Abstract
Background
Commonly used methods of CTG classification do not reliably predict neonatal hypoxic-ischemic encephalopathy (HIE).
Objective
To examine whether a relationship exists between the types of hypoxia as identified on the cardiotocograph using novel physiology-based CTG classification and patterns of injury on neonatal cerebral MRI and later neurodevelopmental outcomes.
Study design
A retrospective study of term-born infants admitted to four neonatal units with HIE as part of a brain injury biomarkers study between January 2014 and December 2015. Intrapartum CTG traces were analyzed by two obstetricians trained in physiological CTG classification, blind to neonatal outcomes. Neonatal cerebral MR images were assessed independently by a neuroradiologist and an expert neonatologist. CTG traces were classified into types of hypoxia and allocated to groups; (1) chronic hypoxia or antepartum injury; (2) gradually evolving or subacute hypoxia; and (3) acute hypoxia.
Results
Of 106 infants recruited to the study, records were available for 58 cases. Of these, CTGs were available for 37. All 37 had abnormal CTGs. Twenty-four infants, all of whom had received therapeutic hypothermia had cerebral MRI. Fourteen of the 24 (58%) infants had abnormal MRI. In group 1 (chronic hypoxia/antenatal injury), total brain injury was most predominant (4/6 infants). Group 2 (gradually evolving/subacute hypoxia) was associated with peripheral brain injury (5/5 infants). Group 3 (acute hypoxia) was associated with basal-ganglia thalamic injury pattern (3/3 infants). Later neurodevelopmental outcomes were available for 35 cases. Infants suspected to have a pre-labor injury on CTG (group 1) had a higher proportion of adverse neurodevelopmental outcomes (4/10, 40%) compared to groups 2 and 3 (4/25, 16%).
Conclusion
Using this novel physiology-based CTG classification, we demonstrate an association between types of hypoxia observed on the CTG and MRI patterns of hypoxic brain injury. Infants with CTG trace suggestive of chronic hypoxia or other antenatal injuries were overrepresented in this cohort and were also more likely to have a poor neurodevelopmental outcome.
Acknowledgments
We would like to give thanks to the following people who kindly helped with data collection; Louise Wheeler, Kingston Hospital, Zoe Lipscombe, Southend University Hospital, Donna Southam, Basildon University Hospital, Lousia Griffiths, Homerton University Hospital, Stephanie Grigsby, Poole Hospital, and Dr. Caroline Everdeen, Royal Surrey County Hospital. A further thanks to Deborah Mckenna, Southend University Hospital, Lisa Relton, Poole Hospital, and Kim Reed, Ekundayo Badmus and Rohit Syed, Barts Health NHS Trust for helping with data acquisition. Finally, we are especially grateful for the families of these infants who participated in the study is probably one of the most difficult times of their lives.
Disclosure statement
The authors report no conflict of interest.