Comparison of Modified Sarnat Staging and Thompson Score in Neonatal Hypoxic-Ischemic Encephalopathy

TL;DR

This study compares two scoring systems for assessing brain injury in newborns due to lack of oxygen, finding they agree well over time but differ in speed and early classifications.

Contribution

The study provides a detailed comparison of Modified Sarnat Staging and Thompson Score in neonatal HIE, including time-efficiency and experience-related improvements.

Findings

Thompson Score is consistently faster than Modified Sarnat Staging by about 55 seconds at all timepoints.

Agreement between the two systems improves over time, with high concordance in severe cases from the start.

Early discordance occurs mainly in mild-moderate classifications, suggesting the need for serial assessments.

Abstract

Background

Neonatal hypoxic-ischemic encephalopathy (HIE) is a major cause of neonatal mortality and long-term neurodevelopmental morbidity. Accurate bedside neurological assessment is critical for severity stratification and timely clinical decision-making, particularly during the early postnatal period when encephalopathy evolves dynamically. This study aimed to evaluate the correlation, agreement, and time-efficiency between the Modified Sarnat Staging (MSS) system and the Thompson Score (TS) in neonates with HIE, to determine the institutional burden of HIE, and to assess experience-related efficiency characteristics associated with repeated application of both scoring systems.

Methodology

A cross-sectional observational study was conducted over an 18-month period in the neonatal intensive care unit (NICU) of Pimpri Chinchwad Municipal Corporation Postgraduate Institute & Yashwantrao Chavan Memorial Hospital, Pimpri, Pune, India. A total of 55 neonates with gestational age ≥35 weeks were enrolled based on the American College of Obstetricians and Gynecologists diagnostic criteria for HIE. The MSS system and the TS were performed at 1, 3, 6, 9, 12, and 24 hours of life, and again at discharge. The time required to complete each assessment was recorded. Severity distribution, correlation (Spearman’s rho), agreement (unweighted and quadratic-weighted kappa), heatmap visualization, and experience-related efficiency trends were analyzed using an available-case methodology.

Results

The prevalence of HIE among neonates admitted to the NICU during the 18-month study period was 3.26% (227 of 6,965 deliveries). MSS classified 9.09% of neonates as mild, 56.36% as moderate, and 34.55% as severe HIE, whereas the TS identified 40% as mild, 18.18% as moderate, and 41.81% as severe cases. Correlation between the two systems was strong across all timepoints (Spearman’s rho = 0.76-0.91). Agreement improved progressively, with quadratic-weighted kappa increasing from 0.647 at 1 hour to 0.789 at 12 hours and 0.877 at discharge. Heatmap analyses demonstrated early discordance primarily involving infants classified as mild by the TS but moderate by MSS, while severe classifications showed high concordance from the outset. The TS was consistently faster by approximately 55 seconds at all timepoints (Wilcoxon p < 10⁻⁹). Both scoring systems demonstrated marked experience-related improvements (rho = approximately -0.98 to -0.999).

Conclusions

The MSS system and the TS demonstrate strong correlation and progressively improving agreement in the assessment of neonatal HIE. The TS offers a consistent time-efficiency advantage, while early discordance at the mild-moderate boundary underscores the importance of serial neurological assessments and complementary use of both scoring systems during early clinical decision-making.

Figures