Contact us / Members' area    

Home > Abstracts > Index > 2018 > Lally PJ, et al


The prognostic accuracy of magnetic resonance biomarkers in neonatal encephalopathy (MARBLE): a prospective multi-centre study

Presented at the Neonatal Society 2018 Spring Meeting (programme).

Lally PJ1, Montaldo P1, Oliveira V1, Soe A2, Atreja G1, Mendoza J1, Kariholu U1, Bassett P3, Harigopal S4, Clarke P5, Satodia P6, Abernethy L7, Bainbridge A8, Shankaran S9, Thayyil S1 for the MARBLE consortium

1 Imperial College London, UK
2 Medway NHS Foundation Trust, UK
3 Stats Consultancy, RVI, UK
4 Newcastle, UK
5 Norwich Hospital, UK
6 UHCW NHS Trust, UK
7 Liverpool Hospital, UK
8 University College Hospital, UK
9 Wayne State University, USA

Background: In neuroprotection trials for neonatal encephalopathy (NE), the clinical manifestations of therapeutic effects can only be reliably assessed several years after an intervention. We examined the prognostic accuracy of early cerebral MR biomarkers in predicting 2-year neurodevelopmental abnormalities following NE treated with hypothermia.

Methods: Consecutive term and near-term babies with neonatal encephalopathy who had therapeutic hypothermia were recruited from eight sites across the UK and USA over 3 years. The MR protocol was conducted on harmonised 3 Tesla scanners, comprising structural T1- and T2-weighted MRI, diffusion weighted MRI, and MR spectroscopy. MR spectroscopy included measurements of absolute metabolite concentrations. An abnormal outcome was defined as death or moderate or severe disability as measured by performance scores in Bayley-III and Gross Motor Function Classification System examinations, and/or the presence of hearing loss, blindness, or persistent seizures.

Results: A total of 223 infants were recruited and underwent MR imaging and spectroscopy at a median (IQR) age of 7 (5) days, with 189 (85%) followed up for a neurological examination at a median (IQR) age of 23 (4) months. Of those who were followed up, 31 (16%) had an abnormal outcome. The thalamic concentration of Nacetylasparate ([NAA]) predicted adverse neurodevelopment with an area under the receiver operating characteristic curve (AUC) of 0.99 (95% CI 0.97-1.00, n=82), sensitivity of 100% (95% CI 74-100) and specificity of 97% (95% CI 90-100).

The ratio of thalamic lactate to NAA had an AUC of 0.97 (95% CI 0.93-1.00, n=160). The mean fractional anisotropy in the posterior limbs of the internal capsules (PLIC) had an AUC of 0.92 (95% CI 0.76-1.00, n=65). The thalamic injury score assessed from T1- and T2-weighted imaging had an AUC of 0.85 (95% CI 0.76-0.95, n=189) (1a). Thalamic [NAA] was the only MR spectroscopy biomarker independently associated with neurodevelopmental outcomes on multivariable analysis (1b).

Conclusion: Cerebral magnetic resonance biomarkers acquired soon after birth, particularly those derived from MR spectroscopy, accurately predict neurodevelopment two years after neonatal encephalopathy. In addition to their clinical utility, these measures could be applied to increase the power of neuroprotection trials, while reducing their duration.

Corresponding author:

Home / The Neonatal Society, founded 1959 / 2001 - 2019