Diagnostic accuracy of the Hammersmith Neonatal Neurological Examination in predicting motor outcome at 12 months for infants born very preterm.

The study followed infants only to 12 months corrected age, and the authors note that motor findings at this age can change by 2 years. The Bayley-III may underestimate motor deficits because it mostly checks motor skill attainment rather than quality of movement. Not all infants completed every assessment: 119 had the early HNNE, 108 had the term-equivalent HNNE, and 104 had the 12-month motor outcome. Infants without a 12-month assessment differed medically, being more likely to have higher birthweight, larger birth head circumference, and fewer days on continuous positive airway pressure. The early reflexes score was adjusted by 0.905 for a missing placing item, although sensitivity analyses with other values changed little. The sample was from one hospital and included only English-speaking families within 200 km, so results may not apply to other settings. The study could not analyse subgroups by postmenstrual age at assessment because of sample size. The same assessor performed the term-equivalent exam, which could introduce bias even though the assessor was blinded to history.

The study was supported by the Cerebral Palsy Alliance Research Foundation, the Financial Markets Foundation for Children, and the National Health and Medical Research Council. No author disclosures are reported in the supplied text.

Do not use the HNNE as a diagnosis or as a reason to stop follow-up. The exam had only moderate predictive validity, and the paper states it should identify infants requiring close monitoring rather than diagnose.