Influence of Pedal Cadence on the Respiratory Compensation Point and Its Relation to Critical Power
Broxterman et al. asked whether the respiratory compensation point (RCP) is a distinct work rate (W) or a distinct metabolic rate (VO2), and whether RCP is mechanistically related to critical power (CP). Ten collegiate male athletes (n=10) performed incremental and constant-power cycling tests at 60 and 100 rpm.
RCP work rate was significantly lower at 100 than at 60 rpm (197 ± 24 W vs 222 ± 24 W, p ≤ 0.05) while RCP VO2 was not different (3.00 ± 0.33 vs 3.12 ± 0.41 L/min).
CP at 60 rpm (214 ± 51 W; VO2 3.01 ± 0.69 L/min) and at 100 rpm (196 ± 46 W; VO2 2.95 ± 0.54 L/min) did not differ significantly from RCP as group means — but RCP and CP were not significantly correlated across individuals.
RCP is therefore a distinct metabolic rate that can be reached at different power outputs, and RCP and CP are not equivalent parameters and should not be used synonymously.
Participants
n=10 collegiate male athletes
Protocol
Within-subject comparison of two pedal cadences (from the abstract): each athlete performed incremental and constant-power cycling tests at 60 and 100 rpm to determine RCP and CP at each cadence.
Outcomes: RCP work rate and VO2, CP and its VO2; RCP and CP compared at each cadence and correlated across individuals.
Key Results (from the abstract)
| Parameter | 60 rpm | 100 rpm | Difference |
|---|---|---|---|
| RCP power | 222 ± 24 W | 197 ± 24 W | lower at 100 rpm, p ≤ 0.05 (25 W) |
| RCP VO2 | 3.12 ± 0.41 L/min | 3.00 ± 0.33 L/min | not significant |
| CP power | 214 ± 51 W | 196 ± 46 W | not significantly different from RCP at either cadence |
| CP VO2 | 3.01 ± 0.69 L/min | 2.95 ± 0.54 L/min | — |
| RCP vs CP across individuals | not significantly correlated | not significantly correlated | — |
Mechanisms
RCP tracks metabolic rate: the same VO2 at RCP was reached at 25 W less external power at 100 rpm. Inference, not tested here: the extra internal work of moving the legs faster is the usual explanation for a higher VO2 per external watt at high cadence; the abstract does not report internal work.
CP as a group mean also came out lower at 100 rpm (214 → 196 W), i.e. the abstract does NOT show CP to be cadence-independent; it shows CP and RCP means to coincide at each cadence while individuals' CP and RCP do not track each other.
Not correlated ≠ not equal on average: two markers can share a mean and still be poor predictors of one another for a given athlete — which is the paper's argument against using RCP as a CP proxy.
Practical Application
In 10 collegiate athletes, the respiratory compensation point was a fixed metabolic rate (VO2 3.00-3.12 L/min) but not a fixed power: it fell from 222 ± 24 W at 60 rpm to 197 ± 24 W at 100 rpm; critical power (214 ± 51 and 196 ± 46 W) matched RCP as a group mean at each cadence, yet RCP and CP were not correlated across individuals, so one cannot stand in for the other (within-subject study, Broxterman 2015).
Do not use RCP (VT2) as a CP proxy or vice versa for an individual: the group means agree, the individuals do not. Derive CP from power-duration data (REF-168, REF-024) and RCP from gas exchange, and keep them as separate markers.
Cadence changes the watts at which a ventilatory marker appears (25 W between 60 and 100 rpm here): a lab ramp done at an unusual cadence gives an RCP wattage that will not match the road. Test at, and prescribe for, the athlete's habitual cadence.
Each marker in its own currency: RCP is a metabolic rate (report and track it as VO2 or as HR if VO2 is unavailable); CP is mechanical (prescribe and track it in watts).
Not from this paper; the base's own practice: a preferred-cadence window (roughly 75-90 rpm) is used for testing; the study tested only 60 and 100 rpm.
Limitations
From the design as the abstract reports it: n=10 collegiate men only; only two cadences (60 and 100 rpm) were tested.
Laboratory ergometer tests with no performance outcome.
Small sample: both the non-significant difference between RCP and CP means and the non-significant correlation between them come from ten participants.
How this entry was checked and changed
Why this level
(mechanistic, small sample)
Content correction 2026-09-27
"CP power: stable / independent of cadence" -> abstract: CP 214 ± 51 W at 60 rpm vs 196 ± 46 W at 100 rpm, described only as not significantly different from RCP; cadence-independence is not claimed. CP values and VO2 added.
"RCP ≈ CP? No consistent relationship" -> abstract: group means not significantly different at either cadence, but RCP and CP not significantly correlated; reworded.
n=10 collegiate male athletes added (was missing); Tier "N/A" -> Tier 2; Sport "cycling, swimming, sprint" -> cycling; internal-work explanation and 75-90 rpm window relabelled inference / base practice.
Cross-references to duplicate stubs repointed to their canonical entries (REF-200->REF-024); the stubs were emptied on 2026-09-27.
2026-09-28: added Protocol and Limitations from the abstract (Europe PMC PMID 25523595).