Abstract Body: Background & Aims: A certain lipoprotein, especially lower low-density lipoprotein cholesterol (LDL-C), has been said to be a risk of developing intracerebral hemorrhage (ICH). However, the correlation between lipoprotein levels and hematoma expansion (HE) following ICH remains unknown. We aimed to elucidate the relation between LDL-C and high-density lipoprotein cholesterol (HDL-C) levels and HE following acute ICH.
Methods: Consecutive patients with acute ICH admitted between July 2012 and November 2023 were retrospectively screened using Kagoshima Medical Center Stroke Database. The inclusion criteria were as follows: 1) onset to door time within 7 days; 2) availability of serum LDL-C and HDL-C levels evaluation on admission; and 3) availability of HE evaluation, defined as an increase of 12.5 mL and a 33% relative increase in hematoma volume on CT performed more than 24 hours after admission compared to baseline. We investigated the association between LDL-C and HDL-C levels and HE by performing Poisson regression analyses with a robust variance estimator adjusted by the following prespecified risk factors for HE: sex, age, use of antiplatelet drugs and anticoagulants before admission, body mass index, and systolic blood pressure on admission. Further, as a sensitivity analysis, we performed the same analysis above by limiting the patients with onset to door time within 24 hours.
Results: We screened 608 consecutive ICH patients. Of them, 538 were identified (279 [52%] males, median age 76 years). HE was observed in 51 patients (9%). In multivariable analysis, lower HDL-C was independently associated with HE (prevalence ration (PR) 0.819, 95% CI 0.677-0.990, p = 0.039, Figure 1A). Further, a negative linear trend was observed between HDL-C quartiles and HE in reference to HDL-C Quartile 4 (Q1: PR 2.910, Q2: PR 1.860, Q3: PR 1.251, Q4: PR 1.000, p = 0.008 for trend, Figure 1B). On the other hand, LDL-C was not associated with HE (PR 1.011, 95% CI 0.935-1.094, p = 0.777). In a sensitivity analysis, lower HDL-C was also independently associated with HE (PR 0.752, 95% CI 0.591-0.956, p = 0.020, Figure 2A) with a negative linear trend (Q1: PR 3.784, Q2: PR 1.658, Q3: PR 1.161, Q4: PR 1.000, p = 0.001 for trend, Figure 2B). LDL-C was not similarly associated with HE (PR 1.030, 95% CI 0.946-1.122, p = 0.493).
Conclusions: In acute ICH patients, lower HDL-C was associated with HE. HDL-C might have a protective role for HE following ICH.