NemoImage 11, Number 5, 2000, Part 2 of 2 Parts ID E al@ DISORDERS - NEUROLOGY Sensorimotor Reorganization in Prenatally Acquired Hemiparesis M. Staudt*t, G. Niemannt, Martin Lo&e*, Michael Erb*, I. Kraegeloh-Mann?, Wolfgang Grodd* *Section Exp. MR of the CNS, University of Tuebingen, Germany iDept. of Child Neurology and Developmental Medicine, University of Tuebingen Introduction Congenital spastic hemiparesis is often caused by prenatal brain lesions acquired during the last trimester of pregnancy (1). The most common neuroimaging finding consists of periventricular T2-hyperintensities and/or asymmetrical ventricular enlargements, encroaching pyramidal tract fibers. We performed functional magnetic resonance imaging (WRI) to monitor the alterations in the organization of the sensorimotor system for hand movement in such hemiparetic patients. Patients and controls 6 patients (age 19-23 years) with congenital hemiparesis (5 right-sided, 1 left-sided) due to unilateral periventricular brain lesions of prenatal origin and 10 healthy, right-handed controls (age 19-50 years) were included in the study. In addition to structural and functional MRI, all patients underwent a neurological examination, including a detailed observation of mirror movements, i.e. involuntary movements of the opposite hand during voluntary movement of one hand. fMRI data were obtained using multislice EPI sequences (1.5 Tesla, TR 8 s, TE 84 ms, 27 axial slices) during alternating epochs (6 scans each) of rest versus opening/closing of the hand (l/set, metronome-paced). Realignment, smoothing, and statistical analysis were performed using SPM99b on a single-subject basis. Activation maps (p < 0.05) were superimposed on the EPI to identify areas of activation according to the individual gyral anatomy. The intensity of activation was estimated using the maximum T-values within anatomically defined regions (contra- and ipsilateral pericentral cortex = Ml/Sl, supplementary motor area = SMA, cerebel- lum). ReSUlt.3 In the patients moving their paretic hand (PH in Fig. 1) activation was observed mainly in contra- and ipsilateral Ml&I, SMA and cerebellum; activation outside these “classical” sensorimotor areas was not more frequent than in controls. However, within Fig. 1. fMRI activation (p < 0.05 corm&ad, SPM 99b) dung movement these areas, considerable differences were observed: of the paretic hand (PH) and the nonparetic hand (NH) in a patient with 4/6 patients showed an extension of Ml/S1 activa- congenital right hemiparesis compared to movement of the right hand (RI-I) and left hand (LH) in a healthy right-handed control. tion posteriorly, beyond the postcentral sulcus in the superior parietal lobe (only 10% in controls). In addition ipsilateral Ml/S1 activation exceeded the normal range in the 4 patients with mirror movements of the nonparetic hand. For movement of the nonparetic hand (NH in Fig. 1) no significant differences compared to controls was found, although mirror movements of the paretic hand were present in S/6 patients. Prenatally acquired periventricular brain lesions lead to reorganization of the sensorimotor system, with increased activation in the superior parietal lobe and in the ipsilateral Ml/S]. Mirror movements were observed both in the nonparetic and in the paretic hand. During voluntary movement of the paretic hand (PH in Fig. I), the mirror movements of the nonparetic hand were associated with an increased activation in the unaffected hemisphere, whereas during voluntary movement of the nonparetic hand (NH in Fig. l), no activation in the affected hemisphere correlating with me mirror movements of the paretic hand was seen. This suggests a different pathomechanism for these phenomenons: The increased activation in the unaffected hemisphere during voluntary movement of the paretic hand could induce mirror movements in the nonparetic hand via the normal crossed pyramidal tract. The lack of activation in the affected hemisphere during voluntary movement of the nonparetic hand points to the unaffected hemisphere as the origin for mirror movements of the paretic hand via ipsilateral, uncrossed projections. Both findings are consistent with tramcranial magnetic stimulation studies showing ipsilateral motor responses only for stimulation of the unaffected hemisphere in hemiparetic children (2). References (1) Niemann G, 1994, Dev Med Child Neurol. 36, 943-950 (2) Carr LJ. 1993, Brain, 116, 1223-1247 s135