The Predictive Accuracy of the General Movement Assessment for Cerebral Palsy: A Prospective, Observational Study of High-Risk Infants in a Clinical Follow-Up Setting

Abstract

Background: Early prediction of cerebral palsy (CP) using the General Movement Assessment (GMA) during the fidgety movements (FM) period has been recommended as standard of care in high-risk infants. The aim of this study was to determine the accuracy of GMA, alone or in combination with neonatal imaging, in predicting cerebral palsy (CP). Methods: Infants with increased risk of perinatal brain injury were prospectively enrolled from 2009–2014 in this multi-center, observational study. FM were classified by two certified GMA observers blinded to the clinical history. Abnormal GMA was defined as absent or sporadic FM. CP-status was determined by clinicians unaware of GMA results. Results: Of 450 infants enrolled, 405 had scorable video and follow-up data until at least 18–24 months. CP was confirmed in 42 (10.4%) children at mean age 3 years 1 month. Sensitivity, specificity, positive and negative predictive values, and accuracy of absent/sporadic FM for CP were 76.2, 82.4, 33.3, 96.8, and 81.7%, respectively. Only three (8.1%) of 37 infants with sporadic FM developed CP. The highest accuracy (95.3%) was achieved by a combination of absent FM and abnormal neonatal imaging. Conclusion: In infants with a broad range of neonatal risk factors, accuracy of early CP prediction was lower for GMA than previously reported but increased when combined with neonatal imaging. Sporadic FM did not predict CP in this study.

Machine Learning of Infant Spontaneous Movements for the Early Prediction of Cerebral Palsy: A Multi-Site Cohort Study

“…Four hundred and fifty infants admitted to one of five participating level III-IV Neonatal Infant Care Units (NICU) in Norway or the United States were enrolled at discharge from the NICU based on extreme prematurity, neonatal neurologic abnormalities, cardiac surgery or medical complexity. Video recordings during the fidgety movements period were taken according to Prechtl's methodology for observation of general movements, and the GMA results for early CP prediction are presented in a different paper [19]. In the study arm presented here, the aim was to develop a novel machine-learning algorithm for early CP prediction based on the same videos.…”

“…Cerebral imaging (cerebral ultrasound (cUS) and magnetic resonance imaging (MRI)) was carried out for clinical purposes following each hospital's guidelines, and a central classification of the results into normal/mildly abnormal or moderately/severely abnormal was carried out based on the local written reports. Lesions known to be associated with later CP were classified as abnormal, and milder abnormalities not associated with later CP were classified as normal (for details about imaging results, see [19]).…”

“…The ability of the CIMA model to predict CP was assessed by sensitivity, specificity, positive and negative predictive values (PPV and NPV) and area under receive operating characteristic curve (AUC). The performance of the CIMA model was compared with the performance of the observational GMA and cerebral imaging (cUS and MRI) [19] in addition to the automated method previously presented by our group using the variation of the spatial center of motion (C SD ) [12,30]. Kruskal-Wallis's test with the post hoc Wilcoxon rank-sum test including Bonferroni correction assessed the significance of the difference in the proportion of CP risk-related movements between the different FM categories assessed by the observational GMA.…”

“…Sixty-two (18.5%) of 336 infants who did not develop CP had a proportion of CP risk-related movements above 50% (false positive; blue bars above the horizontal line in Figure 3). Table 3 shows the predictive values for the current CIMA model, observational GMA [19] and neuroimaging results [19] and the variation of the spatial center of motion (C SD ). The CIMA model had the best sensitivity, NPV and AUC.…”

“…threshold of 50%. The ROC curve and alternative thresholds are provided in Appendix E and crosstables and mean square contingency coefficients are provided in Appendix F. [19]. PPV = positive predictive value; NPV = negative predictive value; GMA = General Movement Assessment.…”