Ginkgo biloba and stroke risk

Thursday February 28 2008

“Could ginkgo cause a stroke?” asks The Daily Mail today. The newspaper says that the herb, taken by thousands of Britons in the hope that it will keep their memory sharp into old age, may do more harm than good.

The herb is extracted from the leaves of the ginkgo biloba tree and was first used medicinally in China more than 5,000 years ago. There is divided opinion on the effectiveness of the herb, and systematic reviews, studies that produce the most reliable results, have found either no benefit or only a small benefit from using it. Among its purported medicinal properties, it is thought to stave off Alzheimer's disease and improve circulation. Reports of adverse effects of the herb have included an increase in bleeding-related complications.

This study was on people over 84 years of age and was set up to establish whether ginkgo extract could delay cognitive impairment in the elderly, not stroke risk. During the study, seven people taking ginkgo had strokes or warning strokes, compared with none in the placebo group. The study was too small reliably to demonstrate any effect that ginkgo may have on dementia. The authors call for further larger studies to clarify the effectiveness of the herb, but serious statistically significant harms, such as stroke, may make future larger trials difficult to justify ethically. 

The Daily Mail headline focused on the increased numbers of stroke in the ginkgo group, but the research paper only advised that the “increased stroke risk will require further close scrutiny in [ginkgo extract] prevention trials”. From this limited information it is not possible to make a definitive statement on the stroke risk of taking Ginkgo.

Where did the story come from?

Dr Hiroko H Dodge and colleagues from the Department of Public Health at the Oregon State University in the US conducted the study. The study was supported by a grant from the National Center for Complementary and Alternative Medicine. The study was published in Neurology , a peer-reviewed medical journal.

What kind of scientific study was this?

This randomised placebo-controlled trial was double blind and ran as a pilot study for 42 months.

The researchers invited 10,700 people aged 84 or older to take part in the study; 636 independent, healthy people who had no complaints of memory loss and had not previously sought assessment for memory loss responded to the invite. Telephone questionnaires were then used to screen these people and exclude those who were already showing signs of dementia. This was followed by a home visit where further cognitive tests, a medical history review and a blood sample were taken. An MRI scan of the brain was also completed. The tests and scan were carried out to make sure they did not have any other illnesses, such as diabetes (on Insulin), angina, heart failure, mental illness or Parkinson’s disease. 

This vetting resulted in the exclusion of over 400 people who had responded to the original invite. This left 134 people to reach the stage where they were randomly allocated to either receiving to ginkgo or a placebo.

The groups were further reduced after 16 participants developed medical conditions, refused to participate or were deemed inappropriate for other reasons. This left 60 people who were given 240mg of gingko each day and 58 in the placebo group who took dummy pills designed to look identical to the ginkgo pills.

The participants were assessed for dementia every year by a neurologist and every six months by a research assistant using the Clinical Dementia Rating (CDR) scale. This assesses six aspects of dementia, such as memory, judgement, hobbies and personal care. These are rated on a five-point scale and then combined to determine an overall dementia score. All participants started at ‘normal’ (CDR=0) and the researchers counted the number of people who progressed to ‘very mild’ dementia (CDR=0.5) using the tool. On this scale, ‘severe’ dementia is given a score of three.

The researchers also counted the number of adverse events and used a range of other dementia measures. They analysed the total number of people who progressed to dementia over the 42 months of follow up and the time that it took for them to develop ‘very mild’ dementia.

What were the results of the study?

Twenty-one people developed ‘very mild’ dementia over the course of the study; 14 in the placebo group and seven in the ginkgo group. However, this difference was not statistically significant. When the researchers analysed the time it took to develop this very mild dementia there was also no significant difference between the groups.

When the researchers looked at the data after the results were known (i.e., secondary analysis), they took into account that only about 69% of people continued to take the medication for the course of the study. They were therefore able to show a significant reduction in the time it took to develop very mild dementia.

Overall, seven people developed a stroke or transient ischaemic attack (a warning stroke that lasts less than 24 hour) during the study. These all occurred in the ginkgo group. This difference was statistically significant.

What interpretations did the researchers draw from these results?

The researchers say that in “unadjusted analyses”, ginkgo extract neither altered the risk of developing very mild dementia nor protected against decline in memory function. They say that in the secondary analysis, when the participants’ adherence to taking the medication was considered, a protective effect of Ginkgo was shown.

The researchers call for larger prevention trials that take medication adherence into account so that the effectiveness of the herb can be clarified. They also warn that the strokes and warning strokes observed in the ginkgo group require further study.

What does the NHS Knowledge Service make of this study?

Systematic reviews of randomised controlled trials (RCTs) of ginkgo have not found the herb to be effective. There is consensus that this sort of study design is appropriate to test the effectiveness of complementary and alternative therapies, and the reliability of the results of this study should therefore be better than for other study designs.

  • The relatively small number of patients and short duration of the study may have affected the ability of the study to detect a true result. The authors estimate that at least 2,800 recruits would be needed to have an 80% chance of detecting a reduction or risk similar to the one found in this pilot study.
  • There were sixteen people who dropped out of the study after randomisation. This is a relatively large number and may have affected the number of people rated as having ‘very mild’ dementia at follow up. A further 29 people died during the study. It is not clear how including or excluding these people will have affected the results.
  • The recruits to this study were healthy and over 84 years. This suggests that the results may not be applicable to younger people and particularly those who already have developed dementia or have risk factors for stroke.

This study author's have emphasised the non-significant benefits for Ginkgo in reducing the onset of very mild dementia, while the study has also showed a significant increase in the risk of stroke. They have called for more studies investigating the benefits, however it now seems that it would be wise and ethical to look at the harms - that is the risk of stroke - in larger studies or systematic reviews first.

Sir Muir Gray adds...

I have found no evidence to convince me of the need to take gingko. There may be benefits, but it is better to take 3,000 extra steps a day, and if you want to keep mentally active take up sudoku.

Analysis by Bazian
Edited by NHS Choices