Reviewed: Ji H, Zhou X, Wei W, Wu W, Yao S. Ginkgol [sic] biloba extract as an adjunctive treatment for ischemic stroke: A systematic review and meta-analysis of randomized clinical trials. Medicine (Baltimore). January 2020;99(2):e18568. doi: 10.1097/MD.0000000000018568.

According to the World Health Organization, roughly 15 million people per year have a stroke. In 2017, strokes were the leading cause of death in China. More than two-thirds of strokes in Asians are ischemic, which result from a blockage in the blood supply to the brain. Current treatments for ischemic stroke are not readily available and are only moderately effective. In China, ginkgo (Ginkgo biloba, Ginkgoaceae) extract (GBE) often is used as an adjunctive treatment for ischemic stroke. These authors conducted a systematic review and meta-analysis of randomized, controlled trials and critical appraisals on the effectiveness and safety of GBE when used for the different phases of ischemic stroke. Recovery from ischemic stroke comprises an acute phase (less than two weeks after symptom onset), a stationary phase (two weeks to six months after symptom onset), and a sequelae, or convalescence, phase (six months or longer after symptom onset).

The authors searched four English and three Chinese databases from the date of inception until September 2018. They included trials of any language or publication type that evaluated the efficacy and safety of GBE in any age, race, gender, or disease phase. Oral GBE treatments in the studies had to be taken for two weeks or longer. Studies that used control groups or conventional medicine groups were included; those that used traditional Chinese medicine or Chinese patent medicines as controls were excluded. The authors looked for study outcomes that included all-cause mortality, adverse events, changes in neurological function, recurrence rates, dependence (ability to carry out basic daily activities), vascular events, and quality of life.

The authors identified 15 trials published between 2010 and 2018 that included patients (N = 1,829) who were randomly assigned to GBE treatments or control/conventional treatments. Sample sizes ranged from 62 to 348 patients per trial. Ten trials enrolled 946 patients in the acute phase of stroke, four trials enrolled 535 patients in the convalescence phase, and one trial enrolled 348 patients in various phases. Most studies compared GBE plus conventional therapy with conventional therapy only, or GBE plus conventional therapy with placebo plus conventional therapy. Among the conventional treatments used were antiplatelet or anticoagulant therapies, neuroprotective agents, nutritional support, and control of serum glucose or blood pressure.

Only one trial was double-blinded, and one trial was single-blinded. In the other 13 trials, GBE was used only in the experimental groups, making it impossible to blind the participants and personnel. Quality of the trials was rated as moderate for evidence of vascular events and recurrence rates and low for evidence of mortality.

Acute Phase of Ischemic Stroke*

None of the trials with patients in the acute phase of ischemic stroke reported all-cause mortality or serious adverse events. In two trials (n = 425) that reported dependence, significant improvements were observed with the use of GBE plus conventional therapy compared with placebo plus conventional therapy (P < 0.001).

Seven trials measured neurological function or clinical effect outcomes. The results showed that GBE plus conventional treatment was more effective than conventional treatment alone in improving neurological function after 14 days of treatment (P < 0.001). In three trials, for which the outcomes were measured by using the National Institutes of Health Stroke Scale (NIHSS), adding GBE to conventional treatment significantly improved NIHSS scores compared with conventional treatment alone or conventional treatment plus placebo (P < 0.001).

None of the acute-phase trials reported recurrence rates. One trial with 106 participants reported dizziness and nausea in one patient each in the GBE group and dizziness in one control patient. In another trial of 88 participants, two participants in the GBE group reported facial flushing. All adverse events disappeared after symptomatic treatment.

Convalescence Phase of Ischemic Stroke

Heterogeneity was minimal in the two trials that reported vascular events in patients during the convalescence phase of ischemic stroke. A meta-analysis of the trials revealed no significant differences in vascular events in the GBE-plus-conventional-treatment group compared with the conventional-treatment-only group.

Two trials assessed dependence in 407 participants. The improvements with GBE plus conventional therapy were greater than with conventional therapy alone or conventional therapy plus placebo (P < 0.001). Two convalescence-phase trials reported recurrence rates. In one of those trials (n = 57), which compared GBE alone with placebo alone, only one patient in the placebo group experienced a recurrent stroke during a four-month follow-up. In the other trial, which was a two-year multicenter trial (n = 348), which compared GBE plus conventional therapy (treatment group) with conventional therapy alone (control group), nine patients in the treatment group and 14 patients in the control group reported recurrent stroke (a nonsignificant between-group difference).

In one trial that assessed mortality, four patients in the GBE group and one patient in the control group died during treatment or during a two-year follow-up period. In another trial, seven patients in the GBE group and nine patients in the control group died during the study; between-group differences were not significant. In four trials with 613 patients, greater improvements in neurological function were observed in patients who used GBE as an add-on to conventional therapy compared with conventional therapy alone (P < 0.001).

One six-month trial with 346 participants reported these “nonserious” adverse events: vomiting, change in blood sugar levels, myocardial infarction, nephritis, sick sinus syndrome, and pneumonia. All events, except for vomiting, were considered to be unrelated to the study treatment.

This review is limited by the small number of studies and their low quality. The authors concluded that GBE with conventional therapy reportedly improves neurological function and dependence compared with conventional therapy alone for patients who have had an ischemic stroke. No beneficial effects of GBE on recurrence rates were found.

* One peer reviewer of this summary noted a discrepancy in the journal article. He wrote: “In the text, the authors say that adding GBE to conventional treatment in the acute stage improved Barthel Index and Neurological Function Deficit Score. However, the Forest plots in Figures 3 and 4 appear to indicate the opposite, the analysis seems to favor control treatment.”