Free
Clinical Science  |   January 1999
Influence of Renal Failure on the Pharmacokinetics and Neuromuscular Effects of a Single Dose of Rapacuronium Bromide 
Author Notes
  • (Szenohradszky) Assistant Professor of Anesthesia. Current position: Assistant Professor, Department of Anesthesia, University of Southern California.
  • (Caldwell) Associate Professor of Anesthesia.
  • (Wright) Assistant Professor of Anesthesia. Current position: Senior Lecturer, University of Newcastle-upon-Tyne.
  • (Brown, Lau) Staff Research Associate.
  • (Luks) Research Fellow.
  • (Fisher) Professor of Anesthesia and Pediatrics.
Article Information
Clinical Science
Clinical Science   |   January 1999
Influence of Renal Failure on the Pharmacokinetics and Neuromuscular Effects of a Single Dose of Rapacuronium Bromide 
Anesthesiology 1 1999, Vol.90, 24-35. doi:
Anesthesiology 1 1999, Vol.90, 24-35. doi:
A RECENT trial suggests that rapacuronium bromide (ORG9487; Organon Inc., W. Orange, NJ, and Organon Teknika, Boxtel, the Netherlands), a new nondepolarizing muscle relaxant, has a rapid onset, possibly comparable to that of succinylcholine, and that recovery after a bolus dose is more rapid than that of other nondepolarizing muscle relaxants. [1] Because patients with renal failure historically have had residual paralysis from nondepolarizing muscle relaxants, [2] clinical trials of each new muscle relaxant have evaluated the influence of renal failure on its time course and pharmacokinetic characteristics. [3-8] Typically, these studies have been performed in patients receiving kidney transplants, [5,7,8] so the elimination of the muscle relaxant by the newly transplanted kidney may have confounded the analyses. In the current study, we administered a single dose of rapacuronium to patients with renal failure and to healthy volunteers to determine its neuromuscular and cardiovascular effects and its pharmacokinetic characteristics. We selected patients with renal failure who underwent non-transplant surgery to eliminate the possibility that a newly transplanted kidney might eliminate rapacuronium and thus confound our analyses.
Methods
After obtaining approval from our local institutional review board and informed consent from each participant, we studied 10 healthy paid volunteers (classified as American Society of Anesthesiologists physical status 1) and 10 patients with chronic renal failure (Table 1). Volunteers underwent anesthesia but no surgery, whereas patients with renal failure underwent minor non-transplant surgical procedures, such as placement of peripheral vascular shunts (n = 4) or peritoneal dialysis catheters (n = 4) or establishment of central venous access (n = 2). One patient with renal failure received phenytoin chronically (plasma concentration before induction of anesthesia, 8 mg/l [therapeutic range, 10-20 mg/l]). Because of the concern that phenytoin might influence the pharmacokinetics, the neuromuscular effects of rapacuronium, or both, [9] data from that patient are reported separately and are not included in any neuromuscular or pharmacokinetic comparisons. The volunteers and remaining patients were not taking any other drugs that might have influenced the neuromuscular response to rapacuronium.
Table 1. Demographic Data for 10 Healthy Volunteers and 10 Patients with Renal Failure
Image not available
Table 1. Demographic Data for 10 Healthy Volunteers and 10 Patients with Renal Failure
×
After an 8-h fast, an intravenous catheter was placed in an upper extremity of the volunteers and patients. Patients received premedication with 1-2 mg midazolam. After administration of 5 [micro sign]g/kg fentanyl, anesthesia was induced with 2-3 mg/kg propofol and the trachea was intubated without the aid of a muscle relaxant. Anesthesia was maintained with propofol infused at approximately 150 [micro sign]g [middle dot] kg-1[middle dot] min-1. If needed for clinical purposes (e.g., to treat hypertension), isoflurane could be administered. An intravenous catheter was placed in the contralateral arm or in the external jugular vein to sample blood. Ventilation was controlled to maintain normocapnia (end-tidal carbon dioxide tension of 30-35 mmHg). The esophageal temperature was maintained at more than 36 [degree sign]C by forced-air warming. Except for two volunteers, arterial blood pressure (Dinamapp; Critikon, Tampa, FL), heart rate, pulse oximetry (N200; Nellcor, Hayward, CA), and end-tidal carbon dioxide tension (Ultima; Datex, Helsinki, Finland) were recorded before and every minute for 5 min after rapacuronium administration, then every 5 min for 25 min. Cutaneous signs of histamine release were sought and the lungs were auscultated regularly to check for wheezing.
After loss of consciousness, a single 5-s 50-Hz supramaximal tetanic stimulus [10] was applied to the ulnar nerve via subcutaneous 27-gauge needles placed at the wrist followed by supramaximal train-of-four (TOF) stimuli every 12 s. The mechanical twitch response of the adductor pollicis muscle was measured with a calibrated force displacement transducer and amplified. Twitch tension was digitized, displayed, and recorded on-line. Evoked twitch tension was also recorded on a strip chart. The first twitch response of each TOF (T1) was stable for more than 5 min before rapacuronium was administered. The ratio of the fourth component to the first component of each TOF was determined.
Rapacuronium (1.5 mg/kg) was administered over 5 s into a rapidly flowing intravenous line. In volunteers and six patients, neuromuscular function was recorded until T1 recovered completely and the TOF ratio was more than 0.9. For the four patients in whom surgery ended before neuromuscular function was recovered completely, 1 mg atropine (n = 3) or 0.4 mg glycopyrrolate (n = 1) and 2 to 3.5 mg neostigmine were administered when T1 was 71-91% of control; neuromuscular function was recorded until T1 was completely recovered and the TOF ratio was more than 0.9.
Venous blood (5 ml) was sampled before and 3, 7, 10, 20, 30, 45, 60, 75, 90, 120, 150, 180, 210, 240, 300, 360, 420, and 480 min after rapacuronium administration. Except for samples obtained more than 2 h after administration of rapacuronium, all blood samples were obtained during a period of less than 10 s; the midpoint of the sampling interval was recorded. For healthy volunteers, urine voided spontaneously was collected during the following time periods after rapacuronium administration: 0-2, 2-4, 4-6, 6-9, 12-18, 18-24, 24-36, and 36-48 h. The patients were anuric. Sodium dihydrogen phosphate was added immediately to blood and urine samples to prevent rapacuronium degradation. Blood was centrifuged within 30 min of sampling, and plasma and urine were stored at -20 [degree sign]C. Concentrations of rapacuronium and its primary 3-OH metabolite, ORG9488, were determined by Corning Hazelton Labs (Hazelton, WI) using an HPLC-MS technique. The assay is linear for concentrations greater than 2 ng/ml for rapacuronium and ORG9488, and the coefficient of variation is less than 11% for rapacuronium and less than 32% for ORG9488.
We determined the pharmacokinetic characteristics of rapacuronium and ORG9488 using a population approach, mixed-effects modeling (NONMEM). We simultaneously analyzed plasma or urine concentration values for all participants to determine typical values for the pharmacokinetic parameters, standard errors of these estimates, and interindividual variability (see appendix 1). In addition, we determined the influence of renal function and other covariates (e.g., demographic characteristics and preoperative laboratory values) on the pharmacokinetic parameters. [11] 
For rapacuronium, two-compartment models included the parameters clearance (Cl), distributional clearance (Cldistribution), and volumes of the central and peripheral compartments (V1and V2, respectively). Three-compartment models included, in addition, a slow distributional clearance (Clslow) and volume of the deep peripheral compartment (V3). We also determined the fraction of rapacuronium eliminated via the kidneys (ƒrenal). Assumptions used in this modeling are described in appendix 1.
For ORG9488, one-compartment models included the parameters clearance (Cl) and volume of distribution (V). Two-compartment models included the parameters clearance (Cl), distributional clearance (Cldistribution), and volumes of the central and peripheral compartments (V1and V2, respectively). Three-compartment models included the additional parameters slow distributional clearance (Clslow) and volume of the deep peripheral compartment (V3). Assumptions used in these models are described in appendix 1. Because urinary recovery of the administered dose of rapacuronium as either rapacuronium or ORG9488 was not complete, we could not estimate the fraction of the administered dose of rapacuronium that converted to ORG9488 (fmetabolized). In turn, we could not estimate the volume of distribution and clearances for ORG9488; instead, all distribution volumes and clearances for ORG9488 are normalized using f (metabolized). However, because the shape of the plasma concentration versus the time curve for ORG9488 is well described, half-life values for ORG9488 are estimated accurately.
All analyses were performed using a model-building approach. Initially, patients with renal failure were assumed to have the same pharmacokinetic parameters as healthy volunteers. Two- and three-compartment models, weight-normalized and non-weight-normalized, were compared to determine the appropriate structural model. The appropriateness of the error model was determined (appendix 1). After the population analysis was performed for each model, the post hoc step of NONMEM was performed. This Bayesian step determines the parameter estimates for each individual compared with the population estimates. These differences are quantified through NONMEM's [Greek small letter eta](eta) terms. We plotted the resulting values for [Greek small letter eta] against the covariates age, weight, height, gender, group (renal failures vs. normal renal function); against the preoperative values for hematocrit and hemoglobin; and against the serum concentrations of creatinine, bilirubin, aspartate aminotransferase, and alanine aminotransferase. After we added a smoother (lowess, a local regression) to each plot, trends were sought by visual inspection. If we observed a relation between a pharmacokinetic parameter and a covariate, we tested this relation in the model. We accepted additional parameters in the model if they statistically improved the objective function of NONMEM (for P < 0.05, 3.8 units for one additional parameter and 6 units for two; for P < 0.01, 6.6 units for one additional parameter and 9.1 units for two) or improved the quality of fit as assessed visually. We performed some analyses using the “centered” option of the first-order conditional estimate method of NONMEM. We calculated half-lives using standard formulas; for three-compartment models, we calculated half-lives iteratively.
We determined the magnitude of maximal twitch depression (expressed as the percentage of depression from the predrug control value) and time from administration of rapacuronium to 90% and maximum twitch depression. We determined the time from administration of rapacuronium to 10%, 25%, and 75% recovery of T1 (based on the T1 value obtained after complete recovery of neuromuscular function); the time from 25% to 75% recovery of T1; and the time to recovery of the TOF ratio to 70% and 80%. Values for patients with renal failure were compared with those from healthy volunteers using the Mann-Whitney U test or the Student's t test for unpaired data. For each group, values for heart rate and systolic, diastolic, and mean blood pressures were analyzed using repeated measures analysis of variance and Dunnett's test. P < 0.05 was considered significant.
Results
Twitch depression was complete except in one patient in whom 88% twitch depression developed (Table 2). Twitch depression at 1 min was less in patients than in volunteers. The time to 90% and maximum twitch depression; twitch tension recovery to 10%, 25%, and 75% of the control value; twitch tension recovery from 25% to 75% of control; and TOF recovery of 70% and 80% did not differ between groups.
Table 2. Magnitude of Twitch Depression and Time Course of Neuromuscular Effects of Rapacuronium in Healthy Volunteers and in Patients with Renal Failure
Image not available
Table 2. Magnitude of Twitch Depression and Time Course of Neuromuscular Effects of Rapacuronium in Healthy Volunteers and in Patients with Renal Failure
×
Plasma concentrations of rapacuronium initially decreased rapidly in both groups (Figure 1). After approximately 60 min, rapacuronium's plasma concentration decreased more slowly in patients with renal failure than in healthy volunteers. For rapacuronium, a three-compartment model was preferred to a two-compartment model (Table 3, models 3 and 4 compared with models 1 and 2), weight normalization improved the objective function (models 2 and 4), and the constant coefficient of error model yielded fits comparable to the error model with two components (model 4 vs. model 5). Permitting the clearance value to differ between the groups markedly improved the quality of the fit (model 6 vs. model 4). Not permitting interindividual variability in V1and Clrapid/Clslowor permitting interindividual variability in clearance to differ between groups did not influence the quality of the fit (models 7 and 8 vs. model 6). However, permitting interindividual variability to differ for each of V2and V3did improve the quality of the fit (model 9 vs. model 6). Permitting each of V (2)(model 10 vs. model 9) and V3(model 11 vs. model 10) to differ between groups improved the quality of fit. Permitting clearance to vary with age and V3to differ with gender also improved the quality of the fit (models 12 and 13 vs. model 11).
Figure 1. Values for the plasma concentration of rapacuronium after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown with a dashed line.
Figure 1. Values for the plasma concentration of rapacuronium after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown with a dashed line.
Figure 1. Values for the plasma concentration of rapacuronium after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown with a dashed line.
×
Table 3. Models Tested for the Pharmacokinetics of Rapacuronium
Image not available
Table 3. Models Tested for the Pharmacokinetics of Rapacuronium
×
Thus the optimal model (Table 4, Figure 2) for rapacuronium had three compartments, and all pharmacokinetic parameters were weight normalized. Clearance was 32% less in patients with renal failure than in the healthy controls. In both groups, clearance decreased 0.909% per year of age compared with the value at age 30 yr. In healthy controls, clearance decreased from 10.25 ml [middle dot] kg-1[middle dot] min-1at age 20 yr to 8.55 ml [middle dot] kg-1[middle dot] min-1at age 40 yr; in patients with renal failure, clearance decreased from 6.98 ml [middle dot] kg-1[middle dot] min-1at age 20 yr to 5.82 ml [middle dot] kg-1[middle dot] min-1at age 40 yr. The steady state distribution volume was 14% less in women than in men. Half-lives varied as a function of renal function, age, and gender. There were no relations between the pharmacokinetic parameters and other covariates. The coefficient of variation of the parameter estimate for the effect of age on clearance was 67%, suggesting that age might not influence clearance. However, fixing the parameter estimate for the effect of age on clearance to different values showed that the decrease in clearance per year of age was at least 0.4%(appendix 1).
Table 4. Parameter Estimates* of Three-Compartment Pharmacokinetic Model for Rapacuronium (Model 13) for Healthy Volunteers and Patients with Renal Failure
Image not available
Table 4. Parameter Estimates* of Three-Compartment Pharmacokinetic Model for Rapacuronium (Model 13) for Healthy Volunteers and Patients with Renal Failure
×
Figure 2. The quality of fit of the pharmacokinetic model to the values for plasma concentration of rapacuronium. The x axis shows time (in minutes) after administration of rapacuronium. The y axis shows the ratio of the measured concentration of rapacuronium to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10) and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected in that the post hoc model permits interindividual variability, but the population model does not.
Figure 2. The quality of fit of the pharmacokinetic model to the values for plasma concentration of rapacuronium. The x axis shows time (in minutes) after administration of rapacuronium. The y axis shows the ratio of the measured concentration of rapacuronium to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10) and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected in that the post hoc model permits interindividual variability, but the population model does not.
Figure 2. The quality of fit of the pharmacokinetic model to the values for plasma concentration of rapacuronium. The x axis shows time (in minutes) after administration of rapacuronium. The y axis shows the ratio of the measured concentration of rapacuronium to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10) and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected in that the post hoc model permits interindividual variability, but the population model does not.
×
In one volunteer, the initial aliquots of urine (0-4 h) were discarded accidentally. In the remaining healthy volunteers, urinary recovery of rapacuronium ranged from 5.8% to 11.4%. The typical value for ƒrenalfor the 10 volunteers was 7.9%(standard error, 0.6%). In the nine volunteers with complete urine collections, more than 90% of the rapacuronium recovered in urine appeared within the first 9 h after its administration.
The plasma concentrations of ORG9488 peaked immediately after administration in healthy volunteers, decreased rapidly, peaked again at approximately 100 min, and decreased monotonically during the rest of the 8-h sampling period (Figure 3). In patients with renal failure, plasma concentrations of ORG9488 reached a plateau at approximately 100 min at 200-400 ng/ml and changed minimally during the rest of the sampling period. For ORG9488, a two-compartment model was preferred to a one-compartment model; there was no additional improvement with a three-compartment model (Table 5, models 14-16). The constant coefficient of error model yielded fits comparable to the error model with two components (model 15 vs. model 17). Permitting clearance and interindividual variability in clearance to differ between groups improved the quality of the fit (model 18 vs. model 17); however, this model demonstrated a bias, necessitating the use of NONMEM's centered option (model 19). Additional models in which Cldistributionhad both a constant and weight-normalized component (model 20), Cldistributionvaried with age (model 21), or Clrenalvaried with gender (model 22) did not improve further the quality of the fit. A model in which interindividual variability was not permitted for V1(model 23) yielded the same fit as model 19. Permitting an administered dose of ORG9488 did not improve the quality of the fit of the model (model 24 vs. model 23). Thus the optimal model (Table 6, Figure 4) for ORG9488 had two compartments, and all pharmacokinetic parameters were weight normalized. Clearance was 85% less in patients with renal failure than in healthy volunteers. In the nine volunteers with complete urine collections, urinary recovery of ORG9488 ranged from 3.3% to 6.8% of the administered dose of rapacuronium. There were no relations between the pharmacokinetic parameters and other covariates.
Figure 3. Values for the plasma concentration of ORG9488 after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown by a dashed line.
Figure 3. Values for the plasma concentration of ORG9488 after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown by a dashed line.
Figure 3. Values for the plasma concentration of ORG9488 after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown by a dashed line.
×
Table 5. Models Tested for the Pharmacokinetics of ORG9488
Image not available
Table 5. Models Tested for the Pharmacokinetics of ORG9488
×
Table 6. Parameter Estimates* of Two-compartment Pharmacokinetic Model for ORG9488 (Model 23)
Image not available
Table 6. Parameter Estimates* of Two-compartment Pharmacokinetic Model for ORG9488 (Model 23)
×
Figure 4. The quality of fit of the pharmacokinetic model to the values for plasma concentration of ORG9488. The x axis is time (in minutes) after administration of rapacuronium. The y axis is the ratio of the measured concentration of ORG9488 to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10), and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected because the post hoc model permits interindividual variability, but the population model does not.
Figure 4. The quality of fit of the pharmacokinetic model to the values for plasma concentration of ORG9488. The x axis is time (in minutes) after administration of rapacuronium. The y axis is the ratio of the measured concentration of ORG9488 to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10), and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected because the post hoc model permits interindividual variability, but the population model does not.
Figure 4. The quality of fit of the pharmacokinetic model to the values for plasma concentration of ORG9488. The x axis is time (in minutes) after administration of rapacuronium. The y axis is the ratio of the measured concentration of ORG9488 to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10), and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected because the post hoc model permits interindividual variability, but the population model does not.
×
In both volunteers and patients, blood pressure decreased and heart rate increased during the 5 min after rapacuronium administration (Table 7). One volunteer and one patient, each of whom had a history of asthma and a recently resolved upper respiratory infection, wheezed approximately 5 min after rapacuronium was administered. In both instances, this manifested initially as an increase in peak airway pressure of more than 10 cm. There were no cutaneous manifestations of histamine release and no significant change in blood pressure or pulse oximetry values. The patient received 2% inspired isoflurane for approximately 5 min. No other bronchodilators were administered, and within 5 min bronchospasm resolved completely in both persons. The patient taking phenytoin as a long-term therapy received 0.5% isoflurane briefly to treat hypertension after neuromuscular function had recovered completely; no other participants received isoflurane.
Table 7. Systolic, Diastolic, and Mean arterial Blood Pressure for 8 of 10 Healthy Volunteers and 10 Patients Given 1.5 mg/kg Rapacuronium
Image not available
Table 7. Systolic, Diastolic, and Mean arterial Blood Pressure for 8 of 10 Healthy Volunteers and 10 Patients Given 1.5 mg/kg Rapacuronium
×
Discussion
We observed that rapacuronium's onset and recovery were similar in patients with renal failure and in healthy volunteers, with the exception of a larger magnitude of twitch depression at 1 min in volunteers compared with patients. These findings indicate that administration of a single dose of rapacuronium to patients with renal failure is not likely to be associated with persistent neuromuscular effects. In addition, the rapid onset of rapacuronium in patients with renal failure may make it useful to facilitate tracheal intubation in these patients, as in healthy patients.
The clearance of rapacuronium was one third less in patients with renal failure than in healthy volunteers. The finding that our patients with renal failure had minimal or no renal function (serum creatinine values were all more than 7 mg/dl) suggests that one third of the elimination of rapacuronium in the healthy volunteers was via the kidney. However, in these healthy persons, less than 12% of administered rapacuronium was recovered from urine and less than 7% was ORG9488. Several explanations may reconcile these findings:
1. The values for the urinary recovery of rapacuronium, ORG9488, or both might be flawed as a result of the volunteers' possibly discarding their samples. However, during the initial 9 h after administration of rapacuronium, when most of the rapacuronium should be eliminated, volunteers were constantly in the presence of an investigator. In addition, the similarity of findings from all volunteers lessens the likelihood of this possibility.
2. Rapacuronium, ORG9488, or both are metabolized in vivo or ex vivo to another compound. However, this process is unlikely to occur ex vivo because urine was acidified immediately when collected. In addition, no other major peaks were identified in the chromatograms.
3. Although less than 12% of rapacuronium is eliminated via the kidneys, renal failure affects hepatic function, thereby decreasing rapacuronium's clearance to a larger extent than explained by ƒrenal. Uremia affects the elimination of other compounds, including propranolol, [12-15] so we consider this the most likely explanation.
In contrast to the small effect of renal failure on rapacuronium's clearance, renal failure was associated with an 85% decrease in the clearance of ORG9488. Two hours after rapacuronium was administered to healthy volunteers, plasma concentrations of rapacuronium and ORG9488 decreased markedly from the peak concentrations associated with paralysis. However, in patients with renal failure, concentrations of ORG9488 decreased minimally during the 8-h sampling period (despite complete recovery of neuromuscular function in a time period similar to that in healthy volunteers). Although little is known about the potency of ORG9488, a study in seven patients suggests that it is slightly more potent than rapacuronium [16] (for which the concentration depressing twitch tension by 50% is approximately 3,500 ng/ml [17]). If so, the concentrations of ORG9488 that we observed in patients with renal failure (200 - 400 ng/ml) are not likely to be of clinical consequence. However, if a patient with renal failure were given repeated doses of rapacuronium, ORG9488 would be likely to accumulate. In turn, recovery will probably be slower than in healthy volunteers. Additional clinical trials are needed to determine if repeated dosing of rapacuronium leads to prolonged neuromuscular effects in patients with renal failure.
We observed that steady state volume of distribution was greater in healthy volunteers than in patients with renal failure, and it was greater in men than in women. The effect of renal failure on volume of distribution differs from our expectation that renal failure might increase the volume of extracellular fluid, and thereby increase the distribution volume of a polar muscle relaxant. The finding of a larger steady state volume of distribution in men than in women was unexpected; however, most previous studies of muscle relaxants have not specifically examined the influence of gender on volume of distribution. Regardless of this, the gender-related difference in volume of distribution influenced half-life values only minimally.
Rapacuronium's low potency leads to its larger dose requirement compared with other nondepolarizing muscle relaxants. Although this low potency is associated with a desirable feature-rapid onset [18,19] -it might also lead to more adverse effects. However, the hemodynamic effects of rapacuronium were minimal in both groups. Bronchospasm occurred in two of our participants, but neither episode manifested within the first several min after rapacuronium was administered. This is in contrast to the rapid onset of adverse respiratory effects that we observed previously with atracurium and vecuronium in persons with asthma. [20] Additional clinical experience is necessary to determine whether persons with asthma are at particular risk with rapacuronium. In addition, the mechanism by which rapacuronium induced bronchospasm in our patients remains to be determined.
During the first several min after rapacuronium is administered, concentrations of its primary metabolite, ORG9488, are larger than might be expected based on the conversion of rapacuronium. These early concentrations of ORG9488 probably result from the administered solution of rapacuronium containing ORG9488 or from immediate conversion of rapacuronium to ORG9488.** However, analyses that permitted an administered dose of ORG9488 failed to improve the quality of the fit.
The results of the current study differ from those with other nondepolarizing muscle relaxants. First, although the fraction of rapacuronium excreted in urine in healthy persons is similar to that for rocuronium, rapacuronium's clearance is 32% less in patients with renal failure and rocuronium's clearance is not decreased by renal failure. [8] Second, the effects of renal failure on elimination of rapacuronium's metabolite were characterized early in the development process, whereas those of vecuronium's potent metabolite (3-desacetylvecuronium) were not. 3-Desacetylvecuronium contributes to both vecuronium's cumulative effect [21] (even with a single dose) and to the prolonged recovery from vecuronium-induced paralysis in critically ill patients. [22] Third, based on findings from a single participant, long-term administration of phenytoin appears not to facilitate recovery from the neuromuscular effects of rapacuronium (Table 2), although it does counteract the renal-failure-induced decrease in clearance of ORG9488 (Figure 3). The lack of effect of phenytoin on the neuromuscular effects of rapacuronium differs from the finding for rocuronium and other nondepolarizing muscle relaxants. [9] Our findings should guide dosing of rapacuronium in patients with renal failure, with or without phenytoin administration.
One limitation of the current study is the small amount of data regarding the recovery of neuromuscular function in patients with renal failure. This resulted because surgery ended before spontaneous recovery was complete in several patients. Additional clinical experience with rapacuronium may reveal a slower recovery profile in patients with renal failure than in healthy controls, even with a single dose.
In conclusion, rapacuronium's onset and recovery are similar in patients with renal failure compared with healthy volunteers, except for a larger magnitude of twitch depression at 1 min in volunteers. Plasma concentrations of rapacuronium are initially similar in patients and volunteers, consistent with the similar neuromuscular effects. However, after 60 min, plasma concentrations of rapacuronium decrease more slowly in patients with renal failure and plasma concentrations of ORG9488 decrease minimally in patients with renal failure, reflecting a decreased clearance of both rapacuronium and ORG9488 in patients with renal failure compared with healthy volunteers. Although the time course of a single dose of rapacuronium is affected minimally by renal failure, recovery from repeated doses likely will be prolonged by renal failure.
Appendix 1. Pharmacokinetic Modeling
Modeling of Interindividual Variability in the Pharmacokinetic Parameters
Interindividual variability was permitted for each of the pharmacokinetic parameters for both rapacuronium and ORG9488. Interindividual variability was modeled by assuming that each individual's pharmacokinetic parameters can be expressed as the sum of the typical value for the population and a factor for that individual. Because interindividual variability tends to assume a log-normal (i.e., skewed) distribution, interindividual variability for clearance was modeled as:Equation 1where Cliis the estimate for clearance for the ithindividual, Cl is the typical value for the population, and [small greek letter eta]iis a random variable with a mean of 0.0. Equation 1can also be written as:Equation 2Interindividual variability for the other pharmacokinetic parameters was modeled in a similar manner. In some models, interindividual variability was assumed to be the same for Cldistributionas for Clslowand for V2as for V3.
Modeling of the Residual Error between Predicted and Measured Concentrations
Residual error between predicted and measured concentrations was initially assumed to have two components-one constant and the other proportional to the predicted plasma concentrations (constant coefficient of variation). This model was chosen because most assays have a constant coefficient of variation when concentrations are significantly greater than the lower limit of quantification of the assay; however, as concentrations approach this limit, error of the assay becomes a larger percentage of the predicted concentration. This error model was compared with error models with only one of these components.
Assumptions Used in the Modeling of the Urinary Elimination of Rapacuronium
The cumulative fraction of rapacuronium eliminated via the kidneys (ƒrenal) was modeled using the following assumptions:
1. Values for the plasma pharmacokinetics of rapacuronium were fixed to the post hoc values determined in the analyses described above (i.e., fitting of the model to urinary excretion of rapacuronium was not permitted to influence the quality of the fit to the plasma concentrations of rapacuronium).
2. Rapacuronium was eliminated from the central compartment.
3. The fraction of rapacuronium eliminated by the kidneys compared with other routes did not vary with the concentration of rapacuronium.
For the participant whose urine sample was discarded, ƒrenalwas modeled using a feature of NONMEM that permits a compartment to be emptied.
Assumptions Used in the Modeling of ORG9488
Pharmacokinetic characteristics of ORG9488 were modeled using the following assumptions:
1. Values for the pharmacokinetics of rapacuronium were fixed to the values determined from the plasma concentrations of rapacuronium (i.e., fitting of the model to ORG9488 concentrations was not permitted to influence the quality of the fit to the rapacuronium plasma concentration values).
2. Conversion of rapacuronium to ORG9488 occurred in the central compartment of rapacuronium and was unidirectional.
3. ORG9488 was eliminated unidirectionally from its central compartment.
4. ORG9488 was distributed to only a single compartment or to central and peripheral compartments.
5. The administered drug contained no ORG9488.
Because each vial of rapacuronium provided for this study contained 0 to 1% ORG9488 (oral personal communication, Viquar Pervaaz, Organon Inc., August 1997), assumption 5 is flawed. The flawed assumption would probably result in the following artifacts in the analysis of the pharmacokinetic characteristics of ORG9488. (1) The measured values of ORG9488 in the period immediately after administration of rapacuronium would be larger than predicted based on conversion of rapacuronium to ORG9488. (2) The values for central compartment volume for ORG9488 would be artifactually decreased. And (3) the distribution half-life for ORG9488 would be artifactually small; however, the terminal half-life for ORG9488 likely would be unaltered.
Therefore, additional analyses were performed that permitted a dose or ORG9488 to be administered to its central compartment.[dagger][dagger]
Likelihood Profile
In one instance, the coefficient for a covariate added to the model had a standard error sufficiently large that the 95% confidence interval for this parameter included zero. This suggested that the parameter was not needed in the model, despite an improvement in the objective function sufficient to justify its inclusion. However, the 95% confidence interval assumes that the distribution of possible parameter estimates is normal, an assumption that is often flawed. To confirm that the parameter should be included in the model, we determined the likelihood profile. [23] Briefly, the analysis was repeated several times with that parameter fixed to values between the optimal estimate and zero, and the resulting objective function was determined. As expected, the objective function increased (i.e., it was associated with a deterioration in the quality of the fit) with all of the fixed estimates compared with the value obtained when the parameter was estimated. The parameter estimate associated with a statistically significant worsening of the objective function, therefore, was determined.
** ORG9487 is stable only at pH 4 (personal, oral, communication, Harrie Joosten, Ph.D., Organon Inc.).
[dagger][dagger] Details are available from Dr. Dennis M. Fisher at
REFERENCES
Kahwaji R, Bevan DR, Bikhazi G, Shanks CA, Fragen RJ, Dyck JB, Angst MS, Matteo R: Dose-ranging study in younger adult and elderly patients of ORG 9487, a new, rapid-onset, short-duration muscle relaxant. Anesth Analg 1997; 84:1011-8
Miller RD, Cullen DJ: Renal failure and postoperative respiratory failure: Recurarization? Br J Anaesth 1976; 48:253-6
McLeod K, Watson MJ, Rawlins MD: Pharmacokinetics of pancuronium in patients with normal and impaired renal function. Br J Anaesth 1976; 48:341-5
Buzello W, Agoston S: Pharmacokinetics of pancuronium in patients with normal and impaired renal function. Anaesthesist 1978; 27:291-7
Fahey MR, Morris RB, Miller RD, Nguyen T-L, Upton RA: Pharmacokinetics of Org NC45 (Norcuron) in patients with and without renal failure. Br J Anaesth 1981; 53:1049-53
Fahey MR, Rupp SM, Fisher DM, Miller RD, Sharma M, Canfell C, Castagnoli K, Hennis PJ: The pharmacokinetics and pharmacodynamics of atracurium in patients with and without renal failure. Anesthesiology 1984; 61:699-702
Cook DR, Freeman JA, Lai AA, Kang Y, Stiller RL, Aggarwal S, Harrelson JC, Welch RM, Samara B: Pharmacokinetics of mivacurium in normal patients and in those with hepatic or renal failure. Br J Anaesth 1992; 69:580-5
Szenohradszky J, Fisher DM, Segredo V, Caldwell JE, Bragg P, Sharma ML, Gruenke LD, Miller RD: Pharmacokinetics of rocuronium bromide (ORG 9426) in patients with normal renal function or patients undergoing cadaver renal transplantation. Anesthesiology 1992; 77:899-904
Szenohradszky J, Caldwell JE, Sharma ML, Gruenke LD, Miller RD: Interaction of rocuronium (ORG 9426) and phenytoin in a patient undergoing cadaver renal transplantation: A possible pharmacokinetic mechanism? Anesthesiology 1994; 80:1167-70
Lee GC, Iyengar S, Szenohradszky J, Caldwell JE, Wright PM, Brown R, Lau M, Luks A, Fisher DM: Improving the design of muscle relaxant studies. Stabilization period and tetanic recruitment. Anesthesiology 1997; 86:48-54
Beal SL, Sheiner LB: Methodology of population pharmacokinetics, Drug Fate and Metabolism: Methods and Techniques. Edited by E Garrett, J Hirtz. New York, Marcel Dekker, 1985, pp 135-83
Fleck C, Borner A, Kretzschmar M, Machnik G, Sprott H, Zimmermann T, Keil E, Braunlich H: Liver function after bilateral nephrectomy. Liver 1992; 12:319-25
Laganiere S, Shen DD: Altered S(-)-propranolol disposition in bilateral ureter-ligated rats. Nephron 1987; 46:305-11
Terao N, Shen DD: Reduced extraction of I-propranolol by perfused rat liver in the presence of uremic blood. J Pharmacol Exp Ther 1985; 233:277-84
Touchette MA, Slaughter RL: The effect of renal failure on hepatic drug clearance. DICP: Ann Pharmacother 1991; 25:1214-24
Schiere S, Proost JH, Wierda JMKH: Pharmacokinetics and pharmacokinetic/pharmacodynamic (PK/PD) relationship of ORG 9488, the 3-desacetyl metabolite of ORG 9487 (abstract). Anesthesiology 1997; 87:A377
Wright PMC, Brown R, Lau M, Fisher DM: A pharmacodynamic explanation for the rapid onset/offset of rapacuronium bromide. Anesthesiology 1999; 90:16-23
Bowman W, Rodger I, Houston J, Marshall R, McIndewar I: Structure:action relationships among some desacetoxy analogues of pancuronium and vecuronium in the anesthesized cat. Anesthesiology 1988; 69:57-62
Kopman AF: Pancuronium, gallamine, and d-tubocurarine compared: Is speed of onset inversely related to drug potency? Anesthesiology 1989; 70:915-20
Caldwell JE, Lau M, Fisher DM: Atracurium versus vecuronium in asthmatic patients. A blinded, randomized comparison of adverse events. Anesthesiology 1995; 83:986-91
Wright PMC, Hart P, Lau M, Sharma ML, Gruenke L, Fisher DM: Cumulative characteristics of atracurium and vecuronium. A simultaneous clinical and pharmacokinetic study. Anesthesiology 1994; 81:59-68
Segredo V, Caldwell JE, Matthay MA, Sharma ML, Gruenke LD, Miller RD: Persistent paralysis in critically ill patients after long-term administration of vecuronium. N Engl J Med 1992; 327:524-8
Bates DM, Watts DG: Nonlinear Regression Analysis and Its Applications. New York, Wiley, 1988, pp 200-31
Figure 1. Values for the plasma concentration of rapacuronium after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown with a dashed line.
Figure 1. Values for the plasma concentration of rapacuronium after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown with a dashed line.
Figure 1. Values for the plasma concentration of rapacuronium after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown with a dashed line.
×
Figure 2. The quality of fit of the pharmacokinetic model to the values for plasma concentration of rapacuronium. The x axis shows time (in minutes) after administration of rapacuronium. The y axis shows the ratio of the measured concentration of rapacuronium to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10) and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected in that the post hoc model permits interindividual variability, but the population model does not.
Figure 2. The quality of fit of the pharmacokinetic model to the values for plasma concentration of rapacuronium. The x axis shows time (in minutes) after administration of rapacuronium. The y axis shows the ratio of the measured concentration of rapacuronium to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10) and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected in that the post hoc model permits interindividual variability, but the population model does not.
Figure 2. The quality of fit of the pharmacokinetic model to the values for plasma concentration of rapacuronium. The x axis shows time (in minutes) after administration of rapacuronium. The y axis shows the ratio of the measured concentration of rapacuronium to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10) and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected in that the post hoc model permits interindividual variability, but the population model does not.
×
Figure 3. Values for the plasma concentration of ORG9488 after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown by a dashed line.
Figure 3. Values for the plasma concentration of ORG9488 after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown by a dashed line.
Figure 3. Values for the plasma concentration of ORG9488 after administration of 1.5 mg/kg rapacuronium are shown for 10 healthy volunteers (left) and 10 patients with renal failure (right). Values for the patient who received phenytoin before anesthesia are shown by a dashed line.
×
Figure 4. The quality of fit of the pharmacokinetic model to the values for plasma concentration of ORG9488. The x axis is time (in minutes) after administration of rapacuronium. The y axis is the ratio of the measured concentration of ORG9488 to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10), and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected because the post hoc model permits interindividual variability, but the population model does not.
Figure 4. The quality of fit of the pharmacokinetic model to the values for plasma concentration of ORG9488. The x axis is time (in minutes) after administration of rapacuronium. The y axis is the ratio of the measured concentration of ORG9488 to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10), and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected because the post hoc model permits interindividual variability, but the population model does not.
Figure 4. The quality of fit of the pharmacokinetic model to the values for plasma concentration of ORG9488. The x axis is time (in minutes) after administration of rapacuronium. The y axis is the ratio of the measured concentration of ORG9488 to the value predicted by the population pharmacokinetic model (left) or the post hoc fit (right). Each line represents a value from a single person. Healthy volunteers are shown by a solid line (n = 10), and patients with renal failure (n = 9) are shown by a dashed line. If the model fit the data perfectly, all lines would lie horizontally at 1.0. The improved quality of fit of the post hoc values compared with those from the population model is expected because the post hoc model permits interindividual variability, but the population model does not.
×
Table 1. Demographic Data for 10 Healthy Volunteers and 10 Patients with Renal Failure
Image not available
Table 1. Demographic Data for 10 Healthy Volunteers and 10 Patients with Renal Failure
×
Table 2. Magnitude of Twitch Depression and Time Course of Neuromuscular Effects of Rapacuronium in Healthy Volunteers and in Patients with Renal Failure
Image not available
Table 2. Magnitude of Twitch Depression and Time Course of Neuromuscular Effects of Rapacuronium in Healthy Volunteers and in Patients with Renal Failure
×
Table 3. Models Tested for the Pharmacokinetics of Rapacuronium
Image not available
Table 3. Models Tested for the Pharmacokinetics of Rapacuronium
×
Table 4. Parameter Estimates* of Three-Compartment Pharmacokinetic Model for Rapacuronium (Model 13) for Healthy Volunteers and Patients with Renal Failure
Image not available
Table 4. Parameter Estimates* of Three-Compartment Pharmacokinetic Model for Rapacuronium (Model 13) for Healthy Volunteers and Patients with Renal Failure
×
Table 5. Models Tested for the Pharmacokinetics of ORG9488
Image not available
Table 5. Models Tested for the Pharmacokinetics of ORG9488
×
Table 6. Parameter Estimates* of Two-compartment Pharmacokinetic Model for ORG9488 (Model 23)
Image not available
Table 6. Parameter Estimates* of Two-compartment Pharmacokinetic Model for ORG9488 (Model 23)
×
Table 7. Systolic, Diastolic, and Mean arterial Blood Pressure for 8 of 10 Healthy Volunteers and 10 Patients Given 1.5 mg/kg Rapacuronium
Image not available
Table 7. Systolic, Diastolic, and Mean arterial Blood Pressure for 8 of 10 Healthy Volunteers and 10 Patients Given 1.5 mg/kg Rapacuronium
×