Cooling Therapy in Neonatal Encephalopathy The unfinished Story Sudhin Thayyil MD, FRCPCH, PhD Head of Academic Neonatology Director, Centre for Perinatal Neuroscience Imperial College London, UK
2 Contents A. Recap of the cooling story so far B. Ongoing work – Magnetic Resonance Biomarkers – Magnetic Resonance Biomarkers – Transcriptomic signatures – HELIX trial
3
4 Healthy Control HIE – Cooled HIE – Normothermic
5 Clinical trials Cool cap Trial: Lancet 2005 (Auckland and Imperial College London) TOBY Trial: NEJM 2009 (Imperial College London) NICHD (USA) Trial: NEJM 2005
6 10 years…and still doesn’t work in all !!! CONTENTS 1. How can we speed up? 2. Why doesn’t it work in all?
7 MR Imaging in neonatal encephalopathy
8 Proton MR spectroscopy normal Naa Lac Lac Lac Cho severe HI Cr Naa
9 Which MR Biomarker Thayyil et al., Pediatrics 2010
MARBLE (Magnetic Resonance Biomarkers in neonatal Encephalopathy) 10 Can MR biomarkers accurately predict neurological outcome 2 years in a multicenter setting? – MR spectroscopy (metabolic injury) – Diffusion MRI (microstructural injury) – Diffusion MRI (microstructural injury) – ‘Conventional’ structural MRI • 223 babies from 8 centres in the UK and USA over 3 years • Bayley III at 2 years of age Lally et al., BMJ Open 2016
MARBLE (Magnetic Resonance Biomarkers in neonatal Encephalopathy) 11
12 Use in future clinical trials Neurodevelopment as NAA: Surrogate endpoint r arm endpoint arm No per arm No per ar Power Power (Mean difference 2.5 mmol/kg, SD 2.) (CER 50%, RRR 20%, alpha 0.05)
Clinical Heterogeneity 13 Maternal illness Co-existent infection and FIRS Growth restriction Encephalopathy Cerebral Palsy Hypoxia- Ischemia Acute versus Sub acute hypoxia
Can transcriptomic signature aid in disease stratification at birth? 14 Number of gene copies (up and down) ‘signature’
Transcriptomic signature 15 NE Babies Controls NE infants total 6305 genes adj pvalue<0.05 Controls
Transcriptomic signature 16
Why do a cooling trial in LMIC? 17 1 million deaths per year from neonatal encephalopathy Lawn et al., Lancet 2005
18 Why do a cooling trial in LMIC? • Clinical heterogeneity – Population differences – Nature and origins of brain injury – Higher perinatal infection & other co-morbidities – Higher perinatal infection & other co-morbidities • Sub optimal intensive care – Public sector versus Private sector issues – Lack of effective transport system
19 Why do a cooling trial in LMIC?
20
21 Phase Change Material for Cooling 54 babies with mild, moderate or severe encephalopathy born at Calicut Medical College, Kerala Thayyil et al., Arch Dis Childhood 2007
22 HELIX feasibility study T ecotherm-Helix Set up I 1. Connect Tecotherm Helix to the electric supply via the power unit 2 2. Connect T ecotherm Helix to the mattress using the hose 4 3 4. Switch on the machine 3. Add pre-mixed coolant till the fluid level is visible (first use only) 6 5 6. Keep a thin cloth over the mattress 5. Add more pre-mixed coolant holding the mattress vertical so that the air bubbles escape and mattress is fully filled with fluid 7 Keep baby on the mattress, Attach the rectal probe to tecotherm helix (red arrow) Insert the rectal probe (4 to 6 cm) (blue arrow) Wrap the mattress around the baby using a small tape (non-sticking) through the mattress eye hole (black arrow) (green arrow) Oliveira et al., BMJ Ped Open 2018
HELIX feasibility study 23 • To examine the feasibility of whole body cooling using a servo controlled cooling device (Tecotherm HELIX) and short term neonatal morbidity • • 82 babies with moderate or severe encephalopathy recruited from 4 82 babies with moderate or severe encephalopathy recruited from 4 centres in India • 22% babies died Oliveira et al., BMJ Ped Open 2018
24 HELIX feasibility study Survivors Non-survivors P-value (n=67) (n=15) Gestational age (weeks) 38.8 ± 1.3 38.0 ± 1.2 0.14 Age at start of cooling (hours) 3.6 ± 1.4 4.1 ± 1.4 0.22 Birth weight (g) 2912 ± 364 2855 ± 388 0.58 10 min Apgar 5.5 ± 0.8 5.0 ± 1.0 0.36 Severe encephalopathy 11/67 (16%) 10/15 (67%) <0.001 55/64 (86%) 13/14 (93%) 0.68 Seizures within 6 h of birth Place of delivery Inborn Inborn 15/65 (23%) 15/65 (23%) 4/13 (31%) 4/13 (31%) Outborn – hospital 48/65 (74%) 9/13 (69%) 0.81 Outborn – home 2/65 (3%) 0/13 (0%) Neonatal complications Gastric bleeds 14/38 (37%) 11/11 (100%) <0.001 Pulmonary haemorrhage 1/38 (3%) 5/11 (45%) 0.001 PPHN 1/38 (3%) 6/11 (55%) <0.001 Thrombocytopenia 42/67 (63%) 10/15 (67%) 1.00 26/67 (39%) 9/15 (60%) 0.16 Metabolic acidosis Treatment duration (days) Invasive ventilation 1.4 ± 1.8 3.3 ± 1.5 0.002 Sedation 0.9 ± 1.5 1.4 ± 1.8 0.40 Inotropic support 3.1 ± 1.3 3.9 ± 0.3 0.04 Anticonvulsive drugs 2.9 ± 1.7 3.9 ± 0.3 0.07 Oliveira et al., BMJ Ped Open 2018
25 HELIX feasibility - Heart rate and Mortality Oliveira et al., BMJ Ped Open 2018
26 HELIX – Main Trial Primary • To examine whether whole body cooling to 33.5˚ C initiated within 6 hours of birth and continued for 72 hours reduces death or neurodisability at 18 to 22 months after neonatal encephalopathy in a ‘real life setting’ in low and middle-income countries. ‘real life setting’ in low and middle-income countries. Secondary • Proton MR spectroscopy Thalamic N-acetyl aspartate level and white matter fractional anisotropy at 1 week • Host gene expression profile • Perinatal infection Thayyil et al., BMJ Open 2017
27 Study design and Centers • 408 babies with moderate or severe encephalopathy – Sion – Mumbai (J Mondkar) – IGH Bangalore (N Benkappa) – ICH, Madras Medical College (Mangalabharathi) – IOG, Madras Medical College (Aresar) – SAT, Trivandrum (Radhika) – MAMC, New Delhi (Ramji) – Kelaniya, Sri Lanka (Rodrigo) – BSMMU, Dhaka (Shahidullah) • Babies randomized (web-based) to whole body cooling (Tecotherm) for 3 days or usual care • 3T MR imaging and spectroscopy within 2 weeks • Detailed infection screening and transcriptomics • Bayley III at 18 to 22 months Thayyil et al., BMJ Open 2017
28 HELIX so far • 288 babies recruited since Aug 2015 (12 to 15 babies per month) • Expected to complete recruitment (120 more) by October 2018 • All recruited cases surviving till 2 weeks had MRI and MRS • • Follow up is challenging Follow up is challenging – Frequent changes in mobile numbers – Migrant families and slum population – Despite this, all except one family in contact – Often the team visited homes, slums, had to fly to different states – Second (external team) for parents upset with recruiting centers • Highly committed team and MAJOR team effort
29 Survey of Cooling in Indian Neonatal Units Hospitals participated in the survey Public Private Total (n=93) (n=25) (n=68) Offers cooling as standard care Offers cooling as standard care 10 (21%) 10 (21%) 37 (79%) 37 (79%) 47 (51%) 47 (51%) Do not offer due to lack of cooling 13 (52% 28 (41%) 41 (44%) devices and trained staff Do not offer as strong evidence from 2 (8%) 3 (4%) 5 (5%) clinical trials are lacking Chandrasekeran, Swamy et al., Indian Pediatrics 2018
Survey of Cooling in Indian Neonatal Units 30 Hospitals offering therapeutic hypothermia Public (n=10) Private (n=37) Total (n=47) Cooling devices Approved cooling devices* 9 (90%) 13 (35%) 22 (47%) Indigenous cooling solutions** 1 (10%) 24 (65%) 25 (53%) Initiation of cooling Initiation of cooling Within 6 hours after birth 10 (100%) 28 (76%) 38 (81%) 0 (0%) 7 (19%) 7 (15%) Upto 12 hours after birth Upto 24 hours after birth 0 (0%) 2 (5%) 2 (4%) Sedation during cooling Intravenous sedation 6 (60%) 20 (54%) 26 (55%) Oral sedation only 2 (20%) 2 (5%) 4 (9%) 2 (20%) 15 (41%) 17 (36%) None Chandrasekeran, Swamy et al., Indian Pediatrics 2018
31 Take home messages • Cooling (33 to 34 C) within 6 h of age for 72 h improves survival without disability after moderate or severe encephalopathy in high income countries • • Safety and efficacy of cooling in LMIC is not known, and hence this Safety and efficacy of cooling in LMIC is not known, and hence this therapy should be considered experiential, until the HELIX trial data is available. • MR spectroscopy biomarkers and transcriptomic profile will help speed up the clinical translation of novel therapies and personalized neuroprotection
Thankyou 32
Recommend
More recommend
