General movements in very low birth weight pre-term infants at time of discharge Analysis of results with respect to neurodevelopmental outcome

Abstract

Background: During the last two decades Prechtl´s method on the qualitative assessment of general movements (GMs) has repeatedly been proven to be a highly sensitive and specific diagnostic tool for the early diagnosis of brain impairment. Compared to other techniques, e.g. magnetic resonance imaging cranial ultrasound, and traditional neurological examination it is quick, non-invasive, non-intrusive and cost-effective. However, despite these advantages longitudinal GM assessment is not yet widely introduced in clinical practice everywhere. Objective: The aim of this study was to describe and analyse the relation between GM assessment performed only once at time of discharge and later neurodevelopmental outcome. Patients and Methods: Between July 2006 and June 2009 GMs of 72 very low birth weight pre-term infants admitted to the Division of Neonatology at the Medical University Graz were filmed once at time of discharge. The recordings were assessed retrospectively according to Prechtl´s method. The results were compared with neurodevelopmental outcome that was measured at the specific time frame of 1-2 years using the Touwen Infant Neurological Examination, the Gross Motor Function Classification System and the Bayley Scale of Infant Development. Results: Our findings widely go along with previously published data as we found a strong relation between certain GM patterns and neurological test results on the one hand and developmental test results on the other hand. 42 (95%) of 44 infants with normal pre-term movements had a normal neurodevelopmental outcome. Cramped-synchronized or poor-repertoire was followed by neurological abnormalities in 10 (36%) of the 28 infants (6 had CP and 4 had other neurological or developmental disorders). Sensitivity and negative predictive value were quite high (83% and 95%, respectively) whereas specificity and especially positive predictive value were somewhat lower (70% and 36%, respectively). Conclusion: Longitudinal GM assessment should be preferred to a single shot assessment at 36 weeks as it has a limited power to predict later neurological outcome properly. In particular, a large number of false positives resulting in low specificity during pre-term period is responsible for the limited use.

Date of Award	2011
Original language	English
Awarding Institution	Medical University Graz
Supervisor	Berndt Urlesberger (Supervisor) & Christa Einspieler (Co-supervisor)