Pharmacokinetics of Cisatracurium in Patients Receiving Nitrous Oxide/Opioid/Barbiturate Anesthesia

After obtaining approval from The New York Hospital--Cornell Medical Center Committee on Human Rights in Research, 20 patients, ranging in age from 23 to 65 yr, consented to participate in the study. All patients were ASA physical status 1 or 2 and were scheduled to undergo elective minor surgical procedures under general anesthesia. Patients were free of neuromuscular, hepatic, renal, pulmonary, or cardiovascular disease. None had a history of unusual susceptibility to neuromuscular blocking agents or of malignant hyperthermia. No patient had received antibiotics, with the exception of penicillin, cephalosporins, or tetracyclines, within the 48 h before enrollment in the study. None had received anticonvulsants, antidepressants, or antihistamines during the week before enrollment in the study. Female patients of childbearing potential were not pregnant as documented by a negative urine pregnancy test the morning of surgery, were using an approved method of birth control, or were scheduled to undergo a hysterectomy.

On consenting to participate in the study, ten patients were assigned to receive 2 times the ED⁹⁵of cisatracurium (0.1 mg *symbol* kg sup -1) and ten patients were assigned to receive 4 times the ED⁹⁵of cisatracurium (0.2 mg *symbol* kg sup -1).

General anesthesia was induced with intravenous midazolam (20-85 micro gram *symbol* kg sup -1), fentanyl (3-7 micro gram *symbol* kg sup -1), and thiopental (2-10 mg *symbol* kg sup -1). The patients' tracheae were intubated before the administration of cisatracurium and anesthesia was maintained with oxygen (30%), nitrous oxide (70%) and additional doses of intravenous fentanyl, midazolam, and thiopental as required to maintain an adequate depth of anesthesia. Mechanical ventilation was adjusted to maintain normocapnea.

Esophageal temperature was maintained between 35.1 degrees Celsius and 36.4 degrees C with warmed intravenous fluids, blankets, and gas humidifiers. After induction of anesthesia, a second large-bore (16-G) intravenous catheter was inserted for venous blood sampling in the arm not containing the initial intravenous catheter.

After a 3-min period of stabilization of the twitch response to neuromuscular stimulation, patients received either 2 (n = 10) or 4 (n = 10) times the ED⁹⁵of cisatracurium (0.1 or 0.2 mg *symbol* kg sup -1, respectively) into the tubing of a rapidly flowing intravenous infusion. Once neuromuscular function had recovered to a train-of-four ratio of greater or equal to 0.70, further neuromuscular block was provided, if required, with intermittent bolus doses of vecuronium (1-4 mg).

Five-milliliter venous blood samples were collected before and at 2, 4, 6, 8, 10, 12, 15, 20, 25, 30, 45, 60, 90, 120, 240, and 480 min after the administration of cisatracurium. Each sample was immediately transferred to a chilled 5-ml vacutainer tube (Becton Dickinson, Franklin Lakes, NJ) containing ethylenediaminetetraacetic acid and centrifuged. Within 3 min of the beginning of sample collection, 1 ml of plasma was decanted, mixed thoroughly with 4 ml of 15 mM sulfuric acid and placed on ice. The samples were stored frozen and later analyzed with a high-performance liquid chromatographic method with fluorescence detection for cisatracurium, laudanosine, and the monoquaternary alcohol metabolite.* The lower limit of quantitation was 10 ng *symbol* ml sup -1 for each analyte. The coefficients of variation for cisatracurium were less than 14% at concentrations between 10 and 2000 ng *symbol* ml sup -1 for each analyte. The presence of the monoquaternary acid metabolite was assessed qualitatively by examination of chromatograms.

Noncompartmental. Noncompartmental analysis was performed using Microsoft Excel 4.0 (Redmond, WA). Cisatracurium plasma concentration versus time data were analyzed by log-linear regression for estimation of the terminal slope (beta). The area under the plasma cisatracurium concentration-time curve from time 0 until the last measured concentration was calculated using the linear trapezoidal rule. The terminal slope was used to extrapolate the area under the plasma cisatracurium concentration versus time curve from the last measured time point to infinity. The total area under the cisatracurium concentration-time curve (AUC⁰→ infinity) was calculated as the sum of the area under the curve from time 0 to the last measured concentration plus the area under the curve extrapolated to time infinity. Cisatracurium clearance was calculated using the equation Cl = Dose/AUC⁰→ infinity, where Dose is the intravenous bolus dose of cisatracurium. The steady-state volume of distribution (V^ss) was calculated as V^ss= Cl *symbol* AUMC⁰→ infinity/AUC⁰→ infinity, where AUMC⁰→ infinity is the area under the first moment curve from time 0 to time infinity (AUMC sub 0 → infinity = AUMC⁰→ t + Cp/beta²+ t sup * Cp/beta) and Cp is the last measured concentration and t is the time of the last measured concentration. The terminal elimination phase half-life of cisatracurium was calculated as the In(2) divided by the terminal slope (beta).

Compartmental. A nontraditional two-compartment model (i.e., with elimination from both the central compartment and the peripheral compartment) was fitted to the plasma cisatracurium concentration-time data from individual patients using NONLIN. [7]The rate constant (k sub 20) describing elimination from the peripheral compartment could not be estimated independently and was, therefore, fixed to 0.0237 min sup -1, the average rate constant describing the in vitro degradation of cisatracurium in plasma from nine healthy volunteers. [6]The NONLIN user-defined subroutine was written (with k²⁰fixed at 0.0237 min sup -1) to generate estimates of Vc and the micro rate constants k¹⁰, k sub 12, and k²¹. The Cl and V^sswere then calculated using the following two equations: Equation 1.

Metabolite Pharmacokinetics. The pharmacokinetics of laudanosine and the monoquaternary alcohol metabolite were evaluated using noncompartmental methods. The maximum concentration (C^max) and the time of occurrence of C^max(T^max) were determined by direct inspection of the plasma concentration-time data for each of the metabolites. The elimination half-life, AUC⁰→ infinity, and the ratio of the AUC⁰→ infinity for each metabolite to AUC⁰→ infinity for cisatracurium were calculated.

Demographic data (age, weight, height, and gender composition), end-tidal carbon dioxide concentration, and temperature data were compared between the two study groups with Student's t test, Fisher's exact test, or Wilcoxon's rank sum test as appropriate. P less or equal to 0.05 was considered to be statistically significant.

Frequency distribution plots and box plots of data were examined to assess the distribution of pharmacokinetic parameter value estimates. Log (ln) transformations were performed on data where appropriate. Values of parameters suspected as being outliers were tested using the Dixon test. [7]Those found to be outliers were omitted from data analysis. Student's t test for unpaired data with unequal variance was performed using Minitab (Minitab Inc., State College, PA) to examine differences in pharmacokinetic values between the dose groups receiving 2 times the ED⁹⁵and 4 times the ED⁹⁵.

As shown in Table 1, the study groups were similar in terms of age and weight. The patients receiving 4 times the ED⁹⁵of cisatracurium tended to be taller than those receiving 2 times the ED⁹⁵of relaxant. All patients studied were within 30% of their ideal body weight. All of the patients receiving 4 times the ED⁹⁵of cisatracurium were men, whereas 40% of the patients receiving the smaller dose of drug were women. This difference in gender composition of the two study groups was not statistically significant (Table 1). The end-tidal carbon dioxide concentration for patients receiving either 2 or 4 times the ED⁹⁵of cisatracurium was not significantly different (31.6+/-0.9 and 30.2+/-1.9 mmHg, respectively). Similarly, patient temperature was the same in the two study groups (35.5+/-0.3 vs. 35.7 +/-0.7 degree C, in the 2 and 4 times the ED⁹⁵groups, respectively).

The cisatracurium plasma concentration versus time data are presented in Figure 2. The dose-normalized mean concentration versus time data are presented in Figure 3. The plasma concentrations of cisatracurium declined in a biexponential fashion. The highest observed cisatracurium concentration occurred at 2 min after dosing for each of the patients (the earliest sampling point). Dose-normalized cisatracurium concentration-time curves were similar, although dose-normalized concentrations were slightly higher in patients receiving the 0.2 mg *symbol* kg sup -1 dose (mean +/-SD dose-normalized C^max: 1400+/-663 and 1760 +/-465 ng *symbol* ml sup -1 for the 0.1 and 0.2 mg *symbol* kg sup -1 dose groups, respectively). This difference did not reach statistical significance (P = 0.18).

The pharmacokinetic parameter values (determined using compartmental and noncompartmental methods) for cisatracurium are presented in Table 2. Whether determined using compartmental or noncompartmental methods, there were no significant differences between the dose groups in Cl, V^ss, and elimination half-life. The power to detect a 20% difference in Cl was 0.96. For all other pharmacokinetic parameters, the power ranged from 0.83 to 1.00. The V^ss(calculated using noncompartmental methods) was underestimated by 17.2+/- 11.9% and 20.2+/-14.4% in the 2 and 4 times the ED⁹⁵groups, respectively. The mean+/-SD AUC⁰→ infinity values (calculated using noncompartmental methods) were 21.07+/-4.05 and 46.81+/-6.66 min *symbol* micro gram sup -1 *symbol* ml sup -1, for the groups receiving 2 and 4 times the ED⁹⁵of cisatracurium, respectively. The ratio of this value for the two groups, therefore, was 2.23. Comparison of the dose-normalized AUC⁰→ infinity between patient groups did not reach statistical significance (P = 0.17).

The laudanosine plasma concentration-time data are shown in Figure 4. The calculated pharmacokinetic variables for laudanosine are presented in Table 3. The C^maxof laudanosine ranged from 13 to 86 ng *symbol* ml sup -1 in the 2 times the ED⁹⁵group and from 66 to 174 ng *symbol* ml sup -1 in the 4 times the ED⁹⁵group. The T^maxranged from 2 to 90 min and from 2 to 60 min after the administration of 2 or 4 times the ED⁹⁵of cisatracurium, respectively.

The monoquaternary alcohol plasma concentration-time profile is shown in Figure 5. The pharmacokinetics of the monoquaternary alcohol are summarized in Table 4. The C^maxfor the monoquaternary alcohol metabolite was 101+/-27 ng *symbol* ml sup -1 and 253+/- 51 ng *symbol* ml sup -1 after administration of the 2 and 4 times the ED sub 95 doses, respectively. The T^maxoccurred from 2 to 25 min after a dose 2 times the ED⁹⁵and from 2 to 15 min after a dose 4 times the ED⁹⁵.

The pharmacokinetics of cisatracurium are independent of dose after doses of 0.1 and 0.2 mg *symbol* kg sup -1 (2 and 4x ED⁹⁵, respectively). Plasma cisatracurium and metabolite concentration data collected in healthy adult patients receiving 0.1 or 0.2 mg *symbol* kg sup -1 doses of cisatracurium are consistent with data from in vitro experiments that suggest that Hofmann elimination is an important pathway for the elimination of cisatracurium. First, laudanosine and the monoquaternary alcohol metabolite, but not the monoquaternary acid metabolite, were detected in plasma from patients receiving 0.1 or 0.2 mg *symbol* kg sup -1 cisatracurium. The absence of the monoquaternary acid metabolite in plasma suggests that hydrolysis by nonspecific plasma esterases is not an important pathway for cisatracurium. Second, values for elimination half-life for cisatracurium in healthy adult patients are similar to the in vitro half-life of cisatracurium in human plasma (mean, 29 min; range, 23-33 min). [6]Therefore, organ-independent Hofmann elimination appears to be a major pathway for the elimination of cisatracurium. Hydrolysis of cisatracurium catalyzed by nonspecific plasma esterases is not an important elimination pathway for cisatracurium. [6].

The pharmacokinetic analyses of cisatracurium are complicated by the finding that cisatracurium undergoes Hofmann elimination, which can occur in the plasma and the tissues. One must assume that elimination occurs only from the central compartment to use noncompartmental pharmacokinetic methods. Violation of this assumption does not affect the estimation of Chlorine, but results in an underestimation of V^ss. [8]For the nontraditional two-compartment model (with elimination from both compartments), one must assume that: (1) the rate constant describing elimination from the peripheral compartment is the same as the average rate constant describing the in vitro degradation of cisatracurium in plasma from nine healthy volunteers, [6]and (2) Hofmann elimination occurs at the same rate in the central and peripheral compartments. These are considered reasonable because (1) Hofmann elimination is pH- and temperature-dependent and the physiologic ranges of pH and temperature are small, (2) the coefficient of variation of the in vitro rate constant in plasma from nine healthy volunteers was small (13%), and (3) the in vitro rate constant does not change substantially when cisatracurium is incubated in buffer or plasma. [6]Violation of the assumptions may have an impact on the estimation of V^ss. Analyses of the data using both methods (noncompartmental and compartmental) have provided insight into the pharmacokinetic characteristics of cisatracurium.

Clearance values were similar when calculated using noncompartmental methods or a nontraditional two-compartment model (with elimination from central and peripheral compartments). This finding is expected because Chlorine is an exit-site-independent parameter (i.e., independent of whether elimination is occurring from only the central or from both the central and peripheral compartments). [8]The V^ssvalues were 17-20% lower when calculated using noncompartmental methods than when calculated using the nontraditional two-compartment model (with elimination from the central and peripheral compartments); this finding is not surprising because V^ssis underestimated when ignoring elimination from the peripheral compartment. [8].

This study demonstrates that the pharmacokinetics of cisatracurium are proportional to dose. This finding is not unexpected because cisatracurium, one of the isomers comprising atracurium, undergoes degradation by Hofmann elimination. Because Hofmann elimination is a nonenzymatic chemical process, one would not anticipate nonlinear metabolism (i.e., a dose-dependent clearance). One would also not anticipate saturable tissue binding because of the limited distribution of cisatracurium in tissues. Small differences (11-12%) in V^ssand Chlorine were observed between patient groups, although these differences were not statistically significant. The 11% smaller V^ssin patients receiving the higher dose can be explained only by saturable tissue binding, a highly unlikely event considering the limited distribution of cisatracurium (i.e., V^ssis not greater than extracellular fluid volume). These small differences more likely represent interpatient variability. The 11% smaller V^sscould account for the 12% lower Cl, a relationship that has been reported for atracurium. [9]Most importantly, even with adequate power, no statistically significant differences between groups were detected in any pharmacokinetic parameters. Thus, as with atracurium, [10]the pharmacokinetics of cisatracurium are proportional to dose.

The clearance of 2 and 4 times the ED⁹⁵of cisatracurium (5.28 and 4.66 ml *symbol* kg sup -1 *symbol* min sup -1, respectively) is consistent with values reported previously for atracurium (5.5 [11]and 6.1 [12]ml *symbol* kg sup -1 *symbol* min sup -1) and for the intermediate-duration steroidal relaxants vecuronium (ranging from 5.3 to 5.6 ml *symbol* kg sup -1 *symbol* min sup -1) [13,14]and rocuronium (ranging from 2.9 to 5.0 ml *symbol* kg sup -1 *symbol* min sup -1) [15,16]in young adults free of hepatic or renal disease receiving 2 times the ED⁹⁰of either relaxant.

Hofmann elimination is a spontaneous temperature-dependent decomposition of the parent compound resulting in the formation of laudanosine. Previous studies have demonstrated that as a result of in vitro decreases of temperature of 14 degrees C, from 37 degrees Celsius to 23 degrees C, the in vitro plasma half-life of atracurium increases from 18 to 49 min. [17]Furthermore, hypothermic patients undergoing cardiopulmonary bypass have a lower atracurium dose requirement, [18]presumably because of an altered response of muscles or because of the relaxant's lower clearance and prolonged elimination half-life. Patients in both study groups were the same temperature, eliminating this as a factor affecting the clearance of cisatracurium.

Hofmann elimination is also a base catalyzed process. In cats, average increases of pH of 0.32-0.40, with either hyperventilation or infusion of sodium bicarbonate (Na²CO³), decreased the depth of atracurium-induced neuromuscular block by approximately twofold. [19]Although pH was not specifically measured in the patients in this study, none had evidence of pulmonary disease and mean end-tidal carbon dioxide concentrations in all the patients in each group were indistinguishable. Furthermore, no patient had evidence of end-organ disease that could have affected acid-base status, such as renal disease.

In compliance with the standards of the Committee on Human Rights in Research at The New York Hospital-Cornell Medical Center, the study was undertaken using venous, rather than arterial, blood. Donati et al. [20]reported on pharmacokinetics and pharmacodynamics of atracurium determined with arterial and venous blood. While they found that relaxant concentrations were greater than 50% greater in the arterial than in the venous blood for the first 3 min after administration of atracurium, the concentration of atracurium in venous samples was approximately 90% of that measured in arterial samples by 20 min. In contrast, the Chlorine of cisatracurium in patients in whom arterial blood was collected was only 4% higher than the Chlorine in patients from other studies in whom venous blood was collected. [21]Most importantly, venous blood was collected in all patients in the current study and therefore is consistent between patient groups.

Although we were able to detect the monoquaternary alcohol as a metabolite of cisatracurium, cisatracurium does not appear to undergo ester hydrolysis. [6]This is in contrast to the elimination of atracurium from the plasma, which occurs by a number of mechanisms including ester hydrolysis as well as Hofmann elimination and organ-dependent elimination. [5,22]Cisatracurium undergoes Hofmann elimination to form laudanosine and a monoquaternary acrylate. The monoquaternary acrylate most likely undergoes hydrolysis by nonspecific plasma esterases to form a monoquaternary alcohol. [6]The lack of ester hydrolysis of cisatracurium may explain our inability to detect monoquaternary acid in any plasma sample.

Although the elimination half-life of laudanosine reported in this study, 3.6 h, is not very different from values calculated previously for laudanosine after the administration of atracurium, 3.3 h, [11]the maximum concentration of laudanosine measured after a 2 times the ED⁹⁵dose of cisatracurium, 38 ng *symbol* ml sup -1, is far less than that reported after administration of approximately 1.5 times the ED⁹⁵of atracurium, 190 ng *symbol* ml sup -1. [11]While direct comparisons are not possible because these values were obtained in different studies, a lower maximum plasma concentration of laudanosine after administration of cisatracurium can be explained, at least in part, by relative potencies. Cisatracurium is 3.5 times more potent than atracurium and thus, fewer molecules are administered with an initial bolus dose. As a result of this, we speculate that, on long-term administration of cisatracurium, by infusion or repeat bolus doses, the maximum plasma concentrations of laudanosine should likely be lower than those observed after administration of atracurium.

The C^maxfor laudanosine ranged from 13 to 86 ng *symbol* ml sup -1 and from 66 to 174 ng *symbol* ml sup -1 in healthy adult patients receiving 0.1 and 0.2 mg *symbol* kg sup -1 cisatracurium, respectively. These concentrations are approximately 30-fold and > 100-fold lower than the plasma laudanosine concentrations associated with cerebral excitatory effects in rabbits [23]or dogs, [24]respectively. In addition, no clinical evidence of cerebral irritation or excitation attributable to cisatracurium has been reported in surgical or in intensive care patients. [25].

In summary, the pharmacokinetics of cisatracurium are independent of dose after doses of 0.1 and 0.2 mg *symbol* kg sup -1 (2 and 4 times the ED⁹⁵, respectively) of cisatracurium. This finding was expected due to its degradation by Hofmann elimination. The pharmacodynamic results are consistent with the dose-proportional pharmacokinetic results from this study. Increasing the dose has a predictable effect on the duration of block. The rate of spontaneous recovery is independent of dose up to doses of 8x ED⁹⁵and duration of infusion. [4].

The authors thank Steve Quessy, Ph.D., Martha Abou-Donia Ph.D., Luann Phillips, Charles James, and Carrie McClung, for their technical assistance and their comments regarding the study design, analysis, and manuscript; and Irusia Kocka, for her editorial assistance in the preparation of this manuscript.

*James C: Personal communication. 1995.