Registered  members with 100+ posts do not see Ads

Clinical Arterial blood gas interpretation

Matt

Emeritus Staff
Emeritus Staff
This topic, like so many in medicine, can be as simple or as complicated as you'd like to make it. For this reason it's best to just stick to the bare bones and be aware of the more complicated bits. Any exam question you get are likely to be answerable with the bare bones, however if you talk to a renal, respiratory or ICU physician on the topic you're always going to be confused.

So, some basic principles:
  1. There are four kinds of acid-base disorder: respiratory alkalosis, metabolic alkalosis, respiratory acidosis and metabolic acidosis
  2. Acidaemia and alkalaemia refer to the pH of the blood, acidosis and alkalosis refer to the pathology that causes acid or alkali, respectively, to accumulate in the blood
  3. All of the acid-base disorders are associated with a degree of compensation unless there is more than one disorder and in theory full compensation is impossible
  4. The finer details of acid-base chemistry and renal/respiratory physiology are not covered here
  5. Finally, the only reason to perform an ABG is to determine the PaO2, everything else can be adequately determined with a venous blood gas

Suggested method for ABG interpretation


  1. Remind yourself of the clinical picture
  2. Determine if there is an acidaemia or an alkalaemia
  3. Determine if the disturbance is respiratory or metabolic, by looking at the pCO2 and HCO3-
    • If there is a metabolic acidosis, calculate the anion-gap and correct for albumin
  4. Determine if the compensation is appropriate
  5. Take note of the PaO2 and consider calculating the alveolar-arterial gradient if relevant

For someone with a little bit of practise the method above is fairly straight-forward. The area of confusion is usually due to compensation and mixed disorders. The way to differentiate the two is to be able to calculate the expected level of compensation based on the degree of acidosis/alkalosis as well as the time when considering respiratory compensation.

First, metabolic compensation, this takes between 3-5 days to reach full compensation.

The normal range for pCO2 is 34-45mm Hg.
  • In the acute setting, for every 10mm Hg increase above 40 you should expect the HCO3- to rise by 1mmol/L.
  • In the chronic setting, for every 10mm Hg increase above 40 you should expect the HCO3- to rise by 3.5mmol/L
  • In the acute setting, for every 10mm Hg decrease below 40 you should expect the HCO3- to fall by 2mmol/L
  • In the chronic setting, for every 10mm Hg decrease below 40, you should expect the HCO3- to fall by 5mmol/L

For, respiratory compensation, it's a bit tricker. Compensation begins within one hour but full compensation takes 12-24 hours.
  • For a metabolic acidosis, expected pCO2 = 1.5x {actual HCO3-} + 8
  • For a metabolic alkalosis the calculation is less reliable but is basically:
    • Expected pCO2 = 0.7x{actual HCO3-} + 20

Tips for identifying a mixed picture

  • The compensation is always in the same direction as the initial chemical change, so if you have an increased CO2 (respiratory acidosis) the compensation is always an increased HCO3-. Where this is not the case, i.e. both the CO2 and the HCO3- are abnormal but in opposite directions there is likely a mixed picture
  • Compensation should never be complete unless you have a mixed disorder. If the pH is normal but either the pCO2 or the HCO3- is abnormal you should suspect a mixed picture
  • The other situation to suspect a mixed picture is if the compensation is inadequate or too extreme based on the above calculations

So, for example, as a basic rule of thumb
  • where pCO2 is elevated and HCO3- is reduced - there is a mixed respiratory and metabolic acidosis
  • where pCO2 is reduced and HCO3- elevated - there is a mixed respiratory and metabolic alkalosis
 

Sam

¿umm?
You should release a textbook Matty
 

Syn

Regular Member
An easy way I reckon to figure out if it's respiratory or metabolic is:

If it's respiratory, the pH and PaCO2 are in opposite directions, e.g. if pH is low, then PaCO2 will be high, hence respiratory acidosis, and if pH is high, then PaCO2 will be low, respiratory alkalosis

If it's metabolic, they'll go in the same direction, if pH is low, then HCO3 will also be low, meaning metabolic acidosis


Then you factor in compensation, which is fairly easy. Look for the thing going in the opposite direction to pH (i.e. since it's going in the opposite direction, it can't be causing the pH changes, so therefore must be due to compensation, e.g. low pH but high HCO3-, for respiratory acidosis)
 

Matt

Emeritus Staff
Emeritus Staff
Calculating the anion gap

Metabolic acidosis is usefully divided into anion gap and non-anion gap acidosis. The usefulness of this calculation is that it narrows the differential diagnosis. In the body the total number of anionic charges is equal to the total number of cationic charges and all of the important cations and anions in the blood have a valency of 1 - that is they only have one positive or one negative charge.


The important anions and cations are Na+, Cl-, HCO3- and, depending on your institution, K+. The anion gap represents the negative charges that balance out the positive charges and, in the blood, are made up of the negatively charged side-chains of proteins.


Thus the anion gap is usefully calculated to be [Na+] - ( [Cl-] + [HCO3-] ) in mmol/L. The normal value for this calculation is 8-16mmol/L. Sometimes K+ is also included and then the normal value is 12-20mmol/L. You can use whichever formula you like so long as you reference the appropriate normal range. Where the anion gap is increased (i.e. it is greater than 16mmol/L) you have a high anion gap metabolic acidosis.


The most common pitfall in calculating the anion gap is not correcting for albumin. Albumin is a protein and contributes significantly to the total anionic charge in the body. Therefore, when the albumin is reduced, as in acute/chronic inflammation or malnutrition, the anion gap needs to correct for this. For every 10g/L reduction in albumin the anion gap should also be reduced by 2.5mmol/L.


The causes of a high-anion gap metabolic acidosis are usefully summarised by the mnemonic MUD PILES which is reviewed in the mnemonics thread here.


The causes of a normal-anion gap metabolic acidosis are usefully divided into two types:
1. Loss of Bicarbonate



  • Gastrointestinal
    • Diarrhoea
    • Fistula (pancreatic, intestinal, biliary)
    • Ileostomy
  • Renal
    • Type 2 Renal Tubular Acidosis (RTA) - proximal
  • Carbonic anydrase inhibitors

2. Impaired renal excretion
  • Type 1 RTA - distal
  • Type 4 RTA - hypoaldosteronism
 

Matt

Emeritus Staff
Emeritus Staff
Calculating the A-a gradient

The A-a gradient is a measure of how efficiently oxygen is passing from the alveoli to the pulmonary capillaries and further arteries. The air in the arteries is supplied completely from the air in the alveoli so the alveolar oxygen will always be higher than the arterial oxygen. The A-a gradient is simply the difference between the two (Alveolar O2 - arterial O2), with a normal range of 15-30mm Hg.

There are two things that decide the arterial O2 and they are the inspired O2 and the capacity of the lungs to exchange gas. A gradient >30mm Hg implies that gas exchange is impaired, assuming the concentration of oxygen is held constant. At sea level the partial pressure of oxygen is ~159mm Hg but this drops to ~150 in the upper airways (because of a higher concentration of water vapour).

In the alveoli O2 is taken up and replaced with CO2, so in order to calculate the alveolar PO2 you have to account for this O2 being consumed. You can estimate how much O2 is being taken up because it is related to the amount of CO2 which is replacing it according to a specific ratio called the respiratory quotient. This respiratory quotient is 0.8 and it basically means that if you multiply the amount of O2 being taken up you get the amount of CO2 that is replacing it.

So Oxygen consumed x 0.8 = CO2 produced.

Therefore CO2 produced / 0.8 = Oxygen consumed
OR
CO2 x 1.2 = Oxygen consumed (since 1/0.8 = 1.2)

Since the CO2 in the alveoli is approximately equal to the CO2 in the blood you can multiply the arterial CO2 by 1.2 to get the oxygen consumed. The alveolar O2 is therefore equal to the inspired O2 - oxygen consumed.

Or Alveolar O2 = 150mm Hg - (PaCO2 x 1.2)

The A-a gradient is the alveolar O2 - arterial O2

And so it is equal to (150mm Hg - (PaCO2 x 1.2) - PaO2).

As I said above the normal range is 15-30mm Hg but this is actually an oversimplification. In a young healthy population a gradient above 10mm Hg would be abnormal, in a middle-aged person a gradient above 15mm Hg would be abnormal, and in an elderly person a gradient above 30mm Hg would be abnormal.

The best way to think about the A-a gradient is the difference between the amount of oxygen that can enter the blood and the amount of oxygen that does enter the blood. If the process of gas exchange between the alveoli and the blood was perfect the A-a gradient would theoretically be zero. In practice it increases with age as per above. The causes of respiratory acidosis with an increased A-a gradient can be considered in terms of those processes that contribute to gas exchange (ventilation, diffusion, and perfusion)

  1. Ventilation
    • Pneumonia
    • Pulmonary oedema
    • Atelectasis
  2. Diffusion
    • Emphysema
    • Interstitial lung disease
  3. Perfusion
    • Pulmonary embolism
    • Anatomic right-to-left shunt
    • COPD/Asthma

Respiration itself occurs in four steps and gas-exchange is the final step. A respiratory acidosis with normal A-a gradient can occur when any of the preceding steps are abnormal. The preceding steps are (1) sensing and signalling - i.e. the response of the medulla to CO2, (2) muscles and motion - that is the ability of the chest to inflate, and (3) a patent airway for air to flow freely through.

Disorders include

  1. Sensing
    • Brain stem injury
    • Drugs (e.g. narcotics, benzodiazepines, alcohol)
    • Oxygen in patients who are CO2 retainers (e.g. COPD)
  2. Signalling
    • Diaphragmatic paralysis
    • Guillain-Barre syndrome
    • Spinal cord injury
    • Polio
    • Lower motoneurone disease
  3. Muscles and motion
    • Flail chest (e.g. rib fractures)
    • Kyphoscoliosis (deformity of spine that restricts breathing)
    • Botulinism (blocks the NMJ inhibiting diaphragm contraction)
    • Myasthenia gravis
    • COPD (lung hyperexpansion)
  4. Free flow
    • Asthma and COPD
    • Croup
    • Epiglottitis
    • Seizures
    • Aspiration

Note that COPD effects all stags of respiration and and the diffusion and perfusion processes of gas exchange.
 
Last edited:

Shizzy

Staff
Administrator
Just to add to that, you can only use the value of 150 if the ABG was taken with the patient breathing room air, which is quite often not the case.

To calculate the true PA[SUB]O2[/SUB] you use the formula [FiO​2 x (P[SUB]atm[/SUB] - PH[SUB]2O[/SUB])] - (PaCO2 x 1.2)

FiO2 = fraction of inspired oxygen (0.21 room air at sea level)
P[SUB][SUP]atm[/SUP][/SUB] = atmospheric pressure (760 at sea level)
P[SUB]H2O[/SUB] = partial pressure of H2O ~47

If you plug those values in you get the 150 - (PaCO2 x 1.2).

However, if your patient is intubated and receiving 100% FiO2 then the PAO2 is 713 - (PaCO2 x 1.2) (makes a big difference when calculating Aa grad). It becomes difficult/messy if you try to estimate how much O2 your patient is receiving via a Hudson mask for instance...
 
Last edited:

Matt

Emeritus Staff
Emeritus Staff
Here's an alternative method presented in this week's issue of the BMJ (BMJ 2013;346:f16).

 Guide to systematic approach to analysis of a report of arterial blood gases

Step 1Assess oxygenationRecord the inspired oxygen concentration. Calculate the P/F ratio,* particularly if patient is receiving supplemental oxygen. Assess haemoglobin saturations, if testing is available
Step 2Assess pHIs the patient acidaemic or alkalaemic?
Step 3Assess sHCO[SUB]3[/SUB]− and base excessAn abnormal base excess and sHCO[SUB]3[/SUB]− indicates a primary or compensatory metabolic acid-base disturbance
Step 4Assess PaCO[SUB]2[/SUB]Is there a primary respiratory acidosis or alkalosis? Is low or high PaCO[SUB]2[/SUB] compensating for a metabolic acidosis or alkalosis respectively? The respiratory system will not normally overcorrect a metabolic acid-base disturbance, and so if this is the case, consider a mixed metabolic and respiratory disorder
Step 5Review additional analytesReview electrolytes, and consider calculation of anion gap to further assess any metabolic acidosis. Haemoglobin, glucose, and lactate concentrations may be available and may be helpful in determining the cause of any acid-base abnormality
Step 6ReassessAfter institution of a management plan, repeat clinical assessment and consider repeat analysis of arterial blood gases to guide further treatment
I think it's very good, nice and concise. Although, for revision and clinical practice purposes it'd be good to read the accompanying article to get your head around the process.

This is also a great table that might be worth keeping from the same article.
acidbase table.jpg
 

Registered  members with 100+ posts do not see Ads

Top