International Journal of ClinicalMonitoring and Computing 9: 17%182,1992. 9 1992KluwerAcademic Publishers. Printedin the Netherlands. A portable target controlled propofol infusion system Gavin NC Kenny & Martin White University Department of Anaesthesia, Royal Infirmary, 8-16 Alexandra Parade, Glasgow G31 2ER, UK Accepted 7May 1992 Key words: computer, control, model, pharmacokinetic, propofol, target Abstract A portable target controlled infusion system for propofol has been developed based on a Psion hand-held microcomputer and the Ohmeda 9000 syringe pump. The system uses a pharmacokinetic model which describes the distribution and elimination of propofol to achieve and maintain any selected target blood concentration. Target blood concentrations of 1/zg/ml, 3/xg/ml and 5/xg/ml were selected in laboratory trials and the cumulative volumes delivered by the Psion system each minute were compared with the theoretical output calculated by the pharmacokinetic model. The results Obtained showed that the computer system delivered volumes which were always within 2% of the theoretical values. This system offers a convenient and simple method of maintaining anaesthesia using propofol. Introduction Propofol has a favourable pharmacokinetic profile for use as a total intravenous anaesthetic [1] but the use of intravenous agents to maintain anaesthesia requires a different technique compared with standard volatile anaesthetics delivered via calibrated vaporisers. Several regimes have been suggested for suitable propofol infusions to maintain anaesthesia [2-4] but these cannot alter accurately the propofol concentration with changing surgical and anaesthetic requirements. The wide range of blood propofol concentrations necessary to produce satisfactory levels of anaesthesia have been reported by Glass and his colleagues [5] who used a computer-based target controlled infusion system. This technique would appear to offer the simplest method of achieving satisfactory levels of anaesthesia when propofol is used as the maintenance agent. We have described previously a target controlled infusion system for propofol [6] which enabled the anaesthetist to aiter the depth of anaesthesia in as simple a manner as that used for volatile agents. The pharmacokinetic parameters which determine the distribution and elimination of propofol [1] were modified [7] and incorporated into a computerised delivery system. The device was capable at all times of calculating the distribution of the drug between the three compartments of a mathematical pharmacokinetic model and allowed the target concentration to be manipulated in any direction at any point in time. This apparatus allowed the anaesthetist to achieve and maintain the required target blood concentration of propofol appropriate to any patient and surgical stimulation. It was based on an Atari 1040ST microcomputer connected to an Imed 929 computer controlled infusion pump via an RS-232 interface. However, the principal disadvantages of this propofol infusion system were: (a) the bulk of the trolley required for the computer (b) electrical power had to be supplied continually to prevent loss of data 180 (c) the Imed 929 pump had the limitation that it could only deliver infusion rates in steps of 1 ml h -~ and volumes to the nearest 1 ml. In spite of these problems, the system performed satisfactorily in routine clinical practice and greatly simplified the use of propofol to induce and maintain anaesthesia. A further development is reported using a hand-held microcomputer interfaced to the Ohmeda syringe pump to produce a portable target controlled infusion system for propofol. Methods The new propofol infusion system is based on the Psion Organiser which is a hand-held microcomputer with an RS-232 interface. The model used is the POS 200 which has a simplified numerical keyboard with no alphabetical keys and is therefore easier to use than the standard version of the Psion Organiser which has a full alphanumeric keyboard. However, a standard Psion Organiser will also function satisfactorily. The Ohmeda 9000 syringe pump uses a range of standard syringes and has selectable 'bolus' infusion rates of 1200, 600 and 300 ml h -1. It can deliver infusion rates in steps of 0.1 ml h -1 and volumes to the nearest 0.1ml. The Psion is connected to the pump via the interface backbar which was produced specially by Ohmeda. This incorporates a rechargable battery source which supplies continuous power for the RS-232 interface in the backbar, electrical supply for the Psion Organiser when mains power is supplied and, in addition, a convenient mounting point for the computer. A standard Ohmeda 9000 backbar fitted with the RS-232 interface can also be used in place of the special backbar. The standard backbar must be supplied with a constant + 12 volts and this is achieved in our system by modifying the Psion communications link to divert power from the mains adaptor. The control program has been written in the Organiser Programming Language and starts automatically when the Psion is switched on. The patient's body weight, age and the initial target blood concentration of propofol are entered into the Psion. The target concentration is achieved by an infusion of propofol at a rate of 1200 ml per hour. The program calculates the predicted blood concentration achieved during this rapid infusion approximately every second and once this concentration is within 1% of the selected target, it is kept at this value by infusing propofol at the rate required to compensate for elimination of the drug and redistribution to the peripheral compartments. Simple controls are used to either stop the infusion or change the target propofol concentration. An increased depth of anaesthesia is achieved by entering a higher target concentration. The system then delivers an infusion rate of 1200 ml/hr until the selected target concentration has been reached. When a lower target is entered, the infusion pump is stopped until the new concentration has been achieved by the processes of elimination and redistribution. The selected target concentration is then maintained automatically. The accuracy of the Psion controlled infusion system was assessed by comparing the cumulative volume delivered by the pump with the theoretical cumulative volume calculated by the pharmacokinetic model used in the Atari system. Target blood propofol concentrations of 1/xg/ml, 3 txg/ml and 5 ~g/ml were set on the Psion controlled infusion system. The pump was fitted with a 50 ml syringe containing water and the outflow delivered by the system was collected and weighed continuously on a digital balance (Sartorius model 1265; accuracy and reproducibility 1 mg). It was found that when slow infusion rates were transmitted frequently to the pump, delivery of the fluid was considerably less than anticipated. This was caused by the requirement for the pump motor to have advanced by a certain amount before receiving the next command to set the pump infusion rate. The fault was corrected by transmitting the required infusion rate to the pump only if it was different from the rate set previously. Three runs at each of the selected blood concentrations were performed the results are shown in Fig. 1. The volumes delivered by the computer controlled system compare favourably with those calculated by the model and were always within 2% of the theoretical values. 181 DOSE (ug) 600000 5ug/ml 500000 400000 3ug/ml 300000 200000 100000 luglml 0 10 20 30 4~0 TIME (rain) Fig. 1. Cummulative output of the Psion-Ohmeda infusion system (circles) compared with the theoretical dose predicted by the pharmacokinetic model (crosses) to achieve three different target blood concentrations of propofol. Discussion The Atari based system which had been developed previously had the capability of rapidly manipulating the target blood concentration of propofol in accordance with the varying anaesthetic and surgical requirements. It was used in a similar manner to that in which a vaporiser was employed to achieve and maintain a given anaesthetic level but was bulky and required continuous mains power. The Psion based system was found to have the problem of frequent transmission of infusion rates to the pump preventing correct delivery. The fault was corrected and the new system functioned satisfactorily thereafter when fitted with a 50 ml syringe. This emphasises the importance of testing any development which uses new software or hardware against a system which has been demonstrated previously to function correctly. The new propofol infusion system has the major advantage over the previous Atari system of much reduced size and complete portability. It can there- fore be used to transport a patient while the depth of anaesthesia or sedation [7] is controlled at the required level. The lack of graphical display is a disadvantage since the relationship of alterations in infusion rates and blood propofol concentrations with time cannot be understook as easily. However, the benefits of cost and size more than compensate for this. Over eighty Psion controlled propofol infusion systems are now in use in eleven countries. Those anaesthetists who have access to the system have commented on its ease of use and the simplicity of maintenance of the required depth of anaesthesia with propofol. It has been reported to have provided good control in difficult anaesthetic situations such as in the presence of a pneumothorax [8] and bronchopleural fistula [9]. A similar infusion system has been developed for the delivery of atfentanil and, together with the propofol system, enables total intravenous anaesthesia to be achieved with relative simplicity and ease of control. 182 References 1. Gepts E, Jonckheer K, Maes V, Sonek W, Camu F. Disposition kinetics of propofol during alfentanil anaesthesia. Anaesthesia 1988; 43 Suppl: 8-13. 2. de Grood PMRM, Ruys AHC, van Egmond J, Booij LHDJ, Crul JF. Propofol ('Diprivan') emulsion for total intravenous anaesthesia. Postgrad Med J 1985; 61: Suppl 3: 65-9. 3. Turtle MJ, Cullen P, Prys-Roberts C, Coates DP, Monk CR, Faroqui MH. Dose requirements of propofol by infusion during nitrous oxide anaesthesia in man. Br J Anaesth 1987; 59: 283-7. 4. Roberts FL, Dixon J, Lewis GTR, Tackley RM, PrysRoberts C. Induction and maintenance of propofol anaesthesia: A manual infusion scheme. Anaesthesia 1988; 43 Suppl: 14-7. 5. Glass PSA, Markham K, Ginsberg B, Hawkins ED. Propofol concentrations required for surgery. Anesthesiology 1989; 71 Suppl: A273. 6. White M, Kenny GNC. Intravenous propofol anaesthesia 7. 8. 9. 10. using a computerised infusion system. Anaesthesia 1990; 45: 204-9. Marsh B, White M, Morton, Kenny GNC. Pharmacokinetic model driven infusion of propofol in children. Br J Anaesth 1991; 67: 41-8. Church JA, Stanton PD, Kenny GNC, Anderson JR. Propofol for sedation during endoscopy: Assessment of a computer-controlled infusion system. Gastrointest Endosc 1991; 37: 175-9. Crofts SL, Hutchison GL. General anaesthesia and undrained pneumothorax. The use of a computer-controlled propofol infusion. Anaesthesia 1991; 46: 192-4. Donnelly JA, Webster RE. Computer-controlled anaesthesia in the management of bronchopleural fistula. Anaesthesia 1991; 46: 383-4. Address for offprints: G.N.C. Kenny, University Department of Anaesthesia, Royal Infirmary, 8-16 Alexandra Parade, Glasgow G31 2ER, UK