Cortical and Internal Watershed Infarcts Might Be Key Signs for Predicting Neurological Deterioration in Patients with Internal Carotid Artery Occlusion with Mild Symptoms.

BACKGROUND: Treatment for acute ischemic stroke due to large vessel occlusion (LVO) with mild symptoms is under discussion. Although most patients have good outcomes, some patients deteriorate and have unfavorable results. Imaging findings that predict the prognosis of LVO with mild symptoms are needed to identify patients who require treatment. In this study, we focused on watershed infarctions (WSIs), because this clinical phenomenon quite sensitively reflects changes in cerebral blood flow. The purpose of this study was to assess positive rates of WSI on MRI findings in patients with internal carotid artery (ICA) occlusion, and compare WSI-positive rates between patients divided according to their clinical course.
METHODS: We retrospectively collected data of 1,531 patients who presented with acute ischemic stroke between June 2006 and July 2019. Among them, we chose symptomatic ICA occlusion patients with a past history of atrial fibrillation who were treated conservatively. We divided these patients into two groups, those with maintenance or improvement in their NIHSS score after hospitalization, and those whose NIHSS score worsened. We compared WSI-positive rates between these two groups.
RESULTS: Thirty-seven of the 1,531 patients were included in this study. Of them, total NIHSS score was maintained or improved in 8 patients (group A), 3 of whom (37.5%) had internal watershed infarctions (IWIs). In group B, consisting of patients whose NIHSS score worsened by >2 at 7 days from symptom onset, 24 (82.8%) had IWIs. Group A thus had statistically lower IWI positivity rates than group B (p = 0.02). Three patients (37.5%) in group A had cortical watershed infarctions (CWIs), while 27 patients in group B (93.1%) had CWIs. Group A thus had a significantly lower CWI positivity rate than group B (p = 0.002).
CONCLUSION: In patients with mildly symptomatic ICA occlusion, CWIs and IWIs might be key signs for predicting neurological deterioration after hospitalization.

Introduction

Since the publication of several randomized control trials, endovascular thrombectomy has become standard therapy for patients with moderate to severe ischemic stroke and large vessel occlusion (LVO) [1, 2, 3, 4, 5, 6, 7, 8, 9]. Based on these trials, current guidelines endorse endovascular thrombectomy as the standard of care for patients with LVO in the anterior circulation (internal carotid artery [ICA] and middle cerebral artery [MCA]) with National Institutes of Health Stroke Scale (NIHSS) scores of 6 or more [10]. This NIHSS cutoff was established because the eligibility criteria of previous RCTs mostly excluded patients with NIHSS scores <6.

The endovascular approach for low NIHSS score emergency LVO has been previously reported; however, data are scarce [11, 12], and endovascular intervention in such cases remains to be established. Therefore, treatment of patients who present with mild neurological symptoms and are also found to have an intracranial LVO pose an exceptionally difficult therapeutic dilemma. In these cases, the clinical outcome is generally poor [13, 14, 15].

Acute ischemic stroke (AIS) with ICA occlusion is usually caused by compromised cerebral perfusion, an embolic event, or a combination of hypoperfusion and embolism [16, 17]. Since watershed infarctions (WSIs), which occur at characteristic locations at the junction between two main arterial territories, correlate with both hypoperfusion and microembolism, we focused on WSIs in this study [16, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29]. Based on these reports, we hypothesized that patients with ICA occlusion with mild symptoms, and whose initial MRI shows fewer WSIs, are likely to have a good clinical course. The purpose of this study was to compare WSI positivity rates in patients with ICA occlusion with mild symptoms between patients whose NIHSS score remained or improved after hospitalization and those whose NIHSS score worsened. This is the first study to research the correlation between WSIs and mild symptoms due to ICA occlusion.

Methods

Study Design

The Ethics Committee of Seirei Mikatahara General Hospital approved the protocol for this study, which was performed in accordance with the principles of the Declaration of Helsinki. Participants in this observational, non-randomized study were identified via a retrospective electronic chart review of AIS patients treated between June 2006 and July 2019 at the Seirei Mikatahara General Hospital.

Patients

We retrospectively collected the data of 1,531 patients who presented with AIS between June 2006 and July 2019. Patients who met the following criteria were selected from among them: (1) those presenting within 24 h of symptom onset, (2) those who underwent emergency brain MRI, including diffusion-weighted imaging (DWI) and magnetic resonance angiography (MRA), (3) who had symptomatic ICA occlusion, and (4) a past history of atrial fibrillation, to extract patients of cardioembolism. ICA occlusion was diagnosed based on the findings of acute MRA. We excluded patients who met the following criteria: (1) those who underwent endovascular treatment, (2) had incomplete early imaging, and (3) early loss to follow-up.

Imaging

According to our stroke imaging protocol, patients were scheduled to undergo an MRI on admission, and follow-up MRI or computed tomography 5–7 days after symptom onset. The initial and follow-up MRI protocol included DWI, fluid-attenuated inversion recovery (FLAIR), and MRA. Imaging studies were performed using a 1.5-T scanner (Provida) from Philips Healthcare (Best, The Netherlands).

The MR images were analyzed by two experienced stroke neurologists blinded to the clinical data. The patterns of DWI lesions were classified as perforating artery infarctions, pial infarctions, cortical watershed infarctions (CWI), and internal watershed infarctions (IWI). According to classic neuropathology studies [18], CWI was defined as infarctions of the white matter between the cortical territories of the anterior cerebral artery, MCA, and posterior cerebral artery. IWI was defined as infarctions of the white matter between the deep and superficial arterial systems of the MCA, or between the superficial systems of the MCA and anterior cerebral artery (Fig. 1). Arterial occlusion was defined as complete loss of the digital flow signal on MRA. The presence of MCA flow and collateral flow of the anterior communicating artery and posterior communicating artery was defined by signal positivity on MRA.

Fig. 1

Classification of watershed infarction: a and b, subtypes of cortical watershed infarction; c, subtypes of internal watershed infarction.

Clinical and Statistical Analysis

We assessed NIHSS scores at 7 days from symptom onset and divided patients into two groups. Patients who satisfied the following criteria were categorized as group A: (1) NIHSS score remained unchanged or improved at 7 days from symptom onset, (2) NIHSS motor scores of the contralateral arm or leg remained as ≤2. Patients who met the following criteria were selected as group B: (1) progressive neurological signs with an increase in NIHSS score of ≥2, and (2) motor score of the contralateral arm or leg increasing to >2 at 7 days. We compared baseline characteristics, brain images, initial NIHSS score, and NIHSS score at 7 days from symptom onset between groups A and B. Primary outcomes were set as IWI and CWI positivity rate, and secondary outcomes were set as collateral flow (anterior communicating, posterior communicating arteries) positivity rate, MCA flow positivity rate, infarctions of the perforating area according to initial MRI, and deteriorated area at 7 days from symptom onset.

Categorical variables were analyzed using the Fisher exact test. Continuous variables were analyzed using the unpaired t test or Mann-Whitney U test, as appropriate. Statistical significance was set at p < 0.05. All statistical analyses were performed using EZR software (Saitama Medical Center, Jichi Medical University, Saitama, Japan) [30].

Results

Clinical Information

A total of 1,531 patients had AIS in the anterior circulation during the study period. Among them, 49 patients had symptomatic ICA occlusion with a past history of atrial fibrillation. A total of 12 of them were excluded because of endovascular treatment (n = 7) and incomplete MRI evaluation (n = 5). Thus, 37 patients (15 men, 22 women) were finally included. Mean patient age was 79.4 years (SD = 10.06) and 8 patients had maintenance or improvement of their total NIHSS score and maintenance of the NIHSS motor score of the contralateral arm or leg at ≤2 at 7 days from symptom onset.

Table 1 shows the general characteristics of the subjects according to their NIHSS score at 7 days from symptom onset. Age, sex, and baseline clinical risk factors did not differ between the two groups. Initial NIHSS scores indicated more mild cases in group A than in group B. There were no statistically significant differences in intravenous tissue plasminogen activator utilization rates between groups A and B.

Table 1

Baseline characteristics of the study groups

Baseline characteristics	Group A (n = 8)	Group B (n = 29)	p
Age (mean ± SD), years	76.5±10.94	80.13±9.82	0.590
Male, n (%)	4 (50.0)	11 (37.9)	0.830
Initial NIHSS score (median, IQR)	5 (3.25)	17 (12.00)	0.000
0–5, n (%)	4 (50.0)	1 (3.4)	0.005
6–10, n (%)	2 (25.0)	4 (13.8)	0.600
>10, n (%)	2 (25.0)	24 (82.8)	0.004
Risk factors, n (%)
Hypertension	4 (50.0)	21 (72.4)	0.440
Diabetes mellitus	1 (12.5)	6 (20.7)	1.000
Dyslipidemia	2 (25.0)	8 (27.6)	1.000
Previous stroke	1 (12.5)	3 (10.3)	1.000
Smoking	3 (37.5)	5 (17.2)	0.330
Contralateral ICAD (>50%)	1 (12.5)	0 (0.0)	0.220
IV-tPA, n (%)	0 (0.0)	4 (13.8)	0.560
Onset to imaging time (median, IQR), h	3 (1)	3 (2)	0.970

Baseline clinical risk factors did not differ between the two groups. The initial NIHSS scores of group A indicated more mild cases than in group B. NIHSS, National Institutes of Health Stroke Scale; ICAD, internal carotid artery disease; IV-tPA, intravenous tissue plasminogen activator.

Primary Outcome Measurements

There were 3 IWI-positive cases (37.5%) in group A and 24 IWI-positive cases in group B (82.8%). Group A thus had a significantly lower IWI positivity rate than group B (p = 0.02), as assessed using Fisher's exact test. There were 3 (37.5%) CWI-positive cases in group A and 27 (93.1%) in group B. Group A had a statistically significantly lower CWI positivity rate than group B (p = 0.002), as assessed using Fisher's exact test (Table 2).

Table 2

Outcome data of the study groups

Outcomes	Group A (n = 8)	Group B (n = 29)	p
Initial MRI
DWI-ASPECTS (median, IQR)	9 (8)	4 (3)	0.000
Acom A flow positivity, n (%)	8 (100.0)	16 (55.2)	0.030
Pcom A flow positivity, n (%)	6 (75.0)	15 (51.7)	0.420
MCA flow positivity, n (%)	5 (62.5)	4 (13.8)	0.010
Perforating area infarction, n (%)	2 (25.0)	25 (86.2)	0.002
Pial area infarction, n (%)	7 (87.5)	27 (93.1)	0.240
IWI, n (%)	3 (37.5)	24 (82.8)	0.020
CWI, n (%)	3 (37.5)	27 (93.1)	0.002
IWI + CWI, n (%)	1 (12.5)	20 (69.0)	0.012
NIHSS score at 7 days (median, IQR)	4.5 (0.75)	21 (17)	0.000
0–5, n (%)	6 (75.0)	0 (0.0)	0.000
6–10, n (%)	1 (12.5)	1 (3.4)	0.390
>10, n (%)	1 (12.5)	28 (96.6)	6.040
MR or CT images at 7 days
DWI, CT-ASPECTS (median, IQR)	8 (7)	2 (0)	0.000
Deteriorated regions
Perforating area, n (%)	2 (25.0)	20 (69.0)	0.040
Pial area, n (%)	3 (37.5)	12 (41.4)	1.000

Group A had significantly lower IWI and CWI positivity rates than group B (p = 0.02, 0.002), as indicated using Fisher's exact test. In group A, there were significantly higher positivity rates of Acom artery flow and MCA flow, and lower rates of infarctions in the perforating area at initial MRI (p = 0.03, 0.01, and 0.002) according to Fisher's exact test. MRI, magnetic resonance imaging; DWI, diffusion-weighted images; ASPECTS, Alberta Stroke Program Early CT Score; Acom A, anterior communicating artery; Pcom A, posterior communicating artery; MCA, middle cerebral artery; IWI, internal watershed infarction; CWI, cortical watershed infarction; NIHSS, National Institutes of Health Stroke Scale; CT, computed tomography.

Secondary Outcome Measurements

Group A had significantly higher rates of anterior communicating artery flow positivity and MCA flow positivity, and lower rates of infarctions of the perforating area in initial MRI (p = 0.03, 0.01, and 0.002, respectively), as assessed using Fisher's exact test. The rate of deteriorated infarctions in the perforating area at 7 days from symptom onset was also statistically significantly lower in group A than group B (p = 0.04), as seen using Fisher's exact test (Table 2).

Discussion

Both IWI and CWI positivity rates were higher in patients whose NIHSS scores worsened by >2 at 7 days from symptom onset as compared to patients with a good clinical course (whose NIHSS score did not change or improved at 7 days from symptom onset). These outcomes supported our hypothesis. In our study, lesions in the perforating area showed statistically greater deterioration in patients whose NIHSS scores worsened by >2 at 7 days from symptom onset. This suggests a correlation between WSI and lesions of the perforating area.

Many reports have implicated hemodynamic impairment as a determinant of IWI, particularly in cases with small, deep infarcts in the centrum semiovale (rosary-like watershed infarcts) [16, 19, 21, 22, 24]. Several characteristics of IWIs have been pointed out, such as vulnerability of the vasculature of the centrum semiovale [23, 25], their association with severe systemic hypotension [21, 23, 25], significant component of an incompletely infarcted brain on pathological studies [31], and impaired cerebral vasoreactivity associated with oxygen metabolism failure [16, 32].

Lenticulostriate arteries (LSAs) are among the largest perforating arteries in the brain. These vessels originate from the MCA and range from 2 to 12 in number [33]. ICA occlusion leads to a decrease in anterograde blood flow in the LSAs. This decrease in blood flow is compensated by retrograde flow from the ICA and the pial branch of the posterior cerebral artery. Since the LSAs are mostly located distally, perfusion pressure is likely to be the lowest for these vessels. Therefore, we considered that hypoperfusion is a significant underlying mechanism of infarction in the territory supplied by the LSA, which is similar to the mechanism underlying IWI.

Conversely, CWI is regarded as being induced by a microembolic mechanism [26, 29] because of a lower level of impairment of cerebral vasoreactivity [27, 28] and oxygen metabolism [16, 20]. There is reportedly a positive correlation between cerebral perfusion and the number of LSAs [34], with occlusion of an individual LSA usually resulting in large ganglionic infarcts [35]. If microembolism of an LSA occurs, blood flow to the basal ganglia decreases encouraging hypoperfusion in the LSA. Therefore, we believe that microembolism of the LSAs results in deterioration of hypoperfusion in the LSA, resulting in a large ganglionic infarct.

The results of this study are subject to some limitations. First, since this was a retrospective study with a small sample size, further studies with a larger sample size are required to confirm our results. Second, we evaluated the initial MRI within 24 h of symptom onset (median time of both groups A and group B: 3 h). There is no evidence suggesting the best timing for initial assessment of WSI.

Conclusion

In patients with mild symptoms who are found to have symptomatic ICA occlusion, presence of cortical and internal watershed infarcts might be key signs for predicting neurological deterioration after hospitalization.

Statement of Ethics

The Ethics Committee of Seirei Mikatahara General Hospital approved the protocol for this study, which was performed in accordance with the principles of the Declaration of Helsinki.

Conflict of Interest Statement

No company had influence on or knowledge of the results of this study.

Funding Sources

The authors received no funding for this work.

Author Contributions

Conception and design: Y. Amano, A. Fujimoto, S. Akamine. Acquisition of data: Y. Amano. Statistical analysis: Y. Amano. Analysis and interpretation of data: H. Sano, H. Kenmochi. Reviewed submitted version of manuscript: A. Fujimoto, H. Sato. Study supervision: S. Akamine.