Place · Level 3
VO2max — endurance ceiling
心肺天花板 · 训练能涨 15-25% · Mandsager 2018 JAMA: 高 VO2max 死亡率 ↓ 5x · 老年下降率 10%/十年
Last updated
Story path
Chapter 1
What is VO2max
What is VO2max
VO2max is maximum oxygen consumption per unit time (mL O₂ / kg / min) — the ceiling of the aerobic system. It's set by 4 serial links: lung gas exchange → cardiac output → blood transport → muscle extraction (this chain is real physiology).
Average adults: men 35-45 / women 30-40Endurance athletes: men 70-85 / women 60-75Extreme talents (XC ski champions): 90+
60-80% is genetically determined (HERITAGE family study, Bouchard 1999); training only moves you 15-25%. An average person starting at 35 may push to 45-50 with extreme training, but will never reach 70.
Mandsager 2018 *JAMA Network Open* (n = 122,007 cardiac stress-test patients): high VO2max (>40 men / >32 women) vs low (<15) → all-cause mortality ↓ 5×, larger than quitting smoking, glycemic control, or blood-pressure control. VO2max is currently one of the strongest modifiable predictors of longevity.
Average adults: men 35-45 / women 30-40Endurance athletes: men 70-85 / women 60-75Extreme talents (XC ski champions): 90+
60-80% is genetically determined (HERITAGE family study, Bouchard 1999); training only moves you 15-25%. An average person starting at 35 may push to 45-50 with extreme training, but will never reach 70.
Mandsager 2018 *JAMA Network Open* (n = 122,007 cardiac stress-test patients): high VO2max (>40 men / >32 women) vs low (<15) → all-cause mortality ↓ 5×, larger than quitting smoking, glycemic control, or blood-pressure control. VO2max is currently one of the strongest modifiable predictors of longevity.
Mechanism · which relay leg is the weak one
The relay metaphor only helps once you know which leg is weakest. The answer is settled and counter-intuitive: in healthy people VO2max is limited not by the muscle's ability to extract oxygen but by the ability to deliver it — specifically, how much blood the heart ejects per beat.Take the number apart first: VO2max = maximal cardiac output x arteriovenous oxygen difference. The first is delivery, the second extraction. And cardiac output = stroke volume x heart rate — with maximal heart rate fairly fixed (and falling with age), what actually varies is stroke volume. The review puts cardiac output at 70-85% of the between-person variance in VO2max.
How do we know it is the delivery end? Three lines of reversal experiment:
Move delivery and VO2max moves: blood reinfusion raises oxygen-carrying capacity and VO2max rises; altitude hypoxia lowers it; beta-blockade caps heart rate, cutting cardiac output, and it falls. All three act on delivery.Overperfuse one small muscle instead: restrict work to single-leg knee extension so the whole circulation serves that one muscle, and its oxygen consumption per unit mass is enormous — far above what it shows in whole-body exercise. Extraction capacity is in surplus; it goes unused because the blood cannot get there.What rises with training: the gain comes mainly from higher maximal cardiac output, not a widened a-vO2 difference.
So which leg does 'training raises it 15-25%' refer to? Mostly the heart — a larger ventricular chamber, higher stroke volume, expanded plasma volume. Which also explains why the ceiling differs so much between people: cardiac plasticity differs to begin with.
The muscle leg is not useless. Mitochondrial density and capillarisation decide how long you can hold what intensity — threshold and endurance performance — not the maximum itself. The two get conflated: VO2max is the ceiling; the threshold decides how close under it you can live, and for how long.
Chapter 2
Training response
Training response
Training gain depends on baseline, genetics, and protocol:
Sedentary populations: +15-25% after 12 weeks of moderate-to-high intensity trainingAlready-trained individuals: +5-10% with the same 12 weeksElites: +1-3% with full-year training'Non-responders': about 15-20% of trained individuals have almost no VO2max change (Bouchard 2011) — genetically determined
HIIT lets you reach the plateau faster but can't exceed it. Short-term high intensity reaches plateau faster than long slow distance, but the long-term ceiling is set by genetics, not the training protocol.
Age-related decline: typical aging ~-10%/decade (Fleg 2005); consistently trained individuals ~-5%/decade; sedentary individuals ~-15%/decade. 'The earlier the better' is mostly about preventing decline, not about unlimited gains.
Sedentary populations: +15-25% after 12 weeks of moderate-to-high intensity trainingAlready-trained individuals: +5-10% with the same 12 weeksElites: +1-3% with full-year training'Non-responders': about 15-20% of trained individuals have almost no VO2max change (Bouchard 2011) — genetically determined
HIIT lets you reach the plateau faster but can't exceed it. Short-term high intensity reaches plateau faster than long slow distance, but the long-term ceiling is set by genetics, not the training protocol.
Age-related decline: typical aging ~-10%/decade (Fleg 2005); consistently trained individuals ~-5%/decade; sedentary individuals ~-15%/decade. 'The earlier the better' is mostly about preventing decline, not about unlimited gains.
Measurement: lab vs field
Direct measurement: VO2max test (treadmill + mask + gas analyser) is the only precise method; lab cost ¥800-2000 in China.Field estimates (in decreasing accuracy):
Cooper test (12-min run): VO2max ≈ (distance in m - 504.9) / 44.731.5-mile run: VO2max ≈ 483 / time + 3.5Resting heart rate: VO2max ≈ 15.3 × (HRmax / HRrest)Apple Watch / Garmin: derived from HR and gait data, ±15% error, useful only for trend
Practically, most people don't need a precise number. 'Six months ago at the same pace over the same distance, HR was 5-10 bpm higher' is a more reliable trend signal.
Chapter 3
Practical implications
Practical implications
Prescription for non-athletes (based on Mandsager 2018):
Phase 1 (first 6 months): 3 sessions/week × 30 min moderate intensity (talk-test pace), VO2max +10-15%, mortality risk already markedly lowerPhase 2 (6-24 months): add Zone 2 work + 1 HIIT session/week, cumulative VO2max +5-10%Maintenance: 150 min moderate or 75 min vigorous per week (ACSM standard)
Two things to avoid:
Blindly chasing the VO2max number — 35 → 45 already captures most of the health benefit; 45 → 55 has steeply diminishing marginal returns'Competitive-izing' VO2max — for non-athletes, a higher number isn't always higher quality of life; balance training time, body weight, and joint load
Cross-cutting links: cardiovascular hypertension type-2-diabetes (all cardio-related conditions use VO2max as a risk / response surrogate).
Phase 1 (first 6 months): 3 sessions/week × 30 min moderate intensity (talk-test pace), VO2max +10-15%, mortality risk already markedly lowerPhase 2 (6-24 months): add Zone 2 work + 1 HIIT session/week, cumulative VO2max +5-10%Maintenance: 150 min moderate or 75 min vigorous per week (ACSM standard)
Two things to avoid:
Blindly chasing the VO2max number — 35 → 45 already captures most of the health benefit; 45 → 55 has steeply diminishing marginal returns'Competitive-izing' VO2max — for non-athletes, a higher number isn't always higher quality of life; balance training time, body weight, and joint load
Cross-cutting links: cardiovascular hypertension type-2-diabetes (all cardio-related conditions use VO2max as a risk / response surrogate).
Chapter 4
Max-HR formula & zones
Max-HR formula & zones
If you want to train by heart-rate zones, the first question is 'what's my max HR?' Gyms and fitness tests usually hand you '220 - age'. That formula comes from a 1971 Fox review article, not a primary study — Fox eyeballed a regression across 10 independent HR studies, with no original data analysis.
Its problems, point by point:
Standard deviation around ±12 bpm — even if the median is right, any individual may be 12+ bpm above or below the predictionThe sample was mostly 20-40-year-old men, with no account for sex, training status, or medicationIt systematically underestimates true max HR in older adults and elite athletes
Tanaka 2001 (JACC, 351 studies, n=18,712) re-ran the regression to get 208 - 0.7 × age: larger sample, covering ages 15-90, with SD ±10 bpm — somewhat more accurate, and notably better for older adults. A 70-year-old gets 150 from the old formula vs 159 from Tanaka, a gap of 9 — so the old formula sets the intensity zones too low for seniors and they end up under-training.
The most accurate route is still direct measurement: an 8-12 minute progressive max-effort test, where peak HR is your true max HR with only ±2-3 bpm error. Hospital cardiopulmonary exercise tests (CPET) never use a formula — they measure directly.
Its problems, point by point:
Standard deviation around ±12 bpm — even if the median is right, any individual may be 12+ bpm above or below the predictionThe sample was mostly 20-40-year-old men, with no account for sex, training status, or medicationIt systematically underestimates true max HR in older adults and elite athletes
Tanaka 2001 (JACC, 351 studies, n=18,712) re-ran the regression to get 208 - 0.7 × age: larger sample, covering ages 15-90, with SD ±10 bpm — somewhat more accurate, and notably better for older adults. A 70-year-old gets 150 from the old formula vs 159 from Tanaka, a gap of 9 — so the old formula sets the intensity zones too low for seniors and they end up under-training.
The most accurate route is still direct measurement: an 8-12 minute progressive max-effort test, where peak HR is your true max HR with only ±2-3 bpm error. Hospital cardiopulmonary exercise tests (CPET) never use a formula — they measure directly.
RPE beats heart rate
There's a more reliable tool in practice: Borg's RPE (rate of perceived exertion) scale, 1-10 revised version, in use for 50 years.Rough mapping:
RPE 3-4 (easy, can hold a full conversation) ≈ 60-70% max HR, i.e. Zone 2RPE 5-6 (moderate, only short phrases) ≈ 70-80%, Zone 3RPE 7-8 (hard, single words) ≈ 80-90%, around lactate thresholdRPE 9-10 (max, can't speak) ≈ 90-100%, the VO2max zone
Why RPE is more reliable than HR in practice: heart rate is contaminated by a long list of variables — caffeine +5-10 bpm, dehydration +5-15, heat +10-20, sleep deprivation +5, excitement +5-10 — and HR lags 60-90 seconds, so during intervals it reflects the previous segment, not the current one. RPE is instantaneous and integrates all of these automatically.
Some people can't use any HR formula at all: β-blockers (metoprolol, atenolol) blunt max HR by 30-40 bpm, leaving only RPE; iron-deficiency anemia keeps max HR near normal while effective output drops sharply. The practical rule is to drive training by RPE and keep HR for retrospective trends ('six months ago the same RPE 7 session was 162, now it's 155' means aerobic fitness improved).
References · 6
- Mandsager, K., Harb, S., Cremer, P., Phelan, D., Nissen, S. E., & Jaber, W. (2018). Association of cardiorespiratory fitness with long-term mortality among adults undergoing exercise treadmill testing. JAMA Network Open, 1(6), e183605. n=122,007 cardio stress-test patients; high VO2max vs low → all-cause mortality ↓ ~5×, effect larger than quitting smoking / glycemic / BP control. 10.1001/jamanetworkopen.2018.3605
- Hawley, J. A., Lundby, C., Cotter, J. D., & Burke, L. M. (2018). Maximizing cellular adaptation to endurance exercise in skeletal muscle. Cell Metabolism, 27(5), 962-976. The physiological basis for low-to-moderate intensity (Zone 2) work as the dominant driver of mitochondrial biogenesis and fat-oxidation capacity. 10.1016/j.cmet.2018.04.014
- Bassett, D. R., & Howley, E. T. (2000). Limiting factors for maximum oxygen uptake and determinants of endurance performance. Medicine & Science in Sports & Exercise, 32(1), 70-84. Definitive review showing cardiac output (Q) is the primary VO2max-limiting factor, accounting for ~70-85% of variance — not peripheral O2 extraction. 10.1097/00005768-200001000-00012
- American College of Sports Medicine. (2018). ACSM's Guidelines for Exercise Testing and Prescription (10th ed.). Wolters Kluwer. www.acsm.org/education-resources/books/guidelines-exercise-testing-prescription
- Tanaka, H., Monahan, K. D., & Seals, D. R. (2001). Age-predicted maximal heart rate revisited. Journal of the American College of Cardiology, 37(1), 153-156. Reanalysis of 351 studies (n=18,712) yields HRmax = 208 − 0.7×age, more accurate than the legacy 220 − age formula (which carries ±10-12 bpm individual error). 10.1016/S0735-1097(00)01054-8
- Robergs, R. A., & Landwehr, R. (2002). The surprising history of the 'HRmax=220-age' equation. Journal of Exercise Physiology Online, 5(2), 1-10. Traces the 1971 Fox-Naughton derivation; the original was never a research-validated equation, and its SD is ±12 bpm at any age.