Share

June 04, 2019; 92 (23) Article

Influence of occlusion site and baseline ischemic core on outcome in patients with ischemic stroke

Huiqiao Tian, Mark W. Parsons, Christopher R. Levi, Longting Lin, Richard I. Aviv, Neil J. Spratt, Kenneth S. Butcher, Min Lou, Timothy J. Kleinig, Andrew Bivard
First published May 1, 2019, DOI: https://doi.org/10.1212/WNL.0000000000007553
Full PDF
Short Form
Citation
Permissions

Make Comment

See Comments

Downloads
271

Share

Abstract

Objective We assessed patient clinical outcomes based on occlusion location, focusing on distal occlusions to understand if occlusion location was an independent predictor of outcome, and tested the relationship between occlusion location and baseline ischemic core, a known predictor of modified Rankin Scale (mRS) score at 90 days.

Methods We analyzed a prospectively collected cohort of thrombolysis-eligible ischemic stroke patients from the International Stroke Perfusion Imaging Registry who underwent multimodal CT pretreatment. For the primary analysis, logistic regression was used to predict the effect of occlusion location and ischemic core on the likelihood of excellent (mRS 0–1) and favorable (mRS 0–2) 90-day outcomes.

Results This study included 945 patients. The rates of excellent and favorable outcome in patients with distal occlusion (M2, M3 segment of middle cerebral artery, anterior cerebral artery, and posterior cerebral artery) were higher than M1 occlusions (mRS 0%–1%, 55% vs 37%; mRS 0%–2%, 73% vs 50%, p < 0.001). Vessel occlusion location was not a strong predictor of outcomes compared to baseline ischemic core (area under the curve, mRS 0–1, 0.64 vs 0.83; mRS 0–2, 0.70 vs 0.86, p < 0.001). There was no interaction between occlusion location and ischemic core (interaction coefficient 1.00, p = 0.798).

Conclusions Ischemic stroke patients with a distal occlusion have higher rate of excellent and favorable outcome than patients with an M1 occlusion. The baseline ischemic core was shown to be a more powerful predictor of functional outcome than the occlusion location, but the relationship between ischemic core and outcome does not different by occlusion locations.

Glossary

ACA=
anterior cerebral artery;
CI=
confidence interval;
CTA=
CT angiography;
CTP=
perfusion CT;
HERMES=
Highly Effective Reperfusion Evaluated in Multiple Endovascular Stroke Trials;
ICA=
internal carotid artery;
INSPIRE=
International Stroke Perfusion Imaging Registry;
IQR=
interquartile range;
MCA=
middle cerebral artery;
mRS=
modified Rankin Scale;
mTICI=
modified thrombolysis in cerebral infarction;
NIHSS=
NIH Stroke Scale;
OR=
odds ratio;
PCA=
posterior cerebral artery

Recently, 8 clinical trials1,–,8 have provided evidence that patients with a large vessel occlusion significantly benefit from endovascular therapy over best medical care in the early time window,9 and 2 have subsequently shown benefit beyond 6 hours.3,4 As a result, endovascular therapy has become a standard of care of patients with a large vessel occlusion. However, alteplase remains the most widely used reperfusion treatment for ischemic stroke, especially for patients with occlusions beyond the M1 segment of the middle cerebral artery.

The increased clinical application of endovascular therapy has had a substantial effect on ongoing clinical trials of thrombolysis (and planning for new thrombolytic trials). Hence, clinical trials of thrombolysis are now less likely to enroll patients with proximal large vessel occlusion unless the trial design allows thrombectomy as a bridging or salvage therapy. This change in case mix for trials will undoubtedly affect their power, sample size, and potentially their ability to detect benefits in clinical outcome. Although a meta-analysis from the Highly Effective Reperfusion Evaluated in Multiple Endovascular Stroke Trials (HERMES) collaboration10 and a combined analysis of registry and randomized trials11 of mechanical thrombectomy in large vessel occlusion suggested possible benefit in the subgroup of patients with M2 occlusion, however, HERMES did not have the power to confirm any benefit or harm in patients with M2 occlusions. In addition, another meta-analysis suggested an increased risk of hemorrhage in M2 occlusions when treated with endovascular thrombectomy.12 Hence, there remains uncertain in subgroup and even greater uncertainty with more distal occlusions. The current and ongoing clinical trials of thrombolysis are likely to see less enrollment of patients with a large vessel occlusion, and possibly a decrease in patients with an M2 occlusion, since there is still the question of a treatment benefit with endovascular therapy in these patients.

Given the changing landscape of stoke therapy, we aimed to assess occlusion site frequency and the clinical and perfusion imaging profiles of patients presenting with differing occlusion sites and assess the relationships between vessel occlusion location and patient outcome. We hoped to provide information of potential relevance to the design of future studies aiming to detect reperfusion treatment response in patient populations that only include distal occlusions.

Methods

Patients

Ischemic stroke patients presenting to a participating hospital within 12 hours of symptom onset who underwent baseline multimodal CT before reperfusion therapy in 5 centers (John Hunter Hospital, New South Wales; Royal Adelaide Hospital, South Australia; Second Affiliated Hospital of Zhejiang University, Hangzhou, China; Sunnybrook Health Science Centre, Toronto; and University of Alberta, Edmonton, Canada) between 2011 and 2016 were prospectively recruited for the International Stroke Perfusion Imaging Registry (INSPIRE). The baseline multimodal CT included noncontrast CT, perfusion CT (CTP), and CT angiography (CTA). Follow-up imaging was performed at 24 hours poststroke, with MRI, which included perfusion-weighted imaging, diffusion-weighted imaging, or magnetic resonance angiography, or multimodal CT. Clinical stroke severity was assessed using the NIH Stroke Scale (NIHSS) at the 2 imaging time points.

IV thrombolysis was used to treat eligible stroke patients according to local guidelines, and patients with proximal large vessel occlusion underwent intra-arterial therapy if the service was available. Functional outcome was assessed using modified Rankin Scale (mRS) score at 90 days after stroke. We restricted our analysis by only including patients potentially eligible for thrombolysis therapy who presented to the hospital within 4.5 hours of onset. Patients were not eligible for alteplase therapy if they had clinical and noncontrast CT contraindications (e.g., intracranial hemorrhage, mild or improving clinical deficit, extensive early ischemic change, and major comorbidities). All patients had multimodal CT before a treatment decision, and this information was used by the treating neurologist to assist the decision on alteplase eligibility.13,14 Patients with an unfavorable pattern on baseline CT perfusion (small or no identifiable volume of perfusion lesion, large ischemic core, lack of perfusion lesion–core mismatch, or lack of vessel occlusion) would have been taken into consideration for treatment even if the patient was eligible on standard clinical grounds.

It was required that patients in the INSPIRE registry have a complete baseline NIHSS score, 90-day mRS score, and baseline CTP and CTA imaging. Patients with vertebral or basilar occlusion were excluded.

Standard protocol approvals, registrations, and patient consents

All patients provided written consent form, and the INSPIRE study was approved by the Hunter New England Health District ethics committees in accordance with Australian NHMRC guidelines (HNEHREC reference no.: 11/08/17/4.01; NSW HREC reference no.: HREC/11/HNE/287; SSA reference no.: SSA/11/HNE/314).

Imaging acquisition

Baseline CT imaging included brain noncontrast CT, CTP, and CTA, obtained with different CT scanners (64, 128, 256, or 320 detectors, with Toshiba [Tokyo, Japan], Siemens [Munich, Germany], or GE [Cleveland, OH] scanners). Axial coverage ranged from 40 to 160 mm. Detailed procedural and scanner details are summarized in table 1.

View this table:
Table 1

Perfusion CT (CTP) acquisition protocols for the 5 International Stroke Perfusion Imaging Registry (INSPIRE) sites involved in this study

All stroke patients, regardless of treatment, underwent follow-up MRI using 1.5T or 3T scanner at 24 hours after the acute imaging. The MRI protocol included an axial gradient-echo T2*-weighted series, diffusion-weighted MRI, perfusion-weighted image, MRI time of flight angiography, as well as fluid-attenuated inversion recovery imaging. Follow-up multimodal CT was performed when MRI was not available in centers or the patient had contraindications to MRI.

Assessment of baseline and follow-up images

All acute CTA images were assessed for the location and the severity of vessel occlusion; follow-up MRI was assessed for the recanalization status by central reading group, with each case being read by 2 experienced readers. The vessel occlusion location was grouped as follows: internal carotid artery (ICA, including ICA terminus T or L type, tandem occlusions), M1 (prebifurcation segment), M2 (from postbifurcation segment in the Sylvian fissure), M3 segment of the middle cerebral artery (opercular branches of MCA), anterior cerebral artery (ACA), and posterior cerebral artery (PCA). Location of perfusion lesion was recorded as frontal operculum, insula, anterior temporal lobe, posterior temporal lobe, internal capsule, anterior cerebral artery territory, and posterior cerebral artery territory. The baseline tissue perfusion status was assessed on CTA using modified thrombolysis in cerebral infarction (mTICI) scale15: normal = mTICI 3, partial = mTICI 2a/2b, or complete = mTICI 1/0. The collateral vessel flow status was assessed using 3-point Miteff scale,16 defined as good, moderate, or poor depending on the extent of contrast visualized distal to vessel occlusion on CTA. Vascular recanalization status was assessed using mTICI scale from follow-up imaging in patients who had partial or complete vessel occlusion at baseline: complete recanalization = mTICI 3, partial recanalization = mTICI 2a/2b, no recanalization = mTICI 1/0. The mTICI scale has a moderate interobserver reliability on single-phase CTA17; therefore, we also provided reperfusion status on follow-up imaging as alternate data.

The CTP images were analyzed retrospectively with commercial software (MIStar; Apollo Medical Imaging Technology, Melbourne, Australia), using single-value deconvolution with delay and dispersion method.18 The threshold for perfusion lesion set to relative delay time >3 seconds and baseline ischemic core was defined by relative cerebral blood flow <30% within the delay time >3 seconds lesion.19 Penumbral volume was calculated from baseline perfusion lesion minus baseline ischemic core volume. Baseline mismatch ratio was the ratio of baseline perfusion lesion to baseline ischemic core volume.13 Patients with target mismatch were classified if they had penumbral volume >15 mL, ischemic core volume <70 mL, and mismatch ratio >1.8.20 The final infarct volume was delineated based on signal intensity and highlighted using an area of interest tool on 24-hour diffusion-weighted MRI. The infarct growth volume was calculated from final infarct volume minus baseline ischemic core. Penumbral salvage was calculated from baseline perfusion lesion minus the final infarct lesion. Complete reperfusion was defined as perfusion lesion reduction >80% from baseline to 24-hour CTP, or 24-hour perfusion lesion of zero (if <80%).

Hemorrhagic transformation was assessed on follow-up MRI or CT based on European Cooperative Acute Stroke Study classifications as hemorrhagic infarction type 1 (small petechiae) or type 2 (more confluent petechiae within the infarcted area) or parenchymal hematoma type 1 (≤30% of the infarcted area with mild space-occupying effect) or type 2 (>30% of the infarcted area with significant space-occupying effect).21 Patients with symptomatic intracerebral hemorrhage were those who had parenchymal hematoma type 2 with NIHSS change ≥4 from baseline to 24 hours.

Statistical analysis

Statistical analyses were programmed using STATA (v13.0; StataCorp, College Station, TX). This study aimed to assess occlusion site frequency and clinical and perfusion imaging profiles at differing occlusion sites,1 and to examine the relationships of occlusion location, baseline ischemic core volume, and 90-day mRS outcome.

Patients' baseline and follow-up imaging and clinical characteristics by occlusion groups were presented as mean and SD, median and interquartile range (IQR), or number with percentages. We compared the imaging and clinical variables of treatment groups at each vessel occlusion site and alteplase-treated patients between vessel occlusion groups. The statistical differences were assessed using 2 independent samples t test, Wilcoxon-Mann-Whitney test, χ2 test, or Fisher exact test where applicable. An additional comparison between treatment groups in each vessel occlusion location was applied to patients who met target mismatch criteria.

To assess the predictors of patient outcome and the relationships between predictors and the outcome, we performed 5 logistic regression models in patients who were treated with alteplase (models 1–5). The outcome variables were dichotomized mRS at 90 days: excellent outcome was defined as mRS 0–1 vs 2–6, favorable outcome was defined as mRS 0–2 vs 3–6, and poor outcome was defined as mRS 5–6 vs 0–4. Vessel occlusion locations were grouped to (1) M1 (reference group), (2) ICA, and (3) distal occlusions including M2, M3, ACA, and PCA. First, univariate logistic regression model was used to assess the predictive ability of (1) vessel occlusion locations; (2) volume of baseline ischemic core (a known predictor of patient outcome); and (3) target mismatch (dichotomized at patient with vs without target mismatch) on clinical response outcomes (model 1–3). Then, a backward stepwise multivariate logistic regression analysis was performed; variables were patient age, time of onset to imaging, baseline NIHSS score, the volume of baseline ischemic core, target mismatch, and vessel occlusion locations (model 4a). Also, we performed 2 additional backward stepwise multivariate logistic regression models in a subgroup of patients who had a distal or an M1 vessel occlusion (model 4b) and a subgroup of patients who had an ICA or an M1 vessel occlusion (model 4c). Models 4b–4c included all variables in model 4a, and collateral status was added to model 4c. To determine if vessel occlusion location or target mismatch alter the relationship between the ischemic core and the outcome, we assessed possible interactions between occlusion locations and baseline ischemic core volume, as well as target mismatch and baseline ischemic core volume (model 5); this model also included all variables mentioned in model 4. Finally, to assess potential treatment effect on different occlusion sites in this nonrandomized study, we tested the interaction of occlusion sites and alteplase treatment using logistic regression analysis (model 6). Areas under the receiver operating characteristic curves were used to discriminate the predictive models. Logistic regression results were presented with odds ratios (ORs) with 95% confidence intervals (CIs) and p values. Values of p < 0.05 were considered statistically significant.

We used Spearman correlation to assess the relationships between the baseline ischemic core, occlusion sites, and the mRS outcomes at 90 days.

As a secondary analysis, we calculated the power to detect the difference at 5% significance level in patients with distal or M1 occlusion between alteplase-treated and untreated patients. We calculated the sample size required for 80% power using the current treatment outcomes. The same power and sample size estimations were applied to a subgroup of patients who met target mismatch criteria. The number of patients who need to be screened to find an eligible case was also calculated based on the current study: 35% of the total patients had a distal vessel occlusion, and 59% met target mismatch criteria; 48% of the total patients had an M1 vessel occlusion, 77% of them met target mismatch criteria.

Data availability statement

Anonymized data used in this study will be shared on request from any qualified investigator.

Results

A total of 1,826 patients from the INSPIRE registry were assessed: 326 were excluded because they were not eligible for alteplase treatment (outside 4.5 hours of symptom onset, premorbid disability, or major early ischemic change on noncontrast CT), 82 were excluded because of imaging artifacts, 390 were excluded because of unidentifiable occlusion location (no visible vessel occlusion and no perfusion lesion), and 83 patients underwent endovascular therapy. Of the remaining 945 patients who were included in the analysis, 157 (17%) had an ICA occlusion, 456 (49%) had an M1 occlusion, 174 (18%) had an M2 occlusion, 87 (9%) had an M3 occlusion, 28 (3%) had an ACA occlusion, and 43 (4%) had a PCA occlusion.

Alteplase-treated vs untreated comparisons

At baseline, alteplase-treated and untreated patients had a similar time of onset to imaging, baseline NIHSS score, volume of perfusion lesion, and ischemic core within each vessel occlusion group. Only patients with M3 occlusions had higher baseline NIHSS in the treated group compared to untreated patients (alteplase: 8, IQR 5–12; untreated: 5, IQR 3–9, p = 0.040), and patients with PCA occlusions had larger baseline ischemic core in the treated group (alteplase: 6 mL, IQR 2–17 mL; untreated: 3 mL, IQR 1–7 mL, p = 0.035, table 2).

View this table:
Table 2

Comparison of baseline clinical and imaging characteristics at differing occlusion locations

On follow-up of the M1 occlusion subgroup, outcome differences between alteplase-treated and untreated patients were evident; alteplase-treated patients showed larger mean salvaged penumbral volumes (alteplase: 58 mL, SD 67 mL; untreated: 37 mL, SD 76, p = 0.009), lower median 24-hour NIHSS (alteplase: 8, IQR 4–16; untreated: 13, IQR 8–18, p < 0.001, table 3), higher rate of mRS 0–1 (alteplase: 37%, untreated: 22%, p = 0.006), and lower rate of mRS 5–6 (alteplase: 21%, untreated: 36%, p = 0.001, table 4). With respect to the risk of hemorrhage after alteplase treatment, thrombolyzed patients with M1s had larger proportion of hemorrhagic transformation than untreated patients (alteplase: 26%, untreated: 7.9%, p < 0.001), and thrombolyzed M1s had higher rates of symptomatic intracerebral hemorrhage (alteplase: 5.2%, untreated: 0%, p = 0.027, table 3).

View this table:
Table 3

Comparison of follow-up clinical and imaging characteristics between thrombolysis-eligible alteplase-treated and untreated patients at differing occlusion locations

View this table:
Table 4

Comparison of 90-day patient outcomes on the modified Rankin Scale (mRS) at differing occlusion locations, with a subgroup of patients who met target mismatch criteria

Patient 90-day outcomes were not significant between treatment groups in ICA occlusions or in distal vessel occlusions. Interestingly, however, the subgroup of alteplase-treated patients with M2 occlusions who also met target mismatch criteria had better outcomes compared to untreated patients (mRS 0–1, alteplase 58%, untreated 28%, p = 0.019; mRS 0–2, alteplase 74%, untreated 50%, p = 0.039, table 4).

Alteplase-treated distal occlusions vs M1s

Significant differences were found in comparisons of alteplase-treated patients with M1 vs distal occlusions. Patients with distal vessel occlusions treated with alteplase had lower median NIHSS score compared to alteplase-treated M1 occlusion patients (distal: 10, IQR 6–13; M1: 15, IQR 12–18, p < 0.001) and smaller ischemic core volumes (distal: 8 mL, IQR 3–14 mL; M1: 16 mL, IQR 7–35, p < 0.001, table 5). On follow-up, patients with distal occlusions treated with alteplase had higher rate of recanalization (distal: 66%, M1: 49%, p = 0.005), higher rate of excellent outcome (distal: 54%, M1: 37%, p < 0.001), and lower rate of poor outcome at 90 days compared to alteplase-treated patients with M1 occlusions (distal: 5%, M1: 21%, p < 0.001). Finally, alteplase-treated distal occlusions had lower rate of all forms of hemorrhagic transformation (distal: 14%, M1: 26%, p < 0.001) and symptomatic intracerebral hemorrhage than those who had M1 occlusions treated with alteplase (distal: 0.5%, M1: 5.2%, p = 0.003, table 5).

View this table:
Table 5

Comparison between baseline and outcome variables for patients who were treated with alteplase

An interaction was observed between vessel occlusion sites and alteplase treatment (p = 0.016, model 6 in table 6).

View this table:
Table 6

Logistic regression analysis of the effect of clinical and imaging variables on the patient functional outcomes in patients who were treated with alteplase (model 1–5) and patients with and without alteplase treatment (model 6)

Logistic regression models on alteplase-treated patients

Baseline ischemic core volume was a stronger univariate predictor of patient outcome at day 90 compared to occlusion location or target mismatch (mRS 0–1, AUCcore 0.83, AUClocation 0.64, AUCtarget MM 0.57, p < 0.001; mRS 0–2, AUCcore 0.86, AUClocation 0.70, AUCtarget MM 0.57, p < 0.001; mRS 5–6, AUCcore 0.87, AUClocation 0.74, AUCtarget MM 0.63, p < 0.001, model 1–3, table 6). In addition, ischemic core was the only independent imaging predictor that remained in logistic models when alteplase-treated patients were assessed (mRS 0–1: OR 0.91, 95% CI 0.89–0.93, p < 0.001; mRS 0–2: OR 0.91, 95% CI 0.90–0.93, p < 0.001; mRS 5–6: OR 1.04, 95% CI 1.03–1.06, p < 0.001, model 4a), and in models of a subgroup of alteplase-treated patients who had an M1 or distal vessel occlusion (model 4b). However, both location site and ischemic core were imaging predictors of outcomes in subgroups of patients with M1 or ICA vessel occlusion (model 4c). A large ischemic core volume is correlated with proximal vessel occlusion (Spearman ρ = 0.43, p < 0.001). Furthermore, no significant interaction was found between occlusion location and baseline ischemic core or between target mismatch and baseline ischemic core (model 5, table 6).

Power calculation for distal occlusions

The penumbral salvage at 24 hours provided a power of 72% with the current sample size for detecting differences between treatment groups in patients with distal occlusions eligible for alteplase treatment. Also, the patient group with target mismatch and distal occlusions demonstrated power of 76% to detect differences in treatment groups. Power for detecting differences between treatment groups in patients with distal occlusions who were eligible for alteplase treatment was low (9%, 8%, and 35% for mRS 0–1, mRS 0–2, and mRS 5–6 at day 90, respectively; table 7).

View this table:
Table 7

Mock power and sample size calculation for all patients who were eligible for alteplase treatment and for patients who also met target mismatch criteria

Discussion

This study analyzed ischemic stroke patients' clinical and imaging characteristics to assess the influence of vessel occlusion location on baseline characteristics as well as imaging and clinical outcomes. This analysis was in the context of the new standard of care for large vessel occlusions (endovascular thrombectomy), and thus sought to identify the effect on ongoing or future clinical trials of IV thrombolysis. We have observed that patients with an M2 or other distal occlusions have a better natural history than patients with an M1 occlusion. In addition, our data show that a difference in outcome with alteplase was observed only among the M1 occlusion group. However, due to the nonrandomized design, the reasons for decisions against treatment were not captured in the registry data. Also, the observed differences in outcome might be the consequence of many factors not available for analysis and whether alteplase treatment alone can be confidently credited with the observed differences is open to some doubt.

The incidence of patients with a distal vessel occlusion in the INSPIRE cohort was fairly high (35%), and we showed that patients with distal vessel occlusions have significantly different natural history compared to patients with a proximal occlusion. Accordingly, a trial of reperfusion excluding large vessel occlusions patients would potentially only be able to enroll 35% of thrombolysis patients due to the remainder going on to thrombectomy. Importantly, the power required to detect a clinically meaningful change in patients with a distal vessel occlusion is significantly reduced compared to M1 occlusions: power at best being 35%–51% when looking at long-term functional outcomes. The low power (9% for detecting treatment difference in excellent outcomes) in our study data in patients with distal occlusion not only suggests caution in interpreting the outcomes between treatment groups, but also supports the poor discriminatory characteristics of mRS for assessing reperfusion therapy response in mild strokes as was demonstrated in the Norwegian Tenecteplase Stroke Trial (NOR-TEST).22 However, in patients with distal occlusions, the use of surrogate outcomes such as imaging biomarkers or early clinical outcomes can provide enhanced power, in our study reaching 72%–76%. In addition, selection using imaging biomarkers such as target mismatch also provides increased study power. Therefore, trialists now need to consider what effect this will have on such studies and whether primary outcomes require changing, or sample sizes require adjusting. In this registry-based and nonrandomized study, we observed that the differences in 90-day mRS outcomes between treatment groups in patients with distal occlusions are too small to detect any significance in the assembled cohort. For example, 75% of alteplase-treated patients with distal occlusion achieved an mRS 0–2, compared to 78% of untreated patients. This 3% difference between treatment groups from this study would therefore require a sample size of 3,134. That is, 3,134 patients with distal occlusions were required to show the rates of patients who achieved mRS 0–2 between treatment groups were statistically different, which may be unrealistic when only assessing 35% of a clinical case mix. Importantly, we found the difference of outcomes between the alteplase-treated and untreated group is more significant in patients with M2 occlusion who met target mismatch criteria, therefore, applying target mismatch as a patient selection criterion may potentially increase the power of the study, but this would also reduce the number of potentially eligible patients.

Although patients with target mismatch were more likely to benefit from alteplase treatment, target mismatch was a weak independent predictor of 90-day outcome. The strong positive correlation between the ischemic core and the occlusion site may have caused the lack of significance of occlusion location as an outcome predictor when core volume is included. That is, patient outcomes at 90 days are more driven by baseline ischemic core than occlusion location or target mismatch. Furthermore, there is no evidence of significant interaction between ischemic core and occlusion location or target mismatch. These results indicate that smaller ischemic core predicts better clinical outcome and that this association does not differ by different occlusion location or whether the patient met target mismatch criteria. This is presumably because occlusion location does not take into account other factors such as collateral flow. However, it is important to note that the baseline ischemic core volume is influenced by the physiology and as such a smaller ischemic core volume is associated with a good collateral status23,24 or a better vessel patency status (partial vessel occlusion or spontaneous recanalization at baseline) and as such the baseline core volume may be a surrogate for a number of moderately powerful prognostic markers of patient outcome.

We observed the highest rate of hemorrhage was found in alteplase-treated patients with M1 occlusions, who also had a relatively large volume of baseline ischemic core and salvaged penumbra. This suggests a higher chance of achieving a favorable clinical outcome comes with a higher risk of hemorrhage in M1 occlusion patients25; however, patients with distal occlusions treated with alteplase had high rate of the favorable clinical outcome but low risk of hemorrhage compared to patients with M1 occlusions. This difference could be explained by the distinction of the baseline ischemic core volume of the 2 occlusion location groups, and it is possible that a large baseline ischemic core may associate with a higher risk of hemorrhage.26

Previous studies used baseline or follow-up noncontrast CT and diffusion-weighted imaging to quantify the ischemic lesion volume when investigating the association of ischemic lesion volume and the follow-up functional outcome.27,–,30 The noncontrast CT provides limited information on patient selection for trials and the prediction of the patient clinical outcome because it cannot accurately rule out stroke mimics or patients who may be harmed by thrombolysis.14 However, CT perfusion has been reported to be more reliable than the noncontrast CT13,31 for the measurement of ischemic core volume, and also proved to be a powerful predictor of patient functional outcome.25,32 Therefore, our results are useful for the treating physician to predict the outcome based on the baseline ischemic core volume and the location of the vessel occlusion, as soon as the baseline multimodal CT imaging was performed.

The study has some limitations that require acknowledgment. (1) Data presented in the study are registry-based and nonrandomized. Many factors (including the imaging findings) were inevitably considered in making a treatment decision in favor of alteplase, which may confound the differences observed in the study. (2) It is an observational study, and there are some differences in baseline characteristics in alteplase-treated and untreated patients across all the study groups. The baseline differences in the treatment groups likely contributed to the no significant improvement observed in treated patients with distal vessel occlusions compared to untreated patients; however, given the excellent natural history of patients with distal occlusions, measuring clinical improvement in this study group is challenging. (3) Our data are from an international registry, and their country's guidelines influenced the local investigator discretion. Also, due to the database design, we were unable to record all information that led to the treating neurologists withholding treatment in such patients, and the unmeasured confounding variables, such as patient frailty or undocumented comorbidities may have influenced the results; for example, the observed selection bias and the differences at baseline in patients with an M3 or PCA occlusion possibly because the investigators may have treated the patients based on their clinical presentation. (4) INSPIRE encourages enrolling consecutive patients, and the requirement of baseline multimodal CT and follow-up MRI, along with clinical data from several timepoints, limited the number of included patients, and this may bias our study cohort against generalizability. (5) The determination of CTA weighting was not available because the imaging data are from a multicenter international registry; however, the CTA weights may not significantly influence the collateral grades based on a recent study.33 (6) Single-phase CTA provides limited spatial and temporal resolution to visualize M3s. To improve the accuracy of vessel occlusion locations, we also used CT perfusion maps to localize. Then, we grouped the MCA M2 or M3, ACA, and PCA into “distal occlusions” in the final regression model to avoid misclassifying M3 occlusions.

The clinical and imaging characteristics of patients with different vessel occlusion locations were analyzed, to find that patients with distal vessel occlusions had better baseline clinical and imaging characteristics and follow-up outcome and this is statistically different compared to patients with M1 occlusion. There is a need to test other thrombolytic agents that have higher efficacy compared to alteplase, to enhance the power to detect the difference between treatment groups in patients with more distal vessel occlusions. Also, the ongoing thrombolysis trials should consider potential influences of endovascular treatment to bring to routine care and adjust their design in order to compensate for the change in case mix that will occur as the rate of thrombectomy increases. The data presented can inform new trial designs for IV thrombolysis, but also provide important information for neurointerventionalists who are using thrombolysis as an accompanying therapy during mechanical thrombectomy. As risks and benefits of the latter technique also depend on the occlusion site, efficacy of IV thrombolysis should be taken into consideration for every stroke patient with large vessel occlusion in the anterior circulation. Furthermore, our models have shown that the baseline ischemic core volume has a stronger effect on patient outcome than does occlusion location. That is, the larger the ischemic core volume, the worse the patient functional outcome; however, the occlusion location limits the maximum size of the ischemic core. The lack of interaction between ischemic core volume and the occlusion location is likely due to a strong mediator such as collateral flow status.

Study funding

Supported by a National Health and Medical Research Council Australia Partnership Project Grant (APP1013719). Christopher R. Levi is funded by a National Health and Medical Research Council Practitioner Fellowship (APP 1043913). Neil J. Spratt is funded by a National Health and Medical Research Council/National Heart Foundation co-funded Career Development/Future Leader Fellowship (APPS1110629/100827). Longting Lin is supported by The Science and Industry Endowment Fund (SIEF) STEM + Business Fellowship.

Disclosure

The authors report no disclosures relevant to the manuscript. Go to Neurology.org/N for full disclosures.

Appendix Authors

Table
Table

Footnotes

  • Go to Neurology.org/N for full disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the article.

  • Editorial, page 1075

  • Received July 24, 2018.
  • Accepted in final form January 29, 2019.

References

  1. 1.↵
    1. Berkhemer OA,
    2. Fransen PSS,
    3. Beumer D, et al
    . A randomized trial of intraarterial treatment for acute ischemic stroke. N Engl J Med 2015;372:11–20.
    OpenUrlCrossRefPubMed
  2. 2.↵
    1. Campbell BCV,
    2. Mitchell PJ,
    3. Kleinig TJ, et al
    . Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med 2015;372:1009–1018.
    OpenUrlCrossRefPubMed
  3. 3.↵
    1. Goyal M,
    2. Demchuk AM,
    3. Menon BK, et al
    . Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med 2015;372:1019–1030.
    OpenUrlCrossRefPubMed
  4. 4.↵
    1. Jovin TG,
    2. Chamorro A,
    3. Cobo E, et al
    . Thrombectomy within 8 hours after symptom onset in ischemic stroke. N Engl J Med 2015;372:2296–2306.
    OpenUrlCrossRefPubMed
  5. 5.↵
    1. Saver JL,
    2. Goyal M,
    3. Bonafe A, et al
    . Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N Engl J Med 2015;372:2285–2295.
    OpenUrlCrossRefPubMed
  6. 6.↵
    1. Mocco J,
    2. Zaidat OO,
    3. von Kummer R, et al
    . Aspiration thrombectomy after intravenous alteplase versus intravenous alteplase alone. Stroke 2016;47:2331–2338.
    OpenUrlAbstract/FREE Full Text
  7. 7.↵
    1. Bracard S,
    2. Ducrocq X,
    3. Mas JL, et al
    . Mechanical thrombectomy after intravenous alteplase versus alteplase alone after stroke (THRACE): a randomised controlled trial. Lancet Neurol 2016;15:1138–1147.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Muir KW,
    2. Ford GA,
    3. Messow CM, et al
    . Endovascular therapy for acute ischaemic stroke: the Pragmatic Ischaemic Stroke Thrombectomy Evaluation (PISTE) randomised, controlled trial. J Neurol Neurosurg Psychiatry 2017;88:38–44.
    OpenUrlAbstract/FREE Full Text
  9. 9.↵
    1. Powers WJ,
    2. Derdeyn CP,
    3. Biller J, et al
    . 2015 American Heart Association/American Stroke Association focused update of the 2013 guidelines for the early management of patients with acute ischemic stroke regarding endovascular treatment: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2015;46:3020–3035.
    OpenUrlAbstract/FREE Full Text
  10. 10.↵
    1. Goyal M,
    2. Menon BK,
    3. van Zwam WH, et al
    . Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials. Lancet 2016;387:1723–1731.
    OpenUrlCrossRefPubMed
  11. 11.↵
    1. Coutinho JM,
    2. Liebeskind DS,
    3. Slater LA, et al
    . Mechanical thrombectomy for isolated M2 occlusions: a post hoc analysis of the STAR, SWIFT, and SWIFT PRIME studies. AJNR Am J Neuroradiol 2016;37:667–672.
    OpenUrlAbstract/FREE Full Text
  12. 12.↵
    1. Saber H,
    2. Narayanan S,
    3. Palla M, et al
    . Mechanical thrombectomy for acute ischemic stroke with occlusion of the M2 segment of the middle cerebral artery: a meta-analysis. J Neurointerv Surg 2018;10:620–624.
    OpenUrlAbstract/FREE Full Text
  13. 13.↵
    1. Bivard A,
    2. Levi C,
    3. Krishnamurthy V, et al
    . Perfusion computed tomography to assist decision making for stroke thrombolysis. Brain 2015;138:1919–1931.
    OpenUrlCrossRefPubMed
  14. 14.↵
    1. Bivard A,
    2. Lou M,
    3. Levi CR, et al
    . Too good to treat? Ischemic stroke patients with small computed tomography perfusion lesions may not benefit from thrombolysis. Ann Neurol 2016;80:286–293.
    OpenUrl
  15. 15.↵
    1. Tomsick T,
    2. Broderick J,
    3. Carrozella J, et al
    . Revascularization results in the Interventional Management of Stroke II Trial. AJNR Am J Neuroradiol 2008;29:582–587.
    OpenUrlAbstract/FREE Full Text
  16. 16.↵
    1. Miteff F,
    2. Levi CR,
    3. Bateman GA,
    4. Spratt N,
    5. McElduff P,
    6. Parsons MW
    . The independent predictive utility of computed tomography angiographic collateral status in acute ischaemic stroke. Brain 2009;132:2231–2238.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Mair G,
    2. von Kummer R,
    3. Adami A, et al
    . Observer reliability of CT angiography in the assessment of acute ischaemic stroke: data from the Third International Stroke Trial. Neuroradiology 2015;57:1–9.
    OpenUrl
  18. 18.↵
    1. Bivard A,
    2. Levi C,
    3. Spratt N,
    4. Parsons M
    . Perfusion CT in acute stroke: a comprehensive analysis of infarct and penumbra. Radiology 2013;267:543–550.
    OpenUrlCrossRefPubMed
  19. 19.↵
    1. Bivard A,
    2. Levi C,
    3. Krishnamurthy V, et al
    . Defining acute ischemic stroke tissue pathophysiology with whole brain CT perfusion. J Neuroradiol 2014;41:307–315.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. Lansberg MG,
    2. Straka M,
    3. Kemp S, et al
    . MRI profile and response to endovascular reperfusion after stroke (DEFUSE 2): a prospective cohort study. Lancet Neurol 2012;11:860–867.
    OpenUrlCrossRefPubMed
  21. 21.↵
    1. Hacke W,
    2. Kaste M,
    3. Fieschi C, et al
    . Randomised double-blind placebo-controlled trial of thrombolytic therapy with intravenous alteplase in acute ischaemic stroke (ECASS II): Second European-Australasian Acute Stroke Study Investigators. Lancet 1998;352:1245–1251.
    OpenUrlCrossRefPubMed
  22. 22.↵
    1. Logallo N,
    2. Novotny V,
    3. Assmus J, et al
    . Tenecteplase versus alteplase for management of acute ischaemic stroke (NOR-TEST): a phase 3, randomised, open-label, blinded endpoint trial. Lancet Neurol 2017;16:781–788.
    OpenUrl
  23. 23.↵
    1. Vagal A,
    2. Menon BK,
    3. Foster LD, et al
    . Association between CT angiogram collaterals and CT perfusion in the Interventional Management of Stroke III Trial. Stroke 2016;47:535–538.
    OpenUrlAbstract/FREE Full Text
  24. 24.↵
    1. Kawano H,
    2. Bivard A,
    3. Lin L, et al
    . Relationship between collateral status, contrast transit, and contrast density in acute ischemic stroke. Stroke 2016;47:742–749.
    OpenUrlAbstract/FREE Full Text
  25. 25.↵
    1. Chen C,
    2. Parsons MW,
    3. Clapham M, et al
    . Influence of penumbral reperfusion on clinical outcome depends on baseline ischemic core volume. Stroke 2017;48:2739–2745.
    OpenUrlAbstract/FREE Full Text
  26. 26.↵
    1. Yassi N,
    2. Parsons MW,
    3. Christensen S, et al
    . Prediction of poststroke hemorrhagic transformation using computed tomography perfusion. Stroke 2013;44:3039–3043.
    OpenUrlAbstract/FREE Full Text
  27. 27.↵
    1. Bucker A,
    2. Boers AM,
    3. Bot JCJ, et al
    . Associations of ischemic lesion volume with functional outcome in patients with acute ischemic stroke: 24-hour versus 1-week imaging. Stroke 2017;48:1233–1240.
    OpenUrlAbstract/FREE Full Text
  28. 28.↵
    1. Yoo AJ,
    2. Chaudhry ZA,
    3. Nogueira RG, et al
    . Infarct volume is a pivotal biomarker after intra-arterial stroke therapy. Stroke 2012;43:1323–1330.
    OpenUrlAbstract/FREE Full Text
  29. 29.↵
    1. Vogt G,
    2. Laage R,
    3. Shuaib A,
    4. Schneider A
    . Initial lesion volume is an independent predictor of clinical stroke outcome at day 90: an analysis of the Virtual International Stroke Trials Archive (VISTA) Database. Stroke 2012;43:1266–1272.
    OpenUrlAbstract/FREE Full Text
  30. 30.↵
    1. Ernst M,
    2. Boers AMM,
    3. Aigner A, et al
    . Association of computed tomography ischemic lesion location with functional outcome in acute large vessel occlusion ischemic stroke. Stroke 2017;48:2426–2433.
    OpenUrlAbstract/FREE Full Text
  31. 31.↵
    1. Wintermark M,
    2. Flanders AE,
    3. Velthuis B, et al
    . Perfusion-CT assessment of infarct core and penumbra: receiver operating characteristic curve analysis in 130 patients suspected of acute hemispheric stroke. Stroke 2006;37:979–985.
    OpenUrlAbstract/FREE Full Text
  32. 32.↵
    1. Bivard A,
    2. Levi C,
    3. Lin L, et al
    . Validating a predictive model of acute advanced imaging biomarkers in ischemic stroke. Stroke 2017;48:645–650.
    OpenUrlAbstract/FREE Full Text
  33. 33.↵
    1. Bisson DA,
    2. Mahmoudian D,
    3. Shatil AS, et al
    . Single-phase CT angiography: collateral grade is independent of scan weighting. Neuroradiology 2019;61:19–28.
    OpenUrl
View Abstract

Letters: Rapid online correspondence

No comments have been published for this article.
Comment

REQUIREMENTS

If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org

Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.

If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.

Submission specifications:

  • Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
  • Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
  • Submit only on articles published within 6 months of issue date.
  • Do not be redundant. Read any comments already posted on the article prior to submission.
  • Submitted comments are subject to editing and editor review prior to posting.

More guidelines and information on Disputes & Debates

Compose Comment

More information about text formats

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Author Information
NOTE: The first author must also be the corresponding author of the comment.
First or given name, e.g. 'Peter'.
Your last, or family, name, e.g. 'MacMoody'.
Your email address, e.g. higgs-boson@gmail.com
Your role and/or occupation, e.g. 'Orthopedic Surgeon'.
Your organization or institution (if applicable), e.g. 'Royal Free Hospital'.
Publishing Agreement
NOTE: All authors, besides the first/corresponding author, must complete a separate Publishing Agreement Form and provide via email to the editorial office before comments can be posted.
CAPTCHA
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.

Vertical Tabs

You May Also be Interested in

Back to top
Advertisement

Related Articles

Topics Discussed

Alert Me