To the Editor: We read with interest the Clinical Case Conference by Vijay A. Mittal, Ph.D., et al (1), published in the September 2010 issue of the Journal. The case concerned a young man with a pineal region tumor who presented prodromal obsessive-compulsive behaviors and psychotic symptoms. We recently provided treatment in a similar case, involving a young patient who had circadian rhythm disturbances in relation to a pineal tumor.

"Ivan" was a 19-year-old Caucasian man who presented to our sleep service 6 years after insertion of a ventriculo-peritoneal shunt, followed by resection of and chemotherapy and radiotherapy for a secreting germ cell pineal tumor, which were performed in 2001. No recurrence of the tumor was shown at follow-up assessment.

Shortly after insertion of the shunt, the patient exhibited insomnia, a severely disturbed sleep-wake cycle, and fragmentation of nocturnal sleep. In June 2004, he developed paranoid ideas and bipolar disorder. Treatment was started, and when the patient was seen by us in December 2007, he was receiving treatment with clorazepate dipotasique (40 mg), olanzapine (7.5 mg), and escitalopram (20 mg). His school performance had been severely affected, and he stayed at home and was socially isolated. He complained of severe fatigue and described his sleep as not restful.

Polysomnography, performed at the beginning of October 2007 under psychotropic treatment, showed light fragmented sleep, and actigraphy showed an irregular sleep-wake cycle, with abnormally raised nocturnal activity. Twenty-four hour urinary 6-sulfatoxymelatonin levels were barely detectable with absence of the typical 24-hour sleep-wake cycle. Treatment with controlled-release melatonin was started in October 2007. The patient's sleep rapidly stabilized, with restoration of a regular sleep-wake cycle and marked improvement of psychotic symptoms, permitting progressive withdrawal of psychotropic medication between January and April 2008. Further sleep analyses, performed in mid-November 2007, showed restoration of slow-wave sleep, with diminished sleep fragmentation, and actigraphy confirmed restoration of a normal sleep-wake cycle. Treatment with controlled-release melatonin was discontinued in 2009, and the patient remains medication- and symptom-free, with a stable sleep-wake cycle.

In our patient, psychiatric symptoms seemed to be triggered by severely disordered sleep. Controlled-release melatonin was used in order to replace the normal physiological secretion of melatonin, which continues throughout the night. Rapid-release melatonin has a half life of 30-60 minutes and is used as a circadian signal (in jet lag, for example) and is not suitable for replacement treatment.

It is interesting that following treatment with controlled-release melatonin and restoration of a normal sleep-wake cycle, not only did psychiatric symptoms cease, but the patient was also able to discontinue melatonin treatment. We hypothesize that pineal gland destruction led to an abnormal sleep-wake cycle, which triggered the psychiatric symptoms, and restoration of the sleep-wake cycle led to resolution of the psychiatric symptoms, permitting the patient to use other signals to synchronize sleep-wake times, thus preventing relapse.