Volatile anesthetics are known to suppress platelet aggregation. In contrast, there is conflicting information regarding the effect of propofol on platelet function. The present study was designed to clarify the effects of propofol on platelet function and the mechanisms underlying these effects.
Propofol or an equivalent volume of 10% Intralipos (as a control) was added to test tubes 5 min before the induction of each reaction. Platelet aggregation induced by epinephrine, arachidonic acid (AA), prostaglandin G2 (PGG2), or STA2 (a thromboxane A2 [TXA2] analog) was measured using an eight-channel aggregometer. To determine type 1 cyclooxygenase activity, AA (0.5 mM) was added to an assay mixture containing type 1 cyclooxygenase, and the concentration of the final product, malonaldehyde, was measured by spectrophotometry. Epinephrine-, adenosine diphosphate-, AA-, and PGG2-induced TXA2 formation was measured using a commercially available radioimmunoassay kit. To estimate TXA2 receptor-binding affinity, 3H-S145, a specific TXA2 receptor antagonist, was added, and the radioactivity of receptor-bound 3H-S145 was determined using a liquid scintillation analyzer. Inositol 1,4,5-triphosphate formation was measured in STA2-stimulated platelets using a commercially available inositol 1,4,5-triphosphate assay kit.
Propofol (40 microM) enhanced, whereas 100 microM suppressed, adenosine diphosphate- and epinephrine-induced secondary aggregation without affecting primary aggregation. Propofol (40 microM) also enhanced, but 100 microM suppressed, AA-induced aggregation. Propofol (100 microM) enhanced PGG2- and STA2-induced aggregation. Propofol (100 microM) suppressed AA-induced TXA2 formation but did not alter that induced by PGG2. Propofol (30-100 microM) suppressed AA-induced malonaldehyde formation, indicating inhibition of type 1 cyclooxygenase activity. Propofol did not alter TXA2 receptor-binding affinity. Propofol (30 and 100 microM) augmented inositol 1,4,5-triphosphate formation in STA2-stimulated platelets.
The present findings clearly indicate that high concentrations of propofol suppress the activity of type 1 cyclooxygenase, the enzyme that converts AA to PGG2. Furthermore, propofol also enhanced STA2-induced inositol 1,4,5-triphosphate formation. These results may explain the inconsistent findings of previous investigators.
ALTHOUGH platelet function directly affects hemostasis, our knowledge regarding the effects of anesthetics on platelet function is still limited. Suppression of platelet function by halothane and sevoflurane has been demonstrated by in vivo 1,2and in vitro 3–8studies. We 9,10have previously shown that the suppressive effect of halothane on platelet aggregation is caused by the suppression of both type 1 cyclooxygenase (COX1) activity and thromboxane A2(TXA2) receptor-binding affinity, whereas that of sevoflurane is caused by COX1 inhibition. Other investigators 7have postulated that the effect of halothane on thrombin-induced platelet aggregation may be mediated by decreased inositol 1,4,5-triphosphate (IP3) or increased cyclic adenosine monophosphate levels. In contrast to volatile anesthetics, however, published information regarding the effect of propofol on platelet aggregation is inconsistent. In vitro studies by Cruz et al. 11and Aoki et al. 12showed that platelet aggregation was suppressed by propofol, but an in vivo study by Turkan et al. , 13no alteration in platelet function was observed during anesthesia with propofol. The present study was undertaken to elucidate the effects of propofol on platelet function and the underlying mechanism to explain this inconsistency. Platelet aggregation, TXA2formation, IP3formation, and TXA2receptor-binding affinity were studied.
Materials and Methods
The study was conducted in accordance with the human research standards of the institutional ethics committee. Venous blood was obtained from 10 healthy volunteers (4 women and 6 men) aged 27 to 43 yr old (mean age, 34.9 ± 4.79) who had taken no drugs known to affect platelet aggregation for at least 2 weeks. The blood was placed in tubes containing a 10% volume of 3.8% weight/volume trisodium citrate and was centrifuged at 160 g for 10 min to prepare platelet-rich plasma (PRP) or at 1,600 g for 30 min to prepare platelet-poor plasma (PPP). The platelet count of the PRP was adjusted to 300,000/μl by adding PPP. PRP and PPP were then stored at room temperature. PRP was used for the aggregation studies and TXA2radioimmunoassay. To prepare washed platelets for the binding assay and IP3measurement, a 10% volume of 100 mM EDTA (p H 7.4) was added to PRP, and the mixture was centrifuged at 900 g for 15 min. The platelet pellet was suspended in buffer A, which was composed of 8 mM Na2HPO4, 2 mM NaH2PO4, 10 mM EDTA, 5 mM KCl, and 135 mM NaCl (p H 7.2), then recentrifuged at 900 g for 15 min. Platelet pellets for the binding assay were finally suspended in buffer B, which was composed of 25 mM Tris-HCl, 1 mM EGTA, 5 mM MgCl2, and 138 mM NaCl (p H 7.5), and the platelet counts were adjusted to 1,000,000/μl by adding buffer B. Platelet pellets for IP3measurement were finally suspended in HEPES buffer, which was composed of 10 mM HEPES, 145 mM NaCl, 5 mM KCl, 0.5 mM Na2HPO4, and 6 mM glucose (p H 7.5), and the platelet counts were adjusted to 1,000,000/μl by adding HEPES buffer. The washed platelets were stored in Ca2+-free buffer at 0°C for 1 h. A few minutes before being used, the Ca2+concentration of the washed platelets suspended in HEPES buffer was adjusted to 1 mM by adding CaCl2.
Propofol and 10% Intralipos (Yoshitomi Pharmaceutical Industries, Osaka, Japan) were appropriately diluted previously by distilled water to adjust adding volume to 1–2 μl. An aliquot of propofol (10–100 μM) or an equivalent volume of 10% Intralipos was added directly to PRP using a micropipet (0.4–2 μl) and incubated at 37°C for a few minutes. The ingredient of Intralipos is essentially the same as the solvent of propofol or Intralipid (commercially available from several companies, including Ohtsuka Pharmacological Company, Tokyo, Japan). Five minutes before analysis, 200-μl aliquots of PRP containing propofol or Intralipos were placed into siliconized glass tubes that were kept at 37°C and stirred continuously before and during the experiments. Platelet aggregation induced by adding (+)-9,11-epithia-11, 12-methano-TXA2(STA2; 0.5–1 μM), arachidonic acid (AA; 0.5–1 mM), prostaglandin G2(PGG2, 25 μM), adenosine diphosphate (ADP; 1–5 μM), or epinephrine (0.5–10 μM) was measured at 37°C by recording the increase in light transmission using an eight-channel aggregometer (MCM Hematracer VI; MC Medical Inc., Tokyo, Japan)(n = 5 each). The light transmission of PPP containing Intralipos, in volume equivalent to that of propofol to be tested, was taken as 100%. Percent aggregation at 7 min after stimulation was analyzed statistically in the presence of propofol or equivolume of Intralipos by unpaired t test.
Radioimmunoassay of TXB2
Platelet aggregation was induced by ADP (1 μM), epinephrine (2 μM), AA (1 mM), and PGG2(25 μM) in the presence of propofol (40 or 100 μM) or an equivalent volume of Intralipos. Seven minutes after adding the aggregating agent, the reaction was terminated by adding one-tenth volume of 100 mM EDTA, precooled to 0°C, and the suspension was immediately placed into an ice bath. The suspension was centrifuged at 10,000 g at 4°C for 2 min to prepare platelet-free plasma, which was stored at −20°C until TXB2determination. The levels of TXB2, a stable metabolite of TXA2, in the platelet-free plasma were determined using a commercially available radioimmunoassay kit.
Assay of COX1 Activity
Because COX1 converts AA to PGG2and then to PGH2, and this is finally converted to malonaldehyde in the presence of hemoglobin and reduced glutathione, 14COX activity can be determined by measuring malonaldehyde formation. 15Aliquots of a reaction mixture containing 100 mM tris/HCl (p H 8.0), 5 mM glutathione, 5 μM hemoglobin, and 10 U COX, together with 10–100 μM propofol or the equivalent volume of Intralipos, were preincubated at 37°C for 5 min before starting the reaction. The reaction was started by adding AA (to a final concentration of 0.5 mM) and terminated after incubation at 37°C for 1 min by adding 0.2 ml 100% trichloroacetic acid in 1 M HCl. After adding 0.2 ml 1% thiobarbiturate solution, the mixtures were placed into a boiling water bath for 20 min, then cooled to room temperature and centrifuged at 2,700 g for 10 min. The absorbance of the clear supernatant at 532 nm was measured using a spectrophotometer (UV-visible recording spectrophotometer, UV-160A; Shimazu Corporation, Kyoto, Japan). A standard curve was constructed for malonaldehyde by adding known amounts of malonaldehyde to a reaction mixture with the aforementioned composition, except for the AA. Malonaldehyde formation was calculated using this standard curve.
Propofol (100 μM) or an equivalent volume of Intralipos was added directly to test tubes containing 100 μl washed platelets and 700 μl buffer B. After incubation for 5 min at 37°C, a [3H]-labeled TXA2receptor antagonist, 5Z-7-(3-endo-([ring-4-3H] phenyl) sulphonylamino-[2.2.1.] bicyclohept-2-exo-yl) heptenoic acid ([3H]-S145; 1–5 nM), was added and incubation continued at 37°C for 30 min. The reaction was terminated by adding 5 ml 5 mM Tris-HCl buffer (p H 7.4) that had been precooled to 0°C. Each mixture was then filtered in vacuo through a Whatman GF/C filter (Whatman International Ltd., Maidstone, United Kingdom) and washed three times with 5 ml precooled Tris-HCl buffer. The radioactivity on each filter (n = 5 each) was determined using a liquid scintillation analyzer (Tri-Carb 1900 CA; Packard Instrument Co., Meriden, CT). A zero control (without Intralipos) was also run to determine the effect of Intralipos alone.
Assay of IP3Formation
Aliquots of washed platelets (120 μl; counts 1,000,000/μl) were stimulated by STA2(final concentration of 1 μM, 30 μl) after preincubation at 37°C with propofol (30 and 100 μM) or an equivalent volume of Intralipos for 5 min. After stimulation for 5, 15, 30, and 60 s, the reaction was terminated by adding 50 μl ice-cold 10% perchloric acid solution. The samples were kept in an ice bath for 30 min and then centrifuged at 1,800 g and 4°C for 10 min. The p H of the supernatant was adjusted to 7.5 with a solution of 1.53 M KOH and 75 mM HEPES. The samples were replaced in the ice bath for 30 min and then centrifuged at 8,000 g and 4°C for 3 min. The IP3levels of the supernatants were measured using a commercially available IP3assay kit.
The drugs used were epinephrine hydrochloride (Sigma, St. Louis, MO), AA, malonaldehyde (Nacalai Tesque Co., Kyoto, Japan), 10% Intralipos, COX, and PGG2(Cayman Chemical Co., MI). 3H-S145 was a gift from Shionogi Research Laboratories (Osaka, Japan), and STA2was a gift from Ono Pharmaceuticals (Osaka, Japan). Propofol was a gift from Zeneca Pharmaceuticals (London, United Kingdom). The radioimmunoassay kit was a thromboxane B23H-assay system, code TRK 780 (Amersham International plc, Buckinghamshire, United Kingdom). The IP3assay kit was a D-myo-Inositol 1,4,5-triphosphate [3H] assay system, code TRK.1000 (Amersham International plc).
All data are expressed as mean ± SD. Data of aggregation study and TXB2formation were analyzed using an unpaired t test. Data of COX activity, TXA2receptor-binding affinity, and IP3formations were analyzed using two-factor analysis of variance and Fisher's protected least significant difference. Differences at P < 0.05 were considered significant.
Platelet Aggregation Study
Adenosine diphosphate (1–5 μM) and epinephrine (0.5–10 μM) induced both primary and secondary aggregation. Primary aggregation induced by ADP and epinephrine was not affected by propofol (1–100 μM). Secondary aggregation induced by them was enhanced by a low concentration (40 μM) but suppressed by a high concentration (100 μM) of propofol (fig. 1A and 1B). AA (1 mM)-induced aggregation was enhanced by 40 μM and completely abolished by 100 μM propofol (fig. 1C). PGG2(25 μM)-induced aggregation was enhanced by 100 μM propofol (fig. 1D). STA2(1 μM)-induced aggregation was enhanced by propofol (40 and 100 μM)(fig. 1E). Table 1summarizes aggregation in percentages, i.e. the light transmission relative to that of PPP containing the equivalent volume of Intralipos, recorded 7 min after stimulation.
The TXB2level in the PRP supernatant was increased markedly by stimulation with ADP (1 μM), epinephrine (1 μM), AA (1 mM), and PGG2(25 μM). Propofol (40 μM) slightly suppressed ADP-, epinephrine-, and AA-induced TXB2formation, and 100 μM propofol suppressed AA-induced TXB2formation markedly. Propofol up to 100 μM did not affect PGG2-induced TXB2formation (fig. 2).
Intralipos alone increased COX1 activity compared with zero control (control without Intralipos;fig. 3). Propofol (30 and 100 μM) suppressed COX1 activity compared with Intralipos alone, and 100 μM propofol suppressed the activity compared with zero control as well (fig. 3).
Intralipos suppressed 3H-S145 binding to platelets in a dose-dependent manner compared with the zero control (control without Intralipos). The effect of propofol (10–100 μM) on binding was not different from that observed with Intralipos alone (fig. 4).
Inositol 1,4,5-triphosphate levels in resting platelets was 1.28 ± 0.52 and 2.41 ± 1.11 pmol/108platelets, respectively, in the absence and presence of propofol (100 μM). The time course of IP3concentrations in platelets stimulated by STA2(1 μM) is shown in figure 5; the IP3level was significantly higher in the presence of propofol (30 and 100 μM) than with Intralipos alone (fig 5).
The plasma concentrations of propofol at which 50% and 95% of patients do not respond to skin incision (Cp50and Cp95, respectively) have been reported to be 85.4 and 153.9 μM (or 15.2 and 27.4 μg/ml), respectively. 16Fentanyl administration decreases Cp50markedly; fentanyl (1 ng/ml) decreased Cp50by 63%. 16Other investigators have shown that Cp50s in the presence and absence of 67% N2O were 30.1 and 45.5 μM (or 5.36 and 8.11 μg/ml), respectively. 17However, because a large fraction of propofol exists in protein-bound form in plasma, 18clinically relevant concentrations of propofol in protein-free buffer may be as low as 1 μM. Thus, our experiments using washed platelets are complicated by having an applied free concentration of high propofol, which is much higher than clinically relevant values. In contrast, in experiments performed in plasma, such as the aggregation study, clinically relevant anesthetic concentrations of propofol will exceed 30 μM, although sedative concentration of propofol will be lower. Thus, the present study indicates that propofol has different effects on platelet aggregation depending on concentration within a clinically relevant range; relatively high concentrations of propofol suppress platelet aggregation, whereas relatively low concentrations enhance aggregation. These complex effects of propofol have probably caused the inconsistency in the estimation of its effects on platelet aggregation by previous investigators.
Propofol altered secondary aggregation without altering primary aggregation when aggregation was induced by weak agonists such as epinephrine and ADP. The binding of these agonists to platelets activates phospholipase A2to release AA. This is then converted by COX1 to PGG2and finally to TXA2(fig. 6 A), which plays a major role in the induction of secondary aggregation. Therefore, the alteration of secondary aggregation alone suggested that propofol affected the formation or function of TXA2. The finding that 100 μM propofol inhibited AA-induced platelet aggregation completely, but significantly enhanced PGG2- and STA2(a TXA2analog)-induced platelet aggregation, suggested that propofol suppresses the activity of COX1, the enzyme that converts AA to PGG2. The level of TXB2, a stable breakdown product of TXA2, was radioimmunoassayed to estimate TXA2levels. In the absence of the anesthetic, the TXB2level did not differ much from those reported previously. 19,20The finding that propofol suppressed AA-induced increase in TXA2but had no effect on PGG2-induced TXA2increase further supports the concept that propofol suppresses COX1 activity. The inhibition of COX1 activity by propofol was confirmed finally by direct measurement of COX1 activity in a platelet-free system.
Propofol is water-insoluble and is now available commercially only dissolved in Intralipos (or Intralipid). Therefore, from the clinical standpoint, the net effect of propofol and Intralipos, not the effect of propofol alone, would be relevant. This study shows that Intralipos itself affects platelet modestly in a different way from that of propofol. Intralipos increased COX1 activity, and this effect was overcome only by a high concentration of propofol. It suggests that, in clinical use, the antiaggregatory effect of propofol may be ameliorated by the effect of Intralipos.
The finding that propofol augmented PGG2- and STA2-induced aggregations and, although only at a low concentration, augmented AA-induced aggregation suggested that, in addition to suppression of COX1 activity, propofol can augment TXA2responsiveness. The TXA2receptor is a member of the seven transmembrane domain family of receptors that interacts with G protein. 21The G protein α subunit that links with this receptor is a member of the Gq family. 22Platelets possess the β and γ forms of phospholipase C. 23Phospholipase C-β is associated with G protein–linked receptors and can be activated by Gq. 23Phospholipase C hydrolyzes phosphatidylinositol 4,5-bisphosphate and produces IP324(fig. 6B). This IP3formation is one of early processes of TXA2-induced actions in platelets, and IP3thus formed mobilizes Ca2+from intracellular stores to trigger aggregation and is then broken down quickly. In this study, we found that supraclinical concentrations (> 30 μM in plasma-free system) increase the STA2-induced IP3production, which could contribute to the augmentation of platelet aggregation.
One possible explanation for these findings would be an increase in TXA2receptor-binding affinity. However, a receptor-binding assay using an isotope-labeled specific TXA2antagonist, 3H-S145, 25,26clearly ruled out this possibility, demonstrating that propofol itself does not alter the affinity and that, moreover, Intralipos suppresses it. The finding that propofol increased IP3levels in STA2-stimulated platelets without affecting TXA2receptor-binding affinity can be explained by any of the three following mechanisms: enhancement of Gq protein, enhancement of phospholipase C activity, or suppression of IP3breakdown. At present we cannot speculate as to which of the aforementioned mechanisms underlies the enhancing effect of propofol on platelet aggregability. Furthermore, in clinical use, this effect of propofol is counteracted to some extent by the Intralipos-mediated decrease in ligand affinity.
Backwell et al. 27showed that general anesthesia with propofol decreased blood loss compared with conventional inhaled agents during endoscopic sinus surgery. Although such clinical results may be coincidental because of a variety of factors, it does suggest the need for further well-controlled clinical studies documenting the effects of varying anesthetic agents in platelet function and bleeding.
We conclude that, in vitro , propofol has both suppressive and enhancing effects on platelet aggregation, and that the suppressive effect is caused by the suppression of COX1 activity, whereas the enhancing effect is caused by increased IP3formation.