Arterial blood gas sampling
PLAB 2 & PRES 3 · practical skills / respiratory · hard · 8 minutes
Candidate brief
You are an FY2 doctor on a respiratory ward. Mr Colin Mercer, 66, has chronic obstructive pulmonary disease and was admitted yesterday with an infective exacerbation. He is on 28% oxygen via a Venturi mask. The registrar wants an arterial blood gas to assess his gas exchange before deciding about non-invasive ventilation. Observations: respiratory rate 26, oxygen saturations 90% on 28% Venturi mask, heart rate 104, blood pressure 138/84 mmHg. He takes warfarin for atrial fibrillation. Please perform the arterial blood gas sample and interpret the result for the examiner. Verbalise each step as you perform it.
What examiners look for
An ABG result without an FiO2 is a number, not a result. "PaO2 8.4" means one thing on air and something entirely different on 28% controlled oxygen — candidates who document the oxygen device and percentage before they start have understood what the test is for. The Allen test is the station's signature safety step. The radial artery is expendable only if the ulnar artery can supply the hand; checking collateral flow takes twenty seconds, and skipping it in an exam mirrors skipping it on a ward. Pressure is part of the procedure, and the dose depends on the patient. Five minutes is the floor; anticoagulation (here, warfarin for AF) extends it — most post-ABG haematomas are aftercare failures, not puncture failures. Sample quality is physiology. An air bubble drags the result toward room-air values; delay lets the cells in the syringe keep metabolising. Expel air, cap, mix, analyse promptly — or the management decision is built on a fiction. Read this gas as a story: acidaemia with a high PaCO2 is respiratory failure; a raised bicarbonate and base excess say the kidneys have been compensating for a while — acute-on-chronic, in a man whose previous admission needed NIV. The gas does not just describe him; it predicts the conversation with the registrar. Local anaesthetic is kindness with evidence behind it: ABG sampling is genuinely among the more painful routine procedures, and offering it (then waiting for it to work) is a marker of a doctor patients trust with the next needle too. The endpoint of a practical station is a decision. Here: maximal medical therapy continues, oxygen stays controlled to 88–92%, and the registrar assesses for NIV now. A candidate who interprets beautifully and escalates nobody has performed a blood test, not practised medicine.
Guideline references
Practise this station
Role-play it with an AI patient and get feedback against the full marking scheme. Create a free account — 3 stations per month free.