1 University Department of Anaesthesia, Glasgow Royal Infirmary, Queen Elizabeth Building, Alexandra Parade, Glasgow G31 2ER, UK. 2 Golden Jubilee National Hospital, Beardmore Street, Clydebank, Glasgow G81 4HX, UK. 3 University of Amsterdam Medical Centre, Meibergdreef 9, 1105AZ, Amsterdam, The Netherlands 4 Present address: Deptartment of Anaesthesia, Good Hope Hospital, Rectory Road, Sutton Coldfield, West Midlands B75 7RR, UK
Corresponding author. Present address: Department of Anaesthesia, Norfolk & Norwich University Hospital, Colney Lane, Norwich, Norfolk NR4 7UY, UK. E-mail: anthony.absalom@nnuh.nhs.uk LMA® is the property of Intavent Limited.
Accepted for publication: May 21, 2003
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. Twenty-nine children aged 115 yr were investigated. General anaesthesia was provided using propofol administered by the Paedfusor system. Accuracy of the system was evaluated by obtaining up to 9 arterial samples for measurement of propofol concentration both during anaesthesia and in the recovery period. Measured arterial propofol concentrations were then compared with values calculated by the Paedfusor.
Results. The predictive indices of median performance error (MDPE), and median absolute performance error (MDAPE) of the Paedfusor system were found to be 4.1% and 9.7%, respectively and the median value for wobble was 8.3%. These values are much better than those found with the adult Diprifusor system.
Conclusion. The Paedfusor performance was found to be within the accepted limits for use as a TCI system.
Br J Anaesth 2003; 91; 50713
Keywords: anaesthetic techniques, i.v.; anaesthetic techniques, i.v. infusions; anaesthetics i.v., propofol; pharmacokinetics; surgery, cardiovascular
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The pharmacokinetic parameters used in the Diprifusor were adapted from those derived from population studies in adults by Gepts2 and were published in 1993.3 Marsh and colleagues3 studied the predictive performance of these adult parameters in 20 children, and found that the measured blood propofol concentrations were significantly less than those predicted by the model. They then used the data to generate a new model, specific to children, that was studied prospectively in a further 10 children and found to predict blood concentrations more accurately.3 Marsh and colleagues findings were consistent with those of other groups of workers who have found that the pharmacokinetics of propofol differ between children and adults.4 5
Schüttler6 later analysed the data from several studies of the pharmacokinetics of propofol in adults and children and produced a model that incorporates age, weight, sampling site (venous or arterial) and administration mode (bolus or infusion) and is valid for patients of all ages. A prototype target controlled infusion (TCI) system for children, the Paedfusor, was developed in Glasgow in 1998, and used routinely for induction and maintenance of anaesthesia at HCI (Scotland) (>600 children). The pharmacokinetic data set it uses is from one of the preliminary models developed by Schüttler before the publication of his final model.6 Table 1 enables a comparison of these pharmacokinetic variablesthe Marsh adult (as used in the Diprifusor) and Marsh paediatric, and the Short, Paedfusor and final Schüttler variables. In both Marsh models and the Short model, the rate constants are fixed while volume variables are linear functions of weight. In the Paedfusor model the central compartment volume and clearance have a non-linear correlation with weight, and in the Schüttler model all variables have a non-linear correlation with age and weight. Thus for ease of comparison the values shown for the Paedfusor and Schüttler models are those for a 5-yr-old child weighing 20 kg.
|
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Non-invasive monitoring was established before induction of anaesthesia. Venous access was established with a 22 or 20 gauge cannula on the dorsum of the hand or forearm and the cannula connected to the Paedfusor. In smaller children with difficult venous access, general anaesthesia was induced with sevoflurane in oxygen-enriched air, before venous and radial or femoral artery cannulae were inserted. Once both venous and arterial access was achieved sevoflurane was discontinued and propofol TCI commenced at 5 µg ml1 with the Paedfusor system. In older children, anaesthesia was induced with the Paedfusor system at an initial propofol target of 5 µg ml1. After loss of consciousness, a second anaesthetist inserted an arterial cannula. If he failed to do this within 3 min of the start of the propofol infusion, that child was excluded from the study. Arterial access was used for both invasive blood pressure monitoring and collection of blood samples. During the procedure, the target propofol concentration was altered on the basis of clinical judgement of the depth of anaesthesia.
The 7 children who underwent cardiac catheterization received fentanyl 1 µg kg1 at induction, and local anaesthetic infiltration of the femoral puncture site with bupivacaine 0.5%. In acyanotic children with no pulmonary hypertension (5 children), a laryngeal mask airway (LMA) was inserted to facilitate spontaneous ventilation with oxygen-enriched air. In those with cyanosis and possible pulmonary hypertension (2 children), the trachea was intubated and ventilation controlled via a circle breathing system. Anaesthesia was maintained with the Paedfusor. One patient required a further dose of fentanyl 1 µg kg1 during the course of a catheterization procedure that lasted longer than expected. At the end of the procedure, the TCI was stopped by setting the target to zero. The children were then monitored and given supplementary oxygen until awake.
In children who underwent cardiac surgery, anaesthesia was maintained with the Paedfusor supplemented by a TCI infusion of alfentanil. The TCI system used to control the alfentanil infusion comprises a Graseby 3400 infusion pump controlled by a microprocessor programmed with the Maitre pharmacokinetic model incorporating age, gender and weight.7 A prototype of this system has been described previously,8 and the existing system has been used in studies of patient maintained analgesia.9 The alfentanil infusion was commenced at induction and continued until after the end of the procedure, with a target blood concentration of 100 ng ml1. Pancuronium 0.040.1 mg kg1 was administered to facilitate tracheal intubation. The lungs were mechanically ventilated with oxygen-enriched air to an end-tidal carbon dioxide tension of 3.54.5 kPa. At the end of surgery, the patient was transferred to the paediatric intensive care unit. The alfentanil and propofol infusions were continued for a further 218 h, usually at lower target concentrations, until the chest drains had been removed, and the patients condition was such that he or she was ready for awakening and tracheal extubation.
Up to 9 arterial blood samples (0.5 ml) were collected from each child for assay of plasma propofol concentration. These were stored in EDTA bottles in a refrigerator and later analysed in batches using gas-liquid chromatography. The limit of detection of the assay was 0.05 µg ml1, and the coefficient of variation 4.2% at 0.5 µg ml1, and 4.9% at 8 µg ml1. Calibration graphs were linear between 0.5 and 8 µg ml1. Samples were collected at the following intervals (sample number in brackets): 3 min (1) and 8 min (2) after starting the propofol infusion; 6090 s after a reduction of the target concentration (3); during stable anaesthesia (4); after >20 min on cardiopulmonary bypass (5); just before (6), and 3 (7) and 8 min (8) after stopping the TCI system; and finally on eye opening (9). Stable anaesthesia (sample 4) was defined as a period when there had been no target change for >30 min. Sample 3 was taken just before the estimated blood concentration had reached the new, lower target concentration. Thus, samples 3, 7, 8 and 9 were taken during periods when the TCI system was not infusing propofol (no flow).
Data analysis
Three patients were excluded from the study. Two were excluded because it was not possible to cannulate the radial artery within 3 min and a third patient died after failed implantation of the coronary arteries. Of the remaining 29 patients, 7 underwent cardiac catheterization, and 22 underwent surgical procedures, all but one of which required cardiopulmonary bypass.
A total of 212 blood samples were obtained and analysed. Predicted and measured propofol concentrations were normally distributed and are presented as a mean (SD). Performance was evaluated using the methods recommended by Varvel and colleagues.10 At each sampling point, the performance error (PE) of the pump was calculated, using Cpred (the propofol concentration predicted by the TCI system), and Cmeas (the measured propofol concentration), as follows:
PE (%) = [(Cmeas Cpred)/Cpred] x 100(1)
The series of performance errors was then used to calculate, for each patient, and for the entire group, the median performance error (MDPE), the median absolute performance error (MDAPE) and the wobble. MDPE is a measure of the bias. It is a signed value and represents the direction (over- or under-prediction) of the performance error, rather than the size of the error (precision), which is reflected in the MDAPE. Wobble is a measure of the intra-individual variability of the PE about the MDPE and was calculated as follows for patient i having Ni samples:
Wobblei = Median {|PEij MDPEi|; j =1...Ni}(2)
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
The MDPE of 4.1% found in the current study is far less than that found in adult studies using the Diprifusor TCI system, where values for bias are of the order of 16%.12 This small bias is put into perspective when one considers that when the end-tidal isoflurane or halothane concentration is used to estimate the arterial concentration, the bias is of the order of 20%.13 14
The bias and precision of the Paedfusor TCI system tend to be worse during no flow periods. These findings contrast with those found in adult studies where the Diprifusor was used, where accuracy was best during no flow periods, and worst with steady-state blood concentrations.12 15 There are two possible reasons why accuracy is worst during no flow with this model in children: either the central compartment clearance is slower than that predicted by the model, or the amount of drug returning to the central compartment is greater than expected. We believe that the latter is the most likely explanation, because accuracy was greatest during stable anaesthesia with no change in propofol target concentration for >30 min, when clearance from the central compartment (V1 x k10) is the main determinant of accuracy. During periods of no flow, plasma concentrations are determined by all the model parameters. Thus the problem is likely to lie with one or more of the other rate constants, but this can only be verified by further studies.
The subjects we studied were a heterogeneous group of patients who were mostly of Middle Eastern origin, whereas the pharmacokinetic variables used were derived from studies of western patients (four European centres and one North American centre6). The Paedfusor and Maitre7 pharmacokinetic models incorporate age and weight, and Maitres model also incorporates gender, but there were other sources of heterogeneity. The majority of patients received alfentanil, but some received fentanyl, some breathed spontaneously whereas others had mechanical ventilation of the lungs, and most, but not all, underwent cardiopulmonary bypass.
In patients who underwent cardiopulmonary bypass, the system bias changed from positive to negative during bypass. This is as expected, for at least two reasons. First, there is a diluting effect of the priming solution used in the bypass machine, and second, the tubing itself effectively causes an increase in the volume of the central compartment. However, the overall effect was small: bias during bypass was only 5.5% (Table 4), and is unlikely to result in a significant change in pharmacodynamic effect (level of consciousness) especially when there is some degree of cooling. The small size of the change in bias on bypass is probably because, even in small children, the volume of the tubing (and thus the priming solution) is small in comparison with the volume of the central and other compartments. For example, in a child weighing 10 kg, the volume of prime would be 270 ml, whereas the volumes of the theoretical central, second and third compartments would be 4600, 1340 and 8200 ml respectively.
There is a paucity of literature on the influence of mode of ventilation on propofol kinetics. In a study in cats it was shown that mode of ventilation did not alter the single (first) pass pulmonary uptake of propofol.16 Interestingly, this same study also showed that a single dose of fentanyl inhibited single pass pulmonary propofol uptake by 40%. Mechanical ventilation has the potential to alter propofol kinetics via alterations in cardiac output and hepatic blood flow, but, as far as we are aware, this has not been demonstrated conclusively.
Other studies have shown significant pharmacokinetic interactions between propofol and opioids. Cockshott17 found that pretreatment with fentanyl 100 µg caused a 50% increase in propofol concentration, and Pavlin,18 when comparing propofol concentrations in subjects receiving propofol alone with those in subjects receiving both propofol and alfentanil infusions, found that a target alfentanil concentration of 40 ng ml1 was associated with a 1929% increase in propofol concentration. It is likely that, had our patients received only propofol, the bias would have been closer to zero, or somewhat negative.
In the study by Pavlin,18 alfentanil concentrations were also significantly higher when it was infused with propofol than when it was infused alone. Mertens and colleagues19 studied the pharmacokinetics of propofol in male volunteers and found significant reductions in elimination clearance (metabolism) and in inter-compartmental clearance rates for alfentanil in the presence of propofol. Proposed mechanisms for these interactions include competition between propofol and opioids for pulmonary binding sites,16 inhibition by propofol of cytochrome P450,20 and haemodynamic alterations caused by propofol.19
In a study of the performance of his model for alfentanil, Maitre21 found that the mean prediction error (bias) was 7.9%, and the mean absolute prediction error was 22.3%. We did not measure alfentanil concentrations in our patients, but given the kinetic interactions mentioned above, it is reasonable to assume that actual blood concentrations may have been higher than the target value of 100 ng ml1.
Swinhoe and Schüttler have stated that acceptable values for an infusion system are a MDPE <1020% and a MDAPE <2040%.12 22 We thus conclude that the Paedfusor performance is well within acceptable limits for clinical use, when combined with sound clinical judgement.
![]() |
Acknowledgement |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Gepts E, Camu F, Cockshott ID, Douglas EJ. Disposition of propofol administered as constant rate intravenous infusions in humans. Anesth Analg 1987; 66: 125663[Abstract]
3 Marsh BJ, White M, Morton N, Kenny GN. Pharmacokinetic model driven infusion of propofol in children. Br J Anaesth 1991; 67: 418[Abstract]
4 Saint-Maurice C, Cockshott ID, Douglas EJ, Richard MO, Harmey JL. Pharmacokinetics of propofol in young children after a single dose. Br J Anaesth 1989; 63: 66770[Abstract]
5 Murat I, Billard V, Vernois J, et al. Pharmacokinetics of propofol after a single dose in children aged 13 years with minor burns. Comparison of three data analysis approaches. Anesthesiology 1996; 84: 52632[ISI][Medline]
6 Schuttler J, Ihmsen H. Population pharmacokinetics of propofol: a multicenter study. Anesthesiology 2000; 92: 72738[ISI][Medline]
7 Maitre PO, Vozeh S, Heykants J, Thomson DA, Stanski DR. Population pharmacokinetics of alfentanil: the average dose-plasma concentration relationship and interindividual variability in patients. Anesthesiology 1987; 66: 312[ISI][Medline]
8 Davies FW, White M, Kenny GNC. Postoperative analgesia using a computerised infusion of alfentanil following aortic bifurcation graft surgery. Int J Clin Monit Comput 1992; 9: 20712[ISI][Medline]
9 Checketts MR, Gilhooly CJ, Kenny GNC. Patient-maintained analgesia with target-controlled alfentanil infusions after cardiac surgery: a comparison with morphine PCA. Br J Anaesth 1998; 80: 74851
10 Varvel JR, Donoho DL, Shafer SL. Measuring the predictive performance of computer-controlled infusion pumps. J Pharmacokinet Biopharm 1992; 20: 6394[ISI][Medline]
11 Short TG, Aun CS, Tan P, et al. A prospective evaluation of pharmacokinetic model controlled infusion of propofol in paediatric patients. Br J Anaesth 1994; 72: 3026[Abstract]
12 Swinhoe CF, Peacock JE, Glen JB, Reilly CS. Evaluation of the predictive performance of a Diprifusor TCI system. Anaesthesia 1998; 53 [Suppl. 1]: 617[ISI][Medline]
13 Frei FJ, Zbinden AM, Thomson DA, and Rieder HU. Is the end-tidal partial pressure of isoflurane a good predictor of its arterial partial pressure? Br J Anaesth 1991; 66: 3319[Abstract]
14 Dwyer RC, Fee JP, Howard PJ, Clarke RS. Arterial wash-in of halothane and isoflurane in young and elderly adult patients. Br J Anaesth 1991; 66: 5729[Abstract]
15 Russell D, Wilkes MP, Hunter SC, et al. Manual compared with target-controlled infusion of propofol. Br J Anaesth 1995; 75: 5626
16 Matot I, Neeley CF, Katz RY, Neufeld GR. Pulmonary uptake of propofol in cats. Effect of fentanyl and halothane. Anesthesiology 1993; 78: 115765[ISI][Medline]
17 Cockshott ID, Briggs LP, Douglas EJ, White M. Pharmacokinetics of propofol in female patients. Br J Anaesth 1987; 59: 110310[Abstract]
18 Pavlin DJ, Coda B, Shen DD, et al. Effects of combining propofol and alfentanil on ventilation, sedation, and emesis in human volunteers. Anesthesiology 1996; 84: 2337[ISI][Medline]
19 Mertens MJ, Vuyk J, Olofsen E, Bovill JG, Burm AGL. Propofol alters the pharmacokinetics of alfentanil in healthy male volunteers. Anesthesiology 2001; 94: 94957[ISI][Medline]
20 Baker MT, Chadam MV, Ronnenburg WC Jr. Inhibitory effects of propofol on cytochrome P450 activities in rat hepatic microsomes. Anesth Analg 1993; 76: 81721[Abstract]
21 Maitre PO, Ausems ME, Vozeh S, Stanski DR. Evaluating the accuracy of using population pharmacokinetic data to predict plasma concentrations of alfentanil. Anesthesiology 1998; 68: 5967
22 Schüttler J, Kloos S, Schwilden H, Stoeckel H. Total intravenous anaesthesia with propofol and alfentanil by computer-assisted infusion. Anaesthesia 1988; 43 [Suppl.]: 27[ISI]