SPECIAL COMMUNICATIONS: Letters to the Editor-in-Chief
The article by Gore et al. (3) documenting a systematic error in oxygen consumption (V̇O2) measurement of a commercial system during exercise has important implications for laboratory quality control efforts. Techniques used to measure V̇O2 have included direct bag collections, mixing chamber, and breath-by-breath (BxB) systems. Each approach has its own set of problems and can lead to inaccuracies in the wrong hands. The mixing chamber and BxB techniques are more amenable to computerization, and thus many commercial systems implement these techniques. Both types of commercial systems may have signal alignment problems. With BxB systems, the gas flow and concentrations are generally measured at the mouth, but the gas concentration signals are delayed by the sum of the transport into the analyzer and the short time needed for the analyzer to respond. The software must shift the gas concentration signals by exactly the total delay time before integrating the signals with flow to determine V̇O2. Calibration or setup procedures in commercial systems must include a routine to determine this delay on a regular basis, as it may change if the tubing ages or is crimped, if the length is changed, or if the gas sampling rate of the gas analyzer changes. It is even plausible that during a test the effective flow rate could vary as a consequence of condensation of water vapor within the lumen of the sampling tube since, according to Pouiselle’s Law, the resistance to flow within a tube is inversely related to the radius raised to the fourth power. Errors in the delay parameter can lead to substantial errors in V̇O2 (1,2,4,5). Though it would seem that this parameter would be easy to define, it turns out in practice that the exact value is often determined empirically. The article by Gore et al. (3) demonstrates errors in BxB gas exchange compared with their computerized and automated bag collection system. However, there could be some misunderstandings generated by this article that should be clarified.
The first potential misunderstanding is the data presented in Table 2, which is a survey of laboratories using the Medical Graphics CPX/D system and their lag times that vary from under 100 ms to over 400 ms. However, this variability may be appropriate and quite normal. Actual lag time will depend on the length of the sample line, the sampling flow of the gas analyzer, and internal delays within the analyzer itself. The important issue is that the software measures this delay accurately. To this end, the authors found that a configuration parameter contained in a file of the Medical Graphics system, “CAL.CFG,” was important in determining how the measured delay time was used by the software. Gore et al. found essentially the same thing we did (5), that varying the value for O2 and CO2 found on lines 202 and 203 of this file effectively fine-tuned the actual delay time used by the software and resulted in systematic changes in the calculated V̇O2 and V̇CO2. However, it appears that when Gore et al. (3) left the parameter at a value of 0 or −60 ms there was still an error in V̇O2 ((3), Fig. 2). We now present these data, replotted to show by extrapolation the “optimal” value for the parameter for all subjects at 100 and 300 W (Fig. 1). Note there is some variability in this value among subjects, but the extrapolated value that would result in zero error should be positive instead of the negative value that, apparently, is shipped with systems from the factory.
The article by Gore et al. (3) leaves one with the impression that there is a permanent error in BxB measurements that cannot be rectified. However, we feel it is important to emphasize that commercial equipment can often be fine-tuned to optimize performance and that users should work with manufacturer’s support teams to perform validations and optimizations. There are multiple sources of error in addition to delay times between analyzers that could be at fault, including calibration of the pneumotachograph, linearity and calibration of the gas analyzers, the handling of water vapor (including performance of gas sample lines designed to eliminate water vapor before the gas enters the analyzer, which are now common on most commercial systems), and finally fine tuning of the delay time parameter. Users should not adjust software setup parameters without close consultation with the manufacturer.
Gore et al. (3) and our work (5) focused on the Medical Graphics CPX/D system. However, any BxB system will potentially be prone to the same errors and should routinely be validated in a similar way. Commercial vendors are encouraged to make this process easier by perhaps providing data analysis routines and step-by-step instructions to help users optimize their own systems.
BxB gas exchange has become commonplace in many laboratories around the globe. Its appeal is in ease of use and the high time resolution of the measurements, but users should be aware that, of the three major techniques available for gas exchange assessment, it is technically the most difficult to implement and may be the most prone to technical failure. The article by Gore et al. served to emphasize this point, but users need to be aware there may be means to optimize system performance.
Kenneth C. Beck, Ph.D.
Christopher J. Gore, Ph.D.
1. Beaver, W. L., N. Lamarra, and K. Wasserman. Breath-by-breath measurement of true alveolar gas exchange. J. Appl. Physiol. 51: 1662–1675, 1981.
2. Beaver, W. L., K. Wasserman, and B. J. Whipp. On-line computer analysis and breath-by-breath graphical display of exercise function tests. J. Appl. Physiol. 34: 128–132, 1973.
3. Gore, C. J., R. J. Clark, N. J. Shipp, et al. CPX/D underestimates V̇O2
in athletes compared with an automated Douglas bag system. Med. Sci. Sports Exerc. 35: 1341–1347, 2003.
4. Hughson, R. L., D. R. Northey, H. C. Xing, et al. Alignment of ventilation and gas fraction for breath-by-breath respiratory gas exchange calculations in exercise. Comput. Biomed. Res. 24: 118–128, 1991.
5. Proctor, D. N., and K. C. Beck. Delay time adjustments to minimize errors in breath-by-breath measurement of V̇O2
during exercise. J. Appl. Physiol. 81: 2495–2499, 1996.