Medical School Lesson Plan – Carbon Monoxide Poisoning

Incomplete combustion of hydrocarbon fuels (gas, coal, wood, petrol) produces CO.

It is usually caused by an accidental lack of oxygen in the environment due to a blocked flue (gas boilers, wood stoves) or burning fuel in an enclosed space (indoor BBQ, car engine running in a garage).

CO Poisoning can affect all systems but classically you get

• Neuro symptoms (headache, nausea, dizziness, visual disturbance, ataxia, seizures);
• Psychocognitive symptoms (memory loss, concentration problems, confusion, personality change);
• Cardiorespiratory symptoms (SOB, chest pain, arrhythmias, hypertension)

Whilst CO Poisoning seems unlikely given a lack of symptoms, occasionally onset of symptoms is delayed and so advice on seeking help if any problems do occur would be wise.

This is also a good moment for some safety advice about having a CO alarm and ensuring heating appliances are regularly serviced, particularly in the autumn, as ventilation pathways can become blocked when appliances have been dormant during the summer months.

CO binds with haemoglobin (HbCO) with over 200 times the affinity of oxygen. This leads to impairment of oxygen delivery to tissues and cellular hypoxia. Dissolved CO in the blood also reaches the tissues and binds even more strongly to mitochondrial cytochromes, impairing cellular respiratory function further.

High concentration oxygen therapy is the mainstay of treatment in the ED.

A well fitted, non rebreather mask with reservoir bag at 10-15 litres per minute of oxygen can potentially reduce the half-life of HbCO from about 6 hours to only 90mins.

For most people with mild symptoms, high flow oxygen for 4-6hrs or until their symptoms resolve, is sufficient medical therapy.

Whilst HbCO levels can be useful for ruling in CO poisoning, unless a sample is taken very early after an acute exposure, they are not good for ruling out CO toxicity.

This is because the HbCO level begins falling immediately the patient is removed from the toxic environment, more quickly if they have been given oxygen therapy (perfectly appropriately). 

A similar issue occurs with chronic, intermittent, low dose exposure, where cellular CO binding can build up whilst HbCO is constantly cleared away. A clue to this may be that although the HbCO level is normal, a lactic acidosis might be present on the blood gas.

Ultimately, this means that CO poisoning is usually a clinical diagnosis, based on a history of potential exposure, appropriate symptomatology and resolution of symptoms with oxygen therapy.

Whilst the diagnosis appears obvious, it is important to consider other causes for an unconscious patient and investigate appropriately.

Thinking specifically of CO Poisoning, a blood gas with HbCO will be necessary, as will an ECG to look for cardiac ischaemia.

A CT of the head may be part of the general work up but is unlikely to show any changes related to CO poisoning at this time.

General supportive care alongside oxygen delivery at the highest concentration available is the management strategy. In this setting intubation and ventilation with 100% oxygen is needed.

A higher ventilation rate can aid CO expulsion but needs to be balanced against the dangers of a low pCO2 on cerebral perfusion.

Whilst this patient clearly has severe CO Poisoning the use of hyperbaric oxygen (HBO) therapy remains controversial.

In theory the best chance of it working is in the initial ‘washout’ phase of treatment (<6hrs) but the inevitable delays of accessing HBO and the logistical problems associated with managing a patient in a hyperbaric chamber, means it is rarely used outside specialist centres. In pregnancy there may be some benefit in using HBO to potentially increase oxygen delivery to the foetus, so cases should be discussed with a Poisons Centre for consideration of treatment.