A Primary FRCA station on MedMock, start to finish

Good morning. We'll start with pharmacology. What do you understand by the term context-sensitive half-time?

It's the time taken for the plasma concentration of a drug to fall by half after you stop an infusion that has been maintaining a steady plasma concentration, where the context is how long the infusion has been running. Hughes and colleagues described it in 1992, because the elimination half-life tells you very little about how quickly a patient will wake after an infusion. Most intravenous anaesthetic drugs behave as if they're distributed in three compartments: a central compartment, which is the plasma and the best-perfused organs, and two peripheral compartments. They're mathematical constructs, but the fast one corresponds roughly to muscle and the slow one to fat. When you stop an infusion, the plasma concentration falls through elimination and through continuing redistribution into the peripheral compartments. After a short infusion they're far from full, so redistribution helps and the concentration falls quickly. After a long infusion they've filled, drug returns from them into the plasma, and the fall is slower. If I draw it, infusion duration goes on the x-axis and context-sensitive half-time on the y-axis. Fentanyl's curve climbs steeply, to several hours after a long infusion. Propofol's rises far less. Remifentanil's is flat, at about three to four minutes whatever the duration, because it's broken down by non-specific esterases in blood and tissues and doesn't accumulate. And the curves don't follow the elimination half-lives: for infusions of up to several hours, sufentanil's context-sensitive half-time is shorter than alfentanil's, although its elimination half-life is longer.

Thank you. How does a target-controlled infusion pump use these ideas?

The pump contains a pharmacokinetic model, usually a three-compartment model with its volumes and rate constants, and a microprocessor that recalculates the infusion every ten seconds to reach and hold the target concentration the anaesthetist sets. To reach a higher target, it gives a bolus to fill the central compartment, then an infusion that starts high and falls exponentially towards the rate that just replaces elimination, as the peripheral compartments fill. That's the bolus, elimination, transfer principle. If the target is lowered, it stops the infusion until the model predicts the concentration has fallen to the new target, then restarts at a lower rate. You can target the plasma or the effect site. The effect site is a theoretical compartment, and the rate constant ke0 describes how quickly it equilibrates with the plasma, which accounts for the delay between a change in plasma concentration and the clinical effect. In effect-site mode the pump deliberately overshoots the plasma concentration to drive drug into the effect site faster. For propofol, the two models most used in adults are Marsh, which uses weight alone, and Schnider, which uses age, weight, height and sex. The Eleveld model was developed from a much wider range of patients, including children, older and obese people. For remifentanil it's the Minto model. The key limitation is that the pump measures nothing: the concentration it displays is a prediction from the model, and the real concentration in an individual patient can be quite different.

Thank you. What steps would you take to reduce the risk of accidental awareness when you give a propofol and remifentanil TCI?

NAP5 found that reports of awareness were more common with TIVA, that most cases were preventable, and that the commonest contributory factor was inadequate education and training. So, first, make sure the drug gets in. I'd use a dedicated TIVA giving set with Luer locks at each end, an antisyphon valve on the drug line and an anti-reflux valve on any fluid line, with the lines joining as close to the patient as possible, and I'd keep the cannula visible throughout wherever that's practical. I'd program the pump only after the syringe is loaded, and check the drug, its concentration, the model and the patient's details, ideally with a second person. Standard concentrations, labelled syringes, attention to alarms, and never letting a syringe run out. Second, monitor the effect. If I give a neuromuscular blocker, the patient can't move to show me they're aware, so I'd use processed EEG monitoring, which the joint Association of Anaesthetists and Society for Intravenous Anaesthesia guidelines recommend whenever a neuromuscular blocker is used with TIVA, as well as quantitative neuromuscular monitoring. And I'd be especially careful at the transitions, such as induction and the move from the anaesthetic room into theatre, when gaps in delivery happen.

Thank you. Let's move on to physiology. How is oxygen carried in the blood?

Mostly bound to haemoglobin, with a small amount dissolved in plasma. Each gram of fully saturated haemoglobin carries about 1.34 millilitres of oxygen, which is Hüfner's constant, and dissolved oxygen follows Henry's law, at about 0.023 millilitres per decilitre of blood for each kilopascal. So the arterial oxygen content is 1.34 times the haemoglobin in grams per decilitre times the saturation, plus 0.023 times the PaO2 in kilopascals. With a haemoglobin of 15 grams per decilitre, a saturation of 98% and a PaO2 of 13 kilopascals, that's about 19.7 millilitres bound plus 0.3 dissolved: roughly 20 millilitres per decilitre, with only about 1.5% of it dissolved. Oxygen delivery is cardiac output times arterial content, so with a cardiac output of 5 litres a minute it's about 1,000 millilitres a minute. Resting consumption is about 250 millilitres a minute, so the extraction ratio is about 25%, and mixed venous blood returns about 75% saturated.

Thank you. Could you draw the oxyhaemoglobin dissociation curve and explain its shape?

Saturation from 0 to 100% on the y-axis, and PO2 in kilopascals on the x-axis. It's sigmoid. I'd mark three points: the P50, the PO2 at which haemoglobin is half saturated, about 3.5 kilopascals in an adult; the venous point, about 5.3 kilopascals and 75%; and the arterial point, about 13 kilopascals and 97 to 98%. The shape comes from cooperative binding. Haemoglobin is a tetramer of two alpha and two beta chains, each with a haem group containing ferrous iron. Deoxyhaemoglobin is in the tense state, with a low affinity for oxygen, and as oxygen binds, the molecule shifts towards the relaxed state, so the remaining sites bind more readily. The shape matters clinically. On the flat upper part, saturation holds up even when the PO2 falls quite a lot, which is a safety margin, but it also means a pulse oximeter won't show a falling PaO2 until you reach the steep part: a saturation of about 90% corresponds to a PaO2 of about 8 kilopascals. On the steep part, the tissues can unload a lot of oxygen for a small fall in PO2.

Thank you. What shifts the curve, and why does that matter?

A shift to the right means lower affinity and a higher P50. It's caused by a rise in hydrogen ion concentration, a rise in PCO2, a rise in temperature and a rise in 2,3-DPG, which is exactly the environment in exercising muscle, so oxygen is unloaded where it's needed. The effect of carbon dioxide and hydrogen ions is the Bohr effect. A shift to the left, higher affinity, comes with alkalosis, hypocapnia, hypothermia and low 2,3-DPG, as in stored blood. 2,3-DPG is made in red cells by a side branch of glycolysis, and it rises with chronic hypoxia, such as at altitude or in chronic anaemia. Fetal haemoglobin is left-shifted because its gamma chains bind 2,3-DPG poorly, which helps the fetus take up oxygen from maternal blood in the placenta. The Haldane effect is the counterpart for carbon dioxide: deoxygenated blood carries more carbon dioxide, as carbamino compounds and because deoxyhaemoglobin is a better buffer of hydrogen ions, which helps load carbon dioxide in the tissues and unload it in the lungs.

Thank you. Finally, how does carbon monoxide affect oxygen carriage?

Carbon monoxide binds haemoglobin with over 200 times the affinity of oxygen, forming carboxyhaemoglobin. That reduces the oxygen content, because those sites can't carry oxygen, so functionally it's like anaemia. It also shifts the curve of the remaining haemoglobin to the left, so the oxygen that is carried is released less readily to the tissues, which makes it worse than an anaemia of the same degree. The PaO2 can be entirely normal, because it reflects dissolved oxygen, not content. Carbon monoxide also binds cytochrome oxidase and impairs oxygen use in the mitochondria. Treatment is high-flow oxygen, which shortens the half-life of carboxyhaemoglobin, with hyperbaric oxygen considered in some severe cases.

Thank you. That's the end of the station.