Blood gas interpretation: quick reference for acid-base disorders

Basic principle

Interpret the blood gas in the same order every time. Skipping a step is a common reason why mixed acid-base disorders are missed. The blood gas describes the physiology at the moment of sampling, but does not by itself identify the cause. The result must always be weighed together with the clinical picture, medication, fluid balance, renal function, electrolytes and supplemental oxygen [1].

Distinguish between:

• Acidaemia and alkalaemia, which describe the current pH of the blood.
• Acidosis and alkalosis, which describe processes that lower and raise pH respectively.

A patient may therefore have both an acidosis and an alkalosis at the same time despite a normal pH. The lungs compensate for a primary metabolic disturbance within minutes by changing ventilation. Renal compensation for a primary respiratory disturbance, by contrast, develops over hours to days [1,2].

Six steps

Rapid classification

It is therefore not enough to ask which of the pCO₂ and the HCO₃ deviates most. What matters is which change can explain the pH, and whether the other variable lies within the expected range of compensation.

Sample type and baseline values

An arterial blood gas is the reference method when the question concerns oxygenation, precise ventilation or severe respiratory failure. A peripheral venous blood gas is easier to obtain and can often be used for an initial assessment of pH and the metabolic component, but it should not be regarded as fully interchangeable with an arterial sample [3].

Venous pH is on average slightly lower than arterial pH. Agreement is usually good for pH and reasonable for bicarbonate and base excess, but considerably poorer for pCO₂ and entirely inadequate for pO₂. The differences increase in severe circulatory failure and marked respiratory failure [3,4].

Common approximate arterial reference ranges in adults are:

• pH 7.35 to 7.45
• pCO₂ 4.7 to 6.0 kPa, corresponding to 35 to 45 mmHg
• HCO₃ about 22 to 29 mmol/L

Always use the local laboratory's reference ranges. Swedish blood gases are usually reported in kPa. 1 kPa corresponds to about 7.5 mmHg.

The bicarbonate reported by the blood gas analyser is usually calculated from the measured pH and pCO₂. Total carbon dioxide in the laboratory electrolyte panel is a separate measurement and consists mainly of bicarbonate. An unexpectedly large difference between these values may reflect a sampling error, storage, an analytical problem or rapidly changing physiology [1,5].

Primary disorders and expected compensation

Compensation is physiologically predictable but not exact. The formulas are therefore ranges, not absolute limits. A simple primary disorder should not produce more compensation than expected. If the compensation is too small or too great, a further primary disorder is likely [2].

Winter's formula in kPa

Calculate first in mmHg and then multiply by 0.133:

Expected pCO₂ in kPa = (1.5 × HCO₃ + 8) × 0.133

The ±2 mmHg range corresponds to approximately ±0.27 kPa.

Example: HCO₃ is 12 mmol/L.

• Expected pCO₂ = 1.5 × 12 + 8 = 26 mmHg
• 26 mmHg corresponds to about 3.5 kPa
• The expected range is about 24 to 28 mmHg, corresponding to 3.2 to 3.7 kPa

Interpretation:

• A measured pCO₂ above 3.7 kPa means a concurrent respiratory acidosis.
• A measured pCO₂ below 3.2 kPa means a concurrent respiratory alkalosis.
• A measured pCO₂ within the range is consistent with adequate respiratory compensation.

Acute or chronic respiratory disturbance

Chronic renal compensation usually takes several days. A high pCO₂ with only a slightly raised HCO₃ therefore suggests an acute respiratory acidosis. A high pCO₂ with a clearly raised HCO₃ and a pH closer to the normal range suggests chronic hypercapnia. In acute deterioration of a chronically hypercapnic patient one often sees acute-on-chronic respiratory acidosis [2].

Example:

• pH 7.22
• pCO₂ 10 kPa, about 75 mmHg
• HCO₃ 30 mmol/L

The pCO₂ is about 35 mmHg above 40 mmHg. In a pure acute respiratory acidosis the HCO₃ would be expected to be around 27 to 28 mmol/L. In a purely chronic disorder it would be expected to be around 37 to 38 mmol/L. A value of 30 mmol/L together with marked acidaemia suggests a predominantly acute component, possibly acute-on-chronic hypercapnia.

Does compensation normalise the pH?

Compensation moves the pH towards the normal range but does not create a new primary disorder. A completely normal pH with clearly abnormal pCO₂ and HCO₃ should therefore raise suspicion of two opposing processes, particularly if the values do not fit a chronically compensated simple disorder.

Anion gap

The anion gap estimates the concentration of anions not included in the routine electrolyte panel.

Anion gap = Na minus (Cl plus HCO₃)

Potassium is usually omitted. If a laboratory includes potassium the reference range is higher. Do not mix formulas and reference ranges.

With modern analytical methods the reference range without potassium is often around 8 to 12 mmol/L, but local laboratories may quote, for example, 8 to 16 mmol/L. Use the local reference range and preferably compare electrolytes drawn at the same time [1,7].

Correction for albumin

Albumin is the most important normal unmeasured anion. Hypoalbuminaemia therefore lowers the anion gap and may conceal an acidosis with accumulation of organic acids.

Corrected anion gap = measured anion gap + 2.5 × (40 minus albumin in g/L) / 10

A practical rule of thumb is to add about 2.5 mmol/L for every 10 g/L that the albumin lies below 40 g/L [7,8].

Example:

• Na 140 mmol/L
• Cl 108 mmol/L
• HCO₃ 18 mmol/L
• Albumin 20 g/L

The measured anion gap is 14 mmol/L. The albumin correction is 5 mmol/L. The corrected anion gap is therefore 19 mmol/L, which indicates a high anion gap acidosis despite an apparently moderate uncorrected gap.

A low anion gap is most often due to hypoalbuminaemia or analytical variation. Less common causes are paraproteinaemia, lithium, and interference from bromide or iodide [1].

Delta gap and delta ratio

In metabolic acidosis with a raised anion gap, ask whether the fall in HCO₃ corresponds to the rise in the anion gap.

Assuming a normal anion gap of 12 mmol/L and a normal HCO₃ of 24 mmol/L:

• Delta anion gap = measured corrected anion gap minus 12
• Delta HCO₃ = 24 minus measured HCO₃
• Delta ratio = delta anion gap / delta HCO₃

The limits are approximate and are influenced by the chosen normal value, albumin, renal function, fluid therapy and the stage of the illness. The delta ratio should therefore support, not replace, clinical judgement.

An alternative is to calculate a corrected bicarbonate:

Corrected HCO₃ = measured HCO₃ + (corrected anion gap minus normal anion gap)

• A corrected HCO₃ below about 22 mmol/L suggests a concurrent normal anion gap acidosis.
• A corrected HCO₃ above about 26 mmol/L suggests a concurrent metabolic alkalosis.

Metabolic acidosis with a raised anion gap

The most common causes are lactate accumulation, ketoacidosis and renal failure. If these do not explain the picture, poisonings and less common organic acids should be considered [1].

Lactate is not synonymous with hypoperfusion

Lactate should be regarded as a warning signal, not as a specific measure of perfusion. A raised lactate may be due to inadequate oxygen delivery, adrenergic stimulation, accelerated glycolysis, mitochondrial dysfunction or reduced hepatic clearance. Sepsis-related hyperlactataemia is often multifactorial [13].

Therefore assess at the same time:

• Blood pressure and pulse pressure
• Peripheral temperature and capillary refill
• Skin mottling
• Urine output
• Level of consciousness
• Oxygenation and haemoglobin
• Liver function
• Drugs and poisonings
• Clinical signs of focal ischaemia

Follow the trend, but do not treat the lactate value alone. A falling lactate may reflect either reduced production or increased metabolism. A normal lactate does not exclude severe shock or mesenteric ischaemia [13].

Metformin-associated lactic acidosis should be considered in a high anion gap acidosis in a patient taking metformin, particularly in acute renal failure, hypoxia, circulatory failure or sepsis. Metformin is often a contributing factor alongside other critical illness rather than the sole cause [14].

Ketoacidosis and euglycaemic ketoacidosis

Measure beta-hydroxybutyrate in blood rather than urine ketones alone. The urine dipstick mainly measures acetoacetate and may therefore underestimate early ketoacidosis, in which beta-hydroxybutyrate predominates.

Diabetic ketoacidosis can occur without marked hyperglycaemia. This applies particularly during treatment with SGLT2 inhibitors, in pregnancy, during fasting or in partially treated ketoacidosis. A normal or only moderately raised glucose therefore does not exclude the diagnosis [12].

During treatment of ketoacidosis the anion gap may normalise while the HCO₃ remains low because of a hyperchloraemic acidosis after large volumes of chloride-rich fluid. Assess the overall clinical improvement and follow blood ketones, not the anion gap alone [12].

Osmolal gap

The osmolal gap is the difference between the measured and the calculated serum osmolality.

When sodium, glucose and urea are reported in mmol/L:

Calculated osmolality = 2 × Na + glucose + urea

Osmolal gap = measured osmolality minus calculated osmolality

If ethanol is present in the blood, its osmotic contribution must be included according to the laboratory's formula. Use the same sampling time for the measured osmolality, electrolytes, glucose, urea and ethanol.

An osmolal gap above about 10 mOsm/kg may raise suspicion of an unmeasured osmole, but is not diagnostic of a toxic alcohol. Ketoacidosis, a raised lactate, renal failure, mannitol, propylene glycol and analytical variation can also raise the gap [15,16].

A normal osmolal gap does not exclude methanol or ethylene glycol. Early after ingestion there is a large amount of unmetabolised alcohol and the osmolal gap may be high while the anion gap is still normal. As the alcohol is metabolised to organic acids the osmolal gap falls while the anion gap rises. Late in the course a severely poisoned patient may therefore have a normal osmolal gap [16,17].

Where there is clinical suspicion, treatment must not be delayed while awaiting methanol or ethylene glycol assays. Fomepizole inhibits alcohol dehydrogenase and can prevent further formation of toxic metabolites. Haemodialysis may be needed in severe acidosis, renal failure, visual disturbance, circulatory failure or other serious organ involvement [17,18]. Contact your national or regional poison control centre and the on-call intensive care consultant immediately.

Salicylate poisoning

The typical blood gas shows a combination of:

• A primary respiratory alkalosis from stimulation of the respiratory centre.
• A metabolic acidosis with a raised anion gap from disturbed cellular energy metabolism and accumulation of organic acids.

The pH may therefore be high, normal or low. An apparently normal pH is not reassuring. Acidaemia increases the proportion of uncharged salicylic acid and thereby its entry into the central nervous system. A falling pH, a rising pCO₂, altered consciousness, pulmonary oedema or a falling minute ventilation are serious warning signs [19].

A single salicylate level may be misleading because of delayed absorption and variable kinetics. Follow serial concentrations and interpret them together with pH, symptoms, renal function and electrolytes. Intubation may be hazardous, since apnoea or inadequate minute ventilation rapidly raises the pCO₂ and lowers the pH [19].

Metabolic acidosis with a normal anion gap

Normal anion gap acidosis is often called hyperchloraemic acidosis. Bicarbonate has then been lost or not regenerated, and chloride has risen to preserve electroneutrality.

Common causes:

• Diarrhoea and other gastrointestinal bicarbonate loss
• Renal tubular acidosis
• Large volumes of isotonic sodium chloride
• Carbonic anhydrase inhibitors, for example acetazolamide
• Early or moderate renal failure
• Ureterosigmoidostomy or other urinary diversion into bowel
• Pancreatic or biliary fistula
• Recovery phase after ketoacidosis
• Hypoaldosteronism or type 4 renal tubular acidosis

Diarrhoea and renal tubular acidosis are the key differential diagnoses. In unexplained hyperchloraemic acidosis, review potassium, renal function, urine pH and medication [9].

The urine anion gap has traditionally been used as an indirect measure of ammonium excretion:

Urine anion gap = urine Na + urine K minus urine Cl

A negative value has been taken to indicate high ammonium excretion, for example in diarrhoea, while a positive value has been taken to indicate impaired renal acid excretion. The method has important limitations, however, and is influenced by diet, electrolyte intake, other urinary anions and whether the patient is in a steady state. Where direct urine ammonium or the urine osmolal gap is available, these may provide better support [10].

Metabolic alkalosis

Metabolic alkalosis arises through loss of acid or administration of alkali, but it persists only if renal excretion of bicarbonate is impaired. Important maintaining factors are volume contraction, chloride depletion, hypokalaemia, reduced glomerular filtration, mineralocorticoid effect and hypercapnia [11].

Urine chloride is often more useful than urine sodium, because bicarbonaturia can drag sodium with it.

Active diuretic treatment can give a urine chloride above 20 mmol/L even though the alkalosis is volume- and chloride-responsive. A low value once the diuretic effect has worn off does not exclude previous diuretic exposure. Interpret the urine chloride in relation to the timing of the last dose [11].

The blood pressure helps:

• Hypotension or orthostatism suggests volume contraction.
• Hypertension together with a hypokalaemic metabolic alkalosis suggests mineralocorticoid excess.
• A normal or low blood pressure with chronic hypokalaemia and a high urine chloride may suggest Bartter or Gitelman syndrome. Gitelman syndrome is typically characterised by hypomagnesaemia and low urinary calcium excretion [27].

In severe alkalaemia, particularly at a pH of 7.55 or above, there is an increased risk of arrhythmia, neurological symptoms, hypokalaemia, hypocalcaemia and impaired tissue oxygenation [11].

Respiratory acidosis

Respiratory acidosis is due to inadequate alveolar ventilation relative to the body's carbon dioxide production.

Common causes:

• COPD exacerbation
• Opioids, sedatives and anaesthetic agents
• Neuromuscular weakness
• Obesity hypoventilation
• Severe asthma with exhaustion
• Airway obstruction
• Chest wall injury or marked restrictive disease
• Incorrect ventilator settings
• Cardiac arrest or very low pulmonary perfusion
• Overfeeding in a ventilated patient

Clinically the change matters more than an isolated pCO₂. A chronically hypercapnic patient may tolerate a high stable value, whereas a rapid rise with a falling pH and declining alertness is acutely dangerous.

In COPD exacerbation, oxygen should be titrated, usually to an SpO₂ of 88 to 92 per cent when there is a risk of hypercapnic failure. Uncontrolled high-dose oxygen can worsen hypercapnia. The mechanisms mainly involve worsened ventilation–perfusion matching and the Haldane effect, not simply loss of a hypoxic respiratory drive [25].

Signs that ventilatory assessment is needed urgently:

• Declining level of consciousness
• Increasing work of breathing or exhaustion
• Rising pCO₂ with a falling pH
• Inability to protect the airway
• Hypoxaemia despite adequate oxygen
• Haemodynamic instability

A rapidly rising pCO₂ with declining alertness is an indication for ventilatory support, not a reason merely to increase the oxygen flow.

Respiratory alkalosis

Respiratory alkalosis is due to increased alveolar ventilation.

Common causes:

• Hypoxaemia
• Pulmonary embolism
• Pneumonia
• Sepsis
• Pain
• Anxiety or panic
• Pregnancy
• Liver failure
• Early salicylate poisoning
• Central nervous system disease
• Over-ventilation on a ventilator
• High altitude

Do not dismiss a respiratory alkalosis as anxiety before hypoxaemia, pulmonary embolism, sepsis and salicylate poisoning have been considered. A low pCO₂ in a severely ill patient may be a sign of vigorous compensatory ventilation. If the pCO₂ subsequently rises towards the normal range without any improvement in the metabolic acidosis, this may mean that the patient is tiring.

Mixed disorders

Suspect a mixed acid-base disorder when:

• The pCO₂ does not fit Winter's formula.
• The HCO₃ does not fit the expected acute or chronic respiratory compensation.
• The pH is normal despite markedly abnormal pCO₂ and HCO₃.
• The delta ratio does not fit an isolated high anion gap acidosis.
• The clinical picture contains several concurrent processes, for example sepsis, vomiting, renal failure and mechanical ventilation.

Common combinations:

Three simultaneous disorders are possible. A patient with septic shock, COPD and vomiting may, for example, have a lactate-related high anion gap acidosis, a respiratory acidosis and a metabolic alkalosis.

Assessment of oxygenation

Acid-base assessment and oxygenation are separate parts of blood gas interpretation. A normal pH says nothing about the adequacy of oxygenation.

PaO₂/FiO₂

PaO₂/FiO₂ = PaO₂ in mmHg divided by FiO₂ as a decimal

Example: a PaO₂ of 60 mmHg and an FiO₂ of 0.30 give a ratio of 200 mmHg.

The global definition of ARDS uses a PaO₂/FiO₂ of no more than 300 mmHg as an oxygenation criterion and also includes patients treated with high-flow nasal oxygen at at least 30 L/min. A diagnosis of ARDS additionally requires an acute onset, bilateral lung involvement, and that the findings are not mainly explained by cardiogenic pulmonary oedema, pleural fluid or atelectasis [20].

The PaO₂/FiO₂ ratio is influenced by PEEP, mode of ventilation and FiO₂. It should therefore not be interpreted in isolation.

A–a gradient

The alveolar oxygen tension can be estimated with the alveolar gas equation:

PAO₂ = FiO₂ × (atmospheric pressure minus water vapour pressure) minus PaCO₂ / respiratory quotient

The A–a gradient is then:

A–a gradient = PAO₂ minus PaO₂

On room air at sea level the following simplification is often used:

PAO₂ ≈ 20 kPa minus PaCO₂ / 0.8

A normal or only slightly raised A–a gradient in the presence of hypoxaemia suggests hypoventilation or a low inspired oxygen fraction. A raised gradient suggests a ventilation–perfusion disturbance, a diffusion defect or a shunt, for example pneumonia, pulmonary oedema, pulmonary embolism or ARDS [21].

The gradient increases with age and is influenced by FiO₂. It is therefore a physiological aid, not a stand-alone diagnostic test.

Carbon monoxide and methaemoglobin

Carbon monoxide

The pulse oximeter may show a normal or near-normal saturation in carbon monoxide poisoning, because conventional pulse oximetry does not reliably distinguish carboxyhaemoglobin from oxyhaemoglobin. The PaO₂ may also be normal, since it measures oxygen dissolved in plasma, not the effective oxygen transport of haemoglobin.

Request COHb by co-oximetry in:

• Exposure to smoke from fire
• Exposure to combustion gases
• Headache, nausea or confusion in several people in the same environment
• Syncope or altered consciousness without a clear explanation
• A raised lactate after a fire in an enclosed space

COHb falls once exposure ends, and faster still after oxygen. A low value on a delayed sample therefore does not exclude earlier clinically significant exposure. COHb also correlates poorly with the severity of symptoms [22].

Methaemoglobin

Methaemoglobinaemia should be considered in cyanosis that does not improve as expected with oxygen, particularly if:

• The SpO₂ tends to sit around 85 per cent.
• The PaO₂ is normal or high.
• There is a clear saturation gap.
• The blood is chocolate brown.
• The patient has been exposed to, for example, dapsone, benzocaine, prilocaine, nitrates, nitrites or nitric oxide.

The diagnosis is made with co-oximetry. The saturation calculated by a conventional blood gas analyser can be misleading [23,24].

Base excess

Base excess is a calculated measure of the metabolic component of an acid-base disorder after standardisation of the pCO₂. A negative value indicates a metabolic acid load or base loss, and a positive value a metabolic alkalosis or chronic renal compensation for hypercapnia.

Base excess is useful for following trends, but does not replace HCO₃, the anion gap or compensation calculations. A negative base excess does not identify the cause of the acidosis and does not distinguish a high anion gap acidosis from a hyperchloraemic acidosis.

Treatment is directed at the cause

Blood gas interpretation should lead to treatment of the underlying cause:

• Restore ventilation in respiratory acidosis.
• Treat hypoxaemia and the underlying lung disease.
• Give fluid, blood, vasoactive drugs or other circulatory support according to haemodynamic assessment.
• Treat sepsis and focal ischaemia.
• Give insulin and fluid in ketoacidosis according to the local care protocol.
• Stop the precipitating drug or toxin.
• Correct chloride and potassium in chloride-responsive metabolic alkalosis where this is clinically appropriate.
• Consider renal replacement therapy in severe renal failure or in a dialysable poisoning [26].

Sodium bicarbonate is not a general treatment for every metabolic acidosis. It can generate carbon dioxide, raise sodium and osmolality, lower ionised calcium and worsen intracellular or respiratory acidosis if ventilation is inadequate. The benefit depends on the cause, the severity of the acidosis, renal function and ventilatory capacity [8].

Specific indications may exist in, for example, salicylate poisoning, certain toxic alcohols, severe hyperkalaemia and selected severe metabolic acidosis with acute kidney injury. Follow local guidelines and involve intensive care, nephrology or your national or regional poison control centre early.

Red flags

The following findings require immediate clinical assessment and treatment alongside continued diagnostic work-up:

• A pH of about 7.20 or below, particularly with circulatory failure, arrhythmia or declining alertness
• A rapidly falling pH
• A rising pCO₂ in a tiring patient
• Marked hyperkalaemia
• A high anion gap acidosis without a clear explanation
• Suspected methanol, ethylene glycol or salicylate
• A raised lactate together with abdominal pain, shock or other suspicion of ischaemia
• A normalised pCO₂ in a patient who had previously been hyperventilating strongly because of a metabolic acidosis
• Cyanosis with a normal PaO₂
• Declining level of consciousness in a hypercapnic patient
• Marked metabolic alkalosis with arrhythmia or neurological symptoms
• Internally inconsistent or physiologically impossible blood gas values

Pitfalls

Applying the wrong rule at step 2

pCO₂ and pH change in opposite directions in primary respiratory disorders. HCO₃ and pH change in the same direction in primary metabolic disorders. The rule that the variable which has moved in the same direction as the pH is the primary one therefore works only for HCO₃, not for pCO₂.

Calling every low HCO₃ a metabolic acidosis

A low HCO₃ may be renal compensation for a chronic respiratory alkalosis. Check the pH, the pCO₂ and the expected compensation.

Forgetting the albumin correction

Hypoalbuminaemia can make the anion gap appear normal despite substantial accumulation of organic acids.

Treating a normal osmolal gap as exclusionary

A normal gap does not exclude a toxic alcohol, particularly late after ingestion.

Being satisfied with a normal pH

A normal pH can conceal two or three concurrent disorders. Always calculate compensation and the delta gap where relevant.

Using the venous pO₂

The venous pO₂ cannot be used to assess arterial oxygenation. Use pulse oximetry or an arterial blood gas depending on the clinical question [3,4].

Treating lactate as synonymous with shock

Lactate can rise with adrenergic stimulation, seizures, beta-2 agonists, hepatic impairment, metformin and other metabolic conditions. At the same time, severe shock can occur without a marked rise in lactate [13].

Disregarding specimen handling

Air bubbles drive the pO₂ towards the oxygen tension of air, and can lower the pCO₂ and raise the pH. A delayed analysis allows blood cells to continue consuming oxygen and glucose and to produce carbon dioxide and lactate. The sample should be taken anaerobically, mixed gently and analysed promptly [5,6].

Not documenting supplemental oxygen

The PaO₂ cannot be evaluated without knowing the FiO₂ or the oxygen delivery device. Also document the mode of ventilation, the PEEP and the time since the last change in ventilator settings or oxygen delivery.

Accepting a blood gas that does not fit the patient

Check:

1. Is the sample genuinely arterial or venous?
2. Is the patient identity correct?
3. Were there air bubbles?
4. Has the sample been diluted by flush solution from an arterial line?
5. How long was it until analysis?
6. Does the HCO₃ agree with the total carbon dioxide in the electrolyte panel?
7. Are the values consistent with the Henderson–Hasselbalch relationship?
8. Does the sample need to be repeated?

Practical on-call algorithm

1. Assess ABC and treat life-threatening problems first.
2. Note the sample type, FiO₂, ventilatory support and time.
3. Read the pH.
4. Identify the primary metabolic or respiratory process.
5. Calculate the expected compensation.
6. Calculate the anion gap and correct for albumin.
7. Calculate the delta gap when the anion gap is raised.
8. Check lactate, blood ketones, glucose, electrolytes, creatinine and relevant toxicology.
9. Calculate the osmolal gap in unexplained high anion gap acidosis or suspected poisoning.
10. Assess oxygenation separately.
11. Compare with the clinical picture and previous values.
12. Repeat the sample if the result is unexpected or physiologically implausible.

Sources

1. Hamilton et al. Understanding Acid-Base Disorders. Ulster Medical Journal 2017. PMID: 29581626
2. Tucker and Johnson. Breathing and balance: Clinical insights and management strategies of respiratory acid-base disorders. Nutrition in Clinical Practice 2025. PMID: 40483586
3. Byrne et al. Peripheral venous blood gas analysis for the diagnosis of respiratory failure, hypercarbia and metabolic disturbance in adults. Cochrane Database of Systematic Reviews 2025. PMID: 40557830
4. Giani et al. The Role of Venous Blood Gas Analysis in Critical Care: A Narrative Review. Medicina 2025. PMID: 40870384
5. Dukić et al. Blood gas testing and related measurements: National recommendations on behalf of the Croatian Society of Medical Biochemistry and Laboratory Medicine. Biochemia Medica 2016. PMID: 27812301
6. Baird. Preanalytical considerations in blood gas analysis. Biochemia Medica 2013. PMID: 23457763
7. Korus et al. Metabolic Acidosis in Patients with Chronic Kidney Disease: Diagnosis, Pathogenesis, and Treatment. Diagnostics 2025. PMID: 40870903
8. Eraky et al. Complexities, Benefits, Risks, and Clinical Implications of Sodium Bicarbonate Administration in Critically Ill Patients: A State-of-the-Art Review. Journal of Clinical Medicine 2024. PMID: 39768744
9. Berend. Review of the Diagnostic Evaluation of Normal Anion Gap Metabolic Acidosis. Kidney Diseases 2017. PMID: 29344509
10. Uribarri and Oh. The Urine Anion Gap: Common Misconceptions. Journal of the American Society of Nephrology 2021. PMID: 33769949
11. Do et al. Metabolic Alkalosis Pathogenesis, Diagnosis, and Treatment: Core Curriculum 2022. American Journal of Kidney Diseases 2022. PMID: 35525634
12. Umpierrez et al. Hyperglycemic Crises in Adults With Diabetes: A Consensus Report. Diabetes Care 2024. PMID: 39052901
13. Müller et al. Lactate: The Fallacy of Oversimplification. Biomedicines 2023. PMID: 38137413
14. See. Metformin-associated lactic acidosis: A mini review of pathophysiology, diagnosis and management in critically ill patients. World Journal of Diabetes 2024. PMID: 38983827
15. Ross et al. Toxic Alcohol Poisoning. Emergency Medicine Clinics of North America 2022. PMID: 35461626
16. Kraut. Diagnosis of toxic alcohols: limitations of present methods. Clinical Toxicology 2015. PMID: 26114345
17. Mégarbane. Treatment of patients with ethylene glycol or methanol poisoning: focus on fomepizole. Open Access Emergency Medicine 2010. PMID: 27147840
18. Alrashed et al. The Perils of Methanol Exposure: Insights into Toxicity and Clinical Management. Toxics 2024. PMID: 39771139
19. Sidlak et al. Acute Salicylate Toxicity: A Narrative Review for Emergency Clinicians. Cureus 2025. PMID: 41049912
20. Matthay et al. A New Global Definition of Acute Respiratory Distress Syndrome. American Journal of Respiratory and Critical Care Medicine 2024. PMID: 37487152
21. Toffaletti and Rackley. Monitoring Oxygen Status. Advances in Clinical Chemistry 2016. PMID: 27717415
22. Moss et al. Diagnosis of carbon monoxide exposure in clinical research and practice: A scoping review. PLOS ONE 2025. PMID: 39908298
23. Iolascon et al. Recommendations for diagnosis and treatment of methemoglobinemia. American Journal of Hematology 2021. PMID: 34467556
24. Cefalu et al. Methemoglobinemia in the Operating Room and Intensive Care Unit: Early Recognition, Pathophysiology, and Management. Advances in Therapy 2020. PMID: 32193811
25. Doğan et al. 2021 Guideline for the Management of COPD Exacerbations: EMAT/TTS Clinical Practice Guideline. Turkish Journal of Emergency Medicine 2021. PMID: 34849428
26. Mirrakhimov et al. The Role of Renal Replacement Therapy in the Management of Pharmacologic Poisonings. International Journal of Nephrology 2016. PMID: 28042482
27. Blanchard et al. Gitelman syndrome: consensus and guidance from a KDIGO Controversies Conference. Kidney International 2017. PMID: 28003083