Volatile vs Total intravenous Anaesthesia for major non-cardiac surgery: a pragmatic randomised triaL (VITAL).

BACKGROUND: Improving outcomes after surgery is a major public health research priority for patients, clinicians and the NHS. The greatest burden of perioperative complications, mortality and healthcare costs lies amongst the population of patients aged over 50 years who undergo major non-cardiac surgery. The Volatile vs Total Intravenous Anaesthesia for major non-cardiac surgery (VITAL) trial specifically examines the effect of anaesthetic technique on key patient outcomes: quality of recovery after surgery (quality of recovery after anaesthesia, patient satisfaction and major post-operative complications), survival and patient safety. METHODS: A multi-centre pragmatic efficient randomised trial with health economic evaluation comparing total intravenous anaesthesia with volatile-based anaesthesia in adults (aged 50 and over) undergoing elective major non-cardiac surgery under general anaesthesia. DISCUSSION: Given the very large number of patients exposed to general anaesthesia every year, even small differences in outcome between the two techniques could result in substantial excess harm. Results from the VITAL trial will ensure patients can benefit from the very safest anaesthesia care, promoting an early return home, reducing healthcare costs and maximising the health benefits of surgical treatments. TRIAL REGISTRATION: ISRCTN62903453. September 09, 2021.

Comparison of low-fresh gas flow technique to standard technique of sevoflurane induction in children-A randomized controlled trial.

BACKGROUND: Although sevoflurane is preferred for inhalational induction in children, financial and environmental costs remain major limitations. The aim of this study was to determine if the use of low-fresh gas flow during inhalational induction with sevoflurane could significantly reduce agent consumption, without adversely affecting induction conditions. METHODS: After institutional ethical committee approval, 50 children, aged 1-5 years, undergoing ophthalmic procedures under general anesthesia, were randomized into two groups-standard induction (Group S) and low-flow induction (Group L). A pediatric circle system with 1 L reservoir bag was primed with 8% sevoflurane in oxygen at 6 L min-1 for 30 seconds before beginning induction. In Group S, fresh gas flow was maintained at 6 L min-1 until the end of induction. In Group L, fresh gas flow was reduced to 1 L min-1 after applying facemask (time = T0). In both groups, sevoflurane was reduced to 5% after loss of eyelash reflex (T1). Once adequate depth of anesthesia was achieved (regular respiration, loss of muscle tone, and absence of movement to trapezius squeeze), intravenous access was secured (T2), followed by insertion of an appropriately sized LMA-Classic™ (T3). Heart rate and endtidal sevoflurane concentration were measured at each of the above time points, and at 15 seconds following laryngeal mask airway insertion (T4). The total amount of sevoflurane consumed during induction was recorded. RESULTS: Sevoflurane consumption was significantly lower in Group L (4.17 ± 0.70 mL) compared to Group S (8.96 ± 1.11 mL) (mean difference 4.79 [95% CI = 4.25-5.33] mL; P < 0.001). Time to successful laryngeal mask airway insertion was similar in both groups. There were no significant differences in heart rate, incidence of reflex tachycardia, or need for rescue propofol. CONCLUSION: Induction of anesthesia with sevoflurane using low-fresh gas flow is effective in reducing sevoflurane consumption, without compromising induction time and conditions.

Efficacy of Malignant Hyperthermia Association of the United States-Recommended Methods of Preparation for Malignant Hyperthermia-Susceptible Patients Using Dräger Zeus Anesthesia Workstations and Associated Costs.

BACKGROUND: The objective of this study was to assess the efficacy and cost of Malignant Hyperthermia Association of the United States-recommended methods for preparing Dräger Zeus anesthesia workstations (AWSs) for the malignant hyperthermia-susceptible patient. METHODS: We studied washout profiles of sevoflurane, isoflurane, and desflurane in 3 Zeus AWS following 3 preparation methods. AWS was primed with 1.2 minimum alveolar concentration anesthetic for 2 hours using 2 L/min fresh gas flow, 500 mL tidal volume, and 12/min respiratory rate. Two phases of washout were performed: high flow (10 L/min) until anesthetic concentration was <5 parts per million (ppm) for 20 minutes and then low flow (3 L/min) for 20 minutes to identify the rebound effect. Preparation methods are as follows: method 1 (M1), changing disposables (breathing circuit, soda lime, CO2 line, and water traps); method 2 (M2), M1 plus replacing the breathing system with an autoclaved one; and method 3 (M3), M1 plus mounting 2 activated charcoal filters on respiratory limbs. Primary outcomes are as follows: time to obtain anesthetic concentration <5 ppm in the high-flow phase, peak anesthetic concentrations in the low-flow phase, and for M3 only, peak anesthetic concentration after 70 minutes of low-flow phase, when activated charcoal filters are removed. Secondary outcomes are as follows: cost analysis of time and resources to obtain anesthetic concentration <5 ppm in each method and a vapor-free Zeus AWS. Sensitivity analyses were performed using alternative assumptions regarding the costs and the malignant hyperthermia-susceptible caseload per year. RESULTS: Primary outcomes were as follows: M3 instantaneously decreased anesthetic concentration to <1 ppm with minimal impact of low-flow phase. M1 (median, 88 minutes; 95% confidence interval [CI], 69-112 minutes) was greater than M2 (median, 11 minutes; 95% CI, 9-15 minutes). Means of peak rebound anesthetic concentrations in M1, M2, and M3 were 15, 6, and 1 ppm, respectively (P < .001). Anesthetic concentration increased 33-fold (95% CI, 21-50) after removing charcoal filters (from 0.7 to 20 ppm). The choice of anesthetic agents did not impact the results. Secondary outcomes were as follows: M3 was the lowest cost when the cost of lost operating room (OR) time due to washout was included, and M1 was the lowest cost when it was not included. When the cost of lost OR time due to washout was considered the estimated cost/case of M3 was US $360 (M1, US $2670; M2, US $969; and a "vapor-free" Zeus AWS was US $930). The OR time and equipment costs represent the largest differentiators among the methods. CONCLUSIONS: Institutions in which demand for OR time has exceeded capacity should consider M3, and institutions with surplus OR capacity should consider M1.

In vitro performance of prefilled CO2 absorbers with the Zeus®.

Low fresh gas flows (FGFs) decrease the use of anesthetic gases, but increase CO2 absorbent usage. CO2 absorbent usage remains poorly quantified. The goal of this study is to determine canister life of 8 commercially available CO2 absorbent prepacks with the Zeus®. Pre-packed CO2 canisters of 8 different brands were tested in vitro: Amsorb Plus, Spherasorb, LoFloSorb, LithoLyme, SpiraLith, SpheraSorb, Drägersorb 800+, Drägersorb Free, and CO2ntrol. CO2 (160 mL min- 1) flowed into the tip of a 2 L breathing bag that was ventilated with a tidal volume of 500 mL, a respiratory rate of 10/min, and an I:E ratio of 1:1 using the controlled mechanical ventilation mode of the Zeus® (Dräger, Lubeck, Germany). In part I, canister life of 5 canisters each of 2 different lots of each brand was determined with a 350 mL min- 1 FGF. Canister life is the time it takes for the inspired CO2 concentration (FICO2) to rise to 0.5%. In part II, canister life was measured accross a FGF range of 0.25 to 4 L min- 1 for Drägersorb 800+ (2 lots) and SpiraLith (1 lot). In part III, the calculated canister life per 100 g fresh granule content of the different brands was compared between the Zeus and (previously published data for) the Aisys. In vitro canister life of prefilled CO2 absorber canisters differed between brands, and depended on the amount of CO2 absorbent and chemical composition. Canister life expressed as FCU0.5 (the fraction of the canister used per hour) was proportional to FGF over 0.2-2 L min-1 range only, but was non-linear with higher FGF: FCU0.5 was larger than expected with FGF > 2 L min-1, and even with FGF > minute ventilation FCU0.5 did not become zero, indicating some CO2 was being absorbed. Canister life on a per weight basis of the same brand is higher with the Zeus than the Aisys. Canister life of prefilled CO2 absorber canisters differs between brands. The FCU0.5-FGF relationship is not linear across the entire FGF range. Canister life of prepacks of the same brand for the Zeus and Aisys differs, the exact etiology of which is probably multifactorial, and may include differences in the absolute amount of absorbent and different rebreathing characteristics of the machines.

COMPARISON OF ISOFLURANE AND SEVOFLURANE FOR SHORT-TERM ANESTHESIA IN MEERKATS (SURICATA SURICATTA)-ARE THERE BENEFITS THAT OUTWEIGH COSTS?

Meerkats ( Suricata suricatta ) are routinely anesthetized with isoflurane in zoo and field settings. Twenty healthy adult meerkats of mixed age and sex held in the Zoological Society of London's collection were anesthetized with 4% isoflurane by face mask for routine health examinations. The procedure was repeated 5 mo later in the same group of animals utilizing sevoflurane at 5% for induction, and again 3 mo later with sevoflurane at 6.5% for induction to approximate equipotency with isoflurane. The speed and quality of induction and recovery were compared between the two volatile anesthetic agents. There was no statistically significant difference in the speed of induction across any of the anesthetic regimes. There was a significant difference in recovery times between isoflurane and 6.5% sevoflurane (427 ± 218 and 253 ± 65 sec, respectively [mean ± SD]). Under the conditions of this study, sevoflurane at 6.5% induction dose resulted in better quality induction and recovery than sevoflurane at 5% induction or isoflurane. The mean heart and respiratory rates during anesthesia were higher using 5% sevoflurane for induction but there was no significant difference in either rate between isoflurane and sevoflurane used at a 6.5% induction dose. This study suggests that sevoflurane at a dose of 6.5% for induction and 4% for maintenance is a safe and effective anesthetic agent in healthy adult meerkats. Rapid return to normal behavior after anesthesia is important in all zoo species but particularly so in animals with a complex social and hierarchical structure such as meerkats. For this species, the advantage afforded by the speed of recovery with sevoflurane may offset the cost in certain circumstances.

Flavored Anesthetic Masks for Inhalational Induction in Children.

OBJECTIVES: To evaluate the clinical efficacy of masking the odor of inhalational agents using fruit flavors on the anxiety behavior and compliance of children for inhalational induction. METHODS: A prospective randomized double blind, placebo controlled study was conducted on 60 unpremedicated children in the age group of 4-12 y. Thirty children received anesthetic masks smeared with a flavor of child's choice while the other 30 children were induced using masks without flavor. Anxiety was assessed using modified Yale Pre-operative Anxiety Scale (mYPAS) in the pre-op room and during inhalational induction. Mask acceptance was graded by Induction Compliance Checklist (ICC). The cost-effectiveness of flavored anesthetic masks was compared to that of commercially available pre-scented masks. RESULTS: The baseline anxiety in the two groups was comparable. The number of children demonstrating high levels of anxiety at anesthetic induction was similar in flavored and non-flavored mask groups (p 0.45). The compliance to mask induction was also equally good (p 0.99). The authors found significant difference in the cost of flavored mask (INR 56.45 per mask) as compared to commercially available pre-scented masks (INR 660 per mask). CONCLUSIONS: The authors observed a placebo effect that reduced the pre-op anxiety in the control group which probably made the quality of induction equivalent with flavored and non-flavored masks. Therefore, using a flavored anesthetic mask is cost-effective than using a commercially available pre-scented mask.

Target-controlled inhalation anaesthesia: A cost-benefit analysis based on the cost per minute of anaesthesia by inhalation.

BACKGROUND: End-tidal target-controlled inhalational anaesthesia (TCIA) with halogenated agents (HA) provides a faster and more accurately titrated anaesthesia as compared to manually-controlled anaesthesia. This study aimed to measure the macro-economic cost-benefit ratio of TCIA as compared to manually-controlled anaesthesia. METHODS: This retrospective and descriptive study compared direct drug spending between two hospitals before 2011 and then after the replacement of three of six anaesthesia machines with TCIA mode machines in 2012 (Aisys carestation®, GE). The direct costs were obtained from the pharmacy department and the number and duration of the anaesthesia procedures from the computerized files of the hospital. RESULTS: The cost of halogenated agents was reduced in the hospital equipped with an Aisys carestation® by 13% as was the cost of one minute of anaesthesia by inhalation (€0.138 and €0.121/min between 2011 and 2012). The extra cost of the implementation of the 3 anaesthesia machines could be paid off with the resulting savings over 6 years. DISCUSSION: TCIA appears to have a favourable cost-benefit ratio. Despite a number of factors, which would tend to minimise the saving and increase costs, we still managed to observe a 13% savings. Shorter duration of surgery, type of induction as well as the way HA concentration is targeted may influence the savings results obtained.

Calculation of volatile anaesthetics consumption from agent concentration and fresh gas flow.

BACKGROUND: The assessment of volatile agents' consumption can be performed by weighing vapourisers before and after use. This method is technically demanding and unavailable for retrospective analysis of anaesthesia records. Therefore, a method based on calculations from fresh gas flow and agent concentration is presented here. METHODS: The presented calculation method herein enables a precise estimation of volatile agent consumption when average fresh gas flows and volatile agent concentrations are known. A pre-condition for these calculations is the knowledge of the vapour amount deriving from 1 ml fluid volatile agent. The necessary formulas for these calculations and an example for a sevoflurane anaesthesia are presented. RESULTS: The amount of volatile agent vapour deriving from 1 ml of fluid agent are for halothane 229 ml, isoflurane 195 ml, sevoflurane 184 m, and desflurane 210 ml. The constant for sevoflurane is used in a fictitious clinical case to exemplify the calculation of its consumption in daily routine resulting in a total expenditure of 23.6 ml liquid agent. CONCLUSIONS: By application of the presented specific volatile agent constants and equations, it becomes easy to calculate volatile agent consumption if the fresh gas flows and the resulting inhaled concentration of the volatile agent are known. By this method, it is possible to extract data about volatile agent consumption both ways: (1) retrospectively from sufficiently detailed and accurate anaesthesia recordings, as well as (2) by application of this method in a prospective setting. Therefore, this method is a valuable contribution to perform pharmacoeconomical surveys.

Labour time required for piglet castration with isoflurane-anaesthesia using shared and stationary inhaler devices.

Isoflurane-anaesthesia combined with an analgesic represents a welfare-friendly method of pain mitigation for castration of piglets. However, it requires an inhaler device, which is uneconomic for small farms. Sharing a device among farms may be an economical option if the shared use does not increase labour time and the resulting costs. This study aimed to investigate the amount and components of labour time required for piglet castration with isoflurane anaesthesia performed with stationary and shared devices. Piglets (N = 1579) were anaesthetised with isoflurane (using either stationary or shared devices) and castrated.The stationary devices were used in a group (n = 5) of larger farms (84 sows/farm on an average), whereas smaller farms (n = 7; 32 sows/farm on an average) shared one device. Each farm was visited four times and labour time for each process-step was recorded. The complete process included machine set-up, anaesthesia and castration by a practitioner, and preparation, collection and transport of piglets by a farmer. Labour time of the complete process was increased (P = 0.012) on farms sharing a device (266 s/piglet) compared to farms using stationary devices (177 s/ piglet), due to increased time for preparation (P = 0.055), castration (P = 0.026) and packing (P = 0.010) when sharing a device. However, components of the time budget of farms using stationary or shared devices did not differ significantly (P > 0.05). Cost arising from time spent by farmers did not differ considerably between the use of stationary (0.28 Euro per piglet) and shared (0.26 Euro) devices. It is concluded that costs arising from the increased labour time due to sharing a device can be considered marginal, since the high expenses originating from purchasing an inhaler device are shared among several farms.

Sevoflurane induction procedure: cost comparison between fixed 8% versus incremental techniques in pediatric patients.

This study compared 2 well-accepted and safe methods of pediatric inhalation induction using sevoflurane. Incremental and fixed 8% induction methods were evaluated for economic outcomes by comparing the amount of liquid sevoflurane consumed. We also tried to establish the relation between cost of induction and demographic parameters in both groups. One hundred pediatric patients scheduled for ophthalmologic examination under anesthesia were randomly divided into 2 equal groups. The amount of sevoflurane consumed in both groups was computed using the Dion method. Although the time to loss of consciousness was significantly lower using the 8% method (75.98 vs 135 seconds), the liquid sevoflurane consumption using the incremental method (2.25 mL) was almost half that of the fixed 8% method (4.46 mL). The overall procedural cost of induction (loss of consciousness plus intravenous cannulation and insertion of a laryngeal mask airway) was also almost double using the fixed 8% method. Use of the incremental method preferably over the fixed 8% method could save almost $18 US for each procedure. The volume of sevoflurane consumed during anesthesia induction was found to be independent of age, weight, or sex of pediatric patients. Both induction methods proved to be equally safe and acceptable to the patients.

[Impact of the decrease of nitrous oxide use on the consumption of halogenated agents]. / Impact de la diminution d'utilisation du protoxyde d'azote sur la consommation d'agents halogénés.

OBJECTIVE: Nitrous oxide (N2O) toxicity and its impact on pollution lead to restrict its use. A decrease of N2O consumption should increase the hypnotic inhaled consumption. This monocentric study estimated consumptions and costs of halogenated agents (HA) and N2O over 5 years when the N2O consumption was reduced. STUDY DESIGN: Retrospective from a computerized database. PATIENTS: Between 2006 and 2010, 34,097 procedures were studied after two meetings exposing the risks of the N2O. METHODS: At the end of anesthesia, consumptions of hypnotic agents (millilitres transmitted by the injectors and the blender) were archived in the database. The annual consumption of agents was obtained by adding the individual consumptions, then divided by the annual number of cases. The costs were given by the hospital pharmacy from invoices. RESULTS: N2O consumption per anesthesia constantly decreased during the study, from 75.1L by act to 22.7L. The sum of the annual consumptions of N2O and air did not change suggesting that total fresh gas flow remained stable. Between 2006 and 2010, the sevoflurane consumption by act increased by 25%, from 16.5 to 20.6mL, and desflurane consumption by 37%, from 46.1 to 63.1mL by patient. The costs of the administration of hypnotic agents remained stable. CONCLUSION: N2O consumption decrease had an impact on the consumption of HA. The cost reduction of the N2O was counterbalanced by the increase of halogenated vapor cost. The profit of the ecological impact of the reduction in N2O use could be quantified.

Economic considerations in the use of inhaled anesthetic agents

Purpose: To describe the components of and factors contributing to the costs of inhaled anesthesia, basis for quantifying and comparing these costs, and practical strategies for performing pharmacoeconomic analyses and reducing the costs of inhaled anesthetic agents. Summary: Inhaled anesthesia can be costly, and some of the variable costs, including fresh gas flow rates and vaporizar settings, are potential targets for cost savings. The use of a low fresh gas flow rate maximizes rebreathing of exhaled anesthetic gas and is less costly than a high flow rate, but it provides less control of the level of anesthesia. The minimum alveolar concentration (MAC) hour is a measure that can be used to compare the cost of inhaled anesthetic agents at various fresh gas flow rates. Anesthesia records provide a sense of patterns of inhaled anesthetic agent use, but the amount of detail can be limited. Cost savings have resulted from efforts to reduce the direct costs of inhaled anesthetic agents, but reductions in indirect costs through shortened times to patient recovery and discharge following the judicious use of these agents are more difficult to demonstrate. The patient case mix, fresh gas flow rates typically used during inhaled anesthesia, availability and location of vaporizers, and anesthesia care provider preferences and practices should be taken into consideration in pharmacoeconomic evaluations and recommendations for controlling the costs of inhaled anesthesia. Conclusion: Understanding factors that contribute to the costs of inhaled anesthesia and considering those factors in pharmacoeconomic analyses and recommendations for use of these agents can result in cost savings (AU)

Volatile anesthetic rescue therapy in children with acute asthma: innovative but costly or just costly?.

OBJECTIVES: To describe volatile anesthesia (VA) use for pediatric asthma, including complications and outcomes. DESIGN: Retrospective cohort study. SETTING: Children's hospitals contributing to the Pediatric Health Information System between 2004-2008. PATIENTS: Children 2-18 years old with a primary diagnosis code for asthma supported with mechanical ventilation. INTERVENTION: Those treated with VA were compared to those not treated with VA or extracorporeal membrane oxygenation. Hospital VA use was grouped as none, <5%, 5-10% and >10% among intubated children. MEASUREMENTS AND MAIN RESULTS: One thousand five hundred and fifty-eight patients received mechanical ventilation at 40 hospitals for asthma: 47 (3%) received VA treatment at 11 (28%) hospitals. Those receiving a VA were significantly less likely to receive inhaled b-agonists, ipratropium bromide, and heliox, but more likely to receive neuromuscular blocking agents than patients treated without VA. Length of mechanical ventilation, hospital stay (length of stay [LOS]) and charges were significantly greater for those treated with VA. Aspiration was more common but death and air leak did not differ. Patients at hospitals with VA use >10% were significantly less likely to receive inhaled b agonist, ipratropium bromide, methylxanthines, and heliox, but more likely to receive systemic b agonist, neuromuscular blocking agents compared to those treated at hospitals not using VA. LOS, duration of ventilation, and hospital charges were significantly greater for patients treated at centers with high VA use. CONCLUSIONS: Mortality does not differ between centers that use VA or not. Patients treated at centers with high VA use had significantly increased hospital charges and increased LOS.

Comparative cost analysis of three different anesthesia methods in gynecological laparoscopic surgery.

In the current study, we assessed and evaluated the costs and benefits of three popular methods of general anesthesia practiced in our department for gynecological laparoscopic surgery in recent years. Sixty adult female patients who underwent elective gynecological laparoscopic surgery under general anesthesia were randomly divided into three groups: group V, group I and group C. In group V, anesthesia was induced intravenously with midazolam, remifentanil, propofol and vecuronium, and maintained with continuous infusion of propofol and remifentanil. In group I, anesthesia was intravenously induced with midazolam, fentanyl, propofol and vecuronium, and maintained with inhaled isoflurane and intravenous bonus of fentanyl. In group C, anesthesia was induced as in group I, but maintained with isoflurane inhalation combined with propofolremifentanil infusion. All patients received vecuronium for muscle relaxation. Perioperative incidences of complications and total anesthesia costs for patients in all groups were recorded. In addition, postoperative satisfaction of the patients was also noted, and similar outcomes of the satisfaction were reported in all 60 patients. Although there was no statistical significance among groups, the incidence of postoperative nausea and vomiting were higher in group C, and the rates of shivering and the needs for analgesics were higher in group V. Anesthesia costs in group I were the lowest. Therefore, it is concluded that the costs of anesthesia induced with midazolam, fentanyl, propofol, vecuronium, and maintained with isoflurane, fentanyl and vecuronium are cheapest, and there is no significant difference in patients' satisfaction and safety among the three above-mentioned methods of anesthesia in our department.

Inhalational or total intravenous anaesthesia: is total intravenous anaesthesia useful and are there economic benefits?

PURPOSE OF REVIEW: The comparison of inhalational and intravenous anaesthesia has been the subject of many controlled trials and meta-analyses. These reported diverse endpoints typically including measures of the speed and quality of induction of anaesthesia, haemodynamic changes, operating conditions, various measures of awakening, postoperative nausea and vomiting and discharge from the recovery area and from hospital as well as recovery of psychomotor function. In a more patient-focused Health Service, measures with greater credibility are overall patient satisfaction, time to return to work and long-term morbidity and mortality. In practice, studies using easier to measure proxy endpoints dominate - even though the limitations of such research are well known. RECENT FINDINGS: Recent study endpoints are more ambitious and include impact on survival from cancer and the possibility of differential neurotoxic impact on the developing brain and implications for neuro-behavioural performance. SUMMARY: Economic analysis of anaesthesia is complex and most published studies are naive, focusing on drug acquisition costs and facility timings, real health economics are much more difficult. Preferred outcome measures would be whole institution costs or the ability to reliably add an extra case to an operating list, close an operating room and reduce the number of operating sessions offered or permanently decrease staffing. Alongside this, however, potential long-term patient outcomes should be considered.

Pharmacoeconomics of volatile inhalational anaesthetic agents: an 11-year retrospective analysis.

With continuously increasing expenditure on health care resources, various cost containment strategies have been suggested in regard to controlling the cost of inhalational anaesthetic agents. We performed a cost identification analysis assessing inhalational anaesthetic agent expenditure at a tertiary level hospital, along with an evaluation of strategies to contain the cost of these agents. The number of bottles of isoflurane, sevoflurane and desflurane used during the financial years 1997 to 2007 was retrospectively determined and the acquisition costs and cumulative drug expenditure calculated. Pharmacoeconomic modelling using low fresh gas flow anaesthesia was performed to evaluate practical methods of cost reduction. The use of isoflurane decreased from 384 bottles during 1997 to 204 in 2007. In contrast, use of sevoflurane increased from 226 bottles during 1998 to 875 during 2007. Desflurane use increased from 34 bottles per year during 2002 (its year of introduction) to 163 bottles per year in 2007. While the inflation-adjusted cumulative expenditure for these inhalational agents (Australian dollars) increased from $132,000 in 1997 to over $326,000 in 2007, an increase of 168%, patient workload over the same period increased by only 11%. Pharmacoeconomic modelling demonstrated that sevoflurane at 2 l/minute costs 19 times more than isoflurane at 0.5 l/minute. For the financial years 1997 to 2007, we found a progressive shift from the cheaper isoflurane to the more expensive agents, sevoflurane and desflurane, a shift associated with marked increases in costs. Low flow anaesthesia with isoflurane is one strategy to reduce costs.