Secondary Logo

Journal Logo

Confusion Between Integration and Receiver Operator Curves?

Drummond, Gordon MD, FRCA

doi: 10.1213/ANE.0000000000001497
Letters to the Editor: Letter to the Editor

Department of Anaesthesia Critical Care and Pain Medicine, University of Edinburgh, Edinburgh, United Kingdom,

Back to Top | Article Outline

To the Editor

Tusman and coauthors compared carbon dioxide elimination with cardiac output measurement in identifying response to fluid administration.1 I was confused by the description they provided of the method used to measure carbon dioxide elimination. I understand that the authors used a device (NICO) that measures simultaneously the gas flow out of the airway opening and the fractional concentration of carbon dioxide in that flow from the airway opening. It is widely accepted that if these measurements are exactly coincident in their timing, then they may be used to calculate carbon dioxide production. The gas flow is multiplied by the carbon dioxide fraction in the exhaled gas, and integrated with respect to time during the expiration, to obtain the volume of carbon dioxide exhaled. In effect, this is the “area under the curve” of the exhaled carbon dioxide versus exhaled volume plot, often called the “volumetric capnogram.” If this value is multiplied by the respiratory frequency, then carbon dioxide elimination can be obtained in milliliters per minute. This method has been commonly used since the Siemens company developed the Servo in-line carbon dioxide analyzer in the 1980s, and forms the basis of the method of volumetric capnography.

The account that Tusman and colleagues give of their method, however, states “The device provides the CO2 eliminated per minute (VCO2 in mL/min) by multiplying the area under the receiver operating characteristic (ROC) curve (AUC) of the volumetric capnography by the respiratory rate.” This confusing statement is followed by a citation2 that relates to measure of mixed expired carbon dioxide but not directly to their method. The authors seem to have confused their account of volumetric capnography with their use of the receiver operator curve to assess their results. It is important that the authors should clarify their description of the method that they used because it is clearly central to the study.

Gordon Drummond, MD, FRCA
Department of Anaesthesia Critical Care and Pain Medicine
University of Edinburgh
Edinburgh, United Kingdom

Back to Top | Article Outline


1. Tusman G, Groisman I, Maidana GA, et al. The sensitivity and specificity of pulmonary carbon dioxide elimination for noninvasive assessment of fluid responsiveness. Anesth Analg. 2016;122:14041411.
2. Badal JJ, Loeb RG, Trujillo DK. A simple method to determine mixed exhaled CO2 using a standard circle breathing circuit. Anesth Analg. 2007;105:10481052.
Copyright © 2016 International Anesthesia Research Society