Atorvastatin reduces cerebral vasospasm and infarction after aneurysmal subarachnoid hemorrhage in elderly Chinese adults

Demographic and clinical data

The demographics and baseline characteristics of patients were similar between the atorvastatin and placebo groups (Table 1). Most of the patients were women with Hunt-Hess grades I to III subarachnoid hemorrhages. The mean treatment duration was 14 days (SD = 2) for the placebo-treated group and 13 days (SD = 2) for the atorvastatin-treated group. Oral and injected nimodipine were administered to all participants receiving atorvastatin and placebo, with no significant difference observed between the two.

The primary endpoint-clinical outcomes

At 6 months, there were no significant differences in GOS classification between the two groups (Table 2 and Figure 3). A higher percentage of patients in the atorvastatin (54.4%, 81/150) group had good recovery compared with patients in the placebo (48.7%, 73/450) group, but statistical significance was not reached (P=0.36, 95% CI 0.72–1.12). Ten (6.6%) patients in the placebo group and 8 (5.3%) in the atorvastatin group died during the study.

Subgroup analyses for a good outcome (GOS 5) showed no effect of gender, age (>75 or <75), aneurysm size, clot size, good or poor clinical condition at admission (Hunt-Hess grades), and different surgical procedure (Table 3 and Figure 4A). Also, heterogeneity test showed that no heterogeneity in age, Hunt-Hess, and surgical procedure subgroups (I²=0); gender (I²=9%) and aneurysm size (I²=28%) subgroups had no significant heterogeneity. Clot size had significant heterogeneity (I²=71%, Figure 4B).

The secondary endpoint-postoperative complications

According to the 30-day all-cause mortality, 7 (4.7%) of 150 patients in the atorvastatin group and 8 (5.3 %) of150 in the placebo group died within 30 days (RR 0.87, P=0.149, 95% CI 0.33–2.35, Figure 5).

Event rates for each of the individual components of the primary composite endpoint are shown in Figure 6. The occurrence of postoperative CVS was significantly lower in the atorvastatin group than in the placebo group (Table 4, Figure 6, 39.3% vs 56%, P =0.004, Relative Risk 1.397, 95% CI 1.11 to 1.76). In this study, 27.3% (47/150) patients in the placebo group and 18.7% (28/150) in the atorvastatin group had delayed vasospasm-related new cerebral infarction (Table 4, Figure 6, P=0.027, Relative Risk 1.327, 95% CI 1.05 to 1.67). Even though group differences in the incidence of DIND were non-significant, the atorvastatin group had 6% lower risk of DIND incidence than the placebo group (Table 4, Figure 6, P=0.207, Relative Risk 1.18, 95% CI 0.92 to 1.52).

Study design

We performed a randomized, parallel-arm placebo-controlled trial in 3 neurosurgical institutions in China (Wuxi Clinical College of Anhui Medical University-904th Hospital of Joint Logistic Support Force of PLA, Renmin Hospital of Wuhan University and Taizhou fourth people's hospital), between October 2014 and Oct 2017. The study was designed to assess the superiority of the intervention. The study protocol was approved by the Anhui Medical University affiliated Wuxi Clinical College Clinical Research Ethics Committees (2014-YXLL-001). The study protocol received Ethics Committee approval from all of the participating centers. We obtained written informed consent from patients whose competence was established by their accurate orientation for time, place, and person, and understanding of the recruiter’s description of the trial, or otherwise from their next of kin or their legal representative. The trial recruitment was decelerated after warnings from the Chinese Food and Drug Administration and the Department of Health of Wuxi in Oct 2014 that the highest approved dose of atorvastatin, ie, 20 mg, was associated with an elevated risk of muscle injury, hepatorenal and kidney dysfunction, or myopathy. The everyday clinical data also was monitored by Hospital Rational and Health drug use review committee. Safety data were reviewed, and it was decided that the study should continue.

Patients were randomly assigned (1:1) to receive 20 mg/day atorvastatin or placebo within 72 h of aSAH and after neurosurgical clipping or coiling (Figure 1). Atorvastatin or placebo were administered orally for up to 14 days after the aSAH. Final follow-up was 6 months after aSAH.

Study patients

The inclusion criteria were as follows: radiological clear diagnosis of spontaneous subarachnoid hemorrhage; Hunt -Hess scale was 1 to 4; either gender; age 60-90 years; could be randomized within 72 h after the onset of SAH; an intracranial aneurysm that was considered to be the cause of the SAH and first onset. The exclusion criteria were as follows: patients unlikely to survive on admission; traumatic SAH; high cholesterol combined with diabetes; long-term use of atorvastatin as treatment group; abnormal liver enzymes, rhabdomyolysis or myopathy; history of mental illness or epilepsy; severe lung disease or multiple organ dysfunction; researchers believed patients were not suitable clinical subjects or participation in other clinical trials

Randomization and masking

A permuted-block randomization was performed using a computer.system with an allocation list generated random numbers (in a 1:1 ratio) using SPSS 14.0 software (SPSS Institute, Hefei, Anhui Medical University) by a statistician not associated with the project team to protect the blinding and integrity of the study, and the results of randomization were sealed in sequentially numbered envelopes and stored at the site of investigation until the end of the study. During the study period, all inclusion patients were randomly assigned to receive either atorvastatin (20mg/day, 14 day) or placebo. A study nurse administered the study drugs according to the randomization sequence. Both the study members and the patients were blinded to the study drug allocation. The randomization envelopes were opaque. If an emergency such as severe hepatic failure occurred, then two experts could request unmasking of the treatment allocation, or adjust or interrupt the study drug if necessary; all situations were documented.

Patient demographics, medical history and relevant investigation results were collected. The severity of the SAH was scored clinically using the Hunt-Hess grading scale and radiologically using the Fisher scale.

Standard of care

In the 3 participating centers, ruptured aneurysms were usually treated (clipping or coiling) within 24 hours after admission. Nimodipine was routinely started on admission and continued for 14 days in both groups. When clinical vasospasm developed, hypertensive treatment with an elevated mean arterial blood pressure of ≥20 mm Hg was started to maintain a relative higher perfusion. We adopted daily Transcranial Doppler (TCD) ultrasonography performed using a 2-MHz probe mounted on a specially designed head frame as a surrogate method to measure CVS. Computed tomographic perfusion (CTP) was performed in two of the participating centers for research purpose. Cerebral digital subtraction angiography (DSA) was not usually performed as it was available in only one hospital.

Procedures

All patients underwent baseline cranial CT and CT angiography(CTA) or DSA. TCD ultrasonography was used to evaluate CVS daily. Cranial CT re-examination was obtained every 24–48 h after the aneurysm was clipped or coiled, for re-examination at discharge, and 6 months after aSAH. Neurological assessment was completed by two nurses every 2 h from initiation of study drug until day 14 using the Glasgow Coma Scale. If the patient had worsening of neurological symptoms, as defined previously [15], then CT, CTA, or DSA was required. At 6 months after randomization, patients were telephone interviewed to obtain GOS scores by a nurse or clinician with no knowledge of the treatment allocation.

All clinical and imaging data were assessed by a masked, independent diagnostic and assessment committee; this committee included two radiologists and two sonographers. Angiographic vasospasm and vasospasm-related new cerebral infarction were confirmed by radiologists. TCD vasospasm was confirmed by sonographers. A clinical assessment committee include one neurosurgeon, one neurointensivist, and one neurocritical care physician; all assessed the primary endpoint together.

The primary endpoint assessed GOS at 6 months after aSAH, dichotomized as good (≥4) or poor (<4) outcome. The secondary efficacy endpoint assessed all-cause mortality at 30 days after aSAH, CVS, vasospasm-related new infarction and DIND within 2 weeks post-aSAH. CVS, vasospasm-related new infarction, and DIND were defined previously [16] as follows: Angiographic vasospasm was defined as focal or generalized reduction of cerebral arterial caliber on conventional cerebral angiogram. TCD vasospasm was defined as any peak systolic middle cerebral artery velocity (PSV_MCA) >200 cm/s and a Lindegaard ratio of >3. Vasospasm-related new infarction was defined where vasospasm was the relevant contributing factor or primary cause, and the development of a new lesion consistent with infarction on CT or MRI. DIND was defined as any 2 or more point fall in Glasgow Coma Scale or unaccountable new focal neurological deficit lasting ≥2 hours.

Statistical analysis

On the basis of previous data we estimated that 212 patients would be required to confirm any effect with an α of 5% and 80% power [17]. For sample size estimation, we assumed that the atorvastain group had a 48% rate of good GOS compared with the placebo group with 46% based on our preliminary trial; 290 patients would be required (80% power and 2-sided,α=0.05, a 10% loss to follow-up). We decided to enroll 300 patients. A research nurse entered all baseline and outcome data in the study database; data were collected on handwritten forms and archived in a password-protected electronic database. Treatment effect was tested by logistic regression adjusted for Hunt-Hess grade (I, II, ≥III) with the Wald χ² test used to determine treatment effect. We described the incidence and relative risk of dichotomous variables for the atorvastatin-treated group relative to the placebo group, with corresponding 95% CIs. The first exploratory end point was the GOS. For the GOS endpoint, if no GOS score was available, a score of 4 (lower moderate disability) was assigned when there was no clinical evidence of prior neurological impairment, and a score of 3 (lower severe disability) was assigned in any other situation when a patient was alive at 6 months. Demographics and safety data are reported as descriptive statistics (means, standard deviation) [5]. Categorical variables were analyzed with the χ² test, continuity correction χ² test or likelihood ratio χ² test. Numeric variables were analyzed by use of an unpaired t test or Mann-Whitney u test. The difference (and 95% CI for the difference) between two medians was calculated with the Hodges-Lehmann estimator.

Statistical analyses were done on SPSS 14.0 software with two-tailed tests wherever appropriate and P values less than 0.05 were considered to be of statistical significance. The Clinical Research Ethics Committee from Anhui Medical University, Wuxi Clinical College (904^th Hospital of PLA) was involved in overseeing the data. The study is registered with http://www.chictr.org.cn, number ChiCTR-IPR-14005395.

Ethics approval and consent to participate

The study protocol was approved by the Anhui Medical University affiliated Wuxi Clinical College Clinical Research Ethics Committee. We obtained written informed consent from the family members of patients whose competence was established by accurate orientation to time, place, and person and understanding of the recruiter’s description of the trial. Otherwise, consent was obtained from the patient’s next of kin or legal representative.