Abstracts

Rationale: Sleep starts are physiological phenomena characterized by sudden and brief muscle contraction in the trunk and limbs at sleep onset. In 1999, Fusco et al. described a case series of repetitive sleep starts (RSS) in three children with epilepsy and spastic quadriplegia, which is the only such case series reported to date. Their patients presented with repetitive "crescendo-decrescendo" muscle activities on electromyography (EMG), which were compatible with epileptic spasms, but had no ictal changes in electroencephalography (EEG). RSS are often misdiagnosed as West syndrome, and children with RSS may unnecessarily be treated with anti-epileptic drugs. To prevent this, we conducted a clinical and electroencephalographic study that clarify the characteristics of RSS. Methods: Children with RSS referred to our facility between November 2006 and December 2017 were included. We defined RSS as follows: 1) cluster of characteristic movements resembling epileptic spasms, with a "crescendo-decrescendo" pattern of EMG, 2) five starts or more in a cluster, 3) events during sleep-wake transition and/or sleep stage N1 and/or N2, 4) no changes in ictal EEG, and 5) video and EMG results that are not consistent with voluntary body movement or myoclonus. Ictal video-EEGs were recorded using at least 19 electrodes according to the international 10-20 system; electrocardiography and/or EMG were also performed. Clinical information, brain magnetic resonance image (MRI), and ictal video-EEG data were retrospectively analyzed. Results: Ten children fulfilled the inclusion criteria. Background diagnoses were perinatal hypoxic-ischemic encephalopathy (n=4), West syndrome of unknown etiology (n=2), and traumatic brain injury (n=1). Of the remaining three children, two were diagnosed with autism spectrum disorder during follow-up, and one had no recognizable neuropsychological disorder at the final follow-up. RSS onset ranged from 3 to 46 months of age. The number of starts in one RSS cluster and intervals between starts ranged from 5 to 547 and <1 to 60 s, respectively; duration of surface EMG activity ranged from 0.3 to 5.4 s. All analyzed parameters were similar between children with (n=5) and without (n=5) corticospinal tract (CST) injury, as observed on brain MRI. RSS disappeared in four children without therapeutic intervention. Although most of the children did not develop epilepsy, focal seizures persisted in two children with West syndrome. Conclusions: This study clarified clinical and electroencephalographic characteristics of RSS. To avoid the unnecessary administration of anti-epileptic therapy, clinicians should be aware of this condition and distinguish it from other conditions that present as involuntary movements or seizures, especially epileptic spasms. RSS were seen in children without CST injury; this may imply that the lack of inhibition of an injured CST is not a prerequisite for the pathophysiology of RSS. Further investigation is needed to elucidate the exact mechanism of RSS. Funding: This study was not supported by any sources of funding.