A meta-analysis of resveratrol protects against cerebral ischemia/reperfusion injury: Evidence from rats studies and insight into molecular mechanisms

Objective: To evaluate the neuroprotective effect of resveratrol (RES) in rat models of cerebral ischemia/reperfusion (I/R) injury. Data sources: PubMed, Embase, MEDLINE, Cochrane Library, and Chinese databases were searched from their inception dates to July 2022. No language restriction was used in the literature search. Date Selection: Studies were selected that RES were used to treat cerebral I/R injury in vivo. Two reviewers conducted literature screening, data extraction and methodological quality assessment independently. Outcome measures: Cerebral infarct volume was included as primary outcome. The secondary outcomes included cerebral water content and neurological deficit scores. Malondialdehyde (MDA) and superoxide dismutase (SOD) were used to evaluate oxidative stress during medication. Results: A total of 41 studies were included, and only a few of them the methodological quality was relatively low. Compared with the control group, RES significantly reduced the cerebral infarct volume (29 studies, standard mean difference (SMD) = −2.88 [−3.23 to −2.53], p < 0.00001) and brain water content (nine studies, MD = −9.49 [−13.58 to −5.40], p < 0.00001) after cerebral I/R injury, then neurological function was improved (15 studies, SMD = −1.96 [−2.26 to −1.65], p < 0.00001). The MDA level (six studies, SMD = −8.97 [−13.60 to −4.34], p = 0.0001) was decreased notably after treatment of RES, while the SOD level (five studies, SMD = 3.13 [−0.16 to 6.43], p = 0.06) was increased unsatisfactory. Consistently, subgroup analysis of cerebral infarct volume suggested that the optimal therapeutic dose is 30 mg/kg (eight studies, SMD = −5.83 [−7.63 to −4.04], p < 0.00001). Meanwhile, 60 min of occlusion (three studies, SMD = −10.89 [−16.35 to −5.42], p < 0.0001) could get maximum benefit from compared with 90 min and 120 min of occlusion. On the other hand, the publication bias cannot be ignored. The pharmacological mechanisms of RES on cerebral I/R injury models as reported have be summarized, which can be used for reference by researchers to further plan their future experiments. Conclusion: RES might have a good neuroprotective effect on cerebral I/R injury in rats, then 30 mg/kg RES may be the optimal dose for treatment, and early administration of RES should be more neuroprotective. Also it need to be further verified through exploration of dose effect relationship, or delay administration or not.