Association of LDL-cholesterol <1.8 mmol/L and statin use with the recurrence of intracerebral hemorrhage

Background: Recent intensive LDL-C lowering trials, including FOURIER, ODYSSEY OUTCOMES, and TST trials, have mostly refuted the concern surrounding statin use, low-density lipoprotein cholesterol (LDL-C) lowering, and intracerebral hemorrhage (ICH) risk. However, the results from these trials may not be fully applied to ICH survivors, as the populations studied were mainly patients without prior ICH, in whom the inherent ICH risk is more than ten times lower than that of ICH survivors. Although available literature on statin use after ICH has demonstrated no excess risk of recurrent ICH, other potential factors that may modify ICH risk, especially hypertension control and ICH etiology, have not generally been considered. Notably, data on LDL-C levels following ICH are lacking. Aims: We aim to investigate the association between LDL-C levels and statin use with ICH risk amongst ICH survivors, and to determine whether the risk differed with patients’ characteristics, especially ICH etiology. Methods: Follow-up data of consecutive spontaneous ICH survivors enrolled in the University of Hong Kong prospective stroke registry from 2011 to 2019 were retrospectively analyzed. ICH etiology was classified as cerebral amyloid angiopathy (CAA) using the modified Boston Criteria or hypertensive arteriopathy, while the mean follow-up LDL-C value was categorized as <1.8 or ≥1.8mmol/L. The primary endpoint was recurrent ICH. The association of LDL-C level and statin use with recurrent ICH was determined using multivariable Cox regression. Pre-specified subgroup analyses were performed, including based on ICH etiology and statin prescription. Follow-up blood pressure was included in all the regression models. Results: In 502 ICH survivors (mean age 64.2±13.5 years, mean follow-up LDL-C 2.2±0.6mmol/L, 28% with LDL-C <1.8mmol/L), 44 had ICH recurrence during a mean follow-up of 5.9±2.8 years. Statin use after ICH was not associated with recurrent ICH (adjusted hazard ratio [AHR] 1.07, 95% confidence interval [CI] 0.57-2.00). The risk of ICH recurrence was increased for follow-up LDL-C <1.8mmol/L (AHR 1.99, 95% CI 1.06-3.73). This association was predominantly observed in ICH attributable to CAA (AHR 2.52, 95% CI 1.06-5.99) and non-statin users (AHR 2.91, 95% CI 1.08-7.86). Conclusion: The association between post-ICH LDL-C <1.8mmol/L and recurrent ICH was predominantly observed in CAA patients and those with intrinsically low LDL-C (non-statin users). While statins can be safely prescribed in ICH survivors, LDL-C targets should be individualized and caution must be exercised in CAA patients.