REM Sleep Behavior Disorder and "Paradoxical kinesis" in patients with Parkinson disease

Abstract & Commentary

By Charles P. Pollak, MD, Professor, Clinical Neurology, Weill College of Medicine. Dr. Pollak is a stockholder for Merck, and is on the speaker's bureau for Merck.

Synopsis: Many patients with Parkinson disease have normalization of movement during REM sleep, similar to the beneficial effects of levodopa while they are awake.

Source: De Cock VC, et al. Restoration of normal motor control in Parkinson's disease during REM sleep. Brain. 2007;130: 450-456.

A peculiar feature of Parkinson disease (PD) that is familiar to most neurologists is the ability for patients to briefly regain the capacity for rapid movement under urgent conditions, such as a dangerous fire from which it is necessary to escape. First described in 1921, the mechanism of this "paradoxical kinesis" remains unknown. It has now been discovered that the capacity for normal movement may also be regained in PD during REM sleep in those PD patients who develop REM sleep behavior disorder (RBD). Normally, the capacity for movement is lost during REM sleep owing to inhibition of motor systems at the brainstem level. In RBD, this inhibition is lost, enabling movements to arise that correspond to the events of REM dreams (dream enactment). In this investigation, these movements were compared to the waking movements of PD patients and were found to be greatly improved,

Of 100 non-demented PD patients that were interviewed, 59 were found to have RBD. Their bed partners were asked to evaluate the quality of movements, voice and facial expression and compare them to the "ON" levodopa state. Fifty-three of the 59 had bed partners who were able to evaluate them, and all reported improvement of at least one component of motor control during RBD. Video monitoring during polysomnography confirmed that movements in REM sleep (talking, laughing, yelling, reaching, gesturing, punching, kicking) were "surprisingly fast, ample, coordinated and symmetrical," though also "jerky, violent and often repetitive". Dreams during RBD often involved fighting or running/fleeing and commonly resulted in injuries to self or co-sleepers. Interestingly, all patients had asymmetrical PD when awake and used the disabled arm, hand and leg more during RBD.


This is an imaginative, wonderfully interesting investigation that makes its point convincingly. RBD is a valuable "experiment of nature," but it is well to remember that it was first created in cats by Sastre and Jouvet by means of brainstem lesions. The cats then appeared to "act out their dreams" whenever they entered REM sleep. This remained a laboratory curiosity until an apparent clinical counterpart was described in 1986 by Schenck and others. PD is one of several disorders that may be associated with RBD. In the present study, RBD was present by history in 60% of the PD patients. In 22%, it had developed before the onset of PD, by a few months to as long as 4 years. The normalized movements of paradoxical kinesis and those associated with RBD seem to differ, because the latter were also observed by the authors when PD patients were acting out dreams with little or no emotional content. RBD dreams are nevertheless usually intensely emotional. The authors speculate that RBD movements have cortical origin and follow the pyramidal tract, bypassing the parkinsonian extra-pyramidal system.