Introduction: Stroke lesion volume on MRI or CT provides objective evidence of tissue injury as a consequence of ischemic stroke. Measurement of “final” lesion volume at 24-h following endovascular therapy (post-EVT) has been used in multiple studies as a surrogate for clinical outcome. However, despite successful recanalization, a significant proportion of patients do not experience favorable clinical outcome. The goals of this study were to quantify lesion growth during the first week after treatment, identify early predictors, and explore the association with clinical outcome. Methods: This is a prospective study of stroke patients at two centers who met the following criteria: (i) anterior large vessel occlusion acute ischemic stroke, (ii) attempted EVT, and (iii) had 3T MRI post-EVT at 24-h and 5-day. We defined “early” and “late” lesion growth as ≥10 mL lesion growth between baseline and 24-h diffusion-weighted imaging (DWI) and between 24-h DWI and 5-day fluid attenuated inversion recovery imaging, respectively. Complete reperfusion was defined as >90% reduction of the volume of tissue with perfusion delay (Tmax>6 s) between pre-EVT and 24-h post-EVT. Favorable clinical outcome was defined as modified Rankin scale (mRS) of 0–2 at 30 or 90 days. Results: One hundred twelve patients met study criteria with median age 67 years, 56% female, median admit NIHSS 19, 54% received IV or IA thrombolysis, 66% with M1 occlusion, and median baseline DWI volume 21.2 mL. Successful recanalization was achieved in 87%, and 68% had complete reperfusion, with an overall favorable clinical outcome rate of 53%. Nearly two-thirds (65%) of the patients did not have late lesion growth with a median volume change of −0.3 mL between 24-h and 5-day and an associated high rate of favorable clinical outcome (64%). However, ∼1/3 of patients (35%) did have significant late lesion growth despite successful recanalization (87%: 46% mTICI 2b/41% mTICI 3). Late lesion growth patients had a 27.4 mL change in late lesion volume and 30.1 mL change in early lesion volume. These patients had an increased hemorrhagic transformation (HT) rate of 68% with only 1 in 3 patients having favorable clinical outcome. Late lesion growth was independently associated with incomplete reperfusion, HT, and unfavorable outcome. Conclusion: Approximately 1 out of 3 patients had late lesion growth following EVT, with a favorable clinical outcome occurring in only 1 out of 3 of these patients. Most patients with no early lesion growth had no late lesion growth. Identification of patients with late lesion growth could be critical to guide clinical management and inform prognosis post-EVT. Additionally, it can serve as an imaging biomarker for the development of adjunctive therapies to mitigate reperfusion injury.

Stroke lesion volume on MRI or CT provides objective evidence of tissue injury following ischemic stroke. Emerging data suggest that therapeutic reperfusion of at-risk tissue may result in minimal growth of the ischemic lesion following successful recanalization. An estimate of lesion volume post-EVT has potential utility in guiding clinical care (e.g., blood pressure management, safety to initiate anticoagulation, prognosis), in addition to serving as an imaging biomarker for clinical treatment trials and as a potential outcome measure by being an early surrogate for final infarct volume. Tourdias and colleagues demonstrated a strong association between the measurement of infarct volume using fluid attenuated inversion recovery imaging (FLAIR) at 5 versus FLAIR at 30 days [1]. Other studies have reported a strong correlation between subacute infarct FLAIR volume within 1 week compared to final infarct FLAIR volume at 30 or 90 days [2, 3]. Largely for convenience, follow-up imaging obtained at 24–48 h post-EVT has routinely been used in trials, registry datasets, and research studies; however, this may be too early and may underestimate lesion growth post-EVT and final infarct volume [4‒8]. As demonstrated by Boers et al. [9] using the pooled Hermes dataset, follow-up infarct volume using mainly NCCT at ∼24-h was found to be an insufficient surrogate for treatment efficacy in EVT.

Prior studies are conflicting on the impact of reperfusion or recanalization on lesion growth over time [4, 5, 10‒13]. Maclellan et al. demonstrated in non-reperfused patients that lesion growth was substantial in late-presenting patients, suggesting that lesion volume measurement at 24-h was not sufficient based on inadequate reperfusion [7]. Federau et al. [14] found that in non-reperfused patients, the lesion volume on diffusion-weighted imaging (DWI) obtained at 24-h underestimated the lesion volume on 5-day FLAIR. The specific timing of lesion volume measurement, which imaging modalities are used, and how lesion reversal, edema formation [15], and hemorrhagic transformation (HT) are accounted for in these measurements complicate the interpretation and comparability of these studies. Validation approaches of final infarct volume as a surrogate have been driven primarily by the association and prediction of clinical outcome [4, 8, 10, 16, 17]. However, despite successful recanalization, prior studies have reported an association with incomplete reperfusion and lesion growth [10, 18]. Secondary injury effects, including HT and hyperemia, on lesion growth are also contributors to poor clinical outcomes [19]. Standardization of both imaging modality and timing of how and when to best measure final infarct volume is a clinical research priority as it impacts the utility of this surrogate endpoint in future clinical trials [20].

The question remains unanswered: in patients who undergo successful endovascular therapy, is 24-h too early to assess acute infarct size including the effects of secondary injury? We hypothesize that lesion growth continues over the first week in a substantial number of patients and is associated with unfavorable outcomes post-EVT. The goals of this study were to quantify lesion growth beyond 24-h, identify predictors of this late lesion growth, and explore potential associations with clinical outcome.

Study Population

The study population consisted of consecutive patients presenting from January 2015 through February 2022. These patients presented to one of two stroke centers and met the following criteria: (i) had an acute ischemic stroke due to an anterior circulation large vessel occlusion of the eICA, iICA, or MCA, (ii) had EVT attempted, and (iii) had MRI post-EVT at 24-h and 5-day.

Variables reported in the study were age, sex, onset (defined as last seen normal time to triage time), admit National Institutes of Health Stroke Scale (NIHSS), pre-admit modified Rankin Scale (mRS), acute intervention including IV or IA thrombolysis, EVT data including modified TICI (mTICI) scores, onset to recanalization time, pre-EVT MRI to recanalization time, single versus multiple pass EVT, and clinical outcomes. Favorable clinical outcome was defined as mRS of 0–2 at 30 or 90 days. Available mRS at 90 days was used, otherwise mRS at 30 days was used.

Imaging Protocol

Established MR imaging protocols were utilized across the two stroke centers. The MRI time points included baseline (pre-EVT when available for patients not screened by CT, otherwise ∼2 h post-EVT was targeted), 24-h, and 5-day post-EVT. Patients were imaged on the same 3T MRI scanner for all time points, either a Siemens Skyra (Siemens AG, Munich, Germany) or Philips Achieva (Philips Healthcare, Best, The Netherlands). A dosage of 0.1 mmol/kg of Gd-DTPA was administered if the dynamic susceptibility contrast perfusion-weighted imaging was obtained at the time point. The protocol sequences evaluated in this study were DWI, scanner-generated apparent diffusion coefficient maps, gradient-recalled echo, pre-Gd-DTPA, and post-Gd-DTPA FLAIR, circle of Willis magnetic resonance angiography, and perfusion-weighted imaging.

Quantitative Imaging Analysis

A semiautomated validated [21] approach using MIPAV (Center for Information Technology, NIH) was performed to calculate the lesion volumes on DWI at baseline (pre-EVT or early post-EVT) and 24-h post-EVT, and on FLAIR at 5-day. Three independent trained raters (RD, CL, ML), blinded to recanalization and reperfusion status, performed these measurements, and the average of two sets of volumes were used in the analyses [19]. Any areas of HT were included in the lesion volume measurements. Areas of chronic infarct as visualized on FLAIR were avoided. Patients with 5-day FLAIR volumes reduced by > 10 mL from 24-h were evaluated as a quality control measure. Inclusion of areas of hyperintense acute reperfusion injury marker (HARM) was minimized by using the pre-Gd-DTPA FLAIR series at 5-day. A fully automated algorithm developed using MIPAV (CIT, NIH), “coretool,” was used to calculate the Tmax>6 s delay perfusion volumes at pre-EVT and at 24-h post-EVT [19]. During the development of “coretool,” the algorithm generated perfusion volumes with Tmax>6 s delay with a strong correlation, R2 = 0.92 (unpublished), compared to the results obtained using RAPID. Complete reperfusion was defined as >90% reduction of the volume of tissue with perfusion delay (Tmax>6 s) between pre-EVT and 24-h-post-EVT or if the 24-h-post-EVT Tmax>6 s volume = 0 mL.

Stroke Lesion Definitions

Early lesion growth was defined as ≥10 mL lesion growth between baseline and 24-h post-EVT DWI lesion volumes. Late lesion growth was defined as ≥10 mL lesion growth between 24-h DWI and 5-day FLAIR. Patients were stratified into groups of less than or greater than 10 mL of lesion growth which was approximately the median early lesion growth (12 mL) for the study population (and also found to be similar to the average volume difference between 24-h DWI and 5-day FLAIR lesion volume data) (online suppl. Fig. I; for all online suppl. material, see https://doi.org/10.1159/000536470).

Qualitative Imaging Analysis

HT grading was performed by consensus across three expert raters (AWH, LLL, ML) with modified ECASS-II criteria using gradient-recalled echo imaging acquired at 24-h post-EVT. HARM rating was also performed by consensus with the same raters, classified as none, minor, or severe HARM using the post-gadolinium FLAIR at 24-h post-EVT. Severe HARM was classified if HARM regions were visualized on ≥10 contiguous slices using the post-Gd FLAIR. Successful recanalization was defined as mTICI 2b/3, 2c was not utilized, based on the angiographic interpretation by the treating interventionist at the end of the EVT procedure.

Statistical Analysis

Variables were reported including baseline and EVT characteristics, imaging volumes and complications, and clinical outcomes. Variables were reported as median (interquartile range 25–75) or percentage as appropriate. Spearman correlation coefficient was calculated between the 24-h DWI and 5-day FLAIR volumes. Linear regression was performed to estimate the variability in the 5-day FLAIR volumes that could be explained by the 24-h DWI volumes.

Univariate analyses were performed to compare patient cohorts using nonparametric tests (χ2 or Fisher’s exact test) and stratification to late lesion growth versus no late lesion growth and favorable outcome versus unfavorable outcome. Binomial logistic regression including stepwise analysis with enter method was performed to identify predictors of late lesion growth and to explore associations with clinical outcome. Variables with significance (p < 0.1) from the univariate analyses were considered in the regression models, with a criterion of p > 0.01 for removal. IBM SPSS Statistics v19.0 was used for all statistical analyses performed.

One hundred twelve patients met criteria with median age 67 years, 56% female, median admit NIHSS 19, 54% received IV or IA thrombolysis, 66% with M1 occlusion, 4% with tandem occlusion, and median baseline DWI volume 21.2 mL (Table 1). For baseline DWI volume, the pre-EVT DWI volume was used in 87 (78%) of patients, while the early post-EVT DWI was used in 25 (22%) of patients. Successful recanalization was achieved in 87%, and 68% had complete reperfusion, with an overall favorable clinical outcome rate of 53% (Table 1). Additional EVT procedural characteristics, clinical and imaging outcomes, and imaging complications were reported in Table 2. Of note, the median early and late lesion growth volumes were 12 mL and 3.8 mL, respectively, across the entire study cohort (Table 2).

Table 1.

Baseline characteristics of study group and cohorts stratified by late lesion growth

VariableAll patients (n = 112)No late lesion growth (n = 73, 65%)Late lesion growth (n = 39, 35%)p value
Baseline characteristics 
Age 67 [56–75] 64 [56–75] 70 [57–78] 0.16 
Sex (female), n (%) 63 (56) 39 (53) 24 (62) 0.36 
Onset (last seen normal to triage time, min) 184 [83–362] 161 [73–417] 199 [98–311] 0.67 
Thrombolysis, n (%) 61 (54) 39 (53) 22 (56) 0.76 
mTICIc, n (%) 7 (6) 3 (4) 4 (10) 0.19 
2a 7 (6) 2a 6 (8) 2a 1 (3) 
2b 42 38) 2b 24 (33) 2b 18 (46) 
55 (32) 39 (54) 16 (41) 
Occlusion sitea, n (%) eICA 3 (3) eICA 0 (0) eICA 3 (8) 0.12 
iICA 15 (13) iICA 10 (14) iICA 5 (13) 
M1 74 (66) M1 49 (67) M1 25 (64) 
M2 20 (18) M2 14 (19) M2 6 (15) 
Pre-admit mRS 0 [0–0] 0 [0–0] 0 [0–1] 0.11 
Admit NIHSS 19 [11–22] 18 [12–22] 19 [11–24] 0.57 
Baseline DWI volume, mL 21.2 [6.8–45.4] 16.1 [5.2–36.7] 28 [13–71] 0.04 
Pre-EVT Tmax>6 s volume, mLb 82.8 [39.9–146.1] 78.1 [39–139] 84 [45–171] 0.32 
VariableAll patients (n = 112)No late lesion growth (n = 73, 65%)Late lesion growth (n = 39, 35%)p value
Baseline characteristics 
Age 67 [56–75] 64 [56–75] 70 [57–78] 0.16 
Sex (female), n (%) 63 (56) 39 (53) 24 (62) 0.36 
Onset (last seen normal to triage time, min) 184 [83–362] 161 [73–417] 199 [98–311] 0.67 
Thrombolysis, n (%) 61 (54) 39 (53) 22 (56) 0.76 
mTICIc, n (%) 7 (6) 3 (4) 4 (10) 0.19 
2a 7 (6) 2a 6 (8) 2a 1 (3) 
2b 42 38) 2b 24 (33) 2b 18 (46) 
55 (32) 39 (54) 16 (41) 
Occlusion sitea, n (%) eICA 3 (3) eICA 0 (0) eICA 3 (8) 0.12 
iICA 15 (13) iICA 10 (14) iICA 5 (13) 
M1 74 (66) M1 49 (67) M1 25 (64) 
M2 20 (18) M2 14 (19) M2 6 (15) 
Pre-admit mRS 0 [0–0] 0 [0–0] 0 [0–1] 0.11 
Admit NIHSS 19 [11–22] 18 [12–22] 19 [11–24] 0.57 
Baseline DWI volume, mL 21.2 [6.8–45.4] 16.1 [5.2–36.7] 28 [13–71] 0.04 
Pre-EVT Tmax>6 s volume, mLb 82.8 [39.9–146.1] 78.1 [39–139] 84 [45–171] 0.32 

aOcclusion site included 5 patients with tandem occlusions, target site for EVT was reported.

bPre-EVT Tmax was calculated for the 80 patients with available pre-EVT perfusion-weighted imaging.

cmTICI was not reported in 1 patient with no late lesion growth.

Table 2.

EVT characteristics and outcomes of study group and cohorts stratified by late lesion growth

VariableAll patients (n = 112)No late lesion growth (n = 73, 65%)Late lesion growth (n = 39, 35%)p value
EVT procedural characteristics 
 Onset to recanalization (min) 308 [225–499] 301 [220–531] 336 [246–488] 0.34 
 Pre-EVT MRIa to recanalization (min) 108 [82–140] 102 [77–148] 117 [88–132] 0.76 
 General anesthesia used in EVT, n (%) 99 (90) 62 (87) 36 (92) 0.43 
 Multiple passes, n (%) 59 (53) 34 (49) 25 (66) 0.09 
 Successful recanalizationb, n (%) 97 (87) 63 (88) 34 (87) 0.96 
Clinical outcomes 
 24-h NIHSS 9 [3–18] 7 [3–13] 17 [9–22] <0.001 
 mRS at 30 or 90 days 2 [1–4] 2 [1–3] 4 [2–5] <0.001 
 Favorable outcome (mRS 0–2)c, n (%) 59 (53) 47 (64) 12 (31) 0.001 
Imaging outcomes 
 Complete reperfusiond, n (%) 76 (69) 56 (78) 20 (53) 0.007 
 Any HT, n (%) 52 (51) 26 (41) 26 (68) 0.007 
 Parenchymal hematoma, n (%) 10 (10) 4 (6) 6 (16) 0.12 
 Any HARM, n (%) 93 (83) 61 (85) 32 (82) 0.72 
 Severe HARM 65 (58) 40 (56) 25 (64) 0.38 
 24-h DWI volume, mL 38.5 [14.9–90.1] 28.6 [11.8–55.8] 79.2 [34.1–131.3] <0.001 
 Early lesion growth, % 59 (53) 30 (41) 29 (74) 0.001 
 Early lesion volume growth, mL 12 [1.7–33.1] 5.9 [−0.8 to 19] 30.1 [9.4–62] <0.001 
 24-h Tmax >6 s volume, mL 0 [0–7.6] 0 [0–3.4] 2 [0–55.4] 0.002 
 5-day FLAIR volume, mL 46.5 [16.6–110.6] 27.8 [11.7–50.7] 116.2 [62.4–168.4] <0.001 
 Late lesion volume growth, mL 3.8 [−2.1 to 15.4] −0.3 [−5.6 to 3.5] 27.4 [15.2–45.7] --- 
VariableAll patients (n = 112)No late lesion growth (n = 73, 65%)Late lesion growth (n = 39, 35%)p value
EVT procedural characteristics 
 Onset to recanalization (min) 308 [225–499] 301 [220–531] 336 [246–488] 0.34 
 Pre-EVT MRIa to recanalization (min) 108 [82–140] 102 [77–148] 117 [88–132] 0.76 
 General anesthesia used in EVT, n (%) 99 (90) 62 (87) 36 (92) 0.43 
 Multiple passes, n (%) 59 (53) 34 (49) 25 (66) 0.09 
 Successful recanalizationb, n (%) 97 (87) 63 (88) 34 (87) 0.96 
Clinical outcomes 
 24-h NIHSS 9 [3–18] 7 [3–13] 17 [9–22] <0.001 
 mRS at 30 or 90 days 2 [1–4] 2 [1–3] 4 [2–5] <0.001 
 Favorable outcome (mRS 0–2)c, n (%) 59 (53) 47 (64) 12 (31) 0.001 
Imaging outcomes 
 Complete reperfusiond, n (%) 76 (69) 56 (78) 20 (53) 0.007 
 Any HT, n (%) 52 (51) 26 (41) 26 (68) 0.007 
 Parenchymal hematoma, n (%) 10 (10) 4 (6) 6 (16) 0.12 
 Any HARM, n (%) 93 (83) 61 (85) 32 (82) 0.72 
 Severe HARM 65 (58) 40 (56) 25 (64) 0.38 
 24-h DWI volume, mL 38.5 [14.9–90.1] 28.6 [11.8–55.8] 79.2 [34.1–131.3] <0.001 
 Early lesion growth, % 59 (53) 30 (41) 29 (74) 0.001 
 Early lesion volume growth, mL 12 [1.7–33.1] 5.9 [−0.8 to 19] 30.1 [9.4–62] <0.001 
 24-h Tmax >6 s volume, mL 0 [0–7.6] 0 [0–3.4] 2 [0–55.4] 0.002 
 5-day FLAIR volume, mL 46.5 [16.6–110.6] 27.8 [11.7–50.7] 116.2 [62.4–168.4] <0.001 
 Late lesion volume growth, mL 3.8 [−2.1 to 15.4] −0.3 [−5.6 to 3.5] 27.4 [15.2–45.7] --- 

aPre-EVT MRI to recanalization time interval was calculated for the 80 patients with pre-EVT MRI and available recanalization times.

bmTICI was not reported in 1 patient.

cmRS at 90 days was used in 92 (82%) of patients; mRS was available at both 30 and 90 days in 89 of these 92 patients, mRS at 90 days only was available in 3 patients, mRS at 30 days only was available in 21 patients.

dComplete reperfusion was not evaluable in 2 patients; qualitative assessment of perfusion lesion at 24-h was performed in 5 patients due to patient motion limiting the Tmax map calculation.

When stratifying by late lesion growth (Table 2), the majority of patients (65%) did not have late lesion growth after 24-h, with a median decrease in lesion volume of 0.3 mL from 24-h to 5-day (Fig. 1a). The lesion volumes in these patients stabilized by 24-h with an early median growth of only 5.9 mL (Table 2). These findings were consistent with the Spearman correlation coefficient of 0.898 (p < 0.01) calculated between the 24-h and 5-day lesion volumes and R2 of 0.80 based on linear regression, indicating that across the entire study cohort, 80% of the “final” lesion volume had evolved by 24-h. These patients without late lesion growth had a high rate of favorable clinical outcome (64%), suggesting substantial neurologic improvement in the setting of high rates of successful recanalization (88%) and complete reperfusion (78%) (Table 2).

Fig. 1.

a Forty-five-year-old female patient without early or late lesion growth: i. baseline EVT DWI volume = 33.5 mL, ii. 24-h DWI volume = 26.3 mL, early lesion volume change of −7.2 mL, iii. 5-day FLAIR volume = 32.6 mL, late lesion growth of 6.3 mL. Patient had admit NIHSS of 20, was treated with IV thrombolysis, had successful recanalization of the M1 segment with single-pass EVT, mTICI 2b at ∼10 h from onset, complete reperfusion at 24-h, HI-2 at 24-h, and favorable outcome with 30 day mRS of 1. b Sixty-three-year-old female patient with early and late lesion growth: i. baseline DWI volume = 24.4 mL, ii. 24-h DWI volume = 131.3 mL, early growth of 106.9 mL, iii. 5-day FLAIR volume = 163 mL, late growth of 31.7 mL. Patient had admit NIHSS of 14, was not treated with IV thrombolysis, had successful recanalization of the M2 segment with multiple pass EVT, mTICI 2b at ∼2 h from onset, incomplete reperfusion at 24 h, HI-1 at 24 h, and unfavorable outcome with 90-day mRS of 4.

Fig. 1.

a Forty-five-year-old female patient without early or late lesion growth: i. baseline EVT DWI volume = 33.5 mL, ii. 24-h DWI volume = 26.3 mL, early lesion volume change of −7.2 mL, iii. 5-day FLAIR volume = 32.6 mL, late lesion growth of 6.3 mL. Patient had admit NIHSS of 20, was treated with IV thrombolysis, had successful recanalization of the M1 segment with single-pass EVT, mTICI 2b at ∼10 h from onset, complete reperfusion at 24-h, HI-2 at 24-h, and favorable outcome with 30 day mRS of 1. b Sixty-three-year-old female patient with early and late lesion growth: i. baseline DWI volume = 24.4 mL, ii. 24-h DWI volume = 131.3 mL, early growth of 106.9 mL, iii. 5-day FLAIR volume = 163 mL, late growth of 31.7 mL. Patient had admit NIHSS of 14, was not treated with IV thrombolysis, had successful recanalization of the M2 segment with multiple pass EVT, mTICI 2b at ∼2 h from onset, incomplete reperfusion at 24 h, HI-1 at 24 h, and unfavorable outcome with 90-day mRS of 4.

Close modal

Nearly 1/3 of patients (35%) demonstrated significant late lesion growth despite successful recanalization (87% TICI 2b/3) (Fig. 1b). These patients exhibited a notable late lesion growth of 27.4 mL between 24-h to 5-day and also showed median early lesion growth of 30.1 mL between baseline and 24-h (Table 2). The baseline clinical and imaging characteristics including age and admit NIHSS were similar between patients with versus without late lesion growth except for baseline DWI volume, which was larger in patients with late lesion growth versus without, 28 mL versus 16 mL (p = 0.04) (Table 1). Based on univariate analysis, patients with late lesion growth had a lower rate of complete reperfusion 53%, and a higher rate of HT 68% (Table 2). The majority (74%) of patients with late lesion growth also had early lesion growth. Similarly, when comparing patients with versus without early lesion growth, there were no differences (not shown) in baseline characteristics including baseline DWI volume which was slightly larger in patients with early lesion growth 22.9 mL versus 15.2 mL (p = 0.07). Using stepwise regression to predict late lesion growth which included baseline DWI volume (p = 0.33), early lesion growth (p = 0.251), and multiple pass EVT (p = 0.884), only incomplete reperfusion (OR 2.85, p = 0.038, 95% CI: [1.058–7.659]), any HT (OR 3.14, p = 0.018, 95% CI: [1.216–8.103]), and unfavorable outcome (OR 3.67, p = 0.008, 95% CI: [1.413–9.548]) remained in the model and were found to be independent predictors of late lesion growth (online suppl. Table 1).

Approximately 1 out of 3 patients with late lesion growth had favorable clinical outcome (Table 2). Figure 2 illustrates the association between late lesion growth and mRS. Online supplementary Figure II displays the association between early lesion growth and mRS. Univariate analysis comparing patients based on favorable versus unfavorable clinical outcome demonstrated significant differences in age, admit NIHSS, baseline DWI volume, and pre-EVT Tmax>6 s volume (online suppl. Table 2). Rates of complete reperfusion and any HT were not significantly different in the patient cohorts based on clinical outcome (online suppl. Table 2). Half of the patients with unfavorable clinical outcome had late lesion growth, with a median lesion increase of 11.7 mL beyond 24-h versus only 1.4 mL increase in patients with favorable outcome (p = 0.003). Only 20% of patients with favorable outcome versus 51% with unfavorable outcome had late lesion growth (online suppl. Table 2). Using stepwise regression to predict unfavorable clinical outcome which included baseline DWI volume (p = 0.389), early lesion growth (p = 0.132), pre-EVT Tmax>6 s volume (p = 0.531), and 24-h Tmax>6 s volume (p = 0.133), only age (OR 1.05, p = 0.036, 95% CI: [1.003–1.102]), admit NIHSS (OR 1.10, p = 0.014, 95% CI: [1.020–1.191]), and late lesion growth (OR 4.848, p = 0.007, 95% CI: [1.530–15.364]) remained in the model and were found to be independent predictors of unfavorable clinical outcome (online suppl. Table 1).

Fig. 2.

Association of late lesion growth and modified Rankin Scale: lesion volume change for patients between 24-h DWI and 5-day FLAIR versus mRS (30–90 days). 1 patient not shown with mRS of 6; off scale with late lesion growth >100 mL.

Fig. 2.

Association of late lesion growth and modified Rankin Scale: lesion volume change for patients between 24-h DWI and 5-day FLAIR versus mRS (30–90 days). 1 patient not shown with mRS of 6; off scale with late lesion growth >100 mL.

Close modal

Approximately 1 out of 3 patients in this study had late lesion growth following EVT, with a favorable clinical outcome occurring in only 1 out of 3 of these patients. In contrast, patients without late lesion growth had a high rate of favorable clinical outcome post-EVT. Estimating the “final” lesion volume measurement at 24-h is likely too early to serve as an imaging biomarker for clinical outcomes post-EVT, as a substantial proportion of patients continue to experience lesion growth beyond 24-h. Late lesion growth was found to be a significant predictor of unfavorable outcome irrespective of complete reperfusion or any HT.

An advantage of the current study is the protocol standardization [20], which involved the utilization of multimodal MRI for all time points. In contrast, other studies often have combined CT and MRI across various time points with most of them lacking a pre-EVT or early post-EVT time point [7, 12‒14, 17, 22, 23]. When comparing our study results to the meta-analysis conducted by Bala et al. [10], we found consistency in the median baseline volume of 21.2 versus 19.5 mL, median final lesion volume of 46.5 versus 37.5 mL, and median lesion growth volume of 12 versus 14.8 mL. Tate et al. [23] reported median early and late lesion growth volumes of 30.3 mL and 6.7 mL, respectively, whereas our study found median early and late lesion growth volumes of 12 mL and 3.8 mL. However, their study utilized mostly CTP “core” for baseline volume measurements instead of DWI and interchanged DWI and CT for 24-h and 5-day time points. Usage of CTP “core” measurements in ∼80% of patients in the Tate et al. [23] study likely led to an underestimation of baseline core, causing an overestimation of early lesion growth at 24-h. Similar to prior studies, we demonstrated a strong association between lesion volumes at 24-h and 5-day [4, 13, 14, 23]. Furthermore, with the same MRI scanner used for each patient, standardized imaging protocols across both participating sites, and a validated quantitative measurement approach, we attempted to minimize methodological errors.

There are limitations to our study including the applicability of the study to the majority of clinical sites that are not able to acquire pre-EVT MRI. We utilized early post-EVT MRI when pre-EVT MRI was not acquired as the baseline time point. Therefore, categorization of early lesion growth in our study included data from both pre-EVT and early post-EVT time points in the calculation in reference to the 24-h. In a recent study, Hernandez-Perez et al. [24] defined early infarct growth as the difference between the pre-EVT and early (∼2 h) post-EVT time points, reporting a median early infarct growth of 4 mL but ranged from 1 to 42 mL depending on mTICI. We acknowledge a limitation of our study is the inter-rater variability in the angiographic interpretation of mTICI by multiple treating interventionists. However, our study focused on successful reperfusion patients, with mTICI 2b/3 in the large majority of patients, while the Hernandez-Perez et al. [24] study specifically looked at the association of infarct growth across the entire range of mTICI. We found no differences in the pre-EVT DWI volumes based on clinical outcome (online suppl. Table 1), but by including the early post-EVT as baseline in some patients, we found a trend for larger baseline DWI volumes in patients with unfavorable clinical outcome, perhaps reflecting “fast” lesion growth in some patients immediately following the EVT procedure. However, this was not an issue for the categorization of late lesion growth, collected consistently at 24-h, and 5-day post-EVT in all study patients, which was the primary focus of this analysis. Finally, assessment of Late lesion growth within 1 week relative to a post-EVT 24-h MRI may be feasible at most clinical sites.

Another limitation is that we did not differentiate between edema including HT-associated versus infarct expansion [15]. In a small heterogenous patient sample, Harston et al. [15] found that anatomic distortion accounted for 20–36% of lesion expansion across 24-h to 5-day. However as summarized above, in comparison to a meta-analysis by Bala et al. [10] involving relatively homogeneous datasets and focusing on EVT effects, the lesion data reported in this study, including volume and growth, aligns well with existing studies.

Ospel et al. [25] recently provided a comprehensive review of potential reasons and treatment opportunities due to the incongruity between successful recanalization and favorable clinical outcome in a large number of acute stroke patients treated with EVT. Post-EVT cerebroprotection is emerging as a potential opportunity to improve clinical outcomes by reducing secondary injury observed within 1 week. While practical considerations favor follow-up imaging evaluation at 24-h, it may be prudent to consider a second follow-up during the first week, especially in patients with evidence of lesion growth within 24-h. The continued growth of the ischemic lesion past 24-h post-EVT presents an opportunity for further therapeutic intervention to alleviate later consequences of the ischemic stroke- and reperfusion-related injury. We purposively used the term lesion in this study to acknowledge the dynamic nature of the ischemic event, especially in the context of reperfusion effects and the uncertainty of defining final infarct at 24-h.

Our study revealed that one third of patients continue to have late lesion growth beyond 24-h with an associated worse clinical outcome, while two-thirds of patients with successful recanalization have no late lesion growth beyond 24-h post-EVT (and minimal lesion growth up to 24-h post-EVT) with an associated favorable clinical outcome. This late lesion growth is associated with incomplete reperfusion, HT, and unfavorable clinical outcome. Identifying these patients who develop late lesion growth could be critical in guiding clinical management and predicting prognosis post-EVT. Moreover, it may assist in patient selection and serve as an imaging biomarker for the development of adjunctive therapies aimed at mitigating reperfusion injury. Understanding and addressing late lesion growth can potentially lead to improved treatment strategies and better outcomes for patients undergoing EVT.

We thank our patients and their families, without whom this research would not have been possible. We also appreciate the clinicians and research and administrative teams for their support of the Natural History of Stroke study.

This is a retrospective analysis of existing coded data. Ethics approval was obtained for this research activity from the NIH Office of Human Subjects Research Protection (OHSR Exempt No. 13285). The OHSR determination of Not Human Subjects Research is based on the interpretation of 45 CFR 46 under “Research Involving Coded Private Information or Biological Specimens.” This retrospective review of patient data did not require ethical approval in accordance with local/national guidelines. Written informed consent from participants was not required in accordance with local/national guidelines.

The authors have no conflicts of interest to declare.

Financial support for this work was provided by the Intramural Research Program of the National Institute of Neurological Disorders and Stroke, National Institutes of Health.

Marie Luby: study implementation and conceptualization, methodology, data acquisition, formal analysis, and writing – original draft, review, and editing; Amie W. Hsia: data acquisition, study implementation and conceptualization, methodology, formal analysis, and writing – original draft, review, and editing; Carolyn A. Lomahan and Yongwoo Kim: data acquisition, methodology, and review and editing; Victoria Uche: data acquisition; Rachel Davis: data acquisition and methodology; Sana Somani: data acquisition, study implementation, formal analysis, and writing – review and editing; Shannon Burton, Rainier Cabatbat, Veronica Craft, Jill B. De Vis, Malik M. Adil, Mariam M. Afzal, Leila C. Thomas, and John K. Lynch: data acquisition and study implementation; William Gandler and Evan S. McCreedy: image processing development; and Lawrence L. Latour: conceptualization, data acquisition, study implementation, formal analysis, resources, methodology, writing – original draft, and writing – review and editing.

The data that support the findings of this study are not publicly available due to privacy reasons but are available from the corresponding author upon reasonable request. Further inquiries can be directed to the corresponding author.

1.
Tourdias
T
,
Renou
P
,
Sibon
I
,
Asselineau
J
,
Bracoud
L
,
Dumoulin
M
, et al
.
Final cerebral infarct volume is predictable by MR imaging at 1 week
.
AJNR Am J Neuroradiol
.
2011
;
32
(
2
):
352
8
. doi: .
2.
Gaudinski
MR
,
Henning
EC
,
Miracle
A
,
Luby
M
,
Warach
S
,
Latour
LL
.
Establishing final infarct volume: stroke lesion evolution past 30 days is insignificant
.
Stroke
.
2008
;
39
(
10
):
2765
8
. doi: .
3.
Krongold
M
,
Almekhlafi
MA
,
Demchuk
AM
,
Coutts
SB
,
Frayne
R
,
Eilaghi
A
.
Final infarct volume estimation on 1-week follow-up MR imaging is feasible and is dependent on recanalization status
.
Neuroimage Clin
.
2015
;
7
:
1
6
. doi: .
4.
Campbell
BC
,
Tu
HT
,
Christensen
S
,
Desmond
PM
,
Levi
CR
,
Bladin
CF
, et al
.
Assessing response to stroke thrombolysis: validation of 24-hour multimodal magnetic resonance imaging
.
Arch Neurol
.
2012
;
69
(
1
):
46
50
. doi: .
5.
Zaidi
SF
,
Aghaebrahim
A
,
Urra
X
,
Jumaa
MA
,
Jankowitz
B
,
Hammer
M
, et al
.
Final infarct volume is a stronger predictor of outcome than recanalization in patients with proximal middle cerebral artery occlusion treated with endovascular therapy
.
Stroke
.
2012
;
43
(
12
):
3238
44
. doi: .
6.
Yoo
AJ
,
Chaudhry
ZA
,
Nogueira
RG
,
Lev
MH
,
Schaefer
PW
,
Schwamm
LH
, et al
.
Infarct volume is a pivotal biomarker after intra-arterial stroke therapy
.
Stroke
.
2012
;
43
(
5
):
1323
30
. doi: .
7.
MacLellan
A
,
Mlynash
M
,
Kemp
S
,
Ortega-Gutierrez
S
,
Heit
JJ
,
Marks
MP
, et al
.
Perfusion imaging collateral scores predict infarct growth in non-reperfused DEFUSE 3 patients
.
J Stroke Cerebrovasc Dis
.
2022
;
31
(
1
):
106208
. doi: .
8.
Seners
P
,
Oppenheim
C
,
Turc
G
,
Albucher
JF
,
Guenego
A
,
Raposo
N
, et al
.
Perfusion imaging and clinical outcome in acute ischemic stroke with large core
.
Ann Neurol
.
2021
;
90
(
3
):
417
27
. doi: .
9.
Boers
AMM
,
Jansen
IGH
,
Brown
S
,
Lingsma
HF
,
Beenen
LFM
,
Devlin
TG
, et al
.
Mediation of the relationship between endovascular therapy and functional outcome by follow-up infarct volume in patients with acute ischemic stroke
.
JAMA Neurol
.
2019
;
76
(
2
):
194
202
. doi: .
10.
Bala
F
,
Ospel
J
,
Mulpur
B
,
Kim
BJ
,
Yoo
J
,
Menon
BK
, et al
.
Infarct growth despite successful endovascular reperfusion in acute ischemic stroke: a meta-analysis
.
AJNR Am J Neuroradiol
.
2021
;
42
(
8
):
1472
8
. doi: .
11.
Federau
C
,
Mlynash
M
,
Christensen
S
,
Zaharchuk
G
,
Cha
B
,
Lansberg
MG
, et al
.
Evolution of volume and signal intensity on fluid-attenuated inversion recovery MR images after endovascular stroke therapy
.
Radiology
.
2016
;
280
(
1
):
184
92
. doi: .
12.
Sarraj
A
,
Hassan
AE
,
Grotta
J
,
Blackburn
S
,
Day
A
,
Abraham
M
, et al
.
Early infarct growth rate correlation with endovascular thrombectomy clinical outcomes: analysis from the SELECT study
.
Stroke
.
2021
;
52
(
1
):
57
69
. doi: .
13.
Bucker
A
,
Boers
AM
,
Bot
JCJ
,
Berkhemer
OA
,
Lingsma
HF
,
Yoo
AJ
, 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
(
5
):
1233
40
. doi: .
14.
Federau
C
,
Christensen
S
,
Mlynash
M
,
Tsai
J
,
Kim
S
,
Zaharchuk
G
, et al
.
Comparison of stroke volume evolution on diffusion-weighted imaging and fluid-attenuated inversion recovery following endovascular thrombectomy
.
Int J Stroke
.
2017
;
12
(
5
):
510
8
. doi: .
15.
Harston
GWJ
,
Carone
D
,
Sheerin
F
,
Jenkinson
M
,
Kennedy
J
.
Quantifying infarct growth and secondary injury volumes: comparing multimodal image registration measures
.
Stroke
.
2018
;
49
(
7
):
1647
55
. doi: .
16.
Abdelkhaleq
R
,
Kim
Y
,
Khose
S
,
Kan
P
,
Salazar-Marioni
S
,
Giancardo
L
, et al
.
Automated prediction of final infarct volume in patients with large-vessel occlusion acute ischemic stroke
.
Neurosurg Focus
.
2021
;
51
(
1
):
E13
. doi: .
17.
Wouters
A
,
Robben
D
,
Christensen
S
,
Marquering
HA
,
Roos
YBWEM
,
van Oostenbrugge
RJ
, et al
.
Prediction of stroke infarct growth rates by baseline perfusion imaging
.
Stroke
.
2022
;
53
(
2
):
569
77
. doi: .
18.
Luby
M
,
Merino
JG
,
Davis
R
,
Ansari
S
,
Fisher
M
,
Hsia
AW
, et al
.
Association of multiple passes during mechanical thrombectomy with incomplete reperfusion and lesion growth
.
Cerebrovasc Dis
.
2022
;
51
(
3
):
394
402
. doi: .
19.
Luby
M
,
Hsia
AW
,
Lomahan
CA
,
Davis
R
,
Burton
S
,
Kim
Y
, et al
.
Post-ischemic hyperemia following endovascular therapy for acute stroke is associated with lesion growth
.
J Cereb Blood Flow Metab
.
2023
;
43
(
6
):
856
68
. doi: .
20.
Warach
SJ
,
Luby
M
,
Albers
GW
,
Bammer
R
,
Bivard
A
,
Campbell
BCV
, et al
.
Acute stroke imaging research roadmap III imaging selection and outcomes in acute stroke reperfusion clinical trials: consensus recommendations and further research priorities
.
Stroke
.
2016
;
47
(
5
):
1389
98
. doi: .
21.
Luby
M
,
Bykowski
JL
,
Schellinger
PD
,
Merino
JG
,
Warach
S
.
Intra- and interrater reliability of ischemic lesion volume measurements on diffusion-weighted, mean transit time and fluid-attenuated inversion recovery MRI
.
Stroke
.
2006
;
37
(
12
):
2951
6
. doi: .
22.
Xie
Y
,
Oppenheim
C
,
Guillemin
F
,
Gautheron
V
,
Gory
B
,
Raoult
H
, et al
.
Pretreatment lesional volume impacts clinical outcome and thrombectomy efficacy
.
Ann Neurol
.
2018
;
83
(
1
):
178
85
. doi: .
23.
Tate
WJ
,
Polding
LC
,
Christensen
S
,
Mlynash
M
,
Kemp
S
,
Heit
JJ
, et al
.
Predictors of early and late infarct growth in DEFUSE 3
.
Front Neurol
.
2021
;
12
:
699153
. doi: .
24.
Hernandez-Perez
M
,
Werner
M
,
Remollo
S
,
Martín
C
,
Cortés
J
,
Valls
A
, et al
.
Early and delayed infarct growth in patients undergoing mechanical thrombectomy: a prospective, serial MRI study
.
Stroke
.
2023
;
54
(
1
):
217
25
. doi: .
25.
Ospel
J
,
Rex
N
,
Kandregula
S
,
Goyal
M
.
The vessel has been recanalized: now what
.
Neurotherapeutics
.
2023
;
20
(
3
):
679
92
. doi: .