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


Intramuscular desferrioxamine slowed the decline in living skills in Alzheimer disease

ACP J Club. 1991 Sep-Oct;115:41. doi:10.7326/ACPJC-1991-115-2-041

Source Citation

McLachlan DR, Dalton AJ, Kruck TP, et al. Intramuscular desferrioxamine in patients with Alzheimer's disease. Lancet. 1991 June 1;337:1304-8.



To evaluate the effects of intramuscular desferrioxamine (DFO; a trivalent ion-specific binding agent) on the decline in living skills in patients with probable Alzheimer disease.


2-year, randomized controlled trial, with behavioral assessments blinded to study assignment.


Not given.


63 patients with probable Alzheimer disease were selected from 1510 persons with memory disorders who were living at home and were < 74 years of age. Inclusion criteria required a computed tomographic scan showing atrophy and no infarcts; Hachinski ischemic scores of ≤ 4; no history of clinically important medical, psychiatric, or neurologic disease; and ability to give informed consent. A caregiver had to be living with the patient. 48 patients completed baseline assessments; 5 patients (10%) did not complete 2 years in the study.


25 patients were randomized to DFO, 500 mg, injected intramuscularly every 12 hours, with 2 rest days per week. After 16 days the dose was reduced to 125 mg, on the same schedule, for 2 years. Ferrous gluconate was given and rest periods were added as needed. 23 patients were randomized to no treatment or to oral lecithin, 500 mg twice daily, as placebo (n = 9). Caregivers were taught to give intramuscular injections at home and were required to keep daily records of medical status.

Main outcome measure

Change in videorecorded home-behavioral assessment (VHB) (48 samples of activities of daily living), which was measured every 6 months. The VHB correlated (r = 0.6) with the Wechsler Adult Intelligence Scale-Revised verbal IQ at baseline. Inter-rater and test-retest reliability were > 80%.

Main results

At entry, no differences existed between the groups on measures of cognition, memory, or aphasia. 5 patients in the no-treatment group died compared with no deaths in the DFO group. All data until death were used to calculate rate-of-change regression slopes for VHB performance. At 2 years, the mean slope for the no-treatment group was -1.72 compared with -0.87 for the DFO treatment group {95% CI for difference -1.64 to -0.67}* (P = 0.038). 4 patients in the DFO group reported appetite loss and 1 had gradual weight loss.


Intramuscular desferrioxamine given for 2 years retarded the decline in living skills in patients with probable Alzheimer disease.

Sources of Funding: National Health and Welfare Canada; the Province of Ontario; Ciba-Geigy.

Address for article reprint: Dr. D.R. McLachlan, Centre for Research in Neurodegenerative Diseases, University of Toronto, Room 226, Tanz Neuroscience Building, Toronto, Ontario M5S 1A8, Canada.

*Numbers calculated from data in article.


The claim of any study to retard the progression of Alzheimer disease deserves careful attention. As the authors wisely suggest, this study requires replication before clinical policy should be changed. Some aspects of the design limit our ability to interpret this study.

Intramuscular DFO was compared with oral placebo or no treatment. The psychologic effect of an oral drug (or no drug) is probably not equivalent to that of an intramuscular drug, possibly biasing the results. The groups were matched on cognitive tests in the baseline period, including the primary outcome measure, videorecorder home-behavioral assessment scores. This instrument may be a valuable method of assessment but it is new and not widely used, and it cannot be judged adequately in this short article. Many data points were missing because of patients who could not be assessed at every time point. Moreover, both groups included patients who did not deteriorate at all during the 2-year time period. Either these patients did not have Alzheimer disease, or the instrument is not sensitive to change. The data would have been strengthened by showing that aluminum excretion was greater in the DFO group than in the control group because brain accumulation of aluminum is the hypothetic mechanism of progression of Alzheimer disease.

The authors are to be congratulated on their provocative study but the use of this toxic drug in patients with Alzheimer disease is not justified until other studies confirm this work.

Peter J. Whitehouse, MD, PhD
John S. Kennedy, MDUniversity Hospitals of Clevelandand Case Western Reserve UniversityCleveland, Ohio, USA