Hypertriglyceridemia predicts good neurological outcome after mechanical thrombectomy in patients with acute ischemic stroke due to large-vessel disease.

INTRODUCTION: Previous studies evaluating the prognostic role of hypertriglyceridemia in acute ischemic stroke (AIS) patients have yielded divergent results. AIM: To determine whether hypertriglyceridemia affects long-term outcome in AIS patients treated with mechanical thrombectomy (MT). MATERIAL AND METHODS: Patients aged ≥ 18 years, with first-ever AIS, treated with MT between 2020 and 2022 in the University Hospital in Krakow, Poland, were included. Serum triglyceride (TG) levels were assessed fasting the next morning after MT. We used Firth's penalized likelihood logistic regression to address the sample size and mediation analysis (Baron-Kenny approach) to assess whether TG's effect was via post-procedural infections (antibiotic use). Interaction analysis examined context-dependent effects, and ROC curves compared the discriminatory ability of triglycerides, glucose and the triglyceride-glucose index (TyG). Our primary outcome measure was the modified Rankin Scale (mRS) 0-2 (good) at 90 days. RESULTS: Among 410 patients (47.8% women, median age 70 [IQR: 62-79] years), hypertriglyceridemia (TG ≥ 1.7 mmol/l) was observed in 22.4%. In the large-vessel disease subgroup (LVD; n = 90), hypertriglyceridemia independently predicted good outcome (multivariable: OR = 5.95, 95% CI: 1.03-34.43, p < 0.05). Importantly, although elevated TG was associated with a 63% reduction in antibiotic use (15.8% vs. 40.9%), formal mediation analysis revealed that this infection-prevention pathway was not statistically significant (p = 0.29) and explained only around 13% of TG's protective effect. Instead interaction analysis revealed context dependency: TG showed a protective association in non-infected patients (p = 0.02), whereas no protective effect was observed in infected individuals. Additionally, the protective effect of TG was masked when combined with glucose in the TyG index (TG: AUC = 0.655 vs. TyG index: AUC = 0.502). No significant associations between TG levels and in-hospital or 30-day mortality were observed. CONCLUSIONS: Our study suggests that in AIS patients with LVD treated with MT, elevated TG levels may predict good outcome through LVD-specific resilience mechanisms, independently of infection prevention.