Diagnostic performance of lung ultrasound and procalcitonin in differentiating early neonatal pneumonia from transient tachypnea of the newborn

BACKGROUND: Early neonatal pneumonia and transient tachypnea of the newborn (TTN) often present with overlapping clinical features in late preterm and term neonates, making early etiological differentiation challenging. OBJECTIVE: To evaluate and compare the diagnostic performance of lung ultrasound alone and in combination with serum procalcitonin for distinguishing early neonatal pneumonia from TTN in neonates born at or beyond 34 weeks' gestation. METHODS: This retrospective observational study included neonates ≥ 34 weeks' gestation admitted with respiratory distress within the first 72 h of life between January 2024 and January 2026. Lung ultrasound (LUS), chest radiography, and serum procalcitonin (PCT) measured at 24 h of life were available for all included neonates. The reference diagnosis was established based on clinical findings, chest radiography, and clinical course. Diagnostic performance of LUS, PCT, and their combination was evaluated using receiver operating characteristic analysis; PCT was primarily analyzed as a continuous variable, and an exploratory post hoc cutoff (≥ 6 ng/mL) was used for the combined model. RESULTS: A total of 78 neonates were included (31 with TTN and 47 with early neonatal pneumonia). Lung ultrasound demonstrated high diagnostic accuracy, with a sensitivity of 91.5%, specificity of 96.8%, and an area under the curve (AUC) of 0.945 (95% CI, 0.894-0.992). Procalcitonin alone showed limited diagnostic performance (sensitivity 53.2%, specificity 67.7%, AUC 0.615; 95% CI, 0.496-0.741). The combined LUS + PCT model yielded a numerically higher AUC of 0.964 (95% CI, 0.926-0.994), with a sensitivity of 93.6% and specificity of 96.8%. However, given the small absolute difference versus LUS alone (ΔAUC = 0.019), this increase should be interpreted cautiously. CONCLUSIONS: In late preterm and term neonates with early respiratory distress, lung ultrasound showed high diagnostic accuracy for identifying early neonatal pneumonia. Procalcitonin alone showed limited discriminatory performance, and its addition to lung ultrasound did not provide robust evidence of a statistically reliable improvement over lung ultrasound alone in this retrospective cohort.