Additional Supporting Information may be found in the online version of this article.

MDS_23645_sm_suppinfofigure1.tif10065KFigure e-1. Surface EMG recordings in orbicularis oculi (O Oc), abductor digiti minimi (Abd dm), intercostalis (T2 and T5 levels), rectus abdominis (T8, T10, and T12 levels), iliopsoas (Ilio ps), and tibialis anterior (Tib ant) muscles obtained from patient 2, showing slow rostral and caudal spreading from a T8 generator of the spontaneous myoclonus and the reflex EMG bursts to motor cortex or supraorbital nerve stimulation. Oblique lines are drawn from jerk onset at different myelomeres to highlight the propagation pattern.
MDS_23645_sm_suppinfofigure2.tif10948KFigure e-2. Diffusion tensor imaging with fiber-tracking of the spinal cord at cervical level. Fiber-tracking reconstruction showed fiber loss in the posterior lemniscal pathways in patient 1 (126 fibers on the right and 143 fibers on the left), but was normal in patient 2 (285 fibers on the right and 228 fibers on the left) according to our control values. DTI was acquired with a 16-direction, single-shot, spin-echo echoplanar sequence, according to the following parameters: repetition time (TR): 3550 ms; echo time (TE): 62 ms, field of view (FOV): 250×250 mm2, section thickness: 2 mm, voxel size: 2×2×2 mm; b values: 0 and 1000 s/mm2, and 60 sections covering the whole cervical spinal canal. DTI-based color maps and fiber tracking were generated by using the TrackVis freeware (version 0.4.2; MGH, Boston, Mass, USA). Tracking was performed in both retrograde and anterograde directions from an initial region of interest situated at the level of the thoracic spinal cord drawn on Fraction Anisotropy (FA) color maps. The propagation was terminated when it reached a voxel with FA < 0.08 or when the angle between two consecutive steps was greater than 50.
MDS_23645_sm_suppinfofigure3.tif1841KFigure e-3. Representation of the neural pathways presumably involved in long-latency abdominal muscle reflexes to supraorbital nerve stimulation (in green), median nerve stimulation (in purple), and motor cortex stimulation (in blue). Blink reflex studies have showed that supraorbital nerve stimulation activates short sensory afferent fibers (from the face) and then cells in the trigeminal sensory nuclei that project to the bulbar reticular formation. The circuit recruited by median nerve stimulation in addition to that recruited by supraorbital nerve stimulation is indicated as a purple dotted line (and contributes to a 25-ms increase in reflex latency, since the latency of cortical evoked potentials to median nerve stimulation at the wrist is about 20 ms and the conduction time for the descending volley from cortex to brainstem could be about 5 ms). The circuit recruited by motor cortex stimulation in addition to that recruited by median nerve stimulation is indicated as a blue dotted line (and contributes to a 30-ms increase in reflex latency, since the latency of abdominal motor evoked potential (MEP) to cortical stimulation is about 20 ms and the difference in the conduction time for ascending sensory volleys between abdominal and wrist origin could be about 10 ms). A thoracic spine generator and the propriospinal pathways (red line and arrows) are presumably the common system involved in the spreading of the reflex responses to the three conditions of stimulation as in the spontaneous propagation of propriospinal myoclonus (PSM). Abdominal reflex latency was shorter following thoracic spine stimulation than supraorbital nerve stimulation (not shown). This is not consistent with a descending control from the brainstem that should logically produce longer reflex latencies to spinal than trigeminal stimulation. Therefore, one may suggest that intraspinal loops involving the propriospinal system could also contribute to the generation of abdominal reflexes in PSM (not shown). However, a simple segmental circuit mediated by proprioceptive afferents from the abdominal muscles (triggered by MEP-associated muscle twitches) is not conceivable because the difference in latency between the abdominal MEPs and reflex responses is not the same following cortical and thoracic magnetic stimulation (120 and 55 ms, respectively).
MDS_23645_sm_suppinfotable1.doc53KTable e-1. Literature data regarding reflex responses elicited in abdominal wall muscles of patients with propriospinal myoclonus in response to mechanical or electrical sensory stimuli

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.