Transcutaneous Versus Blood Gas Measurement of CO2 in Neonates: Variation by Method and Racial Group

Abstract Objective: Transcutaneous carbon dioxide monitors are used to non-invasively monitor blood CO2 levels in acute care settings, however, their accuracy has been shown to vary. We studied the correlation between transcutaneous monitors and arterial, capillary, and venous CO2 blood gas measurements in neonates. We also sought to examine any variability across different racial groups. Design and Setting: Retrospective chart review at a level 4 NICU conducted at Albany Medical Center, Albany, New York, USA. Patients: Twenty-four NICU patients who underwent transcutaneous CO2 monitoring were studied over a 1-year period. Interventions: None Measurements and Main Results: Arterial, venous, and capillary CO2 measurements that occurred within one hour of transcutaneous carbon dioxide monitor measurement were collected. Arterial measurements maintained the highest degree of difference from transcutaneous values (degree of bias 9.3 torr). Venous monitoring and transcutaneous monitoring yielded the lowest degree of bias (-.46 torr), though this sample size was the lowest of our blood gas types. Capillary point-of-care testing values yielded an intermediate bias (4.12 torr). Greater agreement existed when blood gas PCO2 values increased over time for a given patient. No difference in the amount of bias was appreciated between racial groups. Conclusions: Transcutaneous CO2 monitoring in neonatal patients is within the accepted manufacturer range of error when compared to blood gas analysis, and can be considered an accurate, noninvasive method of monitoring. Capillary point of care analysis is an efficient means of confirming transcutaneous monitoring and had better agreement than the more invasive arterial gases.