Total intravenous anaesthesia (TIVA) unsurprisingly refers to the provision of anaesthesia entirely by the intravenous route, with the avoidance of the more commonly used volatile agents. This will usually require the combination of more than one agent, due to the need of multiple actions and the lack of a perfect single anaesthetic agents. The closest to the ‘perfect’ recipe is propofol (anaesthesia) with remifentanil (analgesia).
This different approach to anaesthesia has certain advantages to the use of volatile, and so has a role in some indicated scenarios:
High PONV risk
Anaesthesia in remote environment
‘Non-tube’ airway or thoracic surgery
Avoidance of neuromuscular blockade e.g. neuromuscular disorders.
Neurosurgery - limit intracranial volume
Lower incidence of PONV
Safe in malignant hyperthermia
More rapid recovery
Better CVS stability
Ease of titratability
Risk of awareness e.g. disconnection
Monitoring of anaesthetic depth
Set up time
As with volatile anaesthetic agent, an understanding of how the anaesthetic agent behaves is important to maintain patient safety. With TIVA this will particularly relate to the pharmacokinetics of the drugs used, as this impacts on a lot of the conduct of TIVA. Much of this discussion will focus on propofol.
The EC50 This concept is the parallel of MAC. It is defined as the effective plasma concentration of a drug that will prevent a response from a standard surgical incision in 50% of patients. For propofol this is 6-7 microgram/ml.
The 3 Compartment Model This is a model for describing how propofol (and other drugs) will behave after IV administration. As it is the concentration of the drug in the plasma (and then in the brain) that will produce anaesthesia, this model is essential to help us understanding how IV administration translates into a plasma concentration. The primary physical action is that of diffusion.
After an IV bolus of a drug the drug will equilibrate across ‘compartments’ (i.e. the tissues) in the body. This will result in an exponential decline of the drug in the plasma. Many current models use the concept of 3 compartment system to represent this because of the similar physical characteristics of the compartments. The compartment are:
The circulation i.e. the plasma
Well perfused tissues - muscles, viscera
Poorly perfused tissues - primarily fat
The model relies on calculation of:
The volumes of distribution of these compartments
The rate constants between compartments
The overall elimination rate
Much of these calculations will be estimates, based on input data about the patient:
Once all these factors are taken into account, and a mathematical algorithm applied, estimations of the plasma (or effect site) concentration can be made.
As a simple explanation. Equilibration with compartment 1 can be seen a nearly instantaneous. There will then be ongoing equilibration with compartments 2 and 3, with the drug diffusing down the concentration gradient. With compartment 2 this will be fairly rapid, whilst with compartment 3 it will be slower. Once equilibrium has been reached, the plasma level will continue to drop, depending on the rate of systemic elimination. The diffusion gradient will therefore reverse, with movement of the drug back into the plasma from the compartments, at the same sort of speeds.
This will require administration of the agent intravenously in a manner that takes into account the complex pharmacokinetics of the drug. Whilst boluses can be given, there is much more likely to be profound peaks and troughs in the plasma levels, with corresponding risks of adverse effects and awareness.
More controlled administration could be via:
Target controlled infusion
Manual infusion This uses an set infusion rate that provides a rough account of the pharmacokinetic profile of propofol. The Bristol model is an example of this, employing a 10-8-6 formula:
10mg/kg/h for 10 mins
8mg/kg/h for next 10 mins
6mg/kg/hr for remainder
This will on average lead to a plasma concentration of 3.67 micrograms/ml.
Target Controlled Infusion (TCI) With improved technology, more complicated infusion programmes have been developed. These allow input of the patient’s demographics (age, weight, gender) and can provide a estimated model of the patient’s pharmacokinetics. This allows administration of the appropriate amount of propofol to achieve the desired (calculated) plasma (or effect site) concentration.
Differences between different TCI programmes include:
Plasma site (Cp)
Effect site (Ce)
The different algorithms are similar in design but use different assumptions in their calculations. The Marsh and Schnider models differ in their estimates of the plasma compartment for example (by a factor of around 3!) This results in quite a difference in the calculated loading dose that they will use.
There is the option to set different ‘targets’ for the infusion. Whilst much of the discussion and study relates to plasma levels, it is actually the level of propofol in the brain that will determine anaesthesia. The propofol will therefore be required to cross the blood brain barrier and reach the receptors, which is fast, but not instantaneous. The ‘effect site’ model has compensatory calculations built in to take into account this additional facet. The response to changes that are input to the pump are therefore generally faster. A cited downside is that the systemic effects of the higher plasma levels that occur to achieve this (e.g. CVS compromise).
These calculations are just estimates. And even if correct, as with other anaesthetic agents, the actually result of a specific effect site concentration is dependent on other influencing factors, including:
Coadministered sedatives e.g. opioids
There will be variability in the response to the same plasma/effect site levels between patients. However, an example of possible target levels (microgram/ml) include:
ASA 1-2 (age 55) Induction 6 Maintenance 3-6
ASA 3-4 (age 55) Induction 3.5 Maintenance 2-5
Return of consciousness will often occur at calculated effect site levels of 1-2 mcg/ml.
Whilst other opioids can be used, remifentanil has the near perfect pharmacokinetic profile. It has a very rapid onset and offset. In particularly, its fixed context sensitive half time of 3-5 minutes means that there will not be an impact on plasma levels due to to the length of the infusion. This is problematic with other opioids, as the progressive accumulation of the drug can make the models used increasingly inaccurate and prolong wake up.
Remifentanil will often be combined with propofol in TIVA. In this role it will provide the analgesia, but its potency also provides a degree of muscle relaxation, allowing ventilation of the patient. This may allow muscle relaxants to be only given to facilitate intubation, or even not at all, to reduce the concerns about awareness.
At induction, a dose of 0.5-1 mcg/kg/min is used. Maintenance dose can be very varied, with a quoted range of 0.05-2 mcg/kg/min.
Most of the factors will be identical to other forms of general anaesthesia. A dedicated cannula should be used for the TIVA. This should be visible to help identify disconnection immediately. Use of depth of anaesthesia monitoring e.g. BIS, should be strongly considered.
Some specific features relate to the infusion pump. If a free syringe attached to a loose hanging infusion line is left at a position above the end of the line, it will empty itself fairly rapidly - a process known as siphoning. It’s clear that this could be disastrous. Similarly, deadspace in lines could result in quite a long delay of administration of drug if it is not taken into account. Finally, wrong way flow up a line could also prevent drug administration. Steps to minimise the risk of this include:
Securely locking syringe in pump
One way valves in line
Have connection close to patient
Remove all air from syringes
Keep syringe at a similar level to the patient
Avoid excessively concentrated solutions (which would run at low speed)