Diagnostic Evaluation of the STEPSS Score and Machine Learning Models in Predicting Outcomes in Pediatric Status Epilepticus: A Prospective Observational Study

Background:
Pediatric status epilepticus (SE) is a neurological emergency associated with significant morbidity. The Status Epilepticus Pediatric Severity Score (STEPSS) offers rapid bedside prognostication, but its performance relative to machine learning (ML) models has not been well studied in resource-limited settings.
Methods:
We prospectively enrolled 100 children with SE (median age 3.4 years; 52 % male) admitted to a tertiary center in South India (2023–2024). Clinical features, investigations, and outcomes were recorded. Functional outcome was assessed at discharge using the Pediatric Overall Performance Category (POPC), with unfavorable outcome defined as POPC ≥3. Prognostic accuracy of STEPSS and three ML models—Logistic Regression (LR), Random Forest (RF), and XGBoost—was evaluated using sensitivity, specificity, predictive values, accuracy, and area under the ROC curve (AUC).
Results:
Overall, 30 % of children had unfavorable POPC outcomes. At presentation, 40 % had altered consciousness and 45 % had high-risk seizure types. STEPSS ≥3 was associated with ICU admission, mechanical ventilation, and poor outcome. STEPSS demonstrated good discrimination (sensitivity 79 %, specificity 78 %, NPV 84 %, AUC 0.83). Additional predictors of poor outcome included low SpO₂, hyperglycemia, CT abnormalities (present in 41 % of those imaged), and delay to first AED EEG was performed in 53 % of patients, with abnormalities in 38 % .Among ML models, LR achieved performance similar to STEPSS (AUC 0.82), while RF (AUC 0.88) and XGBoost (AUC 0.91) outperformed it, with XGBoost achieving the highest accuracy (90 %) and the fewest misclassifications. Feature importance analysis highlighted CT abnormalities, treatment delay, blood glucose, and SpO₂ as dominant predictors, with STEPSS also contributing significantly.
Conclusion:
STEPSS remains a practical and reliable bedside triage tool in pediatric SE, particularly in low-resource emergency settings. However, integrating additional clinical indicators into ensemble ML models, especially XGBoost, provides superior prognostic accuracy. A hybrid strategy—STEPSS combined with ML augmentation—offers both interpretability and precision, supporting early risk stratification and informed critical care decisions.