Detection of hypoxaemia in preterm infants: evaluation of a new technology pulse oximeter

Abstract 145 Aims: Pulse oximeters are increasingly used to detect hypoxaemia in preterm infants; however, they are traditionally very prone to movement artifact, resulting in frequent false alarms. A newly developed instrument was recently shown to reduce overall alarm rates by 93%(Bohnhorst B, Poets CF. Intensive Care Med 1998;24;3). We wanted to know whether this presumed reduction in false positive alarms is achieved at the expense of an increased proportion of false negative alarms, i.e. missed or delayed identification of hypoxaemia. Methods: Long-term recordings of transcutaneous PO2(Kontron 7561, Watford, UK) and pulse oximeter saturation (SpO2; Masimo SET, Irvine, CA) were performed in 10 infants (median GA at birth 25 wk (range 24-30), age at study 1 wk (1-11)) with severe apnoea of prematurity. Hypoxaemia was defined as a fall in TcPO2 to ≤40 mmHg. A hypoxaemic episode was classified as identified by the pulse oximeter if there was a fall in SpO2 to ≤85% within 2 minutes of TcPO2 reaching 40 mmHg. Results: 62 falls in TcPO2 to ≤40 mmHg occurred in 8 infants; 61 (98%) were identified by the pulse oximeter (median interval 25 s before the fall in TcPO2 to ≤40 mmH), maximum interval 90 s after it). The SpO2 nadir during the episode where the oximeter had not alarmed was 86%. Conclusion: The reduced alarm rate previously found for the new oximeter evaluated in this study was not achieved at the expense of a delayed or unreliable identification of hypoxaemic episodes.