Ropivacaine is a new long-acting amino-amide local anesthetic. However, there are no data on its use in infants. In the current study, the authors investigated the pharmacokinetics of caudal ropivacaine in 30 infants younger than 12 months.
Two groups of infants (group 1 [n = 15], aged 0-3 months; group 2 [n = 15], aged 3-12 months) were given a caudal bolus dose of 0.2% ropivacaine (2 mg/kg) and a standardized general anesthetic technique. Serial blood samples taken for up to 12 h were analyzed for total and free ropivacaine using high-performance liquid chromatography. Population pharmacokinetic modeling was performed to yield estimates of clearance, volume of distribution, and absorption rate constant. An analysis of covariates on the kinetic parameters also was made.
Median maximum free ropivacaine concentration was significantly higher in group 1 (99 micog/l) than in group 2 (38 microg/l) (P = 0.0002), as was the median free fraction of ropivacaine (10% vs. 5%; P = 0.01). Pharmacokinetic variables of the total population were best described by a one-compartment model with first-order absorption. Mean clearance was 0.31 l.h(-1).kg(-1) (coefficient of variation [CV], 51%), volume of distribution was 2.12 l/kg (CV, 34%), and absorption rate constant was 1.61 h(-1) (CV, 46%). Mean absorption and elimination half-lives were 0.43 and 5.1 h, respectively. Age and percentage of free ropivacaine were significant covariates for clearance. Posterior Bayesian estimates of clearance were significantly higher (38%) in older children.
Total and free plasma ropivacaine concentrations after caudal ropivacaine (0.2%, 2 mg/kg) in infants were within the range of concentrations previously reported in adults and older children. Age and percentage of free ropivacaine were significant covariates of clearance.
CAUDAL epidural block with bupivacaine is one of the most commonly performed regional anesthetic blocks in small children. 1–4It is mainly used as an adjunct to general anesthesia to provide a smooth recovery and good immediate postoperative analgesia. Occasionally, however, caudal anesthesia is also used as the sole anesthetic in infants at risk for complications from general anesthesia and tracheal intubation. 1
The safety profile of the new long-acting amino–amide local anesthetic ropivacaine seems very promising, 5,6and, although the drug is not yet approved for use in children younger than 12 yr, the number of published pediatric studies has increased in recent years. 7–14Clinically, caudal ropivacaine has been shown to provide postoperative analgesia similar to that obtained with caudal bupivacaine, 14and, in the few pediatric studies comprising children aged 1–7 yr, the pharmacokinetic parameters for ropivacaine have been found to be comparable with those reported in adults. 11–13,15,16However, the pharmacokinetics of ropivacaine has not been studied in neonates and infants.
There are marked differences in the pharmacokinetics of drugs in neonates and infants, where maximal physiologic changes occur compared with children and adults. 17For example, the plasma protein (α1-acid glycoprotein, albumin, γ-globulin, and lipoprotein) binding capacity for drugs in neonates and infants may be reduced. Thus, the free fraction of highly protein-bound drugs, such as amide local anesthetics, can be increased, and this, in turn, may alter the drug concentration in the blood. In addition, pharmacokinetic differences in neonates and infants may also arise from differences in body size or immaturity of the organs of elimination. 3,17–19
The principal aims of this study were to describe the plasma ropivacaine concentrations (total and free) and the pharmacokinetics of a single caudal ropivacaine dose (2 mg/kg) in two different age groups of infants younger than 12 months. A secondary aim was to assess the effect of covariates on the pharmacokinetic parameters in this population.
Materials and Methods
After obtaining approval from the Princess Margaret Hospital Ethics Committee and written parental informed consent, 30 infants younger than 12 months, American Society of Anesthesiologists physical status I or II, who were scheduled to undergo elective lower abdominal, genitourinary, or orthopedic surgery, were enrolled in the study. The infants were allocated to two groups according to age (group 1: aged 0–3 months, n = 15; group 2: aged 3–12 months, n = 15). Infants with coagulopathy, neuromuscular disease, lower back problems, local skin infections of the caudal area, or renal–hepatic impairment were excluded. None of the children were prescribed premedication or had local anesthetic cream applied to the skin. Anesthesia was induced with 8% sevoflurane in 100% oxygen, with the children breathing spontaneously via a face mask and Jackson-Rees modification of the T piece. An intravenous catheter was then inserted in a dorsal hand cubital or saphenous vein for blood sampling. Patency of this cannula was ensured by an infusion of 2.5% wt/vol dextrose in 0.45% wt/vol saline at a rate of 5–10 ml/h.
The airway was maintained with a laryngeal mask or an endotracheal tube. If necessary, intravenous vecuronium (0.05–0.1 mg/kg) was administered to facilitate orotracheal intubation.
After the anesthetic induction, the infants were placed in the left lateral position, and a caudal block was performed using an aseptic technique and a 23-gauge needle. After negative aspiration of blood and cerebrospinal fluid, 1 ml/kg of 0.2% ropivacaine (2 mg/ml Naropin, Astra Pharmaceuticals Pty Ltd, NSW, Australia) was injected slowly and in increments into the caudal space while watching vital signs and the electrocardiographic monitor. General anesthesia was maintained with 0.5–1.0% isoflurane and nitrous oxide in 35% oxygen throughout the surgical procedure. If necessary, intravenous fentanyl (1–2 μg/kg) was given. At the end of the surgical procedure, any residual muscular paralysis was reversed with atropine (20 μg/kg) and neostigmine (50 μg/kg), and the infant was extubated. Standard intraoperative and postoperative monitoring equipment was used in all infants to detect any neurotoxic or cardiotoxic effects.
Pain was assessed postoperatively using a 0–10 observational pain score, and intravenous morphine (25 μg/kg) or oral or rectal paracetamol (15–20 mg/kg) was offered when a score of 3 was recorded. Because of the age of the children in this study, no attempt was made to assess the degree of either sensory or muscular block.
Blood Sampling and Analysis
Venous blood samples (1 ml) were taken from a peripheral intravenous catheter (20 or 22 gauge; Insyte, Becton Dickinson Infusion Therapy Systems Inc, Sandy, UT) at 0, 15, 30, 45, 60, 120, 150, 240, 360, 540, and 720 min after the ropivacaine administration or for as long as sampling was possible. After each blood sample was taken, the intravenous catheter was flushed with 1 ml heparinized saline.
Blood samples were subsequently separated by centrifugation at 1,500 g for 5 min, and plasma was stored at −20°C until analyzed for total and free ropivacaine concentrations using high-performance liquid chromatography as previously described. 11Free ropivacaine was not measured in every patient sample because of volume constraints. In 25 patients there was sufficient plasma for analysis of free drug; the median number of samples per patient was 7 (range, 1–10).
The highest total plasma ropivacaine concentration (Cmax) observed was recorded for each infant. The mean free fraction for ropivacaine in plasma (fu) was calculated from the available samples for each patient (n = 13 in group 1, n = 12 in group 2). The highest observed free plasma ropivacaine concentration (Cu,max) also was recorded in these infants.
Compartmental models (1 or 2) with first-order absorption were fitted to the plasma concentration–time data and analyzed using a population pharmacokinetic approach implemented in the program P-Pharm Version 1.5 (InnaPhase, Champs sur Marne, France). A heteroscedastic error variance (1/Y2) model was used to describe residual error, and a log normal distribution was used to describe interpatient variability in clearance (CL), volume of distribution (V), and absorption rate constant (ka). Because bioavailability (F) was not known, estimates of CL and V are relative to F, i.e. , CL/F and V/F. Model validation was conducted as recommended in the P-Pharm manual and also by examination of a plot of measured/predicted concentration data versus time. 20In addition, bias and precision for the overall model, and also median weighted absolute prediction error and median weighted prediction error for the combined data sets, were calculated as described previously. 21–23
Data are summarized as mean ± SD or 95% confidence interval or median (interquartile range) as appropriate. A Mann–Whitney U test was used to compare observed Cmaxdata between the two patient groups, while a two-tailed t test was used to examine differences between mean posterior Bayesian pharmacokinetic parameters for the two groups. Simple linear regression or backward stepwise multiple linear regression analyses (Sigma Stat version 2.03, SPSS Inc., Chicago, IL) were used to investigate correlations between observed and model predicted drug concentrations, between the pharmacokinetic parameters and age, and percentage of free ropivacaine.
Demographic data are summarized in table 1. In both groups the male–female distribution was 14–1. The reasons for surgery in group 1 were inguinal hernia repair (n = 13), malrotation repair (n = 1), and circumcision (n = 1), whereas in group 2 the reasons were inguinal hernia repair (n = 5), hypospadia repair (n = 3), circumcision (n = 3), closure of a colostomy (n = 2), clubfoot repair (n = 1), and orchidopexy (n = 1).
The caudal blocks in all infants studied were considered adequate as they received adequate intraoperative and immediate postoperative analgesia and completed the study without complications or signs of clinical toxicity. Blood pressure, pulse rate, respiratory rate, and oxygen saturation remained stable throughout the study period. One patient in group 1 and three patients in group 2 required intravenous fentanyl during surgery. Overall, postoperative pain scores were mainly less than 3 in both groups (data not shown). In group 2, there were 2 infants who had 1–3 doses of morphine (25 μg/kg) over 5–24 h postoperatively. One infant from group 1 and two infants from group 2 electively received a 24-h morphine infusion (10 μg · kg−1· h−1) that was commenced in the recovery area.
Plasma Ropivacaine Total and Free Concentration Measurements
The total plasma ropivacaine concentrations for infants in groups 1 and 2 are shown in figures 1A and 1B. The median total Cmaxof 748 μg/l (425–1,579 μg/l) in group 1 was similar to the value of 604 μg/l (410–1,278 μg/l) in group 2. The highest individual total ropivacaine concentration (1,579 mg/l) was found in a 76-day-old (4.7-kg) infant 150 min after the dose.
Overall, the free ropivacaine concentrations in group 1 ranged from 10 to 143 μg/l and from 7 to 67 μg/l in group 2. The highest individual free ropivacaine Cmaxof 143 μg/l was found in a 36-day-old (4.2-kg) infant 150 min after dose. The median free ropivacaine concentrations (i.e. , the Cu,max) was significantly (P = 0.0002) higher in group 1 (n = 10), with a Cu,maxof 99 μg/l (44–143 μg/l), than that of 38 μg/l (19–67 μg/l) in group 2 (n = 11). The median percentage of free ropivacaine was significantly greater (P = 0.01) in group 1 (10%; interquartile range, 6.5–14%; n = 12) than in group 2 (5%; interquartile range, 4–6.24%; n = 13).
For the total data set (n = 30), a one-compartment population pharmacokinetic model with first-order absorption and elimination was found to give an optimal fit to the data, as assessed by maximum likelihood, Sigma and Akaike’s information criterion values generated by P-Pharm. Model selection also included consideration of the distribution of parameters and residuals, as well as the correlation of observed and predicted concentration data. Initially, attempts were made to include a lag time in the model, but there were insufficient data to support this additional parameter. The frequency distributions of standardized residuals for the model itself, CL/F, V/F, and kawere not different from normal as assessed by a Kolmogorov-Smirnoff test, and no outliers were detected. Figure 2shows a scatter plot of predicted ropivacaine concentrations (Bayesian estimates) versus observed ropivacaine concentrations, whereas figure 3shows the variation of measured/predicted concentration ratio versus time for all 30 individual patients. Figure 4shows posterior Bayesian predicted total ropivacaine concentration versus time for each of the 30 subjects. Mean CL/F, V/F, and kadescribing the model are given in table 2. The corresponding calculated mean absorption and elimination half-lives were 0.43 and 5.1 h, respectively. For the overall model, mean (95% confidence interval) bias and precision were −22 μg/l (−9,−36 μg/l) and 115 μg/l (56,174 μg/l), respectively. For all data sets combined, median weighted prediction error (25th, 75th percentiles) was −3.9% (−16.7, 7.8%), whereas median weighted absolute prediction error (25th, 75th percentiles) was 12.6% (5.3, 22.2%). Representative posterior Bayesian fits for one patient from each group are shown in figures 5A and 5B. An analysis of the posterior Bayesian pharmacokinetic parameters for the group as a whole and for groups 1 and 2 separately is shown in table 3. For all patients, the pharmacokinetic parameters were similar to the population estimates. A comparison of parameters between groups 1 and 2 indicated that only CL/F was significantly different (P = 0.029), being some 38% higher in the older children.
A preliminary regression analysis (n = 30) indicated, as might be expected, that there was a strong correlation between age and weight (r2= 0.852). Because of this finding, only age and percentage of free ropivacaine were subsequently regressed against either weight-normalized CL/F or V/F using a backward stepwise regression procedure (n = 25, as there was insufficient plasma to measure percentage of free ropivacaine in five infants). Both age (P < 0.001) and percentage of free ropivacaine (P = 0.04) were significant covariates for CL/F (CL/F = 0.0466 + 0.0011 · age in days + 0.0162 · percentage of free; power of test = 97.4%). There were no relations between either V/F or kaand age or percentage of free ropivacaine.
This is the first study of the pharmacokinetics of caudal ropivacaine in infants younger than 12 months. Although mainly addressing pharmacokinetic issues, this study also confirms the effectiveness of caudal ropivacaine reported in other pediatric studies. 11–14Total plasma ropivacaine was best represented by a one-compartment model with first-order absorption. Although this model gave robust parameter estimates, we acknowledge that it has limitations because of the distribution of sampling times and the total duration over which sampling was possible. In particular, we had only three to four samples in the absorption phase, and this could have increased the variability of ka. For some patients, variability in the model was also evident at early time in the measured/predicted concentration–time plot (fig. 3).
Overall, both the total and free ropivacaine levels were comparable to those tolerated by adults (i.e. , 1,000–3,000 μg/l and 10–150 μg/l, respectively). 5,6,15,16The highest individual total ropivacaine concentration (1,579 μg/l) was not associated with any sign of toxicity. Interestingly, adults have tolerated total ropivacaine levels up to 5,200 μg/l during long-term epidural infusion. 15However, it should be noted that we measured venous drug concentrations in the current study and that these may well be somewhat lower than those in arterial blood. Although no toxic events occurred in our study, the number of patients studied is small, and the data should therefore be interpreted cautiously.
The toxicity of extensively plasma protein-bound local anesthetics such as ropivacaine is more likely to be related to the free plasma concentrations rather than the total plasma concentration. 24,25This may be particularly important in young infants, in whom the plasma proteins that bind ropivacaine (mainly α1-acid glycoprotein) are reduced or have lower binding capacity than in older children. 3,17,24–26In our study, the median free Cu,maxwas significantly higher in group 1 (99 μg/l) than in group 2 (38 μg/l). Similarly, the median percentage of free ropivacaine was significantly higher in group 1 (10%) than in group 2 (5%). We did not measure plasma α1-acid glycoprotein because we wished to minimize the amount of blood taken during the study.
The absorption of ropivacaine from the caudal epidural space into the systemic circulation was moderately rapid (mean absorption half-life = 26 min). Nevertheless, our estimate of kawas variable, and inspection of the data in figures 1 and 4shows that, although Cmaxoccurred at around an hour in many patients, peaks as late as 3–4 h were evident in others. In the current study, V/F (2.12 l/kg) was similar to values for Vdreported in our previous studies in older children (range, 1.9–4.2 l/kg). 11,12The derived elimination half-life obtained in the current study (5.1 h) was similar to values previously reported by our group (mean, 4.9 h) in older children (mean age, 3.3 yr) receiving continuous long-term epidural ropivacaine infusion, 11but somewhat longer than we reported previously in older children (mean age, 2.1 yr) after a caudal bolus dose of ropivacaine (mean, 3.9 h). 12In this regard, it is possible that our half-life estimate may have been prolonged by late peaks (continuing absorption) in some patients. As seems logical, both age and percentage of free ropivacaine were significant covariates in the weight-normalized clearance of ropivacaine. Moreover, analysis of the posterior Bayesian estimates of clearance showed significantly greater clearance in the older children. Despite higher free concentrations in younger infants and a lower clearance, median Cmaxwas similar in both age groups. The two subgroups within the study differed slightly in terms of severity of the different surgical procedures and hence the need for supplementary intraoperative and postoperative opioids. However, this difference is unlikely to have been an influence on the disposition of ropivacaine in our study.
Ropivacaine is eliminated predominantly by hepatic metabolism. 27It has an intermediate to low extraction ratio 28with a total plasma clearance dependent on fuand unbound plasma clearance almost exclusively dependent on hepatic enzymatic activity. 15,16Hence, in theory, differences in fumay contribute to interpatient variability in overall clearance. We found that percentage of free ropivacaine was higher in the younger age group. Nevertheless, we found that percentage of free ropivacaine and age were significant covariates for clearance, and that clearance was higher in the older group. However, higher free concentrations–fractions in younger children were accompanied by a lower overall clearance. Thus, hepatic enzyme capacity is the dominant factor in determining clearance in children. Contrary to the findings in adults, 28it appears that clearance of ropivacaine is not limited by available enzyme capacity in the first year of life. Our study also shows that the capacity for hepatic metabolism of ropivacaine is present at an early stage in life. Mean clearance for total ropivacaine (0.31 l · kg−1· h−1) was at the lower end of the range of values (0.35–0.67 l · kg−1· h−1) that we previously reported in children with a mean age of 2.1 and 3.3 yr, respectively. 11,12Although a definitive cutoff age remains to be established, our results suggest that a plateau in the capacity to metabolize ropivacaine in the liver probably occurs at a relatively late stage in infancy (6–12 months).
In dosing the infants, we chose to use a weight-related dose calculation rather than a body surface area–related dose calculation. The choice was made solely on the grounds of simplicity and practicality. Only body weight and not height was necessary, and the approach avoided the potential complications 29of using a nomogram to calculate body surface area.
In conclusion, our study has shown that the pharmacokinetics of ropivacaine in young children is similar to those previously reported in older children and adults. Although no adverse effects were observed in our study, the number of patients was small, and further studies of the possible toxicity of ropivacaine in young children are needed.