Noncontrast Computed Tomography Alberta Stroke Program Early CT Score May Modify Intra-Arterial Treatment Effect in DAWN
Abstract
Background and Purpose—
It is unknown whether noncontrast computed tomography (NCCT) can identify patients who will benefit from intra-arterial treatment (IAT) in the extended time window. We sought to characterize baseline Alberta Stroke Program Early CT Score (ASPECTS) in DAWN (DWI or CTP Assessment With Clinical Mismatch in the Triage of Wake-Up and Late Presenting Strokes Undergoing Neurointervention With Trevo) and to assess whether ASPECTS modified IAT effect.
Methods—
Core lab adjudicated ASPECTS scores were analyzed. The trial cohort was divided into 2 groups by qualifying imaging (computed tomography versus magnetic resonance imaging). ASPECTS-by-treatment interaction was tested for the trial coprimary end points (90-day utility-weighted modified Rankin Scale (mRS) score and mRS, 0–2), mRS 0 to 3, and ordinal mRS. ASPECTS was evaluated separately as an ordinal and a dichotomized (0–6 versus 7–10) variable.
Results—
Of 205 DAWN subjects, 123 (60%) had NCCT ASPECTS, and 82 (40%) had diffusion weighted imaging ASPECTS. There was a significant ordinal NCCT ASPECTS-by-treatment interaction for 90-day utility-weighted mRS (interaction P=0.04) and mRS 0 to 2 (interaction P=0.02). For both end points, IAT effect was more pronounced at higher NCCT ASPECTS. The dichotomized NCCT ASPECTS-by-treatment interaction was significant only for mRS 0 to 2 (interaction P=0.04), where greater treatment benefit was seen in the ASPECTS 7 to 10 group (odds ratio, 7.50 [2.71–20.77] versus odds ratio, 0.48 [0.04–5.40]). A bidirectional treatment effect was observed in the NCCT ASPECTS 0 to 6 group, with treatment associated with not only more mRS 0 to 3 outcomes (50% versus 25%) but also more mRS 5 to 6 outcomes (40% versus 25%). There was no significant modification of IAT effect by diffusion weighted imaging ASPECTS.
Conclusions—
Baseline NCCT ASPECTS appears to modify IAT effect in DAWN. Higher NCCT ASPECTS was associated with greater benefit from IAT. No treatment interaction was observed for diffusion weighted imaging ASPECTS.
Introduction
The DAWN trial (DWI or CTP Assessment With Clinical Mismatch in the Triage of Wake-Up and Late Presenting Strokes Undergoing Neurointervention With Trevo) and DEFUSE 3 trial (Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke 3) established the benefit of intra-arterial treatment (IAT) in carefully selected patients who were treated between 6 to 24 hours and 6 to 16 hours from time last seen well, respectively.1,2 Both trials used computed tomography (CT) perfusion or magnetic resonance imaging (MRI; diffusion with or without perfusion imaging) to identify patients who had small infarcts in the setting of a clinical or imaging mismatch. Yet it remains uncertain whether such advanced imaging is necessary for patient selection in the extended time window. Establishing the utility of noncontrast CT (NCCT) for IAT selection beyond 6 hours would enable more centers to offer thrombectomy in the extended time window, given the wider availability of NCCT at stroke centers.3,4
Baseline infarct size is typically assessed to exclude patients with large infarcts from IAT as well as to predict functional outcome after treatment.5,6 The Alberta Stroke Program Early CT Score (ASPECTS) is a semiquantitative method to determine infarct size on NCCT scans and has been shown to predict functional outcomes and symptomatic hemorrhage after IAT.7 One of the early window IAT trials primarily used baseline NCCT ASPECTS for excluding moderate-large infarcts for patient selection up to 8 hours.8 However, a subgroup analysis of the MR CLEAN trial (Multicenter Randomized Clinical Trial of Endovascular Treatment for Acute Ischemic Stroke in the Netherlands) did not find a significant modification of IAT effect by NCCT ASPECTS within the 6-hour window.9 However, as time progresses, ischemic hypoattenuation on NCCT indicating acute infarction becomes more apparent due to the progressive uptake of water by the infarcted cerebral tissue through the degraded blood-brain barrier.10 Based on this, the interaction between baseline NCCT ASPECTS and treatment effect may be different in the extended time window compared with the traditional time window.
We sought to assess the prognostic value of NCCT ASPECTS in the DAWN trial and to determine whether treatment effect varied based on ASPECTS.
Patients and Methods
The data that support the findings of this study are available from the corresponding author on reasonable request. This study is a post hoc analysis of the DAWN trial results. DAWN was a prospective, randomized, controlled, multicenter, Bayesian adaptive-enrichment trial comparing mechanical thrombectomy with Trevo plus standard medical management to medical management alone in eligible patients with evidence of a clinical-imaging mismatch (CIM) in whom treatment could be initiated between 6 and 24 hours of time last seen well. The trial protocol was approved by the Institutional Review Board at each participating site. All patients or their legal representatives provided written informed consent before randomization. The study methods and patient selection criteria have been previously reported.1
Imaging inclusion criteria were occlusion of the intracranial internal carotid artery or the first segment of the middle cerebral artery (M1) along with evidence of CIM using age-adjusted criteria. The CIM groups were as follows: A=estimated infarct<21 mL, National Institutes of Health Stroke Scale (NIHSS) ≥10, age≥80; B=estimated infarct<31 mL, NIHSS≥10, age<80; C=estimated infarct 31 to <51 mL, NIHSS≥20, age <80.
ASPECTS was scored on baseline and follow-up scans by the core imaging lab, which was blinded to all clinical data. ASPECTS was calculated by subtracting the number of affected regions from a total possible score of 10, such that lower scores correspond to larger infarcts. In this study, we included the 205 DAWN subjects who had core lab-adjudicated baseline ASPECTS, which were used for this analysis. One subject did not have an available baseline NCCT for analysis by the core lab. This was a transfer patient in whom only CT angiography and CT perfusion (CTP) were performed at the treating hospital, and the original NCCT could not be obtained from the referring hospital.
Statistical Analysis
The outcomes evaluated in this post hoc analysis included the trial coprimary end points: utility-weighted modified Rankin Scale score (UW-mRS) and the rate of functional independence (mRS 0–2), both at 90 days. Additional outcomes were 90-day ordinal mRS and 90-day moderate outcome (mRS, 0–3). The prognostic relationship between baseline ASPECTS and UW-mRS was evaluated using Spearman rank correlation, as well as multivariable linear regression adjusting for age, NIHSS, and treatment. Treatment effect was calculated for the IAT group compared with the standard care group using multivariable linear regression for UW-mRS, multivariable logistic regression for mRS 0 to 2 and mRS 0 to 3, and multivariable median regression for better functional outcome on ordinal mRS. Median regression analysis was performed rather than ordinal logistic regression using a proportional odds model to avoid assumptions related to the distribution of mRS scores. Multiplicative interaction terms were used to test the interaction between baseline ASPECTS and treatment effect. The interaction terms were reported both unadjusted and adjusted for age and NIHSS. All analyses were performed with ASPECTS as an ordinal variable and separately as a dichotomized variable (0–6 versus 7–10) and were performed separately for the NCCT data and the MRI diffusion weighted imaging (DWI) data. In subjects who underwent both baseline NCCT and DWI (n=21), they were assigned to the imaging modality that took place closer to randomization. Statistical analyses were performed using SAS software version 9.4 (Cary, NC). Statistical significance was defined as P<0.05.
Results
Baseline Clinical and Imaging Characteristics by ASPECTS Dichotomized Grouping
A total of 206 subjects were enrolled in the DAWN trial, of whom 205 had core lab adjudicated ASPECTS. There were 123 (60%) subjects in the baseline NCCT ASPECTS group and 82 (40%) subjects in the baseline DWI ASPECTS group.
Baseline characteristics for the entire NCCT cohort and for the dichotomized NCCT ASPECTS subgroups are summarized in Table 1. Sixty-two (50.4%) of the 123 subjects were female; mean age was 69.6±14 years; and median NIHSS score was 17 (interquartile range [IQR], 13–21). Eleven (8.9%) subjects received intravenous alteplase, and 65 (52.8%) presented with a wake-up stroke. The median baseline NCCT ASPECTS was 8 (IQR, 7–9): 8 (IQR, 8–9) for CIM A; 7.5 (IQR, 6–9) for CIM B; and 7 (IQR, 5–8) for CIM C. Twenty-six (21.1%) subjects had ASPECTS 0 to 6. Baseline characteristics were generally balanced between the 2 ASPECTS subgroups, except for the percentage of patients with atrial fibrillation, which was higher in the NCCT ASPECTS 7 to 10 group (39.4% versus 11.5%). Baseline CTP-estimated infarct volume was larger in the NCCT ASPECTS 0 to 6 group (15.6±12.3 versus 7.4±9.6; Table 1 and Table I in the Data Supplement).
| Overall Cohort (N=123) | NCCT ASPECTS 0–6 (N=26) | NCCT ASPECTS 7–10 (N=97) | P Value (ASPECTS 0–6 vs 7–10) | |
|---|---|---|---|---|
| Age, y | 0.011 | |||
| Mean±SD (N) | 69.6±14 (123) | 63.4±11.1 (26) | 71.2±14.3 (97) | |
| Median (Q1–Q3) | 71 (59–80) | 64.5 (54–72) | 75 (61–82) | |
| Female | 50.4% (62/123) | 38.5% (10/26) | 53.6% (52/97) | 0.170 |
| Atrial fibrillation | 33.3% (40/120) | 11.5% (3/26) | 39.4% (37/94) | 0.009* |
| Diabetes mellitus | 31.1% (37/119) | 42.3% (11/26) | 28% (26/93) | 0.162 |
| Hypertension | 78.5% (95/121) | 84.6% (22/26) | 76.8% (73/95) | 0.393 |
| Previous ischemic stroke or transient ischemic attack | 7.6% (9/118) | 0% (0/24) | 9.6% (9/94) | 0.201* |
| NIHSS score | 0.855† | |||
| Median (Q1–Q3) | 17 (13–21) | 16.5 (13–20) | 17 (14–21) | |
| Treatment with IV alteplase | 8.9% (11/123) | 7.7% (2/26) | 9.3% (9/97) | 0.801 |
| Infarct volume,‡ mL | <0.001 | |||
| Mean±SD (N) | 9.1±10.7 (123) | 15.6±12.3 (26) | 7.4±9.6 (97) | |
| Median (Q1–Q3) | 5 (1.2–14.7) | 14.9 (4.2–20) | 3 (1–9.7) | |
| Mode of presentation | 0.523 | |||
| Un-witnessed stroke | 34.1% (42/123) | 30.8% (8/26) | 35.1% (34/97) | |
| Wake-up stroke | 52.8% (65/123) | 61.5% (16/26) | 50.5% (49/97) | |
| Witnessed stroke | 13% (16/123) | 7.7% (2/26) | 14.4% (14/97) | |
| Occlusion site (Core Lab adjudicated) | 0.221* | |||
| ICA | 15.4% (19/123) | 26.9% (7/26) | 12.4% (12/97) | |
| M1 | 82.1% (101/123) | 73.1% (19/26) | 84.5% (82/97) | |
| M2 | 2.4% (3/123) | 0% (0/26) | 3.1% (3/97) | |
| Interval between time LKW and randomization, h | 0.326† | |||
| Median (Q1–Q3) | 12.2 (9.6–16.5) | 14.5 (9.5–17.2) | 12 (9.9–15.8) | |
| Interval between time LKW and image acquisition, h | 0.247† | |||
| Median (Q1–Q3) | 10.9 (8–14.9) | 13.4 (8–16.2) | 10.4 (7.8–14.8) | |
| 90-day mRS, LOCF | 0.234 | |||
| 0 | 8.1% (10/123) | 3.8% (1/26) | 9.3% (9/97) | |
| 1 | 13.8% (17/123) | 7.7% (2/26) | 15.5% (15/97) | |
| 2 | 8.1% (10/123) | 3.8% (1/26) | 9.3% (9/97) | |
| 3 | 13.8% (17/123) | 19.2% (5/26) | 12.4% (12/97) | |
| 4 | 21.1% (26/123) | 34.6% (9/26) | 17.5% (17/97) | |
| 5 | 13.8% (17/123) | 19.2% (5/26) | 12.4% (12/97) | |
| 6 | 21.1% (26/123) | 11.5% (3/26) | 23.7% (23/97) |
Unless otherwise specified, χ2 tests were performed for categorical variables and t test for continuous variables. ASPECTS indicates Alberta Stroke Program Early CT Score; CT, computed tomography; ICA, internal carotid artery; LKW, last known well; LOCF, last observation carried forward; mRS, modified Rankin Scale; NCCT, noncontrast CT; and NIHSS, National Institutes of Health Stroke Scale.
*
Fishers exact test.
†
Wilcoxon rank-sum test.
‡
Infarct volume represents the CT perfusion-estimated infarct volume in the primary trial results.
Baseline characteristics for the entire MRI cohort and for the dichotomized DWI ASPECTS subgroups are similarly summarized in Table 2. Fifty-one (62.2%) of the 82 subjects were female; mean age was 70.8±13.2 years; median NIHSS score was 17 (IQR, 13–21). Seven (8.5%) subjects received intravenous alteplase and 48 (58.5%) presented with a wake-up stroke. The median baseline DWI ASPECTS was 8 (IQR, 6–8): 8 (IQR, 8–9) for CIM A; 8 (IQR, 7–8) for CIM B; 5 (IQR, 5–5) for CIM C. Twenty-two (26.8%) subjects had DWI ASPECTS 0 to 6. Baseline characteristics were generally balanced between the 2 groups. Baseline DWI infarct volume was larger in the DWI ASPECTS 0 to 6 group versus the 7 to 10 group (23.7±12.2 versus 11.8±8.7; Table 2 and Table I in the Data Supplement).
| Overall Cohort (N=82) | MRI ASPECTS 0–6 (N=22) | MRI ASPECTS 7–10 (N=60) | P Value (ASPECTS 0–6 vs 7–10) | |
|---|---|---|---|---|
| Age, y | 0.531 | |||
| Mean±SD (N) | 70.8±13.2 (82) | 69.3±13.7 (22) | 71.4±13.1 (60) | |
| Median (Q1–Q3) | 73.5 (63–81) | 72.5 (63–78) | 75 (63–82) | |
| Female | 62.2% (51/82) | 45.5% (10/22) | 68.3% (41/60) | 0.058 |
| Atrial fibrillation | 34.2% (27/79) | 40% (8/20) | 32.2% (19/59) | 0.590* |
| Diabetes mellitus | 23.5% (19/81) | 14.3% (3/21) | 26.7% (16/60) | 0.249 |
| Hypertension | 75.6% (62/82) | 68.2% (15/22) | 78.3% (47/60) | 0.343 |
| Previous ischemic stroke or transient ischemic attack | 6.5% (5/77) | 0% (0/19) | 8.6% (5/58) | 0.325* |
| NIHSS score | 0.830† | |||
| Median (Q1–Q3) | 17 (13–21) | 18 (14–22) | 17 (13–21) | |
| Treatment with IV alteplase | 8.5% (7/82) | 9.1% (2/22) | 8.3% (5/60) | 0.913 |
| Infarct volume, mL | <0.001 | |||
| Mean±SD (N) | 15±11 (82) | 23.7±12.2 (22) | 11.8±8.7 (60) | |
| Median (Q1–Q3) | 14 (6.4–23) | 22 (15.8–32.5) | 11 (3–15.9) | |
| Mode of presentation | 0.631 | |||
| Un-witnessed stroke | 30.5% (25/82) | 22.7% (5/22) | 33.3% (20/60) | |
| Wake-up stroke | 58.5% (48/82) | 63.6% (14/22) | 56.7% (34/60) | |
| Witnessed stroke | 11% (9/82) | 13.6% (3/22) | 10% (6/60) | |
| Occlusion site (Core Lab adjudicated) | 0.883* | |||
| ICA | 25.6% (21/82) | 22.7% (5/22) | 26.7% (16/60) | |
| M1 | 72% (59/82) | 77.3% (17/22) | 70% (42/60) | |
| M2 | 2.4% (2/82) | 0% (0/22) | 3.3% (2/60) | |
| Interval between time LKW and randomization, h | 0.405† | |||
| Median (Q1–Q3) | 13.3 (10.3–15.8) | 14.1 (10.5–16.8) | 13.2 (10.2–15.6) | |
| Interval between time LKW and image acquisition, h | 0.470† | |||
| Median (Q1–Q3) | 12.5 (9.1–15.2) | 13 (9.1–16.2) | 12.3 (9.2–14.8) | |
| 90-day mRS, LOCF | 0.814 | |||
| 0 | 4.9% (4/82) | 9.1% (2/22) | 3.3% (2/60) | |
| 1 | 14.6% (12/82) | 9.1% (2/22) | 16.7% (10/60) | |
| 2 | 14.6% (12/82) | 18.2% (4/22) | 13.3% (8/60) | |
| 3 | 15.9% (13/82) | 13.6% (3/22) | 16.7% (10/60) | |
| 4 | 26.8% (22/82) | 27.3% (6/22) | 26.7% (16/60) | |
| 5 | 9.8% (8/82) | 13.6% (3/22) | 8.3% (5/60) | |
| 6 | 13.4% (11/82) | 9.1% (2/22) | 15% (9/60) |
Unless otherwise specified, χ2 tests was performed for categorical variables, and t test for continuous variables. ASPECTS indicates Alberta Stroke Program Early CT Score; ICA, internal carotid artery; LKW, last known well; LOCF, last observation carried forward; MRI, magnetic resonance imaging; mRS, modified Rankin Scale; and NIHSS, National Institutes of Health Stroke Scale.
*
Fishers exact test.
†
Wilcoxon rank-sum test.
Prognostic Relationship and Treatment Effect Modification by Baseline NCCT ASPECTS
The rank correlation between baseline NCCT ASPECTS and 90-day UW-mRS was weak but significant for the IAT group (ρ, 0.37, P=0.004) and not significant for the control group (ρ, −0.14, P=0.30). In multivariable analysis adjusted for age, baseline NIHSS, and treatment assignment, the association between NCCT ASPECTS and 90-day UW-mRS did not meet statistical significance (β, 0.43, P=0.06).
Table 3 lists the treatment effect of IAT by ordinal and dichotomized NCCT ASPECTS for the trial coprimary end points in addition to 90-day mRS 0 to 3 and ordinal mRS. There was a significant ordinal ASPECTS-by-treatment interaction for the coprimary end points, 90-day UW-mRS (interaction term P=0.04), and mRS 0 to 2 (interaction term P=0.02; Table 3). In general, the magnitude of treatment effect was greater with higher ASPECTS score for these end points. Similar relationships were seen for 90-day mRS 0 to 3 and ordinal mRS, although the interaction terms were not significant.
| Analysis Population | OR* [95% CI] (mRS 0–2 vs 3–6) | P for Interaction† | OR* [95% CI] (mRS 0–3 vs 4–6) | P for Interaction† | Difference of Average UW-mRS* [95% CI] | P for Interaction‡ | Median mRS | P for Interaction§ | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Treatment | Control | P Value | |||||||||
| NCCT | |||||||||||
| ASPECTS 3 | ‖ | ‖ | ‖ | ‖ | 4 | ‖ | |||||
| ASPECTS 4 | ‖ | ‖ | ‖ | 6 | 5 | ‖ | |||||
| ASPECTS 5 | 0.33 [0.02 to 5.03] | 0.023¶ | 2.50 [0.25–24.71] | 0.237¶ | 0.52 [−3.33 to 4.37] | 0.042¶ | 3 | 3.5 | 0.941 | 0.113¶ | |
| ASPECTS 6 | ‖ | ‖ | −3.31 [−8.58 to 1.95] | 0.042¶ | 6 | 4 | 0.035 | 0.113¶ | |||
| ASPECTS 7 | 2.92 [0.55 to 15.56] | 0.023¶ | 0.89 [0.20–3.91] | 0.237¶ | 0.77 [−1.96 to 3.50] | 0.042¶ | 4 | 4 | 0.608 | 0.113¶ | |
| ASPECTS 8 | ‖ | 2.75 [0.67–11.34] | 0.237¶ | 2.24 [−0.42 to 4.90] | 0.042¶ | 3.5 | 4 | 0.165 | 0.113¶ | ||
| ASPECTS 9 | 11.25 [1.65 to 76.84] | 0.023¶ | 27.00 [3.17–230.0] | 0.237¶ | 4.77 [2.07 to 7.47] | 0.042¶ | 1 | 4 | 0.009 | 0.113¶ | |
| ASPECTS 10 | ‖ | ‖ | 2.40 [−8.72 to 13.52] | 0.042¶ | 2 | 3 | 0.767 | 0.113¶ | |||
| ASPECTS 0–6 | 0.48 [0.04 to 5.40] | 0.040# | 3.00 [0.56–16.07] | 0.987# | 0.12 [−2.64 to 2.87] | 0.163# | 3.5 | 4 | 0.786 | 0.731# | |
| ASPECTS 7–10 | 7.50 [2.71 to 20.77] | 0.040# | 2.95 [1.29–6.78] | 0.987# | 2.42 [0.92 to 3.92] | 0.163# | 2 | 4 | 0.004 | 0.731# | |
| MRI | |||||||||||
| ASPECTS 3 | ‖ | ‖ | ‖ | 2 | 4 | ‖ | |||||
| ASPECTS 4 | ‖ | ‖ | ‖ | ‖ | 4 | ‖ | |||||
| ASPECTS 5 | ‖ | 6.00 [0.35–101.6] | 0.282¶ | 2.25 [−2.92 to 7.42] | 0.518¶ | 3 | 4.5 | 0.327 | 0.999¶ | ||
| ASPECTS 6 | ‖ | ‖ | 4.88 [−2.19 to 11.96] | 0.518¶ | 1 | 4 | 0.304 | 0.999¶ | |||
| ASPECTS 7 | 5.00 [0.21 to 117.9] | 0.277¶ | 0.88 [0.06–12.97] | 0.282¶ | 1.27 [−2.59 to 5.13] | 0.518¶ | 4 | 4 | 0.616 | 0.999¶ | |
| ASPECTS 8 | 2.18 [0.35 to 13.76] | 0.277¶ | 3.61 [0.64–20.32] | 0.282¶ | 0.64 [−2.24 to 3.53] | 0.518¶ | 3 | 4 | 0.706 | 0.999¶ | |
| ASPECTS 9 | 10.50 [0.91 to 121.4] | 0.277¶ | 4.50 [0.59–34.61] | 0.282¶ | 2.81 [−1.16 to 6.77] | 0.518¶ | 2 | 4.5 | 0.207 | 0.999¶ | |
| ASPECTS 10 | ‖ | ‖ | ‖ | 1 | ‖ | ‖ | |||||
| ASPECTS 0–6 | ‖ | ‖ | 26.67 [2.31–308.0] | 0.136# | 3.98 [1.30 to 6.65] | 0.175# | 2 | 4 | 0.010 | 0.249# | |
| ASPECTS 7–10 | 5.41 [1.53 to 19.12] | ‖ | 3.50 [1.20–10.19] | 0.136# | 1.73 [0.00 to 3.46] | 0.175# | 3 | 4 | 0.048 | 0.249# | |
ASPECTS indicates Alberta Stroke Program Early CT Score; MRI, magnetic resonance imaging; mRS, modified Rankin Scale; NCCT, noncontrast computed tomography; OR, odds ratio; and UW-mRS, utility-weighted mRS.
*
Treatment compared with control, that is, for OR the control group is the reference level, and the difference of UW-mRS is treatment–control.
†
Unconditional logistic regression.
‡
Linear regression.
§
Median regression.
‖
Results that are not estimable or unreliable.
¶
The interaction term between continuous ASPECTS and treatment arm.
#
The interaction term between dichotomous ASPECTS and treatment arm.
For dichotomized ASPECTS (Table 3), there was a significant treatment interaction only for 90-day mRS 0 to 2 (interaction term P=0.04) with greater treatment benefit seen in the ASPECTS 7 to 10 group (odds ratio, 7.50 [2.71–20.77] versus odds ratio, 0.48 [0.04–5.40] in ASPECTS 0–6). For 90-day UW-mRS, the treatment benefit was numerically higher for ASPECTS 7 to 10 (difference of average UW-mRS [treatment minus control] 2.42 versus 0.12 for ASPECTS 7–10 versus 0–6, respectively; interaction term P=0.16). The point estimates for treatment benefit for 90-day mRS 0 to 3 were nearly identical between the ASPECTS subgroups. The Rankin Scale distributions by treatment and dichotomized NCCT ASPECTS are shown in Figure 1. There was a relatively uniform shift towards better outcomes with treatment in ASPECTS 7 to 10 subjects. A bidirectional effect was observed in the ASPECTS 0 to 6 group, with treatment associated with a greater proportion of mRS 0 to 3 outcomes (50% versus 25%) due mostly to the increase in mRS 3 outcomes, and also a greater proportion of mRS 5 to 6 outcomes (40% versus 25%) owing to a 30% increase in mortality.
After further adjusting for age and baseline NIHSS, none of the interaction analyses were statistically significant (Table II in the Data Supplement).
Prognostic Relationship and Treatment Effect Modification by Baseline DWI ASPECTS
There was no significant correlation between baseline DWI ASPECTS and 90-day UW-mRS in univariate (ρ, 0.02, P=0.89 for IAT; ρ, 0.04, P=0.81 for control) or multivariable analysis adjusted for age, baseline NIHSS, and treatment assignment (β, 0.12, P=0.56).
Table 3 (bottom rows) lists the treatment effect of IAT by DWI ASPECTS. For both ordinal and dichotomized ASPECTS, there was no significant ASPECTS-by-treatment interaction identified for any of the end points. Figure 2 depicts the distribution of mRS scores by treatment and dichotomized DWI ASPECTS. The shift towards better outcomes with IAT were comparable and largely unidirectional in both ASPECTS subgroups. The interaction analyses remained not statistically significant when further adjusted for age and baseline NIHSS (Table I in the Data Supplement).
Discussion
In this post hoc analysis of the DAWN trial, baseline NCCT ASPECTS appeared to modify the benefit of thrombectomy. Specifically, lower ASPECTS on NCCT was associated with reduced treatment effect for the trial coprimary end points. In the analysis of dichotomized NCCT ASPECTS, the ASPECTS 0 to 6 subgroup was associated with a significantly smaller treatment benefit with respect to 90-day mRS 0 to 2 (with a point estimate for the odds ratio <1) and a numerically smaller benefit for UW-mRS. No ASPECTS-by-treatment interaction was observed in the DWI cohort.
Closer examination of the mRS distributions for dichotomized NCCT ASPECTS suggests that treatment benefit in the ASPECTS 0 to 6 group may accrue at mRS 0 to 3 rather than at mRS 0 to 2. However, this is counterbalanced by an increase in mRS 5 to 6 with IAT, resulting in an overall difference in average UW-mRS between IAT and control of 0.12, which is below the accepted minimum clinically important difference of 0.3.11 Given the small number of subjects in this subgroup, these findings require confirmation in future studies.
The differential treatment interaction observed between NCCT and DWI ASPECTS highlights an essential difference in the imaging analysis between these modalities. For the MRI group, ASPECTS was determined on the same DWI sequence that was used for assessment of trial eligibility, whereas in the CT cohort, ASPECTS was determined on imaging (NCCT) different from the qualifying scan (CTP).
The ability of NCCT ASPECTS to identify CTP-selected subjects who are less likely to benefit from IAT supports its potential utility for selecting patients for extended window thrombectomy. Although NCCT ASPECTS and CTP-estimated ischemic core volume are correlated (as demonstrated in this and other studies), they measure distinct pathophysiologic processes. ASPECTS grading scores CT hypoattenuation, which reflects ionic and vasogenic edema from blood-brain barrier breakdown and is a highly specific marker of irreversible tissue injury.10,12 CTP is a dynamic contrast-enhanced imaging technique that measures collateral perfusion and, therefore, does not provide a direct measure of tissue viability. Rather it relies on a user-defined threshold applied to a calculated perfusion parameter (in this case, cerebral blood flow) to estimate tissue that is dead. Numerous studies have shown substantial errors in CTP-estimated ischemic core volume when compared against reference standards, such as concurrent DWI and final infarct volume.13–16 The errors are seen in both directions. CTP core estimates may underestimate the true volume of acute infarction, which might explain why 21% of the CTP-selected cohort in DAWN had NCCT ASPECTS 0 to 6 and responded less favorably to thrombectomy. More importantly, CTP can overestimate acute infarct volume, potentially leading to inappropriate exclusion of patients from IAT.15,16 This was demonstrated in a post hoc analysis of the MR CLEAN trial, in which 42% of the subjects did not meet the CTP mismatch inclusion criteria from the EXTEND-IA trial (Extending the Time for Thrombolysis in Emergency Neurological Deficits–Intra-Arterial) but were shown to have a similar benefit of thrombectomy as those who did.17 It is unknown how many patients who were excluded from DAWN had small to moderate ischemic cores (and high ASPECTS) and would have benefited from thrombectomy. Based on the very low numbers needed to treat for benefit in the trial, this population may have been sizeable. Future studies should examine whether NCCT ASPECTS can further expand the eligible patient population in the extended window. A recent study based on a large prospectively acquired cohort reported strikingly similar outcomes between patients with high ASPECTS (6–10) treated with thrombectomy in the early (0–6 hours) and late (>6 hours) time windows.18 In this analysis, the use of MRI or CTP was not associated with outcome, whereas ASPECTS was an independent predictor. An additional advantage of NCCT is that it is available at all thrombectomy-capable centers.
In the MRI cohort, the scoring of ASPECTS on the qualifying DWI scan helps to explain the lack of treatment interaction. The trial imaging inclusion criteria, along with the high accuracy of DWI for infarct detection, ensured that the core volumes were small despite low ASPECTS. Such discrepancies between ASPECTS and infarct volume are particularly common with DWI due to its high sensitivity, which can lead to point deductions in ASPECTS regions that have only a small infarct burden. However, under different conditions, ASPECTS might provide clinically useful information on infarct localization. In a post hoc analysis of the THRACE trial (Mechanical Thrombectomy After Intravenous Alteplase Versus Alteplase Alone After Stroke), DWI ASPECTS was significantly associated with good outcome in subjects with baseline DWI lesion volume >70 mL, whereas variations in DWI lesion volume were not.19
The primary limitation of this study is the small sample size of the low ASPECTS groups, which is a consequence of the restrictive imaging criteria in DAWN and the DAWN exclusion criterion of infarcts larger than one-third of the middle cerebral artery territory on baseline CT or MRI. As a result, the adjustment of the interaction terms for age and NIHSS was likely underpowered, and there were wide CIs for the treatment effect estimates. Also, we examined multiple clinical end points and treated baseline ASPECTS in both ordinal and dichotomized fashion. Because of the hypothesis-generating nature of this analysis, we did not adjust the α level (ie, statistical significance) for multiple comparisons. Furthermore, the findings must be viewed within the context of the trial selection criteria, and no conclusions can be drawn about patients who do not meet these criteria. Randomized studies are needed to test whether primary NCCT ASPECTS-based selection can expand the eligible thrombectomy population in the extended window and to better characterize the ASPECTS threshold where treatment benefit is lost or negligible.
Supplemental Material
File (str_stroke-2018-024583_supp1.pdf)
- Download
- 277.14 KB
References
1.
Nogueira RG, Jadhav AP, Haussen DC, Bonafe A, Budzik RF, Bhuva P, et al; DAWN Trial Investigators. Thrombectomy 6 to 24 hours after stroke with a mismatch between deficit and infarct. N Engl J Med. 2018;378:11–21. doi: 10.1056/NEJMoa1706442
2.
Albers GW, Marks MP, Kemp S, Christensen S, Tsai JP, Ortega-Gutierrez S, et al; DEFUSE 3 Investigators. Thrombectomy for stroke at 6 to 16 hours with selection by perfusion imaging. N Engl J Med. 2018;378:708–718. doi: 10.1056/NEJMoa1713973
3.
Sheth KN, Terry JB, Nogueira RG, Horev A, Nguyen TN, Fong AK, et al. Advanced modality imaging evaluation in acute ischemic stroke may lead to delayed endovascular reperfusion therapy without improvement in clinical outcomes. J Neurointerv Surg. 2013;5(suppl 1):i62–i65. doi: 10.1136/neurintsurg-2012-010512
4.
Mehta B, Leslie-Mazwi TM, Chandra RV, Chaudhry ZA, Rabinov JD, Hirsch JA, et al. Assessing variability in neurointerventional practice patterns for acute ischemic stroke. J Neurointerv Surg. 2013;5(suppl 1):i52–i57. doi: 10.1136/neurintsurg-2012-010565
5.
Yoo AJ, Verduzco LA, Schaefer PW, Hirsch JA, Rabinov JD, González RG. MRI-based selection for intra-arterial stroke therapy: value of pretreatment diffusion-weighted imaging lesion volume in selecting patients with acute stroke who will benefit from early recanalization. Stroke. 2009;40:2046–2054. doi: 10.1161/STROKEAHA.108.541656
6.
Olivot JM, Mosimann PJ, Labreuche J, Inoue M, Meseguer E, Desilles JP, et al. Impact of diffusion-weighted imaging lesion volume on the success of endovascular reperfusion therapy. Stroke. 2013;44:2205–2211. doi: 10.1161/STROKEAHA.113.000911
7.
Yoo AJ, Zaidat OO, Chaudhry ZA, Berkhemer OA, González RG, Goyal M, et al; Penumbra Pivotal and Penumbra Imaging Collaborative Study (PICS) Investigators. Impact of pretreatment noncontrast CT Alberta Stroke Program Early CT Score on clinical outcome after intra-arterial stroke therapy. Stroke. 2014;45:746–751. doi: 10.1161/STROKEAHA.113.004260
8.
Jovin TG, Chamorro A, Cobo E, de Miquel MA, Molina CA, Rovira A, et al; REVASCAT Trial Investigators. Thrombectomy within 8 hours after symptom onset in ischemic stroke. N Engl J Med. 2015;372:2296–2306. doi: 10.1056/NEJMoa1503780
9.
Yoo AJ, Berkhemer OA, Fransen PSS, van den Berg LA, Beumer D, Lingsma HF, et al; MR CLEAN Investigators. Effect of baseline Alberta Stroke Program Early CT Score on safety and efficacy of intra-arterial treatment: a subgroup analysis of a randomised phase 3 trial (MR CLEAN). Lancet Neurol. 2016;15:685–694. doi: 10.1016/S1474-4422(16)00124-1
10.
Dzialowski I, Weber J, Doerfler A, Forsting M, von Kummer R. Brain tissue water uptake after middle cerebral artery occlusion assessed with CT. J Neuroimaging. 2004;14:42–48.
11.
Kaplan RM. The minimally clinically important difference in generic utility-based measures. COPD. 2005;2:91–97.
12.
Lev MH, Farkas J, Gemmete JJ, Hossain ST, Hunter GJ, Koroshetz WJ, et al. Acute stroke: improved nonenhanced CT detection–benefits of soft-copy interpretation by using variable window width and center level settings. Radiology. 1999;213:150–155. doi: 10.1148/radiology.213.1.r99oc10150
13.
Schaefer PW, Souza L, Kamalian S, Hirsch JA, Yoo AJ, Kamalian S, et al. Limited reliability of computed tomographic perfusion acute infarct volume measurements compared with diffusion-weighted imaging in anterior circulation stroke. Stroke. 2015;46:419–424. doi: 10.1161/STROKEAHA.114.007117
14.
Copen WA, Yoo AJ, Rost NS, Morais LT, Schaefer PW, González RG, et al. In patients with suspected acute stroke, CT perfusion-based cerebral blood flow maps cannot substitute for DWI in measuring the ischemic core. PLoS One. 2017;12:e0188891. doi: 10.1371/journal.pone.0188891
15.
Geuskens RR, Borst J, Lucas M, Boers AM, Berkhemer OA, Roos YB, et al; MR CLEAN trial Investigators (www.mrclean-trial.org). Characteristics of misclassified CT perfusion ischemic core in patients with acute ischemic stroke. PLoS One. 2015;10:e0141571. doi: 10.1371/journal.pone.0141571
16.
Boned S, Padroni M, Rubiera M, Tomasello A, Coscojuela P, Romero N, et al. Admission CT perfusion may overestimate initial infarct core: the ghost infarct core concept. J Neurointerv Surg. 2017;9:66–69. doi: 10.1136/neurintsurg-2016-012494
17.
Borst J, Berkhemer OA, Roos YB, van Bavel E, van Zwam WH, van Oostenbrugge RJ, et al; MR CLEAN Investigators. Value of computed tomographic perfusion-based patient selection for intra-arterial acute ischemic stroke treatment. Stroke. 2015;46:3375–3382. doi: 10.1161/STROKEAHA.115.010564
18.
Nagel S, Herweh C, Pfaff JAR, Schieber S, Schönenberger S, Möhlenbruch MA, et al. Simplified selection criteria for patients with longer or unknown time to treatment predict good outcome after mechanical thrombectomy. J Neurointerv Surg. 2019;11:559–562. doi: 10.1136/neurintsurg-2018-014347
19.
Gautheron V, Xie Y, Tisserand M, Raoult H, Soize S, Naggara O, et al. Outcome after reperfusion therapies in patients with large baseline diffusion-weighted imaging stroke lesions: a THRACE trial (Mechanical Thrombectomy After Intravenous Alteplase Versus Alteplase Alone After Stroke) subgroup analysis. Stroke. 2018;49:750–753. doi: 10.1161/STROKEAHA.117.020244
Information & Authors
Information
Published In
Copyright
© 2019 American Heart Association, Inc.
Versions
You are viewing the most recent version of this article.
History
Received: 22 December 2018
Revision received: 24 May 2019
Accepted: 12 June 2019
Published online: 26 July 2019
Published in print: September 2019
Keywords
Subjects
Authors
Disclosures
Dr Yoo receives research grant support from Stryker, Medtronic, Neuravi/Cerenovus, Genentech, and Penumbra, is a consultant for Cerenovus and Genentech, and has equity ownership interest in Insera Therapeutics, Inc. Dr Jovin has the following disclosures: Consultant at Cerenovus (steering committee/data safety monitoring board–modest), Stryker Neurovascular (Principal Investigator DAWN [DWI or CTP Assessment With Clinical Mismatch in the Triage of Wake-Up and Late Presenting Strokes Undergoing Neurointervention With Trevo]-unpaid), holds Stock at Anaconda, Blockade Medical, Route 92, Corindus, FreeOx Biotech, Viz.ai, Silk Road, Imperative Care. Dr Nogueira reports potential conflicts with Stryker Neurovascular (DAWN Trial Principal Investigator–no compensation, Trevo Retriever Registry Steering Committee–no compensation, Trevo-2 Trial Principal Investigator–modest; Consultant–modest), Medtronic (SWIFT [SOLITAIRE FR With the Intention for Thrombectomy] Trial Steering Committee–modest; SWIFT-Prime Trial Steering Committee–no compensation; STAR [Solitaire Flow Restoration Thrombectomy for Acute Revascularization] Trial Angiographic Core Lab–significant), Penumbra (3D Separator Trial Executive Committee–no compensation), Cerenovus/ Neuravi (ENDOLOW Trial Principal Investigator, EXCELLENT Registry Principal Investigator, Analysis of Revascularization in Ischemic Stroke with EmboTrap (ARISE-2)-2 trial Steering Committee–no compensation, Physician Advisory Board, modest), Phenox (Physician Advisory Board, modest), Anaconda (Physician Advisory Board, modest), Genentech (Physician Advisory Board–modest), Biogen (Physician Advisory Board–modest), Prolong Pharmaceuticals (Physician Advisory Board–modest), Allm Inc. (Physician Advisory Board–no compensation), IschemaView (Speaker, modest), Brainomix (Research Software Use–no compensation), Sensome (Research Device Use–no compensation), Viz-AI (Physician Advisory Board, stock options), Philips (Research Software Use–no compensation, Speaker–modest), and Corindus Vascular Robotics (Physician Advisory Board, stock options). Dr Haussen is a consultant for Stryker and Vesalio. Dr Bonafe is a consultant for Stryker, Medtronic, and Phenox. Dr Yavagal is a consultant for Medtronic, Neural Analytics, Cerenovus, Rapid Medical (Steering Committee Member of TIGER [Treatment With Intent to Generate Reperfusion] Clinical Trial & Consultant); Stryker (funding for patient enrollment in DAWN trial as sponsor of the trial). Dr Ribo is a consultant for Cerenovus, Medtronic, Stryker, Apta Targets, Anaconda Biomed and has equity ownership interest in Anaconda Biomed. Dr Cognard is a consultant for Stryker, Microvention, Medtronic, and Cerenovus. A.E. Hassan is a consultant, speaker and proctor for Medtronic, Stryker, and Microvention and a consultant/speaker for Penumbra, Balt, and GE Heathcare. Dr Saver is an employee of the University of California. The University of California has patent rights in retrieval devices for stroke. Dr Saver has served as an unpaid site investigator in multicenter trials sponsored by Medtronic, Stryker, and Neuravia/Cerenovus for which the University of California Regents received payments on the basis of clinical trial contracts for the number of subjects enrolled. Dr Saver has received contracted hourly payments from Medtronic, Stryker, and Neuravia/Cerenovus and contracted stock options from Rapid Medical for services as a scientific consultant advising on rigorous trial design and conduct. Dr Liebeskind receives support as a consultant as imaging core lab for Stryker, Medtronic, Cerenovus, Vesalio, Neurvana, and Rapid Medical. The other authors report no conflicts.
Sources of Funding
The DAWN trial (DWI or CTP Assessment With Clinical Mismatch in the Triage of Wake-Up and Late Presenting Strokes Undergoing Neurointervention With Trevo) was funded by Stryker Neurovascular, Inc.
Metrics & Citations
Metrics
Citations
Download Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Select your manager software from the list below and click Download.
- Dual-Energy CT in Acute Stroke: Could Non-Contrast CT Be Replaced by Virtual Non-Contrast CT? A Feasibility Study, Journal of Clinical Medicine, 13, 13, (3647), (2024).https://doi.org/10.3390/jcm13133647
- Region‐specific interobserver agreement of the Alberta Stroke Program Early Computed Tomography Score: A meta‐analysis, Journal of Neuroimaging, 34, 2, (195-204), (2024).https://doi.org/10.1111/jon.13184
- Effect of computed tomography vs. computed tomography perfusion on mechanical thrombectomy outcomes within 6 hours, European Radiology, 34, 8, (5331-5338), (2024).https://doi.org/10.1007/s00330-023-10545-y
- Thrombectomy for Stroke With Large Infarct on Noncontrast CT, JAMA, 332, 16, (1355), (2024).https://doi.org/10.1001/jama.2024.13933
- Prognostic Accuracy of N20 Somatosensory Potential in Patients With Acute Ischemic Stroke and Endovascular Thrombectomy, Stroke: Vascular and Interventional Neurology, 3, 5, (2023)./doi/10.1161/SVIN.122.000735
- Neurological Functional Independence After Endovascular Thrombectomy and Different Imaging Modalities for Large Infarct Core Assessment, Clinical Neuroradiology, 33, 1, (21-29), (2022).https://doi.org/10.1007/s00062-022-01202-w
- Predictors of Early Neurological Improvement in Patients with Anterior Large Vessel Occlusion and Successful Reperfusion Following Endovascular Thrombectomy—Does CT Perfusion Imaging Matter?, Clinical Neuroradiology, 32, 3, (839-847), (2022).https://doi.org/10.1007/s00062-022-01147-0
- Imaging as a Selection Tool for Thrombectomy in Acute Ischemic Stroke, Neurology, 97, 20_Supplement_2, (2021).https://doi.org/10.1212/WNL.0000000000012793
- Association of White Matter Lesions and Outcome After Endovascular Stroke Treatment, Neurology, 96, 3, (2021).https://doi.org/10.1212/WNL.0000000000010994
- Clinically Approximated Hypoperfused Tissue in Large Vessel Occlusion Stroke, Stroke, 52, 6, (2109-2114), (2021)./doi/10.1161/STROKEAHA.120.033294
- See more
Loading...
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginPurchase Options
Purchase this article to access the full text.
eLetters(0)
eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.
Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.