Personalization of caffeine therapy for apnea of prematurity : a potential role for sensor technologies?

Tools

Kolukisa Birgec, Burcu and Toprak, Beyza and Mullen, Alexander Balfour (2026) Personalization of caffeine therapy for apnea of prematurity : a potential role for sensor technologies? Sensors, 26 (12). 3962. ISSN 1424-8220 (https://doi.org/10.3390/s26123962)

[thumbnail of Birgec-etal-Sensors-2026-Personalization-of-caffeine-therapy-for-apnea-of-prematurity]
Preview
 Text. Filename: Birgec-etal-Sensors-2026-Personalization-of-caffeine-therapy-for-apnea-of-prematurity.pdf
Final Published Version
License: Creative Commons Attribution 4.0 logo
Download (924kB)| Preview

Abstract

Apnea of prematurity (AOP) remains a critical challenge in neonatal care, with caffeine citrate serving as the cornerstone of pharmacological intervention. However, the current standardized dosing schedule fails to account for significant inter-individual variability in caffeine pharmacokinetics and clinical response. This narrative review explores the transformative potential of integrating wearable sensor technologies and multi-modal data analytics into a closed-loop framework for personalized caffeine therapy. Based on a synthesis of current monitoring literature, we propose a theoretical, comprehensive monitoring system utilizing the area under the respiratory curve (rAUC) as a continuous proxy metric, alongside waveform amplitude analysis aligned with pediatric polysomnography standards. By incorporating emerging metrics such as respiratory rate variability (RRV) and hypoxic burden, the framework enables the objective quantification of respiratory stability. Furthermore, the integration of established neonatal intensive care unit (NICU) parameters for bradycardia and oxygen saturation detection provides a critical cross-validation layer to minimize artifact-induced false alarms. This conceptual model bridges the gap between advanced signal processing and clinical oversight, offering a scalable pathway toward precision dosing. By shifting from reactive to predictive neonatology, sensor-driven optimization can enhance therapeutic efficacy, reduce alarm fatigue, and ultimately improve developmental outcomes for preterm infants.

ORCID iDs

Kolukisa Birgec, Burcu, Toprak, Beyza and Mullen, Alexander Balfour ORCID logoORCID: https://orcid.org/0000-0001-7475-5543;
CORE (COnnecting REpositories)