We retrospectively analyzed a prospective cohort of patients with AIS who underwent MT at Jilin University First Hospital between January 2022 and January 2024. The First Hospital of Jilin University’s Ethics Committee approved the study protocol. The need for informed consent was waived due to the anonymity of the clinical data analysis. The research procedures adhered to the ethical standards outlined in the Declaration of Helsinki. The inclusion criteria were acute large-vessel occlusive stroke in the anterior circulation, confirmed via imaging techniques such as magnetic resonance imaging (MRI), computed tomography (CT), and intraoperative digital subtraction angiography. We excluded patients with (1) spontaneous intracranial hemorrhage (e.g., hypertensive cerebral hemorrhage, ruptured aneurysm hemorrhage) confirmed by preoperative CT/MRI of the head while those with postoperative hemorrhagic transformation (e.g., symptomatic intracranial hemorrhage [sICH]) were retained; (2) acute posterior circulation ischemic stroke; (3) a pre-stroke modified Rankin Scale score ≥2, as these patients were considered to have pre-existing disability, which could confound the results of this study; and (4) missing essential clinical information, such as follow-up information or critical baseline information.

Demographic and baseline clinical characteristics of the patients were collected by reviewing electronic medical records. These including sex, age, cholesterol levels (total cholesterol [TC], TG, high- [HDL-C] and low-density lipoprotein cholesterol [LDL-C]), systolic and diastolic blood pressure (SBP and DBP), homocysteine, urea nitrogen, serum creatinine, BG, glycosylated hemoglobin, and various medical conditions including hypertension, diabetes, atrial fibrillation, transient ischemic attack, wake up stroke, intracerebral hemorrhage, and coronary artery disease.

Additional information included the smoking status, drinking status, blood pressure upon admission, baseline NIHSS score, Alberta Stroke Project Early Computed Tomography Score (ASPECTS) upon admission, site of vascular occlusion, stroke etiology, use of intravenous thrombolysis or tirofiban, anesthesia modality, American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology (ASITN/SIR) score <2, and presence of tandem lesions. Laboratory tests, including BG, glycosylated hemoglobin, and blood pressure measurements, were performed within 24 h of admission. Stroke etiology was categorized based on the Trial of Heparin-like Drug Therapy for Acute Ischemic Stroke Staging. The duration between stroke inception and revascularization was recorded. sICH was defined as intracranial hemorrhage confirmed by cranial CT/MRI within 24 h of surgery, accompanied by an increase of ≥4 points in NIHSS score or new neurological deficits. If the time of onset was unknown, it was considered to be the last known normal time. All imaging was assessed independently by two blinded neuroradiologists. A third investigator assessed and confirmed the results in cases of inconsistent assessment.

TG and BG were collected from fasting venous blood at the time of admission (prior to thrombectomy), before the patients received any intravenous fluids containing glucose. The BG concentration was measured using the enzymatic hexokinase method. TG, TC, LDL-C, and HDL-C, were assayed enzymatically with a fully automated biochemistry analyzer. The TyG index was calculated using the following formula: ln [fasting TGs (mg/dL) × BG (mg/dL)/2] (20). Because TG, BG, and TyG are highly correlated, the model prioritizes the inclusion of the composite indicator TyG to simplify the analysis.

The key efficacy indicator was the 90-day functional independence of patients post-thrombectomy, reflected by an mRS score of 0–2. Secondary efficacy indicators encompassed the 90-day mRS, 24-h postoperative NIHSS, and discharge NIHSS scores. Safety metrics included surgery-related complications (arterial perforation, distal embolization, and arterial entrapment), any intracranial hemorrhage (ICH) within 48 h, sICH within 24 h, and 90-day mortality. The mRS is scored on a 7-point categorical scale assessing neurological disability from asymptomatic (0) to death (6). Functional improvement is defined as a one-grade mRS score reduction (21). The 90-day mRS score was obtained through telephone or outpatient follow-up, and the mRS scores were assessed by a blinded neurologist trained in the neurology department.

To assess quantitative variables, the Shapiro–Wilk test was initially employed to verify normality, revealing that none of the variables in this study were normally distributed. Median and interquartile ranges (interquartile ranges [IQRs]) were used for non-normal distributions, whereas the Kruskal–Wallis H or Wilcoxon Mann–Whitney U tests were employed for comparative analyses.

Qualitative variables are reported as percentages which were subject to one-way comparisons using either Fisher’s exact or chi-square tests. Binary logistic regression analysis was used to evaluate the effect of varying TyG levels on primary, secondary, and safety outcomes. Clinically pertinent factors and significant univariate covariates (p < 0.05) were integrated into the logistic regression models. We adjusted for covariates such as age, sex, SBP, diabetes, coronary artery disease, occlusion site, stroke etiology, baseline NIHSS scores, recanalization rates, and ASPECTS scores. Associations between TyG groups and functional independence were assessed using binary logistic regression. To analyze the associations between TyG and mRS scores, an ordered logistic regression model was used, whereas linear regression was used to evaluate the correlation between TyG and NIHSS scores post-surgery and at discharge. Sensitivity analyses via subgroup analyses were used to validate the consistency in outcomes between the TyG levels and functional independence in patients with diverse baseline characteristics. Figure 1 presents all odds ratios (ORs) with 95% confidence intervals (CIs).

Statistical evaluations were performed using SPSS Statistics v. 25 (IBM Corp.), with significance levels determined bilaterally and statistical significance set at p < 0.05 for all conducted tests. Multiple imputation (n = 5) was employed to handle missing data. The proportion of missing data for key variables ranged from 0 to 10%. No permissible range for missing values was pre-specified during the study design phase, but the decision to use imputation was based on the observed patterns of missingness in the dataset.

A cohort of 356 patients with acute anterior circulation lesions underwent MT from January 2022 to January 2024. The 90-day functional prognosis was obtained by outpatient or telephone follow-up with a mean follow-up time of 90 ± 7 days. Thirty-six patients with missing important clinical data were excluded; consequently, 320 patients (median age, 62 [IQR, 54–70] years) who underwent emergency anterior circulation MT were included, of which 243 were men (75.9%). The median ASPECTS score was 8 (IQR, 7–10), and the median baseline NIHSS score was 13 (IQR, 10–16). The median TyG levels were 9.01 (IQR, 8.58–9.48). The TyG index was categorized into two levels based on the median value: 160 patients in the first ([Q1], 7.67–9.01) group and 160 patients in the second ([Q2], 9.02–11.71) group. Functional independence was achieved by 108 patients, and 39 patients died.

The baseline characteristics of patients grouped by median TyG are presented in Table 1 Significant differences (p < 0.05) were observed between the two groups in terms of age, SBP, DBP, TC, TG, LDL-C, glucose, HbA1C, hypertension, diabetes, baseline NIHSS score, and ASITN/SIR <2 scores. Differences in the remaining indices were not significant. The TyG index was positively associated with TC, LDL-C, TG, SBP, DBP, glucose, HbA1C, hypertension, diabetes, and baseline NIHSS scores and negatively correlated with age and ASITN/SIR <2 scores. Additional comprehensive patient characteristics are outlined in Table 1.

Table 2 shows the association between TyG levels and clinical outcomes across different groups. The percentage of patients achieving an mRS score of 0–2 at 90 days at different TyG levels reflects 90-day functional independence. The unadjusted model revealed that higher TyG indices were correlated with a reduced likelihood of achieving functional independence at 90 days post-event (Table 2). A TyG index in the 9.02–11.71 range was correlated with a decreased likelihood of achieving functional independence (adjusted OR [aOR] = 0.450; 95% CI, 0.257–0.789; p = 0.005) when the TyG index was analyzed as a categorical variable (Q1: 7.67–9.01 and Q2: 9.02–11.71) and adjusted for age, sex, SBP, baseline NIHSS score, baseline ASPECTS score, coronary artery disease, diabetes, stroke etiology, occlusion site, and recanalization rate.

The secondary clinical outcomes comprised the distribution of mRS scores at 90 days (Figure 2 illustrates the distribution of the mRS score in patients grouped by median TyG index), NIHSS scores at 24 h postoperatively, and NIHSS scores at the time of discharge.

No significant differences were observed between groups before covariate adjustment (p > 0.05) (Table 2). When analyzed with an adjusted model, a TyG index of 9.02–11.71 significantly increased the odds of a one-point improvement in mRS scores within 90 days after MT (Q2: aOR = 1.65, 95% CI, 1.06–2.59; p = 0.027). Furthermore, patients with higher TyG (Q2: 9.02–11.71) levels exhibited a significantly increased risk for higher NIHSS scores both at 24 h (Q2 vs. Q1: aOR = 2.13, 95% CI, 0.55–3.72; p = 0.009) and at the time of discharge (Q2 vs. Q1: aOR = 2.04, 95% CI, 0.11–3.96, p = 0.038), as indicated by the adjusted model. These significant results suggest that higher postoperative NIHSS scores in patients with higher TyG levels may indicate a poor prognosis.

No statistically significant differences were observed between safety outcomes (including 90-day mortality, surgery-related complications, sICH within 24 h, and any intracranial hemorrhage within 48 h) and TyG levels in the unadjusted model (Table 2). Similarly, when analyzed with an adjusted model, no statistical significance was observed, indicating that the TyG index may not have a significant association with safety outcomes.

Subgroup analyses were conducted to elucidate the relationship between TyG and pivotal clinical endpoints across diverse demographic subgroups. A significant correlation was identified between TyG levels and functional independence at 90 days after thrombectomy, including the aged ≤62 years (OR = 0.540, 95% CI, 0.301–0.971), NIHSS score >13 (OR = 0.364, 95% CI, 0.180–0.735), hypertension (OR = 0.516, 95% CI, 0.286–0.931), internal carotid artery occlusion (OR = 0.489, 95% CI, 0.256–0.934), and large atherosclerosis (OR = 0.338, 95% CI, 0.174–0.656) groups, as detailed in Figure 1. These patients demonstrated stronger associations between elevated TyG levels and poorer functional outcomes. These findings may reflect increased metabolic and vascular vulnerability in these groups, warranting more focused monitoring and intervention.

This study has certain limitations. Its retrospective, single-center design inherently introduced selective bias and the relatively small sample size limited statistical efficiency. Additionally, while we attempted to control for confounding factors, residual confounding (e.g., dietary habits, statin use) cannot be ruled out. The lack of external validation limits the generalizability of our findings, and future multi-center studies are required to confirm these results.

Patients with elevated TyG may have a higher incidence of ischemic stroke, and elevated TyG may significantly increase stroke recurrence and mortality. However, whether the TyG index can predict clinical outcomes in patients after acute anterior circulation MT remains unknown.

Higher TyG levels may be associated with a worse 90-day functional independence prognosis, suggesting that TyG may serve as a negative predictor for evaluating the prognosis of patients. Higher postoperative NIHSS scores in patients with higher TyG may indicate a poor prognosis.

The findings of this study present novel avenues for future research endeavors. Specifically, doctor should prioritize patients exhibiting elevated TyG indices, employing more aggressive therapeutic interventions and implementing rigorous monitoring protocols. Additionally, further investigation is warranted into the interplay between the TyG index and other biomarkers, as well as their combined influence on stroke prognosis, with the ultimate goal of enhancing patient outcomes.