The Science Journal of the American Association for Respiratory Care

2008 OPEN FORUM Abstracts


William M. Scicolone1, P. R. Mitchell2, Heather L. Kline1

Background: Developmental disorders can threaten the accuracy of SPO2. In order to adequately assess these individuals it was necessary to determine measurement reliability of SPO2 using our existing equipment.

Twenty-five subjects were selected for participation based on nurses' input regarding which individuals were difficult to test for SPO2. The performance of the pulse oximeters was verified by using a Biotek Index II Simulator. The devices were tested at Sp02 96%, heart rate 75 and then at Sp02 78%, heart rate 180. All of the devices in use at this facility were found to be within + 2% accuracy.
Oxygen saturation values for the three measurement sites of finger (Nonin Onyx 2), ear (Nelcor N20) and forehead (Nelcor N560) were compared for obtainability of readings and SpO2 levels at 45 seconds and 120 seconds for the 25 subjects in this investigation. Across the three body sites, the forehead was the only location which yielded readable values for all subjects (25/25) at both 45 and 120 seconds. Measurements at the ear were successfully obtained for 23/25 (92%) subjects and measurements at the finger were obtainable for 20 of the 25 subjects (80%).

In order to examine differences in measurements, missing data were excluded, resulting in 18 subjects for the remaining analyses. Differences in SpO2 values were examined across the three body sites using nonparametric statistical tests due to the sample size for the study and lack of a normal distribution for the data. A significance level of <.01 was set to accommodate for experiment wise error rate (Bonferroni correction). A Friedman test (nonparametric repeated measures analysis of variance) revealed a statistically significant difference between the three sites for the measurements taken at 45 seconds (chi square=11.281, df=2, <.004). For the measurements taken at 120 seconds, there was not a statistically significant difference between sites (chi square=2.62, df=2, < .270). Pairwise comparisons of the finger and forehead sites using a Wilcoxon Test revealed a statistically significant difference between the SpO2 values at the finger and forehead (z= -2.85, < .004) for the measurements at 45 seconds, but not at 120 seconds (z= -1.94, < .053).

While peripheral sites may be accessed more readily and have value in performing routine spot checks on some individuals, forehead sensing has proven to be the most consistent measurement for these participants.