Way to automated determination of the anaerobic threshold
for anaerobicthreshold.org by P. Kvaca [3]


We remember that the anaerobic threshold gained from ventilatory-respiratory values is physiologically defined as the break of ventilation curve in the moment of non-linear ventilation increase. The ventilation increases faster then Oxygen Uptake (VO2) at this change but also the Carbon Dioxide Production (VCO2) and Respiratory Exchange Ratio (RER) increase.

Therefore the basic principle of the non-invasive AT determination is a search for the break in respiratory data [1]. The physician visually judges the parameter relation and he determines the AT. The precision of this AT determination firstly depends on the number of measured points. This number is small for the spiroergometric parameters obtained with the help of cheaper devices measuring in average values. That means that the detection of break in a parameter relation is difficult not only for physicians but it is also impossible this visual procedure to algorithm for automated computer processing as well.

The main problem results from this way of AT determination - every exercise physiologist makes a decision according to own subjective experience, which he gained during his practice and his experience during the AT assessment is applied. It can be pronounced that the AT determination is not so far algorithmic, because the several evaluators determine diverse results from the identical input data (e.g. of one spiroergometric relation) and it exist no chance to meaningfully justify the variability of results.

It would be advisable to lose the dependence on experience and to algorithm the AT determination - the transparent procedure has to guarantee that the same result will be obtained from the same input data. Next advantage of such algorithm would be an easy computer implementation.

Software systems, which are added to the expensive measuring devices, have a procedure of an AT determination implemented. However, if it is looked on the evaluation procedure at the first glance, it is obvious that the above-mentioned problem is not solved. Some methods of the AT determination are offered, but the algorithms of the AT determination on the ground of firmware confidentiality are not described and therefore the way of AT determination value is not known. Thus it is impossible to compare that AT with another AT determined with the help of the different software producer.

Next problem of mentioned software's arises during the evaluation because they mostly ask the user for the division of the relation into two parts. The good news on this generally used methodology is that the algorithm is obvious - the widespread algorithm for the AT determination from the intersection of two regression lines is used [1]. The principle of this method is based on the division of measured points into two groups and the interlacing of each group by regression lines. The point of the intersection is regarded as an anaerobic threshold. The relation between Oxygen Uptake and Carbon Dioxide Production is the most extensive used relation for this type of the AT determination. This way is called "V-slope" method [2]. The name corresponds with the usage of two volume variables (Oxygen Uptake and Carbon Dioxide Production).

The bad news is that the plea for the division actually means the plea for our experience (computer says - tell me where you think that the AT is and I specify it). Hence this method is quite unsuitable for the computer processing because a physician provides the classification into two groups and the division depends on his intuition and his experience. Enlistment of one point in different group can affect considerably the result of the AT determination. The more measured values we have the more serious the problem is, especially for the breath-by-breath analysers [3]. Thus the problem of subjectivity during the AT determination goes back - because it is very probable that it will be a disagreement between two different experienced physicians during the division of measured points.

It is evident that both current mostly used procedures, either the visually determination or the computer processing methods of new expensive breath-by-breath devices, use the physician experience for the AT determination. Hence the AT values without the determination by means of a transparent algorithm are incomparable.

Moreover, this state leads to the disproportion when the clinical research uses the breath-by-breath devices and produces the useful clinical works dealing with the AT determined by possessed software (without transparent AT algorithm). On the other side there exist the practical physicians owning only the cheaper devices measuring only the average data. They are unable to apply the AT research knowledge in practice.

This disproportion began with the start of the last decade in the nineties when the research on the spiroergometric average data was ceased [4]. However, because of the very incomparable price of measuring devices (especially chemical analysers), the producers have not stopped the manufacture of cheaper devices so the supply of these devices is extensive. On the other hand, the demand is also extensive.

Moreover, it exists non-specialised departments, e.g. cardiology departments, which use spiroergometry for the AT determination and the AT value for the prescription of physical activity as a supplemental method. They have no need to buy an expensive breath-by-breath device and with the devices measured only average data are satisfied.

Reference:

[1] Huges EF, Turner SC, Brooks GA. Effects of glycogen depletion and pedaling speed on ”anaerobic threshold”. J Appl Physiol 1982;52(6);1598-1607. [2] Beaver, W.L., Wassermann, K., Whipp, B.J.: A new method for detecting anaerobic threshold by gas exchange. J. Appl. Physiol., 60, pp. 2020-2027, 1986. [3] Kvaca P., J. Radvansky: “Methods for Automated Anaerobic Threshold Determination.” Med Sport Boh Slov, 8/2 (1999), 51-57. [4] Shimizu M, Myers J, Buchanan N, Walsh D, Kraemer M, McAuley P, Froelicher VF The ventilatory threshold: method, protocol, and evaluator agreement. Am Heart J 1991 Aug;122(2):509-16
 
anerobicthreshold.org ©2007