Predictive value of the Thompson score for short-term adverse outcomes in neonatal encephalopathy

Background To explore the predictive value of the Thompson score during the first 4 days of life for estimating short-term adverse outcomes in neonatal encephalopathy. Methods This observational study evaluated infants with neonatal encephalopathy (≥36 weeks of gestation) registered in a multicenter cohort of cooled infants in Japan. The Thompson score was evaluated at 0–24, 24–48, 48–72, and 72–90 h of age. Adverse outcomes included death, survival with respiratory impairment (requiring tracheostomy), or survival with feeding impairment (requiring gavage feeding) at discharge. Results Of the 632 infants, 21 (3.3%) died, 59 (9.3%) survived with respiratory impairment, and 113 (17.9%) survived with feeding impairment. The Thompson score throughout the first 4 days accurately predicted death, respiratory impairment, or feeding impairment. The 72–90 h score showed the highest accuracy. A cutoff of ≥15 had a sensitivity of 0.85 and specificity of 0.92 for death or respiratory impairment, while a cutoff of ≥14 had a sensitivity of 0.71 and a specificity of 0.92 for death, respiratory or feeding impairment. Conclusion A high Thompson score during the first 4 days of life, especially at 72–90 h could thus be useful for estimating the need for prolonged life support. Impact The Thompson score on days 1–4 of age was useful in predicting death and respiratory or feeding impairments. The 72–90 h Thompson score showed the highest predictive capability. Owing to the rarity of withdrawal of life-sustaining treatment in Japan, 43% of infants with persistent severe encephalopathy with a Thompson score of ≥15 at 72–90 h of age could regain spontaneous breathing, be extubated, and survive without tracheostomy. Meanwhile, approximately 50% of infants who survived without tracheostomy required gavage feeding. Our results could provide useful information for clinical decision making regarding infants with persistent severe encephalopathy.