Abstract
Objectives
This study clarified the usefulness of carotid duplex ultrasound (CDU) in evaluating large vessel occlusion (LVO) in patients with acute stroke planned to be treated with mechanical thrombectomy (MT).
Methods
This study was single-center, prospective, observational trial. If the ratio of end-diastolic velocity in the common carotid arteries was ≥1.4, or diastolic flow in the affected internal carotid artery (ICA) was absent on CDU, patients were immediately transferred to the angio-suite without additional cerebrovascular imaging. Clinical parameters, including time metrics and outcomes, were evaluated in participants.
Patients
We enrolled stroke patients with a National Institutes of Health Stroke Scale score ≥6 and Alberta Stroke Program Early CT score ≥6 in whom MT could be initiated within 6 hours of the stroke onset.
Results
Among 140 patients screened during the study period, 48 were ultimately enrolled. Twenty-seven patients were diagnosed with LVO by CDU alone. CDU offered 83% sensitivity and 82% specificity for identifying the occlusion of the ICA or M1 segment of the middle cerebral artery. Among the 29 total patients treated with MT, 20 (67%) showed a modified Rankin Scale score ≤2 at 90 days. The door-to-puncture time was significantly shorter in patients evaluated by CDU alone (34 minutes) than in those evaluated by magnetic resonance angiography after CDU (47.5 minutes, p<0.001).
Conclusion
CDU might reduce the time metrics for early initiation of MT with good sensitivity and specificity in identifying LVO.
Keywords: stroke, mechanical thrombectomy, large vessel occlusion, carotid duplex ultrasound
Introduction
Mechanical thrombectomy (MT) is an established and effective treatment for acute large vessel occlusion (LVO) (1-7). When MT is considered for acute stroke patients, performing non-invasive intracranial vascular imaging as quickly as possible to select candidates for MT is recommended (7,8). Non-invasive vascular imaging is thus a key component in reducing delays to MT.
Computed tomography angiography (CTA) is widely used in the standard imaging for acute LVO because of its short inspection time, good accessibility, and evidence for validation in randomized controlled trials of MT (8). However, ultrasound is non-invasive and can be performed simultaneously during a neurological evaluation or intravenous thrombolysis in the emergency room. Transcranial Doppler (TCD) has been a typical ultrasound modality used to evaluate intracranial vessels, but installation of TCD is a time-wasting and requires skill of the examiner. Although detecting intracranial vessels using transcranial color flow imaging is more manageable than with TCD because of the real-time visualization of intracranial structures with B-mode and color imaging (9), the temporal bone window is often insufficient to evaluate intracranial arteries, particularly in elderly patients and Asian women (10).
Carotid duplex ultrasound (CDU), which is usually used to assess extracranial vessels (11), is the easiest of ultrasound modalities to perform. Although most physicians believe that CDU cannot evaluate intracranial vessel occlusion, this modality can identify proximal intracranial occlusion in the anterior circulation. If the end-diastolic velocity (EDV) ratio (ED ratio) in the common carotid arteries (CCAs) (calculated as the EDV in the contralateral, non-affected CCA divided by the EDV in the ipsilateral, affected CCA) is ≥1.4, intracranial occlusive disease in the anterior circulation can be highly suspected (12,13) (Fig. 1).
Figure 1.
Representative findings on carotid duplex ultrasound (CDU) in patients with intracranial ICA occlusion. Left: End diastolic velocity (EDV) in the left CCA, as the non-affected side, was 27.2 cm/s. Right: EDV in the right CCA, as the affected side, was 6.8 cm/s. EDV ratio (ED ratio) in the CCAs is 4.0, obtained by dividing 27.2 by 6.8.
The safety and efficacy of CDU in vascular imaging for patients to be treated with MT are poorly understood. We therefore planned a study to evaluate the usefulness of CDU as vascular imaging for selecting patients to undergo MT, called Emergent Carotid ultrasound in Hyperacute cerebral vessel Occlusion for SELecting patients to be treated with Endovascular Clot retrieval Treatment (ECHO-SELECT).
Materials and Methods
The data in support of the present findings are available from the corresponding author on reasonable request and with approval from the co-authors and the Kohnan Hospital Ethics Committee.
Study population
This was a single-center, prospective study. Subjects were selected from among consecutive acute stroke patients admitted to Kohnan Hospital (Sendai, Japan) between October 2016 and June 2019. All patients admitted to our institute during this period were examined by neurologists, neurosurgeons, or both, and were screened by routine laboratory tests, as well as brain imaging. Based on the clinical and brain imaging findings, board-certified stroke neurologists specializing in stroke patient care made a diagnosis of ischemic stroke. The severity of neurological deficits was evaluated using the National Institutes of Health Stroke Scale (NIHSS) score on admission (14), and activities of daily living for patients before or after the onset of stroke were measured using the modified Rankin Scale (mRS) (15).
Inclusion criteria in this study were as follows: 1) age ≥18 years old; 2) mRS score of 0 or 1 before the stroke onset; 3) NIHSS score ≥6 on admission; 4) Alberta Stroke Program Early CT Score (ASPECTS) ≥6 (16), as diagnosed by ≥2 stroke physicians; 5) ability to initiate MT within 6 hours after the symptom onset; and 6) ED ratio in the CCAs ≥1.4, or the absence of diastolic flow in the extracranial internal carotid artery (ICA) on CDU. Of note, this sixth criterion for CDU was revised in June 2017, as follows: 6) EDV in the CCAs and ICA able to be evaluated by CDU.
We excluded patients based on the following criteria: 1) vascular imaging already performed in the referring hospital; 2) difficulties with evaluation of ASPECTS on initial computed tomography (CT); 3) symptoms indicative of posterior circulation infarct or mild lacunar syndrome; or 4) other circumstances considered inappropriate by the investigators.
Patients or their relatives provided their written informed consent according to ethical regulations. The Kohnan Hospital Ethics Committee approved the study protocol (280926). This study was registered as follows http://www.clinicaltrials.gov; unique identifier: NCT02989376, URL: https://www.umin.ac.jp/ctr/; unique identifier: UMIN000025032.
Data collection and definitions
All clinical and investigative data were prospectively entered in a standardized fashion by stroke neurologists into the Kohnan Hospital Stroke Registry. In addition to general patient characteristics, including the age, sex, cardiovascular risk factors, mRS before stroke onset, subtype of stroke classification, and initial NIHSS score, the workflow time metrics and use of intravenous recombinant tissue plasminogen activator (rt-PA) or MT were also recorded. The onset-to-door time (OTD) was defined as the duration from the last known time when the patient appeared well and free of index stroke symptoms to the time at which the patient arrived at the emergency room. The door-to-puncture time (DTP) was defined as the duration from when the patient arrived in the emergency room to the time when groin puncture was performed in the angiography suite. The puncture-to-recanalization time (PTR) was defined as the duration from when groin puncture was performed to the time when significant recanalization was obtained. The onset-to-recanalization time (OTR) was defined as the duration from the last known time when the patient appeared well and free of index stroke symptoms to the time when significant recanalization was obtained. Significant recanalization on MT was defined as modified treatment in cerebral infarction (mTICI) grade 2b-3 (17).
Vascular imaging and treatment
All CDU procedures were performed immediately before or just after obtaining non-contrast CT by stroke neurologists, including three registered neurosonographers of the Japan Academy of Neurosonology (R.I., K.F., and Y.Y.), using a LOGIQ e (GE Healthcare, Chicago, USA) and a 12-MHz linear probe. Patients were examined in the supine position. The head was turned away from the ipsilateral side, and then the probe was applied to the ipsilateral neck using the anterior oblique approach. On longitudinal scans, the sample volume was set to the CCA, and the isonation angle was maintained at 60°. First, we measured the EDV of both CCAs. When measuring the flow in the ICA, the probe was moved distal to the CCA to identify the ICA. If the ED ratio in the CCAs was ≥1.4 or the diastolic flow in the ICA was absent on CDU, the patient was transferred to the angio-suite immediately without additional vascular imaging, for an evaluation with digital subtraction angiography (DSA). If the ED ratio in the CCAs was <1.4 and the diastolic flow in the ICA was intact on CDU, additional vascular imaging by magnetic resonance angiography (MRA) was performed. In our center, MRA was employed as the standard vascular imaging modality and could be performed immediately, 24 hours/day, 7 days/week. In cases where MRA was contraindicated, CTA was performed. When LVO diagnosed as an indication for MT was detected on MRA, the patient was transferred to the angio-suite. Based on DSA findings, the final decision was made regarding whether or not to perform MT. If the decision was made not to perform MT, standard medical treatment was provided. Decisions about treatments, including the choice of devices for MT or medications including intravenous rt-PA (0.6 mg/kg, the approved dose in Japan), were determined by the physician in charge.
Outcomes
In the present study, the primary outcome was defined as the rate of mRS 0-2 at 90 days of onset, and secondary outcomes as time metrics including DTP, PTR, and OTR. When physicians could not assess mRS, patients or relatives were contacted via telephone or postal mail to estimate the mRS. The safety outcome was symptomatic intracranial hemorrhage (sICH) within 72 hours after stroke onset, with sICH defined as any intracranial hemorrhage associated with worsening of the NIHSS score by ≥4 points.
Statistical analyses
We conducted univariate analyses to compare clinical and radiological characteristics and outcomes between patients transferred directly to the angio-suite based on CDU (CDU-alone group) and patients evaluated using additional vascular imaging modalities such as MRA (MRA after CDU group). As the inclusion criteria had been changed, we also calculated the sensitivity and specificity of an LVO diagnosis based on CDU alone to identify patients treated with MT only among those enrolled since June 2017.
Categorical variables are presented as numbers and percentages and were analyzed using the χ2 test and Fisher's exact test as appropriate. Continuous variables are expressed as the median and interquartile range (IQR). Continuous variables showing non-normal distributions were analyzed using Wilcoxon's rank-sum test.
All statistical analyses were performed using EZR software (Saitama Medical Center, Jichi Medical University, Saitama, Japan; http://www.jichi.ac.jp/saitama-sct/SaitamaHP.files/manual.html) (18), a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). More precisely, EZR is a modified version of R Commander designed to add statistical functions frequently used in biostatistics. All reported p values are two-tailed, and values of p<0.05 were considered statistically significant.
Results
Study flow and characteristics of enrolled patients
From October 2016 to June 2019, of 248 patients ≥18 years old with an mRS score of 0 or 1 before stroke onset, NIHSS score ≥6 on admission, and time to initiate MT ≤6 hours after the symptom onset, 95 with intracerebral hemorrhage and 13 with ischemic stroke with ASPECTS ≤5 deemed not eligible for screening. Among the 140 patients initially screened, we were unable to enroll 92 patients due to the following reasons: 39 had already been evaluated with vascular imaging at the initial referring hospital; 16 were lacking or incapable of providing informed consent; 13 presented with symptoms suggestive of posterior circulation, lacunar syndrome, or rapid improvement after the initial evaluation; 9 did not meet the criteria for LVO on CDU (excluded up to May 2017, as mentioned in the Methods section); 9 could not be emergently evaluated using CDU on admission (3 due to no examiner, 1 due to an unstable systemic status, and 5 for unknown reasons); and 6 were evaluated by MRA at the physician's discretion, probably due to noncompliance with the research protocol (Fig. 2). Among the 92 patients excluded from this study as described above, 50 were eventually treated with MT. Although the excluded patients were significantly younger than the enrolled patients, the initial NIHSS scores were comparable (Supplementary material). During the study period, 48 patients (median age, 75 years old; 60.4% men; median NIHSS score, 16.5) were ultimately enrolled in this study. The characteristics of these 48 enrolled patients are shown in Table 1.
Figure 2.
Study flowchart. CDU: carotid duplex ultrasound, MT: mechanical thrombectomy, LVO: large vessel occlusion, MRA: magnetic resonance angiography
Table 1.
Characteristics of the 48 Patients Enrolled.
| Characteristics | n=48 | |
| Age (years), median [IQR] | 75 [69-85] | |
| Men, n (%) | 29 (60.4) | |
| Onset-to-door time (min), median [IQR] | 66.5 [50.75-114.5] | |
| Initial NIHSS score, median [IQR] | 16.5 [11-22.25] | |
| ASPECTS, median [IQR] | 9 [8.75-10] | |
| Persistent atrial fibrillation, n (%) | 14 (29.2) | |
| Oral antihypertensive, n (%) | 37 (77.1) | |
| Systolic blood pressure on arrival (mmHg), median [IQR] | 145 [135.5-165.5] | |
| Diastolic blood pressure on arrival (mmHg), median [IQR] | 83 [71-99.5] | |
| Intravenous rt-PA, n (%) | 43 (89.6) | |
| ED ratio ≥1.4, n (%) | 26 (54.2) | |
| ICA occlusion, n (%) | 12 (25) | |
| ICA or M1 occlusion, n (%) | 28 (58.3) | |
| Tandem ICA and M1 occlusion | 0 (0) | |
| Contralateral occlusive disease, n (%) | 6 (12.5) | |
| Ipsilateral carotid stenosis, n (%) | 2 (0.4) | |
| Subtype classification | ||
| Large artery disease | 4 (8.3) | |
| Cardioembolic stroke | 32 (66.7) | |
| Others | 12 (25) | |
| Door-to-needle time (min), median [IQR] | 15 [13-19] | |
| Door-to-puncture time (min), median [IQR] | 39 [30-45] | |
| Puncture-to-recanalization time (min), median [IQR] | 46 [29.5-60.75] | |
| mRS 0-2 at discharge, n (%) | 28 (58.3) | |
| mRS 0-2 at 90 days, n (%) | 29 (60.4) |
IQR: interquartile range, NIHSS: National Institutes of Health Stroke Scale, ASPECTS: Alberta Stroke Program Early CT Score, rt-PA: recombinant tissue plasminogen activator, ED ratio: end-diastolic velocity ratio, ICA: internal carotid artery, mRS: modified Rankin Scale
The diagnosis of LVO and indication for MT
Fig. 2 shows the process for diagnosing LVO and determining the indication for MT. Twenty-seven patients were diagnosed with LVO according to CDU, and the remaining 21 did not show LVO. Among the 27 patients diagnosed with LVO on CDU, 25 were directly transferred to the angio-suite without additional imaging.
On emergent DSA, 11 patients were diagnosed with ICA occlusion, 10 with M1 segment occlusion, 1 with M2 segment occlusion, and 1 with M3 segment middle cerebral artery (MCA) occlusion. Two patients had no treatable vessel occlusion after intravenous rt-PA and showed rapid neurological improvement. In contrast, 23 patients were determined to warrant an evaluation with additional imaging by physicians after CDU, including 2 patients who had already been diagnosed with LVO based on CDU. Among these 23 patients, 17 were deemed suitable for treatment with medical therapy alone including 1 diagnosed with M1 segment occlusion and extensive ischemic core on diffusion-weighted image (DWI) and 4 with M2 segment occlusion. The remaining 6 patients diagnosed with LVO based on MRA were evaluated with emergent DSA (1 diagnosed with ICA occlusion, 5 with M1 segment MCA occlusion). There were no patients evaluated by CTA.
Sensitivity and specificity analyses
Of the 27 patients diagnosed with LVO on initial CDU, including 2 with an additional MRA evaluation, 24 (89%) were treated with MT. Conversely, of the 21 patients not showing LVO on CDU, 5 (24%) were treated with MT. We calculated the sensitivity and specificity of the LVO diagnosis based on CDU only for the 40 patients enrolled since June 2017 (Table 2). The sensitivity and specificity for identifying occlusion of the ICA or M1 segment MCA were 83% and 82%, respectively. Furthermore, the positive predictive value and negative predictive value were 86% and 78%, respectively, while the sensitivity, and negative predictive value for identifying ICA occlusion were both 100%.
Table 2.
The Sensitivity and the Specificity of Diagnosing Vessel Occlusion Based on CDU (n=40 Enrolled since June 2017).
| Site of LVO | Sensitivity (95% CI) | Specificity (95% CI) | Positive predictive value (95% CI) |
Negative predictive value (95% CI) |
||||
|---|---|---|---|---|---|---|---|---|
| ICA occlusion (%) | 100 (76-100) | 60 (52-60) | 46 (35-46) | 100 (87-100) | ||||
| ICA/M1 occlusion (%) | 83 (70-90) | 82 (66-92) | 86 (73-93) | 78 (62-87) | ||||
| ICA/M1/M2 occlusion (%) | 74 (64-79) | 85 (63-95) | 91 (78-97) | 61 (46-69) | ||||
| All LVO treatable for MT (%) | 83 (72-89) | 88 (70-96) | 91 (78-97) | 78 (63-85) |
CDU: carotid duplex ultrasound, CI: confidence interval, LVO: large vessel occlusion, ICA: internal carotid artery, MT: mechanical thrombectomy
Comparing the groups using a univariate analysis
In total, 23 patients diagnosed with LVO based on CDU alone (CDU-alone group) and 6 patients diagnosed with LVO based on MRA, including 1 patient who was additionally evaluated with MRA despite an initial diagnosis of LVO on CDU (MRA after CDU group), were treated with MT (Fig. 2).
We compared the clinical and radiological characteristics and outcomes between the CDU-alone group and MRA after CDU group using univariate analyses (Table 3). Initial NIHSS score was comparable between groups (21 vs. 20; p=0.42). ASPECTS tended to be lower, but not significantly, in the CDU-alone group than in the MRA after CDU group (8 vs. 9; p=0.06). The prevalence of ICA occlusion was 11/23 (47.8%) in the CDU-alone group, compared to just 1/6 (16.7%) in the MRA after CDU group, although this difference was not significant (p=0.35). DTP was significantly shorter in the CDU-alone group than in the MRA after CDU group (34 minutes vs. 47.5 minutes; p<0.001). Rates of mTICI 2b-3 (100% vs. 80%; p=0.18) and sICH (0% vs. 0%; p=1.00) were comparable between the CDU-alone group and the MRA after CDU group. Only one patient had a puncture site complication (pseudoaneurysm requiring ultrasound-guided compression). In addition, the prevalence of patients with mRS 0-2 at 90 days, as the primary outcome, was comparable between the CDU-alone group and the MRA after CDU group (70% vs. 50%; p=0.63).
Table 3.
Univariate Analysis between CDU-alone Group and MRA after CDU Group in Patients Treated with MT.
| Characteristics | CDU-alone (n=23) | MRA after CDU (n=6) | p | |||
|---|---|---|---|---|---|---|
| Age (years), median [IQR] | 73 [69-84] | 84.5 [79.5-85] | 0.26 | |||
| Men, n (%) | 11 (48) | 4 (67) | 0.65 | |||
| Onset-to-door time (min), median [IQR] | 65 [48-90] | 90.5 [50-147.5] | 0.45 | |||
| Initial NIHSS, median [IQR] | 21 [19.25-23.5] | 20 [14.5-23.5] | 0.42 | |||
| ASPECTS, median [IQR] | 8 [8-8.75] | 9 [9-10] | 0.06 | |||
| Persistent atrial fibrillation, n (%) | 7 (30.4) | 2 (30.3) | 1.00 | |||
| Oral antihypertensive, n (%) | 18 (78.3) | 5 (83.3) | 1.00 | |||
| Intravenous rt-PA, n (%) | 21 (91) | 5 (83) | 0.52 | |||
| ICA occlusion, n (%) | 11 (52.2) | 1 (16.7) | 0.35 | |||
| ICA or M1 occlusion, n (%) | 21 (91) | 6 (100) | 1.00 | |||
| Contralateral occlusive disease, n (%) | 3 (13) | 0 (0) | 1.00 | |||
| Ipsilateral carotid stenosis, n (%) | 1 (4.3) | 1 (0) | 1.00 | |||
| Subtype classification | 1.00 | |||||
| Large artery disease | 2 (8.7) | 0 (0) | ||||
| Cardioembolic stroke | 15 (65.2) | 5 (83.3) | ||||
| Others | 6 (26.1) | 1 (6.7) | ||||
| Door-to-needle time (min), median [IQR] | 15 [13-16] | 14 [12-19] | 0.87 | |||
| Door-to-puncture time (min), median [IQR] | 34 [15-40.5] | 47.5 [45.5-50.25] | <0.001 | |||
| mTICI grade ≥2b, n (%) | 23 (100) | 4 (80) | 0.18 | |||
| Onset-to-recanalization time (min), median [IQR] | 143 [121-178.5] | 196 [137.5-272.5] | 0.29 | |||
| Symptomatic ICH, n (%) | 0 (0) | 0 (0) | 1.00 | |||
| Puncture site complication, n (%) | 1 (4.3) | 0 (0) | 1.00 | |||
| mRS 0-2 at discharge, n (%) | 15 (65) | 3 (50) | 0.65 | |||
| mRS 0-2 at 90 days, n (%) | 16 (70) | 3 (50) | 0.63 |
CDU: carotid duplex ultrasound, MRA: magnetic resonance angiography, IQR: interquartile range, NIHSS: National Institutes of Health Stroke Scale, ASPECTS: Alberta Stroke Program Early CT Score, rt-PA: recombinant tissue plasminogen activator, ICA: internal carotid artery, mTICI: modified treatment in cerebral infarction, sICH: symptomatic intracranial hemorrhage, mRS: modified Rankin Scale
Discussion
We demonstrated that CDU for selecting patients to be treated with MT could reduce time metrics for early initiation of MT, providing good sensitivity and specificity for detecting LVO. Particularly for identifying ICA occlusion, both the sensitivity and negative predictive value were 100%. Although the ASPECTS tended to be lower and the prevalence of ICA occlusion higher, but not significantly so, in patients evaluated with CDU alone than in patients evaluated with MRA after CDU, despite these disadvantageous factors, the prevalence of mRS 0-2 at 90 days (primary outcome) was not fewer in patients evaluated with CDU alone than in patients evaluated with MRA after CDU (70% vs. 50%).
Immediate noninvasive intracranial vascular imaging was strongly recommended for patients with acute ischemic stroke to select patients who should be treated with MT (2,3). However, time metrics in the workflow for MT should be as short as possible to maximize the benefits in terms of clinical outcomes (19,20). Simplicity and high sensitivity are required for intracranial vascular imaging in the acute stroke setting. CTA is widely accepted as a standard modality for evaluating LVO in patients with acute stroke because of its short inspection time, good accessibility, and validation from randomized controlled trials for thrombectomy (8). However, even though ultrasonography is a noninvasive modality and can be performed promptly at the bedside during neurological evaluations, placement of an intravenous line, or infusion of intravenous rt-PA, ultrasonography has been considered inappropriate for acute stroke patients, particularly in the acute setting, because of its operator-dependent method and the time wasted in fitting the device for TCD (8).
CDU is generally recognized as suitable for evaluating extracranial vessels, such as carotid arteries (11), but CDU has been reported capable of identifying intracranial occlusion in the anterior circulation. Yasaka et al. reported that an ED ratio >1.3 in the CCAs was useful for detecting intracranial ICA occlusion with 97% accuracy, and an ED ratio >1.3 and <4.0 was also valuable for detecting occlusion in the horizontal segment of the MCA with 93% accuracy in patients with cardioembolic stroke (12). Kimura et al. also demonstrated that an ED ratio ≥1.4 could identify ICA occlusion with high sensitivity and specificity (13). In the present study, an ED ratio in the CCAs ≥1.4 or absence of diastolic flow in the ICA was used as the definition of LVO in the anterior circulation on CDU and showed sufficient sensitivity and specificity for detecting LVO in the acute stroke setting. Furthermore, we showed that CDU could reduce time metrics for early initiation of MT by cutting the initial vascular imaging time. Bilateral assessments for carotid stenosis and measurement of flow velocity, including in the ICA and CCA, take more than a few minutes, but measurement of the ED ratio in the CCAs alone can be completed more quickly. We could not clearly demonstrate improvements in clinical outcomes by selecting MT candidates with CDU, but the difference in imaging characteristics between the CDU-alone group and MRA after CDU group might have affected clinical outcomes. These impacts were not significant, possibly because of the small sample size, but the proportion of patients with ICA occlusion tended to be higher and ASPECTS lower in patients evaluated with CDU alone than in patients evaluated with MRA after CDU (Table 2). If LVO is suspected from emergent CDU, immediate transfer to the angio-suite, skipping conventional vascular imaging, such as CTA or MRA, appears to be an acceptable strategy to minimize delays in initiating MT. When performing MT based on the diagnosis of LVO by CDU, we should be cautious for the lack of information about collateral blood flow from non-occluded vessels to occluded culprit vessels, the inability to evaluate vascular accessibility by imaging of the aortic arch and its branches, and false-positive results with CDU.
Occlusive disease in the contralateral CCA or ICA can reduce the ED ratio and lead to false negative results with CDU. In the present study, there were 6 patients with contralateral stenosis of more than 50% or occlusion (Table 1). Among them, one patient with severe extracranial ICA stenosis had a false-positive result with CDU, probably due to recanalization after intravenous rt-PA. Although a chronic atherosclerotic lesion of the contralateral ICA or CCA might be less likely to affect the ED ratio than an acute embolic occlusion (13), the possibility of false negatives with CDU should be kept in mind.
Direct transfer to the angio-suite by skipping other imaging studies before DSA might represent a reasonable strategy to cut delays in MT. Transferring patients directly to the angio-suite after admission would significantly reduce intrahospital delays in MT. Mendez et al. reported that direct transfer to the angio-suite was an effective strategy to reduce workflow time, potentially significantly increase the odds of achieving favorable outcomes (21). Conventional non-contrast CT was also bypassed in their strategy, with cone-beam CT employed to rule out intracranial hemorrhage. The result of their strategy, however, was that 11 of 96 patients (11.6%) were diagnosed with intracranial hemorrhage, and 7 patients (7.2%) showed no treatable LVO after emergent DSA (16). In 5 pivotal trials for MT, the rate of futile MT was around 10% (1-5). In our series, 3 of the 27 patients (11%) diagnosed with LVO on initial CDU, all three of whom were treated with intravenous rt-PA before DSA, had no treatable LVO on DSA. We considered this false-positive rate in our strategy of CDU for LVO treated with MT to be acceptable. Furthermore, non-contrast CT was not bypassed in our strategy, as futile direct transfer to the angio-suite would be unfavorable for patients with intracerebral hemorrhage, which accounts for 20-30% of acute strokes in Japan (22).
Some limitations should be kept in mind when interpreting the results of this investigation. First, this study was designed as a single-center study. Second, we could not compare CDU alone with CTA as an additional vascular imaging modality after CDU. In our center, MRA is employed as a standard vascular imaging modality. Emergency MRI acquisition is widely available in Japan (23,24). If we had used CTA as a comparator for CDU, the decrease in time metrics for early initiation of MT might have been mitigated. In future prospective studies comparing CDU with CTA as vascular imaging for selecting patients to undergo MT, CDU will have to be performed before transferring the patient to the CT room or immediately after obtaining non-contrast CT in the CT room. Third, we were unable to enroll a sufficient number of participants in this study. The exclusion of patients who had already undergone vascular imaging at the referring center was unfortunate but unavoidable, as additional imaging would have been unethical. Fourth, the evaluation of the ischemic core by DWI in the MRA after CDU group might have underestimated the indication for MT compared to assessment by non-contrast CT in the CDU-alone group. Strict indications for MT based on DWI could have improved clinical outcomes in the MRA after CDU group. Fifth, the sensitivity and specificity of CDU for LVO in this study were slightly lower than described in previous studies (12,13), potentially due to the hyper-acute setting, recanalization by intravenous rt-PA, or the skill of the CDU operators. Furthermore, access to immediate ultrasonography in acute settings might not always be available in most hospitals. CDU is unlikely to become the standard vascular imaging modality for LVO among patients with acute stroke but can be a useful alternative modality when CTA is not immediately available, such as when the CT room is in use or out of service. In addition, CDU has potential utility as a modality for vascular imaging in the prehospital setting.
Conclusion
We demonstrated that an evaluation for LVO with CDU could reduce time metrics for early initiation of MT, with good sensitivity and specificity for identifying LVO. We believe that CDU represents a viable option for evaluating LVO among the various modalities available to select candidates for MT in acute stroke settings. A larger, multi-center study is warranted to confirm the significance of CDU for selecting patients to be treated with MT.
Author's disclosure of potential Conflicts of Interest (COI).
Ryo Itabashi: Honoraria, Bristol-Myers Squibb and Daiichi-Sankyo. Yukako Yazawa: Honoraria, Bayer, Bristol-Myers Squibb, Daiichi-Sankyo, Stryker and Medtronic. Yasushi Matsumoto: Honoraria, GE Healthcare Japan, Medtronic, Otsuka Pharmaceutical, Stryker, Medicos Hirata, Century Medical, Takeda, Kaneka Medics and Fuji Systems. Manabu Inoue: Honoraria, Daiichi-Sankyo, Bayer, Bristol-Myers Squibb, Medtronic and Medicos Hirata.
Financial Support
This trial was funded mainly by the Japan Agency for Medical Research and Development (AMED) (JP19ek0210109h0002).
Supplementary Material
The excluded patients were significantly younger and received significantly less intravenous rt-PA at our institute than those enrolled, but other background factors, including initial NIHSS score, were comparable between the two groups.
References
- 1. Berkhemer OA, Fransen PS, Beumer D, et al. A randomized trial of intraarterial treatment for acute ischemic stroke. N Engl J Med 372: 11-20, 2015. [DOI] [PubMed] [Google Scholar]
- 2. Goyal M, Demchuk AM, Menon BK, et al. Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med 372: 1019-1030, 2015. [DOI] [PubMed] [Google Scholar]
- 3. Saver JL, Goyal M, Bonafe A, et al. Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N Engl J Med 372: 2285-2295, 2015. [DOI] [PubMed] [Google Scholar]
- 4. Campbell BC, Mitchell PJ, Kleinig TJ, et al. Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med 372: 1009-1018, 2015. [DOI] [PubMed] [Google Scholar]
- 5. Jovin TG, Chamorro A, Cobo E, et al. Thrombectomy within 8 hours after symptom onset in ischemic stroke. N Engl J Med 372: 2296-2306, 2015. [DOI] [PubMed] [Google Scholar]
- 6. Goyal M, Menon BK, van Zwam WH, et al. ; HERMES Collaborators. Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet 387: 1723-1731, 2016. [DOI] [PubMed] [Google Scholar]
- 7. Powers WJ, Rabinstein AA, Ackerson T, et al. ; American Heart Association Stroke Council. 2018 guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 49: e46-e110, 2018. [DOI] [PubMed] [Google Scholar]
- 8. Demchuk AM, Menon BK, Goyal M. Comparing vessel imaging: noncontrast computed tomography/computed tomographic angiography should be the new minimum standard in acute disabling stroke. Stroke 47: 273-281, 2016. [DOI] [PubMed] [Google Scholar]
- 9. Itoh T, Matsumoto M, Handa N, et al. Rate of successful recording of blood flow signals in the middle cerebral artery using transcranial Doppler sonography. Stroke 24: 1192-1195, 1993. [DOI] [PubMed] [Google Scholar]
- 10. Tsuchiya T, Yasaka M, Yamaguchi T, Kimura K, Omae T. Imaging of the basal cerebral arteries and measurement of blood velocity in adults by using transcranial real-time color flow Doppler sonography. AJNR Am J Neuroradiol 12: 497-502, 1991. [PMC free article] [PubMed] [Google Scholar]
- 11. Koga M, Kimura K, Minematsu K, Yamaguchi T. Diagnosis of internal carotid artery stenosis greater than 70% with power Doppler duplex sonography. AJNR Am J Neuroradiol 22: 413-417, 2001. [PMC free article] [PubMed] [Google Scholar]
- 12. Yasaka M, Omae T, Tsuchiya T, Yamaguchi T. Ultrasonic evaluation of the site of carotid axis occlusion in patients with acute cardioembolic stroke. Stroke 23: 420-422, 1992. [DOI] [PubMed] [Google Scholar]
- 13. Kimura K, Yonemura Y, Terasaki T, Hashimoto Y, Uchino M. Duplex carotid sonography in distinguishing acute unilateral atherothrombotic from cardioembolic carotid artery occlusion. AJNR Am J Neuroradiol 18: 1447-1452, 1997. [PMC free article] [PubMed] [Google Scholar]
- 14. Lyden P, Brott T, Tilley B, et al. Improved reliability of the NIH Stroke Scale using video training. NINDS TPA Stroke Study Group. Stroke 25: 2220-2226, 1995. [DOI] [PubMed] [Google Scholar]
- 15. van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke 19: 604-607, 1988. [DOI] [PubMed] [Google Scholar]
- 16. Pexman JH, Barber PA, Hill MD, et al. Use of the Alberta Stroke Program Early CT Score (ASPECTS) for assessing CT scans in patients with acute stroke. AJNR Am J Neuroradiol 22: 1534-1542, 2001. [PMC free article] [PubMed] [Google Scholar]
- 17. Zaidat OO, Yoo AJ, Khatri P, et al. ; Cerebral Angiographic Revascularization Grading (CARG) Collaborators. STIR Revascularization Working Group; STIR Thrombolysis in Cerebral Infarction (TICI) Task Force. Recommendations on angiographic revascularization grading standards for acute ischemic stroke: a consensus statement. Stroke 44: 2650-2663, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Kanda Y. Investigation of the freely-available easy-to-use software “EZR” (Easy R) for medical statistics. Bone Marrow Transplant 48: 452-458, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Saver JL, Goyal M, van der Lugt A, et al. Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke: a meta-analysis. JAMA 316: 1279-1288, 2016. [DOI] [PubMed] [Google Scholar]
- 20. Al Banna M, Streib CD. Early endovascular thrombectomy for large-vessel ischemic stroke reduces disability at 90 days. Acad Emerg Med 26: 953-955, 2019. [DOI] [PubMed] [Google Scholar]
- 21. Mendez B, Requena M, Aires A, et al. Direct transfer to angio-suite to reduce workflow times and increase favorable clinical outcome. Stroke 49: 2723-2727, 2018. [DOI] [PubMed] [Google Scholar]
- 22. Hata J, Kiyohara Y. Epidemiology of stroke and coronary artery disease in Asia. Circ J 77: 1923-1932, 2013. [DOI] [PubMed] [Google Scholar]
- 23.OECD Health Statistics 2017. Health care resources [Internet]. [cited 2020 Dec 5]. Available from: https://stats.oecd.org/Index.aspx?DataSetCode=HEALTH_REAC.
- 24. Wintermark M, Luby M, Bornstein NM, et al. International survey of acute stroke imaging used to make revascularization treatment decisions. Int J Stroke 10: 759-762, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
The excluded patients were significantly younger and received significantly less intravenous rt-PA at our institute than those enrolled, but other background factors, including initial NIHSS score, were comparable between the two groups.