Department of Anaesthesiology, Hannover Medical School, Carl-Neuberg-Strasse 1, D-30625 Hannover, Germany*
Accepted for publication: March 8, 2000
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Br J Anaesth 2000; 85: 42430
Keywords: anaesthetics i.v., propofol; blood, leucocytes
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
There is evidence that intravenous anaesthetics can influence the immune system.2 In a previous study, we showed that propofol and its lipid carrier, Intralipid, induce concentration-related inhibition of superoxide anion production during the respiratory burst (RB) of PMNs in vitro.3 The RB enzyme in the plasma membrane of neutrophils catalyses the oxidation of NADPH, leading to the production of superoxide anions which, in turn, is responsible for killing the phagocytosed micro-organisms.4 Other authors found similar inhibitory in vitro effects of propofol on different granulocyte functions.57 Only a few studies have investigated potential immune-compromising effects of intravenous anaesthetics under clinical conditions.8 9 One reason for this may be the difficulties in differentiating between effects on the immune system associated with the anaesthetic itself or the operative trauma, and other side effects occurring during the operation, such as volume shift, loss of body temperature or infusion of blood components. However, the effects of continuous propofol anaesthesia on blood PMNs are still unknown.
Our aim was to evaluate the time-dependent changes in PMN functions during elective interventional neuroradiological procedures with propofol or isoflurane anaesthesia. In this controlled, randomized, blinded study, the induced RB and the phagocytotic capacity of blood PMNs were investigated after 2 and 4 h of general anaesthesia.
Anaesthesia was performed with continuous propofol infusion (1% Disoprivan; Zeneca, Plankstadt, Germany) or isoflurane inhalation (Forene; Abbott, Wiesbaden, Germany). All patients underwent elective interventional neuroradiological procedures for embolization of cerebral arterio-venous malformations (AVM). As the surgical trauma associated with such procedures is minimal compared with that in abdominal or orthopaedic operations, we believe that these procedures are suitable for investigating the potential immune system effects of anaesthetics in humans. Additionally, during embolization there is no significant reduction in body temperature or marked change in blood volume. Infusion of blood components, with possible immune-compromising consequences,10 is unlikely to be necessary.
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Micro-catheterization was performed through an introducer in the femoral artery. After mapping the AVM with digital subtraction angiography, maize protein (Ethibloc occlusion emulsion; Ethicon, Hamburg, Germany) was injected into the feeding vessels.
Sample preparations
Samples (7.5 ml) of venous blood were collected in disposable blood sampling tubes coated with lithium heparin 15 IU ml1 (S-Monovette LH; Sarstedt, Nümbrecht, Germany) before and 2 and 4 h after induction of anaesthesia. After fixing with 3% acetic acid, leucocytes from all samples were counted in a Neubauer cell counting chamber.
Respiratory burst
The protocol of Rothe and colleagues11 12 was used. NADPH oxidase activity was assayed by measuring the intracellular oxidation of dihydrorhodamine 123 (DHR; MoBiTec, Göttingen, Germany) to the green fluorescent dye rhodamine 123 in a flow cytometer. The assay depends upon the incorporation of DHR into the cell. After cell activation, NADPH oxidase catalyses the reduction of oxygen to superoxide anions which are further transformed by dismutation to hydrogen peroxide. The non-fluorescent DHR is oxidized intracellularly in a peroxidase-dependent reaction to rhodamine 123. The amount of rhodamine 123 is proportional to the hydrogen peroxide generated. To isolate leucocytes, 3 ml of heparinized whole blood was layered on to 3 ml of Ficoll Hypaque (density 1.077 g dl1; Biochrom, Berlin, Germany). Nucleated blood cells in the supernatant were harvested without centrifugation. To 1 ml of phosphate-buffered saline (PBS) pH 7.2, without Ca2+ and MgCl2 (Gibco-BRL, Eggenstein, Germany), heated to 37°C, 30 µl of the leucocyte supernatant (containing an average of 5x105 cells ml1) was added; the mixture was incubated with 15 µl of 1 mM DHR. RB was induced by 20 µl of Escherichia coli HB-101 (1x109 ml1; Sigma) or by priming with 10 ng (10 µl of a 1 µg ml1 solution) of recombinant tumour necrosis factor alpha (TNF-; Sigma) for 5 min followed by receptor stimulation with 10 µl of 0.01 mM N-formyl-methionyl-leucylphenylalanine (FMLP; Sigma). After incubation (20 min, 37°C), the reaction was terminated by transferring the samples on to ice. To determine viability, dead cells were counterstained with 1 mM propidium iodide (Serva, Heidelberg, Germany). Samples were stored on ice and subjected to flow cytometry within 30 min (see below). Negative controls without stimulation were included to detect possible preactivation of neutrophils.
Phagocytosis assay
A commercial test kit (Phagotest; Orpegen, Heidelberg, Germany) was used to determine the phagocytotic activity of PMNs. Briefly, 50 µl of heparinized whole blood was incubated (10 min, 37°C) with 5 µl of opsonized and fluorescein isothiocyanate (FITC)-labelled E. coli (1x109 ml1). Negative controls were kept on ice. After addition of 50 µl of quenching solution (Orpegen) and 3 ml of PBS, the samples were centrifuged (5 min, 250g, 4°C). Fifty microlitres of lysis solution (Optilyse; Immunotech, Krefeld, Germany) was added and the mixture incubated for 10 min. Two millilitres of distilled water was added; the samples were then incubated for another 10 min and centrifuged. After washing of the remaining cell pellet and addition of propidium iodide (50 µl, 1 mM), the samples were analysed by flow cytometry.
Flow cytometry adjustment and acquisition
Both of the flow cytometric assays used allow quantification of PMN responses at the single-cell level. The flow cyto meter (Epics XL; Beckman-Coulter, Krefeld, Germany) was equipped with an argon ion laser adjusted to a wavelength of 488 nm. Twenty thousand events were measured for each sample. The rhodamine emission of the RB assay and the FITC emission of the labelled E. coli in the phagocytosis test were measured with the photomultiplier in the green fluorescence channel (FL1; 515545 nm). A photomultiplier for red fluorescence (FL3; 650 nm) was used to measure propidium iodide emission in order to discriminate between living and necrotic cells, or between bacteria and leucocytes. Sideways scatter (SSC) and forward scatter (FSC) were assessed in linear mode, and FL1 and FL3 in logarithmic mode without compensation. The photomultiplier voltage and gains of FSC, SSC, FL1 and FL3 were adjusted for each negative control and remained constant for the matched samples. Data files were stored in list mode and analysed in dot plots using a PC software package (EXPO 2.0; Beckman-Coulter). An example of data processing is shown in Figure 1.
|
Statistical analysis
The percentage of activated PMNs (in the RB assay, the percentage of rhodamine-positive PMNs after induction with E. coli or TNF-/FMLP; in the phagocytosis assay, the percentage of PMNs containing FITC-labelled E. coli) after 2 and 4 h of anaesthesia with propofol or isoflurane was compared with the percentage before induction of anaesthesia (the latter was set at 100%). This net effect was calculated using the formula:
activated PMN % =
Results are presented as mean (SD). The phagocytotic activity of PMNs is also expressed as the percentage of mean green channel fluorescence per cell. Time-dependent intragroup data were evaluated using a two-tailed, paired t-test. At each time, differences between propofol and isoflurane were evaluated using one-factor (group) analysis of variance (ANOVA) for repeated measurement, followed by StudentNewmannKeuls multiple comparison test. P<0.05 was considered significant.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
There were no significant differences in the intra-operative parameters of pulmonary and circulatory function (peripheral oxygen saturation, end-expiratory carbon dioxide concentration, heart rate or arterial and central venous pressure; data not shown) or in decrease in rectal temperature and urine output (Table 1).
The number of leucocytes increased in both groups during the investigation period: at 0, 2 and 4 h of propofol anaesthesia there were 4.1 (2.0), 4.7 (3.3) and 5.3 (3.2) x 106 ml1 leucocytes, respectively; the corresponding values for isoflurane anaesthesia were 4.2 (2.8), 4.7 (2.4) and 5.0 (2.4) x106 leucocytes ml1. There was no significant difference between the groups. The percentage of propidium iodide-positive PMNs, used as a measure of cell necrosis, was low (<5%) in both groups throughout the investigation period.
After 4 h of anaesthesia, the percentage of phagocytic PMNs that were ingesting opsonized FITC-labelled E. coli was slightly lower (propofol: 93.2% (7.0%), P=0.02; isoflurane: 94.3% (9.2%); P=0.03) than before induction of anaesthesia (Figure 2). The phagocytotic activity of each single cell expressed by the mean channel green FITC fluorescence was also reduced in both groups (propofol: 78.4% (12.7%), P<0.001; isoflurane: 89.4% (28.6%), P=0.002). There was no significant difference between propofol and isoflurane anaesthesia in either measure of phagocytosis (Figure 2).
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Phagocytosis
We found a significant reduction in the number of blood PMNs with phagocytotic activity and in the phagocytotic capacity of each single PMN during propofol or isoflurane anaesthesia. It has been shown previously that phagocytosis by blood cells is reduced after anaesthesia and surgery.14 15 Reduced phagocytosis seems to be associated with general anaesthesia as demonstrated by Hole, Unsgaard and Breivik,15 who showed that a decrease in phagocytosis was caused by general anaesthesia but not by epidural anaesthesia.
Recently, Kotani and co-workers reported that the phagocytic capacity of alveolar macrophages decreased during anaesthesia with isoflurane or propofol.16 These findings are complementary to our findings of reduced phagocytosis by blood PMNs after 2 and 4 h of propofol or isoflurane anaesthesia. In both types of anaesthesia, RB activation via viable, non-opsonized E. coli and phagocytosis of opsonized FITC-labelled E. coli were significantly depressed. We can, thus, add to the findings of Kotani et al. by showing that, under general anaesthesia, opsonin-dependent and opsonin-independent phagocytosis is reduced not only in alveolar macrophages but also in blood PMNs. In neither study was this effect dependent on the anaesthetic used; rather, it seems to depend on the conditions of general anaesthesia itself.
Our most notable finding was that propofol anaesthesia did not cause a greater reduction in the phagocytic activity of blood PMNs than isoflurane anaesthesia, as might have been expected from published in vitro results. In two in vitro studies carried out with the same flow cytometric phagocytosis assay applied in the present study, marked inhibition of neutrophil phagocytosis has been demonstrated after incubation with clinically relevant concentrations of propofol.5 17 In both studies the inhibition by propofol was concentration-related. Should this marked in vitro inhibition of phagocytosis by propofol also take place under clinical settings, e.g. after sedation of ICU patients with propofol for several days, this would probably result in reduced microbial killing and an increased risk of infection induced by propofol. In fact, we were unable to show a significant difference in inhibition of PMN phagocytotic capacity after 4 h of intravenous propofol anaesthesia which is, of course, a short time compared with sedation on ICU.
Respiratory burst
In the present study, the RB was measured after challenge with two different stimuli in order to discriminate between phagocytosis- and receptor-dependent activation of RB. The former was carried out assayed by stimulation with phagocytosis of E. coli and the latter by receptor activation with the bacterial peptide FMLP after priming with TNF-. Both pathways increase the intracellular Ca2+ content which activates protein kinase C (PKC) and, hence, NADPH oxidase.4 Inhibition of the RB was estimated from the reduction in the number of PMNs recruited for the RB (percentage of activated PMNs).
The bacterial peptide FMLP is an agonist which mediates the response to bacterial formyl products in the inflammatory environment via a specific formyl peptide receptor on the cell surface. The receptor mediates multiple responses, including chemotaxis, exocytosis of certain enzymes, generation of superoxide anions by the NADPH oxidase complex during the RB and phagocytosis of E. coli.18 19 Like other classical chemoattractant receptors (e.g. complement factor C5a, platelet activating factor (PAF) or lipopolysaccharide (LPS)), the FMLP receptor is linked to a heterotrimeric G protein, leading via intermediate steps to activation of PKC which, in turn, activates NADPH oxidase.20 Priming with cytokines such as TNF- before FMLP activation leads to a high proportion of PMNs being activated. Priming is defined as the exposure of cells to a triggering agent, so that there is a markedly increased response to a second stimulus.21 It has been suggested that priming with TNF-
increases neutrophil FMLP receptor expression.22 23
We found that anaesthesia with propofol, but not isoflurane, decreased RB activity after isolated blood PMNs had been primed with TNF-, followed by FMLP stimulation. This finding is in accordance with the results of Nakagawara and colleagues, who demonstrated that volatile anaesthetics do not seem to interfere with the FMLP receptor, because the affinity of the FMLP receptor for radiolabelled FMLP was not altered in the presence of volatile anaesthetics.24 An in vitro study has demonstrated that isoflurane under clinically relevant concentrations had no effect on FMLP-stimulated RB.25 This study, in particular, was carried out with the same flow cytometry method used in our investigation. The results of the present study and of both cited studies emphasize that isoflurane anaesthesia seems to have only minor effects on the RB of human blood PMNs. In contrast to isoflurane, anaesthesia with propofol seems to interfere with the FMLP receptor or downstream regulation of PMNs. Further studies are necessary to confirm this hypothesis.
In vitro effects of propofol
It is probable that the reported in vitro effects of propofol are caused mainly by its lipid carrier, Intralipid, a long-chain triglyceride (LCT) solution which produces concentration-dependent inhibition of different leucocyte functions in vitro.26 In this previous study we showed that LCT lipids inhibit the RB, whereas a mixture of LCT and saturated medium-chain triglycerides (MCT) augments the RB. In a recent in vitro study, we found that a new propofol preparation dissolved in LCT/MCT lipid (Propofol Lipuro; B. Braun, Melsungen, Germany) gave similar results to standard LCT propofol (Disoprivan; Zeneca).17 It is known that polyunsaturated fatty acids like LCT can be rapidly incorporated into cell membranes,27 thereby increasing membrane rigidity.28 Thus, propofol could inhibit the RB because of a direct effect of its lipid carrier, LCT, on NADPH oxidase in the plasma membrane of PMNs. NADPH oxidase has at least six components. In the resting cell, three of these components are in the cytosol and three in the plasma membrane. After activation by PKC, all components combine in the plasma membrane to produce a fully functional enzyme complex.4 The increase in membrane rigidity following incubation with LCT in vitro could explain the delayed formation of the NADPH oxidase enzyme complex as well as the inhibition of phagocytosis. The close contact between few non-circulating PMNs and fat over an incubation period of at least 20 min. Results in rapid incorporation of LCT fat into the cell membrane could explain the in vitro suppression of the RB and phagocytosis after incubation with LCTpropofol and LCT alone.
Other potential influences
When investigating the effect of anaesthesia on immune function, it is very important to estimate the influence of independent contributions such as operative trauma, mechanical ventilation, co-medication, temperature loss, volume shift or transfusion of blood components. All of our patients were subjected to the same time-dependent alterations of immune function induced by general anaesthesia, mechanical ventilation and embolization of cerebral AVM. In our clinical setting, we attempted to keep these interferences to a minimum. Each embolization was elective, and none of our patients was on steroid or non-steroidal anti-inflammatory medication known to interfere with PMN functions and viability.29 We believe that it is safe to assume that the trauma-induced effects, on immune cells, of interventional embolization of cerebral AVMs is only minor compared with the effects of surgical procedures. Furthermore, the patients in our study did not have significant reduction in body temperature, which has been shown to reduce the release of TNF- and interleukin-1ß.30 Radiographic contrast medium reduces the percentage of granulocytes involved in phagocytosing E. coli but not the number of pathogenic bacteria.31 Furthermore, the impairment was most pronounced immediately after the injection of radiographic contrast medium and then rapidly returned to baseline.32 In the present investigation, a possible influence of contrast medium cannot be excluded, but there was no significant difference between the two groups in the amount of contrast medium administered, so any effect of the medium should have been equal in the two groups.
In conclusion, the phagocytotic activity of blood PMNs decreased significantly during the investigation period for both of the anaesthetics studied. Production of superoxide anions after FMLP activation was markedly lower during propofol anaesthesia than during isoflurane anaesthesia, suggesting that 4 h propofol anaesthesia may reduce FMLP receptor expression on the PMN surface. Additional investigations are necessary to test the hypothesis that propofol interferes with the FMLP receptor or the downstream regulation of human PMNs.
![]() |
Footnotes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Stevenson GW, Hall SC, Rudnick S. The effect of anesthetic agents on the human immune response. Anesthesiology 1990; 72: 54252[ISI][Medline]
3 Heine J, Leuwer M, Scheinichen D, Arseniev L, Jaeger K, Piepenbrock S. Flow cytometry evaluation of the in vitro influence of four i.v. anaesthetics on respiratory burst of neutrophils. Br J Anaesth 1996; 77: 38792
4 Babior BM. The respiratory burst oxidase. Adv Enzymol 1992; 65: 4995[ISI][Medline]
5 Heller A, Heller S, Blecken S, Urbaschek R, Koch T. Effects of intravenous anesthetics on bacterial elimination in human blood in vitro. Acta Anaesthesiol Scand 1998; 42: 51826[ISI][Medline]
6 Mikawa K, Akamatsu H, Nishina K et al. Propofol inhibits human neutrophil functions. Anesthes Analg 1998; 87: 695700.[Abstract]
7 Weiss M, Birkhahn A, Krone M, Schneider EM. Do etomidate and propofol influence oxygen radical production of neutrophils? Immunopharmacol Immunotoxicol 1996; 18: 291307[ISI][Medline]
8 Pirttikangas CO, Perttilä J, Salo M, Vainio O, Liukko-Sipi S. Propofol infusion anaesthesia and immune response in minor surgery. Anaesthesia 1994; 49: 1316[ISI][Medline]
9 Pirttikangas CO, Salo M, Mansikka M, Grönroos J, Pulkki K, Peltola O. The influence of anaesthetic technique upon the immune response to hysterectomy. Anaesthesia 1995; 50: 105661[ISI][Medline]
10 Sillimann CC, Voelkel NF, Allard JD et al. Plasma and lipids from stored packed red blood cells cause acute lung injury in an animal model. J Clin Invest 1998; 101: 145867
11 Rothe G, Oser A, Valet G. Dihydrorhodamine 123: a new flow cytometric indicator for respiratory burst activity in neutrophil granulocytes. Naturwissenschaften 1988; 75: 3545[ISI][Medline]
12 Rothe G, Valet G. Flow cytometric assays of oxidative burst activity in phagocytes. Methods Enzymol 1994; 233: 53948[ISI][Medline]
13 Malech HL, Gallin JI. Neutrophils in human diseases: New Engl J Med 1987; 10: 68794
14 Bardosi L, Tekeres M. Impaired metabolic activity of phagocytic cells after anaesthesia and surgery. Br J Anaesth 1985; 57: 52023[Abstract]
15 Hole A, Unsgaard G, Breivik H. Monocyte functions are depressed during and after surgery under general anesthesia but not under epidural anesthesia. Acta Anaesthesiol Scand 1982; 26: 3017[ISI][Medline]
16 Kotani N, Hashimoto H, Sessler DI et al. Intraoperative modulation of alveolar macrophage function during isoflurane and propofol anesthesia. Anesthesiology 1998; 89: 112532[ISI][Medline]
17 Heine J, Weingaertner N, Jaeger K, Nentwig B, Leuwer M, Piepenbrock S. Anaesthetics dissolved in lipids containing long chain (LCT) and medium chain triglycerides (MCT) stimulate respiratory burst of PMN in vitro. Immunobiology 1999; 200: 5434
18 Fujishima S, Aikawa N. Neutrophil mediated tissue injury and its modulation. Intensive Care Med 1995; 21: 27785[ISI][Medline]
19 Marasco WA, Phan SH, Krutzsch H. Purification and identification of formyl-methionyl-leucyl-phenylalanine as the major peptide neutrophil chemotactic factor produced by Escherichia coli. J Biol Chem 1984; 259: 543039
20 Gerard C, Gerard NP. The pro-inflammatory seven-transmembrane segment receptors of the leukocyte. Curr Opin Immunol 1994; 6: 14045[ISI][Medline]
21 Downey GP, Fukushima T, Fialkow L, Waddell TK. Intracellular signaling in neutrophil priming and activation. Cell Biol 1995; 6: 3456
22 Elbim C, Cholett-Martin S, Bailly S, Hakim J, Gougerot-Pocidalo MA. Priming of polymorphonuclear neutrophils by tumour necrosis factor in whole blood: identification of two polymorphonuclear neutrophil subpopulations in response to formyl-peptides. Blood 1993; 82: 63340[Abstract]
23 Menegazzi R, Cramer R, Patriarca P, Scheurich P, Dri P. Evidence that tumour necrosis factor- (TNF)-induced activation of neutrophil respiratory burst on biologic surfaces is mediated by the p55 TNF receptor. Blood 1994; 84: 28793
24 Nakagawara M, Takeshige K, Takamatsu J, Takahashi S, Yoshitake J, Minakami S. Inhibition of superoxide production and Ca2+ mobilisation in human neutrophils by halothane, enflurane, and isoflurane. Anesthesiology 1986; 64: 412[ISI][Medline]
25 Fröhlich D, Rothe G, Schwall B et al. Effects of volatile anaesthetics on human neutrophil oxidative response to bacterial peptide FMLP. Br J Anaesth 1997; 78: 71823
26 Heine J, Scheinichen D, Jaeger K, André M, Leuwer M. In vitro influence of parenteral lipid emulsions on the respiratory burst of neutrophils. Nutrition 1999; 15: 54045[ISI][Medline]
27 Curtis AS, Shaw FM, Spires VM. Cell surface lipids and adhesion. II. The turnover of lipid components of the plasmalemma in relation to cell adhesion. J Cell Sci 1975; 18: 35773[Abstract]
28 Mandel G, Clark W. Functional properties of EL-4 tumor cells with lipid-altered membranes. J Immunol 1978; 120: 163743[Abstract]
29 Hill J, Alonso A, Thiele GM, Robbins RA. Glucocorticoids blunt neutrophil CD 11b surface glycoprotein upregulation during cardiopulmonary bypass in humans. Anesth Analg 1994; 79: 237[Abstract]
30 Menasché P, Haydar S, Peynet J et al. A potential mechanism of vasodilatation after warm heart surgery: the temperature-dependent release of cytokines. J Thorac Cardiovasc Surg 1994; 107: 2939[ISI][Medline]
31 Farrow R, Roobottom CA, Wells IP, Hurlock N. Effects of radiographic contrast media on leukocyte phagocytosis. Acad Radiol 1994; 1: 24952[Medline]
32 Rasmussen F. The influence of radiographic contrast media on some granulocyte functions. Acta Radiol (suppl) 1998; 419: 735