Presented at the Neonatal Society 2011 Autumn Meeting.
Thayyil S1, Ayer M2, Satheesan K2, Ceebi S2, Guhan B2 on the behalf of Peacock Trial Collaborators
1 University College London, UK
2 Calicut Medical College, India
Background: Therapeutic hypothermia (TH) reduces the brain injury following neonatal encephalopathy measured by magnetic resonance (MR) biomarkers and improves intact survival; however it is not known if TH reduces the brain injury in encephalopathic infants with co-existent perinatal sepsis. Hypoxic ischemic injury in an infection-primed brain (‘dual hit brain injury’) has been shown to produce more severe brain injury than either alone in animal models; it is possible that this injury is less responsive to TH. We examined the effect of TH on brain injury measured by magnetic resonance biomarkers in neonatal unit in south India, which had a high burden infants with encephalopathy and co-existent perinatal sepsis. The study was approved by the Calicut Medical College and UCL ethics committee.
Methods: Within six hours of birth we randomly allocated 33 infants with neonatal encephalopathy to 72 hours whole-body cooling achieved using a mattress of phase changing material or to standard-care, following informed parental consent. All surviving infants had brain MRI during the second postnatal week using a 1.5 Tesla Siemens Avanto scanner (Erlangen, Germany). We obtained T1–weighted 3D fast low angle shot (FLASH) images in the axial plane (512x512x160 pixels; 1mm isotropic resolution; 17ms relaxation time (TR) 6ms echo time (TE), 20º flip angle), 2D fast spin echo (FSE) T2–weighted images (5910ms TR, 110ms TE, 3mm slice thickness, 0.4mm isotropic in-plane resolution) in coronal and sagittal planes, and 21 direction diffusion tensor imaging (DTI; 2800ms TR, 94ms TE, 128 x 128 pixels, 19 contiguous 5mm axial slices, b 0 and 1000s/mm2). The MR images were reported masked to the allocation using a previously defined scoring system.
Results: Clinical characteristics were similar in the cooled (n=17) and standard-care (n=16) groups. Mean (SD) rectal temperatures in the cooled infants were 33.5 (±0.3) ºC and standard-care infants were 36.4 (±0.5) ºC. Four (24%) cooled and two standard-care infants died (13%); the remaining 27 had magnetic resonance imaging. None showed evidence of established brain injury of antenatal onset. No intergroup central white-matter tract based spatial statistics or fractional anisotropy differences were seen.
Table: FA: Fractional anisotropy; PLIC: Posterior limb of internal capsule; ALIC: Anterior limb of internal capsule.
A non significant trend towards higher moderate or severe basal ganglia/thalami injury (46% versus 31%), white-matter injury (62% versus 77%) and cortical injury (46% versus 15%) (p>0.05) was seen in the cooled infants, compared with the infants who received standard care only.
Conclusion: No improvement in the brain injury following TH was seen in this small feasibility study. Possibility of altered responses to hypothermia rescue treatment in encephalopathic infants with co-existent perinatal infection cannot be excluded. Larger studies are required to explore this hypothesis further. Robust qualification of MR biomarkers are also required before these can be used as surrogate outcome measures of neuroprotective therapies; such studies are now in progress (MaRBLE Study).
Corresponding author: s.thayyil@ucl.ac.uk
References
1. Edwards AD et al BMJ 2010;340:c363
2. Rutherford M et al Lancet Neurol 2010;9(1):39-45.
3. MARBLE study– Magnetic Resonance Biomarkers in Neonatal Encephalopathy (NCT01309711)