Results of the search
The electronic search identified 1346 records from databases. Then, after removing duplications and ineligible records, it yielded 655 unique records for the screening process. The manual search did not identify any additional records. Nine RCTs met the inclusion criteria of this systematic review and meta– analysis, including 1464 patients (721 patients in the statin group and 743 patients in the control group) [20– 28] (Figure 1). A summary of the included studies is shown in Table 1.
The mean age of patients is 56.1 years, and the female gender represents about 60.75% of total patients. 50% of aneurysms were anterior circulation aneurysms (ACA) arise from the internal carotid artery (ICA) and its branches, while 8% were posterior circulation aneurysms and 42% was others (including anterior and posterior communicating and anterior choroidal arteries). Hunt & Hess (H & H) grading scale was used in three studies, including 447 patients (223 patients in the statin group and 224 patients in the control group) [22, 27, 28]. 81% of patients in these studies within grade 1 to 3. World Federation of Neurosurgical Societies (WFNS) scale was used in five studies, including 978 patients (479 patients in the statin group and 499 patients in the control group)[21, 23– 26]. 76.67% of patients within grade 1 to 3. Five studies reported Fisher classification of SAH on CT, including 1068 patients (523 patients in the statin group and 545 in the control group)[20, 21, 24, 25, 27]. 84.3% of SAH was within grades 3 to 4. One study reported modified Fisher classification, including 25 patients (13 patients in the statin group and 12 patients in the control group) [26].80% of SAH was within grade 3 to grade 4. Three studies reported using clipping in 48.6% of patients (555 out of 1142) [22, 25, 28]. Also, three studies reported coiling in 45.7% of patients (554 out of 1211) [25, 27, 28].
Five studies investigated simvastatin 80 mg/day as a high dose of statin [20, 22– 24, 26], while four studies used a moderate dose of statins simvastatin 40 mg/day [25], pravastatin 40 mg/day equivalent to simvastatin 20 mg/day[21], pitavastatin 4 mg/day equivalent to Simvastatin 40 mg/day [27], and atorvastatin 20 mg/day equivalent to simvastatin 40 mg/day [28].The statin agent was administered early within 48 hours of ictus [20, 27], within 72 hours [21, 23, 26, 28] and within 96 hours [22, 24, 25] (Table 1).
The risk of bias in included studies
The quality of included studies ranges from moderate to high quality according to the Cochrane risk of bias assessment tool. The summary of quality assessment domains of included studies is shown in Figure 2.
Effects of interventions
Vasospasm
Seven studies considered vasospasm as the primary outcome and reported its incidence in both groups, including 311 patients in the statin group and 311 patients in the control group [21– 24, 26– 28]. Vasospasm was seen in 45.67% (142 out of 311) of patients in the statin group as compared to 59.8% (186 out of 311) in the control group. The incidence of vasospasm was decreased by 45% with the use of statins and the MH odds ratio for this pooled result was 0.55 (95% CI 0.39– 0.81, p value = 0.0003, I2 = 45%, fixed effect model).
Three studies reported different degrees of the severity of the cerebral vasospasm, according to baseline DSA [27], and mean flow velocity of the MCA on TCD [21, 23]. Severe vasospasm was reported in three studies, including 110 patients in the statin group and 110 patients in the control group [21, 23, 27]. Severe vasospasm was seen in 16.36% (18 out of 110) of patients in the statin group compared to 30% (33 out of 110) in the control group. The incidence of severe vasospasm was decreased by 54% with the use of statins and the MH odds ratio for these pooled results was 0.46 (95% CI 0.24– 0.88, p value 0.02, I2 = 0%, fixed effect model). Moderate vasospasm was reported in two studies only including 70 patients in the statin group and 70 patients in the control group [23, 27]. Moderate vasospasm was seen in 31.43% (22 out of 70) patents in the statin group compared to 24.28% (17 out of 70) of the patients in the control group. The use of statins increased moderate vasospasm by 44%, and the MH odds ratio of this pooled result was 1.44 (95% CI 0.68– 3.05, p value = 0.34, I2 = 24%, fixed effect model).
Delayed Neurological Ischemic Deficits (DIND)
Nine studies reported DIND following aSAH [20– 28]. The rate of DIND was 17.48% (126 out of 721) of patients in the statin group as compared to 22.9% (170 out of 742) in the control group. The incidence of DIND was decreased by 29% with the use of statins and the MH odds ratio for this pooled result was 0.71 (95% CI 0.54– 0.92, p value = 0.009, I2 = 47%, fixed effect model)
Delayed Cerebral Ischemia (DCI)
Four studies reported DCI following aSAH[22, 25, 27, 28].The rate of DCI was 15.47% (95 out of 614) of patients in the statin group as compared to 20.6% (131 out of 636) in the control group. The incidence of DCI was decreased by 30% with the use of statins, and the MH odds ratio for this pooled result was 0.70 (95% CI 0.53– 0.94, p value = 0.02, I2 = 27%, fixed effect model).
Functional Outcomes
Seven studies reported functional outcome, the incidence of poor outcome in both the group including 683 patients in the statins group and 704 patients in the control group [21– 23, 25– 28].Poor outcome was seen in 49.48% (338 out of 683) of patients in the statin group as compared to 48.3% (340 out of 704) of patients in the control group. The MH odds ratio for this pooled result was 0.97 (95% CI 0.77– 1.22, p value = 0.79, I2 = 0%, fixed effect model).
Mortality
A mortality rate was reported in seven studies, including 648 patients in the statin group and 669 patients in the control group [21– 26, 28]. The mortality rate was 7.6% (49 out of 648) in the statin group compared to 8.82% (59 out of 669) patients in the control group. The pooled MH odds ratio for this result was 0.85 (95% CI 0.57– 1.26, p value = 0.28, I2= 20%, fixed effect model).
Deterioration causes
Three studies reported hydrocephalus following aSAH requiring ventriculostomy using external ventricular drainage (EVD), including 447 patients in the statin group and 468 patients in the control group [21, 23, 25].The hydrocephalus rate was 20.6% (92 out of 447) of patients in statin group compared to 20.3% (95 out of 468) of patients in the control group. The pooled MH odds ratio was 1.01 (95% CI 0.73– 1.40, p value = 0.63, I2= 0%, fixed d effect model).
Two studies reported rebleeding following initial haemorrhage, including 407 patients in the statin group and 428 patients in the control group [23, 25]. The rebleeding rate was 2.46% (10 out of 407) of patients in the statin group, compared to 2.57% (11 out of 428) of patients in the control group. The pooled MH odds ratio was 0.95 (95% CI 0.4– 2.27, p value = 0.49, fixed effect model).
Three studies reported sepsis following aSAH, including 450 patients in the statin group and 472 patients in the control group [21, 22, 25]. The rate of sepsis was 23.78% (107 out of 450) of patients in the statin group compared to 22.25% (105 out of 472) of patients in the control group. The pooled MH odds ratio was 1.05 (95% CI 0.84– 1.30, p value = 0.8, I2= 0%, fixed effect model).
Adverse effects of statins
Five studies reported increased liver enzymes, including 502 patients in the statin group and 525 patients in the control group [20, 22, 24, 25, 27]. The rate of increased liver enzymes was 4.38% (22 out of 502) of patients in the statin group compared to 6.28% (33 out of 525) of patients in the control group. The pooled MH odds ratio was 0.69 (95% CI 0.4– 1.19, p value = 0.47, I2 = 0%, fixed effect model).
Three studies reported increased creatine kinase (CKP), including 57 patients in the statin group and 59 patients in the control group [20, 22, 24]. The rate of increased CKP was 3.51% (2 out of 57) of patients in the statin group compared to 1.69% (1 out of 59) of patients in the control group. The pooled MH odds ratio was 1.80 (95% CI 0.23– 14.28, p value = 0.6, I2= 0%, fixed effect model).
Two studies reported rhabdomyolysis, including 445 patients in the statin group and 466 patients in the control group [25, 27].The rate of rhabdomyolysis was 0.45% (2 out of 445) of patients in the statin group compared to 0% (0 out of 466) of patients in the control group. The pooled MH odds ratio was 5.19 (95% CI 0.24– 110.69, p value = 0.29, heterogeneity was not applicable, fixed effect model).
Chou et.al. reported increased troponin level in 19 patients with statin group and 20 patients in the control group [22].The rate of increased troponin level was 26.32% (5 out of 19) of patients in the statin group compared to 35% (7 out of 20) of patients in the control group. The pooled MH odds ratio was 0.66 (95% CI 0.17– 2.62, p value = 0.56, fixed effect model).
Kirkpatrick et.al. reported gastrointestinal tract (GIT) hemorrhage and myositis, including 54 patients in the statin group and 54 patients in the control group [25].The rate of GIT hemorrhage and myositis in the statin group were 0% (0 out of 54) and 3.7% (2 out of 54) of patients compared to 3.7% (2 out of 54) and 0% (0 out of 54) of patients in the control group. The pooled MH odds ratio were 0.19 (95% CI 0.01– 4.11, p value = 0.29, fixed effect model) and 5.19 (95% CI 0.24– 110.69, p value = 0.29, fixed effect model), respectively.
A summary of the subgroup analyses is shown in Figure 3.