Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Hydergine is effective for dementia

ACP J Club. 1995 Jan-Feb;122:17. doi:10.7326/ACPJC-1995-122-1-017

Source Citation

Schneider LS, Olin JT. Overview of clinical trials of hydergine in dementia. Arch Neurol. 1994 Aug; 51:787-98.



To evaluate the effect of hydergine in patients with dementia.

Data Sources

Studies were identified through MEDLINE, EMBASE/Excerpta Medica, and 2 proprietary databases using the terms hydergine, ergoloid mesylates, dihydroergocristine, dihydroergocryptine, dihydroergotoxine, and dihydroergocornine. Previously published reviews and unpublished reports were also reviewed.

Study selection

Selected studies concerned hydergine treatment of elderly patients with possible cognitive impairment or dementia. Trials were excluded for the following reasons: The treatment duration was not described; the trial was not double-blind or placebo-controlled; the trial was either a crossover trial without first-period data or an open trial; hydergine was administered intravenously; the patient sample was inappropriate; outcomes were inadequately reported; end points were nonclinical; or the outcome was reported elsewhere. 151 studies were identified, and 47 met inclusion criteria.

Data extraction

Patient status (inpatient or outpatient), diagnosis (dementia consistent with possible Alzheimer disease, vascular dementia, primary dementia of undetermined type, cerebral insufficiency, and cerebral or senile deterioration), age, sex, duration of treatment, dosage, and year of trial publication. Effect sizes were calculated for 3 outcome measures: comprehensive, clinically based ratings; clinical global ratings; and neuropsychological measures.

Main results

The effect size for each study was calculated as the mean difference between 2 outcomes divided by their pooled SD. For the 44 trials that used a comprehensive clinical rating, the effect size (d) was 0.51 (95% CI 0.43 to 0.60), with the largest effect size in the diagnostic group for vascular dementia (d = 0.73, CI 0.54 to 0.93) and the smallest effect size for the group with possible Alzheimer disease (d = 0.16, CI -0.07 to 0.38). For the 23 trials that used global ratings, the effect size was 0.63 (CI 0.52 to 0.75), with no difference among the diagnostic groups; for the 22 trials that used neuropsychological measures, the effect size was 0.27 (CI 0.22 to 0.32), with differences among the 5 diagnostic groups ranging from 0.44 (CI 0.25 to 0.63) for primary dementia to 0.24 (CI 0.1 to 0.37) for possible Alzheimer disease. Hydergine had a greater effect on inpatients in trials with clinical and global ratings and on patients receiving higher dosages with measurements by global and neuropsychological scores.


Hydergine is more effective than placebo in patients with vascular dementia, in inpatients compared with outpatients, and in higher compared with lower dosages. It has modest effects in patients with Alzheimer disease. The conditions for obtaining benefits from hydergine are inadequately defined.

Sources of funding: Sandoz Pharmaceuticals, Inc. and National Institutes of Health.

For article reprint: Dr. L. Schneider, University of Southern California School of Medicine, 1975 Zonal Avenue, KAM 400, Los Angeles, CA 90033, USA. FAX 323-442-3717.


The need to revisit an old medication such as hydergine lies in the frustration of the lack of effective treatments for dementia. The modest clinical effects of tacrine and donepezil, combined with their side effects, may keep an interest in hydergine alive, rather than letting it fall by the wayside with other agents of questionable value. Practitioners relying on clinical trials to define the place of hydergine are left with more doubts than unambiguous conclusions.

Meta-analysis attempts to clarify the clinical effectiveness of a drug when various studies result in conflicting conclusions or when trial sample sizes are small (1). Schneider and Olin have presented a careful overview of the clinical trials of hydergine, using meta-analysis to quantitate its effect. Their objectives are well defined, as are their inclusion and exclusion criteria for selecting studies. Interestingly, data sources included trials identified by the drug manufacturer. The concern for publication bias is addressed by the inclusion of unpublished studies. The assumption that all the study results measured the same true effect (i.e., homogeneity) is questionable. The authors used a homogeneity statistic (Qw) but found it necessary to remove outlier groups for statistical satisfaction. The question of outlier removal and its bias of the effect size remains.

Practical application for this overview is hindered by historical factors of the primary studies. The lack of stringent diagnostic standards, sophisticated statistical analysis, and selective reporting may have biased results in favor of hydergine. Instruments for measuring outcomes used in the clinical trials were remarkable for their variety and sheer number.

Hydergine may be modestly useful in some patients. Still to be defined is which patients, what dose, and how beneficial. This meta-analysis may result in a much-needed, adequately sized, and well-designed clinical trial that will define the role of hydergine.

Charles V. Guida, MD
The Parker Jewish Geriatric InstituteNew Hyde Park, New York, USA


1. L'Abbe KA, Detsky AS, O'Rourke K. Meta-analysis in clinical research.Ann Intern Med. 1987;107:224-33.