Cellular Respiration: How Your Cells Make Energy and What It Means for Athletes

Key Takeaways

• Muscle stores so little ATP that, on its own, it would fuel an all-out sprint for less than 2 seconds; cellular respiration keeps rebuilding it.[1]
• A pooled model of maximal efforts estimated equal aerobic and anaerobic contributions at about 78.6 seconds.[2]
• Endurance, high-intensity and sprint interval training raised pooled markers of muscle mitochondrial content by an estimated 23–27%, with sprint intervals the most time-efficient.[3]
• In a 10-man study, after a high-volume block, two weeks of reduced training volume left every measured mitochondrial parameter except citrate synthase activity no longer significantly different from baseline.[4]
• Lactate is a fuel and a signalling molecule, not a waste product.[5]
• In a small study of non-athletes, iron deficiency was associated with a more glycolytic profile, with no detectable impairment of muscle oxidative phosphorylation.[6]
• In clinical practice, lactate is also measured as a marker of tissue damage, for example in ischaemia and sepsis, and sore muscles after hard training are more often delayed-onset muscle soreness than “lactic acid”.
• In my understanding, fat loss comes down to energy balance and the amount of energy expended, not to a special fat-burning zone or a particular sport.
• Low ferritin without anaemia is an area where medical teaching has been sceptical, yet many athletes report feeling better when it is corrected.

Introduction: Why Your Energy Machinery Matters

Every punch, stride and pedal stroke runs on adenosine triphosphate (ATP). The problem is that skeletal muscle holds only about 5 mmol of ATP per kilogram of wet muscle.[1] At a sprint power output of 900 W, stored ATP alone would theoretically last less than 2 seconds.[1] Cellular respiration is the network of pathways that keeps replacing it, and the metabolic rate can rise more than 100-fold from rest to exercise.[1]

What surprises many athletes is how early the “aerobic” system matters. A 2026 systematic review pooling 102 studies modelled equal aerobic and anaerobic contributions at 78.6 seconds, with a 60-second all-out effort already 42% aerobic.[2]

From a clinical perspective, cellular respiration is rarely the first thing people think about, yet it sits behind many basic clinical situations. It matters in my own sport as well. As more of an interval athlete, I have long had a muscle type that suits anaerobic, lactate-producing efforts better than purely aerobic ones, and anaerobic metabolism has been my natural way of producing energy. Part of this is probably my heavy build combined with an ability to produce force easily. In long aerobic efforts the picture is different, and purely aerobic work has not been my strength.

From a sports medicine perspective, cellular respiration explains why intervals and long sessions both build endurance, why carbohydrate matters for hard sessions, why lactate is not the enemy and why iron status can show up in performance. Below: the biochemistry, then training, fuelling and testing.

What Is Cellular Respiration?

Cellular respiration is how cells release energy from fuel to rebuild ATP. In skeletal muscle, ATP is regenerated through two broad routes [1]:

• Substrate-level phosphorylation (“anaerobic”): the breakdown of phosphocreatine (PCr) and of muscle glycogen, which does not require oxygen.
• Oxidative phosphorylation (“aerobic”): which uses reducing equivalents from carbohydrate and fat metabolism in the mitochondria and depends on the respiratory and cardiovascular systems delivering oxygen.

The anaerobic pathways produce ATP at a much higher rate but have a smaller total capacity than the aerobic pathways.[1]

The Stages of Cellular Respiration

1. Glycolysis. Its regulatory enzyme, phosphofructokinase, is activated by ADP, AMP and inorganic phosphate, the by-products of ATP breakdown, so glycolysis accelerates as soon as muscle works hard. Starting from muscle glycogen, glycolysis yields a net of three ATP molecules and forms lactate.[1]

2. Entry into the mitochondria. Pyruvate produced by glycolysis can be transferred into the mitochondria, where the pyruvate dehydrogenase (PDH) reaction converts it to acetyl-CoA and produces the reducing equivalent NADH. Fatty acids cross the mitochondrial membranes with the help of the carnitine palmitoyl transferase I system and then enter β-oxidation, which produces acetyl-CoA, NADH and FADH2.[1]

3. The citric acid (TCA or Krebs) cycle. The TCA cycle is the major final common pathway for oxidising carbohydrates, lipids and some amino acids. It generates approximately 67% of all reducing equivalents per molecule of glucose.[7]

4. The electron transport chain and oxidative phosphorylation. To make ATP, the respiratory chain needs NADH and FADH2, free ADP, inorganic phosphate and oxygen, and the rise in ADP during exercise is believed to activate it.[1] Oxygen’s final destination is as the terminal electron acceptor in this process, and iron has indispensable roles in oxygen uptake, delivery and this final step.[6]

Cellular Respiration During Exercise: Which Energy System, When?

Energy systems do not take turns: beyond a few contractions, all three pathways contribute simultaneously.[2] The 2026 systematic review by Gastin and Suppiah analysed 311 data points from 102 studies of single maximal efforts and modelled how the anaerobic and aerobic shares change with duration (model-estimated averages for single maximal efforts, not individual measurements) [2]:

Duration of maximal effort	Anaerobic share	Aerobic share
10 s	91%	9%
30 s	75%	25%
60 s	58%	42%
75 s	51%	49%
120 s	38%	62%
240 s	22%	78%
600 s	11%	89%

Equal contributions occurred at 78.6 seconds (95% CI ±1.1 s), and the review found no statistically significant differences between the modelled curves for running and cycling or between the analysed training-status groups.[2] The authors stress that the model’s prediction error was about 12–14%, so the table describes averaged trends rather than precise values for any individual athlete.[2] The review also reported small but significant effects of measurement method and pacing strategy on the modelled contributions, another reason to read the table as a broad trend.[2]

The anaerobic stores themselves are small. Resting PCr (about 75 mmol per kg dry muscle) can be mostly depleted in 10–15 seconds of all-out exercise, and anaerobic glycolytic capacity in 30–90 second efforts is roughly threefold higher, limited by rising muscle acidity rather than by glycogen availability.[1] Meanwhile, 70–100% of VO2max can be reached during an all-out 30-second sprint, and about half of the energy in the last 5 seconds of that sprint is aerobic.[1]

Recovery between efforts is also aerobic. Continued aerobic ATP production regenerates PCr, which can be completely restored within 60–120 seconds, and this matters for repeated sprints in stop-and-go sports.[1]

Anaerobic energy production can also be supported. Some patients use creatine, which supports rapid energy production, and in many it shows up clearly in their results in the gym. It can raise serum creatinine, a blood marker used to assess kidney function. For this reason, when a patient has a raised creatinine in a lab test, I usually ask whether they use creatine. Often the answer is yes. A large amount of muscle mass or a lot of resistance training can also raise it.

Carbohydrate, Fat and Lactate: Fuelling the System

The body stores carbohydrate mainly as glycogen: approximately 100 g in the liver and 400 g in muscle, with 5 g circulating as blood glucose.[8] Which fuel dominates depends mainly on exercise intensity and duration.[1] Carbohydrate oxidation, particularly from muscle glycogen, dominates at higher intensities, while fat oxidation matters more at lower intensities, with the review placing maximal fat oxidation at about 60–65% of VO2max (individual athletes vary widely, as the data below show).[1]

Carbohydrate has two advantages for performance. It is the only fuel that can produce ATP both aerobically and anaerobically, and aerobic ATP production from carbohydrate is approximately 7% more efficient than from fat.[1] Even in marathon and triathlon events lasting 2–2.5 hours, the primary reliance is on carbohydrate oxidation.[1] Depletion of muscle and liver glycogen is a major cause of fatigue in both endurance and high-intensity intermittent activities.[8]

Fat oxidation also varies widely between individuals. In 1,121 athletes tested after at least 5 hours of fasting, maximal fat oxidation averaged 0.59 ± 0.18 g/min but ranged from 0.17 to 1.27 g/min, and it occurred at an average of 49.3 ± 14.8% of VO2max, with a range of 22.6–88.8%.[9] Fat-free mass, VO2max, the intensity of peak fat oxidation, body fat percentage and fasting duration together explained 47% of the variation, leaving more than half unexplained.[9]

Patients, and popular discussion, sometimes talk about a “fat-burning zone” as if there were an intensity that removes fat from the body more effectively than others. The physiology is different. There is a range of intensities in which the body uses a larger share of fatty acids than of carbohydrate, but that is not the same as that range being the best way to lose body fat. Some people also believe that a specific sport burns more fat than others when the goal is weight loss. In my understanding, fat loss comes down to energy balance and the amount of energy expended, not to a particular sport that would burn fat specifically.

Lactate: Part of Cellular Respiration, Not a Waste Product

Lactate was long viewed as a waste product of anaerobic metabolism, but it is now known to form continuously under aerobic conditions. Shuttling between cells that produce it and cells that consume it, lactate serves at least three purposes: a major energy source, the major gluconeogenic precursor and a signalling molecule.[5]

Even with the oxidative pathways active, accelerated glycolysis produces lactate, which accumulates in muscle and blood particularly at higher intensities.[1] Lactate itself is not thought to have a major negative effect on force and power; the acidosis that comes with intense metabolism and ion shifts is considered more important.[1]

Muscle soreness is another area where lactate is often blamed. In my clinical experience, many people believe that sore muscles are caused by lactic acid, or say that their muscles are “full of acid”. The soreness that follows hard training is more often delayed-onset muscle soreness (DOMS), which is not caused by lactate. Lactate also does not stay trapped in the muscle: it is used as fuel, as described above, so the muscle is not left sitting in acid.

How Training Upgrades Cellular Respiration

Endurance training produces two hallmark adaptations: a higher VO2max and greater mitochondrial density in skeletal muscle. Greater mitochondrial density is a major contributor to lower carbohydrate use and lactate production, higher fat oxidation and better endurance performance.[1] Training also shifts the lactate threshold to the right, and the reduction in glycogen use at a given intensity is confined to the muscles that were actually trained.[8]

How should you train to grow mitochondria? A 2024 systematic review and meta-regression pooled five markers of mitochondrial content; the covariate-adjusted model behind the training-type estimates included 343 studies and 5,511 participants.[3] Key findings:

• After adjustment for covariates, mitochondrial content rose to a similar extent with continuous endurance training (23 ± 5%), high-intensity training (27 ± 5%) and sprint interval training (27 ± 7%); the ± values in this review are 95% confidence-interval half-widths.[3]
• Per hour of exercise, sprint interval training was about 2.3 times more efficient than high-intensity training and about 3.9 times more efficient than endurance training at increasing the pooled markers of mitochondrial content, while high-intensity training was about 1.7 times more efficient than endurance training. These figures describe mitochondrial adaptation per training hour, not overall athletic performance.[3]
• More frequent training was associated with larger increases in mitochondrial content, with six sessions per week (the highest frequency compared) showing greater modelled adaptations than four or two.[3]
• Lower initial fitness was associated with larger percentage gains, but well-trained participants still increased mitochondrial content by 6.6 ± 6.5%.[3]
• In the meta-regression, sprint interval training showed larger early adaptations, whereas endurance and high-intensity training kept improving over more weeks.[3]

Intensity may matter even when total work is held constant. In one small within-subject trial, 10 young active men trained one leg with intervals and the other with continuous cycling, with work matched per session (continuous 143 ± 8.4 kJ, interval 144 ± 8.5 kJ).[10] After six sessions over 2 weeks, citrate synthase maximal activity was greater in the interval-trained leg (10.2 ± 0.8 vs. 8.4 ± 0.9 mmol·kg protein⁻¹·min⁻¹), although single-leg VO2peak did not change.[10]

Consistency matters too. In 10 healthy men, 4 weeks of normal-volume interval training (3 sessions per week) changed no mitochondrial parameter, but the following 20 days of twice-daily high-intensity interval training increased mitochondrial respiration and citrate synthase activity by about 40–50%.[4] After 2 weeks of reduced-volume training (5 sessions), every mitochondrial parameter except citrate synthase activity (about 36% above initial) was no longer significantly different from baseline.[4]

Occupational health offers another angle. In many workplaces VO2max is the standard measure of an employee’s physical fitness, and employers often want to invest in fitness and, through it, work ability. In the occupational health service where I work, we also carry out VO2max measurements and fitness programmes. I notice this in myself as well: the better shape I am in, the more work I can do, so improving cellular respiration may well improve work ability. In my clinical experience, some patients, including those who would like to exercise, are in the opposite situation. All their energy goes into getting through the working day, and afterwards they have nothing left. This can easily become a spiral: they skip training, their aerobic capacity declines, and their work ability keeps getting worse.

Iron: A Hidden Link in the Energy Chain

Iron has indispensable roles in oxygen uptake, oxygen delivery and the final electron-transfer step of oxidative phosphorylation,[6] which means iron deficiency may be relevant when investigating unexplained exercise-related symptoms. A 2022 case-control study compared adults with absolute iron deficiency but without frank anaemia (ferritin ≤15 µg/L with transferrin saturation below 16%; mean ferritin 8.3 µg/L) against iron-replete controls, 13 participants per group, and then randomised participants to intravenous (IV) iron or saline.[6]

The findings were nuanced [6]:

• Phosphorus magnetic resonance spectroscopy of exercising calf muscle showed no significant influence of baseline iron status or IV iron on high-energy phosphate metabolism.
• During submaximal cycling, blood lactate fell more slowly in the iron-deficient group (P = 0.005), and IV iron accelerated the fall in venous lactate in the iron-deficient group (P = 0.028).
• IV iron raised the lactate threshold by about 10% regardless of baseline iron status, before any change in haemoglobin.
• The authors concluded that iron deficiency promotes a more glycolytic phenotype without a detectable effect on mitochondrial bioenergetics.

Two caveats matter for athletes. The participants were not selected as athletes, and the study was not powered to rule out a subtle impairment of muscle oxidative phosphorylation.[6]

In my clinical experience, many athletes, and a good share of exhausted patients in general, are acutely aware of iron’s central role in metabolism, and whether iron can be tested is often one of the first things they ask. Ferritin has become a popular fatigue marker, partly through popularisation, and a lot of beliefs circulate around it. In medical school the attitude towards low ferritin in laypeople and sedentary people was rather sceptical. The rule we were given on the haematology course was that if haemoglobin does not fall, iron deficiency does no harm. Many athletes nevertheless report feeling better when iron is replaced, or when a low ferritin that has not yet caused anaemia is corrected. Restless legs syndrome is a separate situation: there, iron deficiency is usually corrected before the restless legs are treated in any other way. Oral iron also causes stomach symptoms in many people, and I often recommend a sucrosomial iron preparation to my patients.

Assessing Cellular Respiration in Athletes

Several measurements capture parts of cellular respiration:

• Lactate threshold testing. Training shifts the lactate threshold to the right,[8] and in the 2022 iron study the lactate threshold rose measurably after IV iron.[6]
• Indirect calorimetry. A graded test after fasting can measure maximal fat oxidation and the intensity where it occurs, which differ widely between athletes.[9]
• VO2max. Estimated increases in pooled mitochondrial-content markers were about 2.2, 2.2 and 3.0 times the increases in VO2max for endurance, high-intensity and sprint interval training, respectively. Changes in the pooled mitochondrial marker were positively but weakly correlated with changes in VO2max (r = 0.283). Associations varied between individual markers: citrate synthase activity showed a stronger correlation (r = 0.453), whereas mitochondrial volume density showed no significant correlation (r = −0.031). VO2max is therefore an imperfect proxy for changes in muscle mitochondrial content.[3]
• Iron status. Ferritin and transferrin saturation defined iron deficiency in the 2022 study;[6] those were research entry criteria, not athlete-specific treatment thresholds.

Phosphorus magnetic resonance spectroscopy can assess muscle energy metabolism non-invasively, but its use for whole-body maximal exercise is limited by cost, accessibility and technical complexity.[2]

In practice, cellular respiration is also measured indirectly in the clinic. VO2max usually tells us about a person’s oxygen uptake and oxygen transport capacity. Lactate, in turn, is a general marker of tissue damage and is often measured where tissues may be short of oxygen: in ischaemic conditions, in the diagnosis of sepsis and in patients with abdominal pain where intestinal ischaemia is a concern. In these situations it serves as an indirect measure of cellular respiration.

Practical Recommendations for Athletes

• Match training to the event’s mix of demands. The review’s authors suggest emphasising phosphagen turnover and glycolytic capacity for short, high-power efforts, and aerobic power and faster oxygen-uptake kinetics as event duration approaches or exceeds the modelled ~75–80 s crossover. Treat that crossover as a broad reference point, not an individual threshold.[2]
• Use both intensity and volume. In the meta-regression, sprint intervals gave the greatest estimated mitochondrial-content adaptation per hour, and higher training frequency and longer intervention duration were also associated with larger gains. These associations do not establish an optimal prescription for every athlete.[3]
• Keep the stimulus going. In a 10-man study, most mitochondrial parameters were no longer significantly different from baseline after two weeks of reduced volume following a high-volume block, although citrate synthase activity stayed about 36% above its starting level.[4]
• Fuel for the work required. Keep carbohydrate availability high for high-intensity sessions and competition. Deliberate low-carbohydrate training has shown inconsistent benefits, can impair acute training intensity and may be best reserved for sessions in which reduced carbohydrate availability is unlikely to compromise the intended workload.[8]
• Do not rely on generic fat-burning zones. Peak fat oxidation occurred anywhere from 22.6% to 88.8% of VO2max between athletes.[9]
• Consider iron status when endurance stalls. Iron deficiency altered lactate handling in otherwise healthy adults.[6] Checking ferritin and transferrin saturation with your doctor is a reasonable step in that situation (clinical synthesis, not a study finding).

Conclusion: Your Energy Machinery Is Trainable

Cellular respiration is not abstract biochemistry. It explains why stored ATP alone would theoretically last less than 2 seconds of sprinting,[1] why a pooled model puts the aerobic–anaerobic crossover of maximal efforts at about 78.6 seconds,[2] and why carbohydrate dominates hard work.[1] The machinery also responds to training: estimated mitochondrial content rose by 23–27% across training types.[3] Keep the stimulus consistent, fuel the sessions that matter, treat lactate as fuel and do not ignore iron. In clinical practice, the same physiology helps to put common beliefs in context: sore muscles are more often DOMS than trapped lactic acid, and fat loss depends on energy balance rather than a special fat-burning zone. Aerobic fitness may also support work ability, something I notice in my own occupational health work.

---

Bibliography

[1] https://doi.org/10.1038/s42255-020-0251-4

[2] https://doi.org/10.1007/s40279-026-02414-7

[3] https://doi.org/10.1007/s40279-024-02120-2

[4] https://doi.org/10.1096/fj.201500100R

[5] https://doi.org/10.1016/j.cmet.2018.03.008

[6] https://doi.org/10.1038/s41598-021-03968-4

[7] https://doi.org/10.2165/00007256-200737120-00005

[8] https://doi.org/10.3390/nu10030298

[9] https://doi.org/10.1249/MSS.0000000000001084

[10] https://doi.org/10.1113/JP272570