Place · Level 3
Renal System
100 万 nephron/肾 · 每天滤 180 L 重吸 99% · 血压 RAAS · 维 D 第二轮羟化 · EPO 红血球指挥棒
Story path
- 1Two 150-g filtersTwo 150-g filters
- 2Glomerulus · the 3-layer sieveGlomerulus · the 3-layer sieve
- 3RAAS · the BP loopRAAS · the BP loop
- 4D activation + EPOD activation + EPO
- 5'High protein hurts kidney' debunked'High protein hurts kidney' debunked
- 6Sodium · not just 'less salt'Sodium · not just 'less salt'
Chapter 1
Two 150-g filters
Two 150-g filters
The kidneys sit on the back of the waist on both sides — each ~150 g, together only 300 g. That's 0.5% of body weight, yet they receive 20–25% of cardiac output, making them the body's most blood-hungry organ (per unit weight, even more than the brain).
This isn't waste, because the kidney's job isn't 'absorb nutrients' — it's to re-filter all the body's blood. About 1.2 L of blood enters per minute, 125 mL of ultrafiltrate is produced, 99% is reabsorbed, leaving ~1 mL/min of final urine.
The functional unit is the nephron:
Each kidney has ~1 million nephrons, and the number doesn't grow over life (essentially fixed at birth)Each nephron is a glomerulus plus 4 sequential tubular segmentsAfter age 40, you lose ~6000 per year; aging, hypertension, and diabetes all accelerate this
Why do we have two kidneys: after a single kidney transplant, the remaining kidney hypertrophies compensatorily and GFR rises 40–50%, but the reserve is gone — which is why living kidney donors need strict long-term-risk screening. 'Two kidneys' doesn't equal '100% redundancy' — it's more like '~200% reserve + each doing half the work normally', and the reserve value mostly shows in old age.
This isn't waste, because the kidney's job isn't 'absorb nutrients' — it's to re-filter all the body's blood. About 1.2 L of blood enters per minute, 125 mL of ultrafiltrate is produced, 99% is reabsorbed, leaving ~1 mL/min of final urine.
The functional unit is the nephron:
Each kidney has ~1 million nephrons, and the number doesn't grow over life (essentially fixed at birth)Each nephron is a glomerulus plus 4 sequential tubular segmentsAfter age 40, you lose ~6000 per year; aging, hypertension, and diabetes all accelerate this
Why do we have two kidneys: after a single kidney transplant, the remaining kidney hypertrophies compensatorily and GFR rises 40–50%, but the reserve is gone — which is why living kidney donors need strict long-term-risk screening. 'Two kidneys' doesn't equal '100% redundancy' — it's more like '~200% reserve + each doing half the work normally', and the reserve value mostly shows in old age.
By the numbers
To see why the kidney is the 'silent master scheduler', lay out the numbers:Blood flow: ~1.2 L/min, ~1700 L of blood passes through both kidneys dailyFiltration rate (GFR): ~125 mL/min, ~180 L of ultrafiltrate per day — equivalent to filtering the body's entire plasma (3 L) 60 times/dayReabsorption: 99.4% — nearly all water, Na, glucose, amino acids reabsorbed; final urine ~1–2 L/dayOxygen use: about 8% of cardiac output's oxygen at rest, but Na/K-ATPase active transport eats 80% of the kidney's own adenosine triphosphate: The cell's universal energy currency — almost everything that costs energy spends it. — making the kidney exquisitely sensitive to ischemia
Beyond making urine, the kidney does these things every day:
Maintains blood pressure: renin–angiotensin–aldosterone system: A hormone chain that controls blood pressure and fluid — when tense it narrows vessels and holds water and sodium. + sodium-water balance (see RAAS scene)Maintains blood volume: ADH regulated by the pituitary, but acts at renal collecting ductsMaintains blood pH: excretes H⁺ + reabsorbs HCO₃⁻, working with the lungs; acidosis / alkalosis is corrected by the kidney within 24–48 hMaintains K⁺ balance: aldosterone + distal tubule + collecting duct (see RAAS). Excess potassium is excreted by the kidney — CKD hyperkalemia is a true emergency, fatal within hours without actionVitamin D second hydroxylation: converts liver-derived 25-hydroxyvitamin D: The storage form of vitamin D in blood — the number measured to check D status. into the 1,25(OH)₂D hormoneEPO (erythropoietin): see EPO sceneGluconeogenesis: contributes ~40% during prolonged fasting (matches the liver — most people don't realize this)
Painless kidneys don't mean healthy kidneys: pain receptors are in the kidney capsule, the parenchyma itself only hurts when stretched (stone blockage / acute pyelonephritis) or the capsule is pulled. Chronic kidney disease (CKD) in early-middle stages is entirely asymptomatic — no pain, no edema, no oliguria — by the time edema, anemia, and oliguria appear, it's usually already G4-G5. This is why it's called the silent killer, and why KDIGO 2024 recommends annual eGFR + urine albumin/creatinine ratio (UACR) testing in everyone with diabetes, hypertension, or age 60+.
Chapter 2
Glomerulus · the 3-layer sieve
Glomerulus · the 3-layer sieve
The glomerulus is a capillary tuft inside Bowman's capsule, one per nephron. The most important physical event in the urinary system happens here: plasma is sieved by molecular size and charge.
Three-layer sieve, from blood side to urine side:
1. Fenestrated endothelium — 70 nm pores, blocks blood cells and large particles
2. Basement membrane (GBM) — negatively-charged glycosaminoglycan layer, blocks albumin (~7 nm + negative); when damaged (diabetes / hypertension / IgA nephropathy), proteinuria appears
3. Podocytes — octopus-like cells with foot processes; the slit between them is ~25 nm, with key proteins nephrin / podocin; damage (minimal change disease / FSGS) causes severe proteinuria
So proteinuria isn't 'just drink more water to flush it out' — it's chemical proof that at least one of the three sieve layers is damaged, usually pointing to diabetic nephropathy / hypertensive nephropathy / primary glomerular disease; a nephrology referral is warranted.
GFR (glomerular filtration rate) is the core kidney number:
Normal adult ~90–120 mL/min/1.73m²<60 sustained 3 months → chronic kidney disease (CKD)<15 → end-stage renal disease (ESRD), requiring dialysis or transplant
Three-layer sieve, from blood side to urine side:
1. Fenestrated endothelium — 70 nm pores, blocks blood cells and large particles
2. Basement membrane (GBM) — negatively-charged glycosaminoglycan layer, blocks albumin (~7 nm + negative); when damaged (diabetes / hypertension / IgA nephropathy), proteinuria appears
3. Podocytes — octopus-like cells with foot processes; the slit between them is ~25 nm, with key proteins nephrin / podocin; damage (minimal change disease / FSGS) causes severe proteinuria
So proteinuria isn't 'just drink more water to flush it out' — it's chemical proof that at least one of the three sieve layers is damaged, usually pointing to diabetic nephropathy / hypertensive nephropathy / primary glomerular disease; a nephrology referral is warranted.
GFR (glomerular filtration rate) is the core kidney number:
Normal adult ~90–120 mL/min/1.73m²<60 sustained 3 months → chronic kidney disease (CKD)<15 → end-stage renal disease (ESRD), requiring dialysis or transplant
Creatinine + eGFR — the 99% misread
Serum creatinine (sCr) is the most common kidney marker on annual physicals, but most people don't know what it actually measures.Where creatinine comes from: muscle creatine spontaneously degrades non-enzymatically into creatinine at ~1–2% per day; in someone with stable muscle mass, creatinine production rate is nearly constant; it's freely filtered, not reabsorbed, barely secreted. So sCr is an indirect marker of kidney excretory function — but it has two pitfalls.
Pitfall 1: muscle mass changes the baseline. A muscular male gym-goer's sCr might be 100–110 μmol/L with normal kidney function; a frail elderly woman's sCr of 60 looks 'normal' but her actual eGFR might be 40 — severely underdiagnosed CKD; an amputee or chronically bedridden patient has lower creatinine production → low sCr → eGFR overestimated.
Pitfall 2: sCr is slow to react. After an acute 50% GFR drop, sCr usually takes 1–2 days to rise slowly to a meaningful level; in early acute kidney injury (AKI), sCr is still in the normal range while the kidney is already failing — which is why newer markers like NGAL / KIM-1 / cystatin C are being pushed in recent years.
eGFR (estimated GFR) is the modern standard: the 2021 CKD-EPI formula (no race coefficient) estimates from sCr + age + sex; more accurate than sCr alone, but still affected by muscle mass. Elderly, very thin, amputee, or athletic patients are best served by adding cystatin C for correction; emergency or AKI scenarios add NGAL or 24-hour creatinine clearance.
Clinical interpretation (KDIGO 2024):
G1 (eGFR ≥90) + no albuminuria: normalG2 (60–89) + no albuminuria: mild decline, still considered normalG3a (45–59): early CKD; about 50% of patients don't know they have itG3b (30–44): mid-stage CKD, needs nephrologyG4 (15–29): prepare for dialysis / transplant discussionsG5 (<15): end-stage renal disease
Reading kidney status from creatinine alone is the most common misreading on physical exam reports — always pair with eGFR, UACR, and trend over time.
After filtration · how the 4 tubular segments fine-tune 180 L
The glomerulus only does coarse filtration — 180 L of ultrafiltrate a day, containing everything: water, sodium, potassium, glucose, amino acids, bicarbonate. The real fine-tuning happens in the four downstream tubular segments, which reabsorb 99% of it while adjusting each electrolyte and acid-base item one by one. Understanding these four equals understanding most of a diuretics pharmacology chapter.Proximal tubule: the reabsorption workhorse — about 65% of sodium and water, nearly all glucose and amino acids, and about 80–90% of bicarbonate are recovered here. It's also where the kidney burns the most adenosine triphosphate: The cell's universal energy currency — almost everything that costs energy spends it. (Na/K-ATPase active transport), so it fails first under ischemia. Carbonic anhydrase here catalyzes HCO₃⁻ recovery — carbonic anhydrase inhibitors like acetazolamide act here (back to the acid-base topic, and the altitude prophylaxis drug)Loop of Henle: builds the hypertonic medullary gradient, the key to concentrating urine and conserving water. Loop diuretics (furosemide) block this segment's Na-K-2Cl transporter, the most powerful diuresis, used in heart failure and pulmonary edemaDistal tubule: fine-tunes sodium and calcium. Thiazide diuretics act here, one of the first-line drugs for hypertensionCollecting duct: the final adjustment station, listening to two hormones at once — aldosterone sets how much sodium is recovered and how much potassium and H⁺ are excreted (back to the renin–angiotensin–aldosterone system: A hormone chain that controls blood pressure and fluid — when tense it narrows vessels and holds water and sodium. scene); ADH (antidiuretic hormone) decides whether to insert the aquaporin water channels, i.e. how concentrated the final urine is
Two commonly confused things become clear here. One is acid-base: the lung adjusts pH within minutes by excreting CO₂, while the kidney uses these segments to excrete H⁺ and recover and regenerate HCO₃⁻ — slow but able to fully correct, completing in 24–48 hours (back to the acid-base lung-kidney duet, the other side of the same mechanism). The other is potassium: blood potassium is almost entirely excreted by the kidney, mainly regulated by aldosterone in the collecting duct — which is why the moment kidney function fails, hyperkalemia immediately becomes a true emergency (fatal arrhythmia possible within hours).
So if a physical exam shows glucose in the urine (with normal blood glucose), amino acids leaking, and low phosphate together, it may point to a proximal tubule problem (such as Fanconi syndrome) rather than plain diabetes — clues only readable once you understand the four-segment division of labor.
Chapter 3
RAAS · the BP loop
RAAS · the BP loop
The renin-angiotensin-aldosterone system (renin–angiotensin–aldosterone system: A hormone chain that controls blood pressure and fluid — when tense it narrows vessels and holds water and sodium.) is one of the body's most classic feedback loops, and the target of 2 of the 5 main antihypertensive drug classes (ACEi and ARB). Understanding this loop equals understanding half a textbook of cardiovascular pharmacology.
The starting point is the juxtaglomerular apparatus (JGA) — specialized smooth-muscle cells on the wall of the glomerular afferent arteriole, plus the macula densa — sensing three things simultaneously:
1. Low pressure in the afferent arteriole → renin release
2. The macula densa senses low Na⁺ concentration in the distal tubule → renin release
3. Sympathetic β1 receptors during stress directly stimulate renin release
The relay chain (the L4 animation walks through it step by step):
Renin is a protease, cleaving liver-synthesized angiotensinogen into angiotensin IACE (angiotensin-converting enzyme), mainly in lung vascular endothelium, converts AT-I to AT-IIAT-II is the true hormonal effector, doing three things at once: directly constricting arterioles to raise blood pressure; stimulating the adrenal cortex zona glomerulosa to secrete aldosterone; stimulating the hypothalamus to produce thirst and ADH releaseAldosterone is a steroid hormone, acting on the distal convoluted tubule and collecting duct to increase Na⁺ reabsorption and K⁺ excretion
Result: sodium and water retention → blood volume rises → blood pressure rises → JGA senses it → renin secretion stops (negative feedback loop).
Open 'deep dive' to see a 4-step animation drawing this loop clearly, plus where ACEi / ARB / MRA each block.
The starting point is the juxtaglomerular apparatus (JGA) — specialized smooth-muscle cells on the wall of the glomerular afferent arteriole, plus the macula densa — sensing three things simultaneously:
1. Low pressure in the afferent arteriole → renin release
2. The macula densa senses low Na⁺ concentration in the distal tubule → renin release
3. Sympathetic β1 receptors during stress directly stimulate renin release
The relay chain (the L4 animation walks through it step by step):
Renin is a protease, cleaving liver-synthesized angiotensinogen into angiotensin IACE (angiotensin-converting enzyme), mainly in lung vascular endothelium, converts AT-I to AT-IIAT-II is the true hormonal effector, doing three things at once: directly constricting arterioles to raise blood pressure; stimulating the adrenal cortex zona glomerulosa to secrete aldosterone; stimulating the hypothalamus to produce thirst and ADH releaseAldosterone is a steroid hormone, acting on the distal convoluted tubule and collecting duct to increase Na⁺ reabsorption and K⁺ excretion
Result: sodium and water retention → blood volume rises → blood pressure rises → JGA senses it → renin secretion stops (negative feedback loop).
Open 'deep dive' to see a 4-step animation drawing this loop clearly, plus where ACEi / ARB / MRA each block.
Why ACEi/ARB are renal protectants
ACEi ('prils') and ARB ('sartans') aren't just antihypertensives, they're also renoprotective drugs — a critical distinction.Mechanism (the L4 animation shows it in detail): AT-II constricts both afferent and efferent arterioles, but the efferent is more sensitive; this raises intraglomerular pressure, which short-term keeps GFR looking normal but long-term damages the glomerulus. ACEi / ARB blocks AT-II → efferent arteriole relaxes → intraglomerular pressure drops → proteinuria decreases → CKD progression slows.
Clinical evidence (A-level):
Diabetic nephropathy + proteinuria: ACEi / ARB recommended even at normal blood pressure (KDIGO 2024)Hypertension + proteinuria: first-line choiceHeart failure: almost everyone is on ACEi or ARB (unless contraindicated)
Key difference between sartans and prils: ACEi blocks ACE, reducing AT-II but also reducing bradykinin breakdown, so some patients develop a dry cough (10–20%, possibly higher in Asians); ARB directly blocks the AT-II receptor, doesn't affect bradykinin, and doesn't cause cough — suitable for ACEi-intolerant patients.
Contraindications and cautions:
Pregnancy: absolutely contraindicated (teratogenic / fetal renal hypoplasia)Bilateral renal artery stenosis: use can trigger acute renal failure — must avoidHyperkalemia (K >5.5): use with caution; stacking with aldosterone antagonists + K supplements can be fatalAcute dehydration / heavy diuresis / NSAID co-administration: the so-called 'triple whammy' — can trigger acute kidney injury, a real emergency requiring immediate drug discontinuation and medical care
MRA (mineralocorticoid receptor antagonists, like spironolactone) block aldosterone at the renin–angiotensin–aldosterone system: A hormone chain that controls blood pressure and fluid — when tense it narrows vessels and holds water and sodium. terminal, used for refractory hypertension, heart failure (HFrEF), and primary aldosteronism. Note: hyperkalemia and male gynecomastia are common side effects of non-selective MRA; selective MRA (eplerenone) has fewer side effects.
Chapter 4
D activation + EPO
D activation + EPO
The kidney doesn't just filter — it's also a factory for 2 hormones:
1. Vitamin D second hydroxylation:
The liver converts D3 to 25-hydroxyvitamin D: The storage form of vitamin D in blood — the number measured to check D status. (storage form, half-life of weeks)CYP27B1 in the renal proximal tubule converts 25(OH)D to 1,25(OH)₂D (calcitriol, the active hormone form)This is the form D actually works as — regulating calcium / phosphorus / bone / immunity / muscleCKD patients, even with adequate D3, often have low 1,25(OH)₂D because of reduced CYP27B1 activity — this is the chemical root of CKD-MBD (mineral-bone disease)
See vitamin-d/kidney L4 for the full animation of CYP27B1 + vitamin D receptor: The cellular 'socket' that vitamin D plugs into to carry out its instructions. + RXR regulating 1000 genes
2. EPO (erythropoietin) — the red cell baton:
Renal cortical interstitial fibroblasts sense tissue oxygen tensionLow O₂ → HIF-α (hypoxia-inducible factor) stabilizes → enters nucleus → transcribes EPO geneEPO enters blood → bone marrow erythroid progenitors differentiate → red cell productionIn normal adults, 90% of EPO comes from the kidney, 10% from the liverHigh altitude / blood loss / chronic hypoxia → EPO ↑ → RBCs ↑
CKD anemia:
In CKD, renal EPO production collapses → anemia (even with adequate iron, B12, and folate)rhEPO (recombinant human EPO, epoetin / darbepoetin) is the standard CKD treatmentPost-2019 HIF-PHI (roxadustat / daprodustat) — oral drugs that stabilize HIF-α, mimicking hypoxia → endogenous EPO secretion + improved iron utilization → a new option for CKD anemia
'Athletic EPO doping' is this same pathway abused — injecting EPO raises red cells → oxygen carrying ↑ → endurance ↑ — but the blood becomes too thick, easily clotting and causing sudden cardiac death — this is the repeating tragedy in pro cycling since the 1990s.
1. Vitamin D second hydroxylation:
The liver converts D3 to 25-hydroxyvitamin D: The storage form of vitamin D in blood — the number measured to check D status. (storage form, half-life of weeks)CYP27B1 in the renal proximal tubule converts 25(OH)D to 1,25(OH)₂D (calcitriol, the active hormone form)This is the form D actually works as — regulating calcium / phosphorus / bone / immunity / muscleCKD patients, even with adequate D3, often have low 1,25(OH)₂D because of reduced CYP27B1 activity — this is the chemical root of CKD-MBD (mineral-bone disease)
See vitamin-d/kidney L4 for the full animation of CYP27B1 + vitamin D receptor: The cellular 'socket' that vitamin D plugs into to carry out its instructions. + RXR regulating 1000 genes
2. EPO (erythropoietin) — the red cell baton:
Renal cortical interstitial fibroblasts sense tissue oxygen tensionLow O₂ → HIF-α (hypoxia-inducible factor) stabilizes → enters nucleus → transcribes EPO geneEPO enters blood → bone marrow erythroid progenitors differentiate → red cell productionIn normal adults, 90% of EPO comes from the kidney, 10% from the liverHigh altitude / blood loss / chronic hypoxia → EPO ↑ → RBCs ↑
CKD anemia:
In CKD, renal EPO production collapses → anemia (even with adequate iron, B12, and folate)rhEPO (recombinant human EPO, epoetin / darbepoetin) is the standard CKD treatmentPost-2019 HIF-PHI (roxadustat / daprodustat) — oral drugs that stabilize HIF-α, mimicking hypoxia → endogenous EPO secretion + improved iron utilization → a new option for CKD anemia
'Athletic EPO doping' is this same pathway abused — injecting EPO raises red cells → oxygen carrying ↑ → endurance ↑ — but the blood becomes too thick, easily clotting and causing sudden cardiac death — this is the repeating tragedy in pro cycling since the 1990s.
CKD-MBD · the Ca-P-bone-vessel cascade
CKD-MBD (Chronic Kidney Disease – Mineral and Bone Disorder) is one of CKD's most underestimated complications, and the real clinical convergence point of multiple atlas stories on calcium / phosphate / D / K2.Mechanism chain (starts in CKD G3, severe in G4–5):
1. Renal phosphate excretion drops → blood phosphate rises
2. CYP27B1 activity drops → 1,25(OH)₂D falls → calcium absorption drops → blood calcium drops
3. Parathyroid senses low Ca + high P + low D → parathyroid hormone: Released when blood calcium dips — it pulls calcium back into the blood from bone, kidney, and gut. persistently rises (secondary hyperparathyroidism)
4. Chronically high PTH → bone loss (especially cortical bone) → pathologic fractures
5. Simultaneously high P + high PTH + low K2 → vascular smooth muscle calcification (transformation toward osteoblast phenotype) → vascular calcification
Result:
Bone: renal osteodystrophy — sharply elevated fracture rateVessels: calcified aorta / coronaries → cardiovascular eventsHeart: most CKD patients die of cardiovascular disease, not kidney failure itself — mainly driven by vascular calcification
KDIGO CKD-MBD interventions:
Phosphate restriction — phosphate additives in processed foods are the hidden killer (sodas / processed meats / instant soups) — natural P absorption ~50%, additive P ~90%Phosphate binders — calcium carbonate / calcium acetate (older), sevelamer (no calcium), lanthanum carbonateActive vitamin D analogs (calcitriol / paricalcitol) — bypass CYP27B1, supplying the active hormone form directlyCalcimimetics (cinacalcet) — fool the parathyroid into 'sensing enough calcium', suppressing PTHK2 (MK-7) + vitamin D + magnesium — some studies show benefit in reducing vascular calcification, but KDIGO hasn't yet listed this as a strong recommendation
This is where the atlas's four lines on Ca / P / D / K2 converge clinically — looking at any single nutrient won't explain this cascade; only the whole map explains why KDIGO manages CKD-MBD as a single integrated disease.
Chapter 5
'High protein hurts kidney' debunked
'High protein hurts kidney' debunked
'High protein hurts your kidneys' is one of the most widely repeated nutrition myths of the past 30 years, originating from research on early CKD patients incorrectly extrapolated to healthy adults. Let's go through the evidence point by point.
Healthy kidneys handle high protein: Devries 2018 AJN meta analyzed 28 RCTs, 1358 healthy adults, comparing high protein (1.5–3.0 g/kg) vs normal (≤1.2 g/kg) — no significant differences in eGFR, sCr, or proteinuria; cohort studies and clinical observation of athletes eating 2–3 g/kg long-term also show no kidney damage; mechanistically, healthy kidneys have abundant filtration reserve, and the 10–30% post-meal rise in GFR is physiologic adaptation, not damage.
CKD patients do need protein restriction, but tiered and professionally managed: CKD G3a–G3b (eGFR 30–59) — KDOQI 2020 recommends 0.55–0.60 g/kg/d low-protein diet plus phosphate restriction to slow progression; CKD G4–G5 pre-dialysis is stricter: 0.4–0.5 g/kg/d plus essential amino acid / ketoacid supplementation; dialysis patients, conversely, raise to 1.0–1.2 g/kg/d because dialysis loses protein heavily. This must be co-managed by a dietitian and nephrologist, otherwise protein-energy wasting (PEW) brings harm earlier than CKD progression.
Creatinine rise doesn't equal kidney damage: high-protein meals + exercise + lifting raise creatinine production → sCr rises → eGFR appears to drop. Real kidney damage shows proteinuria, leukocyturia, hematuria, elevated cystatin C, or imaging changes. Reading kidney health from a single creatinine number on a physical exam is the most common misread (see filtration scene).
Real drivers of kidney damage (ranked by weight):
1. Uncontrolled hypertension — the #1 cause of kidney failure
2. Uncontrolled diabetes — #2
3. NSAID abuse (especially in the elderly / dehydrated)
4. Certain Chinese herbs (aristolochic acid): guan-mu-tong / guang-fang-ji / qing-mu-xiang; mass CKD cases in China and Taiwan, 1990s–2000s
5. Smoking
6. Overweight + sedentary
7. Chronic heavy-metal exposure (Pb, Cd)
Protein intake sits low on this list — for healthy adults, 1.6–2.0 g/kg/d is safe, which is the exercise + elderly muscle-preservation range given in the atlas protein story.
Healthy kidneys handle high protein: Devries 2018 AJN meta analyzed 28 RCTs, 1358 healthy adults, comparing high protein (1.5–3.0 g/kg) vs normal (≤1.2 g/kg) — no significant differences in eGFR, sCr, or proteinuria; cohort studies and clinical observation of athletes eating 2–3 g/kg long-term also show no kidney damage; mechanistically, healthy kidneys have abundant filtration reserve, and the 10–30% post-meal rise in GFR is physiologic adaptation, not damage.
CKD patients do need protein restriction, but tiered and professionally managed: CKD G3a–G3b (eGFR 30–59) — KDOQI 2020 recommends 0.55–0.60 g/kg/d low-protein diet plus phosphate restriction to slow progression; CKD G4–G5 pre-dialysis is stricter: 0.4–0.5 g/kg/d plus essential amino acid / ketoacid supplementation; dialysis patients, conversely, raise to 1.0–1.2 g/kg/d because dialysis loses protein heavily. This must be co-managed by a dietitian and nephrologist, otherwise protein-energy wasting (PEW) brings harm earlier than CKD progression.
Creatinine rise doesn't equal kidney damage: high-protein meals + exercise + lifting raise creatinine production → sCr rises → eGFR appears to drop. Real kidney damage shows proteinuria, leukocyturia, hematuria, elevated cystatin C, or imaging changes. Reading kidney health from a single creatinine number on a physical exam is the most common misread (see filtration scene).
Real drivers of kidney damage (ranked by weight):
1. Uncontrolled hypertension — the #1 cause of kidney failure
2. Uncontrolled diabetes — #2
3. NSAID abuse (especially in the elderly / dehydrated)
4. Certain Chinese herbs (aristolochic acid): guan-mu-tong / guang-fang-ji / qing-mu-xiang; mass CKD cases in China and Taiwan, 1990s–2000s
5. Smoking
6. Overweight + sedentary
7. Chronic heavy-metal exposure (Pb, Cd)
Protein intake sits low on this list — for healthy adults, 1.6–2.0 g/kg/d is safe, which is the exercise + elderly muscle-preservation range given in the atlas protein story.
'Kidney detox / cleanse / tonic teas' — point by point
'Detox and nourish your kidneys' is one of the biggest markets in supplements, but lay out how the kidney works and the claim is nearly a non-question.First, the kidney is already one of the body's detox organs — the blood it filters per minute and the 180 L of ultrafiltrate per day (see the by-the-numbers page) far exceed any scale a tea could influence. In someone with normal kidney function, metabolic waste (urea, creatinine, uric acid) is continuously cleared; there's no backlog waiting for outside help. 'Detox' presumes a pile of toxins that can't get out, but for a healthy kidney that premise doesn't hold.
Second, the real effect of drinking more water is diluting urine and lowering the recurrence risk of certain stones (especially calcium oxalate and uric acid) — this is evidence-based, but the mechanism is diluting crystals in the urine, not 'rinsing the kidney' itself. About 2–2.5 L of urine per day is a reasonable target, but drinking beyond that doesn't make the kidney 'cleaner', and in extremes (pouring in several liters in a short time) can trigger hyponatremia, which has real fatal cases.
Third, some 'kidney tonic teas' actually harm the kidney. The classic is aristolochic-acid-containing herbs (guan-mu-tong, guang-fang-ji, qing-mu-xiang), a confirmed nephrotoxin and carcinogen — China and Taiwan saw many cases progress to uremia from this in the 1990s–2000s (see the protein-myth body). Some weight-loss or diuretic teas with unclear compound ingredients can, taken long-term, disturb electrolytes or directly injure the kidney.
Fourth, a counterintuitive point: for people who already have CKD, blindly drinking large amounts of water offers no protection and, in mid-to-late stages (especially oliguric phases), can worsen water-sodium retention and heart failure. Fluid volume in CKD must be individualized by a nephrologist, not borrowed from the healthy-person 'drink more to detox' logic.
The list that genuinely protects the kidney is plain, mapping right onto the harm factors in the debunk body: control blood pressure and glucose, don't abuse NSAIDs, avoid aristolochic-acid herbs, don't smoke, keep a healthy weight, and check eGFR plus UACR regularly. None of it is 'detox by drinking something' — it's all about reducing the chances of the kidney being damaged over the long run.
Chapter 6
Sodium · not just 'less salt'
Sodium · not just 'less salt'
'Eat less salt' is the #1 health-check advice, but the actual numbers, individual variation, and evidence curve are more complex than the slogan.
International guideline numbers:
WHO: <5 g salt/day (= 2 g Na)AHA 2017: ideal <1.5 g Na, realistic target <2.3 g NaChina DRI 2023: adult <5 g salt/dayDASH RCT + sodium reduction → SBP ↓3 mmHg (on top of already DASH diet)
Global reality:
Chinese adults average ~10–12 g salt/day (2× the WHO upper limit)Soy sauce + pickles + processed foods are China's biggest sodium sourcesUS ~8–10 g (mostly from processed foods / dining out)Italy / France ~8–10 g (bread + processed meat)
Stratified BP response to sodium reduction:
Salt-sensitive (~50% of hypertensives, higher in elderly / Black / obese) — reducing 4 g salt → SBP ↓6–8 mmHgSalt-resistant (~25% of normotensives) — reducing 4 g salt → SBP ↓1–2 mmHg, limited benefitNo easy 'salt sensitivity' test — in practice, try 4 weeks of reduction and see
The 'U-curve' controversy on sodium and mortality:
PURE 2018 Lancet (Mente) — 180,000 people across 18 countries — Na intake <3 g/day AND >6 g/day both linked to higher cardiovascular events and mortality (U-shaped)AHA / WHO push back — PURE used single spot urine to estimate Na (inaccurate); the very-low Na group may have been sicker (reverse causation)2021 Cochrane — reaffirms sodium reduction lowers BP; doesn't address U-curve mortalityConclusion: very low Na (<1.5 g) isn't necessarily better; <2.3 g remains a reasonable target; blanket 'less is always better' has weak evidence
Potassium / sodium ratio matters more than absolute sodium (NASEM 2019):
Eating K-rich foods (fruits + vegetables + legumes + whole grains) has more consistent effects than salt reduction aloneK ≥3.5 g/day + Na ≤2.3 g/day is the combined targetCKD G4–G5 patients can't have more K — hyperkalemia is fatal; this group needs individualization
Practical priority (healthy adults + early hypertension):
1. Avoid processed foods — 80% of sodium comes not from cooking salt but from processed foods
2. Cut soy sauce + oyster sauce + chicken bouillon by half — three hidden Na sources in the Chinese kitchen
3. Double fruit and vegetable intake — boost K + Mg + fiber (DASH approach)
4. No smoking, limit alcohol
5. Regular exercise + weight loss
International guideline numbers:
WHO: <5 g salt/day (= 2 g Na)AHA 2017: ideal <1.5 g Na, realistic target <2.3 g NaChina DRI 2023: adult <5 g salt/dayDASH RCT + sodium reduction → SBP ↓3 mmHg (on top of already DASH diet)
Global reality:
Chinese adults average ~10–12 g salt/day (2× the WHO upper limit)Soy sauce + pickles + processed foods are China's biggest sodium sourcesUS ~8–10 g (mostly from processed foods / dining out)Italy / France ~8–10 g (bread + processed meat)
Stratified BP response to sodium reduction:
Salt-sensitive (~50% of hypertensives, higher in elderly / Black / obese) — reducing 4 g salt → SBP ↓6–8 mmHgSalt-resistant (~25% of normotensives) — reducing 4 g salt → SBP ↓1–2 mmHg, limited benefitNo easy 'salt sensitivity' test — in practice, try 4 weeks of reduction and see
The 'U-curve' controversy on sodium and mortality:
PURE 2018 Lancet (Mente) — 180,000 people across 18 countries — Na intake <3 g/day AND >6 g/day both linked to higher cardiovascular events and mortality (U-shaped)AHA / WHO push back — PURE used single spot urine to estimate Na (inaccurate); the very-low Na group may have been sicker (reverse causation)2021 Cochrane — reaffirms sodium reduction lowers BP; doesn't address U-curve mortalityConclusion: very low Na (<1.5 g) isn't necessarily better; <2.3 g remains a reasonable target; blanket 'less is always better' has weak evidence
Potassium / sodium ratio matters more than absolute sodium (NASEM 2019):
Eating K-rich foods (fruits + vegetables + legumes + whole grains) has more consistent effects than salt reduction aloneK ≥3.5 g/day + Na ≤2.3 g/day is the combined targetCKD G4–G5 patients can't have more K — hyperkalemia is fatal; this group needs individualization
Practical priority (healthy adults + early hypertension):
1. Avoid processed foods — 80% of sodium comes not from cooking salt but from processed foods
2. Cut soy sauce + oyster sauce + chicken bouillon by half — three hidden Na sources in the Chinese kitchen
3. Double fruit and vegetable intake — boost K + Mg + fiber (DASH approach)
4. No smoking, limit alcohol
5. Regular exercise + weight loss