Place · Level 3

Exercise as medicine

肌肉是内分泌器官 · 一次收缩同时调代谢、大脑、炎症 — 这是它能改善 10+ 个慢病终点的分子根源

Story path

1Mechanisms — muscle as an endocrine organMechanisms — muscle as an endocrine organ
2DepressionDepression
3Type 2 diabetesType 2 diabetes
4HypertensionHypertension
5InsomniaInsomnia
6SarcopeniaSarcopenia
7Cancer survivorshipCancer survivorship

Chapter 1

Mechanisms — muscle as an endocrine organ

The 'Fit' in Fitnuhealth lands on this island: exercise is the #1 health lever after nutrition, and in many situations more important than nutrition. The reason it improves so many seemingly unrelated endpoints at once — cardiovascular, glucose metabolism, mood, cognition, musculoskeletal — has a single molecular root: contracting muscle isn't just an engine, it's an endocrine organ.

The main lines (Hawley 2014 review):

Myokines: contracting muscle releases signaling molecules into the blood, including (anti-inflammatory in the exercise context), irisin (inducing BDNF and fat browning), and BAIBA. They cross the blood-brain barrier or act in border regions of the brain — the material basis for exercise improving mood and cognitionGlucose uptake that bypasses insulin: muscle contraction shifts the AMP/ ratio, activates , and recruits GLUT4 glucose transporters straight to the cell membrane, with no insulin signaling needed at all. This is why exercise still lowers blood glucose in insulin-resistant people, and the training effect lasts 24-48 hoursBDNF upregulation: brain-derived neurotrophic factor supports neural plasticity and hippocampal neurogenesis — the shared pathway behind exercise's effects on depression, cognition, and dementiaAnti-inflammatory: long-term exercise downregulates chronic low-grade inflammation (inflammaging), lowering baseline IL-6 and — which doesn't contradict the transient IL-6 rise during acute exercise, that being a hormesis signalMitochondrial biogenesis: endurance stimulus upregulates the PGC-1α pathway, increasing mitochondrial density and fat oxidation capacity, the cellular basis for improved aerobic capacity
This is why exercise isn't merely 'working out' — it's the broadest, cheapest, and strongest medical intervention. The islands below are its prescriptions for specific diseases.

Dose — the all-cause mortality curve

The all-cause mortality dose-response for exercise is clear (Lee 2012 Lancet plus Wen 2011 Lancet): going from fully sedentary up to moderate exercise (150 min/week) lowers all-cause mortality 30-35%; adding more (300 min/week) gains another 5-10%; very high volumes (over 600 min/week) plateau, no longer linear. Extreme volumes (marathon, triathlon) show some mildly negative signals in studies (arrhythmias, atrial fibrillation), but the risk is far less than being sedentary.

A few frequently misread claims, addressed in passing:

'10,000 steps a day': the number comes from 1960s Japanese pedometer marketing, not science. The real evidence is that 4,000-7,000 steps/day already lowers mortality significantly (Paluch 2022 meta)'Just walking is enough': wrong — without strength training, muscle keeps wasting away after 50'Yoga alone is enough': covers flexibility and balance, but cardio and strength are insufficient'HIIT is the most efficient': effective for those with an aerobic base, while complete beginners need to build the base first
If you could pick one intervention to stick to for 30 years, it's not a supplement, nor any 'anti-inflammatory diet' — it's 150 minutes of moderate aerobic per week plus 2-3 resistance training sessions.

Open this on the map

Chapter 2

Depression

The evidence for exercise on depression is strong enough to make it a first-line intervention, not an 'adjunctive measure'. The Cooney 2013 Cochrane review pooled 39 RCTs and 2,326 people, with an effect size of Cohen's d = -0.62 (moderate-to-large) — the same range as the -0.5 to -0.7 typically reported for SSRI antidepressants. Schuch 2016 meta (33 RCTs) revisited it and reached the same conclusion; after correcting for publication bias the effect shrank slightly but remained significant.

So in mild-to-moderate depression, exercise can stand alongside medication and psychotherapy as first-line; severe depression still needs the medication + therapy + exercise combination.

It works through stacking pathways: myokines ( anti-inflammatory, irisin inducing BDNF); BDNF upregulation supporting hippocampal neurogenesis; Default Mode Network (DMN) reorganization, where depressive 'rumination' correlates with DMN overactivation and exercise adjusts its activity pattern; axis normalization, with long-term exercise lowering baseline cortisol; plus the genuinely real post-exercise endorphins and endocannabinoids (the 'runner's high').

Prescription (based on Cooney 2013 plus ACSM 2018): moderate intensity (50-70% max HR, can hold a conversation but a bit short of breath — high-intensity HIIT may acutely worsen anxiety, so not first choice), 3-5 times per week, 30-60 minutes each, 150+ minutes total. Walking, running, cycling, and strength training all work (a resistance-training-only meta shows equivalent effect). Peak effect typically arrives at 8 weeks; the first 2-3 weeks may feel like 'this isn't doing anything', which is normal — don't stop.

Open this on the map

Chapter 3

Type 2 diabetes

The core problem in T2D is that muscle's insulin response is blunted, so blood glucose can't enter muscle and stays in the blood. The magic of exercise: muscle contraction recruits GLUT4 glucose transporters to the cell membrane via a pathway that doesn't require insulin, pulling glucose in directly. Richter 2013 (Physiol Rev) lays out the mechanism — contraction shifts the AMP/ ratio, activates , and triggers GLUT4 translocation, bypassing insulin signaling entirely, with the training effect lasting 24-48 hours.

The clinical data carry weight: the DPP trial (2002 NEJM) showed that in high-risk people, diet plus 150 min/week of exercise cut progression to T2D by 58% — more effective than metformin alone (-31%); the Look AHEAD trial showed strength plus aerobic plus weight loss significantly improving HbA1c, with some patients reversing to non-diabetic status; Reynolds 2016 (Diabetologia) showed that 15 minutes of walking immediately after a meal lowered the postprandial glucose peak by about 30% (vs sitting after a meal).

Prescription (Colberg 2016, ADA official consensus): aerobic 150 min/week moderate or 75 min/week vigorous, no more than 2 consecutive rest days; resistance training 2-3×/week; interrupt sitting by standing and moving 3+ minutes every 30 minutes (especially post-meal). Post-meal walking is the most practical single intervention — 15-minute walks after each of three meals control glucose better than one 45-minute walk, and suit working people who can't carve out a long block.

Note for glucose-lowering medication users: insulin or sulfonylurea users have elevated hypoglycemia risk during and 6-12 hours after training — test glucose pre-training, eat carbs first if below 5 mmol/L, carry glucose tabs during; metformin is training-safe; reducing or stopping medication should be physician-guided, not self-decided.

Open this on the map

Chapter 4

Hypertension

Three exercise types have different blood-pressure effects (Cornelissen 2013, J Am Heart Assoc meta, 93 RCTs, n=5,223):

Aerobic training (3-5×/week, 30-60 min each, 12+ weeks): SBP down 5-8 mmHg, DBP down 3-5 mmHg, with larger effects in hypertensive patientsStrength training (2-3×/week): SBP down 4 mmHg, DBP down 3 mmHg. The early worry that 'weight training raises BP' has been overturned — there's an acute spike during the lift, but the chronic effect is reductionFlexibility training: no significant effect, not recommended as a standalone BP intervention
The most surprising is isometric exercise (Smart 2019, J Hypertens meta): isometric handgrip training (2 min × 4 sets × 3×/week × 8 weeks) lowers SBP 10.4 mmHg and 6.5 mmHg — an effect size that even exceeds the average single-drug effect of an ACE inhibitor or diuretic (~-8 mmHg). Wall-sit isometrics have a similar effect. The mechanism is reduced peripheral resistance, improved endothelial function, and increased nitric-oxide release. Isometric exercise takes under 8 minutes per session, with equipment as simple as a grip device or a wall — adherence is very high, and it suits older adults and those with joint issues especially well.

There's also an acute after-effect: after one moderate-intensity aerobic session, BP stays 5-7 mmHg lower for 22 hours (Pescatello 2004). Cautions: uncontrolled severe hypertension (≥180/110) should see a doctor first; in strength training, exhale on exertion and don't hold your breath (the Valsalva maneuver acutely raises BP markedly); β-blocker users should set intensity by RPE rather than heart rate; don't stop medication on your own.

Open this on the map

Chapter 5

Insomnia

Exercise's effect on insomnia is strong and stable enough to serve as a first-line intervention, not a 'let's try and see' option. Kredlow 2015 (J Behav Med) meta (66 RCTs, acute plus chronic vs control) shows that 4-12 weeks of sustained exercise lowers sleep-onset latency 13 minutes (from 30 to 17), increases total sleep time 19 minutes, improves subjective sleep quality (PSQI) by 0.31 SD (moderate effect), raises the N3 deep-sleep proportion, and improves next-day alertness and mood.

This effect size is similar to cognitive behavioral therapy (CBT-I, the first-line non-drug treatment for insomnia), better than melatonin alone, and comparable to zolpidem-class hypnotics but without the drug side effects. Banno 2018 (PeerJ) meta (9 RCTs in diagnosed insomnia patients) showed PSQI dropping 4.3 after 4-8 weeks of exercise (from 12.5 to 8.2 — clinically significant).

Timing determines how well it works:

Morning training (6-9 AM) is best: bright light, temperature rhythm, and endorphins act together to consolidate the circadian rhythm and advance sleep onset by about 1.5-2 hours — insomniacs should try this slot firstAfternoon (3-7 PM) is second best: peak strength and performance, neutral for sleep, the most practical slot on workdaysEvening 7-9 PM: moderate intensity (walking, jogging, yoga) doesn't hurt sleep, but HIIT plus heavy lifting delays sleep 1-2 hours because core temperature, cortisol, and adrenaline need 2-3 hours to come downLate 9-11 PM: avoid high intensity — even if you can fall asleep, deep-sleep quality clearly drops
Practical for insomniacs: first choice is moderate-intensity morning training 30+ minutes for 4-8 weeks (the first two weeks may feel useless, don't quit early); if evening is your only option, keep it moderate, finish 3+ hours before bed, and take a warm shower afterward (the temperature-rise-then-fall signal aids sleep).

Open this on the map

Chapter 6

Sarcopenia

EWGSOP2 (Cruz-Jentoft 2019, Lancet) redefined sarcopenia as 'progressive systemic skeletal muscle mass loss accompanied by strength or performance decline' — age-related but NOT inevitable. Diagnosis centers on strength (handgrip below 27 kg for men, 16 kg for women, or 5× sit-to-stand over 15 seconds), graded by low muscle mass and poor performance. Prevalence is 5-13% over age 70, 11-50% over 80; clinical consequences are roughly a 3× rise in falls, 2-4× in hip fracture, and 2-3× in mortality.

Core mechanisms: anabolic resistance (elderly muscle's protein-synthesis response is blunted; the same leucine stimulus produces only 60-70% of the young response); motor units are lost at ~1%/year from age 30, with type-II fast-twitch fibers lost faster; IGF-1, testosterone, and growth hormone decline with age; chronic low-grade inflammation accelerates breakdown.

As of now no drug is approved specifically for sarcopenia; the only strong-evidence intervention is resistance training plus protein:

Progressive resistance training (mandatory): 2-3×/week, 60-80% 1RM moderate-to-high intensity (not 'gentle elderly training'), 8-12 reps × 2-3 sets, compounds first, with 12+ weeks before significant muscle-mass improvement is visibleProtein: total 1.0-1.2 g/kg/day (the elderly RDA of 0.8 g/kg is insufficient), 25-30 g high-quality protein per meal (leucine threshold 2.5-3 g), with concentrating it into one or two meals much less effective; animal protein has better bioavailability than plantVitamin D plus calcium: D at 800+ IU/day, with Bischoff-Ferrari 2019 showing D plus resistance training doubles the strength gain of resistance training alone
A few marketing traps: HMB adds little on top of resistance training plus protein; 'elderly-specific protein powder' is marketing, ordinary whey works the same; a 'senior creatine version' isn't needed, regular creatine works. 'Too old to train' is wrong — Fiatarone 1994 proved significant hypertrophy even at 90.

Open this on the map

Chapter 7

Cancer survivorship

The Campbell 2019 ACSM Roundtable (a 17-country expert consensus covering 8 common cancer types) is the primary source for current 'cancer plus exercise' guidance, and its core position is a reversal: the standard 1990s advice was 'cancer patients should rest to conserve energy', while ACSM 2019 plus the 2024 update completely overturned this, writing instead 'avoid inactivity; sustain training where possible and progress recovery gradually'.

Unified prescription: aerobic 150 min/week moderate or 75 min/week vigorous; resistance training 2-3×/week, 8-12 reps, moderate loads; flexibility at least 2×/week; balance training is required for older adults or chemo-induced peripheral neuropathy (CIPN) patients.

Supporting clinical outcomes (Mishra 2012 Cochrane meta, n=3,694): cancer-related fatigue SMD -0.85 (a large effect, the strongest fatigue-intervention evidence at the time); quality of life SMD +0.40; depression and anxiety -0.30 each; breast and colon cancer survivors meeting exercise guidelines have 28-44% lower mortality (Schmid 2014 meta); training groups maintain chemotherapy dose intensity better with fewer interruptions. The mechanisms are immune upregulation (NK cell activity up), insulin/IGF-1 pathway modulation, lower chronic inflammation, improved gut microbiota, and increased cardiovascular reserve.

Schmitz 2009 (NEJM) is the most influential RCT in this field — it overturned the 30-year prohibition on resistance training for breast cancer survivors: 141 breast-cancer-related lymphedema survivors did progressive resistance training, and lymphedema not only didn't worsen but symptom severity actually dropped ~50%, retiring the old post-op '5 lb weight limit on that arm' rule.

Red lines (stop training and seek medical care immediately if these appear): new symptoms (chest pain, severe headache, sudden swelling) — stop training immediately and seek care; platelets below 50 × 10⁹/L — avoid impact or high-intensity resistance training; core temperature reaching 38.5°C — don't train; bone-metastasis patients — avoid impact activities and require physician guidance. The physician's default is now 'train', not 'don't train', but still confirm safety with your doctor before starting.

Open this on the map