Context Sensitive Decrement Times

Chris Thompson, Royal Prince Alfred Hospital, Sydney, Australia Nov 2000

Volume of distribution, clearance and half-life are standard pharmacokinetic parameters for characterising drug offset. They were derived from the application of bi- or tri- exponential curve fitting algorithms to blood level measurements in defined patient groups, and accurately describe the changes in blood level after bolus administration.

While these pharmacokinetic data can be fairly readily applied to loading dose and infusion rate calculations for water-soluble drugs like muscle relaxants, the situation is much more complex with regard to continuous infusions of highly lipid-soluble anaesthetic agents. For example, drugs with long elimination half-lives like thiopentone and propofol are short acting agents, and their intermediate half-lives bear little or no resemblance to clinically relevant offset times. Likewise, the newer narcotics such as fentanyl and alfentanil have quite different offset profiles despite similar fundamental pharmacokinetic profiles.

Context sensitive elimination time 1 is a fairly new pharmacological concept which describes the rate of fall in the effect site drug concentration of a drug after cessation of a continuous infusion of defined duration. It is a new computer generated way of utilising the existing pharmacokinetic data. 'Context', in this sense, refers to the duration of the infusion. All infusions are computer 'titrated' to a fixed effect site concentration, (not a fixed arterial or venous blood concentration).

Hence, the 'one-hour context sensitive half-time' or 'one-hour half-time' of a given drug is the time taken for the effect site concentration of that drug to fall by 50% after cessation of a one-hour infusion. Likewise the '2 hour 80% fall time' (or decrement time) is the time taken for the effect site concentration to fall by 80% after a 2 hour infusion. For most agents, the 'one-hour half-time' will be quite different from the 'two-hour 80% fall time', and will probably be very different from the '6 hour 90% fall time'. The data is usually expressed in the form of a graph showing a family of curves, each indicating a given percentage fall, with the x axis indicating infusion duration and the y axis the time for a given percentage fall. Each drug has a characteristic family of curves, permitting ready comparison of the offset profiles of the different agents.

The percentage effect site concentration fall required for recovery from an anaesthetic is dependent on the level associated with clinical recovery as a proportion of the level associated with adequate anaesthesia. Typically a fall of approximately 80% seems to be required for emergence from anaesthetic agents. This takes about 36 minutes after a two hour propofol-only anaesthetic. If a given patient was had received twice as much propofol as they really needed (over 2 hours), then to get to a residual level of 10%, the 90% fall time will be105 minutes. In contrast, if they were getting half of what they had required (for example a sedative rather than anaesthetic dose) then recovery will occur 10 minutes of cessation of the infusion. The non-linear relationship between these numbers (a four-fold error in drug dosage causing a 10-fold change in recovery times) emphasises the apparently unpredictable nature of context sensitive elimination times &endash; and the importance of avoiding inadvertent over- or under-dosage during the case.

Remifentanil and alfentanil have context sensitive elimination times which, unlike fentanyl, do not increase with prolonged infusion duration. Remifentanil is very well suited for prolonged infusions, since offset will be rapid (50% fall in 3 minutes, 80% fall in 10 minutes) and highly predictable at all times. Kapila 2 has verified the accuracy of these values. Rapid offset may be associated with poor post-operative analgesia.

Alfentanil has very predictable offset, but offset is relatively slow after prolonged infusions or if large falls are required. For any infusion over 2 hours, a 20% fall takes 9 minutes and 40% takes 30 minutes. In contrast, a 60% fall takes 93 minutes, 80% takes 3 hours, and a 90% fall takes 5 hours! Clearly, time to recovery after an alfentanil infusion will primarily be dependent on the degree of over- or under- dosage during the case!

Fentanyl infusions have slower and less predictable offset characteristics. For example, at 2 hours, the 20% fall takes 15 minutes and the 40% fall takes 45 minutes (not much different from alfentanil), whereas the 60% fall takes 210 minutes, 80% takes 10 hours, and a 90% fall takes 18 hours! Offset of Fentanyl is markedly prolonged following anaesthetics of more than 2 hours duration. Fentanyl infusions only become predictable after about 6 hours, with very long offset times (about three to four times longer than alfentanil). While fentanyl is best suited to intermittent bolus techniques, administration during anaesthesia will provide prolonged mild postoperative analgesia.

I have calculated elimination times for inhalational agents using 'GasMan' 3, a commercial computer model of anaesthetic uptake and distribution based on well described kinetic data. I used a semi-closed circle circuit with 6 litres/min fresh gas flow for washout, and this may not be appropriate for postoperative elimination. Bailey 4 has described shorter decrement times for the volatile agents than those calculated from GasMan.

Nitrous oxide has rapid, predictable offset after prolonged administration, with an 80% fall in 9 minutes (to 0.1MAC). Interestingly the 90% fall takes 36 minutes and 95% takes 80 minutes, so there is some hangover. Rapid offset explains, in large measure, it's popularity in neurosurgery.

There is little difference in immediate recovery times between sevoflurane and isoflurane in short cases (one hour 60% recovery takes 7 and 9 minutes, and 80% recovery 12 and 17 minutes, respectively), though 90% recovery times differ significantly (20 and 51 minutes, respectively).

In longer cases, sevoflurane shows clear advantages over isoflurane, particularly for intermediate to late recovery, with 4 hour 80% recovery taking 17 vs 61 minutes and 90% recovery 89 vs 201 minutes for sevo and iso respectively. Note that if 80-90% falls are required at the end of the case, you may have to wait an unexpectedly long time! If you "get it wrong" with sevoflurane, you'll be out of trouble in less than half the time it would take with isoflurane!

Propofol is, surprisingly, even slower in offset than isoflurane! Four hour 80% recovery takes 60 minutes, which is comparable to iso, however 90% recovery takes 210 minutes, or almost four times as long! Rapid offset occurs only if relatively modest falls (40% or less) are needed for recovery (the four hour 40% fall takes only 4 minutes). Prolonged hangover effects are likely if it is infused at excessive doses and/or for long periods. As with the narcotics, some degree of sedation may or may not be a drawback postoperatively, depending on the nature of the sedation. Inadvertent prolonged overdoses will definitely delay recovery. Alfentanil and propofol kinetics are sufficiently different that they should be titrated separately, and to different endpoints.

I do not have comparative data for the barbiturate induction agents, however I expect that methohexitone will be a comparable to propofol and that thiopentone will be the most time dependant and the slowest of all the intravenous agents by a factor of 3 or 4 times.

I am grateful to Charles Minto of Royal North Shore Hospital for assistance with computer modelling for the narcotics and induction agents.

To summarise, to attain adequate anaesthesia intra-operatively and prompt, predictable recovery:

a) Anaesthetic agents should be titrated to the minimum effect site concentration consistent with adequate levels of anaesthesia. It is becoming increasingly clear that surrogate indices of 'depth of anaesthesia' such as blood pressure and heart rate are not as good as processed EEG or Bispectral Index monitors.

b) Choice of agent should be appropriate for the duration of the procedure and should be ceased at the right time, both of which can be predicted from context sensitive elimination data.

c) Nitrous oxide and remifentanil are rapid offset agents, which reduce requirements for other anaesthetic agents such as propofol or sevoflurane, and hence permit rapid emergence within the 40-50% elimination times of the longer agents.

d) Manufacturers should provide complete context sensitive elimination data when introducing new agents to facilitate comparison with existing drugs and to facilitate appropriate dosing schemes.


1. Hughes MA, Glass PS, Jacobs JR: Context-sensitive half-time in multicompartment pharmacokinetic models for intravenous anesthetic drugs [see comments]. Anesthesiology 76:334-41, 1992

2. Kapila A, Glass PS, Jacobs JR, Muir KT, Hermann DJ, Shiraishi M, Howell S, Smith RL: Measured context-sensitive half-times of remifentanil and alfentanil [see comments]. Anesthesiology 83:968-75, 1995

3. Philip J: GasMan v2.0. In: 1996:

4. Bailey JM: Context-sensitive half-times and other decrement times of inhaled anesthetics. Anesthesia and Analgesia 85:681-6, 1997


Comments? please use this form.