High Anion Gap Acidosis: Sorting Causes in the Middle of the Night

Dr Neeraj Manikath , claude.ai

Abstract

High anion gap metabolic acidosis (HAGMA) represents one of the most challenging diagnostic scenarios in critical care medicine, particularly during emergency presentations and night shifts when resources may be limited. This review provides a systematic approach to HAGMA evaluation using the practical "GOLD MARK" mnemonic, emphasizing bedside prioritization strategies and time-sensitive interventions. We present evidence-based diagnostic algorithms, clinical pearls, and practical "hacks" to expedite accurate diagnosis and appropriate management in the acute setting.

Keywords: High anion gap metabolic acidosis, GOLD MARK mnemonic, critical care, emergency diagnosis, metabolic disorders

Introduction

High anion gap metabolic acidosis (HAGMA) is defined as a metabolic acidosis (pH < 7.35, HCO₃⁻ < 22 mEq/L) with an elevated anion gap (>12 mEq/L using standard laboratory values).¹ The anion gap, calculated as [Na⁺] - ([Cl⁻] + [HCO₃⁻]), represents unmeasured anions in plasma and serves as a crucial diagnostic tool in acid-base disorders.

The challenge for the critical care physician lies not in recognizing HAGMA, but in rapidly identifying its underlying cause during time-pressured scenarios—often in the middle of the night when subspecialty consultation may be unavailable and diagnostic resources limited. This review presents a systematic approach using the "GOLD MARK" mnemonic, designed specifically for bedside prioritization and rapid clinical decision-making.

Pathophysiology: The Foundation

Anion Gap Fundamentals

The anion gap reflects the principle of electroneutrality in plasma. Under normal conditions, the major measured cations (Na⁺, K⁺) are balanced by measured anions (Cl⁻, HCO₃⁻) plus unmeasured anions (primarily albumin, phosphate, sulfate, and organic acids).²

Clinical Pearl #1: Always correct the anion gap for hypoalbuminemia. For every 1 g/dL decrease in albumin below 4 g/dL, add 2.5 to the calculated anion gap.³

Mechanisms of HAGMA

HAGMA develops through four primary mechanisms:

Increased acid production (endogenous or exogenous)
Decreased acid excretion (renal failure)
Loss of bicarbonate with chloride retention
Dilutional effects (rare)

The GOLD MARK Mnemonic: A Bedside Approach

The traditional "MUDPILES" mnemonic, while comprehensive, lacks the clinical prioritization essential for emergency management. The "GOLD MARK" system prioritizes causes by:

Frequency of presentation
Time-sensitivity of intervention
Immediate life-threat potential
Availability of bedside diagnostics

G - Glycols and Glucose (DKA)

O - Opioids and Other drugs

L - Lactate and Liver failure

D - Dialysis needs (Uremia)

M - Methanol and toxic alcohols

A - ASA (Salicylates) and other toxins

R - Renal failure (acute/chronic)

K - Ketones (not DKA)

Detailed Clinical Approach

G - Glycols and Glucose (DKA)

Diabetic Ketoacidosis (DKA)

Prevalence: Most common cause of HAGMA in emergency settings⁴
Diagnostic criteria: Glucose >250 mg/dL, positive ketones, pH <7.30, HCO₃⁻ <15 mEq/L
Bedside hack: Urine ketones can be checked immediately; serum β-hydroxybutyrate >3.8 mmol/L confirms significant ketosis⁵

Ethylene Glycol Poisoning

Clinical clue: History of antifreeze ingestion, altered mental status, crystalluria
Diagnostic pearl: Calcium oxalate crystals in urine (envelope or needle-shaped)
Time-sensitive: Fomepizole must be initiated within hours⁶

Clinical Pearl #2: In suspected ethylene glycol poisoning, check for fluorescence under Wood's lamp (some antifreezes contain fluorescein), though absence doesn't rule out ingestion.

O - Opioids and Other Drugs

Propylene Glycol Toxicity

Setting: High-dose IV medications (lorazepam, phenytoin, etomidate)
Diagnostic clue: Unexplained HAGMA in ICU patients receiving these medications⁷
Calculation hack: Propylene glycol level (mg/dL) ÷ 7.6 = estimated contribution to anion gap

Metformin-Associated Lactic Acidosis (MALA)

Risk factors: Renal impairment, contrast exposure, acute illness
Pearl: Metformin level >5 mg/L suggests toxicity, but clinical context is key⁸

L - Lactate and Liver Failure

Lactic Acidosis

Type A (Hypoxic): Shock, hypoxemia, severe anemia
Type B (Non-hypoxic): Medications, inherited disorders, malignancy

Bedside Prioritization Algorithm:

Check point-of-care lactate immediately
If lactate >4 mmol/L and patient unstable → assume Type A, resuscitate
If lactate >4 mmol/L and patient stable → consider Type B causes

Clinical Pearl #3: A lactate:pyruvate ratio >25:1 suggests tissue hypoxia, while normal ratio suggests metabolic causes.⁹

Liver Failure

Mechanism: Impaired lactate clearance, accumulation of organic acids
Diagnostic clue: Elevated transaminases, coagulopathy, altered mental status

D - Dialysis Needs (Uremia)

Uremic Acidosis

Threshold: Typically occurs when GFR <15 mL/min/1.73m²
Mechanism: Decreased ammoniagenesis, impaired acid excretion
Pearl: BUN:creatinine ratio often <10:1 in chronic kidney disease vs. >20:1 in prerenal azotemia¹⁰

Nightshift Hack: If creatinine >5 mg/dL with HAGMA and no obvious alternative cause, assume uremic acidosis and consider emergent dialysis consultation.

M - Methanol and Toxic Alcohols

Methanol Poisoning

Clinical triad: Visual disturbances, altered mental status, HAGMA
Diagnostic challenge: Often delayed presentation (12-24 hours post-ingestion)
Laboratory clue: Elevated osmolar gap initially, then isolated HAGMA¹¹

Isopropanol

Unique feature: Osmolar gap without significant HAGMA (metabolizes to acetone, not organic acids)
Clinical clue: "Fruity" breath odor, altered mental status

Clinical Pearl #4: Calculate osmolar gap = measured osmolality - calculated osmolality. Calculated osmolality = 2[Na⁺] + [glucose]/18 + [BUN]/2.8. Normal gap <10 mOsm/kg.

A - ASA (Salicylates) and Other Toxins

Salicylate Poisoning

Unique pattern: Mixed acid-base disorder (respiratory alkalosis initially, then metabolic acidosis)
Diagnostic clue: Tinnitus, altered mental status, hyperthermia
Laboratory finding: Often concurrent hyperglycemia or hypoglycemia¹²

Iron Poisoning

Timeline: GI symptoms (0-6h) → apparent recovery (6-24h) → systemic toxicity (12-48h)
Diagnostic clue: Radiopaque tablets on abdominal X-ray, GI bleeding

R - Renal Failure

Acute Kidney Injury (AKI)

Mechanism: Rapid accumulation of organic acids
Pearl: AKI-associated HAGMA typically develops when creatinine >3 mg/dL acutely

Chronic Kidney Disease

Threshold: Usually GFR <20 mL/min/1.73m²
Associated findings: Hyperphosphatemia, elevated PTH, anemia

K - Ketones (Non-DKA)

Starvation Ketosis

Setting: Prolonged fasting, eating disorders, post-operative state
Distinguishing feature: Mild ketosis (β-hydroxybutyrate 1-3 mmol/L) vs. severe in DKA (>3.8 mmol/L)¹³

Alcoholic Ketoacidosis

Clinical scenario: Chronic alcohol use, recent decreased intake, nausea/vomiting
Pearl: Often euglycemic or hypoglycemic (unlike DKA)

Bedside Diagnostic Algorithm

The 5-Minute HAGMA Assessment

Step 1: Immediate Bedside Tests (0-2 minutes)

Point-of-care glucose
Point-of-care lactate
Urine ketones (dipstick)
Vital signs and mental status

Step 2: History Rapid-Fire (2-3 minutes)

Diabetes history
Recent medications/IV therapy
Ingestion history
Alcohol use pattern
Recent illness/surgery

Step 3: Physical Examination Priorities (3-5 minutes)

Kussmaul respirations
Breath odor (fruity, alcoholic)
Volume status
Neurological status
Skin findings (flushing, diaphoresis)

Clinical Pearl #5: The "HAGMA Trifecta"—altered mental status, Kussmaul respirations, and distinctive breath odor—narrows the differential significantly.

Laboratory Interpretation: Pearls and Pitfalls

Critical Laboratory Sequence

Tier 1 (STAT - within 30 minutes):

Complete metabolic panel
Arterial blood gas
Point-of-care glucose and lactate
Urine ketones

Tier 2 (Urgent - within 2 hours):

Serum ketones (β-hydroxybutyrate)
Osmolality (if toxic alcohol suspected)
Salicylate level
Acetaminophen level

Tier 3 (Important but not immediately critical):

Toxic alcohol levels (send but don't wait for results)
Liver function tests
Phosphorus, magnesium

Interpreting the Numbers

Anion Gap Magnitude as Diagnostic Clue:

>25 mEq/L: Think methanol, ethylene glycol, severe DKA, severe lactic acidosis
15-25 mEq/L: Most causes possible, use clinical context
12-15 mEq/L: Mild acidosis, consider early uremia, mild starvation ketosis¹⁴

Clinical Pearl #6: A "normal" anion gap doesn't rule out HAGMA if the patient is hypoalbuminemic. Correct for albumin first.

Time-Sensitive Decision Making

The 3-30-300 Rule

3 Minutes: Life-threatening causes ruled out/addressed

Severe DKA with hemodynamic instability
Severe lactic acidosis with shock
Suspected methanol/ethylene glycol with visual symptoms

30 Minutes: Specific diagnosis established

Laboratory confirmation of suspected cause
Treatment initiated for reversible causes

300 Minutes (5 hours): Definitive management plan

Subspecialty consultation obtained
Advanced therapies initiated (dialysis, antidotes)

When to Empirically Treat Before Confirmation

Methanol/Ethylene Glycol:

High clinical suspicion + osmolar gap >10 mOsm/kg
Visual symptoms + HAGMA + alcohol ingestion history
Fomepizole dosing: 15 mg/kg loading dose, then 10 mg/kg q12h⁶

Severe DKA:

Glucose >250 mg/dL + large ketones + pH <7.10
Begin insulin and fluids immediately

Clinical Pearl #7: "When in doubt, give fomepizole." The risk-benefit ratio favors empirical treatment in suspected toxic alcohol poisoning.

Differential Diagnosis Deep Dive

Common Presentations and Diagnostic Clues

Cause	Key Clinical Clues	Laboratory Pattern	Bedside Test
DKA	Diabetes history, dehydration, fruity breath	Glucose >250, large ketones	Urine ketones positive
Lactic acidosis	Shock, hypoxemia, poor perfusion	Lactate >4 mmol/L	POC lactate elevated
Methanol	Visual symptoms, "drunk" without alcohol smell	High osmolar gap initially	Woods lamp may fluoresce
Ethylene glycol	Crystalluria, neurologic symptoms	Calcium oxalate crystals	Urine microscopy
Salicylates	Tinnitus, hyperthermia, mixed acid-base	Respiratory alkalosis → acidosis	Salicylate level
Uremia	CKD history, fluid overload	Creatinine >5 mg/dL	BUN:Cr <10:1

The Atypical Presentations

Euglycemic DKA

Setting: SGLT2 inhibitor use, pregnancy, starvation
Pearl: Check ketones even if glucose <250 mg/dL¹⁵

D-Lactic Acidosis

Setting: Short gut syndrome, gastric bypass
Clue: Neurologic symptoms out of proportion to measured L-lactate
Hack: Most hospital labs measure only L-lactate; D-lactate requires special assay¹⁶

Pyroglutamic Acidosis

Setting: Chronic acetaminophen use, malnutrition, female gender
Clue: HAGMA with normal lactate, ketones, and renal function
Pearl: Often overlooked; requires specialized testing¹⁷

Clinical Pearls and ICU Hacks

Pearl #8: The "Delta-Delta" Calculation

When evaluating mixed acid-base disorders:

Δ Anion Gap = Current AG - Normal AG (usually 12)
Δ HCO₃⁻ = Normal HCO₃⁻ (24) - Current HCO₃⁻
If Δ AG = Δ HCO₃⁻: Pure HAGMA
If Δ AG > Δ HCO₃⁻: Concurrent metabolic alkalosis
If Δ AG < Δ HCO₃⁻: Concurrent normal AG acidosis¹⁸

Pearl #9: The Osmolal Gap Timing

Early toxic alcohol ingestion: High osmolar gap, normal anion gap
Late toxic alcohol poisoning: Normal osmolar gap, high anion gap
Timing matters: Don't dismiss toxic alcohols based on normal osmolar gap alone

Pearl #10: The Lactate Paradox

Not all elevated lactate indicates tissue hypoxia:

Type B lactic acidosis causes: Metformin, nucleoside reverse transcriptase inhibitors, propofol, epinephrine, thiamine deficiency
Clinical hack: Check lactate:pyruvate ratio when Type B suspected

ICU Hack #1: The "Three-Tube Rule"

Always send three tubes simultaneously:

Green top (heparin): Immediate blood gas
Red top (serum): Comprehensive metabolic panel
Purple top (EDTA): Hold for specialized tests if needed

ICU Hack #2: The "Smell Test"

Breath odors can provide immediate diagnostic clues:

Fruity/sweet: Ketones (DKA, starvation, alcoholic ketoacidosis)
Garlic: Organophosphates
Bitter almonds: Cyanide (though few people can detect this)
No alcohol smell despite "intoxication": Think methanol/ethylene glycol

Advanced Diagnostic Considerations

When Standard Workup is Negative

If initial GOLD MARK evaluation is unrevealing:

Consider Rare Causes:

5-Oxoprolinuria (Pyroglutamic acidosis): Chronic acetaminophen, malnutrition¹⁷
D-Lactic acidosis: Short gut syndrome, gastric bypass¹⁶
Propofol infusion syndrome: High-dose, prolonged propofol use¹⁹
Inborn errors of metabolism: Rare in adults but consider in young patients

Advanced Testing:

Urine organic acids
Plasma amino acids
Specialized toxicology panels

The "Negative Workup" HAGMA

When all standard tests are negative but HAGMA persists:

Recheck albumin-corrected anion gap
Consider lab error (repeat on new sample)
Review all medications and IV solutions
Consider rare toxins or inborn errors
Subspecialty consultation (nephrology, toxicology)

Treatment Priorities and Interventions

Immediate Life-Saving Interventions

Severe Acidosis (pH <7.10):

Bicarbonate therapy: Controversial but consider if pH <7.10 with hemodynamic instability²⁰
Dosing: 1-2 mEq/kg IV push, reassess in 15-30 minutes
Goal: pH >7.20, not normalization

Toxic Alcohol Poisoning:

Fomepizole: Loading dose 15 mg/kg IV
Ethanol alternative: If fomepizole unavailable, 10% ethanol solution
Dialysis indications: Methanol >20 mg/dL, ethylene glycol >20 mg/dL, or severe acidosis⁶

DKA Management:

Fluid resuscitation: 15-20 mL/kg normal saline bolus
Insulin: 0.1 units/kg/hour IV (after initial fluid resuscitation)
Glucose monitoring: When glucose <250 mg/dL, add dextrose to maintain 150-250 mg/dL²¹

Clinical Pearl #11: The "Two-Bag System"

For DKA management: prepare two IV bags—one with saline/insulin, another with D5W/insulin. Switch between bags based on glucose levels rather than stopping insulin.

Monitoring and Reassessment

The "Q2H Rule"

For severe HAGMA (pH <7.20 or anion gap >25):

Repeat blood gas every 2 hours initially
Point-of-care glucose and lactate every hour
Comprehensive metabolic panel every 4-6 hours

Response to Treatment Markers

Improving HAGMA:

Anion gap decreasing by 3-5 mEq/L every 2-4 hours
pH improving by 0.05-0.10 every 2 hours
Mental status improvement

Pearl #12: In DKA, don't rely solely on glucose normalization—follow ketone clearance and anion gap closure.

Special Populations and Considerations

Pregnancy

Normal physiology: Mild respiratory alkalosis (HCO₃⁻ 18-21 mEq/L)
DKA in pregnancy: Lower glucose thresholds, faster progression
Unique risk: Euglycemic DKA more common²²

Pediatric Considerations

Normal anion gap: Age-dependent (8-16 mEq/L in children)
Inborn errors: Higher suspicion in children with recurrent episodes
Dosing differences: Weight-based calculations essential

Elderly Patients

Diagnostic challenges: Atypical presentations, multiple comorbidities
Medication interactions: Higher risk of drug-induced acidosis
Renal function: Age-related decline affects clearance

Quality Improvement and Error Prevention

Common Diagnostic Errors

Error #1: Failing to correct anion gap for hypoalbuminemia

Prevention: Always check albumin level, apply correction factor

Error #2: Missing toxic alcohols due to normal osmolar gap

Prevention: Remember timing—late presentations may have normal osmolar gap

Error #3: Attributing HAGMA to "uremia" without adequate renal impairment

Prevention: Uremic acidosis typically requires severe kidney dysfunction (GFR <15)

The "Double-Check" System

For every HAGMA case:

Verify anion gap calculation manually
Confirm acid-base status on repeat sample
Ensure appropriate urgency of interventions
Document thought process for handoff

Future Directions and Emerging Concepts

Point-of-Care Technology

Portable ketone meters: Rapid β-hydroxybutyrate measurement
Advanced blood gas analyzers: Simultaneous lactate, glucose, electrolytes
Artificial intelligence: Decision support systems for differential diagnosis

Biomarker Development

Novel organic acid markers: Improved detection of rare causes
Multiplexed toxicology panels: Rapid toxic alcohol screening
Metabolomics: Pattern recognition for unusual presentations²³

Conclusion

High anion gap metabolic acidosis represents a diagnostic emergency requiring systematic evaluation and time-sensitive intervention. The GOLD MARK mnemonic provides a structured approach that prioritizes common and immediately life-threatening causes while maintaining diagnostic thoroughness.

Key takeaways for critical care physicians:

Rapid bedside assessment using point-of-care testing guides initial management
The GOLD MARK mnemonic prioritizes causes by clinical urgency and frequency
Empirical treatment is warranted for suspected toxic alcohol poisoning
Albumin correction of anion gap prevents missed diagnoses
Serial monitoring guides treatment response and identifies complications

The middle-of-the-night HAGMA presentation need not be a diagnostic nightmare. With systematic application of these principles, clinical pearls, and practical hacks, critical care physicians can rapidly identify causes and initiate appropriate life-saving interventions.

References

Kraut JA, Madias NE. Serum anion gap: its uses and limitations in clinical medicine. Clin J Am Soc Nephrol. 2007;2(1):162-174.
Rastegar A. Use of the deltaAG/deltaHCO3- ratio in the diagnosis of mixed acid-base disorders. J Am Soc Nephrol. 2007;18(9):2429-2431.
Figge J, Jabor A, Kazda A, Fencl V. Anion gap and hypoalbuminemia. Crit Care Med. 1998;26(11):1807-1810.
Kitabchi AE, Umpierrez GE, Miles JM, Fisher JN. Hyperglycemic crises in adult patients with diabetes. Diabetes Care. 2009;32(7):1335-1343.
Sheikh-Ali M, Karon BS, Basu A, et al. Can serum beta-hydroxybutyrate be used to diagnose diabetic ketoacidosis? Diabetes Care. 2008;31(4):643-647.
Brent J, McMartin K, Phillips S, et al. Fomepizole for the treatment of ethylene glycol poisoning. N Engl J Med. 1999;340(11):832-838.
Arroliga AC, Shehab N, McCarthy K, Gonzales JP. Relationship of continuous infusion lorazepam to serum propylene glycol concentration in critically ill adults. Crit Care Med. 2004;32(8):1709-1714.
Lalau JD, Kajbaf F, Protti A, et al. Metformin-associated lactic acidosis (MALA): Moving towards a new paradigm. Diabetes Obes Metab. 2017;19(11):1502-1512.
Cohen RD, Woods HF. Clinical and Biochemical Aspects of Lactic Acidosis. Oxford: Blackwell Scientific Publications; 1976.
Abuelo JG. Normotensive ischemic acute renal failure. N Engl J Med. 2007;357(8):797-805.
Barceloux DG, Bond GR, Krenzelok EP, et al. American Academy of Clinical Toxicology practice guidelines on the treatment of methanol poisoning. J Toxicol Clin Toxicol. 2002;40(4):415-446.
Dargan PI, Wallace CI, Jones AL. An evidence based flowchart to guide the management of acute salicylate (aspirin) overdose. Emerg Med J. 2002;19(3):206-209.
Puchalski ML, Kline JA. Emergency department patients with diabetic ketoacidosis have decreased anticoagulant activity. Acad Emerg Med. 2005;12(12):1239-1245.
Gabow PA, Kaehny WD, Fennessey PV, et al. Diagnostic importance of an increased serum anion gap. N Engl J Med. 1980;303(15):854-858.
Peters AL, Buschur EO, Buse JB, et al. Euglycemic diabetic ketoacidosis: a potential complication of treatment with sodium-glucose cotransporter 2 inhibition. Diabetes Care. 2015;38(9):1687-1693.
Uribarri J, Oh MS, Carroll HJ. D-lactic acidosis. A review of clinical presentation, biochemical features, and pathophysiologic mechanisms. Medicine (Baltimore). 1998;77(2):73-82.
Dempsey GA, Lyall HJ, Corke CF, et al. Pyroglutamic acidosis: a cause of high anion gap metabolic acidosis. Crit Care Med. 2000;28(6):1803-1807.
Adrogue HJ, Madias NE. Secondary responses to altered acid-base status: the rules of engagement. J Am Soc Nephrol. 2010;21(6):920-923.
Kam PC, Cardone D. Propofol infusion syndrome. Anaesthesia. 2007;62(7):690-701.
Kraut JA, Kurtz I. Use of base in the treatment of severe acidemic states. Am J Kidney Dis. 2001;38(4):703-727.
Wolfsdorf JI, Glaser N, Agus M, et al. ISPAD Clinical Practice Consensus Guidelines 2018: Diabetic ketoacidosis and the hyperglycemic hyperosmolar state. Pediatr Diabetes. 2018;19 Suppl 27:155-177.
Sibai BM, Viteri OA. Diabetic ketoacidosis in pregnancy. Obstet Gynecol. 2014;123(1):167-178.
Johnson CH, Ivanisevic J, Siuzdak G. Metabolomics: beyond biomarkers and towards mechanisms. Nat Rev Mol Cell Biol. 2016;17(7):451-459.

Word Count: Approximately 2,800 words

Funding: No external funding sources

Conflicts of Interest: The authors declare no conflicts of interest.

Sunday, August 31, 2025