Intravenous (IV) anaesthetics have been used for decades in animal species as part of a general anaesthetic technique, most commonly to induce general anaesthesia prior to maintenance with inhalational anaesthetic agents. The barbiturate thiopentone has been in widespread use in animals as an induction agent for several decades, primarily due to its rapid onset of action. However, its slow clearance rate which leads to cumulation upon repeated administration makes it unsuitable for use as an anaesthetic for maintenance of anaesthesia for all but the shortest of procedures. In recent years the use of IV agents to maintain general anaesthesia has become more commonplace because of the development of IV anaesthetic agents with pharmacokinetic profiles suitable for use by infusion. Total intravenous anaesthesia (TIVA) is now a popular technique in humans, as it offers advantages over inhalational anaesthesia, eliminating the hazard of atmospheric pollution, and thus reducing operator exposure and environmental exposure, and allowing easy control over the depth of anaesthesia. Total intravenous anaesthesia has also been developed and used in animals, most notably in dogs and horses, and several combinations of anaesthetic / analgesic / hypnotic drugs have been described for use clinically. Of the IV anaesthetic agents in use today in animals, propofol is a drug, which is suitable for use as part of a TIVA regime, due to its rapid clearance and short duration of action. Propofol, supplied as a 1% emulsion, is a short acting drug that is rapidly metabolized. It is rapidly distributed and has a high metabolic clearance. The elimination half-life of propofol in humans and in dogs has been detailed in many studies, which have indicated a wide range in both man and the dog (up to 6h in dogs, 24h in humans) however, in spite of the long half-lives reported, anaesthetised patients exhibit short recovery times, even after relatively long infusion periods. Furthermore the recovery period is generally characterised by its high quality. Some adverse effects have been reported which include muscle twitching and opisthotonos during anaesthesia.

Nolan and Hall (1985) described the use of TIVA with propofol in ponies and reported satisfactory anaesthesia, and good quality recovery. Since then the use of propofol by infusion to maintain anaesthesia has been reported in many species (Nolan & Hall, 1985; Hall & Chambers, 1987; Nolan & Reid, 1993; Correia et al., 1996; Flaherty et al., 1997; Bettschart-Wolfensberger et al., 2001). Propofol causes a transient decrease in arterial blood pressure due mainly to a decrease in peripheral vascular resistance decreased sympathetic outflow, and myocardial depression. Arterial hypotension does not usually result in reflex tachycardia under propofol anaesthesia, and a decrease in heart rate is more commonly observed. Respiratory depression and apnoea are the most common adverse effects associated with IV administration of propofol and work in dogs indicated that when used alone, blood propofol levels in excess of 6 mg/ml are associated with apnoea (Beths et al., 2001). The use of propofol as the sole agent for TIVA is generally unsatisfactory, since the concentration levels required to eliminate responses to surgery induce cardiovascular and respiratory depression. Consequently, it is necessary to supplement the use of propofol with an analgesic drug. These drugs include the opioid analgesics fentanyl and alfentanil and more recently remifentanil, while infusions of propofol with other agents such as ketamine, and the alpha2 adrenoceptor agonist, medetomidine have been described (Correia et al., 1996; Nolan et al., 1996; Flaherty et al., 1997; Hughes & Nolan, 1999).

Delivery of TIVA in animals has been largely by infusion pumps / syringe drivers of different degrees of sophistication, where drugs are infused to a desired clinical effect, although regimes using infusion sets are often used in horses. Blood/plasma levels of individual anaesthetic and adjunct drugs have been measured in some species, and these studies have described pharmacokinetic parameters, which can be used to calculate general infusion rates to achieve desired blood/plasma concentrations (e.g., infusion rate of drug = body clearance x plasma/blood concentration at steady state). The relationship between plasma-blood drug levels and clinical effects in anaesthesia have been studied more widely in humans than in animals, however it is clear that the blood levels of propofol required for surgical anaesthesia are highly variable in humans and in dogs, depending on adjuvant therapy, patient status and procedure (Shafer, 1993; Nolan & Reid, 1993).

Intravenous drugs are sometimes delivered by intermittent bolus injections, where the anaesthetic period is short. Administration by continuous infusion however, is clearly optimal, as it allows the achievement of more stable blood/plasma drug concentrations, which in turn reduces the likelihood of either over dosing or under dosing. Using inhalational anaesthetic agents, actual anaesthetic concentrations can be measured in real time (measurement of end-tidal concentrations of anaesthetic drug from a site in the endotracheal tube), and this information on drug concentrations is continually available to the anaesthetist, thus drug administration is more controlled. However, this task is more difficult for intravenous anaesthesia, because drug plasma/blood concentrations cannot be easily measured in real time.

In human anaesthesia, developments in estimating plasma / blood drug concentrations by using predictions provided by pharmacokinetic and pharmacodynamic models have afforded greater control in the delivery of IV anaesthetics. These data along with developments in mathematical modeling have been translated into the development of computer-controlled delivery of anaesthetic drugs, now termed 'Target Controlled Infusions' (TCI). Using pharmacokinetic parameters of drugs, computer programming may be carried out to permit calculation of the dosage regimens necessary to achieve an effective target blood/plasma concentration. TCI systems allow automatic adjustments of drug delivery (by controlling the infusion rate) to maintain a desired drug concentration, to increase or decrease it. A TCI system comprises a computer programmed with appropriate pharmacokinetic data and a pharmacokinetic model, which controls an infusion pump/syringe driver. The anaesthetist enters some patient details along with the desired drug plasma/ blood concentration. The computer translates predictions from the pharmacokinetic model into instructions to control the infusion device, which then delivers the required infusion rate (Van den Nieuwenhuyzen et al., 2000). There is a continual recalculation of infusion rates, and in this way, accurate control over plasma/ drug concentrations is achieved over the duration of the infusion. In an individual patient, titration of the target concentration is carried out by the anaesthetist/veterinary surgeon, in the same manner as a calibrated anaesthetic vaporizer is adjusted for an inhalational agent, to obtain a desired effect.

Current TCI systems available for use in humans are easy-to use devices, which comprise a fully integrated system with the computer software incorporated into the syringe pump. The 'Diprifusor' TCI system (Astra Zeneca Ltd) was developed as a standardised infusion system for the administration of propofol in humans (Glen, 1998). The associated software consists of a pharmacokinetic model, a set of specific pharmacokinetic variables for propofol and infusion control algorithms. A set of pharmacokinetic parameters for propofol was selected using computer simulation of a known infusion scheme with pharmacokinetic parameters described in published literature. The selected model was then programmed in to the Diprifusor 'computer'. Subsequently validation studies were carried out and information gathered which led to guidance on appropriate target concentrations for the administration of propofol by 'Diprifusor' TCI. Trials of computerised TCI systems studying performance in patients are undertaken for systems and drugs in order to assess the predictive performance of the system incorporating the selected pharmacokinetic variables. This is carried out by comparing the propofol concentrations predicted by the system with the measured drug concentrations in blood samples taken at various time points during anaesthesia, over a range of target concentrations (Varvel et al., 1992; Coetzee et al., 1995). Other TCI systems used in humans include the Aneo TIVAS, the Braun OTCI system and the Fresenius system. These all have models for TCI propofol with additional variations (TCI remifentanil and sufentanil; manual remifentanil), but the pharmacokinetic models used for TCI propofol vary.

TIVA using TCI propofol was recently described in animals for the first time (Beths et al., 2001). A prototype system was developed by selecting a set of parameters that described the pharmacokinetics of propofol in the dog. Using the administration scheme for intravenous propofol in the dog described by Nolan & Reid (1993) and pharmacokinetic parameters from selected published studies as inputs, computer simulation using the program PK-SIM (Specialised Data Systems, Jenkintown, PA, USA) was used to obtain predicted propofol concentration profiles to compare with the measured profile. None of the published pharmacokinetic models of propofol in dogs provided an accurate prediction of the measured blood concentration profile, thus empirical adjustments were made and the performance of the system was investigated in dogs by comparing the predicted concentrations of propofol in venous blood samples with direct measurements of blood propofol concentrations taken at various time points during anaesthesia. Using the methodology described by Varvel et al. (1992) to evaluate the performance of a TCI system, the model developed in dogs was evaluated and the data indicated that that the system developed by Beths and colleagues (2001) performed satisfactorily.

Current trends in anaesthesia suggest that infusions of analgesic drugs such as alfentanil or remifentanil or medetomidine enhance the quality of anaesthesia and recovery, but may influence the pharmacokinetics of propofol (particularly in some groups of patients), and therefore the precision of propofol TCI systems. In veterinary anaesthesia, development of TCI is in its infancy. Pharmacokinetics models are available for propofol and some other analgesic / hypnotic drugs for companion animal species, however these have been described from studies in healthy animals. In order to develop TCI systems in veterinary anaesthesia, validation of these models will be required if the technology is to be exploited. The concentration-effect relationships of all the agents used during TIVA similarly should be understood in order to allow the anaesthetist to select optimal target concentrations. There is considerable work to be done to identify and understand the intraoperative anaesthetic and analgesic requirements in animals to allow the development of TCI systems to deliver optimal infusion schemes that balance intraoperative anaesthetic stability and flexibility and enhance speed and quality of recovery. However, this work is fully justified and potentially very exciting. The ease of use of computerised TCI systems and the control and precision they afford for balanced anaesthesia in the future makes them attractive for development in particular for companion animal species.

References

1.  Beths T, Glen JB, Reid J, Monteiro AM, Nolan AM (2001) Evaluation and optimisation of a target controlled infusion system for administering propofol to dogs as part of a total intravenous anaesthetic technique during dental surgery. Veterinary Record 148, 198-203.

2.  Bettschart-Wolfensberger R, Bowen MI, Freeman SL, Feller R, Bettschart RW, Nolan A, Clarke KW (2001) Cardiopulmonary effects of prolonged anaesthesia via propofol-medetomidine infusion in ponies. American Journal of Veterinary Research 62, 1428-1435.

3.  Bouillon T, Bruhn J, Radu-Radulescu L, Bertaccini E, Park S, Shafer, S (2002) Non-steady state analysis of the pharmacokinetic interaction between propofol and remifentanil. Anaesthesiology 97, 1350-1362.

4.  Coetzee JF, Glen JB, Wilum CA, Boshoff L. (1995) Pharmacokinetic model selection for target controlled infusions of propofol-assessment of 3 parameter sets. Anaesthesiology 82 1328-1345.

5.  Correia D, Nolan AM, Reid, J (1996). Pharmacokinetics of propofol infusions in sheep when used alone and in combination with ketamine. Research in Veterinary Science 60, 213-217.

6.  Flaherty D, Reid J, Welsh EM, Monteiro AM, Lerche P,

7.  Nolan AM (1997). Infusions of propofol or propofol/ketamine in ponies undergoing surgery: a pharmacodynamic study. Research in Veterinary Science 52, 179-184.

8.  Glen JB (1998) The development of 'Diprifusor': a TCI system for propofol Anaesthesia 53, 13-21 Suppl.

9.  Hall LW, Chambers JP (1987) A clinical trial of propofol infusion anaesthesia in dogs. Journal of Small Animal Practice 28, 623-637.

10. Hughes JML, Nolan A (1999) Total intravenous anaesthesia in greyhounds with propofol and fentanyl. Veterinary Surgery 28, 513-524.

11. Nolan AM, Hall LW (1985) Total intravenous anaesthesia in the horse with propofol. Equine Veterinary Journal 17, 394-398.

12. Nolan A, Reid J (1993) Pharmacokinetics of propofol administered by infusion in dogs undergoing surgery. British Journal of Anaesthesia 70, 546-551.

13. Nolan AM, Reid J, Welsh EM, Flaherty D, McCormack R, Monteiro A. (1996) Simultaneous infusions of propofol and ketamine in ponies premedicated with detomidine: a pharmacokinetic study. Research in Veterinary Science 60, 262-266.

14. Shafer SL (1993). Towards optimal intravenous dosing strategies. Seminars in Anaesthesia 12, 222-234.

15. Schuttler J, Kloos S, Schwilden H, Stoeckel H (1998) Total intravenous anaesthesia with propofol and alfentanil by computer assisted infusion. Anaesthesia 43 (Supplement 1), 2-7.

16. Van den Nieuwenhuyzen, MC, Engbers, FHM, Wuyk, J, Burm, AGL (2000) Target controlled infusion systems: role in anaesthesia and analgesia. Clinical Pharmacokinetics, 38, 181-190.

17. Varvel JR, Donoho DL, Shafer SL (1992) Measuring the Predictive Performance of Computer-Controlled Infusion Pumps. Journal of Pharmacokinetics and Biopharmaceutics 20, 63-94.