Renal & Hepatic, the reference

When renal perfusion falls but the tubules are still healthy, aldosterone and vasopressin make them reclaim sodium and water hard. Very little sodium leaks into the urine, and the urine that is passed is concentrated. In tubular necrosis the damaged tubules can no longer conserve sodium or concentrate urine. Urine sodium rises above 40, osmolality drops below 400 and FENa goes above 2%. Diuretics force sodium loss, so these indices cannot be trusted once diuretics have been given.

Low urine osmolality as prerenal. Low osmolality may seem to fit a failing kidney, but it describes dilute urine from tubules unable to concentrate, the intrinsic pattern seen in tubular necrosis. A prerenal kidney still reabsorbs water, so its urine is concentrated.

In prerenal oliguria, sodium is retained (urine sodium ≤20 mEq/L), urine is concentrated (>400 mOsm/kg) and FENa is <1%. Without diuretics, urine sodium >40 with osmolality <400 indicates intrinsic renal disease; FENa >2% means tubules cannot conserve sodium.

Sources: Barash 9eStoelting 8eMayo review

When the kidneys are underperfused but the tubules are intact, they defend blood volume by holding on to sodium and water. Urine is scanty, and urinary sodium falls below 20 mEq/L. Restoring volume restores urine flow. If the ischaemia lasts, it injures the tubules. Tubular necrosis is the extension of that prerenal state. The damaged tubules can no longer reabsorb sodium, so urinary sodium rises above 40 mEq/L. Oliguria and rising urea and creatinine then continue even after blood pressure and volume are back to normal.

Interstitial nephritis. Interstitial nephritis is tempting because it is a real intrinsic renal cause of oliguria with rising urea and creatinine. It fails because the trigger here is hypoperfusion, and the intrinsic lesion that hypoperfusion produces is tubular necrosis. Nothing in the case points to an inflammatory or drug-induced process.

Miller names intravascular volume depletion as the leading cause of postoperative oliguria; a fluid challenge restores urine output when hypotension accompanies it. Barash calls tubular necrosis the commonest ischaemic lesion, an extension of prerenal azotaemia. Faust separates the two by urinary sodium.

Sources: Miller 10eBarash 9eFaust

In surgical patients, injury is most often tubular necrosis. It lowers glomerular filtration, so the kidneys can no longer clear nitrogenous waste, and creatinine and urea climb in the blood. Urine volume depends on something else: the tubules' ability to concentrate. Damaged tubules lose that ability, so urine can keep flowing at a normal rate while waste clearance fails. Volume therefore tracks water handling, not clearance. The laboratory trend of waste products is the marker. Reported mortality varies widely with severity and setting, so no single figure defines the condition.

Oliguria as essential to diagnosis. Oliguria is the textbook picture of kidney failure, so preserved urine flow feels reassuring. It is not. Injured tubules cannot concentrate, so volume can stay normal in nonoliguric failure. Present urine flow also poorly predicts later creatinine change.

The textbooks define acute kidney injury as a fall in filtration with build-up of creatinine and urea. Urine output often drops, but nonoliguric postoperative forms are common. Urine flow, when present, is a weak predictor of subsequent creatinine change.

Sources: Barash 9eStoelting 8eOutside our books: KDIGO Clinical Practice Guideline for Acute Kidney Injury (2012)Miller 10e

Stage 1: creatinine 1.5 times baseline, or urine output under 0.5 mL/kg/h for 6 to under 12 hours. Stage 2: doubled creatinine, or that oliguria for 12 hours or longer. Stage 3: tripled creatinine, creatinine 4 mg/dL or more, dialysis, urine output under 0.3 mL/kg/h over 24 hours, or 12 anuric hours. Either criterion qualifies; the worse decides. A rate of exactly 0.5 is not oliguria. When the urine output has no stated duration, it cannot set the stage; stage by the creatinine ratio.

Judging urine volume without indexing to weight. An hourly urine volume that sounds acceptable for an adult can look like no injury at all. Indexed to weight it falls under 0.5 mL/kg/h, and creatinine more than tripled from baseline reaches stage 3 by itself; the worse criterion sets the stage.

The worse criterion decides. Stage 1: creatinine 1.5–1.9× baseline or up 26.5 µmol/L, or 6–12 h of urine under 0.5 mL/kg/h. Stage 2: creatinine 2.0–2.9×, or that low output past 12 h. Stage 3: creatinine 3× or 4 mg/dL, dialysis, or 12 h anuria.

Sources: Barash 9eMiller 10eOutside our books: KDIGO Clinical Practice Guideline for Acute Kidney Injury (2012)

Each variable tracks a separate failure. Albumin and INR show how much protein the liver still makes, bilirubin shows how well it clears, ascites shows portal hypertension and encephalopathy shows brain toxicity. The worst band of any variable scores 3 points and no variable scores below 1, so a few severe findings can settle the class even when some values are missing. One textbook writes class C as above 10, but the modified table includes 10 in class C, so a total of exactly 10 is class C.

Mid-range bilirubin implies moderate class. Bilirubin is only mid-range and albumin isn't given, so the patient seems moderately ill. But INR above 2.3 and tense ascites score 3 points each, encephalopathy with GCS 11 scores at least 2, and the missing albumin adds at least 1. That makes 11 or more.

The modified table scores albumin, prothrombin time or INR, bilirubin, ascites and encephalopathy from 1 to 3 points each. Total 5–6 is class A, 7–9 is class B and 10–15 is class C. Encephalopathy is graded I to IV, from mild symptoms up to coma.

Sources: Barash 9eMiller 10e

Child-Pugh adds points for bilirubin, albumin, INR, ascites and encephalopathy. Each marker scores higher as it worsens. The total sets the class, not a word like 'mild'. Use INR or prothrombin time, never both. A middle-band value on every marker gives a total of 10, which is class C. Sources quote class C differently: Stoelting gives 75–80%; Barash, pooling decades of studies, gives 60–80%; Miller cites early series up to 82% and notes lower rates in newer work. All put class C far above class B, so expect about 60–80%.

Class B mortality of about 30%. Because the ascites and behavioural change are called 'mild', the patient sounds moderate, so class B's 30% feels right. But the points decide the class, not the adjectives. Every marker scores in the middle band, so the total reaches class C, and class C mortality is about 60–80%.

Stoelting gives surgical mortality of 10%, 30% and 75–80% for classes A, B and C. Barash, across decades of studies, gives 10%, 17–30% and 60–80%. Miller's table puts class C at 10–15 points and says early mortality figures have fallen in recent studies.

Sources: Stoelting 8eBarash 9eMiller 10e

Here, protein loss usually leaves the circulation underfilled, and albumin reverses those signs; renal failure adds pericardial effusion, and clotting tendency adds pulmonary embolism. Stoelting advises seeking tamponade when hypotension in end-stage renal disease resists fluid. A bedside echo can look for an empty, compressed or failing heart, deciding fill or drain; a chest film shows pleural fluid or lung water but cannot decide that. A blood gas for potassium and acid-base, and an ECG, are reasonable whichever image is chosen. Give albumin only once overload is excluded.

Blood gas as the explanatory test. A blood gas is tempting: renal failure brings high potassium and acidosis, and both must be known first. But it grades how sick the child is; it cannot show an effusion, a failing heart or lung water, so it does not explain the shock.

Stoelting notes nephrotic patients can become hypovolaemic with tachycardia and hypotension, which albumin improves, and are prone to pulmonary embolism; in end-stage renal disease, hypotension not corrected by fluid should trigger an immediate search for tamponade.

Sources: Stoelting 8e

A needle passed into kidney tissue will bleed, so a clotting defect rules out the percutaneous route. A solitary kidney was long an absolute bar, but high-resolution real-time ultrasound guidance has made it relative: the concern is not extra bleeding but that a complication could cost the patient their only kidney. One review calls uncontrolled hypertension absolute, the other relative. Both lead to the same practice: control the pressure first. Once controlled, hypertension is no barrier.

Any hypertension contraindicates biopsy. Hypertension appears on every contraindication list for renal biopsy, so any hypertension seems to rule a biopsy out. Both reviews, though, name only uncontrolled pressure. Once hypertension has been treated and controlled, it is no longer a reason to withhold the biopsy.

With ultrasound guidance, solitary kidney no longer mandates open biopsy: bleeding risk is not higher, though a complication could cost the kidney. One review lists clotting disorders and uncontrolled hypertension as absolute contraindications; another lists solitary kidney, infection, uncontrolled hypertension and anaemia as relative.

Sources: Outside our books: Renal Biopsy for Diagnosis in Kidney Disease: Indication, Technique, and Safety (2023)Outside our books: Performing a percutaneous kidney biopsy (2026)

For patients in general, Miller's notes the ulnar artery can be cannulated safely even after a failed attempt on the radial artery of the same arm. End-stage renal disease is a different setting: Stoelting's warns that radial, ulnar, brachial and axillary vessels are the raw material for future fistulae, so puncturing them may cost the patient dialysis access. The fistula arm is excluded because the shunt makes pressure and gas readings from that limb unreliable. The femoral line spares all arm vessels; its price is infection in an immunocompromised patient.

Ulnar artery of the same arm. Miller's explicitly supports the ulnar artery after a failed radial attempt on the same side, so it looks textbook-correct. That advice is for patients in general. In end-stage renal disease the ulnar artery is a likely future fistula site and should be preserved, as should the brachial.

The renal text advises sparing the radial, ulnar, brachial and axillary arteries in dialysis patients, accepts femoral lines despite infection risk, and warns that fistula-limb readings are inaccurate. The general text allows ulnar cannulation after failed same-side radial attempt.

Sources: Stoelting 8eMiller 10e

Volatile agents reduce hepatic blood flow. With isoflurane and sevoflurane, the hepatic arterial buffer response makes up for the fall in portal flow. Halothane abolishes this response and depresses the heart, so it causes the largest fall. Miller finds the buffer also intact with desflurane, but Barash reports that desflurane lowers flow more at 1 MAC. Sevoflurane or isoflurane is therefore safe under both views. Halothane hepatitis is immune-mediated through oxidative halide metabolites. Isoflurane carries minimal risk and sevoflurane produces none.

Isoflurane as the liver-friendly volatile. Isoflurane is the classic liver-friendly volatile and protects hepatic blood flow as well as sevoflurane does. It is not the best choice, because it still forms oxidative halide metabolites, with a small risk of immune hepatitis. Sevoflurane forms none, which matters most in autoimmune or inflamed livers.

Halothane causes the greatest fall in hepatic blood flow and loses the arterial buffer response. Isoflurane and sevoflurane barely reduce flow at 1 MAC, and one text finds desflurane lowers it more. Only sevoflurane generates no oxidative halide metabolites.

Sources: Barash 9eStoelting 8eMiller 10e

Severe post-anaesthetic hepatitis is attributed to an immune reaction: oxidative metabolism yields reactive trifluoroacetyl intermediates that bind hepatocyte proteins, which then act as neoantigens and provoke antibodies. Risk tracks how much of each agent is oxidatively metabolised, so it is highest with halothane, falling through enflurane and isoflurane to remote with desflurane. Sevoflurane forms no such metabolites at all. Barash calls the immune evidence strong; Miller says proof is lacking; both agree newer agents rarely cause it, so halothane is the agent to avoid.

Isoflurane as the riskiest volatile. Isoflurane is tempting because it, like halothane, forms trifluoroacetyl-type metabolites and has been linked to fulminant hepatic necrosis. But it is metabolised far less, so its immune hepatitis risk is minimal; halothane's risk is greatest, making it the agent to avoid.

Antibodies to liver proteins altered by halothane's trifluoroacetyl metabolite are found in halothane hepatitis. Predicted allergic hepatitis incidence is greatest with halothane, minimal with isoflurane, remote with desflurane; sevoflurane makes none of these oxidative metabolites, and severe hepatitis after newer agents is extremely rare.

Sources: Stoelting 8eBarash 9eMiller 10e

Sunken eyes and cold peripheries mean hypovolaemia. Poor tissue and kidney perfusion drives the acidosis and keeps it going. Isotonic crystalloid refills the intravascular space and brings back perfusion and urine flow, so the acid is cleared and the acidosis largely corrects itself. Guidelines prefer crystalloid to colloid for the first volume expansion unless the patient is bleeding. Saline is still acceptable, but Miller warns that its chloride load can add a hyperchloraemic acidosis. A balanced isotonic solution is safer when a lot of volume is needed.

Giving bicarbonate for low bicarbonate. Bicarbonate is tempting because the measured bicarbonate is low. But this acidosis comes from poor perfusion, and a buffer does not replace lost volume. Diabetic ketoacidosis guidance gives fluids first and keeps bicarbonate for extreme acidaemia only (pH below 6.9).

KDIGO suggests isotonic crystalloid rather than colloid as first-line volume expansion when there is no haemorrhagic shock. Barash corrects acidosis by expanding intravascular volume and keeps bicarbonate for pH below 6.9. Miller notes that saline may cause hyperchloraemic metabolic acidosis.

Sources: Outside our books: KDIGO Clinical Practice Guideline for Acute Kidney Injury (2012)Barash 9eMiller 10e

Starch was used as a synthetic alternative to albumin. Trials in septic and critically ill patients then showed more kidney failure and more need for dialysis. Its use was therefore restricted, it left routine practice and sepsis guidance advises against it. The harm comes from the type of colloid, not from oncotic pressure: hyperoncotic starch damages the kidney, while albumin appears protective. A perioperative fluid review found no extra renal harm in elective surgery. That finding does not extend to sepsis, where starch should be avoided.

Hyperoncotic albumin as nephrotoxic. Concentrated albumin is a hyperoncotic colloid, so it seems likely to strain the kidneys more than other fluids. Renal effects depend on the type of colloid: hyperoncotic albumin appears to protect the kidney, and the nephrotoxic colloid is starch.

Critically ill and septic patients given starch needed renal replacement more often, so it was dropped from routine use and is not recommended for resuscitation in sepsis. The excess kidney failure was seen in sepsis, not elective surgery. Albumin, by contrast, looks renoprotective.

Sources: Barash 9eOutside our books: Perioperative fluid therapy in adults and children: a narrative review (2025)Outside our books: KDIGO Clinical Practice Guideline for Acute Kidney Injury (2012)Faust

Mannitol came into use because of what it does to the ischaemic kidney. It raises renal cortical blood flow while the aorta is clamped, limits endothelial swelling and vascular congestion, and causes an osmotic diuresis. For those reasons, giving it before cross-clamping became almost universal, even though no one had proved it worked. Later clinical trials did not show fewer cases of renal failure. So it is still given, as an accepted but unproven practice. Making more urine does not by itself protect the kidney.

Furosemide for its diuresis. Furosemide raises urine output and is sometimes asked for during aortic surgery, so it looks protective. But a brisk diuresis is not protection. A loop diuretic can drain the circulating volume and lower renal perfusion, and it does not improve cortical blood flow during clamping the way mannitol does.

Mannitol 12.5 g/70 kg is almost universally given before cross-clamping, though drug protection is unproven. Its osmotic diuresis improves cortical blood flow and reduces endothelial oedema and congestion. Trials have not shown it lowers renal failure in these patients.

Sources: Miller 10e

Intermittent haemodialysis packs fluid and solute removal into short sessions. Those abrupt shifts lower blood pressure in a patient whose circulation is already struggling. Continuous techniques spread removal across the whole day and avoid these haemodynamic swings. The guideline does not name one modality as best for everyone. It treats the two as complementary and bases the first choice on local availability, experience and haemodynamic status. A stable patient can have either, but instability points to continuous therapy.

Intermittent haemodialysis for efficient clearance. A raised creatinine and urea seem to call for fast, efficient clearance, which intermittent sessions give. But removing fluid and solute quickly causes haemodynamic swings. In a hypotensive, tachycardic post-laparotomy patient this deepens the instability that continuous therapy is chosen to avoid.

The guideline treats continuous and intermittent therapy as complementary but suggests continuous over standard intermittent therapy when haemodynamics are unstable. It bases the initial choice on availability, experience and haemodynamic status. Textbooks describe continuous filtration for unstable critically ill patients because it avoids haemodynamic fluctuations.

Sources: Outside our books: KDIGO Clinical Practice Guideline for Acute Kidney Injury (2012)Miller 10eStoelting 8e

Miller gives 24 hours as the accepted, ideal limit for static cold storage of a kidney, which is why deceased-donor transplant is urgent. Barash notes a graft can still be implanted safely after 24 hours, possibly up to 36, allowing dialysis beforehand. However, longer storage increases the likelihood of delayed graft function, and storage beyond 24 hours raises the recipient's risk of rejection. Treat 24 hours as the working maximum and 36 hours as a stretch limit, not a target.

12 hours, the liver's limit. Twelve hours is the accepted cold storage limit for the liver. Mixing up the rows of the organ limits, or assuming kidneys are fragile, leads to it. In fact the kidney tolerates cold storage longest of the main solid organs, about twice the liver's limit.

Accepted static cold storage limits are kidney 24 hours, liver 12, heart 6 and lung 4. Kidneys can be safely implanted beyond 24 hours, possibly up to 36, though longer cold times bring delayed graft function and times over 24 hours raise rejection risk.

Sources: Miller 10eBarash 9e

Ringer's lactate and Plasma-Lyte cause less acidosis and less hyperkalaemia than normal saline. That matters in a patient whose own kidneys cannot correct either. Relaxation must stay adequate right up to the end of the operation, so twitch monitoring is needed, most of all during closure. The new kidney is denervated, and adrenergic pressors constrict its vessels, so hypotension is not managed mainly with them. Younger, fitter recipients, including some living-donor recipients, need only standard noninvasive monitors.

Routine arterial and CVP lines for all recipients. Transplant feels like major surgery in sick, hypertensive patients, so invasive lines seem automatic. But younger, healthier recipients and selected living-donor recipients are adequately managed with standard noninvasive monitoring. So CVP and arterial lines are chosen case by case; they are not always necessary.

Balanced salt solutions are the preferred crystalloid because they cause less acidosis and hyperkalaemia than saline. Careful neuromuscular monitoring is needed, particularly at closure. Noninvasive monitors suffice for healthier recipients, and adrenergic vasopressors are generally avoided because the graft is denervated.

Sources: Miller 10eBarash 9e

The aim is a brisk diuresis as soon as blood reaches the new kidney. Both drugs take time to circulate, so they are started while the surgeon is still joining the vessels, usually the vein first and then the artery. That way they reach the graft when it is unclamped. Mannitol's osmotic action, given with the loop diuretic, is linked to less delayed graft function and less acute renal failure after transplantation.

Diuretics after ureter is joined. Boosting urine flow once the drainage route is complete feels logical. But the ureter is joined late, after the vascular connections. By then the graft has already been reperfused without the diuretics, and the chance to protect it has gone.

Both diuretics are given ahead of reperfusion to promote urine output. Pooled trial data show mannitol lowers both acute renal failure and delayed graft function. Diuresis begins as soon as the vascular joins start, usually vein first, then artery.

Sources: Barash 9eMiller 10e

An adult kidney is large for a small child's abdomen. It may need to sit in the retroperitoneum, and the abdomen is tight to close. Closing the wall over the graft raises abdominal pressure. The anaesthetist sees this as a climb in peak inspiratory pressure, an early sign of a crowded abdomen that the surgeon must hear about. Chronic peritoneal dialysis has already stretched the cavity, which makes room. Graft vascular thrombosis is a recognised cause of failure in younger children.

Peritoneal dialysis as a hazard here. Peritoneal dialysis seems hazardous because of the indwelling catheter, possible peritoneal scarring and a dialysis-dependent child. With a size mismatch the reverse applies: long-term dialysis has expanded the abdominal cavity, giving the adult kidney more room and making tight closure less likely.

Adult kidneys may need a retroperitoneal site in small children. Chronic peritoneal dialysis may enlarge abdominal volume, but peak inspiratory pressure at closure needs watching; report any rise to the surgeons. Paediatric grafts fail more often than adult ones, with more thrombosis in younger children.

Sources: Barash 9e

The brain drives renal sodium loss and water follows it. The patient becomes hypovolaemic while plasma sodium and osmolality fall, giving a hypo-osmolar, hypovolaemic state with high urine output. Restricting fluid would deepen the volume deficit. SIADH, by contrast, is managed by watching and then restricting water. Diabetes insipidus also causes polyuria, but it loses free water, so sodium rises. For the stable patient, a pituitary surgery review advises normal saline with oral salt. A neuroanaesthesia text keeps hypertonic saline for symptomatic hyponatraemia.

Diabetes insipidus from polyuria. Polyuria after pituitary surgery suggests diabetes insipidus, and a history of desmopressin strengthens that. But diabetes insipidus loses free water and concentrates the plasma, so sodium climbs. Here sodium falls as the patient dries out and then corrects with saline, which is salt wasting.

After pituitary surgery, SIADH is watched and then water-restricted, while salt wasting is repleted with normal saline and oral salt. Salt wasting excretes concentrated sodium with a diuresis despite hypovolaemia. Diabetes insipidus raises serum osmolality. Hypertonic saline is kept for symptomatic patients.

Sources: Miller 10eOutside our books: Perioperative management of endoscopic transsphenoidal pituitary surgery (2020)Barash 9e

Sodium spreads through all body water, not just plasma, so the amount needed equals the desired rise multiplied by total body water. Women hold proportionally less water than men, so use 50% of weight rather than 60%. Each litre of 0.9% saline delivers 154 mEq, so sodium needed divided by 154 gives litres; then divide by hours. Too-fast correction risks osmotic demyelination: Miller's SIADH section caps the rise at 8 mEq/L in 24 hours; its symptomatic-hyponatraemia advice permits slightly more, so staying within 8 is safe under both.

Using male water fraction for a woman. A slightly faster rate looks harmless and close to the computed figure, so rounding up or misremembering the water fraction drifts the rate upward. But a woman's water fraction is 50%; using the male 60% overstates the rate, and any upward error pushes an already fast correction further past safety.

Barash puts body water near 60% of weight in men and 50% in women; sodium distributes through total body water; each litre of 0.9% saline holds 154 mEq of sodium; Miller's limits the serum sodium rise to 8 mEq/L over 24 hours, warning of demyelination.

Sources: Barash 9eFaustMayo reviewMiller 10e

The tubules' job is to reclaim sodium and water. When the tubules are intact, as in dehydration, they conserve both, so urine is concentrated and little sodium escapes. When the tubules are injured, as in acute tubular necrosis, concentration fails and sodium leaks into the urine. Stoelting names urine concentrating ability as the usual test. Miller names FENa as the commonest. These do not conflict: both probe the same reabsorptive work, so either is correct. Creatinine and urea are filtered markers, so they reflect glomerular filtration rate, not tubular handling.

Creatinine clearance. It is the most familiar renal function test, and creatinine even appears in the FENa calculation. But creatinine clearance only estimates how much plasma the glomeruli filter. It says nothing about whether the tubules reabsorb sodium and water correctly.

Stoelting tabulates urea and creatinine clearance as glomerular filtration tests. It puts specific gravity, osmolality and urinary sodium under tubular function, and calls concentrating ability the usual assessment. Miller calls FENa the commonest tubular test: sodium clearance expressed against creatinine clearance.

Sources: Stoelting 8eMiller 10e

Creatinine is not a pure filtration marker. How much is made depends on muscle mass and, in cirrhosis, on liver production. The tubules secrete some of it and some is cleared outside the kidney, and all three vary between patients. So the serum level can stay normal after filtration has already fallen a lot. When production falls, as in cirrhosis, sarcopenia or malnutrition, a low creatinine makes the kidneys look better than they are, and renal dysfunction is missed.

Blame BUN rather than creatinine. BUN is tempting because urea is the other routine renal blood test. But the standard estimating equation is built on serum creatinine, with cystatin C, age and sex added, not urea. Its errors come from the creatinine value, because creatinine generation, secretion and clearance vary.

Anaesthesia texts deem creatinine unreliable: tubular secretion, extrarenal elimination and generation vary, and filtration can fall well before it rises. In cirrhosis, serum creatinine overestimates GFR because reduced hepatic production, sarcopenia and malnutrition lower it. Muscle wasting hides renal dysfunction.

Sources: Miller 10eOutside our books: Acute Kidney Injury and Hepatorenal Syndrome in Patients with Cirrhosis (2023)Barash 9e

The liver makes urea by breaking down amino acids. So anything that increases the protein load raises BUN even when filtration is normal. Digested blood from a GI bleed, a high-protein diet and steroid-driven catabolism all do this. When the liver fails or protein intake is poor, less urea is made and BUN falls. The tubules also reabsorb a large share of filtered urea. That is why BUN alone is a poor marker of GFR.

Sickle cell raises urea through kidney damage. Sickle cell disease is known to damage the kidney, so a rise in urea seems likely. The textbook lists it with liver disease and malnutrition as a cause of low BUN. Tetracycline is the agent that raises BUN without any fall in filtration.

Urea production climbs with catabolism (a gut bleed, corticosteroids), protein-rich feeding and tetracycline, lifting BUN irrespective of filtration. It drops when hepatic synthesis or nutrition is poor, and in sickle cell anaemia. Hepatic origin plus tubular reabsorption make BUN an unreliable GFR gauge.

Sources: Miller 10e

Gastric juice is hydrochloric acid, so vomiting strips chloride without matching sodium. The strong ion difference widens and plasma becomes alkaline. The shrunken circulating volume then switches on renin and aldosterone, and the kidney reabsorbs more bicarbonate, which keeps the alkalosis going. Chloride is the missing piece, so giving it back as saline corrects the disorder. That is why it is called chloride-responsive. Hypokalaemia comes with it, and is also listed separately as a chloride-resistant cause. It contributes, but chloride loss is the primary driver.

Volume depletion as primary cause. Volume depletion is tempting because vomiting does dehydrate, and hypovolaemia drives aldosterone-mediated bicarbonate reabsorption. But that only sustains an alkalosis that already exists. The primary defect is chloride lost with gastric acid. The books call it hypochloraemic alkalosis, and it is corrected by replacing chloride.

Losing gastric hydrochloric acid removes chloride without sodium, causing a hypochloraemic alkalosis that saline corrects. The hypovolaemia that follows raises renin and aldosterone, which increases bicarbonate reabsorption. Hypokalaemia is listed among the chloride-resistant causes of metabolic alkalosis.

Sources: Miller 10eBarash 9eFaust

When blood becomes alkalotic, potassium moves out of plasma into cells. Serum potassium falls even though the body's total store has not changed. In metabolic alkalosis the kidney makes the fall worse, because it trades potassium for hydrogen ions to correct the pH. So the drop is bigger than in respiratory alkalosis. Chloride depletion works in the opposite direction. When serum chloride is low, the tubules reabsorb more bicarbonate, which creates or sustains the alkalosis. Hypochloraemia therefore feeds the alkalosis rather than resulting from it.

Low chloride as result of alkalosis. Vomiting causes low chloride and metabolic alkalosis together, but chloride loss drives the alkalosis, not the reverse: it lets the kidney retain bicarbonate. The alkalosis then lowers potassium, by shifting it into cells and through renal exchange for hydrogen ions.

Alkalotic states shift potassium into cells, lowering serum potassium, more so in metabolic than respiratory alkalosis. The kidney compounds this by exchanging potassium for hydrogen ions to compensate. Low chloride, by contrast, increases tubular bicarbonate reabsorption, causing the alkalosis.

Sources: Barash 9eStoelting 8eMayo review

Read the pH first. An acidaemic gas means the primary process is an acidosis. When bicarbonate has fallen as well, the problem is metabolic. The lungs respond by hyperventilating, which lowers PaCO2 and pulls the pH back towards normal over 12–24 hours. To check whether the compensation is adequate, the expected PaCO2 is 1.5 times the bicarbonate plus 8 (± 2) mmHg. If the gas is in kPa, convert it first: 5.3 kPa equals 40 mm Hg.

Calling it a respiratory alkalosis. A low PaCO2 is genuine hyperventilation, so calling the gas a respiratory alkalosis feels natural. But the pH is acidaemic, and a primary alkalosis cannot produce acidaemia. The falling bicarbonate is driving the pH, and the low PaCO2 is only the body's compensating response.

One text defines metabolic acidosis as acidaemia below pH 7.35, bicarbonate under 21 mEq/L and a compensating fall in PaCO2. Compensation develops within 12–24 hours and is checked with Winter's formula. Another text gives normal PaCO2 as 40 mm Hg, or 5.3 kPa.

Sources: Barash 9eMiller 10e

Large volumes of saline raise plasma chloride and produce a hyperchloraemic metabolic acidosis. This acidosis sits in the extracellular fluid, not inside cells. It pushes potassium out of cells, so plasma potassium rises. In trials, balanced solutions that contain potassium did not cause hyperkalaemia. Sources differ on how much the acidosis matters. Miller's calls its physiological significance unclear and its link to kidney injury unproven. A paediatric guideline says it can worsen outcome. Either way, both favour balanced crystalloids when large volumes are needed.

Lactic acidosis from hypoperfusion. Lactic acidosis is tempting after major haemorrhage, massive transfusion and liver surgery, where poor perfusion seems the obvious cause. But a clearly raised chloride after many litres of saline marks a hyperchloraemic acidosis. Blaming hypoperfusion ignores the chloride, and no raised lactate is there to support it.

Large volumes of normal saline cause hyperchloraemic metabolic acidosis, so they should be avoided. The acidosis drives potassium out of cells, causing hyperkalaemia. Balanced salt solution is probably the better choice. How much the extracellular acidosis matters clinically is still uncertain.

Sources: Miller 10eStoelting 8e

The anion gap is sodium minus chloride and bicarbonate; Miller's form adds potassium to sodium, with its own normal range. If bicarbonate itself is lost, through the bowel or through failed renal handling, chloride takes its place. The sum stays the same, so the gap stays normal and the picture is hyperchloraemic. Ketoacids or lactate are different: they add unmeasured anions that use up bicarbonate without raising chloride, so the gap widens. Cachexia causes ketosis. Neuroleptic malignant syndrome, severe asthma and severe anaemia cause lactate build-up.

Cachexia as bicarbonate-losing normal-gap acidosis. Cachexia sounds like a wasting, 'loss' state, so it seems to fit with bicarbonate depletion from diarrhoea. In fact starvation produces ketoacids. These are unmeasured anions that widen the gap, so cachexia belongs with ketoacidosis, not with gut or renal bicarbonate loss.

Textbook tables list diarrhoea, urinary diversion, gut fistulas and drains, and renal tubular acidosis as normal-gap causes, where bicarbonate is excreted or saline is given. Cachexia-related ketoacidosis and lactic acidosis from neuroleptic malignant syndrome, severe asthma or anaemia raise the gap.

Sources: Barash 9eMayo reviewMiller 10e

The gap widens only when an unmeasured acid anion replaces bicarbonate. Severe anaemia or shock raises lactate; diabetes, alcohol or cachexia produce ketoacids; methanol, ethylene glycol and salicylate add their own anions. Renal failure depends on severity: Mayo counts acute and chronic failure as high-gap, Barash files lesser insufficiency under normal-gap bicarbonate loss, and Miller shows the gap appears once failure is severe enough to retain sulphate and phosphate. Pure bicarbonate loss adds no new anion, whether from the gut, a urinary diversion or recovering ketosis.

ATN assumed always high-gap. ATN is tempting because renal failure appears in every high-gap list. But the gap rises only when failure is severe enough to retain fixed acids. Lesser renal insufficiency is classed as normal-gap bicarbonate loss. Severe anaemia, through lactate, is listed as high-gap without qualification.

Barash lists severe anaemia among the lactic causes of a raised gap. It places urinary diversion, renal insufficiency and ketosis recovery under normal gap. Miller attributes renal acidosis to unexcreted fixed acids in severe failure. Mayo adds toxins such as methanol and salicylates.

Sources: Barash 9eMiller 10eMayo review

When bicarbonate falls, the chemoreceptors drive up ventilation, so PaCO2 should fall by a predictable amount. If PaCO2 is normal or raised despite a low bicarbonate, the lungs are not clearing carbon dioxide. That retained carbon dioxide is a second acid load on top of the metabolic one, not compensation for it. Convert a PaCO2 given in kPa to mm Hg before you compare it with the predicted value. A value near the normal range can still sit well above the predicted one.

Simple metabolic acidosis with compensation. Low pH with low bicarbonate indicates metabolic acidosis, and the lungs are often assumed to be compensating. However, compensation must lower PaCO2 to the predicted value. A normal or raised PaCO2 despite this bicarbonate means compensation has failed, indicating a coexisting respiratory acidosis.

Winter's formula gives the PaCO2 you should expect in metabolic acidosis. A measured PaCO2 above it means compensation is inadequate and there is a respiratory problem as well. Example: lactic acidosis from haemorrhage plus respiratory acidosis from a flail chest.

Sources: Miller 10e

The gap stands for anions the lab does not measure, mainly phosphate, sulphate and anionic proteins. Barash's blood-gas table leaves potassium out. Miller puts potassium with sodium, which raises the result by about the potassium value, so its normal range differs. Organic acids such as ketoacids or lactate use up bicarbonate and add unmeasured anions, so the gap widens. A chloride load lowers bicarbonate but leaves the gap normal. Either formula detects a raised gap if you judge it against its own range.

Subtracting potassium, omitting bicarbonate. This figure comes from subtracting both chloride and potassium from sodium and leaving bicarbonate out. Bicarbonate is always subtracted. Potassium is a cation and, if used at all, is added to sodium, which raises the gap by roughly its value and never lowers it.

Barash's stepwise blood-gas approach subtracts chloride and bicarbonate from sodium, with a normal value below 13. Miller adds potassium to sodium and quotes a normal of 4 to 11. The gap reflects unmeasured anions, which rise with ketoacid or lactate accumulation.

Sources: Barash 9eMiller 10e

Every cause here is a deficiency. Low magnesium raises the risk of ectopic beats and of ventricular tachycardia, fibrillation and torsades, and you correct it by giving magnesium. Low potassium is the other electrolyte trigger for torsades, alongside drugs and ischaemia. Low calcium lengthens the ST segment and so the QT, while excess calcium shortens it. Magnesium also controls calcium: hypocalcaemia caused by magnesium loss only corrects once magnesium is replaced. So when both are low, magnesium is the primary fault.

Hyperkalaemia as the arrhythmia cause. Potassium is the first thought for any arrhythmia under anaesthesia, so excess potassium feels plausible. But torsades and a long QT come from deficiencies: low magnesium, low potassium or low calcium. Once potassium and magnesium are normal, low calcium is the cause left.

One text links low magnesium to ectopy, ventricular tachycardia, fibrillation and torsades, and treats it by giving magnesium. Another lists low potassium and low magnesium as torsades triggers. Low calcium lengthens the QT. Hypocalcaemia from magnesium depletion is treated by replacing magnesium.

Sources: Mayo reviewFaustBarash 9e

Low potassium impairs ventricular repolarisation: the T wave shrinks and a U wave grows. The flattened T fuses with the U wave, so the measured QT is really a QU interval. Miller lists QT prolongation, while Faust shows this lengthening is often only apparent because of TU fusion. Both agree on the trace's appearance. ST depression accompanies these changes. PR lengthening and QRS widening occur only in severe depletion, followed by arrhythmias. Hyperkalaemia causes the same two changes, so they do not distinguish the two.

QRS widening as the hallmark. End-stage renal disease brings hyperkalaemia to mind, whose hallmark is QRS widening towards a sine wave. Here the potassium is low. In hypokalaemia, QRS widening comes late, after T flattening and U waves, so it is neither the defining nor the expected finding.

Miller lists T flattening, QT prolongation, U waves and ST depression in severe hypokalaemia. Faust explains that the long QT is often a flattened T merged into a U wave. Mayo puts PR lengthening and QRS widening after these early signs.

Sources: Miller 10eFaustMayo review

Hypotonic glycine enters the circulation through opened prostatic veins. At first the extra volume raises blood pressure. Then the falling sodium and osmolality push water out of the vessels into the tissues, so the patient becomes effectively hypovolaemic. At the same time the heart muscle is depressed, which shows as a broad QRS and arrhythmias, and the spinal block has removed sympathetic compensation. A vasopressor given for block-related vasodilatation cannot restore a circulation that is leaking and driven by a failing heart, so the pressure stays low.

Bladder perforation. Perforation does cause hypotension, so it looks plausible. But under spinal the patient is awake, and perforation usually shows itself with suprapubic, periumbilical, inguinal or shoulder-tip pain. Its effect on blood pressure also varies, and it can cause hypertension. Refractory hypotension without pain fits the sodium-driven syndrome better.

The sources attribute TURP-syndrome hypotension to third spacing from hyponatraemia and low osmolality, following early hypertension, with depressed cardiac function and sympathetic blockade. Absorbed irrigant masks the usual signs of bleeding, and perforation in an awake patient presents with pain.

Sources: Stoelting 8eMiller 10eFaust

Sodium-free irrigating fluid absorbed during resection dilutes plasma sodium and lowers osmolality, so water moves into brain tissue. The cerebral oedema and raised intracranial pressure that follow cause confusion, agitation, visual upset, seizures and falling consciousness. How fast the sodium drops matters as much as how low it goes. Other metabolic causes can look the same, so electrolytes, glucose and a blood gas are sent together. The sodium result is what confirms irrigant absorption and guides correction.

Urgent CT head for stroke or bleed. Seizures with visual disturbance suggest stroke or intracranial bleed, so a scan seems thorough. Here the cause is metabolic: absorbed irrigant has diluted plasma sodium, which a scan cannot measure. A blood sodium confirms the diagnosis and directs treatment immediately.

The texts link absorbed sodium-free irrigant to confusion, agitation, visual disturbance and seizures through cerebral oedema from hyponatraemia. When TURP syndrome is suspected, they advise measuring plasma sodium and osmolality and drawing electrolytes, glucose and arterial blood gases together.

Sources: Stoelting 8eMiller 10e

Glycine soaks into the circulation through opened prostatic venous channels. It is an inhibitory neurotransmitter, and it suppresses retinal ganglion cell activity. The resulting blindness comes from the retina, so the pupillary light reflex becomes sluggish or absent. When vision fails from a central process, such as cerebral oedema in hyponatraemia or a cortical lesion, the light reflex pathway still works and the pupils react normally. A large absorbed load may also cause encephalopathy, partly from ammonia released as glycine is broken down.

Embolic or ischaemic cortical blindness. An elderly smoker with hypertension and a recent coronary stent invites thinking of an embolic or ischaemic cortical event. But cortical blindness leaves pupillary reflexes intact. Visual loss straight after resection with glycine, with poorly reacting pupils, is a retinal effect of the irrigant.

Absorbed glycine is named as the likely cause of transient blindness after prostate resection. It is thought to act on retinal ganglion cells as an inhibitory transmitter. Pupils are sluggish or unreactive, unlike central causes of visual loss, where light reflexes persist.

Sources: Miller 10eStoelting 8eFaust

Monopolar diathermy needs an irrigant that does not conduct electricity and has no electrolytes, such as glycine. This hypotonic fluid enters the circulation through opened prostatic venous channels. It adds water but no sodium, so serum sodium and osmolality both fall. Water then moves into brain tissue, causing cerebral oedema and raised intracranial pressure. The result is restlessness, confusion, visual upset and seizures. Glycine, an inhibitory transmitter, and its breakdown product ammonia make the neurological picture worse. Severe absorption can also cause circulatory collapse.

Haemorrhage from hypotension and tachycardia. Low blood pressure, a fast heart rate and a low haemoglobin look like haemorrhage. But bleeding does not drop serum sodium sharply or cause early confusion. Profound hyponatraemia with agitation after glycine irrigation points to absorbed irrigant, which can itself cause cardiovascular collapse.

Hypotonic, sodium-free irrigants such as glycine can be absorbed through exposed prostatic veins, causing hyponatraemia and fluid overload. The acute fall in sodium and osmolality causes cerebral oedema. Restlessness and confusion are early warning signs, not signs of inadequate anaesthesia.

Sources: Miller 10eOutside our books: Acute severe iatrogenic hyponatremia (2019)Stoelting 8e

While resection continues, hypotonic irrigant keeps entering the circulation, so no sodium infusion can keep up until the surgeon stops. After that, treatment aims to restore extracellular tonicity. Barash and Faust use hypertonic saline only for symptomatic patients below 120. Faust caps it at a fixed hourly rate. Miller gives a weight-based starting rate of 1 mL/kg/h for moderate symptoms such as confusion and nausea, then rechecks. The safe approach under both is to start with the weight-based rate, never go above the fixed cap, and recheck sodium.

Hypertonic saline at twice the maximum rate, started at once. It names the right fluid for a symptomatic patient with very low sodium, so it looks like definitive treatment. But it is twice the recommended maximum rate. It also skips the first step: irrigant keeps being absorbed until the surgeon stops.

Barash's first step is to tell the surgeon so the operation can be finished or stopped quickly. It reserves hypertonic saline for symptomatic patients with sodium below 120. Faust limits 3% saline to 100 mL/h. Miller starts moderate symptomatic cases at 1 mL/kg/h.

Sources: Barash 9eFaustMiller 10e

Citrate in transfused blood binds ionised calcium, and the liver normally metabolises both citrate and lactate. With the liver out of circulation, neither is cleared, so hypocalcaemia, acidosis and hypomagnesaemia follow in every case. Hypotension is conditional. A bicaval approach cuts venous return and cardiac output. Rapid transfusion lowers ionised calcium enough to give the signs of citrate toxicity: hypotension with a narrow pulse pressure, raised CVP and LVEDP, and a long QT. Liver disease and liver transplantation raise this risk.

Hypovolaemia with low CVP. Hypotension during a bleeding transplant usually means hypovolaemia, where CVP falls, so a low CVP feels natural. Citrate toxicity behaves differently: its recognised signs pair hypotension and a narrow pulse pressure with a raised CVP and LVEDP, not a low one.

Citrate in rapidly transfused blood binds ionised calcium; signs include hypotension, narrow pulse pressure, raised LVEDP and CVP and a prolonged QT. Without the liver, citrate and lactate are not metabolised, causing acidosis, hypocalcaemia and hypomagnesaemia. Bicaval clamping can reduce venous return.

Sources: FaustMiller 10e

When the portal vein clamp is released, cold blood loaded with acid and potassium flows from the graft into the central circulation. Hydrogen ion and potassium levels rise suddenly, preload increases and core temperature falls. This can cause arrhythmias, right heart strain or even arrest. If hypotension persists after recirculation, it is called the post-reperfusion syndrome. It is mainly caused by systemic vasodilation, and sometimes by myocardial depression.

Hypocalcaemia as the reperfusion change. Hypocalcaemia is a genuine transplant derangement, and calcium chloride is given at reperfusion, so it feels right. But low ionised calcium comes from citrate in transfused products while the liver is absent; at reperfusion calcium is given to counter potassium, and the defining new change is acid load.

The neohepatic stage begins when the graft is reperfused through the portal vein. Cold, acidotic, hyperkalaemic blood enters the circulation, so potassium, hydrogen ions and preload all rise. Persistent hypotension after this, the postreperfusion syndrome, is mostly vasodilatory and occasionally due to myocardial depression.

Sources: Miller 10eFaust

During ischaemia the clamped graft builds up vasoactive metabolites while its vessels hold only preservative fluid. When the clamps come off, the graft's vascular bed fills suddenly and takes up to 300 mL out of the circulation. At the same moment, vasodilating substances, preservative solution and venous blood from the legs enter the systemic circulation. Preload and systemic vascular resistance fall together, so the patient becomes hypotensive with a reflex tachycardia. A sudden potassium load reaching the heart can cause cardiac arrest.

Embolism from leg venous blood. Venous blood from the legs does enter the circulation when the clamps come off, so an embolus sounds possible. But this collapse happens predictably at the moment of declamping, and graft filling plus vasodilator washout explain it. Embolism is not the expected mechanism.

Hypotension after unclamping is explained by a sudden increase in intravascular capacity of up to 300 mL and by vasodilating chemicals released from tissue that was ischaemic. Cardiac arrest when the clamp is released is most likely caused by sudden hyperkalaemia.

Sources: Stoelting 8e

Benign postoperative cholestasis begins early and settles within 10 days, so jaundice weeks later needs another cause. Older reviews tie liver disease from parenteral nutrition to use beyond 14 days, and a later series found it can begin sooner, so a patient still fed intravenously fits at any point. The National Halothane Study tracked fatal hepatic necrosis within 6 weeks of a general anaesthetic, so volatile-agent hepatitis fits this timing too. Sickle haemolysis raises unconjugated bilirubin, and bile duct injury requires biliary surgery.

Ligated bile duct. Bile duct ligation tempts because bile duct injury appears among postoperative cholestatic causes. It fails because it needs an operation on the biliary tree, which this picture does not include.

Benign postoperative jaundice begins early and resolves within 10 days. Parenteral nutrition liver disease is described with use beyond 14 days. Halothane hepatic necrosis was studied within 6 weeks of anaesthesia. Haemolysis produces unconjugated hyperbilirubinaemia.

Sources: Outside our books: Review of parenteral nutrition-associated liver disease (2020)Barash 9eOutside our books: Parenteral Nutrition-Induced Liver Function Complications: Incidence, Risk Factors, and Prognosis (2025)

Parenteral nutrition gives a steady hypertonic glucose load into a central vein. The body responds with high insulin output, which is why high blood glucose is common during feeding and may need extra insulin. If the infusion stops abruptly, the glucose supply ends at once but the raised insulin level takes time to fall. For that period insulin keeps driving glucose into cells with nothing coming in to replace it, so blood glucose falls.

Potassium shift from sudden insulin fall. Stopping the glucose load does not make insulin fall straight away, so potassium does not shift out of cells. Insulin stays high after an abrupt stop, so the danger is falling glucose (hypoglycaemia).

The source warns that stopping the infusion suddenly can cause hypoglycaemia, because the body's own insulin stays high for a while. It also notes that high glucose is common while feeding runs, so glucose needs monitoring throughout.

Sources: Stoelting 8e

When pressure rises, the afferent arteriole constricts through a myogenic response. When pressure falls, it relaxes. Local feedback signals tune this tone, so glomerular capillary pressure and filtration barely change. The edges of the plateau are approximate, so books quote them differently. Miller's classic figure, from dog experiments, gives 80–180 mmHg. Miller's key-point summary rounds this to 60–160 mmHg mean pressure. Barash states it in systolic terms as 80–200 mmHg. All three describe the same curve. The figure value is the standard range for blood flow and GFR together.

A wider range starting at 50 mmHg. It looks like a safely wide range, and its lower limit sounds like the familiar cerebral autoregulation floor. But no renal source gives these limits. The classic kidney plateau starts near 80 mmHg, and even the lowest summary figure starts at 60.

Miller's figure shows renal blood flow and filtration rate staying flat between roughly 80 and 180 mmHg. Its summary gives 60–160 mmHg mean pressure, and Barash gives about 80–200 mmHg systolic. Faust links the plateau to myogenic constriction of the afferent arteriole.

Sources: Miller 10eBarash 9eFaust

ANP works against the renin–angiotensin–aldosterone axis. When the extracellular compartment expands, the atrial walls distend and the atrial myocytes secrete it. Acting through cyclic GMP, it relaxes vascular smooth muscle, so it lowers vascular resistance rather than raising it. In the kidney it increases filtration and blocks tubular sodium reabsorption. It also cuts renin release and aldosterone secretion, so less sodium and water are kept. Every response that reduces volume therefore goes the same way: ANP rises, while renin, angiotensin II and aldosterone fall.

Aldosterone rising to reduce volume. Aldosterone is the best-known volume hormone, so it comes to mind first. But it drives distal sodium and water retention, expanding extracellular volume. In hypervolaemia it is suppressed, partly by ANP. A rise in aldosterone therefore cannot be a volume-reducing compensation.

Atrial distension and sodium load release ANP, which vasodilates via cyclic GMP, raises GFR, inhibits renin, opposes angiotensin II and lowers aldosterone. Hypervolaemia is corrected by increased ANP and a suppressed renin–angiotensin–aldosterone axis, all promoting natriuresis.

Sources: Barash 9eMiller 10e

Glucose passes the glomerular filter freely, so the early filtrate carries the plasma glucose concentration. Proximal tubular cells take it back. At a normal filtered load they have spare capacity, so none reaches the urine. Glucose appears in urine only in two situations. In the first, the load exceeds what the tubule can reclaim, as in diabetes mellitus or during intravenous glucose infusion. In the second, proximal tubular function itself fails. Glucose is then lost alongside other proximally conserved solutes.

Sodium, as completely reabsorbed. Sodium is the solute the tubules reabsorb in the greatest quantity, so 'almost all' is easily mistaken for 'all'. However, the kidney always excretes some sodium, adjusted to match intake, so normal urine contains it. Normal urine contains no glucose.

Glucose crosses the glomerulus freely and is taken back proximally. Its presence in urine shows that an excessive load has outstripped the tubules' ability to reabsorb it. In proximal tubular disease, glucose is lost together with potassium, bicarbonate, phosphate, amino acids and water.

Sources: Miller 10eStoelting 8e

A lipophilic molecule can cross the tubular cell membranes on its own, without a transporter. After filtration or secretion, it diffuses back from tubular fluid into blood in the distal tubule. As a result, very little of it ends up in the urine. A water-soluble molecule cannot cross the membrane this way, so it relies on carrier proteins to be reabsorbed. Lipid solubility therefore decides whether reabsorption is passive and distal or carrier-dependent.

Proximal tubule carrier-mediated reabsorption. The proximal tubule is where most filtered substances are reclaimed and where drugs are actively secreted, so it seems the likely site. But the passive, lipid-driven reabsorption of drugs and their active forms takes place further along, in the distal tubule. Carrier-mediated reabsorption applies to water-soluble compounds.

Two processes can reduce renal clearance of secreted drugs: passive reabsorption of lipophilic drugs and carrier-mediated reabsorption of hydrophilic ones. The small filtered and secreted amount of lipophilic drug is readily reabsorbed in the distal tubules, so its net renal clearance becomes negligible.

Sources: Barash 9e

The sodium pump on the blood side keeps intracellular sodium low and potassium high. Sodium therefore enters from the lumen through channels and is pumped out, while potassium leaves the cell into the lumen through its own channels. Sodium enters without an accompanying anion, making the lumen negative relative to blood, which draws more potassium out. Aldosterone increases the activity of both the pump and the luminal potassium channels. Collecting ducts also actively secrete hydrogen ions and ammonia to maintain acid–base balance.

Sodium, as a secreted ion. Sodium is tempting because it always appears next to potassium in accounts of distal exchange. But it travels in the opposite direction. It enters the cell from the lumen and is reabsorbed, and that inward movement is what lets potassium be secreted.

The texts say the collecting ducts actively secrete hydrogen ions and ammonia. Principal cells secrete potassium, driven by the sodium pump, a lumen-negative charge and aldosterone. Sodium is reabsorbed from the tubule.

Sources: FaustMiller 10eBarash 9e

The shared thread is a kidney with little reserve meeting poor perfusion. Miller, drawing on noncardiac cohorts, names male sex, symptomatic heart failure, hypertension, ascites, non-elective and intraperitoneal operations. Stoelting gives a generic perioperative list without sex, adding sepsis, diabetes, hypovolaemia and anaemia. These fit together: Stoelting is broad, Miller is setting-specific, and both name age, CKD, liver disease and heart failure. Female sex appears in cardiac-surgery indices instead. Both describe age only as advanced or increased; neither sets a fixed cut-off.

Female sex as a noncardiac AKI risk. Female sex is tempting because it appears among risk factors in cardiac-surgery AKI indices and in some perioperative stroke data. For noncardiac surgery the sex association runs the other way: Miller lists male sex, so the cardiac-surgery finding does not carry over.

Miller lists, for noncardiac surgery, older age, male sex, symptomatic heart failure, hypertension, liver disease with ascites, CKD, PAD, COPD, non-elective and intraperitoneal surgery; cardiac indices include female sex. Stoelting's perioperative list adds sepsis, diabetes, hypovolaemia and anaemia but omits sex.

Sources: Miller 10eStoelting 8e

Iodinated contrast harms the kidney through its load. The medium is viscous and puts oxidative stress on the proximal tubular cells, which is why limiting the dose and hydrating the patient help. A kidney that is already impaired has the least reserve to absorb that insult, so established kidney disease raises the risk about 20 times. Diabetes, hypertension, gout and nephrotoxic drugs add to the risk, but every source ranks them below existing renal impairment.

Diabetes as main risk factor. Diabetes is the best-known co-risk factor and often travels with kidney disease, so it feels like the main cause. The sources list it only as a secondary factor. Pre-existing renal impairment is ranked above all others and is the factor that really drives the risk.

Barash names chronic kidney disease as the most important predictor, raising risk 20 times, and lists diabetes, hypertension and gout as lesser factors. KDIGO puts pre-existing renal impairment above every other factor. Miller pairs kidney disease with contrast load as the two main contributors.

Sources: Barash 9eOutside our books: KDIGO Clinical Practice Guideline for Acute Kidney Injury (2012)Miller 10e

Rupture means the patient arrives in haemorrhagic shock, and that shock drives the outcome. After elective repair, mortality fell steadily. After rupture, the operative death rate has barely moved in four decades, and counting those who die before reaching hospital, almost nobody survives. Among survivors the kidney is the organ most often injured. The main cause is haemodynamic instability. A long aortic clamp time and atheromatous emboli add ischaemic damage. So renal injury is several times commoner than after elective surgery, but it still happens less often than death.

Renal failure as the commonest outcome. It tempts because the kidney is the classic casualty of shock and aortic clamping, especially when the aneurysm is suprarenal. Renal injury does strike about a quarter of patients after rupture. But perioperative mortality approaches half, so death remains the more likely outcome.

Perioperative mortality for ruptured aneurysm has stayed near 50% for four decades. Including pre-hospital deaths, it exceeds 90%. Kidney injury followed 26% of emergency repairs for rupture, against 4% after well-controlled elective repair.

Sources: Miller 10eBarash 9e

Retained uraemic toxins worsen peripheral insulin resistance. Dialysis removes some, such as guanidino compounds, so insulin sensitivity and glucose tolerance improve. Other problems are not toxin-driven: long-stimulated parathyroid glands become hyperplastic and secrete autonomously, and amenorrhoea results from hypothalamic-pituitary-gonadal disruption that persists on dialysis. Stoelting's nerve chapter states haemodialysis does not reverse neuropathy, whereas transplantation improves conduction within days. Its renal chapter links any gain to baseline symptom severity, so mild symptoms may ease but established disease does not.

Dialysis reverses uraemic polyneuropathy. Uraemic polyneuropathy sounds toxin-driven, so clearing toxins seems bound to help, and some mild symptoms do ease. But haemodialysis does not reliably reverse established neuropathy. Nerve conduction recovers after transplantation, not dialysis, so dialysis cannot be counted on to improve it.

A recent endocrine review reports better insulin sensitivity once dialysis begins, partly from clearing guanidino compounds. The same review describes self-sustaining parathyroid secretion and gonadal-axis disruption in dialysis patients. Stoelting says haemodialysis fails to reverse uraemic polyneuropathy, whereas transplantation improves nerve conduction.

Sources: Outside our books: Endocrine disorders linked to chronic kidney disease: Mechanisms and clinical implications (2026)Stoelting 8e

Dialysis clears potassium efficiently from plasma, but most body potassium sits inside cells. Immediately after a session the extracellular level is low while intracellular stores have not yet moved out to replace it, so an early sample reads hypokalaemic; the value rises as transcellular equilibration occurs. At the same time, the session removes fluid and shifts electrolytes between compartments, leaving the patient acutely volume depleted. Both are expected simultaneously: one is a laboratory reading, the other a haemodynamic state, which is why surgery right after dialysis is discouraged.

Anion-gap acidosis after dialysis. Anion-gap acidosis is the classic metabolic picture of end-stage renal failure, so it feels like the 'renal' answer. But it describes the patient between sessions, when retained acids accumulate; dialysis is what clears them, so it is not the expected finding immediately afterwards.

One text explains that efficient dialytic potassium clearance, with most potassium held inside cells, makes an early post-dialysis sample suggest hypokalaemia before equilibration. Another advises against operating soon after dialysis because of acute volume depletion and electrolyte shifts, including potassium.

Sources: Stoelting 8eMiller 10e

Diseased kidneys make little erythropoietin and cannot excrete acid, so patients become anaemic and acidotic. They compensate for the low oxygen-carrying capacity in two ways: cardiac output rises, and the dissociation curve moves rightward. A right shift lowers haemoglobin's affinity for oxygen, so tissues extract more oxygen at any given capillary partial pressure. The acidaemia adds to this through the Bohr effect. Failing excretion also lets the cations the kidney normally clears build up, so potassium and magnesium rise.

Hypomagnesaemia in renal failure. Magnesium disturbance is a familiar feature of renal disease, so remembering 'magnesium' without its direction makes low magnesium seem plausible. It fails because reduced filtration retains magnesium. The disorder is hypermagnesaemia, just as potassium shows hyperkalaemia rather than hypokalaemia.

The textbook's list of chronic kidney disease features includes raised potassium, magnesium and phosphate, low calcium and metabolic acidosis. It also lists anaemia with a raised cardiac output and a dissociation curve shifted to the right.

Sources: Stoelting 8e

Roughly 50% of neostigmine's plasma clearance is renal. When the kidneys fail, its clearance drops and its elimination half-life lengthens substantially. Atracurium and cisatracurium break down mainly by spontaneous Hofmann degradation in plasma; atracurium is also broken down by ester hydrolysis. Little parent drug is left for the kidney to excrete, so their clearance, half-life and duration of block stay essentially the same in kidney failure.

Fentanyl's possibly longer half-life. Fentanyl is tempting because its elimination half-life can lengthen in chronic kidney disease. That change is only possible, though, and fentanyl leaves no active metabolites. Neostigmine reliably loses about half its clearance, so it is the drug whose action is predictably prolonged.

Renal excretion handles about 50% of neostigmine clearance, and its half-life is clearly prolonged in anephric patients. Atracurium and cisatracurium undergo plasma degradation with minimal renal excretion, so their duration is unaffected by renal failure; fentanyl's half-life only may lengthen.

Sources: Miller 10eMayo reviewBarash 9eStoelting 8e

Rapid breakdown of a large tumour load floods the circulation with potassium, phosphate and uric acid, which the kidneys must clear. Dehydration, pre-existing renal damage and obstructive uropathy are co-morbidities that raise risk. Raised uric acid can itself cause acute renal failure. LDH generally tracks bulky disease and is a high-risk marker, so a high value indicates higher risk rather than excluding the syndrome. Hyperkalaemia and hypocalcaemia can be fatal, so the syndrome is never benign. Anaesthetic teaching advises avoiding dexamethasone in at-risk oncology patients.

Dexamethasone as a harmless cheap antiemetic. Dexamethasone is a familiar, cheap antiemetic, so it seems harmless here. Anaesthesia texts advise steering clear of it in most cancer patients, particularly those at risk of this syndrome. So giving it here is unsafe.

A haematology consensus lists dehydration among the co-morbidities that raise risk, and says LDH reflects bulky disease and is a high-risk factor. Hyperkalaemia and hypocalcaemia can be life-threatening. An anaesthesia text advises avoiding dexamethasone in oncology patients at risk of the syndrome.

Sources: Outside our books: Consensus conference on the management of tumor lysis syndrome (2008)Barash 9eMiller 10e

Chemotherapy or radiation suddenly destroys many tumour cells, and they release their intracellular contents: potassium, phosphate, and nucleic acids that are metabolised to urate. The excess phosphate binds calcium and lowers its serum level, so hypocalcaemia follows the hyperphosphataemia. Urate crystals and calcium phosphate deposits damage the renal tubules, causing acute kidney injury. Renal failure then reduces potassium clearance, which raises the risk of dysrhythmia. The life-threatening pair is high potassium with low calcium.

Low potassium with low calcium. Potassium is a key electrolyte in this syndrome, so recognising 'potassium' alone makes low potassium seem plausible. The direction is wrong: lysed cells release intracellular potassium, which causes hyperkalaemia. Calcium is the only electrolyte that falls.

Both textbooks describe a metabolic crisis after mass tumour cell destruction. Urate, potassium and phosphate rise, and phosphate drives calcium down. Renal failure follows urate crystal or calcium phosphate deposition, and renal dysfunction makes hyperkalaemic dysrhythmias more likely.

Sources: Miller 10eStoelting 8e

Cirrhosis causes splanchnic and systemic vasodilation through excess nitric oxide, prostacyclins, carbon monoxide and endocannabinoids. Resistance falls and the heart compensates by raising output. Arteriovenous shunts let blood bypass the tissue beds, so little oxygen is extracted and venous blood returns highly saturated. Output and venous saturation therefore both go up, and resistance goes down. Patients are often hypotensive yet warm and well perfused. The fault is lost vascular tone, not pump failure, so an alpha agonist fits. Milrinone would only add more vasodilation.

Low mixed venous saturation. A low mixed venous saturation suggests poor perfusion, so it seems to fit a sick, often hypotensive cirrhotic. But output is high and arteriovenous shunting returns oxygen-rich blood unextracted, so the saturation rises instead.

End-stage liver disease is described as hyperdynamic. Cardiac output or index is raised, systemic and peripheral resistance is low, volume is expanded, shunts are present and mixed venous saturation is around 90%. Milrinone's vasodilation is unwanted. Alpha-agonist drugs can help when hypotension needs treatment.

Sources: Barash 9eMayo reviewOutside our books: Acute Kidney Injury and Hepatorenal Syndrome in Patients with Cirrhosis (2023)

Portal hypertension can damage the lung vessels in two different ways. In portopulmonary hypertension, systemic vasodilatation is paired with local production of pulmonary vasoconstrictors. Pulmonary arterial pressure therefore rises, and the right heart hypertrophies and then fails. This shows as distended neck veins, an enlarged liver, and right atrial and ventricular hypertrophy on the ECG. Hepatopulmonary syndrome is a different lesion: the pulmonary vessels dilate and create an intrapulmonary shunt. Its hallmark is shunt hypoxaemia, often eased by lying supine.

Hepatopulmonary syndrome for any hypoxaemia. It is tempting because liver disease with hypoxaemia is the textbook pairing for hepatopulmonary syndrome. But its lesion is vascular dilatation causing shunt. P pulmonale, right axis deviation and a raised JVP point to pulmonary arterial hypertension with right ventricular strain, which is portopulmonary hypertension.

Overt right ventricular failure shows as raised jugular venous pressure and hepatosplenomegaly, with right atrial and ventricular hypertrophy on ECG. Portopulmonary hypertension is pulmonary arterial hypertension complicating portal hypertension. Hepatopulmonary syndrome is liver dysfunction, unexplained hypoxaemia and intrapulmonary vascular dilatation, sometimes with orthodeoxia.

Sources: Stoelting 8eBarash 9eFaust

In established cirrhosis, scarring obstructs portal flow. The resulting portal hypertension sits upstream of the complications that kill: variceal bleeding, ascites, hepatorenal failure and encephalopathy. Miller therefore treats portal hypertension, rather than any single complication, as the main source of illness and death. Stoelting names intracranial hypertension as the chief cause of early death, but only in acute hepatic failure, where cerebral hyperaemia and a leaky blood–brain barrier swell the brain. Decide by setting: chronic decompensated cirrhosis points to portal hypertension, and fulminant failure points to raised intracranial pressure.

Cerebral oedema as the main killer. Encephalopathy brings cerebral oedema to mind, and it is the main early killer in fulminant hepatic failure. This patient, however, has chronic hepatitis C cirrhosis with ascites. That is decompensated chronic disease, and here portal hypertension and its consequences drive mortality.

One text attributes most cirrhosis-related illness and death to portal hypertension. Another names raised intracranial pressure as the chief early killer in acute hepatic failure. A third lists variceal bleeding, ascites, hepatorenal syndrome and encephalopathy as complications of cirrhosis that carry poor survival.

Sources: Miller 10eStoelting 8e

A stent is tunnelled through liver tissue from a hepatic vein into the portal vein, so portal blood drains straight into the systemic veins. This lowers the portosystemic gradient, which shrinks distended oesophageal varices and controls ascites. Systemic pressure is not the target. Because portal blood now bypasses the liver, encephalopathy can worsen, so existing encephalopathy weighs against the procedure. Cardiac failure, tricuspid regurgitation and significant pulmonary hypertension are contraindications. TIPS is usually chosen after endoscopic or drug treatment has failed, and it often serves as a bridge to transplant.

Encephalopathy as an indication. Encephalopathy is a familiar complication of portal hypertension, so decompressing the portal system looks as if it should help. It does the opposite: the shunted blood bypasses the liver and encephalopathy often worsens. Existing encephalopathy is a reason for caution, not an indication.

TIPS is a portal-to-systemic channel relieving portal hypertension. It is indicated for repeated variceal bleeding despite medical therapy and intractable ascites. Heart failure, tricuspid regurgitation and pulmonary hypertension contraindicate it; encephalopathy may worsen afterwards.

Sources: Barash 9eMiller 10eStoelting 8e

Barash puts sudden liver failure above any score-based ranking. Among the rest, priority follows the score. Faust explains that the score combines bilirubin, creatinine and INR to estimate survival in end-stage disease, so a higher score means sicker and more urgent. Children are scored on PELD, and centres can request exception points when the score misses the true severity of a disease.

Highest severity score always outranks acute liver failure. It is tempting because a higher score does mean sicker and does raise priority. But the score only orders children with chronic or end-stage disease. A child in acute failure goes ahead of every score. Children are also scored on PELD, not MELD.

Donor livers go first to acute liver failure, then to the highest MELD or PELD score with a compatible blood group. Exception points can be requested for paediatric diseases the score underrates. A higher MELD score reflects more severe liver failure.

Sources: Barash 9eFaust

Calcineurin inhibitors are mainstays of transplant regimens, and the kidney is their main target. Ciclosporin's recognised long-term toxicities include kidney damage, hypertension and magnesium loss. Ciclosporin and tacrolimus also reduce the kidney's ability to excrete potassium, so they cause hyperkalaemia, not hypokalaemia. Once magnesium is low, the kidney damage gets worse. Low potassium and acidosis add to the injury too, but they are risk factors that make the toxicity worse, not effects of the drug. Years after transplant, the expected chronic finding is therefore hypomagnesaemia.

Hypokalaemia alongside low magnesium. Low magnesium often comes with low potassium, and hypokalaemia is listed among factors that worsen calcineurin-inhibitor kidney injury, so the two seem to go together. But these drugs impair renal potassium excretion and appear among causes of hyperkalaemia. The expected potassium change is a rise.

An adverse-effect table lists ciclosporin with kidney toxicity, hypertension and low magnesium. A hyperkalaemia table names ciclosporin and tacrolimus as drug causes of impaired renal potassium excretion. Low magnesium and potassium make their kidney injury worse.

Sources: Barash 9e

When the clamps come off, preservative fluid flushed from the graft and stagnant venous blood from the legs enter the circulation together. A sudden rise in potassium is the likely cause of cardiac arrest reported at this moment. Afterwards, recipients take cyclosporine or tacrolimus, and both of these cut renal potassium excretion. A newly placed graft that is not yet working well, or is oliguric, cannot clear the extra potassium. Each of these pushes serum potassium up, not down.

Hypokalaemia from graft diuresis. A new graft cannot concentrate urine well, so you might expect a large diuresis to wash out potassium. But the potassium released at reperfusion, the calcineurin inhibitors and any delay in graft function all raise potassium. That makes high potassium, not low, the problem to anticipate.

Cardiac arrest after the graft artery is unclamped is attributed to sudden hyperkalaemia, as preservative solution and leg venous blood enter the circulation. Cyclosporine and tacrolimus, the standard transplant immunosuppressants, are listed among drugs that cause hyperkalaemia by impairing renal excretion.

Sources: Stoelting 8eBarash 9e

Two different people have surgery, and each has a different complication list, so first decide which patient is meant. The recipient's new liver replaces a diseased one. After liver transplant, early heart and lung problems are more frequent than after kidney transplant. Pleural effusion and pulmonary oedema each reach about 47%, and pneumonia 38%. The donor list is different: infection comes first at 13.2%, with pleural effusion close behind at 11.0%. Pneumothorax and pulmonary embolism are uncommon in donors.

Incisional hernia from the large incision. Incisional hernia is tempting because the large upper-abdominal incision makes wound failure seem likely, and it does appear in donor lists. But it is less frequent: in donors it comes after infection and pleural effusion, and it is not among the recipient's leading early complications.

Liver recipients have more early heart and lung complications than kidney recipients, with pleural effusion and pulmonary oedema each up to 47%. In a large donor series, infection was commonest (13.2%), then pleural effusion (11.0%). Hernia, embolism and pneumothorax were less frequent.

Sources: Miller 10e

When skeletal muscle breaks down, it releases myoglobin, a haem pigment, which the kidneys filter into the urine. This gives the urine a dark, cola-like colour, and serum creatine kinase rises sharply at the same time. A general urine-colour table lists myoglobinuria under both red and brown/black. The anaesthesia texts settle which applies here: in muscle breakdown with acidosis they describe cola-coloured urine. So in rhabdomyolysis, expect dark brown urine.

Red urine because myoglobin resembles haemoglobin. Myoglobin is a haem pigment like haemoglobin, and myoglobinuria does appear in the red row of urine-colour tables. That makes red seem reasonable. But the textbook picture of muscle breakdown is cola-coloured urine, and the same table also places myoglobinuria under brown/black.

In malignant hyperthermia, rigidity can progress to rhabdomyolysis. The texts then describe cola-coloured urine together with myoglobinuria and a marked rise in creatine kinase. A renal function table lists myoglobinuria both as a red cause and as a brown/black cause.

Sources: Mayo reviewStoelting 8eFaust

Sarcoid granulomas produce vitamin D activity that is not properly regulated, and this raises serum calcium. Patients with a history of hypercalcaemia have active vitamin D levels that are high relative to the 25-hydroxy storage form. The calcium excess comes from the inflammation itself. Immunosuppression that damps the granulomas therefore corrects it, alongside intravenous fluids. Persistent hypercalcaemia then injures the kidney, so renal dysfunction is a consequence of the high calcium, not a separate low-calcium state.

Hypocalcaemia from renal failure. Hypocalcaemia is tempting because sarcoidosis can lead to renal failure, and chronic kidney disease is linked to low calcium. In sarcoidosis the order is reversed. Excess vitamin D activity raises calcium first, and the kidney damage follows from that hypercalcaemia.

A systematic review found raised serum calcium in 6% of sarcoidosis patients on routine testing, and nearly half of those develop renal failure. About 90% respond when immunosuppression is used to control granulomatous inflammation and the vitamin D-driven hypercalcaemia that comes with it.

Sources: Outside our books: Diagnosis and Detection of Sarcoidosis: An Official American Thoracic Society Clinical Practice Guideline (2020)