REF-120 · Burnley, Doust & Vanhatalo, 2006 · Medicine & Science in Sports & Exercise, Vol. · Paper · PubMed
A 3-min All-Out Test to Determine Peak Oxygen Uptake and the Maximal Steady State
Participants
n=11 habitually active subjects (the abstract gives no VO2max or training history beyond "habitually active").
Protocol
Ramp test; three 3-min all-out cycling tests against a fixed resistance; end-test power = mean power of the last 30 s
Two further submaximal tests lasting up to 30 min at 15 W below and 15 W above the end-test power, with blood lactate and VO2 profiles
Key Results
> ⚠️ CLAIM CORRECTED 2026-08-10. This section previously read "VO2peak validity: r>0.85 vs maximal graded protocol" and "Accuracy: ±5% CP estimation". Neither figure appears in Burnley 2006 — both trace to [REF-039], an entry withdrawn the same day because the paper it cited does not exist. They are removed and replaced with what this study actually reports.
VO2peak in the 3-min all-out test was not different from the ramp test: 3.78 ± 0.68 vs 3.84 ± 0.79 L·min⁻¹ (P = 0.75). That is an equivalence result, not a correlation coefficient.
End-test power 257 ± 49 W — significantly lower than ramp-test peak power (368 ± 73 W) and significantly higher than power at the gas exchange threshold (169 ± 55 W; P < 0.001)
15 W below end-test power: 9 of 11 subjects completed 30 min; 7 of those with a steady-state blood lactate and VO2 profile
15 W above end-test power: blood lactate and VO2 rose inexorably until exhaustion at approximately 13 ± 7 min
End-test power estimates the maximal steady state — this is the paper's central claim and the basis of the protocol's use
The study reports no percentage accuracy for critical-power estimation; any ±% figure attached to this protocol needs its own source
Practical Application
In n=11 habitually active subjects, the end-test power of a 3-min all-out cycling test (257 W) sat between the gas exchange threshold (169 W) and ramp peak (368 W); 15 W below it, 9 of 11 rode 30 min (7 in steady state), while 15 W above it lactate and VO2 rose until exhaustion at ~13 min; the test's VO2peak matched a ramp test (3.78 vs 3.84 L/min, P = 0.75) — one 3-min all-out effort gives both a VO2peak and a usable estimate of the maximal steady state.
Use the 3-minute all-out test as a field CP assessment when laboratory testing is unavailable — baseline, mid-plan, post-plan. Protocol: warm-up, 3 min genuinely all-out from the first second, end-test power (last 30 s) as the CP estimate.
The effort must be all-out from the start. A paced attempt invalidates the test and overestimates the end-test power relative to a true all-out effort (inference from the test's logic; the abstract does not quantify pacing error).
Do not quote a precision figure for this test. The evidence supports "estimates the maximal steady state", not "within ±5%" — and threshold markers disagree with each other by ±22 to ±52 W in the sourced literature (REF-732).
Keep the marker consistent across a season: a number derived this way should be compared with itself, not with a lactate-derived threshold (REF-732).
Limitations
n = 11, habitually active (not trained athletes), single laboratory, 2006.
The ±15 W margin is coarse: 2 of 11 could not complete 30 min even 15 W below the end-test power, and only 7 showed a true steady state.
Validity was established against a ramp test and the maximal steady state in the laboratory; field execution adds pacing error, terrain and motivation on top.
The test is maximal and unpleasant; repeatability depends on the athlete's willingness to fully commit, which varies between attempts.
Cycling only; no equivalent validation is imported here for running or other modalities.
This entry carried unsourced figures for an unknown period. Numeric claims are now taken from the abstract; the full text has not been re-verified.
How this entry was checked and changed
Content correction 2026-09-27
Bold title "Assessment of Critical Torque via 3-Minute All-Out Test" -> the paper's title is "A 3-min All-Out Test to Determine Peak Oxygen Uptake and the Maximal Steady State" (critical torque is a different study).
"Trained subjects", Tier 2-3 -> habitually active subjects, Tier 1. "Self-selected variable pace" -> all-out against a fixed resistance.
Added the abstract's figures: end-test power 257 W vs GET 169 W and ramp peak 368 W; 9/11 completed 30 min at −15 W (7 steady state); exhaustion at ~13 min at +15 W.
Cross-references to duplicate stubs repointed to their canonical entries (REF-129->REF-044); the stubs were emptied on 2026-09-27.