Association Between Hypoxic-Ischemic Brain Injury on Early Head CT Imaging and Neurologic Outcome After Pediatric Cardiac Arrest
Abstract Body: Background: Early quantification of hypoxic-ischemic brain injury (HIBI) after pediatric cardiac arrest may aid in risk stratification, determining candidacy for neuroprotective therapies, and neuroprognostication. However, the relationship between HIBI severity on head computed tomography (CT) scans and outcome remains unclear.
Objective: To determine whether HIBI on early CT scans is associated with outcome after pediatric OHCA.
Methods: Retrospective study of patients £18 years old who had a clinical CT scan performed within 24 hours of OHCA. CT density in Hounsfield units (HUs) was manually measured from 23 regions of interest (ROI) including right and left frontal, parietal, temporal, occipital, and cerebellar cortex, centrum semiovale, caudate, putamen, thalamus, internal capsule, cerebellum white matter, and brainstem. Primary outcome was unfavorable outcome defined as Pediatric Cerebral Performance Category (PCPC) 4-6 at discharge with >=1 change from baseline. Death (PCPC 6) was a secondary outcome. We evaluated the association between CT ROIs and outcomes using multivariable best subset logistic regression. A second model included clinical factors including age, time from ROSC to CT, CPR duration, witnessed arrest, and initial post-arrest lactate.
Results: 131 patients were analyzed (median age 3.7 [IQR 0.9-12.2] years, 62% male, 76% with baseline PCPC=1). 54% (71/131) had unfavorable outcome and 43% (56/131) died. Among patients who died, 54% (30/56) were declared brain dead and 34% (19/56) had withdrawal of life-sustaining technology for poor neurologic prognosis. Patients with unfavorable compared to favorable outcomes more often had witnessed arrests (37% vs 63%, p=0.003), had longer CPR duration (30[15-55] vs 5[2-10] mins, p<0.001), and had higher initial post-arrest lactate (7.5[4.9,11.7] vs 2.5[1.8-4.2] mmol/L, p<0.001). Lower HUs in left putamen and higher HUs in right centrum semiovale were independently associated with increased odds of unfavorable outcome (aOR 1.50[95%CI 1.23,1.82] and 1.68[1.34,2.10], respectively, model AUROC 0.78). Similar results were observed for death (Table 1). After controlling for clinical variables, the same ROIs were associated with unfavorable outcome (putamen aOR 1.72[1.22,2.44] and centrum semiovale 1.77 [1.22,2.56], model AUROC 0.90).
Conclusions: CT density measurements in the putamen and centrum semiovale within 24 hours of pediatric OHCA were independently associated with unfavorable outcome and death.
Donoghue, Emily
( Children's Hospital of Philadelphia
, Bala Cynwyd
, Pennsylvania
, United States
)
Kirschen, Matthew
( Children's Hospital of Philadelphia
, Bala Cynwyd
, Pennsylvania
, United States
)
Vossough, Arastoo
( Children's Hospital of Philadelphia
, Philadelphia
, Pennsylvania
, United States
)
Manteghinejad, Amirreza
( Children's Hospital of Philadelphia
, Bala Cynwyd
, Pennsylvania
, United States
)
Graham, Kathryn
( Children's Hospital of Philadelphia
, Bala Cynwyd
, Pennsylvania
, United States
)
Morgan, Ryan
( Childrens Hospital of Philadelphia
, Philadelphia
, Pennsylvania
, United States
)
Nadkarni, Vinay
( Children's Hospital of Philadelphia
, Bala Cynwyd
, Pennsylvania
, United States
)
Diaz-arrastia, Ramon
( University of Pennsylvania
, Philadelphia
, Pennsylvania
, United States
)
Berg, Robert
( CHILDRENS HOSPITAL OF PHILADELPHIA
, Philadelphia
, Pennsylvania
, United States
)
Topjian, Alexis
( CHILDRENS HOSPITAL PHILADELPHIA
, Wynnewood
, Pennsylvania
, United States
)
Author Disclosures:
Emily Donoghue:DO NOT have relevant financial relationships
| Matthew Kirschen:DO NOT have relevant financial relationships
| Arastoo Vossough:No Answer
| Amirreza Manteghinejad:DO NOT have relevant financial relationships
| Kathryn Graham:No Answer
| Ryan Morgan:DO NOT have relevant financial relationships
| Vinay Nadkarni:DO have relevant financial relationships
;
Research Funding (PI or named investigator):NIH/DOD/AHRQ:Active (exists now)
; Research Funding (PI or named investigator):Laerdal Foundation:Active (exists now)
; Research Funding (PI or named investigator):Zoll Medical:Active (exists now)
| Ramon Diaz-Arrastia:No Answer
| Robert Berg:No Answer
| Alexis Topjian:DO have relevant financial relationships
;
Researcher:NIH:Active (exists now)
; Other (please indicate in the box next to the company name):Elsevier editoral board:Active (exists now)