"That can't possibly be human arterial blood"

This blood gas gets waved around a lot, particularly at unexpecting medical students and other impressionable individuals.

As an isolated result, this one is an abomination. Clinical scenarios with varying levels of improbability are offered to the interpreters. "This is an ABG collected during a cardiac arrest"; "This is a sample of peripheral venous blood from an unwell demented patient"; "This sample suffered prolonged transport time on the way to the laboratory". My favourite has been "This patient presented to emergency complaining of generalised unwellness, dry skin, hoarse speech, ataxia and compulsive brain-eating behaviour".

Data to be interpreteted

Data to be interpreteted



Discussion and Interpretation

In actual fact, this is a sample collected from a cold (4° C) bag of packed red blood cells (AB+, in case you're interested). It was collected moments after their arrival from the blood bank, plastic bag still bearing the dew of their refrigeration. The cells were supplied for a non-urgent transfusion in a reasonably well patient, and though the author has no way to support this view, one can suspect that these cells were taken from the "back of the shelf". In other words, the oldest cells were probably supplied - those nearest to their expiry date, pickled in their own wastes for a month and suffering extensive storage lesions.

Assessment of tension-based and content-based oxygenation indices

Any assessment of "oxygenation" in this case would be a farce, as it has been some weeks since these red cells had even seen a lung. The low pO2 demonstrates that diffusion through the bag wall is rather poor, even accounting for the fact that at 4°C the solubility of oxygen in blood is significantly increased.

The change in pH

There is a horrific acidaemia; the pH is 6.587.

The change in pCO2

The pCO2 is high, likely contributing to the acidosis.

The change in Base Excess

The Actual Base Excess is not reported on this blood gas.

Assessment of compensation

Copenhagen interpretation of acid-base compensation:

In the absence of a base excess value, it is impossible to use the Copenhagen rules.

We don't even get a standard bicarbonate value!

In any case, the standard base excess would not be particularly accurate in this scenario, as it uses a diluted haemoglobin value (50g/L) to account of buffering of the whole extracellular fluid. And these cells have been confined in a small bag with a haematocrit of 64.7% - not much ECF to speak of. Plus, though there is plenty of haemoglobin to buffer with, its buffering capacity has been greatly altered by the low 2,3-DPG levels, acidosis and hypoxia, making it much more difficult to predict the buffering capacity of this blood sample from standardised formulae. As if it were consciously ashamed of this fact, the blood gas analyser has sheepishly concealed its base excess estimate.

Boston interpretation of acid-base compensation:

Having to calculate the actual bicarbonate, we can arrive at a figure of 11.1 mmol/L.

If the blood in the bag were subject to normal Boston bedside rules, we might be able to say that the expected bicarbonate for this scenario 32.1 mmol/L (for an acute respiratory acidosis).

However, it has been stored for weeks ... does that make this a "chronic" respiratory acidosis?

Certainly sounds chronic. The "4 for 10" rule suggests that if the pCO2 is chronically 121mmHg, the bicarbonate should be 56.4mmol/L.

The Boston rules break down in this scenario. The in vivo human CO2 titration experiments on which they are based were perfromed in a living organism. This, however, is an inert sack of concentrated blood, and it does not have kidneys to compensate with. If we had to guess, we would be forced to conclude that the closest bedside rules are those that apply to acute respiratory acidosis, making the "expected" bicarbonate around 32.1 mmol/L.

So, in summary - there is probably a "respiratory-like" acidosis here, coexisting with a metabolic acidosis, but it is hard to use those terms with a straight face in a "patient" with neither lungs nor kidneys.

Assessment of the metabolic component of acidosis

The anion gap is 14.7, if we use the calculated actual bicarbonate value.

The albumin level is unknown, but expected to be near normal, from what little can be said of PRBC albumin content. With this value, the "normal" anion gap should be 12.

The delta ratio is therefore 0.21, suggesting that much of the metabolic acid-base disturbance can be attributed to the hyperchloraemia. Sure, the lactate is dramatically elevated, but so is the potassium - and so the anion-cation balance is preserved.

This assertion rests on the assumption that the normal albumin level is preserved during centrifuge separation, and that the actual bicarbonate value we have calculated is accurate.

Assessment of oxygen-hemoglobin dissociation mechanics

There is an abnormally left-shifted p50(st) and an abnormally right-shifted p50. The right shift is easily explained by the influences of high pCO2 and low pH. The significant left-shift is due to 2,3-DPG depletion, which is a known consequence of erythrocyte storage.

These values are of course all corrected to a normal body temperature: cold PRBCs would actually have a massive left-shift, as hypothermia increases the affinity of haemoglobin for oxygen (by a massive 2200% at 0°C)

Additional remarks

This ridiculous ABG illustrates the limitations of the "classical" methods of assessing acid-base compensation. Both the Boston and the Copenhagen models reveal their deficiencies.