Reviewed: Akhondzadeh S, Sabet MS, Harirchian MH, et al. A 22-week, multicenter, randomized, double-blind controlled trial of Crocus sativus in the treatment of mild-to-moderate Alzheimer’s disease. Psychopharmacol. 2010;207(4):637-643.

The most common form of dementia in elderly people is Alzheimer’s disease. Alzheimer’s causes progressive memory loss, impaired mental function, apathy, disorientation, and physical deterioration. Medications are available to relieve some symptoms, but the effectiveness of these medications is usually modest and temporary. The dried stigmas of saffron (Crocus sativus, Iridaceae) are used in Persian traditional medicine to treat dementia and depression. Animal and laboratory studies support the use of saffron to improve cognitive function, and an unpublished placebo-controlled trial in humans suggests saffron may be effective in people with Alzheimer’s. The purpose of this clinical trial was to compare the efficacy of saffron and donepezil, a drug that is prescribed to treat Alzheimer’s disease, in people with mild to moderate Alzheimer’s.

This double-blind, randomized, controlled trial was conducted by researchers from Tehran University of Medical Sciences in Tehran, Iran. Men and women with mild to moderate Alzheimer’s who were ≥ 55 years of age were enrolled in the study from 2007 to 2009. Subjects were required to be ambulatory and have adequate hearing and vision, and they must have had a history of cognitive decline for at least 6 months and a CT or MRI scan of the brain performed within the last year. No Alzheimer’s medications, ginkgo (Ginkgo biloba, Ginkgoaceae), or saffron may have been used for at least 3 months prior to the study. Those with uncontrolled disease conditions or any psychiatric diagnosis other than Alzheimer’s were excluded from the study.

The trial enrolled 54 subjects who were randomly allocated equally to the saffron group or to the donepezil group using a computer-generated code. Subjects in the saffron group received 15 mg saffron extract daily (1 capsule) for the first 4 weeks and 30 mg saffron extract daily (2 capsules) for an additional 18 weeks. The saffron extract (Green Plants of Life - IMPIRAN Co., Ltd.; Tehran, Iran) was prepared by extracting dried and milled stigmas with 80% ethanol at a 1:15 w/v ratio, followed by drying. The extract contained 0.13-0.15 mg safranal and 1.65-1.75 mg crocin per 15 mg. Subjects in the donepezil group received 5 mg donepezil (1 capsule) daily for the first 4 weeks and 10 mg donepezil (2 capsules) daily for an additional 18 weeks (Aricept®; Pfizer, Inc.; New York, NY). The saffron and donepezil capsules were identical in appearance.

The Alzheimer’s Disease Assessment Scale-Cognitive Subscale and Clinical Dementia Rating Scale-Sums of Boxes were used to measure changes in cognitive performance and clinical profiles. These assessments were performed every 2 weeks. Brief physical examinations were performed and adverse events were collected at each clinic visit. Complete physical examinations and ECGs were conducted at baseline and after 8 weeks and 22 weeks.

The mean age of the participants was 72.7 years in the saffron group and 73.9 years in the donepezil group. There were no significant differences between the 2 groups for age, gender, education level, time since diagnosis, or scores on both rating scales. Of the 54 subjects who were randomized, 47 completed the study. In the saffron group, 1 subject withdrew consent, and 2 subjects were lost to follow-up. In the donepezil group, 1 subject withdrew consent, 2 subjects were lost to follow-up, and 1 subject discontinued for other reasons.

Scores on the Alzheimer’s Disease Assessment Scale-Cognitive Subscale improved by 3.96 points in the saffron group and by 3.77 points in the donepezil group from baseline to 22 weeks. Scores on the Clinical Dementia Rating Scale-Sums of Boxes improved by 0.77 points in the saffron group and by 0.83 points in the donepezil group from baseline to 22 weeks. The improvement in scores for both assessments was not significantly different between the 2 groups. The changes in scores from baseline to 22 weeks were also not statistically significant for either scale in either group. adverse events were similar between the 2 groups with the exception of vomiting, which was reported more often in the donepezil group (P = 0.05). Other minor adverse events in both groups included dizziness, dry mouth, fatigue, and nausea. Also, one saffron user experienced hypomania. None of the subjects dropped out of the study because of adverse events.

In this study, saffron was not different from donepezil in improving scores on 2 assessments of Alzheimer’s and resulted in less vomiting. The authors state that a 3-point change in the Alzheimer’s Disease Assessment Scale-Cognitive Subscale score is clinically meaningful. Based on this, the improvement seen in both groups would be considered clinically meaningful. The authors do not explain what change in the Clinical Dementia Rating Scale-Sums of Boxes score would be clinically meaningful over a 22-week period. However, the authors conclude that this study “provides preliminary evidence of a possible therapeutic effect of saffron extract in the treatment of patients with mild-to-moderate Alzheimer’s disease” and that saffron was safe and well tolerated during 22 weeks of treatment.

While the discussion section states that subjects in the saffron group experienced “statistically significant benefits in cognition after 22 weeks treatment,” the results section states that no significant differences were observed for the change in scores at week 22 compared to baseline for the Alzheimer’s Disease Assessment Scale-Cognitive Subscale (P = 0.85) or the Clinical Dementia Rating Scale-Sums of Boxes (P = 0.83). This discrepancy in interpretation of the statistical significance is puzzling.

According to one HerbalGram peer reviewer, this could be explained by the researchers’ inadvertently leaving out the full results that they described in the statistical analysis section (i.e., leaving out the p values of the t–tests comparing mean changes for saffron and donepezil groups). The reviewer pointed out that the researchers gave p values for ANOVAs and Fisher Exact Tests, but not for the t-tests. The HerbalGram peer reviewer added that, as it stands, the conclusions are not supported by the results, and perhaps the best that can be said is that in this sample, saffron and donepezil were equally effective, but saffron was better tolerated. Should the journal publish a clarification of this omission, it would increase the significance of the findings in this groundbreaking study.

This study appears to be the first randomized, double-blind trial comparing the effects of saffron and an acetylcholinesterase-inhibiting drug in people with Alzheimer’s. The authors of the study point out that the sample size was small and the duration of the study was relatively short for evaluation of Alzheimer’s progression, yet the potentially beneficial effects of saffron on cognitive function observed in this study are consistent with the traditional use of saffron in Persian medicine and with findings from animal and laboratory studies. 

—Heather S. Oliff, PhD