Herein, we have shown that 44% of patients with ACLVO-related AIS failed to achieve long-term functional independence after MT. Similar to other studies [17–20], this study concluded that predictors of poor clinical outcomes in multivariable analysis were related to stroke severity (low-level ASPECTS, high NIHSS score) and baseline characteristics (older age, higher HbA1c concentration, ICA + MCA non-occlusion). Furtherly, we found that absent HMCAS and early imaging signs of infarct were also independent predictors of clinical outcomes after MT.
MT has been found beneficial in most patients with ischemic stroke triggered by ACLVO, gaining wide acceptance and use in clinical practice [4]. However, the clinical merit of this approach is time dependent, so it is important to quickly and efficiently identify qualifying patients. NCCT imaging is generally undertaken to determine prognosis and select therapeutic candidates, achieving early detection of intracranial thrombi and gauging infarct extent [21]. Consequently, NCCT is still the most commonly used diagnostic method during admissions for stroke. Analyzing the relations between imaging indicators on NCCT and clinical prognosis may help delineate patient status and improve treatment effects.
ASPECTS is a 10-point semi-quantitative system devised to assess early ischemic changes on NCCT after anterior circulation AIS. As such, it may be utilized clinically to discern the most appropriate patients for stroke therapy [22]. Our findings indicate that patients are less likely to benefit from MT as the ASPECTS ranking declines. This observation is consistent with results of a subgroup analysis in the randomized, phase 3 MR CLEAN trials, showing a relation between baseline ASPECTS value and intra-arterial treatment effect. Past studies have also demonstrated that ASPECTS status correlates with infarct size, lower values denoting more extensive damage [23]. Higher (vs lower) ASPECTS values seem to reflect significantly better outcomes after endovascular thrombectomy as well [17].
The NIHSS score is another incremental measurement scale for stroke severity, serving as an independent predictor of functional outcomes. Higher NIHSS scores are known to correspond with more severe neurologic disability [22]. In multiple regression analysis of the NASA registry, NIHSS scoring was strongly linked with outcomes of MT, increasing the risk of a poor outcome approximately 4-fold [24]. We have similarly found the NIHSS score is an independent predictor of MT prognosis (OR = 1.169, 95%CI: 1.072–1.275; p < 0.001).
An important finding of the present study is that more patients of the poor (vs good) prognosis group (> 2 vs 0–2 mRS scores) showed early imaging signs of infarct (78.3% vs 46.9%; p = 0.001). In logistic regression, the latter also emerged as strongly predictive of MT outcomes (OR = 2.803, 95%CI: 1.289–6.096; p = 0.009). NCCT changes of cerebral infarction may be subtle during the first few hours after AIS, yet these vague irregularities are likely highly predictive of clinical outcomes [15]. Slightly density of lentiform nucleus and insular or convex cortex may be noted within hours after ischemic stroke onset, at times accompanied by nominal mass effects (such as narrowed sulci). The European Cooperative Acute Stroke Study (ECASS) has confirmed the importance of early CT ischemic changes in predicting intravenous thrombolytic benefit, demonstrating that patients with AIS and early imaging signs of infarct are at a higher risk of hemorrhagic transformation, cerebral edema, and thus poor prognosis [25]. Furthermore, the highly effective reperfusion efforts recorded in multiple endovascular stroke trials (HERMES) suggest that in patients with low ASPECTS values and hypodensity, sICH is four times more frequent, with more than one-third of ischemic MCA are as involved [26].
The HMCAS signals an embolus of MCA segment, seen as increased attenuation/density on NCCT. In the pre-thrombectomy era (or in absence of treatment), its presence was a hall mark of severe cerebral ischemia and poor functional outcomes in patients with strokes [27]. Our multivariate regression model has shown that HMCAS negatively correlates with clinical outcomes in patients with AIS undergoing MT (OR = 0.454, 95%CI: 0.237–0.870; p = 0.017). This particular finding is nonetheless supported by Kiddy et al., having claimed greater odds of disability and death in patients lacking HMCAS on admission NCCT studies [13]. The pathophysiologic underpinnings of our data may be rooted in components of the embolus itself. Upon systematic review, it seems that the HMCAS likely depicts an erythrocyte-predominant clot; whereas in its absence, afibrin/platelet-predominant clot prevails [28]. Kim et al. have suggested that despite the low prognostic implications of HMCAS in clinical outcomes after MT, its absence may imply in situ thrombotic occlusion due to atherosclerotic stenosis [29]. The less calcific components of atherosclerotic plaque ostensibly are devoid of density on NCCT [28]. We suspect that erythrocyte-rich emboli due to atherosclerosis are less calcific than fibrin/platelet-rich emboli and may be removed by MT more efficiently to achieve vascular recanalization.
Clinically, HbA1c determinations reflect average blood glucose levels during prior 3-month intervals (roughly), identifying patients with impaired glucose metabolism and chronic hyperglycemia [30]. High HbA1c concentration also emerged as an independent predictor of poor 90-day functional outcomes in our multivariable logistic regression model (OR = 1.275, 95%CI: 1.073–1.515; p = 0.006). This result is aligned with other research on patients undergoing endovascular thrombectomy, in whom every 10 mmol/mol increase in HbA1c heightened the odds of sICH, mortality, and functional independence by 33%, 26%, and 24%, respectively [18]. Precisely why elevated HbA1c adversely impacts clinical outcomes of thrombolysis or endovascular thrombectomy in this setting is perhaps the potential for post-stroke hyperglycemia to worsen reperfusion injury [11, 31]. Hyperglycemia is apt to enhance intracellular acidosis in the ischemic penumbra, with formation of free radicals. This then results in mitochondrial dysfunction and energy failure, which further exacerbate brain injury. Cerebral auto regulation may also be affected, inducing reperfusion damage and potential hemorrhagic change in infarcts. Moreover, hyperglycemia is an activator of matrix metalloproteinase 9, serving to aggravate malignant CED. Hence, high levels of HbA1c on admission should alert physicians treating post-MT malignant complications (intra cerebral hemorrhage and CED) to the likelihood of poor long-term functional independence.
Although we did not exclude elderly patients, studies on MT often set an upper age limit for participation. The mean age of our subjects overall was 64.0 years, rising significantly in the poor prognosis group (68.2 vs 60.2; p < 0.001). In multivariate logistic regression analysis, age was a strong predictor of clinical outcome in patients with ACLVO-related AIS undergoing MT (OR = 1.081, 95%CI: 1.047–1.115; p < 0.001). Many investigations have corroborated that advanced age is a key predictor of poor prognosis in patients with strokes. Unlike younger counterparts, older patients have poorer clinical outcomes due to more frequent adverse comorbidities and less functional reserve [32]. Relevant studies have shown that endovascular thrombectomy may significantly improve clinical outcomes for advanced age patients of this sort, more so than intravenous thrombolysis [19]. However, higher age is still linked to a lower rate of functional independence and higher rates of sICH and mortality, regardless of treatment modality. Also, vascular access may not be achievable in elderly patients due to aging-related stiffness and to rtuosity of the aorta and large vessels [20].
Our study has certain notable limitations. This was a single-center, observational, and retrospective analysis of patients meeting current MT guidelines. The results are therefore not applicable beyond the boundaries specified. Our sampling of patients with ASPECTS of 0–5 was also undersized; and our analysis did not include MRI related indicators (DWI volume, DWI-ASPECTS and DWI–perfusion-weighted imaging mismatch) because lack of whole pre-treatment DWI data. The blood flow reperfusion and the status of vessel reflows were not completely evaluated because of some patients did not perform CTP examination after MT. A multicenter prospective study addressing more imaging factors is warranted to confirm the results we obtained.