Executive summary
Adrenaline and epinephrine are the same molecule. Coffee or caffeine does not manufacture a special form of adrenaline; rather, caffeine can increase sympathetic–adrenal activation and thereby stimulate the adrenal medulla to release endogenous epinephrine. Resistance exercise activates the same adrenal-medullary endpoint, but through a different initiating signal: central motor command, anticipatory arousal, muscle afferent feedback, mechanical effort, and accumulating metabolic stress. In both cases, sympathetic preganglionic neurons release acetylcholine onto adrenal chromaffin cells, which secrete epinephrine into the circulation. citeturn20search0turn20search4turn20search8
The most important quantitative conclusion is that the epinephrine response to caffeine is strongly dose-dependent and is not necessary for caffeine’s performance benefit. In a classic human dose-response experiment, 3 mg/kg caffeine improved endurance by about 22% without significantly increasing plasma epinephrine, whereas 6 and 9 mg/kg produced progressively greater epinephrine responses; 9 mg/kg produced the largest hormonal/metabolic disturbance but no significant performance improvement. This is strong evidence against the popular idea that caffeine works mainly by generating an “adrenaline surge” or mobilizing fat. citeturn17view0
Resistance exercise, particularly large-muscle, high-effort, high-volume or short-rest training, can provoke a much sharper sympathoadrenal response. In resistance-trained men performing six sets of ten squats at 80% 1RM, epinephrine was already 270% above the resting-control condition before the workout because of anticipatory arousal and then increased a further 512% above the pre-exercise value during the sets. An exhaustive six-set leg-extension protocol at 50% load produced a postexercise epinephrine concentration of 1.9 ± 1.1 nmol/L in its first session. citeturn22view0turn17view5
Thus, ordinary ergogenic caffeine intake and hard lifting are not equivalent “adrenaline exposures.” At approximately one hour after a very large 9 mg/kg caffeine dose, pre-exercise epinephrine in trained subjects was approximately 0.44–0.45 nmol/L versus 0.22–0.31 nmol/L after placebo. By contrast, exhaustive resistance exercise has produced postexercise values around 1–2 nmol/L, and intense endurance exercise can produce substantially higher values. Cross-study comparisons are imperfect, but the available data suggest that hard, fatiguing exercise generally creates the larger acute epinephrine pulse, while caffeine creates a smaller but pharmacologically prolonged sympathetic stimulus. citeturn23search0turn23search2
Caffeine can, however, amplify exercise-associated epinephrine considerably. With 9 mg/kg caffeine before running or cycling, epinephrine at 15 minutes of exercise was approximately 2.51–2.53 nmol/L versus 1.23–1.24 nmol/L with placebo; at exhaustion in another study, 9 mg/kg caffeine yielded 10.1 ± 2.3 nmol/L versus 5.3 ± 0.8 nmol/L with placebo. These experiments involved endurance exercise rather than lifting, so the precise magnitude cannot be transferred to a squat or bench-press workout, and surprisingly little resistance-specific research has directly measured this interaction. citeturn23search0turn23search1
The temporal patterns also differ. Oral caffeine is rapidly absorbed, typically reaching high plasma concentrations within roughly 30–60 minutes and having an average plasma half-life around five hours. Its cardiovascular and arousal effects can therefore outlast a workout by several hours. In contrast, exercise-induced epinephrine begins within minutes—and can begin before the first set through anticipatory arousal—usually rises toward the latter, most demanding portion of the session, and then falls during recovery. Resistance studies frequently find catecholamines still elevated immediately or five minutes after exercise, but precise resistance-specific decay curves are surprisingly poorly characterized. citeturn19search0turn22view0turn17view6turn16search23
For performance, the evidence favors moderate caffeine rather than maximal adrenaline. A large body of sport-nutrition research supports approximately 2–3 mg/kg as often sufficient for resistance exercise, while 3–6 mg/kg is the conventional well-supported performance range. Higher doses such as 9 mg/kg add adverse effects and hormonal stimulation without reliably adding performance. Meta-analyses find small improvements in strength, muscular endurance, movement velocity, and power, and a 2026 human trial found 3 mg/kg improved strength/endurance in resistance-trained men and women. citeturn19search1turn19search18turn19search22turn21search0
For fat loss, neither a caffeine-induced nor a lifting-induced adrenaline spike should be treated as a proxy for body-fat loss. Epinephrine acutely stimulates lipolysis and substrate mobilization, but high-dose caffeine can raise epinephrine without proportionately increasing fat oxidation or performance. Long-term fat loss depends on cumulative energy balance and training adaptations; resistance training is valuable largely because it maintains or builds lean tissue and modestly reduces fat mass, not because each session generates a transient catecholamine peak. citeturn17view0turn23search0turn21search9
For safety, the key distinction is between brief physiological sympathoadrenal activation from exercise and adding an exogenous stimulant on top of the exercise response. Heavy lifting itself can cause very large transient arterial-pressure excursions, particularly with high effort and breath-holding/Valsalva. Caffeine usually raises blood pressure more consistently than heart rate; in people with hypertension, 200–300 mg raised systolic pressure by about 8.1 mmHg and diastolic pressure by 5.7 mmHg in controlled trials, beginning within the first hour and lasting at least three hours. Combining substantial caffeine with maximal or near-maximal lifting therefore deserves particular caution in people with uncontrolled hypertension, symptomatic arrhythmias, exertional symptoms, or marked anxiety. citeturn20search1turn20search29turn16search0turn16search12
The bottom line is that “more adrenaline” is not a desirable training goal. The exercise epinephrine response is a normal, task-coupled component of physiological stress. Caffeine’s useful performance effects occur substantially through adenosine-receptor blockade and central nervous-system effects and can be obtained at doses that cause little or no measurable epinephrine increase. citeturn17view0turn19search1
Scope and physiological framework
A terminology caveat is important. Most mechanistic experiments administer anhydrous caffeine in capsules, not brewed coffee. Coffee delivers caffeine along with chlorogenic acids and numerous other constituents, and its caffeine content varies by bean, preparation, and serving. Consequently, this report interprets “coffee-induced adrenaline” principally as the effect of caffeine ingestion rather than treating coffee as a precisely dosed pharmacological preparation. The established ergogenic literature likewise largely expresses exposure in milligrams of caffeine per kilogram of body mass. citeturn19search1turn21search1
Adrenal-medullary physiology. Epinephrine is synthesized primarily by chromaffin cells in the adrenal medulla. Sympathetic preganglionic neurons innervating the adrenal medulla release acetylcholine, which activates nicotinic receptors on those chromaffin cells and induces systemic catecholamine secretion. This arrangement differs from most sympathetic nerve terminals, where norepinephrine is released locally into the neuroeffector junction. Plasma epinephrine is therefore especially informative about adrenal-medullary activation, whereas circulating norepinephrine more heavily reflects sympathetic nerve-terminal spillover. citeturn20search0turn20search8turn20search16
Caffeine’s initiating mechanism. At customary human doses, caffeine’s major pharmacologically relevant action is competitive antagonism of adenosine receptors, especially A1 and A2A receptors in the central nervous system. Adenosine normally restrains neural activity and transmitter release; blocking that restraint increases wakefulness and changes dopaminergic, noradrenergic and other neurotransmitter signaling. Increased central arousal and sympathetic outflow can subsequently stimulate the adrenal medulla. Mechanisms sometimes mentioned in popular explanations—strong phosphodiesterase inhibition or massive intracellular calcium release—require concentrations substantially beyond those ordinarily produced by coffee and are not considered the principal mechanism at typical dietary/ergogenic exposure. citeturn9search1turn9search3turn4search2turn19search0
Resistance exercise’s initiating mechanism. Lifting weights activates sympathetic output through several overlapping signals: conscious central motor command, anticipation/arousal before effort, mechanically sensitive and metabolically sensitive muscle afferents, reduced local blood flow during forceful contraction, accumulating lactate/H⁺ and other metabolites, and the escalating effort required as fatigue develops. The importance of effort rather than external force alone is illustrated by direct blood-pressure experiments showing that the cardiovascular response tracks relative effort/central command strongly. citeturn16search12turn22view0turn16search23
The two pathways can therefore be represented as follows:
flowchart TD
A["Caffeine ingestion"] --> B["Rapid GI absorption"]
B --> C["Adenosine A1 / A2A receptor antagonism"]
C --> D["CNS disinhibition: arousal, altered neurotransmission"]
D --> E["Increased sympathetic drive"]
F["Resistance exercise"] --> G["Anticipatory arousal + central motor command"]
F --> H["Muscle mechanical / metabolic afferent feedback"]
G --> E
H --> E
E --> I["Preganglionic sympathetic neurons"]
I --> J["Acetylcholine → nicotinic receptors on adrenal chromaffin cells"]
J --> K["Epinephrine released into circulation"]
K --> L["β1: cardiac stimulation"]
K --> M["α / β2: vascular redistribution"]
K --> N["Hepatic & muscle carbohydrate mobilization"]
K --> O["β-adrenergic adipose lipolysis"]
C --> P["Direct CNS effects: alertness, lower perceived effort"]
F --> Q["Local muscle tension & metabolic signaling"]
Q --> R["Training adaptations largely independent of the epinephrine spike"]
A --> S["Caffeine + lifting"]
F --> S
S --> T["Potentially greater combined sympathoadrenal / pressor load"]This scheme matters because the same plasma epinephrine concentration does not imply the same whole-body state. Caffeine simultaneously blocks adenosine receptors throughout multiple tissues; resistance exercise simultaneously generates mechanical tension, local metabolites, changes in muscle blood flow, motor-unit recruitment, and afferent signaling. Epinephrine is one mediator embedded in each state, not the state itself. The glucose-clamp literature demonstrates this directly: caffeine and an epinephrine infusion producing similar plasma adrenaline concentrations did not produce identical effects on glucose disposal. citeturn18search2turn18search9
Magnitude, timing, and dose-response
There is no high-quality head-to-head study that gives the same individuals caffeine alone, lifting alone, and caffeine-plus-lifting under matched conditions while densely sampling plasma epinephrine. Numerical comparisons below therefore represent the best synthesis of separate controlled human studies, not a formal equivalence scale.
| Stimulus and protocol | Observed plasma epinephrine response | Timing | Interpretation |
|---|---|---|---|
| Caffeine, 3 mg/kg, 1 h before running at 85% VO₂max | No statistically significant epinephrine increase versus placebo, despite ~22% greater endurance in this small trained sample. citeturn17view0 | Caffeine given 60 min pre-exercise; samples every 15 min. | Low/moderate ergogenic caffeine need not generate a significant adrenaline surge. |
| Caffeine, 9 mg/kg, before exercise | Before exercise, ~0.44 nmol/L versus 0.22 after placebo in the running trial; ~0.45 versus 0.31 nmol/L in cycling. citeturn23search0 | ~60 min after ingestion. | Very high caffeine approximately doubled pre-exercise epinephrine in one condition, although absolute values remained below those seen during hard exercise. |
| Caffeine, 5 mg/kg, under a hyperinsulinemic clamp | Adrenaline rose to ~0.6 nmol/L. citeturn18search2turn18search9 | Clamp initiated 30 min after ingestion. | Demonstrates a substantial caffeine-associated adrenal response at rest/metabolic challenge, but the clamp makes this different from quiet resting conditions. |
| Resistance exercise: 6 × 10 squats at 80% 1RM in trained men | Pre-exercise epinephrine was 270% above control from anticipatory arousal; during exercise it increased a further 512% above the pre-exercise value. citeturn22view0 | Increase began before the first working set and progressed across the six sets. | Heavy, repeated compound lifting can create very large relative sympathoadrenal activation. |
| Resistance exercise: 6 × 12 bilateral leg extensions at 50% load, exhaustive | Postexercise epinephrine 1.9 ± 1.1 nmol/L in the first bout; 0.7 ± 0.7 in the second; 1.2 ± 1.0 in the fifth. citeturn17view5turn23search2 | Immediately postexercise. | Even moderate load can evoke substantial adrenaline when volume and fatigue are high; repeated exposure changes the response. |
| Heavy whole-body resistance protocol: 3 × 10RM, ten exercises, very short rests | Epinephrine, norepinephrine, dopamine and cortisol were significantly elevated 5 min after exercise. citeturn17view6 | Still elevated at +5 min. | Dense, high-volume lifting produces a persistent immediate-recovery endocrine response. |
| 9 mg/kg caffeine + intense endurance exercise | At 15 min: ~2.51–2.53 nmol/L with caffeine versus ~1.23–1.24 with placebo. citeturn23search0 | During exercise. | High-dose caffeine approximately doubled exercise epinephrine in these trials. |
| 9 mg/kg caffeine + exhaustive prolonged exercise | At exhaustion: 10.1 ± 2.3 nmol/L versus 5.3 ± 0.8 nmol/L with placebo. citeturn23search1 | Exhaustion. | Illustrates how caffeine can amplify a large exercise response; this magnitude should not be assumed for ordinary lifting. |
Caffeine onset and duration. Oral caffeine is rapidly and nearly completely absorbed. Plasma caffeine commonly peaks roughly 30–60 minutes after ingestion, although reported individual time-to-peak values span a much wider range, and its mean plasma half-life in healthy adults is about five hours. Caffeine’s epinephrine response usually becomes evident over the first 30–60 minutes in studies using sufficiently high doses, but the duration of elevated plasma epinephrine itself has not been mapped as consistently as caffeine pharmacokinetics. It is therefore incorrect to infer that epinephrine remains elevated for the full five-hour caffeine half-life. What clearly can persist for hours are stimulant and pressor effects; in hypertensive subjects, the acute blood-pressure elevation from 200–300 mg lasted at least three hours. citeturn19search0turn19search28turn20search1
Resistance-exercise onset and duration. The catecholamine response can begin before the first repetition because anticipatory arousal is sufficient to stimulate the sympathoadrenal system. During hard repeated sets, concentrations typically rise toward the latter portion of the session, especially with circuit-like, high-density protocols. They may remain elevated for at least several minutes afterward, as shown by the Kraemer study’s five-minute postexercise measurement. The exact “duration” is harder to state than for caffeine because catecholamine concentrations change minute by minute and many lifting studies sampled only before and immediately after the workout rather than throughout a 30–60-minute recovery period. citeturn22view0turn17view6turn16search23
Caffeine dose-response is clearer than the resistance-training dose-response. In Graham and Spriet’s 3, 6 and 9 mg/kg trial, plasma caffeine rose dose-dependently; epinephrine did not rise significantly at 3 mg/kg but did at 6 and 9 mg/kg. Crucially, both 3 and 6 mg/kg improved endurance by approximately 22%, whereas 9 mg/kg did not. Contemporary sport-nutrition evidence consequently favors approximately 3–6 mg/kg, with some resistance-exercise benefits occurring at about 1.5–3 mg/kg, rather than pursuing 6–9 mg/kg solely to increase catecholamines. citeturn17view0turn19search1turn19search18turn19search36
For context, the body-mass implications are substantial:
| Body mass | 2 mg/kg | 3 mg/kg | 6 mg/kg | 9 mg/kg |
|---|---|---|---|---|
| 60 kg / 132 lb | 120 mg | 180 mg | 360 mg | 540 mg |
| 80 kg / 176 lb | 160 mg | 240 mg | 480 mg | 720 mg |
| 100 kg / 220 lb | 200 mg | 300 mg | 600 mg | 900 mg |
The FDA’s general-population figure of 400 mg/day as an amount not generally associated with adverse effects in most healthy adults is a population safety reference, not a recommended sports dose and not a guarantee of tolerance. Thus, a 6 mg/kg experimental sports dose already exceeds 400 mg in anyone heavier than about 67 kg. citeturn21search1
For resistance exercise, percentage of 1RM alone does not determine the catecholamine dose. Six exhaustive sets at 50% load generated substantial epinephrine, while six sets of ten squats at 80% 1RM generated very large relative increases. The available literature points to the combined influence of relative effort, proximity to fatigue, number of working sets, amount of muscle recruited, rest intervals, exercise density and anticipatory arousal. Consequently, “80% 1RM causes X nmol/L” would be a false precision. citeturn17view5turn22view0turn17view6
Cardiovascular, metabolic, neuroendocrine, and performance effects
The distinction between epinephrine-mediated and caffeine-mediated effects is critical. Caffeine has direct adenosine-receptor effects, while resistance exercise has large mechanical and local metabolic effects. The following table therefore describes the net physiological state rather than attributing every change to circulating adrenaline.
| Domain | Caffeine | Resistance exercise | Caffeine before lifting |
|---|---|---|---|
| Heart rate | Not reliably increased at rest; sympathetic stimulation can be offset by the baroreflex after BP rises. Exercise/recovery HR effects are variable. citeturn20search17turn18search24 | HR rises acutely with effort and sympathetic activation; magnitude depends strongly on exercise design. citeturn17view6 | May raise HR or delay parasympathetic recovery in some studies, but results are inconsistent. citeturn18search24 |
| Blood pressure | Acute BP elevation is well established; 200–300 mg produced mean +8.1/+5.7 mmHg SBP/DBP in trials of hypertensive adults. citeturn20search1 | Can cause extremely large transient arterial-pressure excursions during hard repetitions, influenced by effort and Valsalva/breathing. citeturn16search0turn16search12 | Effects can be additive: caffeine has been shown to increase the BP response during exercise. citeturn18search20 |
| Glucose / glycogen | High-dose caffeine can impair acute insulin-mediated glucose disposal. At 5 mg/kg, glucose infusion requirement fell ~13%; caffeine-generated adrenaline alone did not fully explain the effect. citeturn18search2turn18search5 | Epinephrine assists hepatic glucose production and muscle glycogen mobilization to meet rapid ATP demand; high-density RT also generates substantial lactate. citeturn20search0turn17view6 | Greater sympathoadrenal drive can favor carbohydrate availability during high-intensity work; this is useful acutely but should not be interpreted as improved insulin sensitivity during the session. citeturn18search2turn23search1 |
| Lipolysis / fat mobilization | Higher doses may raise glycerol/FFA, but 3 mg/kg can improve performance without doing so; even 9 mg/kg can increase epinephrine without a proportionate change in whole-body fat oxidation. citeturn17view0turn23search0 | Catecholamines contribute to adipose lipolysis, but high-intensity lifting itself relies heavily on carbohydrate metabolism. citeturn20search0turn17view6 | A larger adrenaline value does not imply proportionally greater fat loss. citeturn17view0 |
| Cortisol | Acute caffeine increases cortisol; tolerance develops incompletely with habitual exposure. citeturn20search2turn20search6 | Protocol-dependent: very dense heavy RT increased cortisol, whereas a 30-min protocol in young untrained men increased catecholamines without a significant cortisol rise. citeturn17view6turn17view7 | Caffeine can augment cortisol responses around exercise/stress. citeturn20search10turn20search14 |
| Arousal / perceived effort | Adenosine blockade promotes wakefulness and can lower perceived effort; very high doses can produce nervousness, poor concentration and anxiety instead. citeturn19search1turn17view0 | Anticipatory catecholamines may increase readiness and force expression; exercise also has numerous CNS effects independent of adrenaline. citeturn22view0 | Moderate caffeine may improve readiness, pain tolerance and motor output; excessive arousal can become counterproductive. citeturn19search18turn17view0 |
| Strength / endurance / power | Meta-analytic evidence supports small improvements in muscular strength and endurance. citeturn19search26turn19search18 | The training stimulus itself drives long-term neural and muscular adaptation. citeturn21search9 | Current meta-analysis supports improved movement velocity and power; 2026 data also support 3 mg/kg for strength/endurance in trained men and women, though not every individual or study responds. citeturn19search22turn21search0turn21search6 |
Why epinephrine is not the main explanation for caffeine’s ergogenic effect
The most informative evidence is the dose-response dissociation. If caffeine’s performance benefit were primarily caused by epinephrine, the 9 mg/kg condition in Graham and Spriet’s experiment should have produced the greatest performance improvement because it generated the largest catecholamine/metabolic response. Instead, 3 mg/kg improved endurance despite no significant epinephrine increase, 6 mg/kg performed similarly, and 9 mg/kg generated more epinephrine without a significant performance benefit. citeturn17view0
The metabolic clamp data lead to the same conclusion from another direction. Caffeine at 5 mg/kg raised adrenaline to about 0.6 nmol/L and reduced insulin-mediated glucose disposal, but experimentally infusing adrenaline to comparable concentrations did not reproduce the full glucose effect. The investigators concluded that mechanisms additional to adrenaline must be involved. citeturn18search2turn18search9
Current performance evidence is consequently much more compatible with central adenosine blockade, altered perception of effort/pain, enhanced neural drive and muscle activation as major contributors, with catecholamines acting as a parallel or supporting response rather than the sole causal pathway. citeturn19search1turn19search18
Caffeine before lifting
Direct endocrine evidence specifically pairing caffeine with a standardized resistance workout is much thinner than the performance literature. It is therefore reasonable to separate what is well established from what is inferred.
Well established: caffeine can raise epinephrine in a dose-dependent fashion; resistance exercise independently raises epinephrine; caffeine can roughly double exercise epinephrine in intense endurance protocols; and caffeine can improve several resistance-performance outcomes. citeturn17view0turn22view0turn23search0turn19search22
A reasonable physiological inference is that caffeine before a sufficiently stressful lifting session may increase the total sympathoadrenal and cardiovascular load compared with the same workout without caffeine. Evidence from exercise BP studies supports an additive pressor effect. But the exact plasma epinephrine increment during squats, deadlifts or bench pressing after 2–3 mg/kg caffeine has not been characterized well enough to assign a reliable number. citeturn18search20turn18search28
This matters practically because the dose that commonly improves lifting—around 2–3 mg/kg—need not be the dose that maximizes catecholamines. The rational aim is the minimum dose that improves the desired performance outcome, not the maximum dose tolerated. citeturn19search18turn19search36
Acute versus chronic adaptation and population variability
Caffeine tolerance is endpoint-specific rather than all-or-nothing. Repeated caffeine exposure can attenuate some catecholamine, cortisol and blood-pressure responses while leaving other effects partly intact. In a six-week experiment involving habitual 500-mg caffeine supplementation before exercise, the epinephrine response to exercise and to caffeine-plus-exercise became smaller with chronic exposure, although metabolic differences did not disappear completely. Likewise, a controlled study of healthy men and women found that daily caffeine reduced but did not abolish the cortisol response to a subsequent caffeine challenge. citeturn20search3turn20search2turn20search6
This makes habitual intake important when comparing studies. A caffeine-naïve or recently abstinent participant may have a stronger subjective, blood-pressure and hormonal response than a habitual user, yet ergogenic tolerance is less straightforward. Reviews of resistance exercise do not support a simple rule that habitual caffeine users stop receiving performance benefits. citeturn20search7turn19search18
Resistance training also changes the acute adrenal response. Pullinen and colleagues found that epinephrine after an identical exhaustive leg-extension session fell from 1.9 ± 1.1 nmol/L in the first bout to 0.7 ± 0.7 nmol/L in the second bout, despite substantial metabolic strain; the pattern later partly recovered. These data illustrate that prior exposure, recovery status and neural adaptation can alter the hormonal response even before large changes in muscle size occur. citeturn17view5turn23search2
Training status creates another complication: the same external weight becomes a lower relative physiological stress as someone becomes stronger, whereas exercising at a matched percentage of the person’s new maximum may preserve or even reveal a high catecholamine secretory capacity. Thus “trained people produce less adrenaline” is too simplistic; conclusions depend on whether studies match absolute work, relative intensity, fatigue, volume and familiarity. citeturn17view5turn16search11
Sex. Much of the older epinephrine–exercise literature used small male samples: eight men in the Pullinen study, ten resistance-trained men in French et al., and predominantly male samples in many classic caffeine experiments. That limits confidence in sex-specific endocrine estimates. Contemporary resistance-exercise reviews find broadly similar caffeine ergogenicity in women and men, and a 2026 trial reported performance benefits from 3 mg/kg in resistance-trained participants of both sexes. However, similarity in performance response does not prove identical epinephrine kinetics. citeturn17view5turn22view0turn19search18turn21search0
Age. Direct age-stratified plasma-epinephrine data for caffeine-plus-resistance exercise are scarce. For the broader consequences of resistance training, older adults and people with hypertension or cardiometabolic risk frequently show greater, rather than smaller, chronic improvements in resting blood pressure and glucose regulation than young healthy adults. The American Heart Association’s 2023 scientific statement, for example, summarizes mean resting BP reductions around 6/5 mmHg in hypertensive populations undergoing resistance training. citeturn21search9
Genetics and interindividual response. Caffeine pharmacokinetics vary substantially, particularly through CYP1A2-mediated metabolism, and receptor-related genetic variation may also influence subjective or performance responses. Contemporary results are mixed enough that genotype should not currently be used as a deterministic “caffeine responder” test, though the 2026 resistance study provides additional evidence that CYP1A2 genotype may modify the magnitude of some performance responses. citeturn19search28turn21search0
The broader implication is that plasma epinephrine should not be treated as a fixed personal quantity generated by “one cup of coffee” or “one heavy workout.” Dose, habituation, sleep, psychological anticipation, fitness, workload, feeding state, genetics, medication use and sampling time all matter. citeturn17view0turn22view0turn19search28
Safety, fat loss, performance, and recovery implications
Cardiovascular safety
Caffeine’s most reproducible acute cardiovascular effect is a rise in blood pressure, not necessarily tachycardia. A meta-analysis in people with hypertension found that 200–300 mg produced average increases of 8.1 mmHg systolic and 5.7 mmHg diastolic, appearing in the first hour and lasting at least three hours. Individual responses vary considerably. citeturn20search1turn20search29
Heavy lifting creates a different hemodynamic challenge. Direct intra-arterial studies show that arterial pressures during heavy resistance exercise can become extraordinarily high for brief intervals, and the magnitude depends strongly on relative effort, fatigue and breathing/Valsalva behavior. These transient peaks are not equivalent to sustained resting hypertension, and regular resistance training generally lowers resting blood pressure over time, including in hypertensive adults. citeturn16search0turn16search12turn21search9
That acute-versus-chronic distinction is important: the fact that blood pressure and epinephrine become high during a set does not mean lifting causes chronic sympathetic hypertension. The AHA regards appropriately prescribed resistance exercise as safe and beneficial for most healthy people and many stable cardiovascular populations. citeturn21search9
For someone with uncontrolled hypertension, however, combining a substantial caffeine dose with maximal, near-maximal or high-fatigue lifting creates an avoidable overlap of two pressor stimuli. A cautious clinical approach is to achieve BP control and obtain individualized exercise guidance rather than treating a high-stimulant pre-workout as routine. The AHA specifically recommends medical evaluation and modification of training for relevant cardiovascular contraindications and adverse symptoms such as chest pressure, marked dyspnea, dizziness or palpitations. citeturn21search9turn20search1
Arrhythmia and palpitations
Moderate caffeine intake should not be portrayed as universally arrhythmogenic. Contemporary cardiovascular evidence does not support a blanket recommendation that everyone with atrial fibrillation must eliminate coffee or caffeine, and the 2026 American Heart Association scientific statement emphasizes that effects differ by arrhythmia phenotype; randomized data have even suggested less atrial fibrillation with caffeinated coffee while ventricular ectopy can increase in some circumstances. citeturn16search1turn16search17
That population-level reassurance is not a license to ignore individual symptoms. Someone who reliably experiences palpitations, significant ectopy, presyncope or tachyarrhythmia after caffeine—particularly when combined with intense lifting—has a different risk–benefit calculation and should be evaluated. Concentrated/high-dose stimulant use is also very different from ordinary coffee intake. citeturn21search1turn21search9
Anxiety and arousal
Caffeine’s useful alerting effect and its anxiety-producing effect lie on a continuum. A 2024 meta-analysis found increased anxiety with caffeine, particularly at higher doses, and people with panic disorder are especially sensitive: a systematic review found that doses roughly equivalent to about five cups of coffee provoked panic attacks in a large proportion of patients with panic disorder. citeturn16search18turn16search2
This creates an important distinction from lifting-induced arousal. The anticipatory adrenaline rise before a heavy set is usually task-linked and short-lived. Caffeine can maintain the underlying stimulant state for hours because the drug is eliminated slowly. A person prone to panic or somatic hypervigilance may therefore tolerate exercise-induced palpitations well yet react poorly to the prolonged interoceptive sensations of a large caffeine dose. citeturn22view0turn19search0turn16search2
Performance
For most healthy lifters seeking an ergogenic benefit, the evidence argues for a low-to-moderate dose strategy:
Approximately 2–3 mg/kg is a defensible starting range; doses around 3–6 mg/kg have the broadest overall sports-performance evidence, while 9 mg/kg adds side effects without reliably adding performance. In resistance exercise specifically, lower doses appear capable of producing benefits similar to 6 mg/kg in many outcomes. citeturn19search1turn19search18turn17view0
Timing around 45–60 minutes before conventional capsule/coffee ingestion is reasonable when a peak systemic exposure near training is desired, but individual absorption varies. Caffeine’s long half-life means that later-day training requires weighing a small potential performance gain against sleep loss. citeturn19search0turn19search1
The most important practical point is that subjective stimulation is not synonymous with better performance. At high doses, subjects in the classic dose-response work reported nervousness, difficulty concentrating and related adverse sensations even though epinephrine was high. citeturn17view0
Fat loss
Epinephrine acutely activates β-adrenergic pathways that promote lipolysis, so both caffeine and exercise can increase release of fatty acids from adipose tissue. But mobilizing a fatty acid into blood is not the same as oxidizing it, and oxidizing more fat for an hour is not necessarily equivalent to reducing body fat over weeks. citeturn20search0turn17view0
The caffeine experiments demonstrate the problem clearly. The largest caffeine dose produced the greatest epinephrine, glycerol and free-fatty-acid response but the worst performance of the three caffeine doses; another study found a marked epinephrine rise with high-dose caffeine without a corresponding change in respiratory exchange ratio or plasma free fatty acids during exercise. citeturn17view0turn23search0
Resistance training has a considerably stronger long-term rationale for body composition than trying to pharmacologically maximize adrenaline. The AHA’s synthesis reports that in adults with overweight or obesity, resistance training alone increases lean mass and produces modest average reductions in body-fat percentage and total fat mass, although it is not usually sufficient for large reductions in scale weight by itself. citeturn21search9
Therefore, using extra caffeine because “more adrenaline burns more fat” is not well supported. A better rationale for caffeine during a fat-loss phase is indirect: if a tolerated dose helps preserve training quality during an energy deficit, it may help maintain performance and lean mass. That is distinct from claiming that a bigger catecholamine spike itself materially accelerates fat loss. citeturn17view0turn19search18
Recovery and sleep
Exercise-generated epinephrine is coupled to the exercise bout and falls during recovery; caffeine remains pharmacologically active for hours. From a recovery perspective, that is arguably the most important practical difference between the two exposures. citeturn19search0turn17view6
A systematic review and meta-analysis found that caffeine reduced subsequent total sleep time by approximately 45 minutes, lowered sleep efficiency by about seven percentage points, increased sleep-onset latency by about nine minutes and reduced deep sleep. A more recent dose-timing experiment found that a 400-mg dose could impair sleep even when administered many hours before bedtime, whereas 100 mg was considerably less disruptive under the tested conditions. citeturn19search3turn19search7
This creates a trade-off often missed in pre-workout discussions: an evening caffeine dose might improve today’s repetitions while degrading the sleep that supports tomorrow’s recovery. For hypertrophy, strength development and repeated training quality, the net outcome can therefore be worse even when the acute session feels better. The sleep trade-off becomes increasingly important with larger doses and later training. citeturn19search3turn19search7
For most healthy adults, the FDA cites 400 mg/day as an amount not generally associated with negative effects, but susceptibility varies greatly and this figure should not be confused with an optimal single dose, an athletic recommendation, or a safety threshold for people with hypertension, panic disorder, pregnancy, certain medications or cardiovascular disease. citeturn21search1
Evidence limitations, research priorities, and sources
The central limitation of this literature is that the numerical comparison the question naturally invites—“How much adrenaline does coffee generate versus a lifting session?”—has not been tested cleanly enough for a single universal ratio. The most informative caffeine catecholamine studies and the most informative resistance-exercise catecholamine studies use different subjects, laboratories, assays, exercise modes, sampling schedules, pretest caffeine restrictions and training statuses. citeturn17view0turn22view0turn17view5
Sample sizes in the mechanistic literature are also small. The Graham caffeine dose-response study had eight trained athletes; French et al. studied ten resistance-trained men; Pullinen et al. studied eight men; the Fatouros resistance study included 17 young untrained men. These experiments can establish physiological mechanisms but are not ideal for deriving population-wide reference ranges. citeturn17view0turn22view0turn17view5turn17view7
A further methodological problem is hemoconcentration. Dense lifting shifts plasma volume; in the classic Kraemer protocol, estimated plasma volume fell by roughly 17–21% after exercise. A measured rise in plasma hormone concentration can therefore partly reflect less plasma water rather than a proportional increase in total hormone secretion. Modern studies need rigorous plasma-volume correction or direct secretion/clearance techniques. citeturn17view6
“Peak epinephrine” is particularly vulnerable to sampling error because many resistance studies draw blood only at baseline and after the final set. French et al.’s serial pre-set sampling showed how rapidly concentrations can change throughout a workout and demonstrated that a substantial part of the response may occur before exercise. Future protocols should sample densely enough to distinguish anticipatory, intraworkout and recovery kinetics. citeturn22view0
There is also an important coffee-versus-pure-caffeine gap. Mechanistic endocrine work disproportionately uses anhydrous caffeine because it is easy to blind and dose precisely. Real-world coffee has variable caffeine content and additional bioactive compounds, making direct extrapolation imperfect. citeturn19search1turn21search1
Research on women, older adults, people with hypertension and people with anxiety disorders remains much thinner than research in young healthy men, particularly for plasma catecholamine kinetics during resistance exercise. Modern caffeine performance studies are becoming more sex-inclusive, but endocrine equivalence between sexes has not been established simply because performance effects look similar. citeturn19search18turn21search0
The most useful next experiment would be a adequately powered randomized crossover study with four conditions—placebo/rest, caffeine/rest, placebo/resistance exercise and caffeine/resistance exercise—in the same participants. Caffeine arms should include at least 3 and 6 mg/kg; resistance conditions should manipulate load and effort independently, for example matched-volume work at 60% versus 80% 1RM and failure versus non-failure sets. Epinephrine and norepinephrine should be sampled before ingestion, every 15 minutes before exercise, immediately before key sets, immediately after sets, and repeatedly for at least an hour afterward. Continuous beat-to-beat blood pressure and ECG, glucose, lactate, glycerol/free fatty acids, insulin, cortisol, perceived arousal, anxiety, RPE and actual performance should be acquired concurrently.
Such work should stratify or model habitual caffeine exposure, CYP1A2 genotype, sex, age, fitness, hypertension and anxiety susceptibility and correct hormone values for plasma-volume shifts. That design would answer the question far more rigorously than comparing isolated studies from different laboratories.
Selected primary human studies and authoritative reviews used in this report
Graham TE, Spriet LL. Metabolic, catecholamine, and exercise performance responses to various doses of caffeine. Journal of Applied Physiology, 1995. Dose-response trial showing 3 mg/kg ergogenicity without significant epinephrine elevation and progressively greater hormonal responses at 6–9 mg/kg. citeturn17view0
Graham TE, Spriet LL. Performance and metabolic responses to a high caffeine dose during prolonged exercise. Journal of Applied Physiology, 1991. Quantified pre-exercise and during-exercise epinephrine after 9 mg/kg caffeine. citeturn23search0
Lindinger MI et al. Caffeine attenuates the exercise-induced increase in plasma potassium concentration in humans. Journal of Applied Physiology, 1993. Reported exhaustion epinephrine of 10.1 versus 5.3 nmol/L with caffeine versus placebo. citeturn23search1
Battram DS et al. The effect of caffeine on glucose kinetics in humans— influence of adrenaline. Journal of Physiology, 2005. Human clamp experiment separating caffeine’s metabolic effects from those caused by equivalent adrenaline concentrations. citeturn18search2turn18search9
French DN et al. Anticipatory responses of catecholamines on muscle force production. Journal of Applied Physiology, 2007. Serial sampling during six sets of ten squats at 80% 1RM; demonstrated anticipatory and intraworkout epinephrine increases. citeturn22view0
Pullinen T, Huttunen P, Komi PV. Plasma catecholamine responses and neural adaptation during short-term resistance training. European Journal of Applied Physiology, 2000. Quantified postexercise adrenaline during repeated exhaustive resistance sessions. citeturn17view5turn23search2
Kraemer WJ et al. Physiologic responses to heavy-resistance exercise with very short rest periods. International Journal of Sports Medicine, 1987. Demonstrated acute catecholamine/cortisol increases and substantial plasma-volume shifts after dense whole-body resistance exercise. citeturn17view6
Fatouros I et al. Acute resistance exercise results in catecholaminergic rather than hypothalamic-pituitary-adrenal axis stimulation during exercise in young men. Stress, 2010. Demonstrated a resistance protocol in which catecholamines rose while cortisol did not significantly increase. citeturn17view7
MacDougall JD et al. Arterial blood pressure response to heavy resistance exercise. Journal of Applied Physiology, 1985, and subsequent work on determinants of the response. Direct intra-arterial studies establishing the large transient BP excursions of heavy lifting and the importance of effort and breathing mechanics. citeturn16search0turn16search12
Bangsbo J et al. Acute and habitual caffeine ingestion and metabolic responses to steady-state exercise. 1992. Demonstrated attenuation of the caffeine/exercise epinephrine response after six weeks of habitual caffeine exposure. citeturn20search3
Lovallo WR et al. Caffeine stimulation of cortisol secretion across the waking hours in relation to caffeine intake levels. Psychosomatic Medicine, 2005. Controlled evidence that daily caffeine attenuates but does not abolish the cortisol response. citeturn20search2turn20search6
Guest NS et al. International Society of Sports Nutrition position stand: caffeine and exercise performance. Journal of the International Society of Sports Nutrition, 2021. Broad synthesis supporting 3–6 mg/kg, with possible benefits at lower doses and little rationale for 9 mg/kg. citeturn19search1
Grgic J et al. Effects of caffeine on resistance exercise: a review of recent research. Sports Medicine, 2021. Resistance-specific synthesis suggesting low doses around 2–3 mg/kg can be comparably ergogenic to larger doses and that effects are broadly present in both sexes. citeturn19search18
Xiao Y et al. Effects of acute caffeine intake on muscular power during resistance exercise: systematic review and meta-analysis. 2025. Found improvements in resistance-exercise movement velocity and power. citeturn19search22
Montalvo-Alonso JJ et al. CYP1A2 genotype and the ergogenic effect of acute caffeine intake on resistance exercise. 2026. Contemporary evidence that 3 mg/kg can improve resistance performance in trained men and women while illustrating interindividual/genetic variability. citeturn21search0
Mesas AE et al. The effect of coffee on blood pressure and cardiovascular disease in hypertensive individuals: a systematic review and meta-analysis. American Journal of Clinical Nutrition, 2011. Quantified the acute pressor effect of 200–300 mg caffeine in hypertension. citeturn20search1
American Heart Association. Resistance Exercise Training in Individuals With and Without Cardiovascular Disease: 2023 Update. Circulation, 2024 issue. Authoritative synthesis of chronic cardiovascular, glycemic and safety effects of resistance training. citeturn21search9
American Heart Association. Caffeine and Cardiovascular Disease: A Scientific Statement. Circulation, 2026. Current assessment of caffeine’s cardiovascular and arrhythmia implications. citeturn16search1
Gardiner C et al. The effect of caffeine on subsequent sleep: a systematic review and meta-analysis. Sleep Medicine Reviews, 2023. Quantified reductions in sleep duration and efficiency and changes in sleep architecture after caffeine. citeturn19search3
Klevebrant L et al. Effects of caffeine on anxiety and panic attacks in patients with panic disorder: a systematic review and meta-analysis. General Hospital Psychiatry, 2022. Demonstrated pronounced caffeine sensitivity at high doses in panic-disorder populations. citeturn16search2
NCBI/StatPearls. Physiology, Catecholamines; Physiology, Autonomic Nervous System; and Physiology, Noradrenergic Synapse. Authoritative summaries of adrenal-medullary synthesis, sympathetic innervation and acetylcholine-triggered chromaffin-cell secretion. citeturn20search0turn20search4turn20search8