HUMAN HEART
The Opening
The Opening
Put your hand on your chest. Right of center, slightly left. Feel it.
Thump. Thump. Thump.
That's not a sound effect. That's a 300-gram knot of muscle that has been squeezing without rest since before you were born — before your brain had thoughts, before your lungs took their first breath, before you had a name. It has beaten roughly 100,000 times today. It will beat 100,000 times tomorrow. It will not stop until you die.
Here is the engineering brief:
You need a pump that:
├── Delivers 5 liters of fluid per minute at rest, 25 liters during a sprint
├── Runs 24/7 for 80 years with zero scheduled maintenance
├── Generates its own electrical signals — no external power source
├── Routes fluid through 100,000 kilometers of pipe with zero leaks
├── Adjusts output in under one second based on demand
├── Operates inside a sealed cavity with no cooling, no oil changes
└── Costs about 300 grams of muscle tissue and a fistful of electricity
No pump engineered by humans comes close. The best industrial pumps need seals replaced every few thousand hours. The heart runs for 3 billion cycles without a seal change.
You also need it to not stop. Ever. Not during sleep. Not during surgery. Not during a car crash, a fever, a panic attack, or a race. The moment it stops for four minutes, the brain — fed by the same system — begins to die.
The margin for error is zero.
Let's build it.
───
PHASE 1: Pump Without Stopping
You need to move blood. All of it. Continuously. How much, how fast, and how do you do it with two pumps instead of one?
The first design question: what flow rate do you need?
Your body has roughly 37 trillion cells. Every one of them needs oxygen. Oxygen is delivered by red blood cells. Red blood cells travel in plasma. So the question becomes: how much plasma, moving how fast, carries enough oxygen to keep 37 trillion cells alive?
The answer has a name: cardiac output.
CO = SV × HR
SV = stroke volume (blood ejected per beat)
HR = heart rate (beats per minute)
At rest:
SV = 70 mL/beat (left ventricle, typical adult)
HR = 70 beats/min
CO = 70 × 70 = 4,900 mL/min ≈ 4.9 L/min
During intense exercise (trained athlete):
SV → 180 mL/beat
HR → 190 beats/min
CO → 180 × 190 = 34.2 L/min (7× resting output)The heart doesn't just beat faster under stress — each beat ejects more blood (Frank-Starling mechanism: stretch the muscle harder, it contracts harder). Both dials turn up simultaneously.
4.9 liters per minute. In context:
├── A kitchen faucet at full: ~8 L/min
├── Your resting heart: 4.9 L/min
├── An Olympic cyclist's heart under load: ~34 L/min
└── A Tour de France rider's heart weighs 400g — 33% larger than average
Now here is the crucial design decision: you need two pumps, not one.
PULMONARY CIRCUIT SYSTEMIC CIRCUIT
(right side of heart) (left side of heart)
┌──────────────┐
┌──→ lungs ──┐ │ body: head │
│ │ │ arms, gut │
──→ [R.ventricle] [L.atrium] ──→ [L.ventricle] ──→ aorta ──→ ...
│ │ │ │
[R.atrium] ←──── [lungs return] ←── body ←─────── │
│ │
└───────────────── venous return ───────────────┘
Right side pressure: 25 mmHg systolic (gentle push to nearby lungs)
Left side pressure: 120 mmHg systolic (hard push to reach your toes)If you used one pump at 120 mmHg for both circuits, you'd blow out the delicate capillaries in the lungs. The two-pump design lets each side run at the pressure appropriate for its circuit. Both sides pump identical volumes — every drop that leaves the right side must come back through the left.
Why 120 mmHg on the left, not 200 or 60?
You need enough pressure to push blood to the highest point in your body — the brain, roughly 35 cm above the heart. Pressure required to lift a column of blood 35 cm:
P = ρ × g × h
ρ = blood density ≈ 1060 kg/m³
g = 9.8 m/s²
h = 0.35 m (heart to brain, standing)
P = 1060 × 9.8 × 0.35 = 3,635 Pa = 27 mmHg
Add resistance of the entire arterial tree: ~70-80 mmHg
Add safety margin for exertion: ~15 mmHg
─────────
Total: ~120 mmHg ← exactly what we measureThe 120/80 "normal" blood pressure isn't arbitrary. It's the minimum pressure needed to perfuse the brain reliably, plus the resistance of the vascular tree, plus a working margin. See the Gravity article: fighting a fluid column uphill costs exactly ρgh.
When you stand up suddenly, the blood column shift is ~27 mmHg — your pressure sensors (carotid baroreceptors) detect this in milliseconds and signal the heart to increase output before you faint. If that reflex fails — orthostatic hypotension — you black out. Your body is performing a continuous hydraulic calculation every second you're upright.
DESIGN SPEC COMPLETE:
├── Cardiac output: CO = SV × HR = 70 mL × 70 bpm = 4.9 L/min (rest)
├── Range: 4.9–34 L/min (7× dynamic range)
├── Two pumps: right (25 mmHg) for lungs, left (120 mmHg) for body
├── Left side pressure = hydrostatic head + vascular resistance + margin
└── Flow is conserved: right output = left output (every beat, always)
DESIGN SPEC COMPLETE:
├── Cardiac output: CO = SV × HR = 70 mL × 70 bpm = 4.9 L/min (rest)
├── Range: 4.9–34 L/min (7× dynamic range)
├── Two pumps: right (25 mmHg) for lungs, left (120 mmHg) for body
├── Left side pressure = hydrostatic head + vascular resistance + margin
└── Flow is conserved: right output = left output (every beat, always)
───
PHASE 2: Start Your Own Engine
Your heart has no brain. No nerve tells it to beat. Sever every nerve to the heart and it keeps beating. How?
This is the pacemaker problem. Every engine needs a starter. Car engines need a battery. Industrial pumps need a motor controller. But your heart starts itself — and restarts itself if something goes wrong — using nothing but ion channels and geometry.
The trick is a cluster of cells in the upper right atrium called the sinoatrial node. These cells are electrically unstable by design.
Why unstable? The ion channel derivation.
A typical cell has a resting potential of about −70 mV (inside negative). It takes an external signal to make it fire. SA node cells are different: their resting potential is only −60 mV, and they have a special "funny current" (I_f) — a channel that lets Na⁺ and K⁺ leak IN at rest, slowly pushing the voltage upward.
Voltage
(mV)
0 ─────────────────────────────────────────
╱╲ ╱╲ ╱╲
-40 ╱ ╲ ╱ ╲ ╱ ╲
╱ ╲ ╱ ╲ ╱ ╲
-60 ╲ ╱ ╲ ╱
╲ ╱ ← funny current (I_f) drifts upward
-70 ╲ ╱ Ca²⁺ channels open at -40 mV (threshold)
╲ ╱ K⁺ channels open at 0 mV (repolarize)
╲──╱
Phase 4 (drift): I_f + I_Ca,T pushing membrane toward threshold
Phase 0 (upstroke): I_Ca,L fires — rapid depolarization
Phase 3 (fall): I_K repolarizes the cell
Repeat. Automatically. Forever.The SA node fires at ~70 bpm because that's the rate at which the funny current (I_f) depolarizes the cell to threshold. Adrenaline speeds this drift (heart rate up). Vagal tone slows it. The pacemaker rate is a dial, not a fixed clock.
The conduction system: why does the heart contract in sequence, not all at once?
If all 300 grams of heart muscle fired simultaneously, the atria and ventricles would squeeze at the same instant. Blood would have nowhere to go. You need a delay — atria first, then ventricles.
SA node ← origin (fires ~70/min)
│
▼ spreads through atria
[ATRIA CONTRACT] (P wave on ECG)
│
▼
AV node ← mandatory 120 ms delay
(acts as gatekeeper — limits rate)
│
▼
Bundle of His
╱ ╲
Left bundle Right bundle
branch branch
╲ ╱
Purkinje fibers ← fast conduction (4 m/s)
│
▼
[VENTRICLES CONTRACT] (QRS on ECG)
SA → AV delay: ~120 ms
AV → ventricular: ~40 ms
Total atria-to-ventricle delay: ~160 ms ← time for ventricles to fillThe AV node delay is the key design feature. 160 ms is exactly the time the atria need to push their extra 30% of blood volume (atrial kick) into the ventricles before they contract. Remove this delay — as in certain arrhythmias — and you lose atrial kick and cardiac output drops 30%.
What happens if the SA node fails?
The system has backups. Ranked by speed:
Primary: SA node 60-100 bpm ← fastest, takes command
Backup 1: AV node 40-60 bpm ← if SA fails
Backup 2: Bundle of His 30-40 bpm ← if AV fails
Backup 3: Purkinje fibers 20-30 bpm ← last resort
Backup 4: Ventricular cells 15-20 bpm ← barely survivableEach backup has a slower intrinsic rate — so the fastest pacemaker (SA node) always wins by default. An artificial pacemaker implant mimics this: it fires at 70 bpm and the heart follows. It fires only if the SA node is slower — it "backs up the backup."
The Purkinje fibers conduct at 4 m/s — faster than any other cardiac tissue. This ensures the ventricles depolarize from apex to base in under 80 ms, producing a coordinated squeeze that ejects blood upward into the aorta rather than just squashing it sideways.
DESIGN SPEC UPDATED:
├── SA node: self-oscillating via I_f (funny current) — no external trigger needed
├── Rate control: adrenaline (faster drift) / vagal tone (slower drift)
├── AV delay: ~160 ms — mandatory pause for ventricular filling
├── Purkinje fibers: 4 m/s conduction — fast enough to coordinate 300g of muscle in 80 ms
└── Pacemaker hierarchy: 5 levels of backup — SA → AV → His → Purkinje → ventricle
DESIGN SPEC UPDATED:
├── SA node: self-oscillating via I_f (funny current) — no external trigger needed
├── Rate control: adrenaline (faster drift) / vagal tone (slower drift)
├── AV delay: ~160 ms — mandatory pause for ventricular filling
├── Purkinje fibers: 4 m/s conduction — fast enough to coordinate 300g of muscle in 80 ms
└── Pacemaker hierarchy: 5 levels of backup — SA → AV → His → Purkinje → ventricle
───
PHASE 3: Push Through 100,000 km
You have a pump. You have flow. Now you need pipes — 100,000 kilometers of them — running through a body that is, at most, 2 meters tall. And every cell in that body needs to be within 50 micrometers of a capillary.
The pipe network is the hard part. You cannot just run big tubes everywhere — big tubes can't reach inside a muscle fiber. You cannot run only tiny tubes — tiny tubes have catastrophically high resistance. You need a branching network that transitions smoothly from the aorta (2.5 cm diameter) to capillaries (8 micrometers diameter) — a 3,000× reduction in diameter.
The governing equation is Hagen-Poiseuille's law. Let's derive it.
Deriving flow in a pipe from first principles.
Imagine a fluid element inside a cylindrical pipe, radius R, length L, under pressure difference ΔP. The pressure force pushing it forward must equal the viscous drag holding it back.
Consider a fluid cylinder of radius r inside a pipe of radius R.
Pressure force forward: F_P = ΔP × π r²
Viscous shear at radius r — Newton's viscosity law:
τ = η × dv/dr (shear stress = viscosity × velocity gradient)
F_v = τ × 2πrL = η × (dv/dr) × 2πrL
At steady state: F_P = F_v
ΔP × π r² = η × (dv/dr) × 2πrL
Rearranging:
dv/dr = − (ΔP / 2ηL) × r
Integrating from r to R (velocity = 0 at wall):
v(r) = (ΔP / 4ηL) × (R² − r²) ← parabolic velocity profile
Total flow Q = ∫ v(r) × 2πr dr from 0 to R:
Q = (π × ΔP × R⁴) / (8 × η × L)
Or written as resistance:
R_flow = 8ηL / (πR⁴) ← resistance grows as r⁴The r⁴ dependence is the most important number in vascular biology. It is NOT intuitive. Doubling the radius of a vessel does not double the flow — it multiplies flow by 16. Halving the radius does not halve the flow — it divides it by 16.
The r⁴ effect in practice.
Coronary artery stenosis (plaque buildup). The artery narrows. How much does flow drop?
Healthy artery: radius = 1.0 (normalized) → flow = 1.0
25% narrowed: radius = 0.75 → flow = 0.75⁴ = 0.32 (−68%)
50% narrowed: radius = 0.50 → flow = 0.50⁴ = 0.0625 (−94%)
75% narrowed: radius = 0.25 → flow = 0.25⁴ = 0.0039 (−99.6%)
You can lose 25% of your coronary artery diameter and
lose TWO THIRDS of your blood flow to the heart muscle.
Symptoms often don't appear until >70% stenosis — by which
time flow is already reduced by 97%.This is why coronary artery disease is the world's leading killer. The disease is silent — mild narrowing reduces flow dramatically, but the heart compensates by growing collateral vessels. By the time symptoms appear, the narrowing is severe.
Capillary transit: the final 50 micrometers.
At the capillary bed, tube diameter shrinks to 8 μm — barely wide enough for a red blood cell (6-8 μm) to squeeze through. The red blood cell deforms, flattening to a biconcave disc to fit. Why so narrow?
Diffusion time for O₂ across distance d:
t = d² / (2 × D)
D(O₂ in tissue) = 2 × 10⁻⁹ m²/s
d = 50 μm (max tissue distance from capillary):
t = (50 × 10⁻⁶)² / (2 × 2 × 10⁻⁹)
t = 2.5 × 10⁻⁹ / 4 × 10⁻⁹
t = 0.625 seconds ← just fast enough for capillary transit time (~1 s)
d = 200 μm (if capillaries were farther apart):
t = (200 × 10⁻⁶)² / (2 × 2 × 10⁻⁹) = 10 seconds ← too slow, cell diesOxygen diffuses slowly in tissue. The capillary spacing of 50 μm is the maximum distance that keeps diffusion time under 1 second — the transit time of a red blood cell. This is why capillary density in cardiac muscle is roughly one capillary per muscle fiber. The heart can't afford any cell being farther than 50 μm from oxygen supply.
Comparison ladder — the 100,000 km of vessels:
├── Earth's circumference: 40,075 km
├── Earth to Moon: 384,400 km
├── Your blood vessel network: ~100,000 km (2.5 Earth circumferences)
├── Aorta diameter: 25 mm → aorta length: ~40 cm
├── Capillary diameter: 0.008 mm → capillary count: ~37 billion
└── Packed into a 70-kg body. Every vessel is within 2 mm of the next.
DESIGN SPEC UPDATED:
├── Poiseuille: Q = πΔPR⁴ / 8ηL — flow scales as radius to the FOURTH power
├── 50% narrowing = 94% flow reduction (r⁴ makes stenosis catastrophic)
├── Capillary diameter = 8 μm — sized by O₂ diffusion time (t = d²/2D)
├── Max cell-to-capillary distance: ~50 μm — keeps O₂ transit under 1 second
└── Network: 100,000 km in a 70-kg body — 2.5 times Earth's circumference
DESIGN SPEC UPDATED:
├── Poiseuille: Q = πΔPR⁴ / 8ηL — flow scales as radius to the FOURTH power
├── 50% narrowing = 94% flow reduction (r⁴ makes stenosis catastrophic)
├── Capillary diameter = 8 μm — sized by O₂ diffusion time (t = d²/2D)
├── Max cell-to-capillary distance: ~50 μm — keeps O₂ transit under 1 second
└── Network: 100,000 km in a 70-kg body — 2.5 times Earth's circumference
───
PHASE 4: Handle the Pressure Wave
Place two fingers on your wrist. Feel the pulse. That's not blood arriving — blood moves at about 0.5 m/s in arteries. The pulse is a pressure wave, and it travels at 6-9 m/s. You feel the wave long before the blood arrives.
This distinction matters enormously. The heart doesn't push a plug of blood through the vessels like a piston pushing water through a rigid pipe. Arteries are elastic — they stretch when pressure rises and recoil when it falls. The pressure wave is more like a sound wave: information propagating faster than the medium moves.
Deriving pulse wave velocity.
For a thin-walled elastic tube, the speed at which a pressure wave propagates is the Moens-Korteweg equation. Let's build it.
A pressure pulse arrives at a section of artery. It expands the wall radially. The elastic recoil of the wall creates the restoring force that propagates the wave.
Elastic artery: wall tension T = E × h × ε
E = elastic modulus of vessel wall
h = wall thickness
ε = strain (fractional extension)
For a tube radius r under pressure P:
Hoop stress: σ = Pr/h → strain ε = Pr/(Eh)
Wall compliance: C = dV/dP = 2πrL × r/(Eh) = 2πr²L/(Eh)
Pulse wave velocity (fluid-filled elastic tube):
c = √(E × h / (2 × ρ × r))
E = elastic modulus (Pa)
h = wall thickness (m)
ρ = blood density (kg/m³)
r = vessel radius (m)
Aorta (young adult):
E ≈ 300,000 Pa, h ≈ 2 mm, ρ ≈ 1060 kg/m³, r ≈ 12.5 mm:
c = √(300,000 × 0.002 / (2 × 1060 × 0.0125))
c = √(600 / 26.5)
c = √22.6 ≈ 4.8 m/s ← matches clinical measurement (~5 m/s in young aorta)The key dependence: c ∝ √E. If E doubles (stiffer artery), wave speed increases by √2 = 41%. An 80-year-old's aorta has E roughly 10× higher than a 20-year-old's. Pulse wave velocity: 5 m/s at 20 → 12-15 m/s at 80. Clinicians measure pulse wave velocity as a direct readout of arterial stiffness and cardiovascular risk.
The Windkessel model: why the aorta is elastic by design.
If arteries were rigid, every heartbeat would send a violent pressure spike to the capillaries — vessels so fragile (one-cell-thick walls) they'd rupture. Instead, the aorta acts as an elastic reservoir.
RIGID PIPE: ELASTIC AORTA (Windkessel):
Heart beats: Heart beats:
↓↓↓↓↓↓ ↓↓↓↓↓↓
──────────────── Pressure ╭────────╮ ← aorta stretches (stores energy)
│ │ │ BULGE │
│ SPIKE: 120 │ │ │
│ THEN: 0 mmHg │ then heart rests:
└────────────────┘ ╰────────╯ ← recoil pushes steady flow
Pressure: smooth 80-120 mmHg
(diastolic pressure IS the recoil pressure)
Tissue sees: pulsatile death. Tissue sees: steady flow. Alive.The 80 in "120/80" blood pressure is diastolic pressure — the pressure maintained by the aorta's elastic recoil while the heart is relaxing. It's the stored energy of the stretched aorta pushing blood forward between beats. Remove aortic elasticity and diastolic pressure falls to zero — every capillary bed gets slammed then starved 70 times per minute.
Arterial stiffening with age: the cascade.
Elastin fibers (which give arteries their stretch) cannot be replaced once they fragment. Collagen (stiff) gradually takes over. The result:
Age Elastic Modulus PWV Systolic BP Diastolic BP
────────────────────────────────────────────────────────────────
20 ~300 kPa ~5 m/s ~115 mmHg ~75 mmHg
40 ~600 kPa ~7 m/s ~120 mmHg ~80 mmHg
60 ~1,500 kPa ~10 m/s ~130 mmHg ~80 mmHg
80 ~3,000 kPa ~14 m/s ~150 mmHg ~70 mmHg
← systolic climbs (stiffer pipe, bigger spike)
→ diastolic falls (less recoil energy)Isolated systolic hypertension in the elderly is not a disease — it is physics. The stiff aorta can no longer store and release energy smoothly. Systolic pressure rises because the pulse energy has nowhere to go. Diastolic pressure falls because there is no recoil to maintain it. Pulse wave velocity measurement in clinical settings now predicts cardiovascular events better than blood pressure alone.
DESIGN SPEC UPDATED:
├── Pulse wave ≠ blood movement: wave at ~5 m/s, blood at ~0.5 m/s
├── PWV: c = √(Eh / 2ρr) — stiffness sets wave speed
├── Windkessel: elastic aorta converts pulsatile flow to steady capillary flow
├── Diastolic pressure = stored elastic energy of the aorta
└── Aging: E increases 10×, PWV doubles — isolated systolic hypertension is pure physics
DESIGN SPEC UPDATED:
├── Pulse wave ≠ blood movement: wave at ~5 m/s, blood at ~0.5 m/s
├── PWV: c = √(Eh / 2ρr) — stiffness sets wave speed
├── Windkessel: elastic aorta converts pulsatile flow to steady capillary flow
├── Diastolic pressure = stored elastic energy of the aorta
└── Aging: E increases 10×, PWV doubles — isolated systolic hypertension is pure physics
───
PHASE 5: Feed the Pump Itself
The heart pumps blood to the body. But who pumps blood to the heart? This is the bootstrap problem — and it has a single point of failure that kills 9 million people per year.
Every organ gets its blood supply from branches of the aorta. The heart is no exception. Immediately after the aortic valve — the first two branches off the aorta — are the coronary arteries.
┌─── Aorta
│
┌──┴──┐
│Heart│
│ │
LCA ──────┤ ├────── RCA
(left │ │ (right
coronary) └─────┘ coronary)
│ │
LAD (anterior) Supplies right
LCx (lateral) ventricle
│ │
Supplies left ~20% of
ventricle heart muscle
~80% of heart
muscle (including
the interventricular
septum)
LCA = Left Coronary Artery
LAD = Left Anterior Descending (most important vessel in the body)
LCx = Left Circumflex
RCA = Right Coronary ArteryThe LAD — left anterior descending artery — supplies most of the left ventricle, the heart's main pumping chamber. It is sometimes called "the widow-maker." Total occlusion of the LAD produces a massive anterior myocardial infarction with rapid cardiogenic shock. No other single vessel in the body carries such concentrated lethality.
The heart's oxygen extraction problem.
Most organs extract about 25-30% of the oxygen in their blood supply. They keep a reserve — if demand rises, they just extract more. The heart does not have this option.
Organ O₂ extraction at rest O₂ reserve
─────────────────────────────────────────────────────────
Skeletal muscle 25-30% large reserve
Kidney 18% large reserve
Brain 35% moderate
Heart muscle 70-75% almost none
During exercise — heart needs more O₂:
Other organs → extract more from existing flow
Heart → must INCREASE CORONARY FLOW (can only extract ~80% max)
Coronary flow at rest: ~250 mL/min
Coronary flow during max exercise: ~1,000-1,200 mL/min (5× increase)Because the heart already extracts 70-75% of oxygen at rest, it cannot meet increased demand by simply extracting more. The ONLY way to supply more O₂ to a working heart is to increase coronary blood flow. This is why coronary stenosis is so dangerous: the reserve capacity that saves other organs during stress simply does not exist in the heart.
Stenosis math: when does flow become inadequate?
Recall from Phase 3: Q ∝ r⁴. A stenosed coronary artery provides insufficient flow during exercise before it restricts flow at rest. Let's quantify the threshold.
At rest, heart needs 250 mL/min.
Exercise, heart needs 1,000 mL/min (4× reserve).
Stenosis narrows radius to fraction f of normal:
Flow ∝ f⁴
Q_stenosed / Q_normal = f⁴
For angina at rest: need flow < 250/250 = 1.0
f⁴ < 1.0 → always (angina at rest = critical)
For angina only on exercise:
need flow adequate at rest but < 1,000/1,000 during exercise
If f = 0.75: flow = 0.75⁴ = 0.32 × normal
Rest need: 250 mL/min — 0.32 × 800 max = 256 mL/min (just adequate)
Exercise need: 1,000 mL/min — 0.32 × 800 = 256 mL/min (catastrophically short)
f = 0.75 (25% stenosis) = stable angina on exertion
f = 0.50 (50% stenosis) = angina at rest
f = 0 (complete occlusion) = heart attackThe clinical presentation of chest pain on exertion (stable angina) but not at rest means stenosis has reduced coronary radius by roughly 25-40%. The heart has already lost its entire exercise reserve. The artery looks 75% open on an angiogram but is functionally near its limit.
One further wrinkle: the coronary arteries fill during diastole (heart relaxing), not systole (heart squeezing). When the ventricle contracts, it squeezes its own coronary vessels shut — intramyocardial pressure rises above aortic pressure. This is why a racing heart (high rate, less time in diastole) is at higher ischemic risk: less time for coronary filling per minute.
DESIGN SPEC UPDATED:
├── Coronary arteries: first branches off aorta — heart supplies itself first
├── Myocardial O₂ extraction: 70-75% at rest (no reserve — only option is more flow)
├── Coronary reserve: 4-5× resting flow — all consumed by maximal exercise
├── 25% stenosis (r×0.75) → 68% flow reduction via r⁴ — exercise becomes ischemic
└── Coronary filling: diastole only — high heart rate reduces fill time per minute
DESIGN SPEC UPDATED:
├── Coronary arteries: first branches off aorta — heart supplies itself first
├── Myocardial O₂ extraction: 70-75% at rest (no reserve — only option is more flow)
├── Coronary reserve: 4-5× resting flow — all consumed by maximal exercise
├── 25% stenosis (r×0.75) → 68% flow reduction via r⁴ — exercise becomes ischemic
└── Coronary filling: diastole only — high heart rate reduces fill time per minute
───
PHASE 6: Scale Across Species
A shrew's heart beats 1,200 times per minute. An elephant's heart beats 30 times per minute. A blue whale's, 4-8 beats per minute. Different animals, wildly different rates — but they all get roughly the same number of heartbeats in a lifetime. Why?
This is one of the most beautiful patterns in all of biology: Kleiber's law combined with the scaling of heart rate. From a mouse to a whale — 7 orders of magnitude in body mass — a single equation describes them all.
Deriving the heart rate scaling law.
Start with metabolic rate. Max Kleiber measured metabolic rate B across 60+ species in 1932 and found:
B = 70 × M^0.75 (metabolic rate in kcal/day, M in kg)
Why 0.75? Competing theories:
├── Fractal supply network (West, Brown, Enquist 1997):
│ networks that minimize transport losses in branching trees → 3/4
├── Surface area hypothesis (older): B ∝ M^(2/3) — doesn't fit data
└── Empirical: 0.75 fits mouse-to-whale data far better than 2/3
For the heart specifically:
Cardiac output CO must supply O₂ at the metabolic rate.
CO ∝ B ∝ M^0.75
But CO = SV × HR, and stroke volume scales with heart SIZE:
Heart mass ∝ M^1.0 (heart is constant fraction ~0.6% of body mass)
Heart volume ∝ M^1.0
Stroke volume ∝ M^1.0
Therefore:
HR = CO / SV ∝ M^0.75 / M^1.0 = M^(-0.25)
Empirically fitted:
HR = 241 × M^(-0.25) (M in kg, HR in beats per minute)This is not an approximation. It is a law that holds across 8 orders of magnitude of body mass — from the 2-gram shrew to the 150,000-kg blue whale — within a factor of 2. The exponent -0.25 is as fundamental to biology as the inverse square law is to physics.
Test it against real animals.
Animal Mass (kg) Predicted HR Measured HR
──────────────────────────────────────────────────────
Shrew 0.003 828 bpm ~1,200 bpm
Mouse 0.020 560 bpm ~500-600 bpm
Rat 0.300 330 bpm ~300-400 bpm
Cat 4.0 191 bpm ~150-200 bpm
Human 70.0 83 bpm ~70-80 bpm
Horse 500.0 44 bpm ~36-44 bpm
Elephant 5,000.0 24 bpm ~25-35 bpm
Blue whale 150,000.0 11 bpm ~4-8 bpm
Human: 241 × 70^(-0.25) = 241 × 0.345 = 83 bpm ← predicted
Resting heart rate of a healthy adult: ~70 bpm ← measured. Close.The shrew diverges high because small mammals lose heat faster (surface area/volume ratio — see the Dinosaur article) and need higher metabolic rates to stay warm, pushing heart rate above the pure scaling prediction. Cold-blooded animals don't follow this law at all — it applies only to endotherms.
Every mammal gets 1 billion heartbeats.
Lifespan scales as:
T ∝ M^0.25 (years, empirically)
Heart rate scales as:
HR ∝ M^(-0.25)
Lifetime beats:
N = HR × T ∝ M^(-0.25) × M^0.25 = M^0 = constant
Calculating the constant:
Human: 80 years × 365 days × 24 hours × 60 min × 70 bpm
= 2.95 × 10⁹ ≈ 3 billion beats
Mouse: 2 years × 525,600 min/year × 500 bpm
= 5.26 × 10⁸ ≈ 500 million beats
Elephant: 70 years × 30 bpm × 525,600 min/year
= 1.1 × 10⁹ ≈ 1 billion beats
Range: ~1–3 billion beats across all mammals.
Humans get roughly 3× what most mammals get — our lifespan
runs ahead of the scaling prediction (brains, medicine, low predation).The billion-beat rule is remarkable: a shrew lives 2 years at 1,200 bpm and gets roughly 1 billion beats. An elephant lives 70 years at 30 bpm and gets roughly 1 billion beats. The metabolic clock runs at the same number of beats regardless of size. Humans cheat this budget — medicine, cooked food, and reduced predation push us to 3 billion, but our hearts wear out on the same schedule as any other mammal's.
Comparison ladder — heart rate vs. mass:
├── Blue whale heart: size of a small car, beats 4-8 times/min
├── Human heart: size of a fist, 70 bpm
├── Hummingbird heart: 1,200 bpm in flight — faster than a dentist's drill
├── Etruscan shrew: fastest mammal heartbeat — 1,511 bpm (measured)
└── All of them: same physics, same equation, same ~1 billion beats
DESIGN SPEC UPDATED:
├── Kleiber's law: B = 70 × M^0.75 (metabolic rate)
├── Heart rate scaling: HR = 241 × M^(-0.25) (holds across 8 orders of magnitude)
├── Lifespan scaling: T ∝ M^0.25 — same exponent, opposite sign, cancels
├── Lifetime beats: N = HR × T ∝ M^0 = constant (~1-3 billion for all mammals)
└── Humans cheat the budget: medicine extends lifespan beyond the scaling prediction
DESIGN SPEC UPDATED:
├── Kleiber's law: B = 70 × M^0.75 (metabolic rate)
├── Heart rate scaling: HR = 241 × M^(-0.25) (holds across 8 orders of magnitude)
├── Lifespan scaling: T ∝ M^0.25 — same exponent, opposite sign, cancels
├── Lifetime beats: N = HR × T ∝ M^0 = constant (~1-3 billion for all mammals)
└── Humans cheat the budget: medicine extends lifespan beyond the scaling prediction
───
PHASE 7: Survive a Heart Attack
A coronary artery closes. Complete occlusion. Blood stops flowing to a section of heart muscle. You have about 20 minutes before that muscle is dead. What happens inside, second by second, and can you stop it?
This is the engineering problem of failure mode analysis. The heart attack is not one event — it is a cascade, and understanding the cascade is what makes treatment possible.
The ischemia cascade: 20 minutes to death.
Time 0: Coronary artery occludes (usually plaque rupture + clot)
0–10 sec: O₂ in myocyte mitochondria exhausted
ATP synthesis stops
Cells switch to anaerobic glycolysis
10–60 sec: ATP drops from 5 mM to ~2 mM
Na/K-ATPase pump fails (needs ATP)
Na⁺ floods into cells → cells swell
K⁺ leaks out → extracellular K⁺ rises → arrhythmia risk
1–5 min: Contractile failure — affected region stops squeezing
ECG shows ST elevation (injury current)
Ca²⁺ floods into cells (ATP-dependent Ca pump fails)
Reversible damage window begins to close
5–20 min: Acidosis (lactate accumulates)
Mitochondria swell, inner membrane disrupts
Cytochrome C leaks → apoptosis signal
20–40 min: Cell death begins (coagulation necrosis)
Irreversible infarction expanding from subendocardium
(innermost layer — highest O₂ demand, least collateral supply)
1–6 hours: Wavefront of necrosis expands outward through wall
Full-thickness infarction (transmural MI) if untreated
6+ hours: Necrotic tissue replaced by fibrotic scar
Scar cannot contract — permanent loss of pump functionThe "door-to-balloon" time — from hospital arrival to opening the artery with a stent — is the single most important metric in cardiac emergency medicine. Every 30 minutes of additional ischemia kills roughly 0.5% more of the heart muscle. Modern targets: under 90 minutes. World-class centers: under 60 minutes.
The reperfusion paradox: restoring blood flow can kill cells.
Counterintuitively, restoring flow to ischemic tissue causes additional damage — reperfusion injury. The mechanism:
During ischemia:
├── Na⁺ overloads cell (Na/K pump failure)
├── Cell acidifies (pH drops to ~6.5)
└── Mitochondria stressed but intact — still salvageable
Blood returns — suddenly:
├── pH normalized rapidly → Na/H exchanger exports H⁺, imports Na⁺
├── Na⁺ overload worsens → Na/Ca exchanger exports Na⁺, imports Ca²⁺
├── Ca²⁺ floods mitochondria → mitochondrial permeability transition pore (mPTP) opens
├── mPTP opening → uncontrolled energy dissipation → cell death
└── Reactive oxygen species (ROS) burst: O₂ returning reacts with damaged mitochondria
Result: up to 40-50% of final infarct size occurs at the moment of reperfusion —
not during the ischemia itself.
Paradox: restoring flow causes damage.
Solution: cool the tissue (slows all chemistry) + mPTP inhibitors (experimental)Reperfusion injury explains why simply opening the artery is not always sufficient. Hypothermia before reperfusion (cooling the heart to 32-34°C) slows the Ca²⁺ cascade and reduces mPTP opening. The treatment is now in trials but already used in cardiac surgery: cardioplegia solutions stop the heart in cold, high-potassium, high-Mg fluid — deliberately induced cardiac arrest to prevent reperfusion injury during bypass.
CPR physics: what does chest compression actually do?
Bystander CPR is performed on a patient with no cardiac output. What mechanism sustains circulation?
Two competing mechanisms:
CARDIAC PUMP THEORY (direct compression):
Compressing sternum ~5 cm at 100/min squeezes heart between
sternum and spine → valves close → blood ejected
Requires: sufficient compression depth to compress heart
THORACIC PUMP THEORY (chest-as-bellows):
Compression raises intrathoracic pressure
→ all intrathoracic vessels compressed equally
→ venous valves prevent backflow
→ net forward flow out of chest
In practice: BOTH mechanisms contribute
Compression generates ~30% of normal cardiac output
(just enough to keep brain alive, not enough for consciousness)
Rate: 100-120/min (optimized for blood flow vs. recoil time)
Depth: 5-6 cm (deeper = more output, but rib fracture risk)
Force required:
F = P × A (pressure × chest area)
Pressure needed: ~50-80 mmHg above ambient
Chest area: ~600 cm² = 0.06 m²
F = 80 mmHg × 0.06 m² = 10,666 Pa × 0.06 ≈ 640 N
≈ 65 kg of force per compression — a heavy person's full weight leaning downThe 30% cardiac output from CPR buys time — it keeps the brainstem oxygenated but cannot restore consciousness. Survival depends on defibrillation (if VF) or coronary reperfusion (if MI) within minutes. CPR is a bridge, not a cure. Without it, that bridge collapses and the brain dies before help arrives.
DESIGN SPEC UPDATED:
├── Ischemia cascade: 0 sec (O₂ exhausted) → 60 sec (pump failure) → 20 min (cell death)
├── Critical window: 20 minutes from occlusion to irreversible cell death begins
├── Reperfusion injury: 40-50% of infarct size occurs AT reperfusion (Ca²⁺ flood + ROS)
├── CPR: ~30% cardiac output via thoracic pump + direct cardiac compression
└── CPR force: ~640 N (65 kg of weight) at 100-120/min, 5-6 cm depth
DESIGN SPEC UPDATED:
├── Ischemia cascade: 0 sec (O₂ exhausted) → 60 sec (pump failure) → 20 min (cell death)
├── Critical window: 20 minutes from occlusion to irreversible cell death begins
├── Reperfusion injury: 40-50% of infarct size occurs AT reperfusion (Ca²⁺ flood + ROS)
├── CPR: ~30% cardiac output via thoracic pump + direct cardiac compression
└── CPR force: ~640 N (65 kg of weight) at 100-120/min, 5-6 cm depth
───
PHASE 8: Replace It
The heart fails. Irreversibly. 6 million Americans live with heart failure right now. The waiting list for a transplant is 3,500 people long. Only ~3,500 donor hearts become available per year. The math doesn't work. So we build alternatives.
This is the most ambitious engineering problem in medicine: replacing an organ that has no off switch, no backup mode, and no tolerance for a single missed beat.
Option 1: The LVAD (Left Ventricular Assist Device)
An LVAD is a pump implanted inside the chest, with its inlet cannula sewn into the left ventricle apex and its outlet cannula grafted to the aorta. A cable exits through the skin to an external battery controller.
The biological heart: The LVAD:
──────────────────────────────────────────────
Pulsatile flow (70/min) Continuous flow (centrifugal)
Self-powered (ATP) External battery (12-16 hr)
Self-repairing No self-repair
Self-regulating (Frank-Starling) Speed controlled by algorithm
300 g tissue ~400 g titanium impeller
No mechanical failure for 80 yr Bearing wear, thrombosis, infection
Modern LVAD (HeartMate 3):
├── Flow: 3-10 L/min (continuously adjustable)
├── Speed: 3,000-9,000 rpm
├── Power: ~4-6 watts
├── Battery life: 12-16 hours (requires daily charging)
└── 2-year survival: ~80% ← competitive with heart transplant
The HeartMate 3 has no mechanical bearings (magnetically levitated rotor).
No contact = no wear = no thrombosis from mechanical surfaces.
Mean time to failure measured in YEARS, not months.Modern LVADs create near-continuous (non-pulsatile) flow. Patients on LVADs often have no palpable pulse — blood pressure cuffs don't work. They are measured by Doppler. The kidney, brain, and liver function normally on non-pulsatile flow, disproving a century-old assumption that pulsatile flow was physiologically essential.
Option 2: Xenotransplantation — pig hearts.
Pigs are anatomically similar to humans. Pig heart valves have been used successfully in humans for 50+ years. Why not the whole heart?
The problem is the hyperacute rejection response. A human immune system, confronted with a pig organ, destroys it within minutes. The culprit: a pig-specific sugar molecule on cell surfaces, Gal antigen (galactose-alpha-1,3-galactose), which human antibodies attack immediately.
Pig genome edits needed to survive in a human:
KNOCK OUT (remove):
├── GGTA1 — removes Gal antigen (main antibody target)
├── CMAH — removes Neu5Gc (another pig-specific antigen)
└── B4GalNT2 — removes third pig carbohydrate antigen
KNOCK IN (add human genes):
├── hCD46 — human complement regulatory protein
├── hCD55 — inhibits complement attack
├── hCD59 — blocks membrane attack complex
├── hTBM — human thrombomodulin (prevents clotting)
└── hEPCR — endothelial protein C receptor
Result: 10-gene edited pig → "humanized" heart
January 2022: David Bennett Sr. received the first 10-gene-edited pig
heart. He survived 60 days — remarkable for a first attempt.
Cause of death: porcine cytomegalovirus (CMV) in donor organ.
Next attempt: pre-screen donors for CMV. Expected lifespan: months to years.The 2022 xenotransplant used a pig from Revivicor with 10 genome edits. The surgery itself was straightforward — pig heart anatomy is nearly identical to human. The barrier is immunology: every cell surface molecule that differs between pig and human is a target for rejection. Genome editing is systematically removing these differences one gene at a time.
Option 3: The transplant shortage — the math that kills.
Annual heart failure new diagnoses: ~900,000
Patients currently on transplant list: ~3,500
Donor hearts available per year: ~3,500
Transplants performed per year: ~3,700
Patients who die waiting per year: ~20-30% of wait-list per year
(median wait time: 6-12 months)
If xenotransplantation succeeds:
Pig heart supply: theoretically unlimited (weeks to raise a donor)
Pig heart cost: ~$500,000 (editing + screening) — vs. $1M+ for transplant system
Production time: 6 months (pig growth) vs. waiting for brain-dead donor
LVAD bridge to transplant:
60-70% of transplant recipients now have an LVAD implanted first.
LVAD buys 1-3 years while waiting for a donor.
Without LVAD: ~30-50% die before a heart becomes available.
With LVAD: ~80% survive to transplant.The engineering solution to the shortage is not more donors — it is making donors irrelevant. LVADs already replace transplants in a growing fraction of patients (destination therapy — permanent LVAD instead of bridge to transplant). If xenotransplantation achieves 1+ year survival, the transplant list becomes a historical artifact.
Full artificial heart: total replacement.
The SynCardia Total Artificial Heart (TAH) replaces both ventricles with pneumatically driven polyurethane sacs. Two external compressor drives push air to inflate the sacs 80 times per minute. The patient carries a ~6 kg driver system.
Metric Biological heart SynCardia TAH
──────────────────────────────────────────────────────
Weight 300 g ~160 g (implanted)
Driver system None (self-powered) ~6 kg external
Flow 4-9 L/min 4-9.5 L/min
Pulse 70 bpm (adaptive) 70-150 bpm (fixed)
Infection risk Low High (skin exit site)
1-year survival ~85% ~55% (bridge use)
Mobility Full Restricted (driver size)
Battery life Infinite (ATP) 4 hours portable
Used as bridge to transplant only — not destination therapy.
~1,700 patients worldwide have received it.The total artificial heart solves the most extreme heart failure — both ventricles gone, no LVAD possible — but the 6 kg driver is the constraint. Miniaturization is the frontier: wireless power transmission through skin (transcutaneous energy transfer) could eliminate the external cable entirely, reducing infection risk to near zero.
DESIGN SPEC UPDATED — THE FULL SYSTEM:
├── LVAD: magnetically levitated continuous-flow pump, 4-6W, 2-yr survival ~80%
├── Non-pulsatile flow: kidney/brain/liver function normally — no pulse needed
├── Xenotransplantation: 10-gene-edited pig heart — hyperacute rejection solved (CMV problem remains)
├── Transplant shortage: 3,700 transplants/year vs. 900,000 new failure patients
└── Total artificial heart: bridge use only — driver weight/infection remain barriers
DESIGN SPEC UPDATED — THE FULL SYSTEM:
├── LVAD: magnetically levitated continuous-flow pump, 4-6W, 2-yr survival ~80%
├── Non-pulsatile flow: kidney/brain/liver function normally — no pulse needed
├── Xenotransplantation: 10-gene-edited pig heart — hyperacute rejection solved (CMV problem remains)
├── Transplant shortage: 3,700 transplants/year vs. 900,000 new failure patients
└── Total artificial heart: bridge use only — driver weight/infection remain barriers
───
FULL MAP
Human Heart
├── Phase 1: Pump Without Stopping
│ ├── Cardiac output: CO = SV × HR = 70 mL × 70 bpm = 4.9 L/min (rest)}
│ ├── Range: 4.9–34 L/min (7× dynamic range)}
│ ├── Two pumps: right (25 mmHg) for lungs, left (120 mmHg) for body}
│ ├── Left side pressure = hydrostatic head + vascular resistance + margin}
│ ├── Flow is conserved: right output = left output (every beat, always)}
│ ├── Cardiac output: CO = SV × HR = 70 mL × 70 bpm = 4.9 L/min (rest)}
│ ├── Range: 4.9–34 L/min (7× dynamic range)}
│ ├── Two pumps: right (25 mmHg) for lungs, left (120 mmHg) for body}
│ ├── Left side pressure = hydrostatic head + vascular resistance + margin}
│ └── Flow is conserved: right output = left output (every beat, always)}
│
├── Phase 2: Start Your Own Engine
│ ├── SA node: self-oscillating via I_f (funny current) — no external trigger needed}
│ ├── Rate control: adrenaline (faster drift) / vagal tone (slower drift)}
│ ├── AV delay: ~160 ms — mandatory pause for ventricular filling}
│ ├── Purkinje fibers: 4 m/s conduction — fast enough to coordinate 300g of muscle in 80 ms}
│ ├── Pacemaker hierarchy: 5 levels of backup — SA → AV → His → Purkinje → ventricle}
│ ├── SA node: self-oscillating via I_f (funny current) — no external trigger needed}
│ ├── Rate control: adrenaline (faster drift) / vagal tone (slower drift)}
│ ├── AV delay: ~160 ms — mandatory pause for ventricular filling}
│ ├── Purkinje fibers: 4 m/s conduction — fast enough to coordinate 300g of muscle in 80 ms}
│ └── Pacemaker hierarchy: 5 levels of backup — SA → AV → His → Purkinje → ventricle}
│
├── Phase 3: Push Through 100,000 km
│ ├── Earth's circumference: 40,075 km
│ ├── Earth to Moon: 384,400 km
│ ├── Your blood vessel network: ~100,000 km (2.5 Earth circumferences)
│ ├── Aorta diameter: 25 mm → aorta length: ~40 cm
│ ├── Capillary diameter: 0.008 mm → capillary count: ~37 billion
│ ├── Packed into a 70-kg body. Every vessel is within 2 mm of the next.
│ ├── Poiseuille: Q = πΔPR⁴ / 8ηL — flow scales as radius to the FOURTH power}
│ ├── 50% narrowing = 94% flow reduction (r⁴ makes stenosis catastrophic)}
│ ├── Capillary diameter = 8 μm — sized by O₂ diffusion time (t = d²/2D)}
│ ├── Max cell-to-capillary distance: ~50 μm — keeps O₂ transit under 1 second}
│ ├── Network: 100,000 km in a 70-kg body — 2.5 times Earth's circumference}
│ ├── Poiseuille: Q = πΔPR⁴ / 8ηL — flow scales as radius to the FOURTH power}
│ ├── 50% narrowing = 94% flow reduction (r⁴ makes stenosis catastrophic)}
│ ├── Capillary diameter = 8 μm — sized by O₂ diffusion time (t = d²/2D)}
│ ├── Max cell-to-capillary distance: ~50 μm — keeps O₂ transit under 1 second}
│ └── Network: 100,000 km in a 70-kg body — 2.5 times Earth's circumference}
│
├── Phase 4: Handle the Pressure Wave
│ ├── Pulse wave ≠ blood movement: wave at ~5 m/s, blood at ~0.5 m/s}
│ ├── PWV: c = √(Eh / 2ρr) — stiffness sets wave speed}
│ ├── Windkessel: elastic aorta converts pulsatile flow to steady capillary flow}
│ ├── Diastolic pressure = stored elastic energy of the aorta}
│ ├── Aging: E increases 10×, PWV doubles — isolated systolic hypertension is pure physics}
│ ├── Pulse wave ≠ blood movement: wave at ~5 m/s, blood at ~0.5 m/s}
│ ├── PWV: c = √(Eh / 2ρr) — stiffness sets wave speed}
│ ├── Windkessel: elastic aorta converts pulsatile flow to steady capillary flow}
│ ├── Diastolic pressure = stored elastic energy of the aorta}
│ └── Aging: E increases 10×, PWV doubles — isolated systolic hypertension is pure physics}
│
├── Phase 5: Feed the Pump Itself
│ ├── Coronary arteries: first branches off aorta — heart supplies itself first}
│ ├── Myocardial O₂ extraction: 70-75% at rest (no reserve — only option is more flow)}
│ ├── Coronary reserve: 4-5× resting flow — all consumed by maximal exercise}
│ ├── 25% stenosis (r×0.75) → 68% flow reduction via r⁴ — exercise becomes ischemic}
│ ├── Coronary filling: diastole only — high heart rate reduces fill time per minute}
│ ├── Coronary arteries: first branches off aorta — heart supplies itself first}
│ ├── Myocardial O₂ extraction: 70-75% at rest (no reserve — only option is more flow)}
│ ├── Coronary reserve: 4-5× resting flow — all consumed by maximal exercise}
│ ├── 25% stenosis (r×0.75) → 68% flow reduction via r⁴ — exercise becomes ischemic}
│ └── Coronary filling: diastole only — high heart rate reduces fill time per minute}
│
├── Phase 6: Scale Across Species
│ ├── Blue whale heart: size of a small car, beats 4-8 times/min
│ ├── Human heart: size of a fist, 70 bpm
│ ├── Hummingbird heart: 1,200 bpm in flight — faster than a dentist's drill
│ ├── Etruscan shrew: fastest mammal heartbeat — 1,511 bpm (measured)
│ ├── All of them: same physics, same equation, same ~1 billion beats
│ ├── Kleiber's law: B = 70 × M^0.75 (metabolic rate)}
│ ├── Heart rate scaling: HR = 241 × M^(-0.25) (holds across 8 orders of magnitude)}
│ ├── Lifespan scaling: T ∝ M^0.25 — same exponent, opposite sign, cancels}
│ ├── Lifetime beats: N = HR × T ∝ M^0 = constant (~1-3 billion for all mammals)}
│ ├── Humans cheat the budget: medicine extends lifespan beyond the scaling prediction}
│ ├── Kleiber's law: B = 70 × M^0.75 (metabolic rate)}
│ ├── Heart rate scaling: HR = 241 × M^(-0.25) (holds across 8 orders of magnitude)}
│ ├── Lifespan scaling: T ∝ M^0.25 — same exponent, opposite sign, cancels}
│ ├── Lifetime beats: N = HR × T ∝ M^0 = constant (~1-3 billion for all mammals)}
│ └── Humans cheat the budget: medicine extends lifespan beyond the scaling prediction}
│
├── Phase 7: Survive a Heart Attack
│ ├── Ischemia cascade: 0 sec (O₂ exhausted) → 60 sec (pump failure) → 20 min (cell death)}
│ ├── Critical window: 20 minutes from occlusion to irreversible cell death begins}
│ ├── Reperfusion injury: 40-50% of infarct size occurs AT reperfusion (Ca²⁺ flood + ROS)}
│ ├── CPR: ~30% cardiac output via thoracic pump + direct cardiac compression}
│ ├── CPR force: ~640 N (65 kg of weight) at 100-120/min, 5-6 cm depth}
│ ├── Ischemia cascade: 0 sec (O₂ exhausted) → 60 sec (pump failure) → 20 min (cell death)}
│ ├── Critical window: 20 minutes from occlusion to irreversible cell death begins}
│ ├── Reperfusion injury: 40-50% of infarct size occurs AT reperfusion (Ca²⁺ flood + ROS)}
│ ├── CPR: ~30% cardiac output via thoracic pump + direct cardiac compression}
│ └── CPR force: ~640 N (65 kg of weight) at 100-120/min, 5-6 cm depth}
│
├── Phase 8: Replace It
│ ├── LVAD: magnetically levitated continuous-flow pump, 4-6W, 2-yr survival ~80%}
│ ├── Non-pulsatile flow: kidney/brain/liver function normally — no pulse needed}
│ ├── Xenotransplantation: 10-gene-edited pig heart — hyperacute rejection solved (CMV problem remains)}
│ ├── Transplant shortage: 3,700 transplants/year vs. 900,000 new failure patients}
│ ├── Total artificial heart: bridge use only — driver weight/infection remain barriers}
│ ├── LVAD: magnetically levitated continuous-flow pump, 4-6W, 2-yr survival ~80%}
│ ├── Non-pulsatile flow: kidney/brain/liver function normally — no pulse needed}
│ ├── Xenotransplantation: 10-gene-edited pig heart — hyperacute rejection solved (CMV problem remains)}
│ ├── Transplant shortage: 3,700 transplants/year vs. 900,000 new failure patients}
│ └── Total artificial heart: bridge use only — driver weight/infection remain barriers}
│
└── CONNECTIONS
├── Kidney → blood pressure regulation, renin-angiotensin system, fluid balance
└── Nuclear Reactor → heat management, continuous operation, redundancy design
───