Resting State Functional MRI Connectivity Association with Consciousness, Mortality, and Two-Year Outcomes in Neonatal Acute Brain Injury
Boerwinkle, V. L.; Sussman, B. L.; Manjon, I.; McGary, A.; Lucia, M.; Broman-Fulks, J.; Hunter, S.; Wyckoff, S.; Allred, K.; Tom, D.
Show abstract
BackgroundAn accurate and comprehensive test of integrated brain network function is needed for neonates during the acute brain injury period to inform on morbidity. In our first term neonatal acute brain injury (ABI) study we demonstrated resting state functional MRI (RS) acquired within 31 days of life, results in disrupted connectivity of the resting state fMRI networks, incrementally associated with consciousness, mortality, cognitive and motor development, and ongoing concern for seizures at 6 months post-gestation. In this retrospective cohort study, we evaluate extended 2-year outcomes in the same patients. MethodsStudy subjects included the same 40 consecutive neonates from our prior study, with resting state functional MRI acquired within 31 days after suspected brain insult from March 2018 to July 2019. Acute-period exam and test results were assigned ordinal scores based on severity as documented by respective treating specialists. Analyses (Fisher exact, Wilcox Sum-Rank test ordinal/multinomial logistic regression) examined association of resting state networks with demographics, presentation, neurological exam, electroencephalogram, anatomical MRI, magnetic resonance spectroscopy, passive task functional MRI, and outcomes of NICU and all mortality, outpatient development measured by exam and the Pediatric Cerebral Performance Category Scale (PCPC), motor development and tone, and ongoing concern for seizure at up to 42 months of age. All statistical tests were 2-sided, with statistical significance and CI adjusted using a Bonferroni correction to account for multiple test comparisons for each network and other modality. ResultsSubjects had a mean (standard deviation) gestational age of 37.8 (2.6) weeks, follow-up median age follow-up median age (interquartile range) 30.5 (23.6, 36.7) months, 68% were male, with a diagnosis of hypoxic ischemic encephalopathy (60%). Of the 40 patients, three died prior to discharge, and another four between 6-42 months, and 5 were lost to follow-up. Of the followed, findings at birth included mild distress (46%), moderately abnormal neurological exam (34%), and consciousness characterized as awake but irritable (37%). Significant associations after multiple testing corrections were detected for resting state networks: basal ganglia with PCPC (odds ratio [OR], 9.54; 99.4% confidence interval [CI], 1.89-48.1; P = 0.0003), NICU mortality (OR, 57.5; 99% CI, 1.35->999; P = 0.006), outpatient mortality (OR, 65.7; 99% CI 1.47->999; P = 0.005), and motor tone/weakness (OR, 17.8; 99% CI, 2.2-143; P = 0.0004); language/frontoparietal network with developmental delay (OR, 3.64; 99% CI, 1.02-13.05; P = 0.009), PCPC (OR, 3.98; 99% CI, 1.09-14.45; P = 0.006), and all mortality (OR, 9.2; 99% CI, 0.91-92.6; P = 0.01; default mode network with developmental delay (OR, 4.14; 99% CI, 1.19-14.43; P = 0.003); PCPC (OR, 4.1; 99% CI, 1.2-14.2; P = 0.004), NICU mortality (OR, 20.41; 99% CI, 0.89-468; P = 0.01), and motor tone/weakness (OR, 3.35; 99% CI, 1.01-11.12; P = 0.009); and seizure onset zone with concern for seizures (OR, 4.02; 99% CI, 1.0-16.15; P = 0.01). Of the other acute phase tests, only anatomical MRI was showed association with and outcome, concern for seizure (OR, 2.40; 99% CI, 0.94-6.13; P = 0.01). ConclusionsThis study provides level 3 evidence (OCEBM Levels of Evidence Working Group) demonstrating that in neonatal acute brain injury, the degree of abnormality of resting state networks is associated with mortality, ongoing concern for seizure and 2 year outcomes. These findings suggest RS is feasible and safe to implement in a busy tertiary neonatal ICU and the findings are of at least equivalent value to other standard of care diagnostics. HighlightsO_LICognition is incrementally associated with the DMN, Lang/FP, and the BG at two years in neonatal ABI. C_LIO_LIMotor outcomes are incrementally associated with the BG and DMN at two years in neonatal ABI. C_LIO_LISeizure outcomes are incrementally associated with rs-SOZ at two years in neonatal ABI. C_LIO_LIMortality after discharge is incrementally associated with the DMN and BG. C_LIO_LICompared to EEG, MRS, and task-fMRI, only anatomical MRI had 2-year association with outcomes, on-going concern for seizure at 2 years in neonatal ABI. C_LIO_LIRS is feasible and safe to implement in a busy tertiary neonatal ICU and the findings are of at least equivalent value to other standard of care diagnostics. C_LI
Matching journals
The top 7 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Brief Apnea and Hypoventilation Reduces Seizure Duration and Shifts Seizure Location for Several Hours in a Model of Severe Traumatic Brain Injury. 94%
- Ablation of Apparent Diffusion Coefficient Hyperintensity Clusters and Mesial Temporal Lobe Epilepsy Improves Seizure Outcomes after Laser Interstitial Thermal Therapy 92%
- Uncovering Spatiotemporal Dynamics of the Corticothalamic Network during Seizures 91%
Similar papers in this journal
- Characterization of Clinical MRI Findings in Moderately-Late Preterm Infants Diagnosed with Cerebral Palsy: A Single Center Retrospective Study 94%
- Use of magnetic resonance imaging in neuroprognostication after pediatric cardiac arrest: Survey of current practices 92%
- Use of Computational Phenotypes for Predicting Genetic Subgroups of Cerebral Palsy 92%
Similar papers in this journal
- Lesioned hemisphere-specific phenotypes of post-stroke fatigue emerge from motor and mood characteristics in chronic stroke 90%
- Thalamohippocampal atrophy in focal epilepsy of unknown cause at the time of diagnosis 90%
- The effectiveness of Cenobamate in patients previously treated with Vagus Nerve Stimulation for drug resistant epilepsy 89%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.