Pulse oximetry may overestimate oxygen saturation in PICU patients

Image: PD

1. Pulse oximetry (SpO₂), on average, overestimated oxygen saturation as measured by arterial blood gas CO-oximetry (SaO₂) in critically ill children on full ventilator support with SpO₂ below 97%.

2.  SpO₂ showed variable bias and precision dependent on SpO­_2­­­ measurement, with inferior performance in SpO₂ ranges below 91% compared with those above.

Evidence Rating Level: 2 (Good)

Study Rundown: Pulse oximetry (SpO₂) is often used to assess oxygenation and as a surrogate for arterial blood gas CO-oximetry (SaO₂) when making clinical decisions in hospitalized children. Smaller studies have previously demonstrated evidence of a bias for SpO₂ in children to overestimate oxygen saturation. This study sought to add to these findings through a large multicenter trial comparing SpO₂ with SaO₂ in 225 children on full ventilator support. The study found that SpO₂ overestimated SaO₂, but variably so. SpO₂ measurements were less accurate below 91% compared with above. The trial was limited by its observational format and potential selection bias due to inclusion of SpO₂ and SaO₂ comparisons only among children in which arterial blood gas testing was deemed necessary by the medical team. Regardless, these findings call for more information about the reliability of pulse oximetry in children, beyond the current accuracy trials performed on healthy adult volunteers, in order to better guide clinicians in the appropriate use of this clinical tool.

Click to read the study, published today in Pediatrics

Relevant Reading: Pulse Oximetry in Pediatric Practice

In-Depth [prospective, observational study]: This multicenter PICU-based, prospective, observational study of 225 children, from birth to 18 years of age who were mechanically ventilated with SpO₂ less than 97% garnered 1980 SpO2/SaO2 pairs for analysis. Analysis looked at bias (SpO₂ – SaO₂), local bias (the bias over ranges of SpO₂), precision (a standard deviation above and below the mean bias) and accuracy root mean squared (A­_rms ,combines bias, precision, and the number of samples obtained and is reported to regulatory bodies for pulse oximetry). The study also used multivariate models to control for confounding variables such as congenital cyanotic heart disease or prolonged capillary refill. The bias between pulse oximetry and arterial blood gas was greatest when the pulse oximeter readings were between 81% and 85%, (336 samples, median 6%,mean 6.6%, A­_rms 9.1%).  There was the least bias in readings between 91% and 97% SpO₂ (901 samples, median 1%, mean 1.5%, A­_rms 4.2).

By Laurel Wickberg and Leah H. Carr

More from this author: PG-13/R rated movies similar in violence and risk behavior volume, Many smokeless tobacco users also smoke, Emergency department use of radiography in asthma increasing, Antivirals linked to reduced mortality among children with influenza, Regular bedtime helps behavior in children

©2012-2013 2minutemedicine.com. All rights reserved. No works may be reproduced without expressed written consent from 2minutemedicine.com. Disclaimer: We present factual information directly from peer reviewed medical journals. No post should be construed as medical advice and is not intended as such by the authors, editors, staff or by 2minutemedicine.com. PLEASE SEE A HEALTHCARE PROVIDER IN YOUR AREA IF YOU SEEK MEDICAL ADVICE OF ANY SORT.