Question 4.2

The following venous blood results are from a 52-year-old female who has had a prolonged ICU course following extensive surgery for resection of a pelvic sarcoma, complicated by sepsis and multi-organ dysfunction.

Venous Blood Gas


Patient Value

Normal Adult Ran e


7.32 - 7.43


42 mmH (5.5 kPa)

27 - 50 (3.5 -6.6)


44 mmH (5.8 kPa)

36 —44 (4.7 — 5.8)


11 mmol/L*

22 - 38

Base Excess

-18 mmol/L*

02 Saturation

70 - 80


140 mmol/L

135 - 145


3.8 mmol/L

3.5 - 5.2


119 mmol/L*

95 - 110

Anion Gap

14 mmol/L

7 - 17

Calcium Ionised

1.30 mmol/L

1 .12 - 1 .32


10.6 mmol/L*

3.0 — 5.4


1.0 mmot/L

< 1.5

Haemo lobin



115 - 160


9.3 mmol/L*

3.0 - 8.0


244 mol/L*

45 - 90

a) Describe the acid-base disturbance in the above results           (10% marks)

b) Give possible explanations. (20% marks)

[Click here to toggle visibility of the answers]

College Answer


  • Hyperchloraemic metabolic acidosis (normal/non-anion gap) with respiratory acidosis (no respiratory compensation acceptable)


  • Renal tubular acidosis (probably Type 1) secondary to nephrotoxicity (sepsis, drugs, obstruction)
  • GI losses (complicated abdomino-pelvic surgery)
  • Resuscitation with hyperchloraemic fluids
  • Respiratory depression secondary to opiates (or any reasonable explanation)



Let us dissect these results systematically.

  1. The A-a gradient cannot be estimated, as the gas is venous.
  2. There is acidaemia.
  3. The PaCO2 is not compensating for the acidaemia.
  4. The SBE is -18, suggesting a severe metabolic acidosis
  5. The respiratory compensation is inadequate- the expected PaCO2 (11 × 1.5) + 8 = 24.5mmHg, according to the Boston rules. Thus, there is also a respiratory acidosis.
    (Copenhagen rules can also be applied, and yield an expected PaCO2 of 22 mmHg)
  6. The anion gap is 14, according to the college.
    Assuming the albumin is normal, the AG is raised by 2.
    The delta ratio is therefore (2 / 13) = 0.15

Thus, this is a pure NAGMA and a respiratory acidosis.

A word of thanks is offered to one of the commentators for bringing attention to the fact that the college have used conventional arterial gas interpretation rules to interpret this venous sample. How valid is this practice? Let us reason though it.

Consider that mixing CO2 with blood will yield a predictable increase in HCO3- no matter where that blood is kept. Thus, the change in bicarbonate associated with acute hypercapnoea should be consistent on both sides of the circulation. However, the venous side of the circulation has a higher CO2. There may arise situations during which the venous CO2 is higher by a significant degree, giving rise to the impression that the respiratory compensation for metabolic acidosis is inadequate. Of course, that blood still has not yet passed through the pulmonary circulation, and so one cannot convincingly say that the respiratory system has failed at compensation (it hasn't even had a go yet). 

Thus, one might expect the venous gas to occasionally give a false impression of a mixed metabolic and respiratory acidosis (or, to fail to reveal a respiratory alkalosis). One can imagine that this might happen in states where the arterio-venous CO2 difference is somehow elevated. In fact, this has been demonstrated experimentally, in dissertations by Murphy (1982) and Berner (1983) who used exsanguinating dogs. The arteriovenous CO2 gap is thought to be a reasonably good marker of circulatory efficacy in shock states, good enough to guide therapy (Mallat et al, 2016). In contrast, it would appear that either there is probably little difference between arterial and venous samples during states with normal circulation, as demonstrated by Ilkiw et al (2008) whose dogs were not exsanguinating. 

So it it fair to say that the patient in this SAQ has a respiratory acidosis on the basis of a venous gas? Probably not, one should think. Sepsis and multi organ system failure do not sound like they describe a state of circulatory adequacy. However, the college examiners decided not to mention it. They seem to have just used the standard rules anyway. In their defense, it should be pointed out that in order for the corresponding arterial blood sample to have adequate compensation, the A-v CO2 difference would have to be about 18 mmHg, consistent with severe shock (it's meant to be under 6 mmHg).

In summary, the savvy exam candidate will be aware of this caveat, and be prepared to discuss it if questioned, but will probably be safe to ignore it in future venous blood gas questions.


The college only asked for possible explanations of the acid-base disturbance. The generic causes of NAGMA are given below. Any of them could potentially be applicable.


Ilkiw, Jan E., R. J. Rose, and I. C. A. Martin. "A Comparison of Simultaneously Collected Arterial, Mixed Venous, Jugular Venous and Cephalic Venous Blood Samples in the Assessment of Blood‐Gas and Acid‐Base Status in the Dog." Journal of veterinary internal medicine 5.5 (1991): 294-298.

SIGGAARD‐ANDERSEN, Ole, and Ivar H. Gøthgen. "Oxygen and acid‐base parameters of arterial and mixed venous blood, relevant versus redundant." Acta Anaesthesiologica Scandinavica 39.s107 (1995): 21-27.

Griffith, K. K., et al. "Mixed venous blood-gas composition in experimentally induced acid-base disturbances." Heart & lung: the journal of critical care 12.6 (1983): 581.

Berner, Barbara J. "The Use of mixed venous blood to assess acid-base status in states of decreased cardiac output when respiration is controlled." (1983).

Murphy, Janet A. "The use of mixed venous blood for assessment of acid-base status in states of decreased cardiac output." (1982).

Mallat, Jihad, et al. "Use of venous-to-arterial carbon dioxide tension difference to guide resuscitation therapy in septic shock." World journal of critical care medicine 5.1 (2016): 47.