Near resolution of sleep related rhythmic movement disorder after CPAP for OSA

1. Introduction 

Sleep related rhythmic movement disorder (RMD) is characterized by repetitive, stereotyped, and rhythmic motor behaviors that involve large muscle groups and occur predominantly during drowsiness and light sleep, though they can appear in any sleep stage [1]. Diagnosis is often but not always straightforward, especially when RMD involves movement patterns other than the most classic body rocking or head banging, or when the patient presents initially in adulthood rather than childhood. We describe an adult diagnosed with RMD only after being treated for restless legs syndrome (RLS) and referred for evaluation of obstructive sleep apnea. Furthermore, treatment of this patient’s obstructive sleep apnea dramatically reduced the RMD.

2. Case description 

A 38-year-old man was referred to our sleep laboratory because of a history of loud snoring and witnessed apneic episodes. The patient denied any history of daytime sleepiness, and his Epworth Sleepiness Scale score was only 2 (on a scale of 0 to 24) [2]. He did have a history of repetitive bilateral leg rolling movements noted nightly by his wife since they were married 17 years prior. The patient typically had these movements for extended periods before falling asleep, but also occasionally during the day while fully awake but sedentary. The patient was aware of these movements while awake during the day, but not fully aware of them at night. He did not feel refreshed upon waking up in the morning. He did not experience any urge to move his legs, or unpleasant sensations in his legs. He had been treated nonetheless by his primary care physician for presumed RLS with ropinirole (1mg each night) without any clinical improvement.

3. Video and image analysis 

A diagnostic video-polysomnogram (V-PSG) revealed severe obstructive sleep apnea with an apnea-hypopnea index of 87 events per hour of sleep and a minimum oxygen saturation of 73% [3]. His periodic limb movement index was 0.

The V-PSG also showed frequent rhythmic movements of the lower extremities and pelvis (see online video). These movements occurred at a rate of about 1.5Hz, and involved side-to-side rolling of the legs in unison, sometimes more prominently on the left, with movement often extending into the pelvis as well. The movements were not associated with any single body position. Each bout of movements lasted between 7 and 20second. The movements were noted mostly during wakefulness, awakenings, and brief arousals, though they also occurred at times during NREM and REM sleep (Fig. 1A). For much of the study, the movements appeared to be nearly time-locked with arousals, i.e., they commenced with the onset of each arousal and resolved quickly in synchrony with transition of the EEG back to stage 1 or another sleep stage (Fig. 1B).

View Large Image Download to PowerPoint View Large Image Download to PowerPoint View Large Image Download to PowerPoint

Fig. 1. (A) Rhythmic movements often occurred at the end of apneas, as demonstrated on this 1min epoch recorded during REM sleep. (B) Rhythmic movements appeared to be time-locked with arousals at the termination of obstructive apneas, as demonstrated on this 2min epoch recorded during stage 1 sleep. (C) During the CPAP titration study, at the optimal setting of 12 cm of water, the patient had fewer arousals; many, but not all, now showed no associated rhythmic movements. E2-M1, electro-oculography, right eye; E1-M2, left eye. Chin1–Chin2, surface EMG of the chin. F4-M1, right frontal EEG (international 10–20 EEG electrode placement system); F3-M2, left frontal; C4-M1, right central; C3-M2, left central; O2-M1, right occipital; O1-M2, left occipital. ECG2–ECG1 and ECG2–ECG3, electrocardiographic leads. LAT1–LAT2, left anterior tibialis surface EMG; RAT1–RAT2, right anterior tibialis surface EMG. NPRE, nasal pressure; N/O, oronasal thermocouple. MFLO, mask flow. ORAL, oral leak. THOR and ABD, thoracic and abdominal excursion (piezo-electric belts). SpO2, oxyhemoglobin saturation (%). Pleth, pulse waveform from oximeter.

The patient underwent a follow-up continuous positive airway pressure (CPAP) titration study. At a CPAP setting of 12cm of water, the patient showed dramatic improvement in apnea and arousals and nearly complete resolution of the rhythmic movements (Fig. 1C). Treatment at home with CPAP for two weeks at the time of this writing eliminated snoring and witnessed apneas and also dramatically improved the RMD by the patient’s report.

4. Discussion 

Our patient’s presentation with restless sleep and abnormal leg movements led to his initial diagnosis and treatment for RLS, despite the lack of any urge to move the legs or unpleasant leg sensations [1], any response to a dopamine agonist, or any clinical consequences from his involuntary leg movements [4]. This patient’s polysomnogram later showed no evidence of periodic limb movements (PLMs), normally observed in 80–90% of patients with RLS [1]. Other diagnostic considerations also could have included epilepsy, but the polysomnogram showed no epileptiform activity or seizures on a limited EEG montage, and the movements occurred in REM sleep, in addition to NREM, which would be unusual for seizures [5], [6]. Extrapyramidal disorders such as Parkinson’s disease might also produce rhythmic movements during brief awakenings in a patient with a relevant history.

Our RMD patient had severe obstructive sleep apnea, and CPAP greatly ameliorated his leg movements. Rhythmic leg movements have been noted previously in association with arousals from sleep, in at least one patient with severe sleep apnea and another with mild sleep apnea, but RMD was not diagnosed and response to CPAP was not reported [7]. Descriptions of alternating leg muscle activation (ALMA) have focused on EMG findings similar to those of our patient, again in association with arousal, but prolonged movements prior to sleep onset were not noted, RMD could not be diagnosed, bilateral events were alternating rather than synchronous, and ALMA did not change with CPAP [8]. Mayer and colleagues recently reported that in three of five RMD patients who happened to have sleep apnea, arousals at the termination of apneas frequently seemed to trigger the movements [4]. These authors also mentioned that for three patients, RMD frequency was reduced, and for two RMD duration was reduced, during recordings made while CPAP was administered [4]. Our report now extends these observations to detail an intimate link in our patient’s case between RMD and obstructive sleep apnea; to provide the first video demonstration of RMD time-locked to arousals, and to document acute as well as longer term near resolution of symptomatic RMD on CPAP. Together, these findings suggest that RMD can be triggered by other primary sleep disorders or sometimes confused with other sleep disorders. Thorough clinical evaluation combined with V-PSG may prove important to accurate diagnosis of RMD.

4. Disclosure 

The authors report no conflicts of interest.