Skip to main content

A few people had a go. Well done!

This is a tough one. This is my roundabout way of working it out. Probably not as slick as it should be but….it does the job.  It IS a triple acid base abnormality and the diagnosis is probably salicylate toxicity.

Sure the patient has a fever and is vomiting and may be septic, but lets look at the solution.


pH is 7.8- so it is an alkalosis


Lets look for compensation

If I use the numbers we use in the Fellowship Course:

Met Acid                pCO2 = 1.5 x Bic + 8  or  pCO2 = last 2 digits of pH (if pH>7.1)

Met Alk                  pCO2 = 1    x Bic + 8  or  pCO2 = last 2 digits of pH (if pH<7.6)

Resp Aci                ∆10mm pCO2  =  ∆1mmol Bic acutely (or 0.05 pH)

                                                                  ∆4mmol Bic chronically

Resp Alk                 ∆10mm pCO2  =  ∆2mmol Bic acutely (or 0.05 pH)

                                                                    ∆5mmol Bic chronically

Given that the pCO2 is decreased by a min of 25mmHg, we would expect in a pure respiratory alkalosis, HCO3 to change by 4-5mmol acutely or 10 mmol chronically, ie go to 20 or lower. As the HCO3 is higher, it indicates a METABOLIC ALKALOSIS

We can confirm this by calculating as below, but don’t really need to:

∆Bic = 25-Bic                      (25 = Normal Bic)

∆AG = AG-10                       (2-10 = Normal AG)

∆Bic >> ∆AG   then  Mixed High and Normal Anion gap Acid

if  ∆Bic << ∆AG     then  Additional Metabolic Alkylosis

In this case :

∆Bic = 25-Bic = 25-23 =2     ∆AG = AG-10 = 21   as  ∆Bic << ∆AG there is also a METABOLIC ALKALOSIS

If we measure the ANION GAP, it is: (119) – (65+23) = 31, so there is a raised anion gap.

METABOLIC ACIDOSIS is really the only thing that can give such a raised AG

Anion Gap: (119) – (65+23) = 31

The A-a gradient is:

= (FiO2(Patm-PH2O)- PaCO2/0.8) -PaO2

= (0.4(760-47)-15/0.8)-192

= 74.4- so a much increased A-a gradient

The calculated osmolarity is 2(Na) +U + Glc = 238+10.3+4.5 = 252.8mmol/L

We cannot calculate the GAP but a usual osmolarity is 270-290 mmol/L. There it is not a raised aosmolarity that we would ascribe to alcohols.

There is also hyponatraemia, hypokalaemia and hypocloraemia and there is renal renal impairment.


Causes of high anion gap metabolic acidosis are:

M- methanol  X– as low osmolarity

U-Uraemia X– Urea is not that high

D-DKA – X– Not likely as BSL is not high

P – Paraldehyde – X -Unlikely

I- Iron, isoniazid – X -Could be but no real reason

L -Lactate – Could be because of the vomiting

E – Ethylene Glycol X– unlikely as has low osmolarity.

S -Salicylate OD – could be- remember the patient is giving no history, which is strange.

So from this group we have Lactate and Salicylates

What are some of the causes of Respiratory Alkalosis?

The 2 big ones are infection and salicylates

The metabolic alkalosis is probably secondary to the vomiting.

Certainly sepsis and dehydration must be considered. Can salicylates cause this?

Lets look at the phases of salicylate overdose:

Phase 1 (12 hours)

There is hyperventilation from respiratory center stimulation. This leads to respiratory alkalosisPotassium and sodium bicarbonate are excreted in the urine. This phase may last as long as 12 hours.

Phase 2 (12-24 hours)

There is paradoxic aciduria.

Phase 3

The patient can become dehydrated and hypokalaemic. There is a metabolic acidosis.

Nausea and vomiting can be the earliest signs of salicylate toxicity. Hyperthermia may indicate severe toxicity.

So in conclusion, certainly we need to exclude sepsis and other causes, however on this ABG and electrolytes, salicylate toxicity seems the most likely cause!


  • mohassan says:

    your calculation of anion gap was wrong -formula -2XNa -Cl- HCO3
    I didn’t understand your osmolar gap calculation – as I understand it is difference between lab measured osmolality and calculated osmolality . and also if you calculate delta gap it was high which indicate metabolic alkalosis as well .

    • Resus says:

      Thanks for the comment, lets see.
      My calculation for anion gap was correct DOOHH! NO IT ISN’T YOU ARE CORRECT (I have substituted Glc for HCO3- It is still a raised AG, but the wrong numbers) My apologies and thanks for the eagle eye and to Domini as well!i.e. ANION GAP, it is: (119 2.5) – (65 23) = 33.
      The Osmolarity calculation is straightforward. There are several ways to calculate it, I use this one. It is true that the osmolar gap is the difference between measured and calculated. The calculation of osmolarity assists, in that we know that the range is about 275-295 mosm/kg. If the gap is >10 we can use this, however, this calculation can assist us.

      The delta gap is a good one to use, I haven’t used it in this case, but you should if you understand it and it helps.

  • BDBiles says:

    Nice answer.

    Just a clarification.
    Up in the paragraph that starts ‘let’s look for compensation….’ you say, ‘Given the BE is high and the HCO3 is high, there is also a METABOLIC ALKALOSIS.’. The base excess is obviously high at 10, the bicarbonate is low normal at 23.
    Did you mean te bicarbonate is high relative to te expected/predicted bicarbonate of 19 (if there were no third disturbance)?
    Or perhaps I’ve missed something….
    Thanks, Ben.

    • Resus says:

      Exactly. The bicarbonate is supposed to be say 19, but is higher and even with a low normal bicard of 23, there is a big BE.

  • lt1978 says:

    Looking forward to the Kamikaze course soon. Could I please have a clarification as I found part of the answer slightly confusing, and slightly different to how I have been taught to approach these?

    1. I agree with pH 7.8 this is clearly an alkalemia (I have been advised to write this rather than alkalosis, but maybe this is just semantics!)

    2. I agree that obviously there is a respiratory component as the PCO2 of 15 is low.

    The next bit is where I get a bit confused…

    3. You say “The HCO3 is 23 and thus at the lower end of normal, so there is some compensation. Therefore there is a METABOLIC ACIDOSIS also( The high anion gap, shown below, seals that deal for me)”. THEN you mention the metabolic alkalosis.

    But at this stage, I would look to assess adequacy of compensation and see that the HCO3 at 23 (although at the lower end of normal) is actually higher than we would expect if compensation was adequate – we would expect that it would be 19 (assuming this is an acute process). As you noted in the sentence “expect about a 2mmol/L decrease in HCO3 for every 10mmHg of CO2, ie about a 4mmol/L decrease, down to potentially 19”.

    So, this means that there is an additional metabolic alkalosis.

    I haven’t been advised to use the base excess as the primary means of determining the metabolic component of the disorder.

    Also: I can’t see how, other than by noting the high AG (which is my next step) how I can say that there is an additional metabolic acidosis.

    Please can you help?



    • Resus says:

      Hi Leanne
      Sorry to confuse you. I use the BE as a personal approach, to give me an indication that not all things are making sense. I agree that you should keep using the AG. Wait till the Kamikaze. It will all make sense, not only when we go over these, but when you see the videos in the virtual environment.

  • lt1978 says:

    Thanks for that.

Leave a Reply