The maximal metabolic steady state represents the highest metabolic rate at which exercise can be maintained almost exclusively by oxidative metabolism. The oxygen uptake (V̇O2) associated with this critical intensity of exercise resides approximately halfway between the V̇O2 at which blood lactate concentration exceeds its resting value (i.e., lactate threshold [LT]) and the V̇O2max (1–3). Representing the upper limit of sustainable exercise, the maximal metabolic steady state is a key parameter for the evaluation of aerobic exercise intensity and a valuable component for personalized exercise prescription. However, the incidence with which it is used within the exercise and sport science community is relatively low due primarily to the cumbersome methods required for its detection and because, at present, it may not be identified from the ramp-incremental exercise protocols that commonly precede exercise interventions in both research and practical settings (1,4). For this reason, exercise prescription continues to be based on percentages of maximal values—a framework that assumes, incorrectly, that i) exercise at a given fraction of V̇O2max elicits universal metabolic stress characteristics and ii) the V̇O2 versus power output (PO) relationship from ramp exercise reflects that of constant-load exercise (1,2,5–7).
Metabolic demands exceeding the maximal metabolic steady state (i.e., within the severe-intensity domain) necessitate heightened rates of phosphorylation at the substrate level resulting in the depletion of stored energy and accumulation of fatigue-inducing metabolites intramuscularly (e.g., inorganic phosphate) and in plasma (e.g., lactate and hydrogen ions [H+]) (8,9). We have demonstrated that the V̇O2 at the respiratory compensation point (RCP) closely approximates the maximal metabolic steady state and the boundary between heavy and severe intensity domains (2,10–14). During incremental exercise, when V̇O2 surpasses a finite value around 75%–85% V̇O2max, ventilation increases disproportionately from the rate of carbon dioxide excretion (V̇CO2), causing arterial CO2 tension (PCO2) to fall. Driven by a sudden disruption in the balance between metabolically derived anion production and clearance from blood plasma (15–17), this rapid hyperventilatory response and associated hypocapnia constrains the rise in arterial [H+]. Within an individual, the V̇O2 at which this “respiratory compensation point” occurs is highly reproducible in test–retest designs using the same (11,18,19) and different incremental protocols (13,20,21), indicating an inextricable link between the hyperventilatory response and the metabolic rate at which it is initiated.
Although the V̇O2 at RCP from ramp-incremental exercise coincides with the metabolic rate associated with the maximal metabolic steady state (2,10–14), identifying the PO (or speed) at which this important boundary occurs is not presently possible. This is because the V̇O2 versus PO relationship from ramp exercise relates poorly to that of constant-load exercise (1,4,7). The misalignment of V̇O2 from PO occurs immediately at ramp onset: changes in muscle V̇O2 are detected at the mouth after a delay equivalent to muscle–lung circulatory time and the kinetics component of muscle V̇O2 (i.e., mean response time [MRT]) (22). The MRT, and thus the magnitude of the misalignment (equivalent to 10–15 W for an MRT of 20–30 s from a 30-W⋅min−1 ramp), remains fixed up to the LT because, below this intensity, V̇O2 kinetics and gain (∆V̇O2/∆PO) are constant (23–25). Above LT, V̇O2 kinetics slow and gain heightens progressively (6,25,26), causing V̇O2 to become exceedingly misaligned from ramp PO (1,13,27). Paradoxically, these V̇O2 response dynamics are masked during most ramp protocols (13) and manifest as a near linear V̇O2–PO relationship (26). As a corollary, incremental exercise overestimates the constant power required to elicit the V̇O2 at the RCP (1,4,13). This conundrum has prevented the appropriate use of the RCP concept in exercise prescription and has, as a consequence, generated skepticism for its suitability as a heavy- to severe-intensity domain surrogate (28).
To overcome some of these challenges, we previously developed a simple “step–ramp” protocol that accurately and reliably quantified the MRT, and thus the interval by which ramp V̇O2 data need to be “left shifted” in relation to PO (29). Despite its validity, this protocol is only effective at aligning V̇O2 versus PO relationships of ramp and constant exercise paradigms below the LT and within the moderate intensity domain where V̇O2 kinetics are invariant. With the objective of estimating also the PO associated to the maximal metabolic steady state in a single visit, we tested whether appending a heavy-intensity step transition to our “step–ramp” protocol (i.e., “step–ramp–step” [SRS]) could be used. With this approach, the V̇O2–PO values associated with LT and from the heavy step could be used to estimate, by linear regression, the PO at RCP. We hypothesized that exercise at the PO estimated from the SRS protocol would accurately elicit the V̇O2 at RCP. In addition, if the RCP is indeed synonymous with the maximal metabolic steady state, we hypothesized that prolonged exercise at this PO would exhibit performance and physiological response characteristics associated with the heavy-intensity domain whereas exercise above would not.
After providing written informed consent, 10 recreationally active individuals (5 men and 5 women; mean ± SD, age = 30 ± 7 yr, body mass = 67 ± 10 kg, height = 175 ± 8 cm) participated in this study, which was approved by the Conjoint Health Research Ethics Board of the University of Calgary and conformed to the standards established in the latest revision of the Declaration of Helsinki. All participants were lifelong nonsmokers and were free of any medical conditions and treatments that could interfere with metabolic and cardiorespiratory responses to exercise.
Methodological Approach to Determining the PO that Elicits the V̇O2 at RCP
The SRS protocol includes step transitions performed before and after ramp-incremental exercise to facilitate estimation of the PO that would elicit the V̇O2 at RCP and thus establish, in a single visit, the confines of the intensity domains. Specifically, the first step is performed within the moderate-intensity domain (100 W) and the second within the heavy-intensity domain (~50% POpeak). The ramp-incremental test provides the V̇O2 at the LT and RCP. The moderate step transition is used to align the V̇O2–PO relationship in the moderate domain and thus identify the PO at LT (29). Assuming V̇O2 gain increases linearly within the heavy-intensity domain (1), the PO (x), V̇O2 (y) coordinates of LT, and the heavy step transition establish the V̇O2–PO relationship in the heavy domain for constant-load exercise. Linear regression between these points gives, by extrapolation, an estimation of the PO that would be expected to elicit the V̇O2 at RCP (see Figs. 1 and 2).
Equipment and Measurements
During all exercise trials, gas exchange and ventilatory variables were measured breath by breath by metabolic cart (CPET; Cosmed, Rome, Italy). The system consisted of a facemask and low dead space turbine as well as O2 and carbon dioxide (CO2) gas analyzers; these were calibrated with a 3-L syringe and a gas mixture encompassing fractional gas concentrations of 16% O2, 5% CO2, and balance N2, respectively. At select intervals, capillary blood samples were drawn from a pinprick of the finger and immediately analyzed for blood lactate concentration ([La−]b) (Biosen C-Line; EKF Diagnostics, Barleben, Germany). All exercise was performed on an electromagnetic-braked cycle ergometer (Velotron; RacerMate, Seattle, WA).
Participants visited the laboratory on four to five occasions during which they performed: 1) an SRS protocol and 2) three to four constant-power trials. Each visit was scheduled at the same time of the day (±30 min) and at least 48 h but not more than 72 h from the previous visit. Participants self-selected their cadence (range, 75–95 rpm) and were instructed to maintain this cadence throughout the protocol and for all subsequent visits. Exercise intolerance was defined as the moment at which participants could not maintain their self-selected cadence for greater than 10 s despite strong verbal encouragement.
On the first day, participants performed the following in series: a moderate-intensity step transition, a ramp-incremental protocol, and, after a period of recovery, a heavy intensity step transition (see schematic in Fig. 1). The moderate step transition (MOD) consisted of cycling for 2 min at 20 W followed by 6 min at 100 W. The ramp-incremental protocol was initiated immediately after completion of MOD and included a 4-min baseline at 50 W and 30 W·min−1 ramp to volitional exhaustion. The heavy step transition (HVY) was performed 30 min after completion of the ramp and consisted of 12 min of cycling at 50%–65% of POpeak from the ramp. For each individual, the %POpeak was selected during the 30-min recovery period after examining the ramp data to ensure that the HVY PO would engender a V̇O2 in excess of the estimated LT.
All constant-power tests were preceded by 4 min of cycling at 50 W. On separate days, participants cycled for (i) 12 min at a PO equivalent to HVY from the SRS protocol, (ii) 30 min at the SRS-estimated RCP PO, (iii) 30 min at 5% above this PO, and in some cases (see below) (iv) 30 min at 10% above SRS-estimated RCP PO. During the 30-min trials, [La−]b was measured at 5-min intervals. In each participant, exercise responses were classified as “heavy-intensity” if participants were able to complete 30 min of exercise and there was no appreciable change between three successive time points (e.g., 20th, 25th, and 30th min) in both V̇O2 (≤120 mL·min−1 [30,31]) and [La−]b (>1 mM ), with this “steady state” being maintained until end exercise. “Severe-intensity” responses were classified as non–steady state in either V̇O2 or [La−]b (by criteria mentioned above) or an inability to complete 30 min. In cases where severe-intensity responses were not observed, an additional 30-min trial was performed at 10% above or below the SRS-predicted RCP.
Before analyses, all breath-by-breath gas exchange and ventilatory data were processed by removing aberrant breaths identified as those positioned more than 3 SD from the local mean. Thereafter, data were linearly interpolated on a second-by-second basis and bin averaged into 5-s bins (30,31).
To estimate the constant PO that elicits the V̇O2 at RCP, the following three-step approach, exemplified in Figure 2, was used:
- 1. Identification of the V̇O2 at LT and RCP. Ramp-incremental exercise provided identification in each participant of V̇O2max, LT, and RCP. The V̇O2max was considered as the highest value computed from a 20-s rolling aggregate. Interpretation of the profiles of ventilation (V̇E), gas exchange (V̇O2 and V̇CO2), respiratory exchange ratio, and end-tidal partial pressure of O2 and CO2) and their combination (V̇E/V̇CO2, V̇E/V̇O2) plotted against V̇O2 were judged independently by two investigators, after which estimates of LT and RCP were compared and a consensus reached. After ruling out potential confounding effects of CO2 storage on LT detection by confirming absence of an early change-point in RER versus V̇O2 relationship (33), the LT was determined as the V̇O2 at which V̇CO2 and V̇E began to increase disproportionately in relation to V̇O2, with a systematic rise in end-tidal PO2, whereas end-tidal PCO2 and V̇E/V̇CO2 were stable (3). RCP was determined as the V̇O2 at which end-tidal PCO2 began to fall after a period of isocapnia, corroborated by the second and first breakpoints in the V̇E- and V̇E/V̇CO2–V̇O2 relationships, respectively (17).
- 2. Determination of the MRT and PO at LT. A linear regression was used to fit the V̇O2 response (vs PO) from the onset of its systematic rise (visually determined) to the end of exercise (or onset of plateau if detected). The steady-state V̇O2 corresponding to MOD performed before the ramp was computed as the average of all breaths within the last 2 min. Thereafter, the ramp PO corresponding to the steady-state V̇O2 value from MOD was calculated using the linear V̇O2 versus PO relationship from ramp-incremental exercise. Finally, the difference in PO (W) between the ramp PO at the MOD V̇O2 and 100 W provided the “ramp-to-constant PO correction factor” (or MRT) and was calculated as follows:
where the intercept (mL·min−1) and the slope (mL·min−1· W−1) are obtained from the linear regression of the ramp V̇O2 response. This factor was subtracted from the ramp PO corresponding to LT to determine the constant PO that would elicit, in steady state, the V̇O2 at LT (29).
- 3. Estimation of the PO eliciting the V̇O2 at RCP. The steady-state V̇O2 corresponding to HVY was calculated as the average of all breaths within the last 2 min. Next, linear regression was applied to the PO, the V̇O2 (x, y) coordinates corresponding to the MRT-corrected LT, and the HVY to obtain the heavy domain slope (slopeHVY) and an intercept (interceptHVY). These parameters were then used to calculate the PO that would elicit, during constant-power exercise, the V̇O2 at RCP (i.e., SRS-predicted RCP) as follows:
A spreadsheet that demonstrates how the SRS-predicted RCP is calculated from all measured input variables is included as a supplement (see spreadsheet, Supplemental Digital Content 1, RCP power output calculation, https://links.lww.com/MSS/B957).
Data are presented as mean ± SD. A paired sample t-test was used to compare the steady-state V̇O2 between HVY trials performed within the SRS protocol and on a separate day. The V̇O2 values at RCP, the SRS-predicted RCP, and at 5% above SRS-predicted RCP were compared by repeated-measures ANOVA, and their agreement was assessed by Bland–Altman analyses (with 95% limits of agreement); these analyses were also used to compare the end-trial [La−]b during constant-power exercise at the SRS-predicted RCP and at 5% above SRS-predicted RCP. Coefficient of variation (CV) and Lin’s concordant coefficient (CCC) were used to evaluate the variance and the concordance, respectively, between the ramp-identified V̇O2 at RCP and the steady-state V̇O2 from constant-power exercise at the SRS-predicted RCP. The CCC was interpreted as follows: almost perfect agreement = CCC > 0.99, substantial agreement = 0.99 < CCC > 0.95, moderate agreement = 0.95 < CCC > 0.90, and poor agreement = CCC < 0.90 (34). A V̇O2 of ±120 mL·min−1 was considered as the minimally important difference for steady-state exercise (30)—i.e., a difference less than this value is within the range of measurement variability and, within an individual, is not considered significant. Statistical significance was set at an α level <0.05.
Table 1 displays the measured parameters and variables during the MOD, ramp, and HVY portions of the “SRS” protocol, and exemplar V̇O2 data from a representative participant are depicted in Figure 2. The mean V̇O2max was 3653 ± 643 mL·min−1; LT and RCP occurred at 64% ± 3% and 85% ± 3% of this value, respectively. [La−]b immediately after ramp exercise was 10.0 ± 3.0 mmol·L−1. On average, the MOD and HVY rides corresponded to 28% ± 6% and 52% ± 5% of POpeak, respectively, representing, in MOD, 64% ± 13% of the PO at LT and, in HVY, 84% ± 6% of the PO associated with the SRS-estimated RCP. These PO engendered steady-state V̇O2 approximating 47% ± 8% (MOD) and 73% ± 4% (HVY) of V̇O2max. During HVY, the V̇O2 at 12 min (2690 ± 510; range, 3410 to 1880 mL·min−1) was different (P = 0.048) from that at 6 min (2670 ± 520; delta range, 80 to −20 mL·min−1) but not at 8 min (2680 ± 510; range, 50 to −40 mL·min−1, P = 0.484). Compared with the HVY performed during the SRS protocol (i.e., 30 min after ramp exercise), the steady-state V̇O2 from the HVY completed on a separate day was not different (bias = −2 mL·min−1; LOA = −150–140 mL·min−1; CV = 1.4% ± 1.1%; P = 0.930; see also Table 1).
There was no difference (P = 0.129) and high intraparticipant agreement between the ramp-identified V̇O2 at RCP and the steady-state V̇O2 measured during constant-power exercise at the SRS-estimated RCP (bias = −26 mL·min−1, r = 0.99, CCC = 0.99 [0.95–1.00], CV = 1.5% ± 1.4%; see Table 1 and Fig. 3). In 8 of 10 participants, the difference between measured and SRS-predicted V̇O2 was less than 80 mL·min−1 and below the limits of detectable difference (i.e., ≤120 mL·min−1). All participants completed 30 min of constant-power exercise at the SRS-estimated RCP achieving stable metabolic responses with end-exercise V̇O2 corresponding to 86% ± 4% V̇O2max (see Table 1) and [La−]b reaching 5.4 ± 1.8 mmol·L−1. With exercise at 5% above this PO, 8 of 10 exhibited response profiles for V̇O2 and [La−]b that were indicative of severe-intensity exercise. Of these, three participants failed to complete 30 min with V̇O2 approximating V̇O2max at end exercise (range, 97%–99%V̇O2max) and both V̇O2 and [La−]b not attaining steady state, and five participants completed 30 min with [La−]b failing to attain a steady state and V̇O2 achieving steady state in some (n = 2) but not all; end-exercise V̇O2 ranged from 90% to 97% V̇O2max. A representation of the response pattern is displayed along with the group mean in Figure 4. Two participants completed 30 min of exercise at 5% above the SRS-predicted RCP achieving steady state in both V̇O2 and [La−]b. However, both individuals were unable to complete 30 min of exercise during an additional constant-power bout set a 10% above the SRS-predicted RCP. Group mean end-exercise V̇O2 and [La−]b achieved at 5% above the SRS-estimated RCP were 3462 ± 633 mL·min−1 (95% ± 3% of V̇O2max) and 8.5 ± 2.2 mmol·L−1, respectively. [La−]b values during this trial were, on average, not different from the peak [La−]b recorded at the end of the ramp exercise (P = 0.14).
A divergence exists in the V̇O2–PO relationships derived from ramp versus constant-power exercise paradigms. This disparity, which is exacerbated at higher %V̇O2max, has precluded the appropriate application of ramp-incremental protocols and the RCP concept for the determination of the constant-power associated with maximal metabolic steady state. To overcome this challenge, we designed and tested a simple SRS protocol that enables prediction of the PO required to recreate, during constant-power exercise, the metabolic rate (i.e., V̇O2) associated with the RCP. Novel findings in a group of healthy young individuals are that i) constant-power exercise at the SRS-predicted RCP elicits, with a high degree of intraindividual accuracy and in steady state, the ramp-identified V̇O2 at RCP; and ii) constant-power exercise at 5% above the SRS-predicted RCP elicits, in the majority of the participants, non–steady-state V̇O2 and [La−]b response profiles consistent with severe-intensity domain exercise. These findings support the validity of the SRS protocol for establishing, in a single visit, the maximal metabolic steady state and the suitability of the RCP as a surrogate demarcator of the heavy- to severe-intensity boundary.
The SRS protocol is designed specifically to estimate the constant-power required to elicit the V̇O2 associated with RCP and, in doing so, identify the maximal metabolic steady state. On average, constant-power exercise at the SRS-estimated RCP yielded a steady-state V̇O2 that was within 30 mL·min−1 of the ramp-identified V̇O2 target (see Fig. 3). In 8 of 10 participants, the difference between estimated and measured was less than 80 mL·min−1 and well below the limits of detectable difference (i.e., ±120 mL·min−1) for steady-state V̇O2. The high level of agreement between estimated and measured V̇O2 (bias = −26 mL·min−1, r = 0.99) also supports the assumption that the V̇O2 versus PO relationship within the heavy-intensity domain is linear (1) and, importantly, suggests that this test could be used to predict the constant power for any V̇O2 within the heavy-intensity domain. In two participants, the SRS-predicted RCP yielded V̇O2 values that were 120 and 240 mL·min−1 less than intended. Although further investigation of these individuals was not explored, we suspect that an overestimation of LT might explain the difference. Overestimations in LT will lead to underestimations of the SRS-predicted power for any target V̇O2; the converse is true for underestimations of LT. Importantly, these individuals do not undermine the validity of the approach but highlight the need for correctness in the estimation of LT to maintain its exactness.
The high accuracy of the SRS prediction model enabled us to test, for the first time, whether the V̇O2 at RCP identifies the maximal metabolic steady state, a contentious hypothesis debated in a recent Medicine and Science in Sports and Exercise contrasting perspectives (4,28). Based on the consistent finding that the V̇O2 at RCP is remarkably constant within an individual (13,20,21), our hypothesis was that the RCP signifies a consistent physiological response initiated by a fixed metabolic rate (4), and thus, a discrete PO defining the RCP must exist within the constant-power paradigm. Consistent with this hypothesis, we observed that prolonged exercise at the SRS-derived PO associated with RCP exhibited performance and physiological response characteristics compatible with heavy-intensity domain, whereas exercise at 5% above this PO did not (see Fig. 4). Indeed, all participants were able to complete 30 min of exercise at their SRS-predicted RCP, with both V̇O2 and [La−]b achieving steady state. By contrast, with constant-power exercise at 5% above this intensity, 8 of 10 participants exhibited physiological response profiles indicative of severe-intensity domain exercise (8,35–37). Of the two participants able to attain a steady state above the SRS-predicted RCP, additional 30 min constant-power test set at 10% above the SRS-predicted RCP indicated that, for the first participant, the V̇O2 corresponding to RCP was correct but the V̇O2 at LT may have been slightly overestimated, whereas, in the second, the V̇O2 at RCP may have been slightly underestimated. Concurrently, it is also possible that the V̇O2 at the maximal metabolic steady state may itself vary on a day-to-day basis as suggested previously (38,39). Nevertheless, our findings support an equivalence between the maximal metabolic steady state and the RCP of ramp-incremental exercise for the majority of the participants. Even in the two cases where a disagreement existed, the difference was within 5% of the expected value, which is the typical error commonly accepted for other measures of critical intensities of exercise (39).
The strengths of the SRS protocol are that it is relatively simple to execute, its variables are easy to identify and measure, the estimation of RCP PO involves straightforward computations, it is not dependent on ramp slope, and, thus, it can be easily adopted by anyone who uses ramp-incremental exercise. Provided all identified (LT, RCP) and measured (MOD, HVY) variables are correct, the SRS-predicted RCP should estimate very closely the constant power that elicits the V̇O2 at RCP. That said, some sources of potential error that might affect the accuracy of SRS-predicted RCP should be considered. First, the MOD PO, decided a priori, must elicit a V̇O2 that is above the ramp-incremental baseline V̇O2 but below LT. In healthy young individuals, 100 W would, in most cases, satisfy these requirements but may not in those less conditioned. In such cases, lowering both the ramp baseline (e.g., to 20 W) and the MOD PO (e.g., to 55 W) may be advised. Second, the PO at HVY must elicit a V̇O2 that is between LT and RCP. In young healthy populations who performed 25–30 W·min−1 ramp-incremental protocols, such a V̇O2 can be obtained with >90% probability at 50% POpeak (2). Conversely, within a 30-min recovery period, a quick scan of the ramp data could facilitate a more personalized estimate (e.g., 30% of the difference between ramp power at LT and V̇O2max or 50% between ramp powers at LT and RCP). Furthermore, we show that the steady-state V̇O2 at this intensity is not affected when performed 30 min after ramp-incremental exercise. Third, as described above, LT and RCP must be identified correctly. Lastly, the length of the protocol may be perceived as an issue particularly in clinical scenarios where time and resources are constrained. That said, in its entirety, the SRS protocol did not in any participant exceed 75 min and involved 30–40 min of total exercise time or 22 additional min of exercise time compared with contemporary ramp-incremental protocols. Reductions in postramp recovery time (e.g., from 30 to 15 min) and in the HVY duration (requisite steady-state V̇O2 was attained by 6–8 min in most participants) are possible targets to reduce total protocol time (e.g., by 20 min).
Although largely unappreciated, the limitations of ramp-incremental exercise protocols for exercise prescription are well known (1,13,25,27,40). Yet they remain, as routine practice for laboratories worldwide, the standard for exercise intensity assignment in research and practical settings. Applying simple measurements and computations, the SRS protocol overcomes the exercise prescription limitations of traditional ramp protocols by facilitating precise determination of the PO associated with the LT and the maximal metabolic steady state. Thus, within a single visit, one can discern with accuracy the PO associated with the two thresholds that partition the three exercise intensity domains. Although further testing is required, these findings indicate that this method could correctly predict the PO for any V̇O2 within the moderate- and heavy-intensity domains and provide unprecedented accuracy for aerobic exercise prescription (see spreadsheet, Supplemental Digital Content 1, RCP power output calculation, for specific details and calculations, https://links.lww.com/MSS/B957). Unlike other similarly intended single visit protocols that incorporate maximal sprint bouts postramp exercise (41,42), the SRS requires only submaximal efforts before and after ramp exercise, making it potentially more suitable for a wide range of populations that are less capable of reproducing maximal efforts or in whom such efforts are contraindicated. Such a protocol could assist in circumstances where accuracy in the prescription of exercise intensity is required.
Here we demonstrate that by determining the constant-power equivalent of the RCP using the SRS protocol, one can identify with a reasonable degree of accuracy the PO associated with the maximal metabolic steady state. Although further testing of this concept is required, these findings suggest that this protocol could be applied to determine the constant PO required to elicit any steady-state V̇O2 within the moderate- and heavy-intensity domains.
This study was supported by grants given to Juan M. Murias by the Natural Sciences and Engineering Research Council of Canada (RGPIN-2016-03698) and the Heart and Stroke Foundation of Canada (1047725). Daniel A. Keir was supported by a postdoctoral fellowship from the Canadian Institute for Health Research (CIHR).
None conflicts of interest, financial or otherwise, are declared by the authors.
The results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Keir DA, Paterson DH, Kowalchuk JM, Murias JM. Using ramp-incremental V̇O2
responses for constant-intensity exercise selection. Appl Physiol Nutr Metab
2. Iannetta D, Inglis EC, Mattu AT, et al. A critical evaluation of current methods for exercise prescription in women and men. Med Sci Sport Exerc
3. Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol
4. 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 Sport Exerc
5. Whipp BJ. Domains of aerobic function and their limiting parameters. In: Steinacker J, Ward S, editors. The Physiology and Pathophysiology of Exercise Tolerance
. New York: Plenum Press; 1996. pp. 83–9.
6. Whipp BJ, Mahler M. Dynamics of pulmonary gas exchange during exercise. In: West JB, editor. Pulmonary Gas Exchange, Vol II, Organism and Environment
. New York: Academic Press; 1980. pp. 33–96.
7. Davis JA, Whipp BJ, Lamarra N, Huntsman DJ, Frank MH, Wasserman K. Effect of ramp slope on determination of aerobic parameters from the ramp exercise test. Med Sci Sports Exerc
8. Poole DC, Ward SA, Gardner GW, Whipp BJ. Metabolic and respiratory profile of the upper limit for prolonged exercise in man. Ergonomics
9. Jones AM, Wilkerson DP, DiMenna F, Fulford J, Poole DC. Muscle metabolic responses to exercise above and below the “critical power
” assessed using 31P-MRS. Am J Physiol Regul Integr Comp Physiol
10. Keir DA, Fontana FY, Robertson TC, et al. Exercise intensity thresholds: identifying the boundaries of sustainable performance. Med Sci Sports Exerc
11. Mattioni Maturana F, Keir DA, Mclay KM, Murias JM. Can measures of critical power
precisely estimate the maximal metabolic steady state? Appl Physiol Nutr Metab
12. Iannetta D, Fontana FY, Maturana FM, et al. An equation to predict the maximal lactate steady state from ramp-incremental exercise test data in cycling. J Sci Med Sport
13. Iannetta D, de Almeida Azevedo R, Keir DA, Murias JM. Establishing the V̇O2
versus constant-work rate relationship from ramp-incremental exercise: simple strategies for an unsolved problem. J Appl Physiol
14. Inglis EC, Iannetta D, Keir DA, Murias JM. Training-induced changes in the respiratory compensation point, deoxyhemoglobin break point, and maximal lactate steady state: evidence of equivalence. Int J Sports Physiol Perform
15. Meyer T, Faude O, Scharhag J, Urhausen A, Kindermann W. Is lactic acidosis a cause of exercise induced hyperventilation at the respiratory compensation point? Br J Sports Med
16. Wasserman K, Beaver WL, Sun X-G, Stringer WW. Arterial H(+) regulation during exercise in humans. Respir Physiol Neurobiol
17. Whipp BJ, Davis JA, Wasserman K. Ventilatory control of the “isocapnic buffering” region in rapidly-incremental exercise. Respir Physiol
18. Iannetta D, Qahtani A, Mattioni Maturana F, Murias JM. The near-infrared spectroscopy-derived deoxygenated haemoglobin breaking-point is a repeatable measure that demarcates exercise intensity domains. J Sci Med Sport
19. Mattu AT, Iannetta D, MacInnis MJ, Doyle-Baker PK, Murias JM. Menstrual and oral contraceptive cycle phases do not affect submaximal and maximal exercise responses. Scand J Med Sci Sports
20. Leo JA, Sabapathy S, Simmonds MJ, Cross TJ. The respiratory compensation point is not a valid surrogate for critical power
. Med Sci Sport Exerc
21. Scheuermann BW, Kowalchuk JM. Attenuated respiratory compensation during rapidly incremented ramp exercise. Respir Physiol
22. Boone J, Bourgois J. The oxygen uptake response to incremental ramp exercise: methodogical and physiological issues. Sports Med
23. Spencer MD, Murias JM, Kowalchuk JM, Paterson DH. Effect of moderate-intensity work rate increment on phase II τVO2
, functional gain and Δ [HHb]. Eur J Appl Physiol
24. Keir DA, Robertson TC, Benson AP, Rossiter HB, Kowalchuk JM. The influence of metabolic and circulatory heterogeneity on the expression of pulmonary V̇O2
kinetics in humans. Exp Physiol
25. DiMenna FJ, Jones AM. “Linear” versus “nonlinear” V̇O2
responses to exercise: reshaping traditional beliefs. J Exerc Sci Fit
26. Keir DA, Benson AP, Love LK, Robertson TC, Rossiter HB, Kowalchuk JM. Influence of muscle metabolic heterogeneity in determining the V̇O2
p kinetic response to ramp-incremental exercise. J Appl Physiol
27. Rossiter HB. Exercise: kinetic consideration for gas exchange. Compr Physiol
28. Broxterman RM, Craig JC, Richardson RS. The respiratory compensation point and the deoxygenation break point are not valid surrogates for critical power
and maximum lactate steady state. Med Sci Sports Exerc
29. Iannetta D, Murias JM, Keir DA. A simple method to quantify the V˙O2
mean response time of ramp-incremental exercise. Med Sci Sport Exerc
30. Keir DA, Murias JM, Paterson DH, Kowalchuk JM. Breath-by-breath pulmonary O2
uptake kinetics: effect of data processing on confidence in estimating model parameters. Exp Physiol
31. Lamarra N, Whipp BJ, Ward SA, Wasserman K. Effect of interbreath fluctuations on characterizing exercise gas exchange kinetics. J Appl Physiol
32. Heck H, Mader A, Hess G, Mucke S, Muller R, Hollmann W. Justification of the 4-mmol/l lactate threshold. Int J Sports Med
33. Whipp BJ. Physiological mechanisms dissociating pulmonary CO2
exchange dynamics during exercise in humans. Exp Physiol
34. McBride G, Bland JM, Altman DG, Lin LI. A proposal for strength-of-agreement criteria for Lin’s concordance correlation coefficient. NIWA Client Rep
35. Murgatroyd SR, Ferguson C, Ward SA, Whipp BJ, Rossiter HB. Pulmonary O2
uptake kinetics as a determinant of high-intensity exercise tolerance in humans. J Appl Physiol
36. Keir DA, Copithorne DB, Hodgson MD, Pogliaghi S, Rice CL, Kowalchuk JM. The slow component of pulmonary O2 uptake accompanies peripheral muscle fatigue during high-intensity exercise. J Appl Physiol
37. Sawyer BJ, Morton RH, Womack CJ, Gaesser GA. V̇O2max
may not be reached during exercise to exhaustion above critical power
. Med Sci Sports Exerc
38. Jones AM, Burnley M, Black MI, Poole DC, Vanhatalo A. The maximal metabolic steady state: redefining the ‘gold standard.’ Physiol Rep
39. Poole DC, Burnley M, Vanhatalo A, Rossiter HB, Jones AM. Critical power
: an important fatigue threshold in exercise physiology. Med Sci Sport Exerc
40. Mezzani A, Hamm LF, Jones AM, et al. Aerobic exercise intensity assessment and prescription in cardiac rehabilitation. J Cardiopulm Rehabil Prev
41. Murgatroyd SR, Wylde LA, Cannon DT, Ward SA, Rossiter HB. A “ramp-sprint” protocol to characterise indices of aerobic function and exercise intensity domains in a single laboratory test. Eur J Appl Physiol
42. Constantini K, Sabapathy S, Cross TJ. A single-session testing protocol to determine critical power
and W. Eur J Appl Physiol