201 Assessment of Gestational Age in Very Preterm Neonates Using Cerebellar Measurements at Cranial Ultrasound - What is the Best Approach?

Abstract

Background and aims: Clinical assessment of gestational age (GA) can be challenging. Several ultrasound approaches to estimate GA using cerebellar measurements have been reported, claiming to be simpler and more accurate than clinical assessment. We compare the accuracy of 3 previously described cerebellar measurements for estimating the GA in VLBW infants. Methods: We studied prospectively VLBW infants under 32 weeks GA defined by certain dates, IVF date or early fetal ultrasound. We excluded infants with IUGR, neurological problems or scan abnormality except isolated GLH/transient flares. Ultrasound acquisition and cerebellar measurements were made by author AG. Measurements of the vermis anterior-posterior diameter (APD, Cuddihy 1999), transverse cerebellar diameter via anterior (TCDa, Makhoul 2000) and mastoid fontanelles (TCDm, Davies 2001) were obtained. Estimated GA was calculated using published equations, and compared to known GA using intraclass correlation coefficient (ICC). Results: We studied 60 infants. Mean birth GA was 28.4 weeks and mean postnatal age at scan was 1 week. ICC was 0.801 for APD, 0.123 for TCDa and 0.802 for TCDm (p< 0.001). Best estimates of GA were obtained with TCDm (difference SD 1 week) and APD (difference SD 1.3 weeks). Conclusions: APD and TCD gave good estimates of GA and can be recommended. We obtained similar measurements of TCD via the two approaches, but only the equation for the mastoid fontanelle gave a good estimate of GA with excellent ICC. Studies assessing the use of cerebellar measurements for estimating GA in growth restricted, neurologically abnormal and older infants are needed.