11/08 Update:

A major rigorous study by Steven DeKosky's group showed that "daily doses of a commonly used, standardized ginkgo biloba leaf extract taken over six years did not delay the development of dementia or Alzheimer disease in 1,545 treated seniors who started with no or minor memory loss." The results are in the November 19 issue of JAMA.

Earlier research indicated that cognitive functioning, psychiatric symptoms, and activities of daily living all improve wit Ginkgo, when compared to placebo.
Napreyenko A, Ihl R, Schlaefke S, Hoerr R. Ginkgo biloba Extract EGb 761® in the Treatment of Dementia with Neuropsychiatric Features: A Randomized, Placebo-controlled, Double-blind Trial. Poster, 21st International Conference of Alzheimer's Disease International, Istanbul, Turkey, September 2005.
A recent German study of a special extract of Ginkgo has been shown to delay decline in ways that might delay nursing home placement (Haan, J & Horr, R Delay in progression of dependency and need of care of dementia patients treated with Ginkgo special extract EGb 761 Wien Med Wochenschr 2004 Nov; 154:511-514) other recent results suggest no benefits for prevention of AD (Journal of the American Medical Association, August, 2002- blind placebo controlled study found no significant memory, learning or concentration benefits).

Early Study: Ginkgo Biloba and Alzheimer's Disease
Le Bars PJ, et al. A Placebo-controlled, double-blind, randomized trial of an extract of Ginkgo bilboa for dementia. JAMA 278:1327-1332, 1997.

This study was conducted at 6 centers in the United States, and used a 52 week, double-blind, placebo-controlled, parallel-group randomized design. Patients had to be older than 45 and have a diagnosis of dementia (MMSE 9-26) due to either Alzheimer's disease (AD) or vascular disease. Patients were assessed by the ADAS-Cog for cognitive changes, by the Geriatric Evaluation by Relative's Rating Instrument (GERRI) for activities of daily living, and by the Clinicians Global Impression of Change (CGIC) for the physician's assessment of change. Data were analyzed on an intent to treat basis, with the last assessment carried forward, because of a low proportion of patients completing the study, and the unequal distribution of dropouts among treatment arms. One-hundred sixty-six patients were enrolled in the Ginkgo bilboa (EGb) arm and received 120 mg of the compound daily, and 161 were enrolled in the placebo arm. However, only 78 (47%) and 59 (36.6%) of patients in each respective arm completed the protocol. Patients had to reach 12 weeks of treatment to be included in the analyses.

For the ADAS-Cog there was no significant change in the EGb arm (0.1 unit), but a significant worsening was seen for the placebo group (-1.5), p = .04 between groups. For the GERRI, mild improvement was seen in the EGb group (-.06) and mild worsening was seen for the placebo group (.08), p = .004. No difference was seen in the CGIC for either group.

A greater proportion of patients receiving EGb showed improvement, while a greater proportion of the placebo group showed worsening for both the ADAS-Cog and GERRI. Thus, groupwise changes were not due to dramatic improvement or decline in a few outlying patients. The adverse events were similar in the two groups.

Comment: Ginkgo bilboa has been used for a number of years in Europe to treat a variety of neurological disorders, and is approved in Germany for the treatment of dementia. This study represents the first well controlled clinical trial in the United States to examine the efficacy and safety of this plant extract in the treatment of dementia. The main limitation of the study was the high proportion of patients dropping out, and the unequal distribution of these dropouts among the treatment arms. The authors discuss the possible influence of these dropouts on the results and conclude that this should have helped the placebo arm more than the EGb arm. Other limitations are the use of an intent to treat analysis, lack of effect seen in the CGIC rating, and limited assessment of doses in the EGb arm. Overall, however, the results are comparable though somewhat less than the improvements seen with the acetylcholinesterase inhibitors. One limitation not discussed by the authors is the potential variability in efficacy of different formulations of the extract. At present there is no objective assessment of the potency of these formulations in the United States. Despite these limitations, the results suggest that further studies are warranted and the use of Ginkgo bilboa should be considered as an adjunctive agent in the treatment of AD or vascular dementia. DRG