Melatonin Administration to Blind People: Phase Advances and Entrainment

Robert L. Sack, Alfred J. Lewy, Mary L. Blood, James Stevenson, L. Donald Keith

Research output: Contribution to journalArticlepeer-review

216 Scopus citations


The purpose of this study was to test the phase-shifting and entraining effects of melatonin in human subjects. Five totally blind men were found in a previous study to have free-running endogenous melatonin rhythms. Their rhythms were remarkably stable, so that any deviation from the predicted phase was readily detectable. After determination of their free-running period and phase, they were given exogenous melatonin (5 mg) at bedtime (2200 hr) for 3 weeks, in a double-blind, placebo-controlled trial. The effects on the endogenous melatonin rhythm were assessed at intervals ranging from several days to 2 weeks. Exogenous administration of melatonin phase-advanced their endogenous melatonin rhythms. In three of the subjects, cortisol was shown to be phase-shifted in tandem with the melatonin rhythm. A sixth subject [one of the coauthors (JS)] was previously found to have free-running cortisol and temperature rhythms and was plagued by recurrent insomnia and daytime sleepiness. He had tried unsuccessfully to entrain his rhythms for over 10 years. After he took melatonin (7 mg at 2100 hr), his insomnia and sleepiness resolved. Determination of his endogenous melatonin rhythm after about a year of treatment demonstrated endogenous rhythms that appeared normally entrained. The treatment of blind people with free-running rhythms has many advantages for demonstrating chronobiological effects of hormones or drugs.

Original languageEnglish (US)
Pages (from-to)249-261
Number of pages13
JournalJournal of biological rhythms
Issue number3
StatePublished - Sep 1991
Externally publishedYes

ASJC Scopus subject areas

  • Physiology
  • Physiology (medical)


Dive into the research topics of 'Melatonin Administration to Blind People: Phase Advances and Entrainment'. Together they form a unique fingerprint.

Cite this