We observed that most staff members of our anaesthesia department constantly use propofol in one of the two distinctly different fashions: a ‘conservative’ variant with handling of manually driven perfusor pumps, in which the dosage is expressed in mg h−1, and a ‘modern’ variant in which the propofol is administered in ‘target controlled infusion’ (TCI) mode, in which effect site concentrations are calculated according to a pharmacokinetic algorithm. As nearly all staff members adhered strictly to their dosage habits, we saw the possibility to evaluate their anaesthesia records and to check whether these two distinguishable regimens have different clinical and pharmaco-economical effects.
The available literature on the comparison of conservative propofol administration using manual infusion vs. modern application with TCI is contradictory. One investigation revealed a lower drug consumption and a shorter awakening time for TCI as compared with manual infusion.1 In contradiction, in other studies, a higher drug consumption has been found for TCI application.2,3 This is explained by the finding that TCI algorithms provide initially a higher infusion rate (decreasing by time) than manual pumps are usually set. This specific imbalance reverses to the opposite during longer periods of application. The decreasing infusion rate in TCI is intended to avoid accumulation of propofol, whereas manually controlling anaesthesiologists usually do not reproduce this process.2,4 Therefore, the duration of the procedure strongly influences the average consumption that way, that as compared with manual infusion, TCI is consuming more propofol in short procedures and less propofol in long procedures (calculated for the same amount per body weight and hour). Also, it seems plausible that the more predominantly constant drug levels during TCI application would cause less variable haemodynamic changes, although this assumption has been contradicted by some authors.4,5
A possible difference in average drug consumption has direct pharmaco-economical consequences. Even more, the duration for emergence from anaesthesia and awakening has a strong influence on the transition times between successive interventions as well as the length of stay in a postanaesthesia care unit (PACU). The occurrence of postoperative side effects may also contribute to the type of postanaesthesia care, therefore yielding direct and indirect economical implications.6,7 We were interested to recognize possible clinical and pharmaco-economical differences in a routine clinical setting for a typical and often performed operation such as gynaecological laparoscopy. We considered the restriction to just one operation type as a prerequisite to obtain standardized conditions for the retrospective mode of investigation. The retrospective approach on its part eliminates the bias caused by the unavoidable knowledge of the participants that their performance is under observation.
The data collection and analysis have been approved by the Data Protection Officer of the Hospital in compliance with the condition to use anonymized patient data only. The investigated anaesthetists gave their consent to evaluate their records. The institutional ethics committee also approved the investigation.
To ensure collection of representative data and to obtain the best possible comparability of the two prospected study groups, in the first instance, the observed anaesthesiologists had to be chosen from a group of suitable candidates from our department of anaesthesia. First, we recognized anaesthesiologists with more than 2 years of professional experience in anaesthesia whom we assumed to have sufficiently established views of anaesthetic dosage and clinical practice and who could be clearly distinguished by their behaviour patterns concerning propofol administration as either clearly ‘conservative’ or ‘modern’. From a total of 100 medical doctors who are enrolled in our department of anaesthesiology, 50 have more than 2 years of professional experience in anaesthesia. From these, 16 were assigned to work predominantly in the gynaecological operation unit for up to 6 months. Among these, we finally found 12 colleagues who could be recognized as having a stable propofol administration habit, of whom six were assigned to group C (for ‘conservative’ behaviour) and other six to group M (for ‘modern’ behaviour). Then, from each chosen anaesthesiologist, their eight most recent anaesthesia records were fetched from the archive, which met the inclusion criteria: general anaesthesia for gynaecological laparoscopy, patients with American Society of Anesthesiologists (ASA) classes 1–3, age 16–65 years, duration of anaesthesia for at least 90 min and total intravenous anaesthesia (TIVA) with propofol and opioids without additional use of other anaesthetic agents. Exclusion criteria were cases with intraoperative use of vasoactive drugs and/or average crystalloid and colloid fluid intake exceeding 10 ml kg−1 h−1. The anonymized anaesthesia protocols were photocopied and then distributed by group allocation. In this way, a total of 96 anaesthesia protocols were obtained for evaluation. Of these, 48 were derived from six ‘conservative’ anaesthesiologists (group C) and another 48 originated from six ‘modern’ anaesthesiologists (group M). In group C, propofol was administered by an initial bolus dose followed by a manually controlled infusion with a conventional perfusor pump (Asena; Alaris Cardinal Health, Dublin, Ohio, USA). The patients received routinely an oral premedication with midazolam 7.5 mg 45–60 min before induction of anaesthesia. The anaesthesias performed in group M consisted of TCI infusion based on the locally popular Schnider algorithm with effect site target control. For this purpose, TIVA-TCI (Alaris Cardinal Health) perfusor pumps were used. In both groups, control of anaesthetic dosage was performed according to the personal experience of the involved anaesthetists by monitoring the haemodynamic course and closely following the ongoing intervention in terms of invasiveness and surgical stimulation.
The following data were extracted and analysed:
- Biometric data of the patients: age, size, weight, ASA class, preoperative heart rate (HR) and blood pressure (BP).
- Surgery and anaesthesia: diagnosis and surgical intervention, duration of anaesthesia and surgery and time to wake up which was defined as time between stop of anaesthetic medication and extubation. Extubation was performed when the patient opened his eyes and commenced a sufficient spontaneous breathing.
- Intraoperative haemodynamics: minimum, maximum and average HR (average of all readings), HR variability (standard deviation of all measurements), minimum, maximum and average mean BP (average of all measurements) and variability of mean BP (standard deviation of all readings). The mean BP values were calculated from the systolic and diastolic values.
- Drug consumption: total quantity of consumed propofol, fentanyl and remifentanil. The values per time and body weight (mg kg−1 h−1) were calculated.
- Postoperative adverse events during the first 4 h after end of anaesthesia: number of postoperative nausea and vomiting (PONV) episodes and phases with additional need of pain treatment [rest pain with intensity >3 on a verbal rating scale (VRS) from 0 to 10].
The course of haemodynamic parameters was recorded in 5-min intervals. Because of the different time lengths of the surgical procedures, we extracted 10 absolute measurement times (the first 10 readings in 5-min intervals from beginning the anaesthesia), and a further seven were introduced relatively to characteristic moments of the anaesthesia course such as M for the middle of the anaesthesia, M − 5 and M + 5 for measurements 5 min before and after the middle of the anaesthesia, L for the last reading after the end of anaesthesia, and L − 15, L − 10 and L − 5 for the three last 5-min intervals before. This way, a comparison of all anaesthesias with their most characteristic time periods was possible.
The obtained data were collected in an Excel file and sorted by group affiliation. The applied statistics software was SPSS (version 13; SPSS Inc., Chicago Illinois, USA). The statistical analysis was performed for normally distributed parameters with the t-test and for nonnormally distributed parameters with the Mann–Whitney U test. The level of statistical significance was set at a P value of less than 0.05.
Regarding the length of professional experience (expressed in years of performing anaesthesia; mean/range), there was no significant difference between the two studied populations of anaesthetists [group C: mean 11 (range 2–21) years of anaesthesia practice, three women, three men; group M: mean 9 (range 2–19) years of anaesthesia practice, two women, four men]. The chosen 12 anaesthetists were a representative part of the local staff composed of 16 individuals. All general anaesthesias were typical for the local procedures for laparoscopic operations in gynaecology. After oral premedication with 7.5 mg midazolam, anaesthesia was induced with fentanyl, propofol and atracurium or rocuronium as follows. Propofol was specifically applied according to the habits of the two recognized groups of anaesthetists: in group C, an initial bolus dose of 1.5–2 mg kg−1 was given, and in group M, a TCI target effect site concentration of 4–5 μg ml−1 was set. To maintain anaesthesia, propofol infusion was continued in group C with 5–10 mg kg−1 h−1, whereas in group M, the effect site concentration was varied between 2.8 and 4 μg ml−1. Neuromuscular relaxation was applied under relaxometric control without yielding statistically different dosages between the groups. After the initial dose of 0.1–0.2 mg fentanyl, repetitive boluses from 0.1 mg up to a maximum of 0.5 mg were administered according to the subjectively estimated clinical needs. In a later stage of the intervention, the administration of fentanyl was ceased and analgesia was continued with remifentanil (0.05–0.5 μg kg−1 min−1). Concerning the total dose of opioids, there was no statistically significant difference between the two groups, but we observed a tendency to more fentanyl and less remifentanil use in group C and the opposite in group M.
For the biometric data of the patients, there was no statistically significant difference between the two groups. The same was observed for the baseline values for HR and BP. The operation durations in both groups were comparable, as well as the duration of anaesthesia. Awakening time was on average 5 min significantly shorter in group M (Table 1). The course of intraoperative HR and BP did not differ significantly (Fig. 1). The propofol consumption showed only small differences in the sense of a slightly lower amount in group M (Table 2).
There have been repeated attempts to demonstrate that TCI technology is superior to manually controlled infusion of propofol. The supporters who fortify this assumption occasionally found less haemodynamic variability, shorter awakening time and lower propofol consumption.1,2,8 However, even though some of these differences attained statistical significance, because of their modest dimension, the clinical relevance of this has to be discussed. The reported positive effects on awakening time were perceivable in the early postoperative period, showing (statistically not significant) less pain medication and nausea. However, further lasting outcome-related advantages were not perceivable. Also, other authors could not find relevant differences in the parameters mentioned above.9–12
In our patients, anaesthetic dosages were chosen by the investigated anaesthesiologists according to their views about the actual clinical needs. There was no anaesthetic depth measurement (e.g. bispectral index) used. In both groups, the anaesthetists relied on the conventional means of assessment of anaesthesia, predominantly the haemodynamic profile in the context of the actual surgical stimulation. However, the very similar haemodynamic course in the two groups suggests that steering the anaesthesia as well as the confounding factors, such as volaemia, was comparable in the two groups.
The data shown here reflect on the usual clinical procedures at the site of the investigation and are not biased by the observation. The authors are aware that a retrospective study cannot produce as stringent data as a randomized prospective investigation. On the contrary, this shortcoming may be compensated by a thorough standardization of inclusion criteria and the avoidance of influencing the investigated anaesthetists by the observation itself. We consider the fact that the retrospective evaluation of anaesthesia protocols was an essential precondition to gain unbiased data, which otherwise can be achieved only by strict double blinding. In particular, it seems impossible to obtain valid information about behavioural issues if the individuals (here the involved anaesthetists) know that they are under observation.
The awakening time after ‘modern’ application mode was significantly shorter by 5 min. This agrees very well with the observations of Hoymork and Raeder.13 At the first glance, 5 min is a short period of time, but considering up to successive three operations per day per table, and having this in four operating rooms, there will be eight transitions of 5 min each per day, which sum up to a total of 40 min, sufficient for an additional surgical procedure. This would substantially contribute to increase efficiency of an operating unit.6 Due to partial unavailability of retrospective data on the postoperative course, we cannot show here the lengths of stay in the postoperative care unit (PACU). At least we can refer to Tagliente,7 who stated that a shorter awakening time is usually associated with a shorter stay in a postoperative care unit, thus contributing further to increased perioperative efficiency.
The observed small differences in drug consumption proved to be statistically not significant. In group M, the trend was for less propofol (−4.6%), more fentanyl (+15.6%) and more remifentanil (+24.1%). Comparing drug costs, we found that the average total amount of consumed propofol was 1057 mg in group C and 1109 mg in group M. This means that in both cases, two ampoules of 500 mg along with another one of 200 mg have been opened and the remaining surplus wasted. Thus, the same propofol amount was prepared for and caused the same drug expenditures in both groups. In case of fentanyl, the difference was 512 μg in group C vs. 495 μg in group M. This means that one ampoule of 100 μg might be opened in group C, which costs only 1.38€ in our department. More remifentanil per case was used in group M (547 μg) as compared with group C (464 μg), but one 1-mg ampoule would have been opened in both cases, thus not causing any difference in costs.
In the available literature, it is not clearly specified which influence the opioid comedication had on propofol consumption. Our results are in the range of 6.5 mg kg−1 h−1, which is about 15–20% lower than in other studies, in which opioid selection and quantity were not adequately enough reported to be interpreted in a useful way.2 The comparison between the two groups yielded no significant difference in propofol consumption. The observed differences in opioid dosage patterns did not affect propofol consumption. The same is true for the duration of anaesthesia. According to the study by Russell et al.,4 shorter durations are characterized by higher propofol consumption when adopting the TCI mode, whereas longer durations consume more drug when this is applied by manual infusion. The anaesthesia durations in our patients (average 155 min in group C and 168 min in group M) seem to balance these antidromic effects, thus neutralizing the difference in the total propofol consumption.
The subsequent postoperative courses up to 4 h showed a tendency to lower incidence of minor postanaesthetic problems such as pain or PONV episodes in group M, but these differences attained no statistical significance.
At present, there are two distinct habits to apply propofol with opioid comedication based on behavioural patterns, to which their respective individuals remain consistently attached. The pharmacological and pharmaco-economical outcome of these two different approaches for TIVA for gynaecological procedures of intermediate duration is similar. The only statistically significant short-term outcome parameter that could be detected in this respect is a few minutes' shorter awakening time when propofol was used according to the locally adopted TCI mode. This difference may have a potential economical benefit, if it can be translated to shorten the transition time between multiple operations in a busy surgical unit.
None of the authors has any conflict of interest.
1 Passot S, Servin F, Allary R, et al
. Target-controlled versus manually-controlled infusion of propofol for direct laryngoscopy and bronchoscopy. Anesth Analg 2002; 94:1212–1216.
2 Breslin DS, Mirakhur RK, Reid JE, Kyle A. Manual versus target-controlled infusions of propofol. Anaesthesia 2004; 59:1059–1063.
3 Lehmann A, Boldt J, Thaler E, et al
. Bispectral index in patients with target-controlled or manually-controlled infusion of propofol. Anesth Analg 2002; 95:639–644.
4 Russell D, Wilkes MP, Hunter SC, et al
. Manual compared with target-controlled infusion of propofol. Br J Anaesth 1995; 75:562–566.
5 Gale T, Leslie K, Kluger M. Propofol anaesthesia via target controlled infusion
or manually controlled infusion: effects on bispectral index as a measurement of anaesthetic depth. Anaesth Intens Care 2001; 29:579–584.
6 Suttner S, Boldt J, Schmidt C, et al
. Cost analysis of target-controlled infusion-based anesthesia compared with standard anesthesia regimens. Anesth Analg 1999; 88:77–82.
7 Tagliente TM. Pharmacoeconomics of propofol in anesthesia. Am J Health Syst Pharm 1997; 54:1953–1962.
8 Triem JG, Röhm KD, Boldt J, Piper SN. Comparison of a propofol-based anesthesia regimen using optimated-target-controlled-infusion (OTCI) and manually-controlled infusion (MCI) technique. Anasthesiol Intensivmed Notfallmed Schmerzther 2006; 41:150–155.
9 Hunt-Smith J, Donaghy A, Leslie K, et al
. Safety and efficacy of target controlled infusion
(Diprifusor) vs manually controlled infusion of propofol for anaesthesia. Anaesth Intens Care 1999; 27:260–264.
10 Rehberg B, Ryll C, Hadzidiakos D, Baars J. Use of a target-controlled infusion system for propofol does not improve subjective assessment of anaesthetic depth by inexperienced anaesthesiologists. Eur J Anaesthesiol 2007; 24:920–926.
11 Struys M, Versichelen L, Thas O, et al
. Comparison of computer-controlled administration of propofol with two manually controlled infusion techniques. Anaesthesia 1997; 52:41–50.
12 Weninger B, Czener S, Steude U, Weninger E. Comparison between TCI-TIVA, manual TIVA and balanced anaesthesia for stereotactic biopsy of the brain. Anasthesiol Intensivmed Notfallmed Schmerzther 2004; 39:212–219.
13 Hoymork SC, Raeder J. Why do women wake up faster than men from propofol anaesthesia? Br J Anaesth 2005; 95:627–633.