Does a Split-Week Gestational Age Model Provide Valuable Information on Neurodevelopmental Outcomes in Extremely Preterm Infants?

Our primary objective for this follow-up study was to compare the neurodevelopmental outcomes of a surviving cohort of infants using a split-week gestational model (early versus late) gestational age (GA) and the standard completed GA categorization. Neurodevelopmental outcomes using a split-week GA model defined as early (X, 0-3) and late (X, 4-6), with X being 23-26 weeks GA, were compared to outcomes using completed weeks GA. In total, 1012 infants were included in the study. Statistically significant differences were noted in outcomes between the early and late split of the gestational week at 23 weeks (early vs. late), with 13.3% vs. 54.5% for no neurodevelopmental impairment, and 53.3% vs. 22.7% for significant impairment (p = 0.034), respectively. There were no differences seen in the split week model for 24, 25, and 26 weeks. A trend towards improved neurodevelopmental outcomes was seen with each increasing gestation week. The split-week model did not provide additional information for pregnancies and infants between 24 and 26 weeks gestation. It did, however, provide information for counsel for infants at 23 weeks gestation, showing benefits in the late versus early half of the week.

Split-week gestational age model provides valuable information on outcomes in extremely preterm infants.

AIM: To compare composite outcomes of neonatal mortality or morbidity using a split-week gestational age (GA) model to completed weeks GA maturity at 23-26 weeks gestation. METHODS: This was a retrospective cohort study of infants born at 23-26 weeks GA. Outcomes using a split-week GA model defined as early (X, 0-3) and late (X, 4-6) with X being 23-26 weeks GA were compared to outcomes using completed weeks GA, with a similar comparison between the late split of the preceding week (X, 4-6) and early split of the subsequent week (X + 1, 0-3). RESULTS: A total of 1345 infants were included in the study. Statistically significant differences were noted in outcomes between the early and late split of the gestational week at 24 (early vs late, 85.6% vs 73.0%), 25 (69.6% vs 56.6%) and 26 weeks (55.9% vs 37.4%), but not at 23 weeks GA (95.2% vs 94.5%). No statistically significant differences were noted between the late vs early part of the subsequent week (23, 4-6) vs (24, 0-3), and (24, 4-6) vs (25, 0-3) GA. CONCLUSION: Neonatal outcome estimates using a split week model differs from that based on the use of completed weeks of gestational maturity.

Prediction of long-term neurodevelopmental outcome in preterm infants using trajectories of general movement assessments.

OBJECTIVE: To assess the predictive value of trajectories and individual assessment of quality of general movements (AQGM) for identification of neurodevelopmental impairment (NDI) at 18-24 months corrected age (CA) in infants <30 weeks gestational age and/or birth weight <1500 g. METHODS: In this retrospective cohort study, AQGM at 6 weeks and 3 months CA were scored and categorized as normal (N) or abnormal (A). AQGM measures were compared with degree of NDI and Bayley Scales of Infant Development, Third Edition (BSID-III) composite motor and cognitive scores. 'Persistently abnormal' AQGM included both mildly abnormal (MA) and definitely abnormal (DA) assessments. A "modified AQGM" where MA assessments were considered normal variant/transient injury was used to conduct post-hoc analysis. RESULTS: Across 244 cases, persistently abnormal AQGM trajectory predicted the level of NDI (OR 2.5, 95% CI 1.2, 5.1) compared to AQGM trajectory that normalized. However, using the "modified AQGM", persistently DA trajectories were associated with significantly lower BSID-III composite motor and cognitive scores (p < 0.001 and p = 0.039, respectively). CONCLUSION: Categorizing MA assessments as transient injury increased the predictive value of AQGM trajectories and significantly predicted lower cognitive and motor scores at 18-24 months CA.