ABSTRACT
BACKGROUND: Assessment of general movements (GMs) in preterm infants is qualitative and potentially subjective. Accelerometers provide quantitative data that could overcome the problems of the GMs assessment. STUDY AIMS: To determine if quantitative measures (obtained from four tri-axial accelerometers) correlate with GMs assessments performed in the preterm period at 28- or 32-weeks postmenstrual age (PMA). STUDY DESIGN: Prospective observational study. Tri-axial accelerometers were applied to the dorsum of each hand and foot at 28- and 32-weeks PMA. Simultaneous video recordings of the babies' spontaneous movements were made to assess GMs. SUBJECTS: Eligible babies were born <28-weeks PMA or had a birth weight of <1000 g. Babies were recruited before they reached 33-weeks PMA. OUTCOME MEASURES: GMs assessments were made offline on the video recordings. Forty-six quantitative motor parameters were calculated during the same periods of activity and compared with GMs assessments. RESULTS: At 28-weeks PMA, 24/43 (55.8 %) babies had abnormal GMs. At 32-weeks PMA, 26/57 (45.6 %) had abnormal GMs. The inter-rater reliability of the GMs was poor. When comparing MDS measures between; infants with normal and those with abnormal GMs, at 28-weeks PMA, 7/46 parameters were significantly different, and at 32-weeks PMA, 19/46 parameters were significantly different. CONCLUSION: Isolated use of quantitative movement measures, obtained from four tri-axial accelerometers before hospital discharge, correlate with the GMs assessments at both 28-weeks and 32-weeks PMA. Accelerometers may provide a useful screening tool for abnormal GMs in preterm infants and could overcome issues with inter-rater reliability.