Secondary Logo

Journal Logo

Propofol-Associated Dilation of Rat Distal Coronary Arteries Is Mediated by Multiple Substances, Including Endothelium-Derived Nitric Oxide

Park, Kyung W. MD; Dai, Hai B. MD; Lowenstein, Edward MD; Sellke, Frank W. MD

Cardiovascular Anesthesia

Previous in vitro studies on the effect of propofol on coronary arteries have shown variable results, ranging from constriction to no effect to dilation.Although most of these studies reported that the observed effect is endothelium-independent, propofol also releases nitric oxide from cultured porcine endothelial cells. The present study examines the direct effect of propofol in rat distal coronary arteries in vitro, especially in regard to endothelial dependence and involvement of the adenosine triphosphate (ATP)-sensitive potassium channels (KATP channels). Forty-three subepicardial arteries (size 91.1 +/- 15.8 micro meter) from Wistar rats were studied in vitro in a no-flow, pressurized (40 mm Hg) state, using an optical density video detection system. After preconstriction with the thromboxane analog U46619 1 micro Meter, relaxation responses to increasing concentrations of propofol (10-6-10 (-4) M) were measured after 1) endothelial denudation, 2) pretreatment with the nitric oxide synthase inhibitor NG-nitro-L-arginine (L-NNA), 3) pretreatment with the cyclooxygenase inhibitor indomethacin, 4) pretreatment with the KATP channel blocker glibenclamide, or 5) no intervention (control). Propofol produced a significant concentration-dependent vasodilation of the U46619-preconstricted coronary arteries. This effect was significantly attenuated by endothelial denudation, pretreatment with L-NNA, or indomethacin, but was not affected by glibenclamide. We conclude that propofol has a direct vasodilatory effect on distal coronary arteries in rats. This effect is primarily endothelium-dependent and is mediated by multiple substances, including nitric oxide (NO) and a vasodilatory prostanoid. The effect is not mediated by opening of the KATP channels.

(Anesth Analg 1995;81:1191-6)

Department of Anesthesia and Critical Care and Department of Surgery, Beth Israel Hospital, Harvard Medical School, Boston, Massachusetts.

Section Editor: Kenneth J. Tuman.

Supported in part by US Public Health Service Grant R29 HL-46716 and by a grant from Beth Israel Anesthesia Foundation.

Presented in part at the 1995 Society of Cardiovascular Anesthesiologists annual meeting in Philadelphia, PA, May 1995.

Accepted for publication July 20, 1995.

Address correspondence and reprint requests to Frank W. Sellke, MD, Department of Surgery, Division of Cardiothoracic Surgery, Beth Israel Hospital, Dana 905, 330 Brookline Ave., Boston, MA 02215.

Propofol was originally popularized in ambulatory surgery for rapid recovery and low incidence of nausea and vomiting. Recently it has been suggested for use in cardiac surgery [1] and for sedation after coronary artery surgery [2]. Yet only a few studies examining the direct effect of propofol on coronary vessels have been performed. These have produced variable results. Introna et al. [3] demonstrated, in canine coronary artery rings, that propofol produced an endothelium-independent biphasic change in tension--constriction below 10-5 M and relaxation above the supraclinical concentration of 10-4 M. In similar preparations, Coughlan et al. [4] demonstrated that propofol had no significant vasomotor effect in clinical concentrations, but a relaxant effect only at supraclinical concentrations. On the other hand, Yamanoue et al. [5] showed in porcine coronary arteries that propofol produced endothelium-independent relaxation even in clinically meaningful low concentrations (10-7-10-6 M) and suggested that propofol acts as a calcium-channel blocker. Although these studies suggested that propofol-induced relaxation was endothelium-independent, there is evidence [6] that propofol can stimulate nitric oxide release from cultured porcine aortic endothelial cells.

The adenosine triphosphate (ATP)-sensitive potassium channels (KATP channels), first discovered by Noma [7] in cardiac myocytes, have since been described in arterial smooth muscle as well [8]. These channels are believed to play an important role in maintenance of coronary vascular tone [9], coronary microvascular autoregulatory responses [10], and in vascular response to ischemia [11].

In the present study, we investigated the vasomotor effect of propofol in rat distal coronary arteries and examined the endothelial dependence of the observed effect. Additionally, we examined the role of the KATP channels in propofol-induced vasomotion.

Back to Top | Article Outline


In accordance with institutional Animal Care Committee standards, Wistar rats of either sex, weighing 100-150 g, were anesthetized by injecting ketamine 40 mg/kg and xylazine 5 mg/kg, intraperitoneally. Subepicardial microvessels were prepared as described previously [12]. Each vessel was placed in a vessel chamber, cannulated with dual micropipettes (50-75 micro meter in diameter), and secured with 10-0 sutures. The vessel was bathed continuously with Krebs buffer, gassed with 95% O2-5% CO2 mixture, and maintained at 37 degrees C and pH of 7.4. The total volume of Krebs buffer circulating in the vessel chamber, buffer reservoir, and the connecting tubing was 100 mL. PO2 in the vessel chamber exceeded 400 mm Hg. As the vessel was studied in a no-flow state, the pressure in the micropipettes was maintained at 40 mm Hg to provide distention. The vessel was visualized and its internal lumen diameter measured and recorded, as previously described [12,13].

In order to establish the stability of our vessel preparation, 11 vessels (equilibrated size 99.5 +/- 14.6 micro meter), including 5 vessels denuded of endothelium, were monitored for changes in internal diameter. The vessel diameter equilibrated within 10 min and remained stable thereafter for at least 3 h.

Five endothelium-intact vessels (size 109.0 +/- 7.6 micro meter) were equilibrated for 30 min. Then, each vessel was subjected successively to the thromboxane analog U46619 1 micro Meter, acetylcholine 10 micro Meter, and sodium nitroprusside 1 micro Meter. After rinsing with fresh Krebs solution and reequilibration, the vessel was then subjected to KCl 100 mM. The vessel was rinsed again and equilibrated. At 90 mins, the vessel was subjected to the same set of interventions and changes in internal diameter were monitored.

Similarly, for five endothelium-denuded vessels (size 88.4 +/- 13.5 micro meter), responses to U46619, acetylcholine, sodium nitroprusside, and KCl after 30 min of equilibration versus after 90 min of equilibration were compared.

After equilibration of each microvessel for at least 20-30 min in the vessel chamber, a baseline internal diameter was measured. The vessel was then preconstricted with the thromboxane analog U46619 1 micro Meter. Propofol was added to the Krebs buffer solution in its commercially available 1% Intralipid Registered Trademark (Zeneca Pharmaceuticals, Wilmington, DE) emulsion (56 mM or 10 mg/mL of propofol, 100 mg/mL of soybean oil, 22.5 mg/mL glycerol, and 12 mg/mL egg lecithin), cumulatively to obtain final concentrations of approximately 1.4, 4.2, 14, 42, and 140 micro Meter (0.25, 0.75, 2.5, 7.5, and 25 micro gram/mL, respectively). Because of the presence of lipid in the solution and using the octanol:water partition coefficient for propofol of 5012:1 (per manufacturer's insert), the corresponding calculated aqueous phase concentrations of propofol were approximately 1.3, 3.1, 6.2, 8.8, and 10.4 micro Meter (0.22, 0.55, 1.1, 1.6, and 1.9 micro gram/mL, respectively). Higher concentrations of propofol could not be studied due to optical interference of the turbid lipid solution that made it impossible to measure vessel diameters. At each concentration, the internal diameter was measured and the percent relaxation from U46619-induced preconstriction was calculated (propofol control group: n = 10, baseline size 84.7 +/- 13.9 micro meter, range 56-100). Vessels for vehicle control (n = 6, size 107.0 +/- 8.0 micro meter, range 97-123) were exposed to equivalent volumes of Intralipid Registered Trademark and changes in the internal diameter were measured.

Additional vessels were preincubated with the nitric oxide synthase inhibitor NG-nitro-L-arginine 10 micro Meter (n = 6, size 87.4 +/- 19.7 micro meter, range 60-114), the cyclooxygenase inhibitor indomethacin 10 micro Meter (n = 7, size 85.1 +/- 13.9 micro meter, range 69-108), or the KATP channel blocker glibenclamide 1 micro Meter (n = 6, size 90.0 +/- 7.8 micro meter, range 83-108). After preconstriction with U46619 1 micro Meter, the vessels were exposed to increasing concentrations of propofol as described above. Changes in the internal diameter were measured. Additional vessels (n = 8, size 95.6 +/- 13.4 micro meter, range 68-107) were denuded of the endothelium by repeatedly abrading the luminal surface with a human hair. Endothelial denudation was confirmed after preconstricting the vessel with U46619 1 micro Meter and then observing lack of dilation to acetylcholine 10 micro Meter, but presence of normal dilation to sodium nitroprusside 10 micro Meter. These vessels were then flushed with fresh Krebs solution, reequilibrated, and then exposed to propofol, as described above.

At the end of each experiment, the vessel chamber was flushed with fresh Krebs buffer and the vessel re-equilibrated at 37 degrees C. KCl was then added to a final concentration of 100 mM and the internal diameter was measured. Only those vessels that constricted by at least 15% to KCl at the end of each experiment were considered still viable and included for data analysis. Forty-three vessels (size 91.1 +/- 15.8 micro meter, range 56-123) from 23 rats, not counting the vessels used to test the stability of the preparation, met this criterion and are the subject of this report.

All drugs were dissolved in ultradistilled water, except that U46619 was dissolved in ethanol and water and glibenclamide was dissolved in dimethyl sulfoxide and water. Neither ethanol nor dimethyl sulfoxide had vasomotor effects at the concentrations used in this study.

Vasomotor responses of the vessels after 30 min of equilibration versus 90 min of equilibration were compared with Student's t-test (two-tailed). Whether propofol or the vehicle had any concentration-dependent effect on the vessel diameter was determined by a one-way analysis of variance (ANOVA) (Scheffe's linear contrast). The concentration response curves to propofol under various conditions were compared by multiway repeated-measures ANOVA, with post-hoc multiple pairwise comparison when the initial ANOVA yielded P < 0.05. Significance was considered as P < 0.05. All statistics were calculated using True Epistat Registered Trademark software (Epistat Services, Richardson, TX).

Back to Top | Article Outline


Both endothelium-intact and endothelium-denuded vessels reached their equilibration diameters within the first 10 min and thereafter showed neither spontaneously vasoconstrictive nor vasodilatory tendencies.

Responses of the endothelium-intact and endothelium-denuded vessels to U46619, acetylcholine, sodium nitro-prusside, and KCl after 30 min of equilibration versus after 90 min of equilibration were not significantly different Figure 1 and Figure 2. Although the endothelium-intact vessels dilated to acetylcholine, the endotheliumdenuded vessels did not dilate significantly to acetylcholine after constriction with U46619. Both types of vessels dilated in response to sodium nitroprusside. Rinsing and reequilibration between interventions returned the vessel diameter to the baseline.

Figure 1

Figure 1

Figure 2

Figure 2

Propofol produced a significant (P < 0.001), concentration-dependent dilation of U46619-preconstricted rat coronary arteries. The vehicle produced a minimal, but statistically significant (P < 0.05) dilation of the vessels. The difference between propofol and its vehicle was significant (P < 0.001) Figure 3; Table 1.

Figure 3

Figure 3

Table 1

Table 1

Propofol-induced dilation of U46619-preconstricted arteries was significantly attenuated by endothelial denudation (P < 0.001), pretreatment with L-NNA (P < 0.001), or pretreatment with indomethacin (P < 0.001). However, none of the three measures totally abolished propofol-induced dilation. Propofol produced mild, but statistically significant dilation of the vessels even after endothelial denudation (P < 0.01). Finally, pretreatment with glibenclamide did not affect the response of the vessels to propofol significantly (P = 0.22) Figure 4; Table 2.

Figure 4

Figure 4

Table 2

Table 2

Back to Top | Article Outline


Propofol, touted for its rapid onset and offset of anesthesia and favorable side effect profile, has been used extensively in ambulatory surgery. In addition, its use has been suggested in cardiac surgical settings and postcardiac surgical intensive care units [1,2] as well as for cardioversion [14]. The rapid onset and offset of anesthesia from propofol is due to its high clearance. Clinically relevant blood concentrations of propofol include 0.8-1.0 micro gram/mL on awakening from propofol anesthesia [15-18], 1-2 micro gram/mL on long-term sedation in the intensive care unit [2,19], at least 2.5 micro gram/mL for satisfactory hypnosis [15], and 3-11 micro gram/mL for maintenance of satisfactory anesthesia [18]. With a bolus induction dose of propofol, blood propofol concentrations of 14-18 micro gram/mL are attained [20]. Based on these data, one can estimate that propofol blood concentrations from approximately 1 to nearly 20 micro gram/mL may be encountered in the clinical practice of anesthesia. Propofol protein binding is extensive and unbound fraction is at most 3% [21]. In the clinical range, the unbound propofol would be 0.60 micro gram/mL (3.4 micro Meter) or less. The range of propofol concentrations used in this study, 0.25-25 micro gram/mL, encompasses the clinically meaningful total blood concentrations of propofol, from the subclinical to the supraclinical. The range of aqueous phase concentrations of propofol in this study, 0.22-1.9 micro gram/mL, however, represents the upper end (and higher) of the clinical range of the unbound fraction of propofol. It is not known whether it is only the unbound propofol that exerts a vasomotor effect, or whether any bound fraction, perhaps in a liposome, may have a vasomotor effect [22].

In the present study, we have shown that (a) in the concentration range used, propofol produces a concentration-dependent vasodilation of distal coronary arteries, (b) this effect is not due to the vehicle, (c) this effect is primarily endothelium-dependent and appears to be mediated by multiple substances, including nitric oxide (NO) and a vasodilatory prostanoid, and (d) the KATP channels do not play a significant role in propofol-induced vasodilation.

Previous studies, performed with much larger vessels than those in the present study, have been consistent with an endothelium-independent vasodilatory effect of propofol. Chang and Davis [23] demonstrated, in rat thoracic aortic ring preparations, that propofol produced an endothelium-independent relaxation of KCl-constricted or phenylephrine-constricted rings and attenuated Ca2+-induced contraction of the rings exposed to Ca2+-free media and depolarized with KCl, concluding that propofol acted like a Ca2+-channel blocker. However, the EC (50) for these effects represented the upper end of the clinical range or the supraclinical--50 micro Meter, 800 micro Meter, and approximately 30 micro Meter, respectively. Park et al. [22] studied rat thoracic aortic and pulmonary artery rings and demonstrated an endothelium-independent relaxation of phenylephrine-preconstricted rings by supraclinical concentrations (aqueous concentrations of 22-66 micro Meter) of propofol. They also observed in vessels with intact endothelium that indomethacin attenuated propofol-induced relaxation. This suggested that propofol released a vasodilatory prostanoid from the endothelium. This finding is corroborated by our own observation. Introna et al. [3] demonstrated in canine proximal coronary arteries that propofol produced an endothelium-independent constriction in clinically relevant low concentrations (10-7-10-5 M), but dilation in higher concentrations. Both the constrictive and dilatory effects were abolished or attenuated by verapamil or a decrease in extracellular Ca2+ concentration. On the other hand, Yamanoue et al. [5] demonstrated, in porcine proximal coronary artery rings, that propofol produced an endothelium-independent relaxation even in the clinically relevant concentration range of 3 times 10-7 to 3 times 10-6 M. Coughlan et al. [4] were the first to distinguish between proximal (1.3-2.5 mm) and distal (250-500 micro meter) coronary arteries in dogs, and showed that propofol had no significant vasomotor effect in the clinical concentration range but produced greater dilation of the distal arteries than the proximal ones in higher concentrations. Endothelial dependence of the vasomotor effect was not examined. The only previous study to demonstrate endothelial NO production by propofol has been that of Petros et al. [6]. They showed that propofol, in supraclinical concentrations of 3 times 10-5 to 1 times 10-3 M, increased cyclic guanosine monophosphate production of cultured porcine aortic smooth muscle cells in the presence of endothelial cells and this effect was blocked by NG-nitro-L-arginine. Our study provides evidence that propofol produces primarily endothelium-dependent vasodilation of distal coronary arteries, and that the mediators may include both NO and a vasodilatory prostanoid. In addition, there appears to be an endothelium-independent effect of propofol as well in the high normal to supraclinical range, since endothelial denudation did not totally abolish propofol-induced dilation. One possible explanation accounting for the differences between our observation and those of others is that in distal coronary arteries where propofol produces greater dilation [4], the relative importance of endothelium-dependent dilation may be increased. Differences in species studied and experimental methods may also be contributory. Advantages and limitations of our methods in comparison to studies of vascular rings and measurements of vessel tension have been discussed previously [12].

The KATP channels, first described in cardiac myocytes [7], have since been described in many other tissue types, including vascular smooth muscle [8]. These channels are activated by decreases in intracellular ATP such as from ischemia or inhibitors of cell metabolism (e.g., cyanide, dinitrophenol) [24]. The channels are selectively blocked by the antidiabetic sulfonylureas and weakly, nonselectively blocked by the tetraethylammonium ion [25]. In addition, the hyperpolarizing actions of acetylcholine and vasoactive intestinal polypeptide are blocked by the sulfonylureas [8], suggesting that the endothelium-dependent hyperpolarizing factor (EDHF) released by these vasodilators may work by opening the KATP channels.

The relationship between the KATP channels and intravenous anesthetics has not been examined to date. In the present study, we found that propofol-induced vasodilation is not altered by glibenclamide, and therefore is not mediated by the KATP channels. Introna et al. [3] also reported that the biphasic vasomotor effect of propofol is not affected by the nonspecific K channel blocker tetraethylammonium ion. Since the KATP channels appear to be implicated in vasodilation mediated by EDHF [8], propofol probably does not release EDHF.

In summary, we have shown in in vitro preparations of rat distal coronary arteries that propofol produces a concentration-dependent vasodilation and that this effect is primarily endothelium-dependent and mediated by multiple substances, including NO, a vasodilatory prostanoid, but probably not EDHF.

Back to Top | Article Outline


1. Russel GN, Wright EL, Fox MA, et al. Propofol-fentanyl anaesthesia for coronary artery surgery and cardiopulmonary bypass. Anaesthesia 1989;44:205-8.
2. McMurray TJ, Collier PS, Carson IW, et al. Propofol sedation after open heart surgery. A clinical and pharmacokinetic study. Anaesthesia 1990;45:322-6.
3. Introna RPS, Pruett JK, Yodlowski EH, Grover E. Direct effects of propofol (2,6-diisopropylphenol) on canine coronary artery ring tension. Gen Pharmacol 1993;24:497-502.
4. Coughlan MG, Flynn NM, Kenny D, et al. Differential relaxant effect of high concentrations of intravenous anesthetics on endothelin-constricted proximal and distal canine coronary arteries. Anesth Analg 1992;74:378-83.
5. Yamanoue T, Brum JM, Estafanous FG. Vasodilation and mechanism of action of propofol in porcine coronary artery. Anesthesiology 1994;81:443-51.
6. Petros AJ, Bogle RG, Pearson JD. Propofol stimulates nitric oxide release from cultured porcine aortic endothelial cells. Br J Pharmacol 1993;109:6-7.
7. Noma A. ATP-regulated K+ channels in cardiac muscle. Nature 1983;305:147-8.
8. Standen JB, Quayle JM, Davies NW, et al. Hyperpolarizing vasodilators activate ATP-sensitive K+ channels in arterial smooth muscle. Science 1989;245:177-80.
9. Samaha FF, Heineman FW, Ince C, et al. ATP-sensitive potassium channel is essential to maintain basal coronary vascular tone in vivo. Am J Physiol 1992;262:C1220-7.
10. Komaru T, Lamping KG, Eastham CL, Dellsperger KC. Role of ATP-sensitive potassium channels in coronary microvascular autoregulatory responses. Circ Res 1991;69:1146-51.
11. Nichols CG, Lederer WJ. Adenosine triphosphate-sensitive potassium channels in the cardiovascular system. Am J Physiol 1991;261:H1675-86.
12. Park KW, Dai HB, Lowenstein E, et al. Heterogeneous vasomotor effect of isoflurane on rabbit coronary microvessels. Anesthesiology 1994;81:1190-7.
13. Halpern WG, Osol G, Coy G. Mechanical behaviour of pressurized in vitro prearteriolar vessels determined with a video system. Ann Biomed Eng 1984;121:463-79.
14. Valtonen M, Kanto J, Klossner J. Anaesthesia for cardioversion: a comparison of propofol and thiopentone. Can J Anaesth 1988;35:479-83.
15. Wessen A, Persson PM, Nilsson A, Hartvig P. Concentration-effect relationships of propofol after total intravenous anesthesia. Anesth Analg 1993;77:1000-7.
16. Servin F, Cockshott ID, Farinotti R, et al. Pharmacokinetics of propofol infusions in patients with cirrhosis. Br J Anaesth 1990;65:177-83.
17. Servin F, Farinotti R, Haberer J-P, Desmonts J-M. Propofol infusion for maintenance of anesthesia in morbidly obese patients receiving nitrous oxide. Anesthesiology 1993;78:657-65.
18. Short TG, Aun CST, Tan P, et al. A prospective evaluation of pharmacokinetic model controlled infusion of propofol in paediatric patients. Br J Anaesth 1994;72:302-6.
19. Albanese J, Martin C, Lacarelle B, et al. Pharmacokinetics of long-term propofol infusion used for sedation in ICU patients. Anesthesiology 1990;73:214-7.
20. Fassoulaki A, Farinotti R, Servin F, Desmonts JM. Chronic alcoholism increases the induction dose of propofol in humans. Anesth Analg 1993;77:553-6.
21. Kirkpatrick T, Cockshott ID, Douglas EJ, Nimmo WS. Pharmacokinetics of propofol (diprivan) in elderly patients. Br J Anaesth 1988;60:146-50.
22. Park WK, Lynch C III, Johns RA. Effects of propofol and thiopental in isolated rat aorta and pulmonary artery. Anesthesiology 1992;77:956-63.
23. Chang KSK, Davis RF. Propofol produces endothelium-independent vasodilation and may act as a Ca2+ channel blocker. Anesth Analg 1993;76:24-32.
24. Ashcroft FM. Adenosine 5 prime-triphosphate-sensitive potassium channels. Ann Rev Neurosci 1988;11:97-118.
25. Robertson DW, Steinberg MI. Potassium channel modulators: scientific applications and therapeutic promise. J Med Chem 1990;33:1529-41.
© 1995 International Anesthesia Research Society