ICU Diuretics Pharmacology — Comprehensive (Loop, Thiazide, K-Sparing, Acetazolamide)

PROXIMAL TUBULE — acetazolamide (carbonic anhydrase inhibitor):

• Carbonic anhydrase (on the luminal brush border) catalyses CO2 + H2O → H2CO3 → H+ + HCO3-. The H+ is secreted into the lumen in exchange for Na+ (Na+/H+ exchanger).
• Acetazolamide inhibits carbonic anhydrase → less H+ available → less Na+/H+ exchange → Na+ and HCO3- are lost in the urine.
• Result: alkaline urine + metabolic acidosis (HCO3 loss). Weak diuresis because downstream segments (TAL, DCT) reabsorb most of the escaped Na+.
• Clinical: corrects metabolic alkalosis (e.g. chronic diuretic use, vomiting), altitude sickness (induces metabolic acidosis → stimulates ventilation → ↑oxygenation), glaucoma (↓CSF/aqueous humour production), altitude acclimatisation.
• Toxicity: metabolic acidosis (intended in alkalosis treatment), hypokalaemia (increased distal Na+ delivery → increased K+ secretion), renal stones (alkaline, calcium-rich urine), paraesthesia. [1]

THICK ASCENDING LIMB (TAL) — loop diuretics (the powerhouse):

• The TAL reabsorbs ~25% of filtered Na+ via the Na-K-2Cl (NKCC2) cotransporter on the luminal membrane. It is impermeable to water — the "diluting segment" — so it generates the corticomedullary osmotic gradient that drives water reabsorption in the collecting duct.
• The TAL also has a paracellular Ca2+ and Mg2+ reabsorption pathway (driven by the lumen-positive potential from K+ back-leak through ROMK).
• Loops block NKCC2 → (a) abolish the osmotic gradient → massive water loss; (b) lose the lumen-positive potential → increased Ca2+ and Mg2+ excretion → hypocalcaemia + hypomagnesaemia.
• This is why loops are the MOST POTENT diuretic (maximal natriuresis, up to 20-25% of filtered Na+) AND why they uniquely cause hypocalcaemia + hypomagnesaemia (the paracellular divalent cation loss).
• Result: powerful diuresis, Na+/K+/Cl-/Ca2+/Mg2+ loss, and (with volume contraction) hypochloraemic metabolic alkalosis. [1]

DISTAL CONVOLUTED TUBULE (DCT) — thiazides:

• The DCT reabsorbs ~5-10% of filtered Na+ via the Na-Cl (NCC) cotransporter (sensitive to thiazides). It is also the main site of active Ca2+ reabsorption (via the TRPV5 channel, driven by basolateral Na+/Ca2+ exchange and Ca-ATPase).
• Thiazides block NCC → more distal Na+ delivery → increased ENaC-mediated Na+ reabsorption in the collecting duct (and thus more K+ and H+ secretion → hypokalaemia + alkalosis).
• The calcium effect: thiazides increase Ca2+ reabsorption (the mechanism is complex — partly volume contraction increasing proximal Ca reabsorption, partly direct upregulation of DCT apical Ca channels) → HYPERcalcaemia / hypercalciuria reduction — opposite to loops. This is exploited to treat idiopathic hypercalciuria and calcium oxalate stones.
• Result: moderate diuresis, HYPERcalcaemia, and the classic thiazide triad: hypokalaemia + hyponatraemia + hyperuricaemia. [1]

CORTICAL COLLECTING DUCT (CCD) — potassium-sparing diuretics:

• The principal cell of the CCD reabsorbs Na+ via the epithelial Na+ channel (ENaC) on the luminal membrane; this creates a lumen-negative potential that drives K+ secretion (via ROMK) and H+ secretion (by intercalated cells). Aldosterone amplifies all of this (↑ENaC, ↑Na+/K+-ATPase, ↑ROMK).
• Two ways to block it: (a) aldosterone receptor antagonists (MRAs) — spironolactone, eplerenone — compete with aldosterone at the mineralocorticoid receptor → ↓ENaC activity → ↓K+ and H+ secretion → retain K+ (hyperkalaemia) and H+ (metabolic acidosis); (b) ENaC blockers — amiloride, triamterene — directly block the apical Na+ channel → same downstream effect.
• Weak diuresis (only 1-3% of Na+) — used NOT primarily for diuresis but to counteract loop/thiazide-induced hypokalaemia and for aldosterone-mediated disease (HFrEF, Conn syndrome, cirrhotic ascites).
• Result: HYPERkalaemia, metabolic acidosis, and (spironolactone only) gynaecomastia, impotence, menstrual irregularities (spironolactone binds androgen and progesterone receptors — eplerenone is more MRA-selective and avoids these).

ASSESS VOLUME STATUS AND PERFUSION FIRST (Stevenson phenotyping):

• Classify the patient as WET or DRY (congestion) and WARM or COLD (perfusion).
• Diuretics are for the WET patient. If WET + COLD (cardiogenic shock), you need inotropes/vasopressors FIRST, then diurese — diuretics alone in a cold, hypoperfused patient worsen AKI.
• Signs of congestion: raised JVP, pulmonary crackles/orthopnoea, peripheral oedema, ascites, lung ultrasound B-lines, IVC >2 cm with no collapse. [1]

CONVERT ORAL LOOP TO IV AND GIVE A STARTING DOSE:

• If the patient was on an oral loop, give 1-2.5x the total daily oral dose as IV (DOSE trial high-dose arm = 2.5x). IV:PO ratio for furosemide is ~2:1 (so 80 mg PO → ~40 mg IV equivalent, but in decompensation use at least the IV equivalent of the home dose, often more).
• Example: home furosemide 80 mg PO daily → start IV furosemide 40-80 mg bolus, assess 2-hour natriuresis/urine output. [1]

USE THE 6-HOUR URINE OUTPUT / SPOT SODIUM TO GUIDE TITRATION (the key to avoiding diuretic under-dosing):

• Measure urine output at 6 hours after the first dose and/or a spot urine sodium at 1-2 hours.
• Good response (urine output >100-150 mL/h at 6 h, or spot Na+ >50-70 mmol/L): continue current dose q12-24h.
• Poor response: DOUBLE the next dose (furosemide dose-response is logarithmic — small increments don't work). The DOSE trial used 2.5x.
• This stepwise doubling is critical — most diuretic "failure" is under-dosing. [1]

CHOOSE BOLUS OR CONTINUOUS INFUSION (DOSE trial — equivalent):

• The DOSE trial (2011) randomised to bolus vs continuous infusion → no difference in symptoms or renal function at 72 h. Either is acceptable.
• Continuous infusion is preferred by some for: very high doses (avoids peak-related ototoxicity), haemodynamically unstable patients (smoother), or when boluses cause large fluid shifts.
• Bolus is preferred for: ease, ability to assess dose-response (a bolus gives a discrete natriuretic response you can measure), lower nursing complexity.
• Rate limit for furosemide infusion: <4 mg/min (or dilute, infuse over ≥20 min for bolus doses >80 mg) to avoid ototoxicity.[1]

SET A DAILY NEGATIVE FLUID BALANCE TARGET:

• Target net negative balance per day depends on congestion severity (typically 1-3 L/day net negative).
• Avoid over-diuresis: if creatinine rises, re-assess — is it from over-diuresis (give a cautious fluid challenge, ease off diuretics) or from worsening congestion/cardiorenal (continue/intensify diuresis)? A small creatinine rise with effective decongestion is acceptable and often resolves. [1]

MONITOR ELECTROLYTES DAILY (or more often):

• Potassium (target 4.0-5.0 in HF — replete aggressively; hypokalaemia → arrhythmia and digoxin toxicity).
• Magnesium (loop-induced hypomagnesaemia worsens hypokalaemia and is often overlooked — replete to keep >0.8 mmol/L).
• Sodium, creatinine, urea.
• Replace K+ and Mg2+ ENTERALLY when possible (slow-release KCl, magnesium aspartate/glycerophosphate); IV only for severe deficits. [1]

IF RESISTANT — ESCALATE TO SEQUENTIAL NEPHRON BLOCKADE:

• Defined as failure to achieve adequate diuresis despite adequate loop doses (e.g. <100 mL/h after an adequate IV loop dose).
• Add a thiazide: metolazone 2.5-5 mg PO (preferred because it retains activity even at low GFR, unlike other thiazides) OR IV chlorothiazide 500-1000 mg.
• Give ~30 min before the loop dose. Expect a powerful synergistic diuresis. Monitor K+, Mg2+, Na+, creatinine every 6-12 h — the combination can cause profound electrolyte disturbance and AKI (volume depletion).[5]
• See the dedicated diuretic-resistance section below.

CONSIDER ADJUNCTS:

• Acetazolamide 250-500 mg if metabolic alkalosis has developed from loop therapy (corrects the alkalosis and may restore loop sensitivity).
• Hypertonic saline + high-dose furosemide (the 'sodium-loading' strategy) — emerging data (e.g. SMAC-HF) suggests improved diuresis, but not standard.
• Ultrafiltration — reserved for refractory cases; CARRESS-HF showed pharmacologic stepwise therapy was superior to ultrafiltration for cardiorenal syndrome with worsening renal function — ultrafiltration NOT first-line.
• Vasopressin V2 receptor antagonists (tolvaptan) — aquaretics, for dilutional hyponatraemia; limited role.

Confirm it is TRUE resistance (not under-dosing):

• Check the dose is adequate. The loop dose-response curve is logarithmic — if response is poor, double the dose (e.g. 40 → 80 → 160 → 240 mg IV furosemide as bolus or split).
• Use the 6-hour urine output (>150 mL/h = good; <100 mL/h = poor) and spot urine sodium (>50-70 mmol/L = good) to quantify response.
• Ensure the drug is reaching the tubule: IV loop given to a patient with intestinal oedema and poor oral absorption will out-perform the same oral dose. [1]

Exclude CORRECTABLE causes of reduced loop efficacy:

• NSAIDs / COX-2 inhibitors — block prostaglandin-mediated afferent arteriole dilation, reducing renal blood flow and luminal furosemide delivery → resistance. STOP them.
• Hypoalbuminaemia (e.g. nephrotic, cirrhosis) — furosemide is highly protein-bound; low albumin means more free drug is metabolised/reaches non-target sites → give albumin + furosemide (evidence mixed but commonly used).
• Renal hypoperfusion (low cardiac output, sepsis) — the kidney is underperfused; the answer is inotropes/vasopressors/resuscitation, not more diuretic. Diuresing a 'cold/shock' kidney worsens AKI.
• Severe renal impairment — even loops lose efficacy at very low GFR (less luminal delivery); higher doses or combination needed. [1]

SWITCH or COMBINE loop agents:

• If oral absorption unreliable (intestinal oedema), switch to IV.
• Consider switching furosemide → torasemide/bumetanide (better bioavailability, longer half-life) — some patients respond better to one loop than another (no robust trial, but reasonable). [1]

ADD A THIAZIDE — sequential nephron blockade (the definitive step):

• Metolazone 2.5-5 mg PO (preferred — active even at GFR <30) OR chlorothiazide 500-1000 mg IV OR hydrochlorothiazide 25-50 mg PO (only if GFR reasonable).
• Give 30-60 min BEFORE the loop dose. The thiazide blocks the hypertrophied DCT (NCC) while the loop blocks the TAL (NKCC2) → synergistic natriuresis.
• Expected outcome: a dramatic diuretic response over 12-24 h (often several litres). Be ready for it. [1]

MONITOR INTENSIVELY (the dangerous part):

• Check K+, Mg2+, Na+, creatinine every 6-12 hours for the first 24-48 h.
• Anticipate and pre-empt: aggressive K+ and Mg2+ replacement; hold the combination if creatinine rises sharply or over-diuresis (hypotension, rising urea).
• Stop once euvolaemic (JVP normalised, B-lines resolved, weight stable). [1]

REFRACTORY CASES — consider:

• Hypertonic saline + high-dose loop (the 'sodium loading' / SMAC-HF approach) — emerging.
• Vasopressin V2 antagonist (tolvaptan) — aquaretic for dilutional hyponatraemia, modest decongestion.
• Ultrafiltration (veno-venous) — last resort; CARRESS-HF (2012) showed it was inferior to stepwise pharmacologic therapy for cardiorenal syndrome with worsening renal function (more adverse events, creatinine rose). Reserve for truly refractory fluid overload unresponsive to maximal pharmacologic therapy.
• Dopamine ("renal-dose") — NO benefit, do not use.