Metabolic Alkalosis
pH > 7.45 with elevated HCO₃⁻. Most common acid-base disorder in hospitalized patients. Classify by urine chloride: saline-responsive (UCl < 20) vs saline-resistant (UCl > 20). Replace chloride, not just volume.
🔍 Overview
Pathophysiology
Two requirements: (1) Generation -something creates the alkalosis (vomiting, diuretics, etc.), (2) Maintenance -something prevents the kidney from excreting the excess HCO₃⁻ (volume depletion, Cl⁻ depletion, hypokalemia, mineralocorticoid excess).
Causes, CLEVER PD Mnemonic
- C, Contraction alkalosis (volume depletion: diuretics, GI losses, burns)
- L, Licorice (glycyrrhizin inhibits 11β-HSD2 → apparent mineralocorticoid excess) / Laxative abuse (chronic/surreptitious → volume contraction + hypokalemia, mimics Bartter/Gitelman)
- E, Endocrine: Conn's (primary hyperaldosteronism), Cushing syndrome, Bartter, Gitelman
- V, Vomiting / NG suction (#1 cause overall)
- E, Exogenous alkali: milk-alkali syndrome, massive citrate (transfusion), IV NaHCO₃
- R, Refeeding alkalosis (intracellular shift of phosphate/K⁺ drives H⁺ into cells)
- P, Post-hypercapnic (rapid correction of chronic respiratory acidosis leaves retained HCO₃⁻)
- D, Diuretics (loop, thiazide, both active use and post-diuretic states)
Clinical tip: Most bedside cases come down to Vomiting and Diuretics. If neither fits and BP is elevated → think Endocrine (aldosterone/cortisol excess). If the alkalosis is unexplained and the patient was recently extubated → think Post-hypercapnic.
Classification by Urine Chloride
| UCl < 20 (Saline-Responsive) | UCl > 20 (Saline-Resistant) |
|---|---|
| Vomiting / NGT suction (#1 cause) | Primary hyperaldosteronism |
| Post-diuretic use (drug cleared / last dose >24h ago) | Current (active) diuretic use |
| Post-hypercapnic | Cushing syndrome |
| Chloride-losing diarrhea (rare) | Bartter/Gitelman syndrome |
| Treatment: NS + KCl | Treatment: Treat underlying cause |
💡 Why KCl (not K-phos or K-acetate) in saline-responsive alkalosis?
The chloride in KCl is doing just as much work as the potassium. Here's why the specific salt matters:
- Cl⁻ is the "anion companion" the kidney needs to dump HCO₃⁻. The distal tubule (pendrin transporter) exchanges intracellular HCO₃⁻ for luminal Cl⁻. No Cl⁻ delivered → no HCO₃⁻ excreted.
- K-acetate and K-citrate are metabolized to HCO₃⁻ in the liver -giving these to an alkalotic patient makes the alkalosis worse.
- K-phosphate is fine for K⁺ but provides phosphate, not chloride -doesn't help the kidney excrete bicarbonate.
- Hypokalemia itself perpetuates alkalosis via H⁺/K⁺ shift and renal H⁺ wasting → K⁺ must be repleted, and giving it as Cl⁻ fixes two deficits with one order.
Bottom line: NS + KCl works because it replaces both missing anions (Cl⁻) and cations (K⁺) with a form the kidney can use to dump HCO₃⁻. Swapping in K-acetate defeats the treatment.
Causes by Mechanism
| Mechanism | Chloride-Responsive (UCl < 20) | Chloride-Resistant (UCl > 20) |
|---|---|---|
| GI H⁺ loss | Vomiting, NG suction, villous adenoma (rare) | , |
| Renal H⁺ loss | Post-diuretic use (drug cleared, last dose >24h ago), post-hypercapnic | Hyperaldosteronism, Cushing, Bartter/Gitelman, licorice ingestion, current (active) diuretic use |
| HCO₃⁻ gain | Citrate in massive transfusion, milk-alkali syndrome | , |
| Contraction | Diuretics (volume loss concentrates HCO₃⁻) | , |
| Intracellular H⁺ shift | Hypokalemia (H⁺ moves into cells as K⁺ moves out) | , |
Contraction Alkalosis, Detailed Mechanism
Contraction alkalosis occurs through three synergistic mechanisms: (1) Concentration effect, loss of Cl⁻-rich, HCO₃⁻-poor fluid (e.g., from loop diuretics) concentrates the remaining HCO₃⁻ in a smaller plasma volume. (2) RAAS activation, volume contraction activates the renin-angiotensin-aldosterone system, which increases proximal tubule Na⁺/HCO₃⁻ reabsorption and distal H⁺ secretion. (3) Cl⁻ depletion, without adequate Cl⁻, the kidney cannot excrete HCO₃⁻ (HCO₃⁻ reabsorption in the proximal tubule is linked to Cl⁻ availability). This is why NS (which provides both Na⁺ and Cl⁻) corrects the alkalosis, it expands volume, suppresses RAAS, and provides Cl⁻ for renal HCO₃⁻ excretion.
Clinical pearl, Bartter vs Gitelman: Bartter syndrome mimics chronic loop diuretic use: defective NKCC2 in the thick ascending limb → Na⁺/K⁺/Cl⁻ wasting, metabolic alkalosis, hypokalemia, normal BP. Gitelman syndrome mimics chronic thiazide use: defective NCC in the distal convoluted tubule → same picture as Bartter plus hypocalciuria and hypomagnesemia. Both autosomal recessive channelopathies. Distinguish from surreptitious diuretic abuse with a urine diuretic screen.
Timing matters, same drug, two patterns: Diuretics can appear on either side of the table depending on when UCl is sampled. Active/current use (drug still blocking Na/Cl reabsorption) → ongoing urinary Cl wasting → UCl >20 (saline-resistant pattern). Post-diuretic / remote use (drug cleared, typically >24h since last dose) → kidneys resume normal Cl reabsorption, but the alkalosis persists via volume contraction and K⁺/H⁺ depletion → UCl <20 (saline-responsive). Always ask when the last dose was given before interpreting the UCl.
Post-Hypercapnic Alkalosis, Detailed Mechanism
Setup, chronic respiratory acidosis: In chronic CO₂ retention (COPD, obesity hypoventilation, neuromuscular disease, OSA), PaCO₂ stays elevated for days to weeks. The kidneys compensate by retaining HCO₃⁻ (~3.5 mEq/L rise in [HCO₃⁻] per 10 mmHg rise in PaCO₂) so the pH drifts back toward normal. The patient looks like: high PaCO₂, high HCO₃⁻, near-normal pH, a stable compensated respiratory acidosis.
The flip, rapid CO₂ correction: If PaCO₂ is dropped quickly (typically iatrogenic, aggressive mechanical ventilation targeting a "normal" PaCO₂ of 40, or rapid NIV initiation), the compensatory HCO₃⁻ doesn't leave fast enough. The kidneys need 1–2 days to excrete the excess bicarb, but PaCO₂ can be normalized in minutes. Result: pH overshoots → frank metabolic alkalosis unmasked on top of now-normal ventilation.
Why it persists (the "maintenance" problem): These patients are often concurrently volume-depleted (diuretics for cor pulmonale, poor PO intake) and Cl⁻-depleted, so the kidney can't efficiently excrete the retained HCO₃⁻ even once PaCO₂ is normal. The alkalosis sticks around until Cl⁻ and volume are replaced.
Clinical consequences, why this matters: A post-hypercapnic alkalosis can cause respiratory depression (alkalemia suppresses central respiratory drive → hypoventilation → CO₂ re-accumulates → failure to wean from the vent). It can also trigger hypokalemia, hypocalcemia (ionized), seizures, and arrhythmias. Critically, in a ventilated COPD patient it becomes a self-perpetuating weaning failure cycle.
Prevention & treatment:
- Don't over-ventilate, target the patient's baseline/home PaCO₂ (often 50–65 mmHg in chronic CO₂ retainers), not 40. Check old ABGs or discharge summaries.
- Lower minute ventilation if alkalosis develops (reduce rate or tidal volume) → allow PaCO₂ to rise.
- Replace chloride, NS (not LR) gives Na⁺ and Cl⁻; the kidney needs Cl⁻ to excrete HCO₃⁻.
- Correct K⁺ and Mg²⁺, hypokalemia perpetuates the alkalosis via distal H⁺ secretion.
- Acetazolamide 250–500 mg IV/PO is the pharmacologic option if volume overload prevents saline, induces renal HCO₃⁻ wasting (watch for hypokalemia).
- Rarely: HCl infusion via central line for severe pH >7.55 refractory to the above.
Bedside pearl: Before intubating a COPD exacerbation, look at prior ABGs. A patient whose baseline PaCO₂ is 60 and HCO₃⁻ is 32 does not want a PaCO₂ of 40, that will leave them with HCO₃⁻ 32 and PaCO₂ 40 = pH ~7.52 (post-hypercapnic alkalosis by iatrogenesis). "Permissive hypercapnia" in chronic CO₂ retainers isn't just lung-protective, it's also alkalosis-protective.
🚨 Management
Treatment Based on Urine Chloride
- Saline-responsive (UCl < 20): IV NS (provides Cl⁻ for kidney to excrete HCO₃⁻) + KCl repletion. Fix the volume and chloride deficit.
- Saline-resistant (UCl > 20): Treat underlying cause (e.g., spironolactone for hyperaldosteronism, stop diuretics)
- Severe/refractory: Acetazolamide (Diamox) 250–500 mg IV -forces renal HCO₃⁻ wasting
- Life-threatening (pH > 7.55): Consider HCl infusion (0.1–0.2 N) via central line, or dialysis
Hypokalemia perpetuates metabolic alkalosis. When K⁺ is low, the kidney reabsorbs K⁺ in exchange for H⁺ secretion → paradoxical aciduria. Replace K⁺ aggressively.
🔑 The 3 C's of Saline-Responsive Alkalosis Repletion
Three deficits maintain a chloride-responsive metabolic alkalosis. You must replace all three -fixing volume alone won't correct it.
- Chloride (Cl⁻) -the kidney can only excrete HCO₃⁻ when paired with a reabsorbable anion. Cl⁻ depletion = HCO₃⁻ retention. Give: NS (154 mEq/L Cl⁻).
- Contraction (volume) -volume depletion activates RAAS → proximal Na⁺/HCO₃⁻ reabsorption. Until ECF volume is restored, the kidney "clings" to bicarbonate. Give: isotonic crystalloid.
- K⁺ (potassium) -hypokalemia drives intracellular H⁺ shift, renal H⁺ secretion, and ammoniagenesis. The alkalosis will not resolve without K⁺ repletion. Give: KCl 10-40 mEq/h (KCl, not K-phos or K-acetate -chloride matters).
Mnemonic: "Chloride, Contraction, K⁺" -or remember that NS + KCl covers all three in one order.
Treatment Algorithm
| Step | Condition | Action |
|---|---|---|
| 1 | Check urine Cl⁻ | UCl < 20 → saline-responsive. UCl > 20 → saline-resistant. |
| 2a | Saline-responsive | IV NS 125–250 mL/h + KCl 10–40 mEq/h. Goal: replace Cl⁻ deficit and correct volume. |
| 2b | Saline-resistant | Identify and treat underlying cause: spironolactone for hyperaldosteronism, stop offending diuretics, dexamethasone suppression for Cushing. |
| 3 | Refractory (HCO₃⁻ > 40 despite above) | Acetazolamide 250–500 mg IV q6–12h, forces renal HCO₃⁻ wasting. Monitor K⁺ (causes K⁺ loss). |
| 4 | Severe / life-threatening (pH > 7.55) | HCl infusion 0.1–0.2 N via central line at 100–200 mL/h. Or: NH₄Cl, or hemodialysis with low-bicarbonate bath. |
HCl infusion is caustic, must go through a central line. Calculate Cl⁻ deficit: 0.5 × weight(kg) × (103 – measured Cl⁻). Replace half over 12–24 hours. Monitor ABG q2–4h during infusion.
🧪 Workup
- ABG -confirm primary metabolic alkalosis (↑ pH, ↑ HCO₃⁻, compensatory ↑ PaCO₂)
- Urine chloride -key classification tool. < 20 = saline-responsive, > 20 = saline-resistant
- BMP -K⁺ (usually low), Cl⁻ (usually low)
- Urine electrolytes -Na⁺, K⁺, Cl⁻
- If saline-resistant: renin, aldosterone, cortisol
💊 Medications
| Drug | Dose | Indication |
|---|---|---|
| IV Normal Saline | 125–250 mL/h | Saline-responsive alkalosis -provides Cl⁻ |
| Potassium Chloride (KCl) | 10–40 mEq/h IV (max 40 mEq/h via central) | K⁺ repletion -essential for correction |
| Acetazolamide (Diamox) | 250–500 mg IV q6–12h | Refractory alkalosis -carbonic anhydrase inhibitor → renal HCO₃⁻ wasting |
| Spironolactone (Aldactone) | 25–100 mg PO daily | Hyperaldosteronism-related alkalosis |
📋 On Rounds
Pimp Questions
Why is urine chloride more useful than urine sodium in metabolic alkalosis?
In metabolic alkalosis, the kidney may be "obligated" to excrete Na⁺ with HCO₃⁻ (bicarbonuria) even in volume-depleted states, making urine Na⁺ unreliable. Urine Cl⁻ is not affected by this obligatory loss and accurately reflects volume status. UCl < 20 = volume/Cl⁻ depleted → will respond to saline. UCl > 20 = not volume depleted → look for mineralocorticoid excess.
What is "contraction alkalosis" and how does it occur?
When volume contracts (diuretics, vomiting), the same amount of HCO₃⁻ is dissolved in less plasma volume → concentration effect. Additionally, volume depletion activates RAAS → increased proximal tubule Na⁺/HCO₃⁻ reabsorption → maintenance of alkalosis. The kidney "can't let go" of HCO₃⁻ because it's trying to retain Na⁺ and volume. Giving NS provides Na⁺ and Cl⁻, allowing the kidney to excrete HCO₃⁻.
Why does hypokalemia cause paradoxical aciduria?
When K⁺ is depleted, the kidney's principal cells in the collecting duct reabsorb K⁺ instead of secreting it, and intercalated cells secrete more H⁺ to maintain electroneutrality. This produces acidic urine despite systemic alkalosis -"paradoxical aciduria." It also means the kidney is generating NEW HCO₃⁻, perpetuating the alkalosis. K⁺ must be repleted to correct the alkalosis.
A patient on furosemide has UCl of 35. Does this mean the alkalosis is saline-resistant?
Not necessarily. Current diuretic use is the major exception to the UCl framework. Furosemide blocks the Na-K-2Cl cotransporter in the loop of Henle, causing obligatory Cl⁻ wasting in the urine → UCl > 20 even though the underlying alkalosis IS saline-responsive. The key: if the patient is actively taking diuretics, UCl is unreliable. Stop the diuretic, wait 48–72h, then recheck UCl. Alternatively, treat empirically with NS + KCl and reassess.
How does acetazolamide correct metabolic alkalosis, and what is its main side effect?
Acetazolamide inhibits carbonic anhydrase in the proximal tubule, blocking the conversion of CO₂ + H₂O → H₂CO₃ → H⁺ + HCO₃⁻. This prevents HCO₃⁻ reabsorption, causing renal HCO₃⁻ wasting (bicarbonaturia). Main side effect: hypokalemia, the increased distal Na⁺ delivery drives K⁺ secretion. Always supplement K⁺ when using acetazolamide. Also causes metabolic acidosis if over-used.
What is the expected respiratory compensation for metabolic alkalosis?
The expected PaCO₂ = 0.7 × [HCO₃⁻] + 21 (± 2). The body compensates by hypoventilating to retain CO₂. However, respiratory compensation is limited, PaCO₂ rarely exceeds 55–60 mmHg because hypoxemia from hypoventilation triggers a competing drive to breathe. If PaCO₂ is higher than expected, suspect a concurrent respiratory acidosis (e.g., COPD + vomiting).
How do you differentiate Bartter from Gitelman syndrome?
Both cause metabolic alkalosis with hypokalemia and are saline-resistant (UCl > 20). Bartter = defect in Na-K-2Cl cotransporter (loop of Henle) → mimics loop diuretic. Presents in childhood with polyuria, normal/high urinary calcium, and hypercalciuria. Gitelman = defect in NaCl cotransporter (DCT) → mimics thiazide diuretic. Presents in adolescence/adulthood with milder symptoms, hypocalciuria, and hypomagnesemia.
Why does vomiting cause metabolic alkalosis if the stomach just regenerates the HCl?
When parietal cells secrete H⁺ into the stomach lumen, they generate HCO₃⁻ on the basolateral side (“alkaline tide”). Normally, this HCO₃⁻ is neutralized when pancreatic secretions release HCO₃⁻ into the duodenum to neutralize gastric acid. With vomiting, the H⁺ is lost externally and never reaches the duodenum → the alkaline tide HCO₃⁻ accumulates without being offset.
Clinical Examples
📋 Case 1, Post-Emetic Alkalosis
Patient: 34F with hyperemesis gravidarum at 10 weeks gestation. Vomiting 8–10x/day for 2 weeks. Lethargic, dry mucous membranes. HR 112, BP 88/52. Labs: pH 7.56, PaCO₂ 48, HCO₃⁻ 38, K⁺ 2.6, Cl⁻ 78, Na⁺ 132.
Assessment:
- Primary metabolic alkalosis (pH 7.56, HCO₃⁻ 38)
- Expected PaCO₂ = 0.7 × 38 + 21 = 47.6, matches (appropriate respiratory compensation)
- Urine Cl⁻: 8 mEq/L → saline-responsive
- Severe hypokalemia (2.6) and hypochloremia (78)
Treatment:
- IV NS at 200 mL/h, volume resuscitation + Cl⁻ replacement
- KCl 40 mEq in 1L NS x 3, aggressive K⁺ repletion (cannot correct alkalosis without fixing K⁺)
- Ondansetron 4 mg IV q8h, antiemetic to stop ongoing losses
- Monitor: BMP q6h, urine Cl⁻ at 24h (should rise > 20 once repleted, confirming correction)
Result: After 3L NS + 120 mEq KCl over 24h: pH 7.44, HCO₃⁻ 26, K⁺ 3.8, Cl⁻ 98. Alkalosis corrected.
Key lesson: Classic saline-responsive alkalosis. The triad of vomiting + low UCl + hypokalemia = give NS + KCl aggressively. Must fix K⁺ to fix the alkalosis.
📋 Case 2, Diuretic-Induced Alkalosis
Patient: 68M with CHF (EF 25%) on furosemide 80 mg BID. Admitted for dyspnea. Labs: pH 7.52, PaCO₂ 50, HCO₃⁻ 36, K⁺ 2.9, Cl⁻ 82, Cr 1.4. Urine Cl⁻: 42 mEq/L.
Assessment:
- Metabolic alkalosis from chronic loop diuretic use (contraction alkalosis + Cl⁻/K⁺ wasting)
- UCl is 42, but patient is on active furosemide → UCl is unreliable. This is saline-responsive masquerading as saline-resistant.
- Challenge: cannot give aggressive NS to a patient with EF 25%, will worsen volume overload
Treatment:
- Hold or reduce furosemide, stop the ongoing Cl⁻/K⁺ losses
- Acetazolamide 250 mg IV q12h, forces renal HCO₃⁻ wasting without volume loading (ideal for CHF patients who cannot tolerate NS)
- KCl 40 mEq PO q8h + IV KCl as needed, K⁺ repletion essential
- Spironolactone 25 mg daily, K⁺-sparing diuretic, also blocks aldosterone-driven H⁺ secretion
- Monitor: Daily weights, I/Os, BMP q8h, telemetry (K⁺ 2.9 = arrhythmia risk)
Key lesson: In CHF patients, you cannot just give NS. Acetazolamide is the key tool, it corrects the alkalosis without adding volume. Always pair with K⁺ repletion since acetazolamide causes additional K⁺ wasting.
📋 Case 3, Primary Hyperaldosteronism (Conn Syndrome)
Patient: 52F with resistant hypertension (on 3 agents including amlodipine, losartan, HCTZ). Incidental labs show K⁺ 2.8 and HCO₃⁻ 34. BP 168/102 despite medications. No vomiting, no diarrhea.
Assessment:
- Metabolic alkalosis with hypokalemia in the absence of GI losses or diuretics other than HCTZ
- Urine Cl⁻: 38 mEq/L → saline-resistant
- Resistant HTN + spontaneous hypokalemia + metabolic alkalosis = screen for primary hyperaldosteronism
Workup:
- Step 1: Morning aldosterone/renin ratio (ARR). Hold HCTZ for 2 weeks, switch losartan to verapamil (does not affect RAAS). Result: aldosterone 28 ng/dL, PRA 0.3 ng/mL/h → ARR = 93 (cutoff > 30).
- Step 2: Confirmatory test, oral sodium loading or saline infusion test. Aldosterone fails to suppress → confirmed primary hyperaldosteronism.
- Step 3: CT adrenals → 1.8 cm left adrenal adenoma. Adrenal vein sampling confirms lateralization.
Treatment:
- Laparoscopic left adrenalectomy, unilateral adenoma with confirmed lateralization
- Pre-op: Spironolactone 50 mg BID + KCl supplementation to normalize K⁺ and BP
- Post-op: HTN resolved on 1 agent. K⁺ and HCO₃⁻ normalized. Alkalosis corrected.
Key lesson: Resistant HTN + hypokalemia + metabolic alkalosis = think primary hyperaldosteronism (Conn syndrome). It is the most common secondary cause of HTN (5–10% of all HTN). Screen with aldosterone/renin ratio. Saline-resistant alkalosis that does not correct with NS, must treat the underlying cause.
⚡ Summary
It Takes Two Steps
Generation, then maintenance. Something must add bicarbonate or lose acid, and then the kidney must fail to excrete the excess, which it does when volume, chloride or potassium is depleted, or when mineralocorticoid activity is high. Treating only the generating event without fixing the maintenance factor is why the alkalosis persists.
Urine Chloride Splits the Differential
Not urine sodium. UCl below 20 = saline-responsive: vomiting, nasogastric suction, prior diuretic use, contraction alkalosis, post-hypercapnic. UCl above 20 = saline-resistant: primary hyperaldosteronism, Cushing syndrome, active diuretic use, Bartter and Gitelman, severe potassium depletion, licorice. Urine sodium is unreliable here because bicarbonaturia drags sodium out with it even in a volume-depleted patient.
Replace Chloride, Not Just Volume
Chloride is the anion the kidney needs in order to excrete bicarbonate: without it, the collecting duct cannot exchange, so the bicarbonate is retained no matter how much volume is given. Normal saline plus potassium chloride is the treatment for the saline-responsive group. Use KCl, not potassium phosphate or acetate, because acetate is metabolized to bicarbonate and makes the alkalosis worse.
Most Bedside Cases Are Two Things
Vomiting and diuretics. If neither fits and the blood pressure is high, think endocrine, aldosterone or cortisol excess. If it is unexplained and the patient was recently extubated, think post-hypercapnic. CLEVER PD is the fuller list: Contraction, Licorice, Endocrine, Vomiting, Excess alkali, Refeeding, Post-hypercapnia, Diuretics.
The Same Drug Sits on Both Sides
Diuretics give a high urine chloride while the drug is still working and a low one once it has worn off, so the classification depends on when you sampled. Ask when the last dose was given before concluding that a patient is saline-resistant.
Post-Hypercapnic Alkalosis Blocks Weaning
The kidney retained bicarbonate to compensate for chronic CO₂ retention; correcting the CO₂ quickly leaves the bicarbonate behind. The resulting alkalemia suppresses central respiratory drive, so the patient hypoventilates, the CO₂ climbs again and the wean fails. Prevention is not over-ventilating: target the patient's own baseline PaCO₂, often 50 to 65 mmHg in a chronic retainer, not 40. Check an old blood gas or discharge summary.
Fix the Potassium or Nothing Works
Hypokalemia both causes and sustains metabolic alkalosis, by shifting hydrogen ions into cells and stimulating renal ammoniagenesis and bicarbonate reabsorption. Potassium repletion is part of the treatment, not an afterthought, and in severe depletion it is the limiting step.
The Saline-Resistant Group Needs Different Treatment
Saline will not correct these, and volume loading a hyperaldosteronism patient just raises the blood pressure. Mineralocorticoid excess → spironolactone or eplerenone. Ongoing gastric losses → a proton pump inhibitor to reduce acid loss at source. Severe or symptomatic alkalosis with volume overload → acetazolamide, which produces a bicarbonate diuresis but wastes potassium, so replace it. Dialysis for severe alkalosis in renal failure.
Related Topics
Acid-Base DisordersCalciphylaxis (Calcific Uremic Arteriolopathy)Chronic Kidney Disease (CKD)Contrast-Induced NephropathyDialysis IndicationsHypernatremia