Recovery from Anesthesia

Recovery from anesthesia is determined by the pharmacokinetic principles that govern the rate of decrease in drug from the effect compartment, once drug administration is terminated, as well as by the pharmacodynamics of the drug. Although the terminal elimination half-life is often interpreted as a measure of how short or long lasting a drug is, the rate at which a drug decreases is dependent on both elimination and redistribution of the drug from the central compartment. The contribution of redistribution and elimination to the rate of decrease in drug concentration varies according to the duration for which the drug has been administered. ^{[78] [82]}

In 1985, Schwilden^[83] developed a mathematical model to relate the time course of offset of the inhaled anesthetics to the duration of anesthetic drug delivery. Similarly, Fisher and Rosen^[84] demonstrated how accumulation of muscle relaxants in peripheral volumes of distribution results in slowed recovery with increasing duration of administration. They introduced two measures of the time course of recovery, the time for twitch tension to recover from 5% to 25% and the time for twitch tension to recover from 25% to 75% (also see Chapter 13 ).

Since then, the time for the plasma concentration to decrease by 50% from an infusion that maintains a constant concentration (e.g., the infusion given by Equation 14) has been termed the "context-sensitive half-time" ( Fig. 12-17 ),^[82] with the context being the duration of the infusion. The 50% decrease was chosen both for tradition (e.g., half-lives are the time for a 50% decrease with a one-compartment model) and because, very roughly, a 50% reduction in drug concentration appears to be necessary for recovery after the administration of most intravenous hypnotics at the termination of surgery. Depending on the circumstances, decreases other than 50% may be clinically relevant. Additionally, sometimes it is the plasma concentration that is of interest, and sometimes it is the effect-site concentration that is of interest. A more general term is the context-sensitive "decrement time,"^[85] in which the decrement in concentration is specifically noted, as is the compartment in which the decrease is modeled (plasma or effect site). For example, the relationship between infusion duration and the time required for a 70% decrease in fentanyl effect-site concentration is the "context-sensitive 70% effect-site decrement time."

Context-sensitive effect-site decrement times for varying percent decreases in alfentanil, fentanyl, and remifentanil concentration are illustrated in Figure 12-18 . To determine when an infusion should be termined (to enable awakening of the patient at the end of surgery), the clinician needs to bear in mind the decrease in concentration necessary for recovery, the duration of the infusion (the context), and the context-sensitive effect-site decrement time required for the necessary decrease.

Figure 12-17 Context-sensitive half-times as a function of infusion duration (the "context") derived from pharmacokinetic models of fentanyl, sufentanil, alfentanil, propofol, midazolam, and thiopental. (From Hughes MA, Glass PSA, Jacobs JR: Context-sensitive half-time in multicompartment pharmacokinetic models for intravenous anesthetic drugs. Anesthesiology 76:334–341, 1992.)

456

Figure 12-18 Context-sensitive effect-site decrement times for alfentanil, fentanyl, sufentanil, and remifentanil showing the time required for decreases in a given percentage (labeled for each curve) from the maintained effect-site concentration after termination of the infusion.

Context-sensitive decrement times are fundamentally different from the elimination half-life. With monoexponential decay, each 50% decrease in concentration requires the same amount of time, and this time is independent of how the drug was given. Such is not true for the context-sensitive half-time. First, as the name is intended to imply, the time to achieve a 50% decrease is absolutely dependent on how the drug was given, with infusion duration being the context to which the name refers. In addition, small changes in percent decrement can result in surprisingly large increases in the time required. As can be seen from Figure 12-18 , in some situations the time required for a 60% decrease in drug concentration can be more than twice the time required for a 50% decrease.

Context-sensitive decrement times are based on the assumption that the plasma or effect-site concentration is maintained at a constant level. This is rarely the case clinically, but maintenance of constant concentrations is a necessary assumption to provide a unique mathematical solution to the time required for a given percent decrement in plasma or effect-site concentration. Because plasma and effect-site concentrations are rarely kept constant, it is important that context-sensitive decrement times be used as general guidelines for interpreting the pharmacokinetics of intravenous drugs and not as absolute predictions for any given case or infusion regimen. Automated drug delivery systems can provide more precise predictions of the time required for the plasma or effect-site concentration to decrease to any desired concentration based on the actual drug dosing in the individual patient. This information provides the clinician with guidance for the most appropriate time to terminate the infusion.

Context-sensitive decrement times focus on the role of pharmacokinetics in recovery from anesthesia. Pharmacodynamics plays an important role in recovery as well. Bailey^[86] used integrated pharmacokinetic/pharmacodynamic models to define the "mean effect time" as the average time to responsiveness after maintenance of anesthesia at the 90% probability of unresponsiveness. The mean effect time demonstrates that when drugs have a very shallow concentration-versus-response relationship, concentrations must decrease by a significant fraction to provide adequate emergence, which delays recovery from anesthesia. In contrast, recovery is hastened by a steep concentration-versus-response relationship, in which emergence from anesthesia occurs after a relatively small fractional decrease in concentration. Most intravenous hypnotics have a fairly steep concentration-versus-response relationship.

Pharmacodynamic drug interactions also play a role in recovery from anesthesia. Interaction relationships predict that the same anesthetic state can be achieved by different ratios of two drugs. One way of selecting the best ratio might be the combination that offers the most rapid recovery. For example, when an opioid is combined with a hypnotic, recovery from anesthesia depends on the opioid and hypnotic concentrations, the rate of decrease in both drugs, and the relative synergy between them for loss of response to noxious stimulation (i.e., the state maintained during anesthesia) versus the relative synergy

457

Vuyk and coworkers^[87] modeled the predicted time to awakening when propofol is combined with fentanyl, sufentanil, alfentanil, or remifentanil on the basis of the interaction between propofol and these opioids to provide adequate anesthesia and the interaction between propofol and opioids on emergence from anesthesia, as seen in Figure 12-20 and Figure 12-21 . Recovery times vary according to the opioid selected and the relative balance of opioid and propofol during maintenance of anesthesia. For example, in the upper left of Figure 12-20 is a simulation of emergence from a propofol/fentanyl anesthetic of 10 minutes' duration. The simulations assume a steady concentration of fentanyl and propofol throughout the anesthetic, similar to the underlying assumption of context-sensitive decrement times. The red curve in the lower plane is the interaction curve between fentanyl and propofol; it ranges from no fentanyl and 12 µg/mL of propofol on the left to 1.8 µg/mL of propofol and 6 ng/mL of fentanyl on the right. In theory, any point along this curve would provide equivalent maintenance of anesthesia. When the infusion is turned off after 10 minutes of anesthesia, the concentrations of both drugs decrease. The decreasing concentrations of propofol and fentanyl when the infusion is turned off can be found by the upward lines drawn from different points on the interaction curve, with the distance away from the lower plane representing time. Taken together, these upward lines represent a "recovery surface." The red line drawn on the recovery surface shows the points at which the fentanyl-propofol interaction model predicts emergence.

Figure 12-19 Interaction between hypnotics and opioids for prevention of movement after a noxious stimulus and for awakening and adequate spontaneous ventilation at the end of a surgical procedure. From this interaction it can be seen that the time to recover at the end of a procedure is dependent on the concentration of both drugs used during surgery and the time for both to decrease below that required for consciousness and adequate spontaneous ventilation (i.e., their context-sensitive decrement times).

Figure 12-20 shows that after 10 minutes of maintaining 1.8 µg/mL of propofol and 6 ng/mL of fentanyl (right edge of the interaction curve), it takes approximately 12 minutes for the concentrations of both drugs to decrease enough to permit emergence. However, if one maintains concentrations of 3.5 µg/mL of propofol and 1.5 ng/mL of fentanyl (toward the middle of the interaction curve), emergence can be expected just 8 minutes after the infusions are turned off. Examination of the curves for 60, 300, and 600 minutes of fentanyl/propofol anesthesia suggests that the fentanyl target concentration that provides the most rapid emergence is approximately 1.0 to 1.5 ng/mL, which requires a propofol concentration of approximately 3.0 to 3.5 µg/mL to maintain adequate anesthesia. These values are consistent with the earlier observation that most of the MAC reduction benefit of fentanyl occurs within the analgesic range and that exceeding this range is of little benefit during anesthesia. In similar simulations, Vuyk and colleagues demonstrated that maintaining alfentanil and sufentanil concentrations in excess of the analgesic range (i.e., approximately 80 ng/mL for alfentanil and 0.15 ng/mL for sufentanil) is of little clinical benefit but can be expected to delay recovery from anesthesia. A second conclusion from these simulations is that if the patient demonstrates inadequate anesthesia, to prevent prolongation of recovery, it is preferable to increase the hypnotic concentration than to increase the opioid concentration beyond the analgesic range.

The situation is different for remifentanil because of its unusual pharmacokinetic properties (see Fig. 12-21 ) (also see Chapter 11 ). The extraordinary clearance of remifentanil results in a very rapid offset of opioid drug effect when a remifentanil infusion is terminated. In Figure 12-21 , the lower plane again shows equivalent anesthetic states during maintenance with an opioid, remifentanil, and propofol. High doses of remifentanil permit a modest reduction in the dose of propofol necessary for

458

Figure 12-20 Simulation of the interaction of propofol and fentanyl in preventing a somatic response at skin incision and time to recovery. On the x axis is the fentanyl concentration, and on the y axis, the propofol concentration. The curve in the lower plane shows the propofol-fentanyl interaction required to provide adequate anesthesia. When the infusion is turned off, the concentrations of each drug decrease, as shown on the z axis. The curve drawn on the recovery surface shows the time to emergence from anesthesia for combinations of fentanyl and propofol after an anesthetic of 10 minutes' (A), 60 minutes' (B), 300 minutes' (C), or 600 minutes' (D) duration. Note that the optimal combination for the most rapid recovery is a propofol concentration of 3.0 to 3.5 µg/mL of propofol combined with 1.5 ng/mL of fentanyl. As the concentration of propofol or fentanyl increases, the time for recovery increases. In addition, the longer the duration of drug infusion, the longer that recovery takes, especially if the optimal combination is not used. Adapted from Vuyk et al. ^[87]

The offset of effect of isoflurane and sevoflurane is similar to that of propofol. The offset of desflurane effect is slightly quicker than that of propofol. We can draw some logical extrapolations from these propofol/opioid simulations to anesthetics consisting of volatile anesthetics combined with opioids. When fentanyl, alfentanil, or sufentanil is combined with a volatile anesthetic, the most rapid recovery occurs when the opioid concentration is maintained at an analgesic concentration equivalent to 1 to 1.5 ng/mL of fentanyl ( Table 12-5 ). Concurrently, the volatile anesthetic should be administered at the lowest concentration required to provide adequate anesthesia but no less than an end-tidal concentration of 0.3 MAC, the MAC value for return of consciousness. If the patient demonstrates signs of inadequate anesthesia, it is preferable to increase the volatile anesthetic because this agent has less of an effect on prolonging the wake-up time than does increasing the opioid (with the exception of remifentanil) and is more likely to ensure that awareness does not occur.