Heterogeneity of the Relative Benefits of TICI 2c/3 over TICI 2b50/2b67

Purpose Incomplete reperfusion after mechanical thrombectomy (MT) is associated with a poor outcome. Rescue therapy would potentially benefit some patients with an expanded treatment in cerebral ischemia score (eTICI) 2b50/2b67 reperfusion but also harbors increased risks. The relative benefits of eTICI 2c/3 over eTICI 2b50/67 in clinically important subpopulations were analyzed. Methods Retrospective analysis of our institutional database for all patients with occlusion of the intracranial internal carotid artery (ICA) or the M1/M2 segment undergoing MT and final reperfusion of ≥eTICI 2b50 (903 patients). The heterogeneity in subgroups of different time metrics, age, National Institutes of Health Stroke Scale (NIHSS), number of retrieval attempts, Alberta Stroke Programme Early CT Score (ASPECTS) and site of occlusion using interaction terms (pi) was analyzed. Results The presence of eTICI 2c/3 was associated with better outcomes in most subgroups. Time metrics showed no interaction of eTICI 2c/3 over eTICI 2b50/2b67 and clinical outcomes (onset to reperfusion pi = 0.77, puncture to reperfusion pi = 0.65, onset to puncture pi = 0.63). An eTICI 2c/3 had less consistent association with mRS ≤2 in older patients (>82 years, pi = 0.038) and patients with either lower NIHSS (≤9) or very high NIHSS (>19, pi = 0.01). Regarding occlusion sites, the beneficial effect of eTICI 2c/3 was absent for occlusions in the M2 segments (aOR 0.73, 95% confidence interval [CI] 0.33–1.59, pi = 0.018). Conclusion Beneficial effect of eTICI 2c/3 over eTICI 2b50/2b67 only decreased in older patients, M2-occlusions and patients with either low or very high NIHSS. Improving eTICI 2b50/2b67 to eTICI 2c/3 in those subgroups may be more often futile. Supplementary Information The online version of this article (10.1007/s00062-021-01131-0) contains supplementary material, which is available to authorized users.

It remains less clear which patients with intraprocedural eTICI 2b50/2b67 reperfusion would benefit from rescue therapy and in which patients additional retrieval attempts may be futile/unnecessary and expose patients to unnecessary risks [11]. Recent work indicated that time metrics (particularly time from groin puncture to reperfusion) play a critical role in the effectiveness of complete reperfusion [12].
With the intention to explore a potential heterogeneity of the relative benefits of eTICI 2c/3 over eTICI 2b50/2b67, we analyzed outcome differences across different reperfusion grades in clinically important subpopulations.

Study Population
We performed a retrospective analysis of our institutional database including all patients with occlusion of the intracranial internal carotid artery (ICA) or the M1/M2 segment, undergoing endovascular stroke treatment and final reperfusion degree of ≥50% (≥eTICI 2b50) between 1 st January 2010 and 31 st December 2018. Patients with incomplete/insufficient angiographic documentation of preinterventional and postinterventional digital subtraction angiography (DSA) runs as well as incomplete clinical history were excluded (Fig. 1).
In the first step, three neuroradiologists independently evaluated the eTICI scores for 142 randomly selected patients based on preprocedural and postprocedural DSA runs, with both an anterior-posterior and a lateral projection. They were blinded to other clinical, imaging, and outcome data. Since good interrater reliability was achieved, the remaining cases (n = 761/903) were evaluated independently by core laboratory judication in the same fashion.
Collateral status was graded by two neuroradiologists based on the preinterventional DSA runs, blinded to all clinical and outcome information. Collaterals were graded according to the American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology (ASITN/SIR) scale: 0-No collaterals visible to ischemic site, 1-Slow collaterals to the periphery of ischemic site, with persistence of some of the defect, 2-Rapid collaterals to the periphery of ischemic site, with persistence of some of the defect, and to only a portion of the ischemic territory, 3-Collaterals with slow but complete angiographic blood flow of the ischemic bed by the late venous phase, 4-Complete and rapid collateral blood flow to the vascular bed in the entire ischemic territory by retrograde perfusion [14].

Statistical Analysis
Baseline and outcome parameters between the groups were compared using the Kruskal-Wallis test for continuous variables or the Fisher's exact test for categorical variables. Data are displayed as median (interquartile range [IQR]) and n (%) if not otherwise specified. Interrater reliability was assessed using Krippendorff's alpha for multiple raters. Krippendorff's alpha is a coefficient representing the interrater reliability for any number of raters and any kind of variables. Values were interpreted as follows: 0 = poor; 0.01-0.20 = slight; 0.21-0.40 = fair; 0.41-0.60 = moderate; 0.61-0.80 = substantial and 0.81-1.0 = almost perfect agreement.
The distribution of clinical outcomes (mRS at 90 days after the intervention) for each eTICI was depicted graphically. Furthermore, binary logistic regression modeling was performed to assess the association of eTICI and clinical outcome (dichotomized into mRS ≤2 = good outcome versus mRS >2 = poor outcome [15]) and to account for potential prespecified confounders based on the literature (age, sex, NIHSS on admission, number of retrieval attempts, site of intracranial occlusion, general anesthesia, time from onset to groin puncture and ASPECTS).
The cases were then grouped into eTICI 2b50/2b67 and eTICI 2c/3. A multiplicative interaction term (eTICI* variable) was used to test for interaction with the following variables: continuous time metrics (onset to reperfusion time, puncture to reperfusion time and onset to puncture time), age on admission, NIHSS on admission, number of retrieval attempts, ASPECTS, ASITN/SIR DSA collateral score and site of occlusion.
To assess heterogeneity, subgroups were defined based on their quartiles for time metrics, age, and NIHSS; for ASPECTS as 0-4, 5-7, and 8-10 [16]; for collaterals ASITN/SIR 0/1, 2, and 3/4 (due to the small number of cases with ASITN/SIR 0 and 4, respectively); and for the number of retrieval attempts as 1, 2, and ≥3. Finally, marginal effects were plotted for time metrics and age as well as point estimates for ASPECTS.
All statistical analyses were conducted using R (Version 4.0.2, R Foundation for Statistical Computing, Vienna, Austria) [17]. A two-tailed p-value of <0.05 was considered statistically significant.

Population
A total of 903 patients with occlusion of the intracranial ICA or M1/M2 segments (median age 75 years, IQR 63-82 years, 51% female) were included. Baseline characteristics are shown in Table 1. Group sizes were almost balanced, only the eTICI 2b50 group was smaller. With higher degrees of reperfusion, there was a trend towards shorter time from symptom onset to groin puncture (eTICI 2b50

Heterogeneity of the Association of TICI2c/3 and good Clinical Outcomes with Strata of Time Metrics
Time metrics as continuous variables showed no interaction with dichotomized eTICI scores (eTICI 2b50/2b67 vs. eTICI 2c/3; p for interaction: time from onset to reperfusion p = 0.77, time from groin puncture to reperfusion p = 0.65, time from onset to groin puncture p = 0.63). The adjusted odds ratios (aOR) for mRS 90-day ≤2 stratified by dichotomized eTICI scores slightly changed between different quartiles but showed no clear trend (Fig. 3).
The change of probability of a good clinical outcome (mRS-90d ≥2) over time courses stratified by dichotomized eTICI scores are shown in Fig. 4 and show stable superiority of eTICI 2c/3 over eTICI 2b50/2b67.
We also observed a decreasing proportion of patients with eTICI 2c/3 reperfusion with increasing time metrics,
The change of probability of a good clinical outcome (mRS ≤2) over age (Fig. 5) and ASPECTS (Fig. 6) stratified by dichotomized eTICI scores again showed stable superiority of eTICI 2c/3 over eTICI 2b50/2b67 with the exception of older patients and patients with ASPECTS 10.

Discussion
This study has the following main findings: 1) better eTICI scores were associated with better clinical outcomes across several subgroups, including a large range of early ischemic changes and various time courses, and 2) a decreasing beneficial effect of eTICI 2c/3 over eTICI 2b50/2b67 was only found in older patient groups, more distal occlusions and patients with either low or very high NIHSS.

Effect of near Complete/Complete Reperfusion in Subgroups
In line with previous work [1,10], we found better outcomes for patients with ≥eTICI 2b67 reperfusion, when compared to patients with eTICI 2b50 reperfusion.
We did not find interactions of various time courses with clinical outcome after 90 days, contradicting the results of the study of Kitano et al. [12]. They found decreasing association of good clinical outcome of mTICI 3 reperfusion over mTICI 2b after ≥80 min puncture to reperfusion time. It should be noted that our quartiles were different, with the upper quartile being >65 min puncture to reperfusion time, indicating differences between the two cohorts; however, with prolonged time courses, we found overall decreasing beneficial effects of successful reperfusion and also a decreasing likelihood of eTICI 2c/3 reperfusion. These findings make it reasonable to assume that longer onset to admission and/or longer procedure times promote tissue infarction and worse clinical outcomes as described before [18][19][20][21].
We found a trend of decreasing beneficial effect of complete recanalization in older patient groups, with the biggest leap being in the group with patients older than 82 years. This phenomenon is well known and probably relies on multiple factors, such as higher prevalence of comorbidities, higher rates of polypharmacy, and reduced neuronal plasticity [22][23][24]. Moreover, there is generally a less stringent association of successful reperfusion, initial neurological improvement and mid-term functional outcome in older patients [25].
Another interesting finding is that patients with either low NIHSS (≤9) or very high NIHSS (>19) did not benefit from eTICI 2c/3 as much as the group with NIHSS 10-19. A possible explanation for this could be that patients presenting with low NIHSS were more likely to have good to excellent collateral status and patients presenting with very high NIHSS were more likely to have poor collateral status. Previous studies showed the association of adequate collaterals and good tissue outcome [26,27] and our results are in line with these findings, as the difference of eTICI 2b50/2b67 to eTICI 2c/3 reperfusion did not play a crucial role in patients with adequate collaterals. Also in favor of this hypothesis, patients in our cohort with NIHSS ≤5 benefited less from eTICI 2c/3 than patients with NIHSS 6-9. On the other hand, patients with poor collaterals benefited significantly more from a final eTICI 2c/3 than patients with good to excellent collaterals, underlining the importance of complete or near complete reperfusion especially in patients with poor collaterals.
While a beneficial effect of eTICI 2c/3 on the mRS at 90 days was present in ICA/T and M1 occlusions, it was absent in M2 occlusions. Obviously, eTICI 2b50/2b67 perfusion deficits after M2 occlusions are significantly smaller compared to the range of possible perfusion deficits in eTICI 2b50/2b67 reperfusion after e.g. M1 occlusions. Other possible explanations are different collateral situations in medium vessel occlusions and smaller amounts of affected brain tissue and therefore having a lower probability to affect more eloquent brain areas. In this sense, reliable measurement of the volume of remaining hypoperfused tissue during mechanical thrombectomy (e.g. with flat-panel CT perfusion) could help in deciding whether to extend or discontinue the procedure.

Possible Implications on Patient Selection
Regarding extended mechanical thrombectomy for improving eTICI 2b50/2b67 to eTICI 2c/3, it is still not possible to draw clear boundaries for proper patient selection. In an individual case-based decision the operator may weigh in the likelihood of the improvement being clinically futile, which seems to occur more often in older patients, M2 occlusions and patients with either low or very high NIHSS.

Limitations
The retrospective analysis of data from a single center implicates limitations of our study and generalizability of our results on other cohorts has to be validated. Furthermore, we found shorter time from symptom onset to groin puncture and lower number of retrieval attempts in patients with higher degrees of reperfusion, possibly biasing our findings. While we have corrected for these confounders in all presented logistic regression analyses, there was high collinearity between these variables and the possibility of residual and hidden confounding remains.
The results of our study should be interpreted with caution. It is not clear whether secondary established eTICI 2c/3 (after initial eTICI 2b50/2b67 reperfusion) over K eTICI 2b50/2b67 has the same beneficial effects as first pass eTICI 2c/3, but recent reports seem promising [28,29]. On the other hand, a lack of benefit of primary eTICI 2c/3 reperfusion over eTICI 2b50/2b67 indicates little to no benefit in secondary established eTICI 2c/3 reperfusion.
A particular strength of our study is the consistent core laboratory judication of the eTICI scores of our large sample size.

Conclusion
The beneficial effect of eTICI 2c/3 over eTICI 2b50/2b67 was stable over various time metrics. It only decreased in older patient groups, M2 occlusions, patients with either low or very high NIHSS and patients with good to excellent collaterals. Improving eTICI 2b50/2b67 to eTICI 2c/3 in those subgroups may be more often futile, which should be acknowledged when making individualized decisions regarding potential improvements from eTICI 2b50/2b67 to eTICI 2c/3.

Supplementary Information
The online version of this article (https:// doi.org/10.1007/s00062-021-01131-0) contains supplementary material, which is available to authorized users. Ethical standards This study was performed in line with the principles of the Declaration of Helsinki. Institutional Review Board approval was obtained. Consent to participate: written informed consent was obtained from all subjects (patients) in this study. Consent for publication: not applicable.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4. 0/.