Reply: On the Methodology of the Conconi Test

The aim of our study was to investigate the relationship between “anaerobic thresholds” obtained from incremental exercise testing and the maximal lactate steady-state (MLSS) in elite cyclists. Because the vast majority of cyclists use heart rate as an index of exercise intensity, we were particularly interested to find out the threshold procedure yielding the most precise prediction of heart rate corresponding with the MLSS. Dr. Grazzi and his co-workers criticize the test protocol we have used to eventually conclude that the “Conconi threshold” (deflection of the heart rate/power curve) is an inadequate method to predict MLSS exercise intensity. We acknowledge the criticism that we did not use the test procedures as originally described by Grazzi and his co-workers [[5]]. Thus, we cannot exclude that this procedure still might yield a good prediction of MLSS. However, it must at the same time be emphasized that experimental evidence is entirely missing which proves that the test protocol by Grazzi et al. allows to adequately predict MLSS heart rate or power output. In fact, the reliability and validity of the Conconi threshold principle has been a matter of debate at several earlier occasions [[2], [3], [7], [8]].

Furthermore, we did not use the test protocol as described by Grazzi et al. with the express purpose to optimize the validity of our study. Grazzi et al. have justified their test procedure [[5]] by the premise that “increments in power output during cycling are obtained by increasing cadence, as well as the force applied during each pedalling stroke”. However, this premise is false. Indeed, it is inherent to the gear system of a race bicycle that different power outputs can be generated while maintaining constant cadence. In fact, the gear system serves to uncouple power from cadence. Thus, during training and competition as a rule power output can largely vary, whilst cadence only slightly fluctuates. In addition, heart rate during cycling at a given power output depends on the cadence used, with higher cadence resulting in higher heart rate [[4]]. It is thus inadequate to design an exercise test procedure requiring cyclists to increase power output by increasing cadence. This procedure forces cyclists into an unaccustomed temporal co-ordination pattern (so-called “souplesse”) resulting in a shifted power heart rate relationship which is not relevant to monitoring training intensity. Therefore, it is important that cyclists use their “normal” cadence during testing. This is not possible with the test procedure proposed by Grazzi et al. [[5]].

Our above rationale applies to road cycling. However, on the track cyclists use a fixed gear which, indeed, implies that the only means of increasing power output is increasing cadence. Grazzi et al. use the observation that track cyclists sprint at a cadence exceeding 160 revolutions per min, as an argument to prove the point that power output during cycling is the simple exponent of cadence. They pretend that such high cadences are optimal to develop maximal power. Again, this contention is incorrect. It has been well established that maximal power output during cycling occurs at cadences between 90 and 120 rpm. Beyond 120 rpm maximal power drops (Fig. [1], Koninckx et al., unpublished observations) [[1], [6], [9]]. Why then do track sprinters develop cadences exceeding 160 rpm? Simply because the selection of the gear ratio results from the compromise between going for maximal sprint power (high gear ratio) yet maintaining sufficient potential to accelerate (low gear ratio), the latter also being crucial to victory. In fact, road cyclists seldom sprint at cadence exceeding 105 rpm (53 · 11 gear ratio, 60 - 65 km · h^-1), which is compatible with the power velocity curve depicted in Fig. [1].

Fig. 1 Relationship between maximal power output and cadence in elite cyclists (Koninckx et al., unpublished observations). Young elite cyclists (n = 16) competing at national and international level (20.4 ± 0.6 years; height: 181 ± 2 cm; body weight: 71.0 ± 1.9 kg) were tested on their own race bicycle that was mounted on a self-constructed ergometer allowing to function in an isokinetic modus. Maximal power output was measured during short maximal sprints (5 s) at cadences from 40 to 140 revolutions per min. Cadences were administered in random order. The figure shows that power peaks at ∼ 120 revolutions per min and drops at either lower or higher cadences.

Dr Grazzi and his co-workers use a case report to support their points. We acknowledge that Francesco Moser in 1988 exhibited a Conconi threshold that corresponded surprisingly well with his 1-hour time trial work rate. Along this line, we also could select a case report from an elite cyclist to demonstrate that there is an absolute mismatch between the Conconi threshold as determined in this study, and the maximal lactate steady-state. However, this would be scientifically incorrect because this selection would be biased and not representative for the general trend in the population considered. And this trend says that in most cyclists the individual so-called Conconi threshold, if possible to determine at all, mismatches with individual MLSS and lacks reliability [[2], [8]].

Reference

• 1 Baron R, Bachl N, Petschnig R, Tschan H, Smekal G, Pokan R. Measurement of maximal power output in isokinetic and non-isokinetic cycling. A comparison of two methods.  Int J Sports Med. 1999;  20 532-537
• 2 Bourgois J, Vrijens J. The Conconi test: a controversial concept for the determination of the anaerobic threshold in young rowers.  Int J Sports Med. 1998;  19 553-559
• 3 Conconi F, Grazzi G, Casoni I, Guglielmini C, Borsetto C, Ballarin E, Mazzoni G, Patracchini M, Manfredini F. The Conconi test: methodology after 12 years of application.  Int J Sports Med. 1996;  17 519
• 4 Gotshall R W, Bauer T A, Fahrner S L. Cycling cadence alters exercise hemodynamics.  Int J Sports Med. 1996;  17 17-21
• 5 Grazzi G, Alfieri N, Borsetto C, Casoni I, Manfredini F, Mazzoni G, Conconi F. The power output/heart rate relationship in cycling: test standardization and repeatability.  Med Sci Sports Exerc. 1999;  31 1478-1483
• 6 Hintzy F, Belli A, Grappe F, Rouillon J D. Optimal pedalling velocity characteristics during maximal and submaximal cycling in humans.  Eur J Appl Physiol Occup Physiol. 1999;  79 426-432
• 7 Hofmann P, Pokan R, von Duvillard S P, Seibert F J, Zweiker R, Schmid P. Heart rate performance curve during incremental cycle ergometer exercise in healthy young male subjects.  Med Sci Sports Exerc. 1997;  29 762-768
• 8 Jeukendrup A E, Hesselink M K, Kuipers H, Keizer H A. The Conconi test.  Int J Sports Med. 1997;  18 393-396
• 9 Sargeant A J. Human power output and muscle fatigue.  Int J Sports Med. 1994;  15 116-121