Question 19.1

A 60-year-old male with alcoholic cirrhosis and atrial fibrillation on regular flucloxacillin and paracetamol is admitted to your ICU post-variceal banding. His coagulation screen is displayed below:

Parameter	Patient Value	Adult Normal Range
Prothrombin time (PT)	28.0 sec*	12.0 – 16.5
International normalised ratio (INR)	2.4*	0.9 – 1.3
Activated partial thromboplastin time (APTT)	21.0 sec	27.0 – 38.5
Fibrinogen	2.1 g/L	2.0 – 4.0
D-Dimer	0.5 mg/L	< 0.5

Give three likely causes for the coagulation abnormalities. (10% marks)

The arterial blood gas and biochemistry results from the same patient are given below:

Parameter	Patient Value	Adult Normal Range
FiO2	0.21
pH	7.23*	7.35 – 7.45
pO2	90 mmHg (12 kPa)
pCO2	22.0 mmHg (2.93 kPa)*	35.0 – 45.0 (4.60 – 6.00)
SpO2	92%
Bicarbonate	9.0 mmol/L*	22.0 – 26.0
Base Excess	-16.7 mmol/L*	-2.0 to +2.0
Lactate	1.3 mmol/L	0.5 – 1.6

Parameter	Patient Value	Adult Normal Range
Sodium	135 mmol/L	135 – 145
Potassium	4.0 mmol/L	3.5 – 5.0
Chloride	100 mmol/L	95 – 105
Bicarbonate	9.0 mmol/L*	22.0 – 26.0
Glucose	6.0 mmol/L	3.5 – 6.0
Urea	7.0 mmol/L	3.0 – 8.0
Creatinine	120 μmol/L*	45 – 90

Comment on the acid base status and give the three most likely causes for your findings. What further tests would you order to distinguish between them? (30% marks)

[Click here to toggle visibility of the answers]

College answer

i)   Warfarin, Vitamin K deficiency, liver disease, paracetamol toxicity
ii)   High anion gap metabolic acidosis. (AG 26, delta ratio 0.93). Likeliest causes are ketosis (alcoholic or starvation), pyroglutamic acidosis and ethanol/methanol toxicity.
iii)   Serum/urine ketone levels, measured osmolarity and plasma / urine assay for 5-oxoproline (pyroglutamate).

Examiners Comments:

When asked for a specific number of responses (e.g. 'three causes of') please supply this number of responses. Extra responses will not gain extra marks. If there are more causes, then list the most likely. Many candidates did not appear to pay attention to the mark allocation and gave insufficient detail in sections of the question worth the most marks.

Discussion

1) An isolated normal PT in a middle-aged alcoholic with known AF? What could it be??

Warfarin therapy
Vitamin K deficiency
Liver disease
Isolated Factor VII deficiency

2) A systematic approach to this gas:

The patient is normoxic: the A-a gradient is not particularly raised (0.21 x 713) - (22 x 1.25) - 90 = 32.23 mmHg
There is acidaemia
The CO₂ is appropriately low
There is a severe metabolic acidosis (the SBE is -16.7)
The compensation is perfectly appropriate: the expected PaCO₂ is (9 × 1.5) + 8 = 21.5, or 23.3 mmHg by the Copenhagen rules.
The anion gap is elevated: (135) - (100 + 9) = 26, or 30 when calculated with potassium
The delta ratio, without using potassium and assuming a normal anion gap is 12 and a normal bicarbonate is 24, would therefore be (26 - 12) / (24 - 9) = 0.93. This gives you the impression that the normal anion gap component of this acidosis is playing at best a trivial role.

Thus, this is a well-compensated severe metabolic acidosis with a high anion gap. And it's not due to lactate or uraemia. To apply a well-worn mnemonic, the differentials are

Methanol or other toxic alcohol poisoning
Ketoacisosis
- Diabetic
- Alcoholic
- Starvation
Pyroglutamic acidosis
D-lactic acidosis
Salicylate toxicity

The investigations, therefore, are:

Serum osmolality
Serum ketones (blood rather than urine, to detect the β-hydroxybutyrate)
Urinary 5-oxoproline level
Urinary oxalate level

References

EMMETT, MICHAEL, and ROBERT G. NARINS. "Clinical use of the anion gap."Medicine 56.1 (1977): 38-54.

Salem, Mahmoud M., and Salim K. Mujais. "Gaps in the anion gap." Archives of internal medicine 152.8 (1992): 1625-1629.

Kraut, Jeffrey A., and Nicolaos E. Madias. "Serum anion gap: its uses and limitations in clini

[Submit a comment or correction]