Association of admission LDL cholesterol with clinical profile and 3-months outcomes after ischaemic stroke

Abstract

Abstract Background and aim We investigated whether admission LDL-C levels are associated with clinical profile and outcomes after acute ischaemic stroke by comparing patients stratified into five LDL-C groups. Patients and methods We retrospectively analyzed consecutively included patients ≥18 years admitted to the Bernese Stroke Centre (02/2015–03/2023). Patients were stratified into five LDL-C groups: >3.0 mmol/L (n=493), 2.6–3.0 mmol/L (n=225), 1.8–2.6 mmol/L (n=490), 1.4–1.8 mmol/L (n=254) and <1.4 mmol/L (n=172). LDL-C was measured within 24 h of admission. Outcomes included successful reperfusion, symptomatic intracranial haemorrhage, favourable and excellent functional outcomes at 3 months, and 3-month mortality. Separate complete-case multivariable logistic regression models evaluated LDL-C as a continuous and five-level categorical exposure, adjusting for prespecified clinical covariates. Results Among 1634 patients, those with LDL-C <1.4 mmol/L were older (median 78 vs. 73 years; p<0.001), more often male (59% vs. 44%; p<0.001), and had a greater cardiovascular and renal comorbidity burden than those with LDL-C >3.0 mmol/L. Favourable and excellent 3-month functional outcomes differed across LDL-C categories (favourable: 28.7–41.9%; p = 0.023; excellent: 23.1–35.9%; p = 0.027), as did survival (61.0–79.5%; p<0.001), lowest in the <1.4 mmol/L group. 96 symptomatic intracranial haemorrhages occured (3.9–7.6%; 96 events; p = 0.228). After adjustment, continuous LDL-C was not associated with successful reperfusion (aOR per 1-mmol/L increase, 0.98; 95% CI 0.83–1.16), favourable 3-month outcome (aOR 1.09; 95% CI 0.96–1.23), or excellent 3-month outcome (aOR 1.11; 95% CI 0.97–1.26); the corresponding categorical models were also non-significant (global p = 0.411, p = 0.193, and p = 0.269, respectively). In contrast, higher continuous LDL-C was associated with lower odds of 3-month mortality (aOR 0.85; 95% CI 0.74–0.97; p = 0.019). The categorical model was also significant (global p=0.007); compared with LDL-C >3.0 mmol/L, mortality odds were higher with LDL-C 1.8–2.6 mmol/L (aOR 1.54; 95% CI 1.09–2.18) and <1.4 mmol/L (aOR 2.14; 95% CI 1.34–3.43). Conclusions Low admission LDL-C identified a clinically vulnerable phenotype and was associated with higher 3-month mortality after adjustment, but not with successful reperfusion or functional outcomes. These observational findings do not establish a direct adverse effect of low LDL-C. No significant association with sICH was detected, although clinically relevant differences cannot be excluded.