Journal Logo

SPECIAL COMMUNICATIONS: Contrasting Perspectives

The Respiratory Compensation Point and the Deoxygenation Break Point Are Not Valid Surrogates for Critical Power and Maximum Lactate Steady State


Author Information
Medicine & Science in Sports & Exercise: November 2018 - Volume 50 - Issue 11 - p 2379-2382
doi: 10.1249/MSS.0000000000001699
  • Free

The physiological threshold separating “sustainable” and “nonsustainable” exercise intensities has undoubtedly, in practical terms, dictated human activity since the origin of man. A preponderance of evidence, from nearly a century of research, has repeatedly identified the critical power (CP)/maximum lactate steady-state (MLSS) parameters as a robust quantification of this physiological threshold. The CP/MLSS defines the highest exercise intensity for which the energetic requirements can be met primarily by oxidative metabolism. As such, CP/MLSS represents the highest exercise intensity for which a steady-state for many crucial variables can be achieved within the exercising skeletal muscle (e.g., pH, inorganic phosphate, and phosphocreatine), the peripheral circulation (e.g., blood lactate, blood pH, skeletal muscle blood flow), and the systemic organism (e.g., oxygen consumption and neuromuscular fatigue development). Thus, the physiological underpinnings of CP/MLSS are of the upmost importance for understanding the mechanistic bases of exercise tolerance in both health and disease.

With increasing regularity, it has been argued that the respiratory compensation point (RCP)/deoxyhemoglobin-myoglobin break point (deoxy-BP) parameters are surrogates for critical power (CP)/maximum lactate steady-state (MLSS). This contention primarily relies on the premise that the work and metabolic rates associated with these parameters are, on average, not statistically different. This “absence of evidence” of a difference has been argued to be “evidence of absence” to support equivalence between the RCP/deoxy-BP and CP/MLSS (1,2). However, in addition to somewhat questionable logic, this argument largely disregards the disparate physiological mechanisms, sequential expression, and inconsistent relationship among these parameters. Moreover, this thesis is inconsistent when these parameters are evaluated across the organism and across diverse species. It has also been argued that the selection of assessing RCP/deoxy-BP or CP/MLSS can be determined by the availability of measurement instruments. However, we are firm believers in the concept that one should let one’s intentions determine the method and not vice versa, such that the physiological relevance of the parameter should dictate the means of assessment. Regardless, substantiation of this argument of equivalence, if incorrect, would greatly impede the advancement of our understanding of the mechanistic bases for exercise tolerance in health and disease. It is our contention, that, in fact, there is substantial “evidence of absence” of RCP/deoxy-BP being physiological equivalents to CP/MLSS and, thus, here we argue, based on current evidence and a mechanistic physiological framework, that the RCP/deoxy-BP are not valid surrogates for CP/MLSS.


The physiological bases of RCP/deoxy-BP must be considered when evaluating these parameters as potential surrogates for CP/MLSS. The RCP defines an exercise intensity at which frank hyperventilation ensues (i.e., increased E/CO2, decreased PaCO2 and PETCO2), for which a multitude of culprits have been identified. Indeed, body temperature and other factors (e.g., catecholamines, blood osmolarity) can directly influence ventilation and have been demonstrated to increase ventilation in a time-dependent fashion at exercise intensities below CP/MLSS (3). The primary stimulus for the RCP has been traditionally attributed to the progressively worsening metabolic acidosis during incremental exercise. For hyperventilation to occur with this stimulus, there must be an accumulation of hydrogen ions in the blood sufficient enough to overwhelm the buffering capacity and be detectable by the receptors modulating ventilation. As muscle pH, blood pH, and blood lactate all attain steady-state for exercise intensities up to CP/MLSS, the RCP, by definition, must occur at an exercise intensity above CP/MLSS. Thus, the numerous stimuli that can influence RCP clearly dissociate this parameter from CP/MLSS.

The deoxy-BP is a relatively newly defined parameter that designates an exercise intensity at which the near-infrared spectroscopy-derived deoxygenated hemoglobin/myoglobin concentration begins to level-off during ramp incremental exercise. In terms of the equivalence between the deoxy-BP and CP/MLSS, there are several potential issues to consider. First, the majority of near-infrared spectroscopy measurements are constrained to the superficial regions of the interrogated muscle. As the deoxy-BP has been most often observed in the quadriceps muscles, this means that more glycolytic muscle fibers are the primary muscle fibers being assessed, whereas the deeper more oxidative muscle fibers are neglected. Using a specially designed near-infrared spectrometer, Okushima et al. (4) assessed the oxygenation characteristics of the superficial and deep muscles of the quadriceps during cycling exercise. Interestingly, this study revealed that the deoxy-BP is only expressed in the superficial muscles of the quadriceps, and not in the deep muscles. Second, the deoxy-BP parameter is also not detectable in all subjects and is often arbitrarily modeled beyond the measured data due to the curvilinear shape of the response. These issues dampen enthusiasm for the deoxy-BP being considered equivalent to CP/MLSS.


Investigation of the order in which CP/MLSS and RCP/deoxy-BP are expressed has provided important insight into the equivalence of these parameters. Specifically, very recently, Caen et al. (5) elegantly demonstrated a sequential ordering of these parameters, in terms of work and metabolic rates, with CP/MLSS preceding the RCP/deoxy-BP. Importantly, this sequential ordering was maintained even after endurance exercise training. Consistent with this conclusion, further examination of the data from, perhaps, the first study to identify equivalence between RCP/deoxy-BP and CP/MLSS (1), actually documented a dissociation between these parameters, despite a lack of statistical difference between the average RCP/deoxy-BP and CP/MLSS. Indeed, the RCP/deoxy-BP occurred at an intensity approximately 20% greater than CP/MLSS.


For the RCP/deoxy-BP to be surrogates for CP/MLSS, these parameters should be consistently and strongly related. Despite occurring at similar work and metabolic rates, on average, the relationship between these parameters is inconsistent at best. Correlations between RCP/deoxy-BP and CP/MLSS in the literature have ranged from r < 0.01 to r > 0.90 (1,5–9). Indeed, as already noted, the primary study concluding equivalence between RCP/deoxy-BP and CP/MLSS (1) failed to demonstrate any relationship between the parameters despite a lack of statistical difference. Because of the inherent variability of the RCP/deoxy-BP (6–8), when a statistically significant correlation with CP/MLSS has been documented, it has been with a high degree of intrasubject variability, such that the RCP/deoxy-BP, in most data sets, will underestimate or overestimate CP/MLSS by more than 10% (5–8).


Examining the expression of RCP/deoxy-BP and CP/MLSS across the organism and across species provides a unique opportunity to substantiate RCP/deoxy-BP as surrogates for CP/MLSS. The CP/MLSS is robustly expressed across the spectrum of whole-body to isolated-muscle exercise, as well as across a diverse array of species (e.g., horse, rat, mouse, ghost crab, and lungless salamander; for review, see (10)). If the RCP/deoxy-BP parameters are surrogates for CP/MLSS, these parameters should be equally robust; however, this is not the case. Although the RCP/deoxy-BP are expressed, in most participants, during whole-body exercise, they are not evident during small-muscle mass exercise. For example, during handgrip or single-leg isometric exercise, there is minimal change in ventilation, and therefore, there is no RCP, whereas CP can be determined in these exercise models. Additionally, there are several animal species that do not express the RCP during exercise, but still demonstrate CP/MLSS. The exemplar of this is the horse, which due to respiratory constraints during exercise is incapable of expressing the RCP until after the cessation of exercise, but CP/MLSS is still present (11–13). Thus, the examination of these parameters across the organism and across species appears to have revealed the elusive “Black Swan” for this predominantly “absence of evidence” debate.


When assessing whether the RCP/deoxy-BP parameters are valid surrogates for CP/MLSS, more detailed analyses, beyond the statistical comparison of group mean data, are requisite. Specifically, for these parameters to be equivalent, there must be an underlying mechanistic link between them, as evidenced by a strong, consistent relationship. However, it appears, from a mechanistic physiological framework, the stimuli for the RCP/deoxy-BP must occur once the CP/MLSS is surpassed. In support of this, a sequential ordering has been demonstrated with RCP/deoxy-BP occurring at higher exercise intensities than CP/MLSS. Additionally, the relationship between RCP/deoxy-BP and CP/MLSS is inconsistent at best, and the high degree of intrasubject variability in this relationship precludes accurately predicting CP/MLSS from the RCP/deoxy-BP. Finally, but of significant importance, the CP/MLSS parameter is robust across the organism, from whole-body exercise to isolated muscle exercise, as well as across species, whereas the same is not true for the RCP/deoxy-BP. Thus, based on current evidence and a mechanistic physiological framework, the RCP/deoxy-BP are not valid surrogates for CP/MLSS.


Keir and colleagues (14) contend that the (RCP/deoxy-BP are valid surrogates for CP/MLSS and that evidence to the contrary arises primarily from: 1) methodological issues with the translation between incremental and constant-intensity paradigms and 2) a high probability of error in the determination of CP. We contend, however, that an in-depth consideration of these arguments, in conjunction with a mechanistic, physiological framework, actually lend support to the RCP/deoxy-BP not being valid surrogates for CP/MLSS.

Argument 1: Methodological issues with translation between incremental and constant-intensity paradigms

We concur with Keir and colleagues (14) with regard to the importance, and potential issues, of methodology when attempting to translate oxygen uptake and work rate values between incremental and constant-intensity paradigms. The basis of their argument is that information gleaned from an incremental test does not directly translate to a constant-intensity test and vice versa. If this is the case, it is perplexing why the authors continue to propose the use of parameters derived from an incremental test (RCP/deoxy-BP) as surrogates for a constant-intensity threshold (CP/MLSS). Looking beyond this, we contend that amassing evidence supports that RCP/deoxy-BP are not equivalent to CP/MLSS, and that this is not simply the result of the stated methodological issue. For example, we have previously circumvented the concern regarding mean response time for incremental tests by determining the oxygen uptake values associated with both CP/MLSS and RCP/deoxy-BP during constant power tests (6). Although the group mean oxygen uptake values were not statistically different between CP/MLSS and RCP/deoxy-BP, there was, importantly, no relationship between these parameters due to the remarkably high intrasubject variability. The lack of a strong and consistent relationship among CP/MLSS and RCP/deoxy-BP in the literature precludes the acceptance of these parameters as equivalent. Thus, as stated in our perspective, there is ample evidence that RCP/deoxy-BP are not valid surrogates for CP/MLSS, when the data are considered beyond group mean values.

Argument 2: A high probability of low accuracy in the determination of CP

First off, the statement that there is a high probability of low accuracy in the determination of CP is unfounded. This argument ignores extensive evidence validating the accuracy of CP in demarcating the threshold between “sustainable” and “nonsustainable” exercise intensities. One of the most robust approaches to assess the accuracy of CP is to conduct constant-intensity tests just above and below CP. Importantly, these assessments have consistently demonstrated CP to be a valid demarcation between “sustainable” and “nonsustainable” exercise (10,15). Indeed, the data presented by Keir and colleagues (14) in Figure 1 clearly demonstrate that, despite potential variability between incremental and constant-intensity exercise, CP/MLSS robustly defines sustainable exercise (i.e., attainment of steady-state oxygen uptake and 30+ min of exercise), whereas RCP/deoxy-BP does not (i.e., attainment of peak oxygen uptake and attendant task-failure). Thus, substantial efforts have been made to determine how closely CP/MLSS identifies the physiological responses expected at the critical metabolic rate, while efforts to validate RCP/deoxy-BP, in a similar manner, have not been performed.

Physiological mechanisms versus methodology

The arguments put forth by Keir and colleagues (14) are purely methodological in nature. However, the physiological determinants of these parameters and a mechanistic link between them are, perhaps, most important to establish whether RCP/deoxy-BP are valid surrogates for CP/MLSS. If these parameters are, indeed, surrogates, there must be a strong, consistent relationship between them. We and others have demonstrated repeatedly that this relationship is lacking (6–8). RCP, as viewed in the paradigm of incremental exercise, is predicated on the accumulation of hydrogen ions and the associated respiratory buffering of the increasing acidemia. However, when encouraged by very slow ramp rates, the RCP can be expressed at moderate work rates closely approximating gas exchange threshold, indicating that this exercise parameter is protocol dependent (3,16). Although Keir and colleagues (14) recognize that CP/MLSS represents that highest intensity that can be sustained by oxidative metabolism without progressive depletion of high energy phosphates or continuous lactate accumulation, they then state that the RCP/deoxy-BP also reflects this boundary. However, as in our perspective, we contend that these two statements are very difficult to reconcile, as the RCP/deoxy-BP occurs, physiologically, in the non–steady-state. Moreover, these parameters are clearly dissociated during small muscle mass and isolated muscle exercise, where CP/MLSS occur without RCP/deoxy-BP. Unique insight into human physiology is often provided by animal models, where it has been demonstrated, in some of these models, that CP/MLSS occurs without the expression of RCP/deoxy-BP (11–13). Thus, when the physiological mechanisms responsible for these parameters are considered, and experimentally tested, there is reasonable evidence to suggest that RCP/deoxy-BP are not valid surrogates for CP/MLSS.


It is our contention that, based on current evidence and in conjunction with a mechanistic, physiological framework, RCP/deoxy-BP are not valid surrogates for CP/MLSS. Certainly, there are numerous examples of RCP/deoxy-BP not being statistically different from CP/MLSS, when assessed as group mean values, which appears to confirm the equivalence of these parameters. However, these “White Swans” do not provide definitive proof, unlike the, already highlighted, “Black Swans,” for whom only one instance of dissociation between these parameters is needed, and has been documented, to confirm a lack of equivalence. Furthermore, we argue that, due to different physiological stimuli, sequential ordering, inconsistent relationships, and a lack of congruency across exercise paradigms and organisms, RCP/deoxy-BP and CP/MLSS are clearly not equivalent.


1. Keir DA, Fontana FY, Robertson TC, et al. Exercise intensity thresholds: identifying the boundaries of sustainable performance. Med Sci Sports Exerc. 2015;47(9):1932–40.
2. Keir DA, Fontana FY, Robertson TC, et al. Response. Med Sci Sports Exerc. 2015;47(9):1998–9.
3. Whipp BJ. Control of the Exercise Hyperpnea. In: Hornbein TF, editor. Regulation of Breathing. New York: Dekker; 1981. pp. 1069–138.
4. Okushima D, Poole DC, Rossiter HB, et al. Muscle deoxygenation in the quadriceps during ramp incremental cycling: deep vs superficial heterogeneity. J Appl Physiol. 2015;119(11):1313–9.
5. Caen K, Vermeire K, Bourgois JG, Boone J. Exercise thresholds on trial: are they really equivalent? Med Sci Sports Exerc. 2018;50(6):1277–84.
6. Broxterman RM, Ade CJ, Barker T, Barstow TJ. Influence of pedal cadence on the respiratory compensation point and its relation to critical power. Respir Physiol Neurobiol. 2015;208:1–7.
7. Broxterman RM, Ade CJ, Craig JC, WIlcox SL, Schlup SJ, Barstow TJ. The relationship between critical speed and the respiratory compensation point: coincidence or equivalence. Eur J Sport Sci. 2015;15(7):631–9.
8. Leo JA, Sabapathy S, Simmonds MJ, Cross TJ. The respiratory compensation point is not a valid surrogate for critical power. Med Sci Sports Exerc. 2017;49(7):1452–60.
9. Dekerle J, Baron B, Dupont L, Vanvelcenaher J, Pelayo P. Maximal lactate steady state, respiratory compensation threshold and critical power. Eur J Appl Physiol. 2003;89:281–8.
10. Jones AM, Vanhatalo A, Burnley M, Morton RH, Poole DC. Critical power: implications for determination of V˙O2max and exercise tolerance. Med Sci Sports Exerc. 2010;42(10):1876–90.
11. Bayly WM, Hodgson DR, Schulz DA, Dempsey JA, Gollnick PD. Exercise-induced hypercapnia in the horse. J Appl Physiol. 1989; 67(5):1958–66.
12. Lauderdale MA, Hinchcliff KW. Hyperbolic relationship between time-to-fatigue and workload. Equine Vet J Suppl. 1999;30:586–90.
13. Padilla DJ, McDonough P, Kindig CA, Erickson HH, Poole DC. Ventilatory dynamics and control of blood gases after maximal exercise in the thoroughbred horse. J Appl Physiol. 2004;96(6):2187–93.
14. Keir DA, Pogliaghi S, Murias JM. The respiratory compensation point and the deoxygenation break point are valid surrogates for critical power and maximum lactate steady state. Med Sci Sports Exerc. 2018;50(11):2375–8.
15. Poole DC, Ward SA, Gardner GW, Whipp BJ. Metabolic and respiratory profile of the upper limit for prolonged exercise in man. Ergonomics. 1988;31(9):1265–79.
16. Wasserman K, Hansen JE, Sue DY, et al. Principles of Exercise Testing and Interpretation: Including Pathophysiology and Clinical Applications. 5th ed. Philadelphia: Lippincott Williams & Wilkins; 2012.
Copyright © 2018 by the American College of Sports Medicine