Intravenously Administered (Nonopioid) Anesthetics

Traditionally, these intravenous anesthetics were used solely for induction of anesthesia as a single intravenous bolus. Only with the clinical introduction of propofol have these agents been infused for maintenance of anesthesia, in which case assessment of the depth of anesthesia becomes more relevant.

Assessing Depth during Induction of Anesthesia

Induction of anesthesia often consists of a rapid intravenous bolus injection of a hypnotic (e.g., propofol, thiopental, etomidate). Plasma concentrations peak within a half to 1 minute and decline rapidly on redistribution of the drug. The rapidly changing plasma concentrations cause a corresponding fluctuation in the degree

Figure 31-14 Interaction surface for fentanyl and isoflurane reflecting suppression of the autonomic response. The surface is based on the isoflurane interaction reported by McEwan and colleagues,^[23] but modified to increase the C₅₀ of isoflurane in the absence of fentanyl by 1.3 and adjusting the isoflurane C₅₀ in the presence of an effect-site fentanyl concentration of 1.36 (corresponding to a 1.5-mg/kg bolus) to 0.4 MAC, as reported by Daniel.^[37] The black arrow shows the reduction in MAC_BAR demonstrated by Daniel and associates.^[37]

Clinical end points useful in assessing the depth of anesthesia during induction include loss of verbal responsiveness, loss of eyelid reflex, and loss of corneal reflex. Typical stimulation occurring during induction of anesthesia includes laryngoscopy and intubation, which constitute profoundly noxious stimuli. Frequently, response to these two procedures cannot be eliminated completely with just the intravenously administered hypnotic. In one study, administration of thiopental (6 mg/kg) was followed by an average increase in systolic blood pressure of 53 mm Hg on laryngoscopy and intubation.^[93] In another study, administration of thiamylal (4 mg/kg) was followed by an increase in mean arterial blood pressure from 92 mm Hg (control) to 136 mm Hg on laryngoscopy.^[94] Because most intravenous hypnotics do not provide significant analgesia, the hemodynamic response to major noxious stimuli is great, even when large doses are given. Thus, assessment of the depth of anesthesia with the use of clinically relevant noxious stimuli such as laryngoscopy and intubation requires the concurrent administration of other analgesic drugs (opioids or nitrous oxide) to provide reasonable and clinically acceptable hemodynamic control.

Most research on estimating the depth of anesthesia induced by intravenous anesthetics has focused on the relationship between dose and response. For example, Brett and Fisher^[95] reported that the dose of thiopental associated with a 50% probability of no movement in response to a firm squeeze of the trapezius muscle was 3 to 7 mg/kg for adults and more than 7 mg/kg for infants 1 to 11 months of age. It is possible that a larger initial volume of distribution or a more rapid redistribution (both pharmacokinetic mechanisms) in infants accounts for the difference in dose requirement. It is also possible that brain sensitivity to thiopental differs for infants and adults. Dose-response studies cannot differentiate between these two very different mechanisms.

Dose-response studies can have meaningful scientific and clinical value if they are performed with multiple measures of drug effect, concomitant variables, and appropriate statistical data analysis. Avram and associates^[96] demonstrated this point by successfully examining the thiopental induction dose requirement using an EEG end point (3 to 5 seconds of isoelectric signal), clinical end points (dropping of a syringe), and measurement of multiple covariates (age, gender, lean body mass, cardiac output). Avram's group^[96] concluded that age and either lean or total body weight are the most important predictors of thiopental dose requirement, with gender and cardiac output being less important. Jacobs and Reves^[97] provided editorial comment on this article and used

1244

Assessing Depth during Maintenance of Anesthesia

Becker^[98] presented one of the first studies to quantitate the relationship between plasma concentrations of an intravenous anesthetic (in this case, thiopental) and clinical measures of depth of anesthesia. First studying patients anesthetized with 67% nitrous oxide and thiopental, Becker found that the corneal reflex and movement response to a firm squeeze of the trapezius muscle correlated highly with the movement response to surgical stimulation (cervical dilation or skin incision). He then related the plasma concentration of thiopental to three clinical signs (loss of eyelid reflex, loss of corneal reflex, and absence of movement in response to squeezing the trapezius muscle) in another group of patients given thiopental/oxygen anesthesia. Anesthesia was induced with thiopental, 2 to 2.5 mg/kg, followed by an intravenous infusion of 1 to 1.5 mg/kg/min. Patients were observed for the three clinical signs. Arterial blood samples drawn at these clinical end points were analyzed for both total plasma concentration and free (or unbound) plasma concentrations of thiopental. Plasma levels of thiopental gathered under these pseudo-steady-state conditions, especially free or unbound plasma levels, are believed to be accurate predictors of brain levels of the drug and are therefore good indicators of the depth of anesthesia. The eyelid reflex was lost at significantly lower levels of thiopental than the corneal reflex or movement response was. Similar plasma concentrations of thiopental were needed for loss of the corneal reflex and for loss of the movement response to squeeze of the trapezius muscle, both of which had been found to correlate highly with loss of movement in response

Figure 31-15 A, Move/no move versus serum thiopental concentration for five different clinical stimuli. Each bar indicates the serum thiopental concentration and response to stimuli applied to an individual patient. B, Predicted probability of no movement versus serum thiopental concentrations obtained by using logistic regression of the data indicated in A. The bars indicate the 95% confidence bounds of the estimate of serum thiopental concentration that produces a 50% probability of no movement response. I, laryngoscopy/intubation; L, laryngoscopy; S, trapezius muscle squeeze; T, tetanic nerve stimulation; V, verbal. (From Hung OR, Varvel JR, Shafer SL, et al: Thiopental pharmacodynamics. II. Quantitation of clinical and electroencephalographic depth of anesthesia. Anesthesiology 77:237, 1992.)

Hung and colleagues^[99] proposed a conceptual approach to examining intravenous anesthetic pharmacodynamics that involved clinical measures. A computer-driven infusion pump using a pharmacokinetic model of thiopental disposition was used to rapidly achieve and then maintain constant thiopental plasma concentrations in 26 surgical patients. A low (10 to 40 µg/mL) and then high (40 to 90 µg/mL) constant thiopental plasma concentration was achieved and then maintained for 10 minutes. After allowing 5 minutes for blood-brain equilibration, the following sequential clinical stimuli were applied to the patients at 1-minute intervals: verbal command, 50-Hz electrical tetanus, trapezius muscle squeeze, laryngoscopy, and laryngoscopy followed by intubation. Purposeful movement response was used as the measure of clinical response. Logistic regression was used to relate the measured constant thiopental plasma concentration to the presence or absence of movement and to estimate the Cp₅₀ value. Figure 31-15 displays the thiopental plasma concentration versus response/no response in the data, along with the logistic regression characterization. Prevention of movement during intubation required significantly higher thiopental concentrations (78.8 µg/mL) relative to laryngoscopy (50.7 µg/mL), trapezius muscle squeeze (39.8 µg/mL), electrical tetanus (30.3 µg/mL), and verbal responsiveness (15.6 µg/mL). These data confirm the utility of purposeful movement as a measure of the depth of anesthesia for intravenous anesthetics and demonstrate that

1245

Kazama and colleagues^[26] examined the pharmacodynamics of propofol alone and then of propofol with fentanyl by using the defined noxious stimuli/movement responses that Hung and associates^[99] used for thiopental and the hemodynamic methodology developed by Zbinden and colleagues ^[92] for isoflurane. Kazama and coworkers found propofol Cp₅₀ values for the following defined stimuli: loss of verbal responsiveness, 4.4 µg/mL; electrical tetanus, 9.3 µg/mL; laryngoscopy, 9.8 µg/mL; skin incision, 10.0 µg/mL; and intubation, 17.4 µg/mL. With the addition of a steady-state fentanyl plasma concentration of 1 or 3 ng/mL, there was only a minimal decrease in the propofol Cp₅₀ for loss of verbal responsiveness: 11% for 1 ng/mL and 17% for 3 ng/mL. For the other, more intense noxious stimuli (tetanus, laryngoscopy, skin incision, and intubation), a much greater decrease in propofol Cp₅₀ occurred. Fentanyl, 1 ng/mL, decreased Cp₅₀ values 31% to 34%, whereas fentanyl, 3 ng/mL, decreased Cp₅₀ values 50% to 55%. Further increases in fentanyl plasma concentrations did not lead to additional decreases in propofol Cp₅₀ values. Kazama and colleagues found that the systolic blood pressure response to propofol alone was profound, similar to what Zbinden and coauthors^[92] described. The addition of fentanyl to propofol attenuated the systolic blood pressure increases in a dose-dependent manner. The conceptual methodology used by Kazama and associates allows for characterization of anesthetic depth with multiple, defined noxious stimuli. The combination of the hypnotic anesthetic propofol with the analgesic effects of fentanyl allows a clinically successful anesthetic to be given to patients with quantitative methodology.

Quantitation of the clinical depth of anesthesia for the combination of an opioid and an intravenous anesthetic was reported by Vuyk and colleagues.^[100] The pharmacodynamic methodology developed by Ausems and coworkers^[39] was used to examine the interaction of a constant plasma concentration of propofol (in place of nitrous oxide) as alfentanil was titrated to clinical responses. Propofol had a significant interaction with alfentanil, thus decreasing the total dose of alfentanil and the alfentanil plasma concentrations needed for adequate anesthesia. Table 31-4 indicates how the total dose of alfentanil needed decreased from 22.8 mg with 66% nitrous oxide to 10.3 mg with a steady-state propofol plasma concentration of 4 µg/mL. The mechanism of this decreased dose requirement was a marked potentiation of alfentanil such that alfentanil plasma concentrations needed to achieve the same degree of pharmacologic effect were two to four times lower with propofol.

In a subsequent study, Vuyk and coworkers used this same methodology, but they randomized the surgical subjects to receive a range of constant propofol blood concentrations.^[24] With blood propofol concentrations increasing from 2 to 10 µg/mL, the alfentanil Cp₅₀ decreased from 170 to 25 ng/mL for laryngoscopy, from 280 to

TABLE 31-4 -- Interaction of propofol and nitrous oxide with alfentanil in lower abdominal surgery

	Alfentanil Cp50 (ng/mL) in Conjunction with
Event	Propofol	Nitrous Oxide
Steady-state propofol or nitrous oxide concentration	4.0 ± 0.6 µg/mL (range, 3.2–4.9)	66%
Intubation	92 ± 20 *	429 ± 42 *
Skin incision	55 ± 16 *	101 ± 16 *
Intra-abdominal dissection	66 ± 38 †	206 ± 65 †
Total alfentanil dose (mg)	10.3 ± 6.1 †	22.8 ± 6.4 †
Modified from Vuyk J, Lim T, Engbers FHM, et al: Pharmacodynamics of alfentanil as a supplement to propofol or nitrous oxide for lower abdominal surgery in female patients. Anesthesiology 78:1036, 1993.

*±SE.
†±SD.

In clinical practice, intravenously administered anesthetic drugs are frequently combined with other drugs that provide additional analgesia (opioids, nitrous oxide, potent inhaled anesthetics). As indicated earlier, large intravenously administered doses of thiopental or propofol are less than effective in eliminating the hemodynamic response to relevant clinical stimuli such as laryngoscopy and intubation. ^{[26] [93] [94]} Fentanyl decreases the anesthetic requirement for thiopental or propofol by providing antinociceptive effects that the intravenous hypnotics do not provide.^{[26] [102]} Clinically, the hemodynamic response to laryngoscopy, intubation, or skin incision is most commonly used to assess depth of anesthesia. The use of muscle relaxants to ease endotracheal intubation precludes use of the movement response. Because laryngoscopy and intubation are single events, if clinical depth is inadequate (e.g., in the event of a profound hemodynamic response), additional intravenous anesthetics, opioids, or maintenance anesthetic drugs are rapidly administered. When precise hemodynamic control becomes important (as in coronary artery disease), larger doses of opioids are used instead of intravenously administered anesthetics.