Comparison of coagulation and blood loss during anaesthesia with inhaled isoflurane or intravenous propofol

N. L. Law, K. F. J. Ng, M. G. Irwin and J. S. F. Man

Department of Anaesthesiology, F Block, Queen Mary Hospital, Pokfulam Road, Hong Kong*Corresponding author

Accepted for publication: August 20, 2000


    Abstract
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Propofol has been reported to affect blood coagulation. This prospective, randomized study compared coagulation and blood loss during anaesthetic maintenance with target-controlled intravenous propofol infusion vs. inhaled isoflurane. Thirty-eight ASA I–III patients undergoing head and neck surgery were allocated randomly to receive either inhaled isoflurane at end-tidal concentration 1–1.5% (group I, n=20) or target-controlled infusion (TCI) of propofol at target concentration 2–5 µg ml–1 (group P, n=18). Thrombelastography® on recalcified whole blood was performed pre-induction, and at 15, 30, 60, 90, 120 min post-induction and 30 min after anaesthesia in both groups. Blood loss was estimated from weighing swabs and the volume in suction bottles. Induced hypotension was not used, and perioperative body temperature was similar between groups. There were no significant differences in thrombelastographic coagulation (R-time, K-time, maximum amplitude and angle) or fibrinolytic variables (lysis index at 30 and 60 min) at all times between groups. Total blood loss was also not significantly different (median group I: 350 ml, range 20–1200 ml; group P: 200 ml, range 50–800 ml). Shortening of R-time and widening of angle developed over time in both groups (P<0.05 groups I and P, repeated measures ANOVA). We conclude that maintenance of anaesthesia with propofol TCI at 2–5 µg ml–1 does not cause detectable coagulation changes on thrombelastography® nor increase surgical blood loss when compared to inhaled isoflurane.

Br J Anaesth 2001; 86: 94–8

Keywords: anaesthetics i.v., propofol; anaesthetics inhalational, isoflurane; coagulation; thrombelastography®; blood loss


    Introduction
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
The use of intravenous (i.v.) propofol infusion for maintenance of anaesthesia is gaining in popularity, particularly with the introduction of commercial target-controlled infusion (TCI) systems. A number of recent studies have raised the concern that propofol may have adverse effects on blood coagulation and fibrinolysis14 and, particularly, impair platelet aggregation.13 The lipid emulsion solvent of propofol may also affect platelet function.5 6 Such data would suggest that the use of propofol infusion for anaesthetic maintenance might be undesirable in operations where bleeding may become a problem. However, previous studies on propofol and coagulation were performed in vitro1 3 4 except one in vivo study, 2 where no coagulation measurement was made during propofol infusion. It has also been shown that some inhaled agents may have inhibitory effects on coagulation.1 7 8 In this study, we decided to investigate the effects of propofol on coagulation under three conditions that we feel are more clinically relevant: (1) in vivo during surgery with ongoing infusion of propofol; (2) with propofol administered in its usual form, that is dissolved in its lipid emulsion solvent; and (3) not looking at the coagulation effects of propofol in isolation, but comparing it with an alternative for anaesthetic maintenance, inhaled isoflurane.


    Methods
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
The investigation was approved by the Institutional Research Ethics Committee. With informed written consent, ASA I–III patients scheduled for elective head and neck operations were recruited. These patients were undergoing resection of head and neck tumours followed by reconstructive plastic surgery. Those with a history of abnormal bleeding, or with abnormal preoperative prothrombin time, activated partial thromboplastin time or platelet count or who were taking drugs that may affect coagulation were excluded. Patients who required fibreoptic intubation prior to induction of anaesthesia were also excluded.

Eligible patients were allocated randomly to one of two groups: group I, maintenance of anaesthesia with inhaled isoflurane at an end-tidal concentration of 1–1.5%; group P, maintenance of anaesthesia with TCI propofol (1% pre-filled syringe AstraZeneca, UK) using a Diprifusor® pump (Graesby, UK) with a target plasma concentration of propofol between 2 and 5 µg ml–1. Patients were unpremedicated and anaesthesia was induced in both groups using i.v. fentanyl 2–2.5 µg kg–1 and propofol 2–2.5 mg kg–1. Rocuronium was administered for neuromuscular block and continued according to response to train-of-four peripheral nerve stimulation. Patients’ tracheas were intubated and lungs mechanically ventilated throughout surgery to an end-tidal carbon dioxide (PE'CO2) between 4–5 kPa. Morphine was used for analgesia and both groups also received 30% oxygen in nitrous oxide. At the end of surgery, all patients had their neuromuscular block antagonized and resumed spontaneous breathing via either a tracheostomy or their natural airway. Continuous pulse oximetry, electrocardiogram, PE'CO2, end-tidal anaesthetic agent, radial arterial pressure and core temperature were monitored throughout the operation. All i.v. fluids were warmed, and a warm air blanket (Bair Hugger 505®, Augustine Medical Inc., MN, USA) was applied to every patient unless it caused interference to the surgical field. There was no induced hypotension and patients’ systolic arterial pressures were maintained at 100 mm Hg or 70% of the preoperative value, whichever was higher. Hypotension was treated with i.v. crystalloid fluid loading or intravenous boluses of ephedrine as appropriate.

The variables to be observed or measured in this study were: (1) serial thrombelastography® (TEG®); (2) intraoperative blood loss; (3) serial body temperature; and (4) patient characteristics and surgical details. Whole arterial blood samples were taken for TEG® measurement (CTEG® model 3000, Haemoscope Corporation, Skokie, IL, USA) before induction and at 15, 30, 60, 90 and 120 min after induction. The last sample was taken in the recovery room 30 min after the patient first opened their eyes to verbal command. Blood samples were collected with the double syringe technique9 from the radial arterial line. The first 6 ml of each sample was discarded. Samples of whole blood (3.5 ml) were then collected in a bottle containing 3.2% sodium citrate (blood to citrate 1:9 by volume; Vacuette®, Greiner GmbH, Germany) and stored at room temperature until measurement. Because of the limited number of TEG® channels available, there were slight variations in the exact measurement time for individual samples. Recalcification was performed immediately before measurement by adding 15 µl of 0.2 M calcium chloride solution to 335 µl of whole blood from the citrate bottle.9 Mixing was accomplished by lowering the piston of the TEG® into the plastic cup 10 times. No celite was added for the TEG® measurement and all measurements were made within 3 h of sample collection. All conventional TEG® variables including R-time, K-time, maximum amplitude (MA), angle ({alpha}), and lysis index at 30 (LY30) and 60 (LY60) min were recorded.

Total blood loss was recorded at the end of surgery by measuring the volume collected in the suction bottles and weighing swabs collected from the operative field.

Statistics and power analysis
Statistical analysis was performed using the software programme Statistica release 4.5 (StatSoft®, Tulsa, OK, USA). Intergroup comparisons of parametric data such as the TEG® parameters, volume of fluids administered and body temperature were performed using two-tailed unpaired t-test at each time point of observation. Blood loss was compared using the Mann–Whitney U test. The change in various TEG® variables over time was compared to the preoperative value in each group by repeated measures ANOVA. Where a statistically significant difference was detected, further pairwise comparisons were made between individual time points and the preoperative values using two-tailed paired t-test. The significance level was set at a P value of <0.05.

Based on our previous experience, the average blood loss in these procedures was around 400 ml. A sample size of 20 patients in each group would give us an 80% power to detect a difference in blood loss of 100 ml between the two groups, assuming the standard deviation of blood loss would be around 120 ml and at the current {alpha}-error level. This sample size would also have more than 80% power to detect the magnitude of propofol related fibrinolysis reported by Kohro and colleagues.4


    Results
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
A total of 40 patients were recruited. However, only the data from 38 patients were used for analysis, as there were two patients in whom data collection was incomplete. There were 20 patients in group I and 18 patients in group P. Patient details and surgical information is displayed in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1 Patient characteristics (mean (SD) or actual numbers). No significant differences between groups
 
It has been previously reported that during maxillary sinus surgery, blood loss is significantly correlated to body weight and operation duration.10 We have, therefore, compared total blood loss, and blood loss adjusted for patients’ body weight and total operation time between the two groups. There was no statistically significant difference in blood loss between the two groups (Fig. 1), although blood loss in group I tended to be slightly higher (median total blood loss group I: 350 ml, range 20–1200 ml; group P: 200 ml, range 50–800 ml).



View larger version (26K):
[in this window]
[in a new window]
 
Fig 1 Total blood loss and blood loss per kg body weight per hour of operation. There were no significant between group differences (P=0.185 for total blood loss; P=0.274 corrected for body weight and operation duration. Mann–Whitney U test).

 
There were also no statistically significant differences in any TEG® variables, including clotting parameters (R-time, K-time, MA and {alpha}) and fibrinolytic (LY 30 and LY 60) variables between the two study groups at all time points after induction of anaesthesia. The rate of clot formation was increased in both groups, with shortening of R-time and K-time and widening of {alpha}, as time progressed (group I: P<0.05 for R-time and {alpha}; group P: P<0.05 for R-time, K-time and {alpha}, repeated measure ANOVA). These results are displayed in Table 2. There was no difference in body temperature between the two groups at all measurement points.


View this table:
[in this window]
[in a new window]
 
Table 2 Change of TEG® parameters over time (mean (SD)). *P<0.05 compared with preoperative value (ANOVA); normal ranges R: 15–23 mm; K: 4–15 mm; MA: 34–46 mm; {alpha}: 36–48° (Operator’s manual, TEG®, Haemoscope Corp.)
 

    Discussion
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
The pharmacokinetic properties and recovery characteristics of propofol have encouraged its use for i.v. maintenance of anaesthesia. There are, however, a number of studies showing that propofol has an inhibitory effect on coagulation and platelet aggregation and that it enhances fibrinolysis.14 The lipid emulsion solvent of propofol may also affect platelet function.5 6 It is possible, therefore, that the benefits of infusing propofol dissolved in lipid emulsion for maintenance of anaesthesia during delicate neurosurgical or ophthalmological procedures may be offset by the potential risk of increased bleeding. However, the studies which have demonstrated the adverse effects of propofol on coagulation were mostly in vitro studies,1 3 4 or were referring to propofol concentrations higher than that usually maintained in patients’ plasma.1 3 4

In this investigation we have attempted to determine the existence, or otherwise, of such a potential risk by using a prospective randomized trial comparing anaesthetic maintenance with i.v. propofol dissolved in lipid emulsion and that with a conventional inhaled agent. We chose isoflurane for comparison in our study not only because it is popular, but also because many other inhaled agents such as halothane and sevoflurane have been shown to have antiplatelet or other adverse coagulation effects themselves.1 7 8 11–13 Isoflurane has not been found to possess these properties. Anaesthetic induction was accomplished by i.v. propofol in both groups as the airway irritant effect of isoflurane made this agent undesirable for this purpose. With an initial fast distribution half-time of around 3 min,14 the initial bolus of propofol is unlikely to have affected the first intergroup comparison at 15 min post-induction.

One of the limitations of previous studies investigating the coagulation effects of propofol is the lack of control over the actual plasma concentration of propofol during the time of measurement. During in vitro studies, propofol or intralipid was usually only administered at a fixed rate15 16 or as a single bolus.17 The concentrations of propofol used during in vitro studies, were usually well above those administered to maintain anaesthesia clinically.3 4 TCI systems, such as the Diprifusor®, use a pharmacokinetic model to predict the initial bolus dose and infusion rates to achieve and maintain a given target blood propofol concentration. There is good evidence to show that the bias and inaccuracy of this system are reasonably low and the predicted concentration correlates well with the actual plasma level.18 Our results should be an accurate reflection of the changes that might occur during normal use of this drug.

Most studies on the effect of propofol on coagulation have identified an inhibitory effect of propofol on platelet aggregation1–3 or an enhancement of fibrinolysis4 as the underlying mechanism. Although platelet aggregometry is an excellent tool for monitoring platelet aggregation, it requires a large sample volume for measurement and is time consuming. It was, therefore, not desirable in our study where repeated measurements were necessary on the same patient. Moreover, changes in platelet function as seen on aggregometry may not necessarily correlate with changes in surgical blood loss.19 20 In contrast, the TEG® is a simple bedside tool which allows rapid measurement of global coagulation including fibrinolysis with a relatively small sample volume. TEG® has been reported to predict blood loss in a variety of procedures,2123 and has also been used successfully to identify the coagulation effects of propofol.4 We believe, therefore, that it is an appropriate coagulation monitor for this study.

There was no difference observed in TEG® variables between the two groups of patients during the first 2 h of anaesthesia and at the end of surgery in our study. We also did not find any difference in blood loss either in terms of absolute value, or adjusted to patients’ body weight and operation duration. Despite the in vitro evidence of propofol’s adverse effects on coagulation, propofol does not appear to increase blood loss during surgery. There were studies which showed less blood loss where propofol anaesthesia was compared with maintenance with inhaled agents for endoscopic sinus surgery, both retrospectively15 and in a prospective randomized study.16 Propofol maintenance was also associated with reduced blood loss for termination of pregnancy,24 25 although this may be related to its effects on uterine muscle. Propofol vs. inhaled agent maintenance gave rise to similar blood loss during Caesarean section.26 Although there was no statistically significant difference in blood loss between the two groups in our study blood loss did tend to be slightly higher in the isoflurane group. With the support of the TEG® findings and the elimination of possible confounders, such as intergroup differences in arterial pressure or body temperature in our study, we conclude that i.v. maintenance of anaesthesia with propofol does not adversely affect coagulation to a clinically significantly extent, as compared to use of isoflurane.

The results of early in vitro studies strongly suggest that propofol possesses antiplatelet properties.13 A recent study, however, indicated that the effect of propofol might differ in high and low concentrations.27 In particular, the results of this study suggested propofol might enhance secondary aggregation by ADP and epinephrine at concentrations around 5 µg ml–1. This could partly explain the different observations in our study compared with earlier in vitro studies. Another issue which is relevant to our study, and other in vivo studies, is the natural development of a hypercoagulable state associated with surgical stress and tissue trauma.28 29 This was suggested in our study by shortening of R-time, K-time and widening of {alpha} in most samples collected intra- and postoperatively in both groups. The antiplatelet and profibrinolytic effects of propofol are probably mild at the concentrations used clinically, and will not be detectable with the development of such natural hypercoagulability. However, it is important to note that the changes in TEG® variables over time in both groups in our study may not be totally attributable to ‘real’ hypercoagulability. Shortening of R- and K-time, and widening of {alpha} have been demonstrated to occur in recalcified citrated blood over time30 and would contribute to at least part of the changes observed. However, this should not have affected the validity of intergroup comparisons. In any particular patient, the delay in measurement was related to the time required to complete measurement of a previous sample. As a result, although there should be more delay in the recovery room sample as compared with the pre-induction sample for example, the delay at any particular sampling point should be similar.

We conclude that maintenance of anaesthesia with i.v. infusion of propofol dissolved in lipid emulsion at target concentrations of 2–5 µg ml–1 is not associated with increased risk of bleeding compared with maintenance with inhaled isoflurane. We believe that the use of this technique is safe even during procedures where an increased risk of bleeding should be strictly avoided, provided patients do not have other conditions that may interfere with coagulation. Further studies are required to investigate the long-term effect of propofol infusion, such as for sedation in ICU, the safety of using propofol in the presence of other drugs that are known to have an inhibitory effect on coagulation such as non-steroidal anti-inflammatory agents and heparin, and the effect of propofol in patients with disorders of haemostasis.


    References
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
1 Dogan IV, Ovali E, Eti Z, Yayci A, Gogus FY. The in vitro effects of isoflurane, sevoflurane, and propofol on platelet aggregation. Anesth Analg 1999; 88: 432–6[Abstract/Free Full Text]

2 Aoki H, Mizobe T, Nozuchi S, Hiramatsu N. In vivo and in vitro studies of the inhibitory effect of propofol on human platelet aggregation. Anesthesiology 1998; 88: 362–70[ISI][Medline]

3 deLaCruz JP, Carmona JA, Paez MV, Blanco E, Sanchez-deLaCuesta F. Propofol inhibits in vitro platelet aggregation in human whole blood. Anesth Analg 1997; 84: 919–21[Abstract]

4 Kohro S, Yamakage M, Omote T, Namiki A. In vitro effects of propofol on blood coagulability and fibrinolysis by the use of thromboelastograph technique. Acta Anaesthesiol Scand 1999; 43: 217–9[ISI][Medline]

5 Aviram M, Deckelbaum RJ. Intralipid infusion into humans reduces in vitro platelet aggregation and alters platelet lipid composition. Metabolism 1989; 38: 343–7[ISI][Medline]

6 Jarnvig IL, Naesh O, Hindberg I, Behnke O, Bregengaad C, Wilhjelm B. Platelet responses to intravenous infusion of Intraplipid in healthy volunteers. Am J Clin Nutr 1990; 52: 628–31[Abstract]

7 Hirakata H, Ushikubi F, Toda H, et al. Sevoflurane inhibits human platelet aggregation and thromboxane A2 formation, possibly by suppression of cyclooxygenase activity. Anesthesiology 1996; 85: 1447–53[ISI][Medline]

8 Kohro S, Yamakage M. Direct inhibitory mechanisms of halothane on human platelet aggregation. Anesthesiology 1996; 85: 96–106[ISI][Medline]

9 Khurana S, Mattson JC, Westley S, O’Neill WW, Timmis GC, Safian RD. Monitoring platelet glycoprotein IIb/IIIa-fibrin interaction with tissue factor-activated thrombelastography. J Lab Clin Med 1997; 130: 401–11[ISI][Medline]

10 Yoshikawa T, Sano K, Kan T. Clinical assessment of anesthesia and estimated blood loss during maxillary sinus surgery. Anesth Prog 1989; 36: 242–8[Medline]

11 Dalsgaard NJ, Risbo A, Simmelkjaer P, Gormsen J. Impaired platelet aggregation and increased bleeding time during general anaesthesia with halothane. Br J Anaesth 1981; 53: 1039–42[Abstract]

12 Hirakata H, Nakamura K, Sai S, et al. Platelet aggregation is impaired during anaesthesia with sevoflurane but not with isoflurane. Can J Anaesth 1997; 44: 1157–61[Abstract]

13 Hirakata H, Ushikubi F, Narumiya S, Hatano Y, Nakamura K, Mori K. The effect of inhaled anesthetics on the platelet aggregation and the ligand-binding affinity of the platelet thromboxane A2 receptor. Anesth Analg 1995; 81: 114–8[Abstract]

14 Gepts E, Camu F, Cockshott ID, Douglas EJ. Disposition of propofol administered as constant rate intravenous infusions in humans. Anesth Analg 1987; 66: 1256–63[Abstract]

15 Blackwell KE, Ross DA, Kapur P, Calcaterra TC. Propofol for maintenance of general anesthesia: a technique to limit blood loss during endoscopic sinus surgery. Am J Otolaryngol 1993; 14: 262–6[ISI][Medline]

16 Pavlin JD, Colley PS, Weymuller EA Jr, vanNorman G, Gunn HC, Koerschgen ME. Propofol versus isoflurane for endoscopic sinus surgery. Am J Otolaryngol 1999; 20: 96–101[ISI][Medline]

17 Turkan H, Suer AH, Beyan C, Yalcin A. Propofol does not affect platelet aggregation. Eur J Anaesthesiol 1996; 13: 408–9[ISI][Medline]

18 Swinhoe CF, Peacock JE, Glen JB, Reilly CS. Evaluation of the predictive performance of a ‘Diprifusor’ TCI system. Anaesthesia 1998; 53 Suppl 1: 61–7[ISI][Medline]

19 Boldt J, Zickmann B, Ballesteros BM, Stertmann F, Hempelmann G. Influence of five different priming solutions on platelet function in patients undergoing cardiac surgery. Anesth Analg 1992; 74: 219–25[Abstract]

20 Lazenby WD, Russo I, Zadeh BJ, et al. Treatment with desmopressin acetate in routine coronary artery bypass surgery to improve postoperative hemostasis. Circulation 1990; 82: 413–9[ISI]

21 Williams GD, Bratton SL, Riley EC, Ramamoorthy C. Coagulation tests during cardiopulmonary bypass correlate with blood loss in children undergoing cardiac surgery. J Cardiothorac Vasc Anesth 1999; 13: 398–404[ISI][Medline]

22 Davis CL, Chandler WL. Thromboelastography for the prediction of bleeding after transplant renal biopsy. J Am Soc Nephrol 1995; 6: 1250–5[Abstract]

23 Ereth MH, Nuttall GA, Klindworth JT, et al. Does the platelet-activated clotting test (HemoSTATUS) predict blood loss and platelet dysfunction associated with cardiopulmonary bypass? Anesth Analg 1997; 85: 259–64[Abstract]

24 Nathan N, Peyclit A, Lahrimi A, Feiss P. Comparison of sevoflurane and propofol for ambulatory anaesthesia in gynaecological surgery. Can J Anaesth 1998; 45: 1148–50[Abstract]

25 Hall JE, Ng WS, Smith S. Blood loss during first trimester termination of pregnancy: comparison of two anaesthetic techniques. Br J Anaesth 1997; 78: 172–4[Abstract/Free Full Text]

26 Abboud TK, Zhu J, Richardson M, Peres-daSilva E, Donovan M. Intravenous propofol vs thiamylal-isoflurane for caesarean section, comparative maternal and neonatal effects. Acta Anaesthesiol Scand 1995; 39: 205–9[ISI][Medline]

27 Hirakata H, Nakamura K, Yokubol B, et al. Propofol has both enhancing and suppressing effects on human platelet aggregation in vitro. Anesthesiology 1999; 91: 1361–9[ISI][Medline]

28 Caprini JA, Arcelus JI, Laubach M, et al. Postoperative hypercoagulability and deep-vein thrombosis after laparoscopic cholecystectomy. Surg Endosc 1995; 9: 304–9[ISI][Medline]

29 Ng KF, Lo JW. The development of hypercoagulability state, as measured by thrombelastography, associated with intraoperative surgical blood loss. Anaesth Intens Care 1996; 24: 20–5[ISI][Medline]

30 Camenzind V, Bombeli T, Seifert B, Jamnicki M, Popovic D, Pasch T, Spahn DR. Citrate storage affects thrombelastograph® analysis. Anesthesiology 2000; 92: 1242–49[ISI][Medline]