Postoperative opioid-induced respiratory depression monitoring in the context of supplemental oxygen: current limitations and emerging machine-learning approaches

Opioid-induced respiratory depression (OIRD) may develop after extubation and remain unrecognized during intermittent postoperative observation. Supplemental oxygen is commonly used to prevent or treat postoperative hypoxemia, but it may preserve peripheral oxygen saturation despite impaired ventilation and thereby delay oximetric warning. Experimental evidence suggests that hyperoxia may also reduce peripheral chemoreceptor drive, although the clinical importance of this effect during routine postoperative low-flow oxygen therapy remains uncertain. This narrative Mini Review examines how supplemental oxygen affects the detection and interpretation of postoperative OIRD, evaluates current respiratory monitoring and machine-learning evidence, and considers requirements for clinical translation. Continuous oximetry and capnography detect more respiratory abnormalities than intermittent assessment, but monitor-defined episodes do not necessarily represent clinically important or opioid-attributable respiratory depression. Their interpretation requires consideration of technical signal validity, clinical significance, and opioid contribution. Existing machine-learning studies have predicted threshold-defined respiratory events, classified individual breaths, distinguished normal from abnormal monitoring segments, or quantified ataxic breathing, supporting the technical feasibility of automated signal interpretation and respiratory phenotype discovery. Among the studies reviewed, none used a prospectively adjudicated clinical OIRD outcome or explicitly modeled oxygen delivery as a time-varying exposure. A testable development pathway would integrate respiratory time series with oxygen delivery, medication administration, and changing clinical context, followed by external validation across devices, institutions, and healthcare settings. Clinical value should ultimately be judged by earlier appropriate intervention and reduced respiratory harm without an unacceptable alarm burden, rather than by improved prediction of threshold-defined episodes alone.