1997 OPEN FORUM Abstracts
0-16 L/MIN OXYGEN FLOWMETERS ARE NOT ACCURATE IN DELIVERING LOWFLOW THERAPY.
Boynton B. CRTT, Jung S. RRT, Nelson S. RRT, Potts C. CRTT, Rankin D. CRTT, Reed M. CRTT, Salzwedel K. RRT, Schuelke D. CRTT, Stewart T. CRTT, Weathers P. CRTT, Ward J. RRT, Helmholz H., Jr. MD, Plevak D. MD. Rochester Community College and Mayo Foundation, Rochester, MN
Background: Some patients with chronic obstructive pulmonary disease (COPD) require accurate administration of low flow oxygen. For proper patient care, we felt that oxygen flowmeters should be accurate to within 10% error at flows ranging from 1-3 L/min. We performed this investigation to see if there was sufficient agreement between the settings on 0-16 L/min oxygen flowmeters and measurements made by a calibration analyzer. Methods: Oxygen flow measurements were made by a Timeter RT 200 Calibration Analyzer (Allied Healthcare Products Inc., St. Louis, MO) The accuracy of the analyzer was confirmed using a Tissot water-sealed spirometer. Following a pilot study accomplished to determine the number of flowrneters needed for sufficient statistical power, a random selection of 74 Ohio/Ohmeda oxygen flowmeters currently in clinical use were tested at three flows: 1, 2, and 3 L/min. Flow settings were made visually and verified by three observers; the center of the indicator ball was used as the landmark. The oxygen flow analyzer measurement was recorded after the reading had stabilized. Bland-Altman plots were used to analyze the data. Statistical agreement between the oxygen flowmeter settings and the analyzer measurements would have been confirmed if 95% of the data points (two standard deviations) were contained within a 10% margin of error for each flow level. Results: The standard deviation of data points at 1 L/min was 0.15 (see Fig. 1), at 2 L/min was 0.18 and 3 L/min 0.31. Standard deviations of data exceeded the 10% margin of error in all three flow settings. Conclusion: 0-16 oxygen flowmeters are not accurate at flows from 1 to 3 L/min and should be used with caution in patients with COPD.
(See original for figure)