Arterial blood gases

Editor, – I read with concern the article on arterial blood gases (Aust Prescr 2010;33:124-9). I believe the emphasis of arterial blood gases over venous blood gases is no longer representative of what is being taught and practised in acute care medicine.

Venous blood gases are easier to obtain, hurt less, are safer and provide extra information about tissue oxygen use that arterial blood gases do not. In combination with a pulse oximeter reading, venous blood gases can be used to guide clinical decision making in the majority of situations where arterial blood gases have previously been thought to be necessary. Venous blood gases are therefore better than arterial blood gases most of the time.

Arterial blood gases are now rarely obtained from patients in emergency departments, especially children, unless there is repeated sampling from an arterial line, usually inserted for haemodynamic monitoring. This is because venous blood gases (along with pulse oximetry) provide adequate information for the majority of acute paediatric and adult clinical scenarios, including sepsis, asthma, chronic lung disease, toxicology, diabetic ketoacidosis, and therapy adjustments for invasive and non-invasive ventilation. Reviews in the literature aim to educate that venous blood gases can replace arterial blood gases in most acute care clinical scenarios.^1,2

Decisions involving oxygenation can be made with information from a pulse oximeter, unless there is poor waveform. Modern pulse oximeters are accurate +/– 2% down to saturations as low as 70%. Given this accuracy, it is questionable concerning the value of arterial verses venous blood gases and pulse oximetry to assess the need for domiciliary oxygen therapy.

Although local anaesthetic reduces the pain of arterial blood gases sampling without decreasing success rates, a better option is to just not do them at all.

Lindsay Bridgford
Director of Emergency Medicine Training
Maroondah Hospital
Ringwood East, Vic.

Dr Abhishek Verma and Dr Paul Roach, authors of the article, comment:

We acknowledge that in the acute setting, sampling venous blood is sufficient to obtain information about a patient's acid–base and ventilation status. Combined with pulse oximetry, venous blood gases are useful in a variety of clinical scenarios. However, there are some important caveats. It is essential to obtain a good waveform for pulse oximetry if the result is used for estimating the oxygen saturation and partial pressure. Yet, in several acute situations – for instance, sepsis, trauma or cardiac arrest – peripheral circulation may be inadequate so it is difficult to obtain any information about potential hypoxaemia. Pulse oximetry can also be influenced by other factors, such as if the patient is vasoconstricted due to inotrope use or is excessively moving or shivering. Also when a patient presents with toxic gas exposure or carbon monoxide poisoning, a falsely high oximetry reading may confound the recognition of severe tissue hypoxaemia. Taking an arterial blood gas sample in these instances ameliorates the problems of estimating the oxygenation entirely.

Pulse oximetry, while being a far less invasive method of determining the state of oxygenation than arterial blood gas analysis, does rely on an understanding of the physiology of the oxygen–haemoglobin disassociation curve. These are concepts that many medical students and junior doctors are not always cognisant of, and so the interpretation of oxygenation status from an arterial blood gas sample remains important.

Current Australian guidelines still require arterial blood gas analysis before domiciliary oxygen can be legally prescribed. Accordingly, the performance and interpretation of arterial blood gases remains a very important skill for a clinician.