analyses

All statistical analyses were conducted using statistical software package SPSS 25.0 for Windows (IBM, Armonk, NY, USA). Receiver operating characteristic (ROC) analysis was performed to specify the associations of viscosities (systolic and diastolic) with the FPR and determine the best cutoff values for FPR prediction. All patients were dichotomized based on the first-pass effect (FPR vs. non-FPR) and best cutoff value of diastolic viscosity. Differences between FPR and non-FPR were analyzed using Student’s t-test or Mann-Whitney U test for continuous variables and the χ2 test or Fisher’s exact test for categorical variables. In addition, multivariable analysis with binary logistic regression was performed to determine independent variables for non-FPR and p-values < 0.20 in univariable analysis were entered into a backward multivariable analysis. Multiple linear regression analyses were performed to evaluate the independent association of DBV with other variables. Two-tailed p-values≤0.05 were considered to indicate statistically significant difference.

3 Results

3.1 Association of systolic and diastolic blood viscosity with first-pass reperfusion

In ROC analysis, the best cutoff threshold of SBV was 3.65 (cP) to discriminate between FPR and non-FPR (area under the curve, AUC 0.772, 95% confidence interval, CI 0.710–0.833; p < 0.001, FPR: SBV < 3.65, 49/107 (45.8%) vs. SBV≥3.65, 44/131 (33.6%); p = 0.002). In the same category, 10.55 (cP) was the best cutoff value for FPR (AUC 0.792, 95% CI 0.730–0.853; p < 0.001, FPR: DBV < 10.55, 52/104 (50.0%) vs. DBV≥10.55, 41/134 (30.6%); P < 0.001) (Fig. 1).

Fig. 1

Receiver operating characteristic (ROC) analyses of the association between systolic blood viscosity (SBV) and diastolic blood viscosity (DBV) with first-pass effect were performed to discriminate between first-pass reperfusion (FPR) and non-FPR. The best cutoff values were identified at SBV of 3.65 (cP) and DBV of 10.55 (cP), respectively; area under the curve, (AUC) 0.772, 95% confidence interval, CI 0.710–0.833; p < 0.001, and AUC 0.792, 95% CI 0.730–0.853, p < 0.001, respectively.

3.2 Baseline characteristics, laboratory findings, and outcomes

Among 238 patients, 93 had FPR (39.1%, mean age 69.6 years, 54.8% male) and the other 145 patients had non-FPR (60.9%, mean age 68.0 years, 56.6% male). Mean SBV and DBV values were significantly higher in non-FPR patients than in FPR patients (non-FPR:FPR, SBV = 4.00 (standard deviation, SD 0.71):3.43 (SD 0.47), DBV = 12.59 (SD 4.43):9.50 (SD 2.21), p = < 0.001 and <0.001, respectively; Fig. 2A and B). However, significant differences in other laboratory findings, risk factors, stroke etiology, and pre-interventional details were not observed between FPR and non-FRP patients. Median procedure time was shorter in FPR patients (25 min, interquartile range, IQR 13–37) than in non-FPR patients (57 min, IQR 29–85) because more stent retrievals were attempted in non-FPR patients (P < 0.001). Non-FPR patients (24.8 mm, SD 9.8) showed significantly longer mean thrombus length than FPR patients (19.3 mm, SD 7.7; p < 0.001; Fig. 2C). The prevalence of successful recanalization and favorable clinical outcome were higher in FPR patients (100% and 68.8%) than in non-FPR patients (93.1% and 53.1%; p = 0.002 and 0.004, respectively; Table 1). Subjects were classified into high (≥10.55 cP) or low (<10.55 cP) DBV patients based on the best cutoff DBV value in ROC analysis; 134 (56.3%) patients had high DBV and 104 patients (43.7%) had low DBV. The mean hemoglobin and hematocrit levels were significantly elevated in patients with high DBV (13.81 g/dL, SD 2.19, and 40.57%, SD 6.00) compared with patients with low DBV (13.19 g/dL, SD 1.96, and 38.91%, SD 5.89; p = 0.023 and 0.033, respectively). Patients with high DBV showed significantly longer median procedure time (51 min, IQR 17–80) and mean thrombus length (25.7 mm, SD 9.4) than patients with low DBV (36 min, IQR 16–61; 18.7 mm, SD 7.9; p = < 0.001 and < 0.001, respectively). Patients with high DBV had higher median number of stent passages (3, IQR 1–5) and lower FPR rate (30.6%) compared with patients with low DBV (1, IQR 1–3 and 50.0%; p = < 0.001 and <0.001, respectively) (Table 2).

Fig. 2

Box plots of systolic blood viscosity (SBV), diastolic blood viscosity (DBA), and thrombus length between first-pass reperfusion (FPR, gray square) and non-FPR (white square). (A) The non-FPR patients had a higher mean SBV level than FPR patients (p < 0.001). (B) The box plots of the non-FPR patients show a significantly elevated DBV compared with FPR patients (p < 0.001). (C) The mean thrombus length was significantly longer in non-FPR patients than in FPR patients (p < 0.001).

Table 1

Comparisons of baseline characteristics and outcomes in patients with mechanical thrombectomy, according to first-pass reperfusion

Variables	FPR	Non-FPR	p value
	(n = 93, 39.1%)	(n = 145, 60.9%)
Blood viscosity, mean (SD)
Systolic blood viscosity, cP	3.43 (0.47)	4.00 (0.71)	<0.001^*
Diastolic blood viscosity, cP	9.50 (2.21)	12.59 (4.43)	<0.001^*
Risk factors
Age, mean (SD)	69.6 (13.5)	68.0 (13.9)	0.382
Men, n (%)	51 (54.8)	82 (56.6)	0.894
Hypertension, n (%)	52 (55.9)	70 (48.3)	0.288
Diabetic mellitus, n (%)	24 (25.8)	33 (22.8)	0.642
Atrial fibrillation, n (%)	31 (33.3)	44 (30.3)	0.669
Coronary artery disease, n (%)	21 (22.6)	29 (20.0)	0.745
Prior stroke or TIA, n (%)	9 (9.7)	19 (13.1)	0.537
Smoking, n (%)	18 (19.4)	29 (20.0)	0.998
Dyslipidemia, n (%)	30 (32.3)	52 (35.9)	0.580
Body mass index ≥25 kg/m², n (%)	21 (22.6)	40 (27.6)	0.448
Laboratory findings, mean (SD)
White blood cells, ×10⁹/L	9.20 (3.19)	9.78 (3.41)	0.189
Red blood cells, ×10¹²/L	4.32 (0.60)	4.38 (0.94)	0.580
Hemoglobin, g/dL	13.57 (1.89)	13.51 (2.25)	0.840
Hematocrit, %	40.03 (5.11)	39.77 (6.52)	0.700
Platelets, ×10⁹/L	216.1 (60.4)	229.1 (99.6)	0.260
Total protein, g/dL	6.73 (0.77)	6.72 (0.73)	0.909
Blood urea nitrogen (BUN), mg/dL	17.33 (7.85)	16.68 (6.48)	0.302
Creatinine (Cr), mg/dL	0.92 (0.89)	0.82 (0.40)	0.268
BUN/Cr	21.06 (8.04)	19.94 (6.78)	0.249
Fibrinogen, mg/dL	298.4 (113.1)	276.9 (102.3)	0.224
D-dimer, mg/L	4.00 (3.86)	4.01 (3.75)	0.992
Fibrin degradation products, μg/mL	12.07 (10.72)	12.38 (11.42)	0.906
High-sensitivity C-reactive protein, mg/L	0.76 (0.55)	0.80 (0.52)	0.207
Total cholesterol, mg/dL	176.3 (47.8)	173.6 (45.8)	0.661
High-density lipoprotein cholesterol, mg/dL	50.9 (13.8)	49.1 (13.3)	0.291
Low-density lipoprotein cholesterol, mg/dL	107.3 (35.2)	104.5 (33.1)	0.537
Triglyceride, mg/dL	124.8 (77.5)	129.9 (122.0)	0.722
Stroke etiology, n (%)
Cardio-embolic	40 (43.0)	58 (40.0)	0.709
Atherosclerosis	27 (29.0)	40 (27.6)	0.821
Dissection	0 (0.0)	2 (1.4)	0.218
Other or Undetermined	26 (30.0)	45 (31.0)	0.889
Arterial occlusion site, n (%)
Middle cerebral artery	55 (59.1)	81 (55.9)	0.703
Distal internal carotid artery	14 (15.1)	23 (15.9)	0.832
Proximal internal carotid artery	13 (14.0)	23 (15.9)	0.508
Posterior circulation	9 (9.7)	15 (10.2)	0.661
Anterior cerebral artery	2 (2.1)	3 (2.1)	0.913
Pre-interventional details
Left hemisphere stroke, n (%)	57 (61.3)	81 (55.9)	0.423
Intravenous thrombolysis, n (%)	28 (30.1)	54 (37.2)	0.268
ASPECTS, median (IQR)	9 (8–10)	9 (8–10)	0.998
Initial NIHSS, median (IQR)	10 (3–17)	9 (2–18)	0.640
Symptom to puncture time (min), median (IQR)	222 (82–376)	218 (76–382)	0.472
Procedure detail and outcomes
Procedure time (min), median (IQR)	25 (13–37)	57 (29–85)	<0.001^*
Thrombus length (mm), mean (SD)	19.3 (7.7)	24.8 (9.8)	<0.001^*
Successful recanalization, n (%)	93 (100.0)	135 (93.1)	0.002^*
Favorable clinical outcome, n (%)	64 (68.8)	77 (53.1)	0.004^*
Symptomatic intracerebral hemorrhage, n (%)	4 (4.3)	15 (10.3)	0.140
Mortality at 3months, n (%)	6 (6.5)	16 (11.0)	0.382

^*p≤0.05. Abbreviations: FPR, First pass reperfusion; SD, Standard deviation; TIA, Transient ischemic attack; ASPECTS, Alberta Stroke Program Early CT Score; IQR, interquartile range; NIHSS, National Institutes of Health Stroke Scale; Successful recanalization, TICI 2b or 3; TICI, Thrombolysis in Cerebral Infarction scores; Favorable clinical outcome, modified Rankin Scale (mRS) score of 0–2 at 3 months.

Table 2

Comparisons of demographics in patients with mechanical thrombectomy, according to level of diastolic blood viscosity (DBV)

Variables	Low DBV	High DBV	p value
	(<10.55)	(≥10.55)
Number of patients, %	104 (43.7)	134 (56.3)
Risk factors
Age, mean (SD)	68.6 (13.4)	69.9 (13.4)	0.158
Men, n (%)	56 (53.8)	77 (57.5)	0.228
Hypertension, n (%)	53 (51.0)	69 (51.5)	0.935
Diabetic mellitus, n (%)	25 (24.0)	32 (23.9)	0.977
Atrial fibrillation, n (%)	33 (31.7)	42 (31.3)	0.949
Coronary artery disease, n (%)	23 (22.1)	27 (20.1)	0.210
Prior stroke or TIA, n (%)	8 (7.7)	20 (14.9)	0.105
Smoking, n (%)	16 (15.4)	31 (23.1)	0.144
Dyslipidemia, n (%)	34 (32.7)	48 (35.8)	0.681
Body mass index ≥25 kg/m², n (%)	25 (24.0)	36 (26.9)	0.656
Laboratory findings, mean (SD)
Systolic blood viscosity, cP	3.26 (0.28)	4.18 (0.65)
White blood cells, ×10⁹/L	9.89 (3.51)	9.16 (3.04)	0.227
Red blood cells, ×10¹²/L	4.31 (0.84)	4.39 (0.82)	0.429
Hemoglobin, g/dL	13.19 (1.96)	13.81 (2.19)	0.023^*
Hematocrit, %	38.91 (5.89)	40.57 (6.00)	0.033^*
Platelets, ×10⁹/L	222.9 (66.2)	224.9 (99.6)	0.857
Total protein, g/dL	6.72 (0.80)	6.73 (0.71)	0.922
Blood urea nitrogen (BUN), mg/dL	16.70 (7.07)	16.03 (7.09)	0.471
Creatinine (Cr), mg/dL	0.93 (0.92)	0.81 (0.24)	0.152
BUN/Cr	20.49 (7.74)	20.29 (6.97)	0.832
Fibrinogen, mg/dL	287.2 (117.5)	282.1 (99.1)	0.769
D-dimer, mg/L	4.20 (6.03)	3.91 (6.46)	0.789
Fibrin degradation products, μg/mL	12.48 (13.07)	12.16 (14.21)	0.899
High-sensitivity C-reactive protein	0.75 (0.49)	0.82 (0.61)	0.194
Total cholesterol, mg/dL	174.7 (43.6)	174.6 (48.8)	0.976
High-density lipoprotein cholesterol, mg/dL	50.2 (13.1)	49.4 (13.8)	0.671
Low-density lipoprotein cholesterol, mg/dL	107.7 (34.3)	104.7 (33.7)	0.644
Triglyceride, mg/dL	120.4 (72.4)	133.7 (127.1)	0.340
Procedure details
Symptom to puncture time (min), median (IQR)	220 (78–378)	218 (80–376)	0.887
Procedure time (min), median (IQR)	36 (16–62)	51 (17–80)	<0.001^*
Thrombus length (mm), mean (SD)	18.7 (7.9)	25.7 (9.4)	<0.001^*
Number of stent passage, median (IQR)	1 (1–3)	3 (1–5)	<0.001^*
Successful recanalization, n (%)	100 (96.2)	128 (95.5)	0.893
First pass reperfusion (FPR), n (%)	52 (50.0)	41 (30.6)	<0.001^*

^*p≤0.05. Abbreviations: SD, Standard deviation; TIA, Transient ischemic attack; IQR, interquartile range; Successful recanalization, TICI 2b or 3; TICI, Thrombolysis in Cerebral Infarction scores.

3.3 Multivariable and multiple regression analysis

To identify parameters independently associated with FPR failure, univariable and multivariable binary logistic regression analyses were performed. In univariable analysis, longer thrombus length (≥20 mm), high SBV (≥3.65 cP), and high DBV (≥10.55 cP) were associated with non-FPR. In multivariate analysis, only high DBV (≥10.55 cP) was independently associated with FPR failure (odds ratio, OR 2.82, 95% CI 1.64–4.22; p = 0.001) (Table 3). In simple correlation analysis, DBV was significantly associated with SBV, hematocrit level, and BUN/Cr ratio. Similarly, SBV, hematocrit level, and BUN/Cr ratio had a significant predictive power for increased DBV in multiple regression analysis (p = < 0.001, 0.004, and 0.002, respectively) (Table 4).

Table 3
Multivariable analysis of parameters associated with failure of first-pass reperfusion (FPR)

Variable Univariable analysis Multivariable analysis

OR (95% CI) p value OR (95% CI) p value

Age (>70 yr) 1.45 (0.78–3.25) 0.169 1.40 (0.86–4.02) 0.201

Sex 1.67 (0.81–3.49) 0.249

Hypertension 1.72 (0.82–3.59) 0.153 1.52 (0.78–3.14) 0.222

Diabetic mellitus 1.31 (0.56–3.07) 0.534

Atrial fibrillation 1.20 (0.58–2.49) 0.631

Coronary artery disease 1.62 (0.88–3.52) 0.172 1.40 (0.76–2.88) 0.238

Prior stroke or TIA 1.17 (0.41–3.33) 0.775

Smoking 1.10 (0.47–2.60) 0.482

Dyslipidemia 0.84 (0.42–1.71) 0.644

Body mass index ≥25 kg/m² 1.69 (0.64–2.46) 0.336

Left hemisphere stroke 0.78 (0.32–1.82) 0.834

Intravenous thrombolysis 0.82 (0.58–1.22) 0.092 0.90 (0.72–1.08) 0.102

Thrombus length (≥20 mm) 1.50 (1.14–3.01) 0.027^* 1.32 (0.88–2.92) 0.088

High SBV (≥3.65 cP) 2.08 (1.28–4.50) 0.019^* 1.68 (0.92–3.02) 0.062

High DBV (≥10.55 cP) 3.42 (1.48–6.62) <0.001^* 2.82 (1.64–4.22) 0.001^*

Variable	Univariable analysis	Multivariable analysis
Age (>70 yr)	1.45 (0.78–3.25)	0.169	1.40 (0.86–4.02)	0.201
Sex	1.67 (0.81–3.49)	0.249
Hypertension	1.72 (0.82–3.59)	0.153	1.52 (0.78–3.14)	0.222
Diabetic mellitus	1.31 (0.56–3.07)	0.534
Atrial fibrillation	1.20 (0.58–2.49)	0.631
Coronary artery disease	1.62 (0.88–3.52)	0.172	1.40 (0.76–2.88)	0.238
Prior stroke or TIA	1.17 (0.41–3.33)	0.775
Smoking	1.10 (0.47–2.60)	0.482
Dyslipidemia	0.84 (0.42–1.71)	0.644
Body mass index ≥25 kg/m²	1.69 (0.64–2.46)	0.336
Left hemisphere stroke	0.78 (0.32–1.82)	0.834
Intravenous thrombolysis	0.82 (0.58–1.22)	0.092	0.90 (0.72–1.08)	0.102
Thrombus length (≥20 mm)	1.50 (1.14–3.01)	0.027^*	1.32 (0.88–2.92)	0.088
High SBV (≥3.65 cP)	2.08 (1.28–4.50)	0.019^*	1.68 (0.92–3.02)	0.062
High DBV (≥10.55 cP)	3.42 (1.48–6.62)	<0.001^*	2.82 (1.64–4.22)	0.001^*

^*p≤0.05. Abbreviations: OR, Odds ratio; CI, Confidence interval; TIA, Transient ischemic attack; SBV, Systolic blood viscosity; DBV, Diastolic blood viscosity.

Table 4

Multiple regression analysis for correlation between diastolic blood viscosity and other variables

Variables	Simple correlation		Multiple regression		Ad. R²
	R	p value	β	p value
Age, yr	0.078	0.234	0.051	0.350
Body mass index, kg/m²	0.035	0.530	0.028	0.521
Systolic blood viscosity, cP	0.726	<0.001^*	0.769	<0.001^*	0.824
Hemoglobin, g/dL	0.075	0.776	0.071	0.785
Hematocrit, %	0.426	<0.001^*	0.441	<0.001^*	0.648
Platelets, ×10⁹/L	0.061	0.297	0.067	0.215
Total protein, g/dL	–0.023	0.700	–0.021	0.712
Blood urea nitrogen/Creatinine	0.502	<0.001^*	0.525	<0.001^*	0.655
Fibrinogen, mg/dL	0.024	0.688	0.020	0.737
D-dimer, mg/L	–0.195	0.341	–0.186	0.336
Fibrin degradation products, μg/mL	0.261	0.210	0.048	0.368
hsCRP, mg/L	0.347	0.227	0.208	0.281
Total cholesterol, mg/dL	0.039	0.866	0.049	0.766
HDL, mg/dL	–0.053	0.502	–1.361	0.176
LDL, mg/dL	0.068	0.741	0.044	0.425
Triglyceride, mg/dL	0.030	0.699	0.019	0.731

^*p≤0.05. Abbreviations: Ad. R², Adjusted R Square; β, Standardized coefficient beta; hs CRP, High-sensitivity C-reactive protein; HDL, High-density lipoprotein cholesterol; LDL, High-density lipoprotein cholesterol.

4 Discussion

In the present study, a strong association between elevated DBV and FPR failure was observed in the cohort of patients who underwent MT for LAO. High DBV (≥10.55 cP) values contributed to increased non-FPR. Furthermore, non-FPR patients had higher SBV and DBV levels than FPR patients. Patients with high DBV had a higher median number of stent passages and lower FPR rate than patients with low DBV. The co-related factors for the elevated DBV were increased SBV, hematocrit level, and BUN/Cr ratio.

The association of blood viscosity with the occurrence of stroke was reported in previous studies that included healthy subjects or patients at high risk of ischemic stroke. Song et al. reported that DBV is correlated with lacunar infarct (adjusted OR 1.01, CI 1.00–1.02, p = 0.033) and baseline DBV was significantly higher in patients with small artery occlusion (274.7 mP = 27.47 cP) than in patients with LAO (214.5 mP) or cardio-embolic stroke (252.0 mP; p = 0.037). [24] Grotemeyer et al. showed subjects with small vessel disease had significantly higher mean plasma viscosity values (1.29 mPas) than patients with large vessel disease or cardio-embolic stroke (1.22 mPas; p < 0.001). [7] Li et al. suggested that silent infarct patients had higher DBV (10.34 mPas) and SBV (4.46 mPas) values than patients without silent infarct (8.98 and 4.29 mPas), and increased DBV was associated with risk of silent cerebral infarct (OR 2.025, CI 1.750–2.343, p < 0.001). [42] In the present study, only patients with MT for LAO were evaluated and difference in viscosity based on the difference of stroke subtypes was not observed. Despite the differences in study population and classification criteria, higher DBV (≥10.55 cP) in the present study was an independent factor for inhibition of FPR. Failed FPR patients had higher mean DBV (12.59 cP) and SBV (9.50 cP) values than FPR patients. Furthermore, patients with high DBV (≥10.55 cP) had more prevalence of FPR than patients with low DBV (<10.55 cP).

Increased blood viscosity at low or high shear stress (equivalent DBV or SBV, respectively) may affect the various pathogenesis in cerebral vessels. First, erythrocyte aggregation with deformability could increase blood viscosity, and elevated blood viscosity might decreases the cerebral blood flow [43, 44]. In patients with small vessel disease with lacunar infarction, elevated blood viscosity can aggravate the flow disturbance and trigger endothelial remodeling and luminal occlusion, and DBV may be more responsible than SBV in this process [24, 42]. Similar pathogenesis could be occurred in cases with occlusion of large artery; elevated DBV in LAO may induce more flow stasis and blood cell accumulation thereby increasing the thrombus integration. Increased thrombus length by thrombus integration had no relevant effect on recanalization or neurological outcome. However, it is associated with increased number of stent passages and failure of FPR, and result in increase of the procedure time [45 –47]. In the present study, the mean thrombus length in patients with high DBV (25.7 mm) was longer than in patients with low DBV (18.7 mm). In addition, patients with high DBV required more stent passages and longer reperfusion procedure time, resulting in reduced FPR prevalence. The FRR is known as a factor related to good clinical outcome in patients with MT [48]. In contrast, more number of stent passage result in increase of parenchymal hematoma after MT [49]. Therefore, blood viscosity with predictability for FPR can be a useful prognostic factor in patients with MT. Second, inflammatory reactions which are positively correlated with blood viscosity, could induce vascular damage, formation of pathologic erythrocytes, and further contribute to reduced deformability and thrombus formation [50 –52]. However, there were no significant differences in inflammatory markers of our study, such as white blood cell count and high-sensitivity C-reactive protein, based on blood viscosity.

The most important determinants of blood viscosity are fibrinogen and hematocrit levels, which play pivotal roles in aggregation of red blood cells and erythrocyte deformability [17, 24]. In the present study, patients with high DBV had higher hemoglobin and hematocrit levels than patients with low DBV, however, fibrinogen levels were not different. Furthermore, in multiple regression analysis, increased SBV, hematocrit level, and BUN/Cr ratio were associated with elevated DBV. Hematocrit level and BUN/Cr ratio are indicative of dehydration and represented by increased blood viscosity [53, 54]. Therefore, dehydration in stroke patients should be avoided because not only poor outcome but also elevated DBV may occur [54]. In addition, lipoproteins were reported to influence plasma or blood viscosity in previous studies, and viscosity correlated with increased low-density lipoprotein/high-density lipoprotein ratio and triglycerides [35 , 56]. However, there was no significant the association between lipoproteins and blood viscosity in our study, and further analyses are needed.

The present study had several limitations. First, the retrospective non-blinded analysis of a relatively small sample size without randomization is prone to introduction of bias and can allow errors in data interpretation. Second, plasma viscosity, which influences the whole blood viscosity with tissue perfusion, and is a well-documented as a cardiovascular risk factor [43 , 58]. However, our study is lacking in information about plasma viscosity. Third, patients did not fast before venous blood samples were obtained, thus, blood viscosity could be easily affected by hydration therapy or drinking water. Fourth, dynamic changes with repeated measurements of blood viscosity were not included. Other unmeasured and residual confounding data may have contributed to the results. However, every effort was made to adjust for the possibility of false results.

5 Conclusion

In the present study, elevated blood viscosity on hospital admission was associated with FPR failure in patients who underwent MT for LAO. In particular, DBV may play a role in FPR and elevated DBV could be a predicting factor for FPR failure. In addition, longer thrombus length and requirement of increased number of stent passages for reperfusion were identified in patients with high DBV compared with patients with low DBV. Furthermore, increased SBV, hemoglobin level, and BUN/Cr ratio resulted in elevated DBV. Further investigations are required to verify the results of the present study and identify the underlying mechanisms.

Footnotes

Acknowledgments

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. All study protocols received full approval from local ethical committee. This study was approved by institutional review board

References

Malek

, Alper

, Izumo

. Hemodynamic shear stress and its role in atherosclerosis. JAMA. 1999;282(21):2035–42.

Tamariz

, Young

, Pankow

, Yeh

, Schmidt

, Astor

, et al. Blood viscosity and hematocrit as risk factors for type 2 diabetes mellitus: the atherosclerosis risk in communities (ARIC) study. Am J Epidemiol. 2008;168(10):1153–60.

Noh

, Seo

, Jeong

, Park

, Kim

, Noh

, et al. Blood viscosity in subcortical vascular mild cognitive impairment with versus without cerebral amyloid burden. J Stroke Cerebrovasc Dis. 2014;23(5):958–66.

Coull

, Beamer

, de Garmo

, Sexton

, Nordt

, Knox

, et al. Chronic blood hyperviscosity in subjects with acute stroke, transient ischemic attack, and risk factors for stroke. Stroke. 1991;22(2):162–8.

Tsuda

, Satoh

, Kitadai

, Takahashi

. Hemorheologic profiles of plasma fibrinogen and blood viscosity from silent to acute and chronic cerebral infarctions. J Neurol Sci. 1997;147(1):49–54.

Furukawa

, Abumiya

, Sakai

, Hirano

, Osanai

, Shichinohe

, et al. Increased Blood Viscosity in Ischemic Stroke Patients with Small Artery Occlusion Measured by an Electromagnetic Spinning Sphere Viscometer. J Stroke Cerebrovasc Dis. 2016;25(11):2762–9.

Grotemeyer

, Kaiser

, Grotemeyer

, Husstedt

. Association of elevated plasma viscosity with small vessel occlusion in ischemic cerebral disease. Thromb Res. 2014;133(1):96–100.

Lowe

, Rumley

, Norrie

, Ford

, Shepherd

, Cobbe

, et al. Blood rheology, cardiovascular risk factors, and cardiovascular disease: the West of Scotland Coronary Prevention Study. Thromb Haemost. 2000;84(4):553–8.

Sweetnam

, Thomas

, Yarnell

, Beswick

, Baker

, Elwood

. Fibrinogen, viscosity and the 10-year incidence of ischaemic heart disease. Eur Heart J. 1996;17(12):1814–20.

10.

Jung

. From hemorheology to microcirculation and regenerative medicine: Fahraeus Lecture 2009. Clin Hemorheol Microcirc. 2010;45(2-4):79–99.

11.

Linderkamp

, Stadler

, Zilow

. Blood viscosity and optimal hematocrit in preterm and full-term neonates in 50- to 500-micrometer tubes. Pediatr Res. 1992;32(1):97–102.

12.

Eckmann

, Bowers

, Stecker

, Cheung

. Hematocrit, volume expander, temperature, and shear rate effects on blood viscosity. Anesth Analg. 2000;91(3):539–45.

13.

Grotta

, Ostrow

, Fraifeld

, Hartman

, Gary

. Fibrinogen, blood viscosity, and cerebral ischemia. Stroke. 1985;16(2):192–8.

14.

Varlet-Marie

, Brun

, Raynaud de Mauverger

, Fedou

. Exercise-induced changes in hematocrit and hematocrit/viscosity ratio in male rugby players. Clin Hemorheol Microcirc. 2016;64(4):817–26.

15.

Kim

, Cho

, Lee

, Park

, Moon

, Hur

, et al. Analytical performance evaluation of the scanning capillary tube viscometer for measurement of whole blood viscosity. Clin Biochem. 2013;46(1-2):139–42.

16.

Pop

, Duncker

, Gardien

, Vranckx

, Versluis

, Hasan

, et al. The clinical significance of whole blood viscosity in (cardio)vascular medicine. Neth Heart J. 2002;10(12):512–6.

17.

Cho

, Cho

, Rosenson

. Endothelial shear stress and blood viscosity in peripheral arterial disease. Curr Atheroscler Rep. 2014;16(4):404.

18.

Jeong

, Cho

, Duey

, Rosenson

. Cardiovascular risks of anemia correction with erythrocyte stimulating agents: should blood viscosity be monitored for risk assessment? Cardiovasc Drugs Ther. 2010;24(2):151–60.

19.

Cho

, Cho

. Hemorheology and microvascular disorders. Korean Circ J. 2011;41(6):287–95.

20.

Lee

, Durairaj

, Mehra

, Wenby

, Meiselman

, Alexy

. Hemorheological abnormalities in stable angina and acute coronary syndromes. Clin Hemorheol Microcirc. 2008;39(1-4):43–51.

21.

Wasilewski

. Hemorheological abnormalities in stable angina and acute coronary syndromes. Clin Hemorheol Microcirc. 2009;41(1):81–2.

22.

Fisher

, Meiselman

. Hemorheological factors in cerebral ischemia. Stroke. 1991;22(9):1164–9.

23.

Szapary

, Horvath

, Marton

, Alexy

, Demeter

, Szots

, et al. Hemorheological disturbances in patients with chronic cerebrovascular diseases. Clin Hemorheol Microcirc. 2004;31(1):1–9.

24.

Song

, Kim

, Lee

, Yun

, Choi

, Kim

. Elevated blood viscosity is associated with cerebral small vessel disease in patients with acute ischemic stroke. BMC Neurol. 2017;17(1):20.

25.

Schneider

, Ringelstein

, Zeumer

, Kiesewetter

, Jung

. The role of plasma hyperviscosity in subcortical arteriosclerotic encephalopathy (Binswanger’s disease). J Neurol. 1987;234(2):67–73.

26.

Schneider

, Wobker

, Willmes

, Jung

, Kiesewetter

. Do different ischemic brain lesions have different hemorheological profiles? Klin Wochenschr. 1986;64(8):357–61.

27.

Goldstein

, Rypins

. Blood vessel modeling. Int J Biomed Comput. 1991;29(1):23–9.

28.

Goyal

, Menon

, van Zwam

, Dippel

, Mitchell

, Demchuk

, et al. Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet. 2016;387(10029):1723–31.

29.

Hacke

, Kaste

, Bluhmki

, Brozman

, Davalos

, Guidetti

, et al. Thrombolysis with alteplase 3 to 4. 5 hours after acute ischemic stroke. N Engl J Med. 2008;359(13):1317–29.

30.

, Sung

, Lee

. Bridging Intravenous Thrombolysis Before Mechanical Thrombectomy for Large Artery Occlusion May be Detrimental with Thrombus Fragmentation. Curr Neurovasc Res. 2020;17(1):18–26.

31.

Deshaies

. Tri-axial system using the Solitaire-FR and Penumbra Aspiration Microcatheter for acute mechanical thrombectomy. J Clin Neurosci. 2013;20(9):1303–5.

32.

Amarenco

, Bogousslavsky

, Caplan

, Donnan

, Hennerici

. Classification of stroke subtypes. Cerebrovasc Dis. 2009;27(5):493–501.

33.

Kowal

, Marcinkowska-Gapinska

. Hemorheological changes dependent on the time from the onset of ischemic stroke. J Neurol Sci. 2007;258(1-2):132–6.

34.

Aono

, Ohkubo

, Kikuya

, Hara

, Kondo

, Obara

, et al. Plasma fibrinogen, ambulatory blood pressure, and silent cerebrovascular lesions: the Ohasama study. Arterioscler Thromb Vasc Biol. 2007;27(4):963–8.

35.

Irace

, Carallo

, Scavelli

, Esposito

, De Franceschi

, Tripolino

, et al. Influence of blood lipids on plasma and blood viscosity. Clin Hemorheol Microcirc. 2014;57(3):267–74.

36.

Vaya

, Gimenez

, Alis

, Murado

, Fuster

. Influence of lipids on blood and plasma viscosity. Clin Hemorheol Microcirc. 2014;58(4):551–3.

37.

Kim

, Yi

, Lee

, Sung

. Association of High-sensitivity C-reactive Protein with Patient Prognosis Following Mechanical Thrombectomy for Acute Ischemic Stroke. Curr Neurovasc Res. 2020.

38.

Baskurt

, Boynard

, Cokelet

, Connes

, Cooke

, Forconi

, et al. New guidelines for hemorheological laboratory techniques. Clin Hemorheol Microcirc. 2009;42(2):75–97.

39.

Kiesewetter

, Jung

, Wenzel

, Muller

. Variance, factors of influence and clinical relevance of plasma viscosity. Biomed Tech (Berl). 1991;36(10):241–7.

40.

Zaidat

, Yoo

, Khatri

, Tomsick

, von Kummer

, Saver

, et al. Recommendations on angiographic revascularization grading standards for acute ischemic stroke: a consensus statement. Stroke. 2013;44(9):2650–63.

41.

Maus

, Brehm

, Tsogkas

, Henkel

, Psychogios

. Stent retriever placement in embolectomy: the choice of the post-bifurcational trunk influences the first-pass reperfusion result in M1 occlusions. J Neurointerv Surg. 2019;11(3):237–40.

42.

, Cao

, Li

, Wang

. Increased whole blood viscosity is associated with silent cerebral infarction. Clin Hemorheol Microcirc. 2015;59(4):301–7.

43.

Cecchi

, Marcucci

, Poli

, Antonucci

, Abbate

, Gensini

, et al. Hyperviscosity as a possible risk factor for cerebral ischemic complications in atrial fibrillation patients. Am J Cardiol. 2006;97(12):1745–8.

44.

Santos-Galduroz

, Bueno

, Yamaga

, Armani

, Galduroz

. Influence of blood viscosity to cerebral blood flow in older humans compared to young subjects. Clin Neurophysiol. 2012;123(1):117–20.

45.

Shu

, Riedel

, Meyne

, Jansen

, Jensen-Kondering

. Successful recanalization in acute basilar artery occlusion treated with endovascular therapy is independent of thrombus length. J Neurointerv Surg. 2017;9(11):1047–52.

46.

Seker

, Pfaff

, Wolf

, Schonenberger

, Nagel

, Herweh

, et al. Impact of thrombus length on recanalization and clinical outcome following mechanical thrombectomy in acute ischemic stroke. J Neurointerv Surg. 2017;9(10):937–9.

47.

, Lee

, Sung

. Clinical Usefulness of Waiting after Stent Deployment in Mechanical Thrombectomy: Effect of the Clot Integration. World Neurosurg. 2018;119:e87–e93.

48.

Zaidat

, Castonguay

, Linfante

, Gupta

, Martin

, Holloway

, et al. First Pass Effect: A New Measure for Stroke Thrombectomy Devices. Stroke. 2018;49(3):660–6.

49.

Bourcier

, Saleme

, Labreuche

, Mazighi

, Fahed

, Blanc

, et al. More than three passes of stent retriever is an independent predictor of parenchymal hematoma in acute ischemic stroke. J Neurointerv Surg. 2019;11(7):625–9.

50.

Charakida

, O’Neil

, Masi

, Papageorgiou

, Tousoulis

. Inflammatory disorders and atherosclerosis: new therapeutic approaches. Curr Pharm Des. 2011;17(37):4111–20.

51.

Vaya

, Hernandez-Mijares

, Bonet

, Sendra

, Sola

, Perez

, et al. Association between hemorheological alterations and metabolic syndrome. Clin Hemorheol Microcirc. 2011;49(1-4):493–503.

52.

Yedgar

, Koshkaryev

, Barshtein

. The red blood cell in vascular occlusion. Pathophysiol Haemost Thromb. 2002;32(5-6):263–8.

53.

Holsworth

, Jr . , Cho YI, Weidman J. Effect of hydration on whole blood viscosity in firefighters. Altern Ther Health Med. 2013;19(4):44–9.

54.

Rowat

, Graham

, Dennis

. Dehydration in hospital-admitted stroke patients: detection, frequency, and association. Stroke. 2012;43(3):857–9.

55.

Sloop

, Garber

. The effects of low-density lipoprotein and high-density lipoprotein on blood viscosity correlate with their association with risk of atherosclerosis in humans. Clin Sci (Lond). 1997;92(5):473–9.

56.

Rosenson

, Shott

, Tangney

. Hypertriglyceridemia is associated with an elevated blood viscosity Rosenson: triglycerides and blood viscosity. Atherosclerosis. 2002;161(2):433–9.

57.

Koenig

, Hombach

, Ernst

, Sund

, Mraz

, Keil

. Plasma viscosity as a cardiovascular risk factor. Circulation. 1992;86(3):1045.

58.

Jung

, Mrowietz

, Hiebl

, Franke

, Pindur

, Sternitzky

. Influence of rheological parameters on the velocity of erythrocytes passing nailfold capillaries in humans. Clin Hemorheol Microcirc. 2011;48(1):129–39.