Comparison of the vasorelaxant effect of nitroprusside and nitroglycerin in the human radial artery in vitro

Abstract

Aims

In recent years the radial artery (RA) has been re-introduced for coronary artery bypass grafting (CABG). However, the potential for vasospasm remains a clinical problem when this vessel is employed and effective vasodilator agents are required to combat vasospastic events. This in vitro study was designed to compare the vasodilator effects of sodium nitroprusside (SNP) and nitroglycerin (NTG) in the human RA.

Methods

Human RA segments (n = 70) were taken from vessels employed for grafting in patients undergoing CABG. Concentration-relaxation curves for SNP and NTG were established in RA which had been precontracted with various vasoconstrictors (potassium chloride [K⁺], the thromboxane A₂ mimetic agent U46619 or endothelin-1 [ET-1]).

Results

Both SNP and NTG caused complete relaxation and EC₅₀s were similar except that NTG was 6.2-fold more potent than SNP in U46619-induced contraction (−7.50 ± 0.16 vs−6.71 ± 0.38 log m, P = 0.04). After treatment with verapamil and NTG solution during harvesting, the RA segments responded with reduced maximal relaxation to NTG (84.9 ± 3.9%, compared with 98.8 ± 0.8% in the control, P = 0.004). The vessel became less sensitive to NTG (Ec₅₀: −6.29 ± 0.4 vs −7.50 ± 0.16 log m, P = 0.01). In investigations carried out with SNP, tolerance was only seen in the magnitude of the relaxation (87.4 ± 4.7%vs 99.2 ± 0.6% in the control, P = 0.03).

Conclusions

Both NTG and SNP are potent vasodilators in the RA. NTG may have more potent effects in certain situations (constriction related to thromboxane A₂). However, tolerance to NTG may develop. A cross tolerance to SNP may exist but the effect is weak so that SNP may be preferable to NTG as a vasodilator in the RA postoperatively. Other vasodilators may be the drugs of choice under such circumstances.

Introduction

In coronary artery bypass grafting (CABG) surgery, arterial grafts have been used with increasing frequency because the long-term patency is expected to be superior to that seen with venous grafts. In comparison with arterial grafts such as the gastroepiploic artery and the inferior epigastric artery, which have been employed for grafting in recent years, the radial artery (RA) was initially employed as a graft for CABG in 1971 [1]. However, it was soon abandoned because of the relatively high incidence of vasospasm and low patency rates encountered [2, 3]. With increased knowledge of the propensity of this artery [4, 5] to vasospasm and of how to overcome the spasm using pharmacological agents [3, 6], this arterial graft is now regaining popularity [3, 7, 8].

The aetiology of vasoconstriction in the RA [4, 5] and in other arterial grafts remains unknown and many vasoconstrictor substances may be involved [9, 10]. Vasodilators such as calcium antagonists, ACE-inhibitors, long-acting nitrates, and phosphodiesterase III inhibitors have been used in different indications in patients undergoing CABG. Vasodilator agents have also been used to alleviate vasospasm in the RA [3, 4, 6–8]. In fact, the use of calcium antagonists is one of the key components in the re-introduction of the RA for CABG surgery [3]. In addition, nitrovasodilator drugs such as nitroglycerin (NTG) have been used for both local [6] or systemic administration in patients receiving RA grafts. However, the interactions between different vasoconstrictors and nitrovasodilators in this vessel have not been well studied. Two nitrovasodilators, NTG and sodium nitroprusside (SNP), are commonly used in patients receiving the RA during the postoperative period. SNP is given, if necessary, immediately postoperatively for its antihypertensive effect since hypertension frequently occurs early after CABG and NTG is used in ischaemic heart disease mainly because of its effect of dilating epicardial conductance arteries to increase collateral blood flow to the ischaemic myocardium and to decrease left ventricular preload [11]. Both NTG and SNP have relaxant effects on the RA but such effects have not been compared in this vessel. The present study was therefore designed to investigate the interactions between nitrovasodilators and various vasoconstrictors in the human RA with emphasis on the comparison between SNP and NTG.

During the harvesting of the RA, the artery is treated with vasodilators such as papaverine [3, 7, 8] or, in our clinic [6], verapamil plus NTG (VG) solution to overcome vasospasm. Tolerance to nitrovasodilators is a well-recognized clinical phenomenon [12, 13] and if this exists in the RA, the clinical significance would be that postoperative therapy with nitrovasodilators would be less effective. The second part of the study was designed to investigate whether tolerance to nitrovasodilators exists in the RA after it has been treated with a combination of NTG and verapamil solution for 30 min during surgery.

Methods

Seventy human RA segments were collected from patients undergoing CABG with RA grafts. Approval to use discarded RA tissue was given by the Hospital Ethics Committee. Discarded RA segments were collected in a container with oxygenated Krebs solution maintained at 4° C, and immediately transferred to the laboratory. The RA was transferred into a glass dish and dissected out from its surrounding connective tissue. During this period, the artery was frequently washed with Krebs’ solution. The vessels were cut into 3 mm long rings and suspended on wires in organ baths [14]. The number of rings taken from each patient varied from 2–4. The Kreb’s solution was aerated with a gas mixture of 95% O₂-5% CO₂ at 37° C.

Organ-bath technique

A technique that allowed the equilibration of the vascular rings at a pressure comparable with that experienced in vivo was used. The details of the technique have been previously published [14]. Briefly, each ring was stretched in progressive steps to determine its length-tension curve. An iterative fitting program (VESTAND 2.1, Yang-Hui He, Princeton University, NJ) was used to determine the pressure and the internal diameter. When the transmural pressure in the rings reached 100 mmHg, the stretching procedure was stopped and the rings were released to 90% of their internal circumference at 100 mmHg. This degree of the passive tension was then maintained throughout the experiment. The endothelium was preserved by cautiously dissecting and mounting the rings since it is known to play a modulatory role in the contractility of the human RA [5].

Protocol

After the stretching procedure, the rings were equilibrated for at least 45 min. The following experimental protocols were employed:

Relaxation

SNP- or NTG-induced relaxation was studied in rings contracted with the depolarizing agent K⁺ (25 mm), the thromboxane A₂ mimetic U46619 (10 nm), or endothelin-1 (ET-1, 10 nm). The concentrations employed were submaximal as determined from the curve fitting equation [5, 14]. Cumulative concentration-relaxation curves to SNP or NTG were then established. Only one concentration-relaxation curve was obtained from each RA ring. From 5–10 rings (taken from at least three patients), a mean concentration-relaxation curve was constructed.

Tolerance to nitrovasodilators after VG treatment

The RA segments taken from sections which had been immersed in VG solution for 30 min during surgery were used for this study. The relaxant effect of NTG or SNP was compared with that obtained in arteries which had not been exposed to VG solution (control) in U46619- constricted RA. During the preparation of the vascular rings, they were repeatedly washed with Krebs’ solution.

The components of the VG solution were [6]: verapamil hydrochloride 5 mg, NTG 2.5 mg, heparin 500 Units, 8.4% NaHCO₃ 0.2 ml and Ringer’s solution 300 ml yielding a concentration of 30 μmol l⁻¹ of verapamil and NTG in an isotonic solution with pH 7.4.

Data analysis

The EC₅₀s were calculated from each concentration-contraction (or relaxation) curve by a curve-fitting equation: E=MA^p/(A^p+K^p) where E is response, M is maximal contraction (or relaxation), A is concentration, K is Ec₅₀ concentration, and p is the slope [14]. A computerized program was used for curve-fitting [14]. From this equation, the mean EC₅₀ value ± s.e.mean was calculated in each group. Data were expressed as mean and 95% confidence intervals. An unpaired t-test was used to test statistical significance between different constrictors and dilators regarding the maximal response or EC₅₀.

Materials

Drugs used were: endothelin-1 (Peptides International, Louisville, Kentucky); nitroglycerin (SoloPak Laboratories, Franklin Park, IL); sodium nitroprusside (F. Hoffmann-La Roche & Co. Ltd, Basle, Switzerland). Other chemicals were purchased from Sigma, St Louis, MO, USA. Stock solution of endothelin-1 and U46619 was kept frozen until required.

Results

Resting vessel parameters [14]

The mean internal diameter of the 70 rings at an equivalent transmural pressure of 100 mm Hg was 2.5 ± 0.1 mmHg (mean ± s.e.mean) as determined from the normalization procedure. When the rings were relaxed to a resting diameter of 90% of this value, the equivalent transmural pressure was 69.1 ± 2.0 mmHg, and the resting tone was 2.4 ± 0.2 g.

Relaxation by SNP or NTG in the RA precontracted by K⁺, U46619, or ET-1

Both SNP and NTG caused a full or nearly full relaxation in K⁺,− U46619- or ET-1-mediated precontraction (Figure 1). NTG was marginally more potent than SNP in the U46619-induced contraction (−7.50 ± 0.16 vs−6.71 ± 0.38 log m, P = 0.04, 95% CI: −1.56, −0.02 log m). However, there was no difference in the reversal of the contraction induced by either K⁺ (−6.49 ± 0.42 vs−6.19 ± 0.22 log m, P = 0.6, 95% CI: 0.70, 22.65 log m) or ET-1 (−6.70 ± 0.31 vs−6.80 ± 0.12 log m, P = 0.7, 95% CI: 1.27, 27.56 log m).

Figure 1.

Mean concentration (−log m)−response (% relaxation) curves for nitroglycerin (NTG, •, n = 6) and sodium nitroprusside (SNP, ○, n = 5) in the radial artery precontracted by potassium chloride (K⁺, 25 mm, 1a), thromboxane A₂ mimetic U46619 (10 nm, n = 9 for NTG and n = 6 for SNP, 1b), and endothelin-1 (ET, 10 nm, n = 6 for NTG and n = 7 for SNP, 1c). Vertical error bars are 1 s.e.mean of mean values.

Tolerance of nitrovasodilators after exposure to VG solution

After treatment with VG solution during harvesting, the maximal relaxation induced by NTG was 84.9 ± 3.9%, compared to 98.8 ± 0.8% in the control (95% CI: 5.1, 22.7%, P = 0.004) and the vessels were less sensitive to NTG (EC₅₀ −6.29 ± 0.4 vs −7.50 ± 0.16 log m, 95% CI: −2.13, −0.30 log m, P = 0.01) (figure 2a). The maximal relaxation to SNP was reduced (87.4 ± 4.7%vs 99.2 ± 0.6% in the control; 95% CI: 1.3, 22.3%, P = 0.03). However, there were no significant changes in the sensitivity to SNP (EC₅₀−6.48 ± 0.26 vs−6.80 ± 0.45 log m, P = 0.5, 95% CI: 0.47, 15.36 log m). (figure 2b).

Figure 2.

Mean concentration (−log m)−response (% relaxation) curves for nitroglycerin (NTG, n = 10, 2a) and sodium nitroprusside (SNP, n = 6, 2b) in the radial artery treated with verapamil and NTG solution (VG) topically during surgery (○) compared with the control (non-treated, •, n = 9 for NTG and n = 6 for SNP). The precontraction was induced by thromboxane A₂ mimetic U46619 (10 nm). *P < 0.05; **P < 0.01.

Discussion

The results of this study show that in the human RA, a vessel which is regaining popularity in CABG surgery, 1) both SNP and NTG induced a full or near-full relaxation in K⁺−, TXA₂ (U46619)-, or ET-1-mediated contraction and NTG may be marginally more potent than SNP in thromboxane A₂-mediated contraction; 2) tolerance to nitrovasodilators exists after pretreatment with NTG and verapamil; and 3) there may be a weak cross-tolerance between NTG and SNP.

The RA contracts with many vasoconstrictors [4, 5]. Those used in the present study, were chosen since during cardiopulmonary bypass plasma concentrations of TXA₂ and ET-1 have been reported to be increased [15, 16] and these two potent vasoconstrictors may constrict arterial grafts during perioperative period. The depolarizing agent K⁺-mediated contraction was employed as a non-specific vasoconstrictor.

NTG attenuates the vasoconstrictor effect of TXA₂ and ET-1 in the human internal mammary artery (IMA) [17–20] but its effects in the human RA have not been reported previously. Similarly, the effect of SNP in the RA has not been well studied.

This investigation demonstrates that both SNP and NTG are effective vasodilators in the human RA constricted by various agents. This has potential clinical implications with regard to the treatment of graft spasm after CABG since both SNP and NTG are frequently used during the postoperative period. NTG is 6.2-fold more potent than SNP in relaxing vessels which have been constricted with U46619. However, there is no major difference between SNP and NTG with regard to the relaxation of arteries constricted by the other vasoconstrictors. The action of these two vasodilators are mediated by the release of nitric oxide (NO).

We recently reported that in the human IMA [20] NTG is more potent than SNP in relaxing phenylephrine-induced contraction but both have similar vasorelaxant effects in vessels preconstricted with angiotensin II or ET-1 [20]. This is in accordance with the results of the findings in the present study in the human RA.

Tolerance to NTG is well recognized both clinically and experimentally [12, 13]. It has been proposed that NTG-induced tolerance affects at least two major sites in the cascade of events between the initial site of NTG action and guanylate cyclase activation [13]. It was reported that 20 min incubation with NTG resulted in tachyphylaxis to NTG in porcine coronary arteries [21]. In our previous studies [17–20, 22], we have repeatedly observed that the effect of NTG is less potentin vessels which have been exposed to VG solution than in untreated vessels when it is used to prevent contraction and we suspect that this is related to rapid tolerance. We were therefore interested in examining whether tachyphylaxis to NTG develops in the RA following exposure to VG solution which contains NTG and may render the graft less sensitive to NTG during the perioperative period. Although it is unknown what role the other vasodilator verapamil in the VG solution plays on the tolerance to NTG, to our knowledge, there is no report on the cross tolerance between verapamil and NTG. The results support our hypothesis and show that, following treatment with VG solution, the RA was less responsive to NTG as demonstrated by the reduced maximal relaxation as well as the decreased potency (higher EC₅₀) of the drug.

Another aim of this study was to investigate whether the VG solution also influences the responses to SNP, another commonly used nitrovasodilator reported to show cross tolerance with NTG [23], although other studies found that the vasodilator effect of SNP is only marginally suppressed in NTG-tolerant arteries [24]. The present study confirmed this cross-tolerance which is evidenced by the reduced maximal relaxation to SNP after exposure to VG solution. However, the cross-tolerance is weak and the potency of SNP was maintained as shown by the unchanged EC₅₀ of this vasodilator. This suggests that SNP may still have a satisfactory vasodilator effect in the RA postoperatively.

The findings from this study may be clinically relevant. For example, in a patient on high dose NTG before CABG, tolerance to NTG may develop and reduce the effect of NTG perioperatively but the artery may still respond well to SNP although its effect may be slightly reduced due to cross-tolerance. Alternatively, other vasodilators such as calcium channel blockers may be considered in situations when spasm of the RA graft occurs where we [6] and others [3] have reported vasodilator effects in the RA.

In conclusion, our results show that both NTG and SNP are potent vasodilators in the RA. NTG may have more potent effects under certain conditions (contraction related to TXA₂). However, tolerance to NTG may develop. Cross-tolerance to SNP may exist but the effect is weak so that SNP may be superior to NTG with regard to the vasodilator effect in the RA postoperatively. These effects should be taken into account in the perioperative management in cases of RA grafting. Alternatively, other vasodilators may be considered when vasospasm occurs.