Repetitive Ballistic Movements as a High-Risk Comorbidity in a Ventricular Assist Device Patient

PRACTICAL IMPLICATIONS

Parasomnias can be a life-threatening but modifiable comorbidity in patients with implantable ventricular assist device.

A patient with a left ventricular assist device (LVAD) presented with violent nocturnal limb movements during sleep. This sleep-related movement disorder—repetitive ballistic movements (RBMs)—underscores the importance of identification of nocturnal movements, their relationship to sleep, and to potential catastrophic consequences in susceptible patients.

Case

A 38-year-old man with anxiety disorder, substance use disorders complicated by cognitive impairment, and end-stage systolic heart failure presented with violent movements during sleep. His bed partner described repetitive jerking and more complex behaviors such as punching or writhing. He remained unresponsive during events, but maintained partial recollection. He also had snoring and apneas. Sleep scheduling was erratic, with uneven bedtimes and awakenings resulting in ∼4 hours of sleep a night.

Although problematic in themselves, the behaviors brought a new potential severe morbidity after LVAD implantation and chronic anticoagulation; episodes could cause the disconnection of power or tubing (a “driveline fracture”) resulting in either cessation of circulation or exsanguination.

Polysomnography showed decreased sleep efficiency, slow wave and REM sleep, and moderate obstructive sleep apnea (OSA) (table). A high arousal index was associated with apneas and limb movements. A trial of continuous positive airway pressure (CPAP) failed because of anxiety. Therapeutic trials with melatonin, gabapentin, and iron (ferritin was 49 ng/mL) were ineffective.

Table.

Polysomnography Results in a Man With Repetitive Ballistic Movements

With ongoing symptoms, he was admitted to the epilepsy monitoring unit. Examination demonstrated disorientation to year and reason for admission. Video-EEG with additional limb leads captured typical episodes (figure, video 1). The movements consisted of (1) complex, semipurposeful movements that appeared Akathitic such as shifting positions in bed, punching pillows, or pulling at bedclothes; (2) RBM, usually confined to legs, as repetitive flexion and extension at the hip as if swimming, or repetitive kicking with flexion and extension at the knee. The EEG showed that episodes emerged in either hypnogogic or hypnopompic transitions or during light sleep (figure). No epileptiform discharges were present. No abnormality of REM atonia occurred; in fact, most sleep consisted of only light sleep with a paucity of putative REM or deep sleep.

Figure. Video-EEG Sample in a Man With Repetitive Ballistic Movements.

Thirty-eight-second sample of video-EEG; standard 10–20 electrode placement with additional lower leg EMG (left tibialis anterior 1-2, right tibialis anterior 1-2). Repetitive ballistic movements of legs (arrows) seen during a period of light sleep.

Video 1

“Video-EEG demonstrates the patient's cycle of 1) akathisia prior to sleep, 2) sleep transition, 3) repetitive ballistic movements of limbs in sleep, 4) arousal.”Download Supplementary Video 1 ^{(17.2MB, mov)} via http://dx.doi.org/10.1212/000752_Video_1

Based on findings of RBMs, elements of somnambulism inferred from the monitoring, and lack of evidence of epilepsy, the treatment plan focused on improving sleep hygiene, stabilizing sleep state, and decreasing the arousal threshold. Clonazepam was titrated to 2 mg as a means of raising the arousal threshold and was associated with marked improvement in the patient's sleep quality and sleep-related movements.

Discussion

This case demonstrated 2 important findings. First, RBMs occur in sleep-wake transitions or light sleep. Similar to our earlier case,¹ observers describe dramatic, sometimes violent, repetitive movements that can be mistaken for epileptic seizures. Second, the current case emphasizes that diagnosis of arousal disorders, such as RBM or somnambulism, can lead to specific treatments that may avoid dire clinical consequences.

RBM resemble rhythmic movement disorder (RMD) such as jactatio capitis nocturna (head banging).² RBM shares with RMD (1) a tendency for occurrence in cognitively impaired adults such as those with intellectual disability or autism spectrum disorder (we note that these can occur in intellectually normal adults as well), (2) quasi-rhythmic movements of 0.5–2 Hz, and (3) expression during both wakefulness and sleep-wake transitions and light sleep.² RMD is associated with OSA and periodic limb movements, both of which are seen in this patient.

Although the patient was partially nonadherent, improvements in RBM occurred with treatments designed to decrease arousals as seen in our earlier patient.¹ Greater than >85% of patients with injurious parasomnias improved on treatment with benzodiazepines.³ Ideally, CPAP desensitization, dopamine agonists, or further clonazepam could have been attempted.

Diagnosis and attempts at specific treatments are important because of the risk of injury associated with arousal disorders. This patient was in particular jeopardy because his episodes could cause an LVAD driveline fracture.⁴ In a case series of 10 patients with driveline fractures, only 1 patient survived without heart transplantation, LVAD exchange/explant, or death.⁴ He was also at risk of traumatic hemorrhage on anticoagulation (the incidence of intracranial hemorrhage in LVAD patients is 2%^5,6).

In summary, non-REM movement parasomnias such as RBM present challenges in diagnosis and treatment. Spontaneous development of these movements is rare in adults. A multidisciplinary approach with the use of clinical neurophysiology techniques may be required to differentiate non-REM parasomnic episodes from REM behavior disorder, epileptic seizures, or psychogenic events and uncover other sleep disorders. Diagnosis of reversible causes of sleep fragmentation is important in vulnerable patients.

Appendix. Authors

Study Funding

No targeted funding reported.

Disclosure

The authors report no disclosures relevant to the manuscript. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.