185 - Can Total Sarnat Score Predict Outcome in Moderate or Severe Hypoxic-Ischemic Encephalopathy (HIE)?

Poster Number: 185
Publication Number: 185.137

Girija Natarajan, Children's Hospital of Michigan, Detroit, MI, United States; Scott A. McDonald, RTI International, Raleigh, NC, United States; Seetha Shankaran, Wayne State University School of Medicine, Austin (AUS), TX, United States; Abhik Das, RTI International, Rockville, MD, United States; Abbot Laptook, The Warren Alpert Medical School of Brown University, Providence, RI, United States; Sonia Bonifacio, Stanford University School of Medicine, Palo Alto, CA, United States; Elizabeth K. Sewell, Emory University; Children's Healthcare of Atlanta, Atlanta, GA, United States; Lina Chalak, UT Southwestern.edu, DALLAS, TX, United States; NICHD Neonatal Research Network, Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, MD, United States

Background: The NICHD Induced Hypothermia (IH) and Optimizing Cooling (OC) randomized controlled trials (RCTs) used the modified Sarnat staging at ≤6 hours of age to categorize HIE as moderate or severe, groups with disparate outcomes. The predictive accuracy of the Total Sarnat Score (TSS) generated by adding numerical scores in each category (2 for moderate and 3 for severe abnormality) in moderate/severe HIE has not been validated in large RCTs.

Objective: To examine the association of TSS, modified Sarnat staging and each of 6 components of the neurologic examination with death/disability at 18 months of age among infants with moderate/severe HIE.

Design/Methods: This analysis included participants (n=572) of the IH [Therapeutic hypothermia (TH) vs. normothermia] and OC (randomized to 4 TH groups: 33.5℃-72 and 120 hours, 32℃-72 and 120 hours) RCTs. Infants with missing/incomplete neurologic exam (n=11), missing primary outcome (n=19) or RCT entry by seizures alone (n=14) were excluded. The primary outcome was death or moderate/severe disability, as defined in the trials (using Bayley-III for OC and Bayley-II for IH). Statistical analysis for TSS, TSS tertiles (6-10; 11-14; 15-18) and modified Sarnat staging included logistic and linear regression, adjusting for center, trial, and temperature (normothermia, standard TH, longer/deeper cooling). Forward-selection was used for the individual exam components, with p< 0.2 for model entry.

Results: The cohort (n=528) included 374 (71%) moderate and 154 (29%) severe HIE; 37% were IH and 63% were OC participants. There were significant differences between moderate and severe HIE groups in clinical characteristics [Table1]. IH participants comprised 33% of moderate HIE vs. 46% of severe HIE; a third of the cohort received standard TH. Death/disability at 18 months of age occurred in 26% of moderate HIE and 71% of severe HIE. The median (IQR) TSS of infants with moderate and severe HIE were 11 (9-13) and 16 (16-17) respectively. Figure 1 shows the modified Sarnat staging (A) and 18-month outcomes (B) at each TSS. Associations between TSS, TSS tertiles and death/disability and Bayley-III scores were all significant and AUCs were comparable to associations between modified Sarnat staging and outcomes [Table 2]. All 6 neurologic exam components resulted in significant R2 and AUC for death/disability; tone did not add to the prediction with the other 5 components present.

Conclusion(s): TSS and modified Sarnat staging in the first 6 hours after birth were comparable and significantly predictive of 18-month outcome in moderate/severe HIE.