Randomized Controlled Trial of Early Versus Delayed Statin Therapy in Patients With Acute Ischemic Stroke

ASSORT Trial (Administration of Statin on Acute Ischemic Stroke Patient)

Shinichi Yoshimura, MD, PhD; Kazutaka Uchida, MD; Takashi Daimon, PhD; Ryuzo Takashima, BA; Kazuhiro Kimura, PhD; Takeshi Morimoto, MD, PhD, MPH

Disclosures

Stroke. 2017;48(11):3057-3063. 

In This Article

Discussion

Early statin administration did not have favorable effects on mRS disability at 90 days compared with delayed administration in patients with acute ischemic stroke and dyslipidemia. Early versus delayed statin administration did not affect major adverse cardiovascular or cerebrovascular events or safety outcomes. This neutral effect was similar across subgroups including age, baseline LDL-C or high-sensitivity C-reactive protein values, imaging classification, neurological symptoms at onset, intravenous tPA or statin use, stroke history, and stroke classification.

Immediate statin administration after ischemic stroke was reported to decrease infarct size and improve physical function in animal models.[17–20] The neuroprotective effects of statin on acute ischemic stroke has been reported as antioxidation, anti-inflammation, vasodilation, antithrombosis, angiogenesis, synaptogenesis, and neural progenitor cell migration to the injury site.[20–24] The clinical effects of statin on recurrence of ischemic stroke or survival in patients with acute ischemic stroke were reported. The SPARCL trial (The Stroke Prevention by Aggressive Reduction in Cholesterol Levels) showed that statin administration at 2 to 3 months after onset was associated with more event-free patients with ischemic stroke or transient ischemic attack.[25] A retrospective observational study reported that 1-year survival rates among patients with ischemic stroke who were administered statins before or during admission were higher than the counterparts.[6]

About disability outcomes, the subanalysis of SPARCL Trial also showed that statin administration at nonacute phase was associated with reduced disability measured by mRS.[26] Another non-RCT found that statin withdrawal for 3 days after ischemic stroke onset in patients previously on statins was associated with significantly worse disability status after 90 days compared with patients receiving statins immediately after onset.[4] Thus, statin withdrawal at acute phase or nonadministration at nonacute phase should be associated with worse outcomes in patients with acute ischemic stroke, but the clinical significance of early administration to naive patients should be evaluated.

The median NIHSS on admission was 3, and the almost half of prevalent cause of ischemic stroke in our study was lacunar, whereas in the previous study the median NIHSS was 12, and atherothrombosis was the major cause and lacunar was only 18%.[4] Although the enrolled patients were similar to the epidemiology of Japanese patients,[27] such differences in stroke severity or cause might be associated with the attenuated results. Statin administration in patients with transient ischemic attack or mild stroke had failed to show beneficial effects on stroke recurrence within 90 days.[28] Thus, neuroprotective effect of statin may be more apparent in patients with moderate severity and warrants further study. Thus, the clinical significance of statin administration for acute phase ischemic stroke is still uncertain, especially in patients with more severe disability or other causes, and use of intravenous tPA or endovascular thrombectomy. These area of uncertain should be confirmed with well-conducted randomized trials.

This study had some limitations. First, the enrolled patients had relatively less severe stroke. In addition, we used mRS as primary outcome measure because mRS was most frequently used in stroke trials. Because mRS was not sensitive to evaluate the mild disability, statin administration in patients with acute ischemic stroke might be beneficial if patients with severe disability were enrolled or more sensitive outcome measurements such as Fugl-Meyer or grip strength were used. Second, this was an open-label randomized trial. Although individuals performing primary outcome assessments were blinded to treatment allocation, concomitant therapies such as rehabilitation might differ between hospitals. However, the effect of such differences on results should be small because hospital was included in the stratification variable at randomization and all patients received the statin at discharge. Third, we enrolled patients with preexisting dyslipidemia, and the doses of statins were relatively lower than the regular doses in the Western countries. However, the THRaST study (The Thrombolysis and Statins) reported that atorvastatin 10 to 20 mg/d had similar effect to 40 to 80 mg/d on neurological outcomes, and the type of statin also did not matter.[7] Indeed, Asian cohorts was reported to be more sensitive to statin than the Western cohorts so that lower doses provided similar LDL-lowering effect compared with the higher dose in Western cohorts.[29,30] Therefore, the doses and types of statin were rational in our study, but the higher dose should be attested in the future studies. Final, the sample size was based on the common OR of 2 calculated from the distribution of mRS at 90 days in the previous observational study. However, the observed OR of our study was 1.1 for 1 decrement of mRS. The resultant effect size was smaller than expected, and effect size in observational study was generally larger because of prescribing bias and others.[9] The animal studies also showed positive effect of statins for recovery from stroke. The injection of statin in these animal studies might enhance the effect on the recovery from stroke in addition to the maximum doses used.[20,21,24] In addition, accumulation of milder disability (mRS score of 0–2) decreases the power of detection of efficacy in the neuroprotective trial.[31] However, OR of 1.59 for excellent outcome in our study was clinically important, and further study with appropriate sample size should be conducted to attest the role of early statin administration for acute ischemic stroke.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....