Temperature Changes Stimulate Contraction in the Human Radial Artery and Affect Response to Vasoconstrictors

CARDIOVASCULAR

Temperature Changes Stimulate Contraction in the Human Radial Artery and Affect Response to Vasoconstrictors Aung Y. Oo, FRCS (CTh), Alan R. Conant, PhD, Michael R. Chester, MRCP, MD, Walid C. Dihmis, FRCS (CTh), and Alec W. M. Simpson, DPhil The Cardiothoracic Centre, The National Refractory Angina Centre, and Department of Human Anatomy and Cell Biology, University of Liverpool, Liverpool, United Kingdom

Background. Radial artery conduits are increasingly used in coronary artery bypass grafting as an additional arterial graft to the internal thoracic artery. Their reactive nature remains a concern, often necessitating the routine use of topically applied vasodilators, such as glyceryl trinitrate, papaverine, phenoxybenzamine, or calcium channel antagonists, in theatre. During preparation prior to surgery and grafting, radial artery conduits are exposed to cooling and rewarming. We investigated how these temperature changes would affect radial artery contractility and how commonly used topical treatments might be used to prevent this. Methods. Human radial artery was obtained excess to surgery and arterial sections used in organ bath tension experiments or for the culture of smooth muscle cells from medial explants. Results. The radial artery responded to rapid cooling by the addition of 22°C buffer with contraction. Gradual cooling, over a 20 to 30 minute period, reduced basal

tension and the response to potassium chloride (KCl) and noradrenaline. Subsequent rewarming from 22°C to 37°C reestablished contraction at precooled levels and led to an elevation of the basal tension. Increases in tension measured in the radial artery were paralleled by increases in intracellular calcium in smooth muscle cells. Contraction induced by rapid temperature changes could be blocked by glyceryl trinitrate but not by phenoxybenzamine. Papaverine and calcium channel blockers had only limited activity. Conclusions. Temperature changes commonly encountered in theatre during the preparation of radial artery grafts are likely to cause contraction. If rapid temperature change cannot be avoided during graft preparation, then topically applied glyceryl trinitrate will block these responses.

I

esterase inhibitors such as papaverine and milrinone, and calcium channel antagonists such as diltiazem and verapamil [6]. In addition, the irreversible ␣ adrenoceptor antagonist phenoxybenzamine is now routinely employed by many surgeons against catecholamineinduced contraction [7]. The patients own blood is the preferred storage solution [8] but there is no clear consensus on the optimal conditions which should be used to store radial artery grafts after harvest [6]. During preparation of the pedicled internal thoracic artery, it is kept inside the chest cavity and thereby maintained at body temperature. In contrast, the excised radial artery is stored in a receiver for upward of an hour before being grafted onto the target coronary artery. Depending on whether the graft is placed in a solution at theatre temperature or in a warmed solution that is allowed to cool gradually, it is subjected to either an immediate or slow cooling to theatre temperatures (18°C to 22°C). In addition, the graft then undergoes a second temperature change as it is rewarmed to 37°C after grafting. Cooling acts as a vasodilator in internal thoracic arteries, saphenous veins, aorta, coronary arteries, and pulmonary arteries [9 –13]; but in renal arteries, pulmo-

n coronary artery bypass grafting (CABG) the internal thoracic (mammary) artery is the first conduit of choice as an arterial graft [1], but with a trend toward total arterial revascularization other vessels such as the radial artery are being used. Midterm to long-term follow-up studies have demonstrated excellent patency rates, establishing the radial artery as a useful additional graft to the internal thoracic artery [2– 4]. The radial artery was initially rejected as a bypass conduit because of a high level of spontaneous contraction or spasm [5]. With changes in technique, where the graft was dissected en bloc together with its pedicle and dilated using blood and papaverine rather than mechanically using metal probes, the incidence of spasm was reduced [5]. However, spasm continues to remain a problem for surgeons using the radial artery [2– 4] and many routinely use topical vasodilators during surgical preparation to reverse contraction [6]. These include nitrovasodilators such as glyceryl trinitrate, phosphodi-

Accepted for publication Aug 18, 2006. Address correspondence to Dr Conant, The Cardiothoracic Centre, Liverpool NHS Trust, Thomas Drive, Liverpool L14 3PE, UK; e-mail: [email protected].

© 2007 by The Society of Thoracic Surgeons Published by Elsevier Inc

(Ann Thorac Surg 2007;83:126 –33) © 2007 by The Society of Thoracic Surgeons

0003-4975/07/$32.00 doi:10.1016/j.athoracsur.2006.08.035

nary veins, cerebral arteries, and in some studies using coronary arteries, cooling led to contraction [10, 13–15]. In coronary and cerebral arteries after cooling, when the vessel was rewarmed either by stepwise increases in temperature or in one single step, a second larger contraction was observed [14, 16]. The highly contractile nature of the radial artery [17] may make this vessel particularly sensitive to changes in temperature. No study to date has addressed this issue. Our study set out to investigate whether temperature changes experienced by radial artery grafts during preparation could lead to contraction. Because contraction is initiated by a rise in the intracellular calcium concentration in the smooth muscle of the vascular wall, we studied the effects of temperature changes on smooth muscle calcium homeostasis.

Material and Methods Materials The glyceryl trinitrate solution and papaverine solution were purchased from David Bull Laboratories (DBL, Warwick, UK) and Martindale Pharmaceuticals (Romford, UK), respectively. Diltiazem was purchased from Sigma Chemicals (Poole, UK) and stock solutions were

OO ET AL TEMPERATURE AND RADIAL ARTERY CONTRACTION

127

made up from powder prior to each experiment. Noradrenaline was purchased from Abbott Laboratories Ltd (Queenborough, UK). The 2-aminoethoxydiphenyl borate was purchased from Tocris Bioscience (Bristol, UK) and ryanodine from Merck Biosciences Ltd (Nottingham, UK). All other chemicals and reagents were purchased from Sigma Chemicals (Poole, UK) and VWR International (Leicester, UK). All tissue culture reagents were purchased from Invitrogen Ltd (Paisley, UK).

Tissue Preparation Approval for this study from the Liverpool Research Ethics Committee was given in December 2000. The radial artery sections were obtained, with informed consent, from 25 patients with a mean age of 65 ⫾ 4 years (22 male and 3 female) undergoing CABG at the Cardiothoracic Centre, Liverpool, UK. Depending on the practice of the surgeons, concerned samples provided were treated in theatre with either papaverine (0.6 mg/mL; n ⫽ 10) or phenoxybenzamine (1 mg/mL; n ⫽ 15). Sections of radial artery surplus to surgical requirements were collected from theatre into Dulbecco’s modified eagles medium (DMEM; Invitrogen Ltd, Paisley, UK) on ice and transferred immediately to the research laboratory.

Fig 1. Effect of rapid cooling on contractility of the radial artery. Representative traces showing the two types of response, either transient (A) or sustained (B) elicited by the addition of solution at 22°C as indicated by the vertical arrow. Average peak tension is shown in Figure C following rapid cooling. Sections of radial artery were pretreated with no drug (control; n ⫽ 7), 1 mM ethylene glycol bis-2-aminoethyl etherN,N=,N⬙,n=-tetraacetic acid (EGTA; n ⫽ 6), 0.5 mg/mL (2.2 mM) glyceryl trinitrate (GTN; n ⫽ 4), or 0.5 mg/mL (1.3 mM) papaverine (Pap; n ⫽ 4). The response was significantly reduced in treatment groups marked with an asterisk. (KCl ⫽ potassium chloride.)

CARDIOVASCULAR

Ann Thorac Surg 2007;83:126 –33

CARDIOVASCULAR

128

OO ET AL TEMPERATURE AND RADIAL ARTERY CONTRACTION

Organ Bath Tension Measurements Arterial rings (2 to 3 mm) were suspended in 25 mL 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES)-buffered-saline (HBS) composed of the following: sodium chloride (NaCl), 145 mM; potassium chloride (KCl), 2.5 mM; dibasic sodium phosphate (Na2HPO4), 1 mM; magnesium sulfate (MgSO4), 1 mM; HEPES, 10 mM; D-glucose, 10 mM; and either calcium chloride (CaCl2), 1mM or ethylene glycol bis-2-aminoethyl etherN,N=,N⬙,n=-tetraacetic acid (EGTA), 1 mM; pH 7.4 at 37°C, gassed with 100% oxygen. Rings were mounted between two fine wire stirrups connected to a force transducer and changes in isometric force were recorded on a Lectromed 5220 medium gain amplifier, connected to a Lectromed MT6 chart recorder (Lectromed Ltd, Letchworth, UK). Before studying the effects of temperature change all arterial rings were subjected to a pretensioning protocol at 37°C based on that of Chanda and colleagues [18], and Bond and colleagues [19]. The tension of each ring was set at 3g for one hour, then subsequently relaxed to 1g for 30 minutes. Arterial rings were stimulated with 60 mM KCl to test their contractility. After washout the effects of cooling were investigated using two protocols. Rapid cooling was investigated by the addition of 22°C HBS to the sample chamber resulting in an immediate drop in temperature. Slow cooling, over 20 to 30 minutes, was investigated by reducing the temperature of the circulating water jacketing the organ chambers. Arterial rings were treated with papaverine (0.5 mg/mL) and glyceryl trinitrate (0.5 mg/mL) at concentrations equivalent to those topically applied to grafts in theatre. Diltiazem (4.5 ␮g/mL; 10 ␮M) was used at a concentration sufficient to reverse KCl-induced contraction according to the published data [20]. Drugs were added five minutes prior to the stimulus and responses were compared with untreated control rings from the same sample.

Ann Thorac Surg 2007;83:126 –33

from 340/380 nm ratio signals. The Kd values used for fura-2 at 37°C and 22°C were 224 nM and 145 nM, respectively [22], and a Kd value of 190 nM at 30°C was derived using standard calcium solutions. A rapid temperature drop was initiated by the addition of 22°C HBS. Slower cooling (20 to 30 minutes) and subsequent rewarming (5 to 10 minutes) were achieved using the circulating water bath. Glyceryl trinitrate, 2-aminoethoxydiphenyl borate, and ryanodine, were added 5, 10, and 20 minutes, respectively, before the stimulus.

Data Analysis Data are presented as mean ⫾ standard error of the mean, where “n” represents the number of independent cell batches or samples studied. Data from arterial samples are expressed as per cent response to 60 mM KCl. Statistical comparisons were undertaken using a one way analysis of variance and a p value of 0.05 using the

Measurement of Intracellular Calcium ([Ca2⫹]c) in Human Radial Artery Smooth Muscle Cells (hRASMCs) The hRASMCs were grown from medial explants as described previously [21]. Briefly, after removal of the intima, the tunica media was gently chopped and the pieces washed in culture medium (DMEM plus 10% fetal bovine serum [FBS] and antibiotic) and left to adhere for two hours. Cells began to migrate out of the explants within 5 to 12 days and grew to confluency within one month. Responses were defined on cells used within the first four passages, confirmed as smooth muscle by positive immunoreactivty to ␣ smooth muscle actin (Vector, Burlingame, CA) and a profile of agonist-induced [Ca2⫹]c responses, including noradrenaline, angiotensin II, and endothelin-1 [21]. hRASMCs cultured to confluency on glass coverslips were loaded with the calcium sensitive dye, fura 2-AM, at 22°C for 120 minutes and mounted in a thermostatted chamber maintained by a circulating water bath. The fura 2-loaded cells were excited alternately at 340 and 380 nm with the emission collected at 510 nm. Photometric data were generated from images of individual cells and the [Ca2⫹]c calculated

Fig 2. The [Ca2⫹]c response to rapid cooling in radial artery smooth muscle cells (hRASMCs). Representative traces of single fura 2-loaded hRASMCs are shown in Figure A. Cells were maintained at 37°C in buffer and a temperature drop initiated by a rapid exchange of solution cooled to 22°C (indicated by the vertical arrow). The sample chamber was water jacketed at 37°C so that temperature returned to 37°C as indicated by the insert. Figure B shows average basal (e) and peak () [Ca2⫹]c values in cells subjected to a rapid temperature drop and cells treated with buffer adjusted to pH 7.4 at 37°C (pH; n ⫽ 3), 0.5 mM ethylene glycol bis-2-aminoethyl ether-N,N=,N⬙,n=-tetraacetic acid (EGTA; n ⫽ 3) and 0.05 mg/mL glyceryl trinitrate (GTN; n ⫽ 3). Peak [Ca2⫹]c was significantly different from basal values prior to cooling in sample groups marked with an asterisk.

OO ET AL TEMPERATURE AND RADIAL ARTERY CONTRACTION

129

Table 1. Effect of Cooling on Contraction Elicited by Noradrenaline or KCl in the Radial Artery

Tension at 37°C (g) Tension at 30°C (% 37°C) Tension at 22°C (% 37°C)

Basal (n ⫽ 3)

Noradrenaline (n ⫽ 3)

KCl (n ⫽ 3)

1.24 ⫾ 0.18

2.51 ⫾ 0.67

3.78 ⫾ 0.28

78.3 ⫾ 8.1a

88.1 ⫾ 10.0

80.8 ⫾ 14.4

72.3 ⫾ 8.5a

58.9 ⫾ 7.9a

72.3 ⫾ 10.7a

Arterial rings were stimulated with either 2 ␮m noradrenaline or 60 mM potassium chloride (KCl) at 37°C and the tension recorded. After agonist washout, rings were gradually cooled to 30°C and subsequently to 22°C and the responses repeated at each temperature. Data are expressed as a percentage of the initial response in the same ring at 37°C and a significant reduction (p ⬍ 0.05) in response was observed in groups marked with an a.

program Arcus QuickStat Biomedical (Hearne Scientific Software, Dublin, Eire).

Results The contractile capacity of arterial rings was evaluated by the addition of 60 mM KCl, which increased tension from an average basal value of 1.10 ⫾ 0.08g to 3.16 ⫾ 0.17g (n ⫽ 25). Data were not significantly different (p ⬍ 0.05) when collected from proximal (n ⫽ 9) or distal (n ⫽ 16) arterial samples. Depending on the practice of the surgeon concerned, they were either treated with phenoxybenzamine (n ⫽ 15) or papaverine (n ⫽ 10) in theatre. Papaverine-treated sections had an average basal tension of 1.21 ⫾ 0.10g, which increased to 3.69 ⫾ 0.31g upon addition of KCl and the average tension of phenoxybenzamine-treated samples increased from 0.95 ⫾ 0.10g to 3.24 ⫾ 0.20g upon addition of KCl. There was no significant difference (p ⬍ 0.05) between the data collected from these two groups. Phenoxybenzamine-treated arterial samples failed to respond to 1 ␮m noradrenaline.

Table 2. Effect of Cooling on [Ca2⫹]c Responses to Noradrenaline and KCl in Cultured Radial Artery Smooth Muscle

2⫹

[Ca ]c at 37°C (nM) [Ca2⫹]c at 30°C (% 37°C) [Ca2⫹]c at 22°C (% 37°C)

Basal (n ⫽ 8)

Noradrenaline (n ⫽ 3)

KCl (n ⫽ 5)

169 ⫾ 13

602 ⫾ 161

412 ⫾ 53

93.5 ⫾ 8.0

93.9 ⫾ 14.2

81.5 ⫾ 8.4

55.7 ⫾ 5.1a

42.9 ⫾ 8.8a

41.0 ⫾ 4.0a

Fura 2-loaded radial artery smooth muscle cells (hRASMCs) were stimulated with either 2 ␮m noradrenaline or 60 mM potassium chloride (KCl) at 37°C and the [Ca2⫹]c response recorded. The temperature was then reduced to either 30°C or 22°C and the response repeated. Data are expressed as a percentage of the initial response at 37°C and a significant reduction (p ⬍ 0.05) in response was observed in groups marked with an a.

Fig 3. Response of radial artery to rewarming from 22oC to 37oC. Sections of radial artery slowly cooled, over 90 minutes to 22oC, were rewarmed to 37oC using a temperature-controlled water bath. A representative trace is shown in Figure A with rewarming initiated at time ⫽ 0. Mean data for control and drug-treated rings are shown in Figure B. Samples of radial artery, slowly cooled to 22oC, were pretreated with no drug (control; n ⫽ 15), or 0.5 mg/mL glyceryl trinitrate (GTN; n ⫽ 5), 0.5 mg/mL papaverine (Pap; n ⫽ 5), 4.5 ␮g/mL diltiazem (Dil; n ⫽ 7) for 5 minutes prior to rewarming. The tension was significantly greater than control values prior to cooling in groups marked with an asterisk. (EGTA ⫽ ethylene glycol bis-2-aminoethyl ether-N,N=,N⬙,n=-tetraacetic acid; KCl ⫽ potassium chloride.)

Effect of a Rapid Temperature Drop on Radial Artery Tension Radial artery grafts stored at room temperature prior to grafting are subject to a rapid temperature drop. In arterial rings the addition of 22°C HBS resulted in an immediate, transient increase in tension (Fig 1A) giving an average peak value 18.8 ⫾ 3.0% KCl (n ⫽ 13) with tension returning to basal values within 90 seconds. Control additions of 37°C HBS did not affect resting tension. In six of the samples tested basal tension remained elevated by 20.3 ⫾ 7.1% for 15 to 20 minutes (Fig 1B). Responses obtained using HBS pH 7.4 at 22°C (n ⫽ 5) were not significantly different from those using standard HBS, thereby excluding the contribution of a change in pH. Cooling-induced contraction was abol-

CARDIOVASCULAR

Ann Thorac Surg 2007;83:126 –33

CARDIOVASCULAR

130

OO ET AL TEMPERATURE AND RADIAL ARTERY CONTRACTION

Ann Thorac Surg 2007;83:126 –33

ished by glyceryl trinitrate but not by EGTA or papaverine (Fig 1C). In samples treated with phenoxybenzamine a rapid temperature drop gave peak responses 23.0 ⫾ 2.9% KCl (n ⫽ 8).

Effect of a Rapid Temperature Drop on hRASMCs In hRASMCs with a resting [Ca2⫹]c of 225 ⫾ 11 nM (n ⫽ 15) at 37°C, the rapid addition of HBS at 22°C led to an immediate, transient increase in [Ca2⫹]c lasting for 90 to 120 seconds, in 56 ⫾ 7% of the cells tested. The average peak [Ca2⫹]c response was 381 ⫾ 32 nM (Fig 2). The rapid addition of 37°C HBS did not alter the baseline [Ca2⫹]c, whereas the addition of HBS (pH 7.4 at 22°C), led to a transient [Ca2⫹]c response of 446 ⫾ 86 nM (n ⫽ 3), demonstrating that the change in temperature was the initiating stimulus. In the presence of glyceryl trinitrate, the addition of cooled HBS did not raise [Ca2⫹]c significantly above basal values, but in the presence of EGTA the response was maintained (Fig 2B). The effect of papaverine was not tested due to its highly fluorescent properties. The [Ca2⫹]c responses induced by the rapid addition of cooled HBS were unaffected by the presence of 30 ␮M ryanodine (n ⫽ 4), but were abolished in the presence of 100 ␮M 2-aminoethoxydiphenyl borate (n ⫽ 4), a cell permeable inositol 1,4,5 trisphosphate receptor antagonist [23].

Effect of Gradual Cooling on Vasoconstrictor Responses in Radial Artery Rings and VasoconstrictorInduced [Ca2⫹]c Responses in hRASMCs Clinically, gradual cooling results from either a controlled reduction in the patient core temperature to 30°C or graft storage in warmed solutions allowed to cool to theatre temperatures. The contraction elicited by both KCl and noradrenaline was reduced by cooling, as was the basal tension (Table 1). To control for any loss of response over the course of the experiment, rings maintained at 37°C and stimulated in parallel with KCl gave responses 98.5 ⫾ 5.5% of the initial response. Responses were also recorded to noradrenaline after gradual rewarming to 37°C, giving values 92.4 ⫾ 2.1% of the initial response at 37°C. In hRASMCs, cooling reduced both basal [Ca2⫹]c and the average response to noradrenaline or KCl (Table 2).

Effect of Rewarming on Radial Artery Tension After cooling to 22°C, rewarming led to an increase in tension beyond that of the initial baseline at 37°C (Fig 3A) giving an average maximal response 43.4 ⫾ 9.8% KCl (n ⫽ 15) sustained for at least 20 minutes. In the presence of glyceryl trinitrate, papaverine or diltiazem rewarming did not increase tension beyond values obtained prior to cooling (Fig 3B). Phenoxybenzamine-treated samples gave an average maximal response 57.7 ⫾ 12.0% KCl (n ⫽ 9).

Effect of Rewarming on hRASMC [Ca2⫹]c Cultured hRASMCs slowly cooled to 22°C responded to rewarming to 37°C with an increase in basal [Ca2⫹]c from 113 ⫾ 9 nM at 22°C to 226 ⫾ 22 nM at 37°C (n ⫽ 9). This

Fig 4. Rewarming-induced calcium responses in radial artery smooth muscle cells (hRASMCs). Representative traces of single fura 2-loaded cells at 22oC, rewarmed to 37oC, using a temperaturecontrolled water bath, are shown in Figure A. Figure B shows the average number of cells (% total) exhibiting repetitive intracellular calcium spikes when pretreated with no drug (control; n ⫽ 9), 0.5 mM ethylene glycol bis-2-aminoethyl ether-N,N=,N⬙,n=-tetraacetic acid (EGTA; n ⫽ 4); 0.05 mg/mL glyceryl trinitrate (GTN; n ⫽ 5), or 4.5 ␮g/mL diltiazem (Dil; n ⫽ 4) for 5 minutes prior to rewarming. Significant differences when treatment groups were compared with control are shown by an asterisk.

[Ca2⫹]c rise was unaffected by any of the treatments used but was accompanied by spontaneous [Ca2⫹]c oscillations in 37 ⫾ 11% of the cells (Fig 4A). These rewarminginduced [Ca2⫹]c oscillations were abolished by EGTA and reduced by glyceryl trinitrate (Fig 4B).

Comment Radial arteries are often exposed to changes in temperature during their preparation prior to grafting. These results demonstrate that changes in temperature, similar to those experienced in theatre, have the potential to induce contraction. When arteries were exposed to an acute temperature drop from 37°C to 22°C this led to a brief contraction independent of any manipulation of the buffer or change in pH. In a proportion of the samples tested this led to a sustained elevation of the basal tension. Gradual cooling acted as a vasodilator; however, subsequent rewarming did not merely reestablish the basal tone but increased tension, on average twofold, above values obtained prior to cooling.

The [Ca2⫹]c responses in cultured smooth muscle cells paralleled the changes in tension observed in arterial rings and allowed further investigation of the mechanisms involved. A rapid temperature drop from 37°C to 22°C induced a transient rise in hRASMC [Ca2⫹]c similar in time frame and duration to the cooling-induced contraction observed in radial artery. Both responses were not blocked by EGTA indicating that calcium release from the sarcoplasmic reticulum could account entirely for cooling-induced contraction. The use of selective antagonists demonstrated that it was the inositol 1,4,5 trisphosphate receptors and not the ryanodine receptors that were involved in this response. Rewarming-induced contraction was completely abolished by EGTA and reduced by diltiazem, indicating that calcium influx through the L-type calcium channel was involved. Calcium channel blockade has also been shown to inhibit rewarming-induced contraction in saphenous vein [12]. Rewarming of hRASMCs slowly cooled to 22°C, back to 37°C, induced a rise in basal calcium which was accompanied, in a proportion of the cells studied, by spontaneous [Ca2⫹]c oscillations. In common with the rewarminginduced contraction in the arterial sections, these oscillations were inhibited by glyceryl trinitrate and abolished by EGTA, implicating calcium influx as a causal factor. Interestingly, diltiazem did not abolish rewarming-induced [Ca2⫹]c oscillations in hRASMCs, indicating that while the L-type calcium channel was involved in the response in the intact artery its contribution was reduced in hRASMCs. This would suggest that calcium channels other than the L-type channel may also be affected by rapid rewarming. To demonstrate how a reduction in temperature would affect both calcium release from intracellular stores and calcium influx through plasma membrane channels, we investigated how cooling would affect the response to noradrenaline, a G-protein coupled receptor agonist, and KCl, an activator of L-type calcium channels. The ␣1 adrenoceptor is the dominant adrenergic receptor in the human radial artery [24] and ␣1 adrenoceptor-mediated contraction in human saphenous veins and skin arteries at 22°C is much lower than that recorded at 37°C [25, 26]. Current through the cloned human L-type calcium channel is also reduced by cooling [27]. A reduced temperature slows the mysosin-ATPase responsible for mediating contraction [28], but in our studies cooling reduced both the vasoconstriction in arterial rings and the rise in [Ca2⫹]c in smooth muscle induced by KCl and noradrenaline. Therefore, a reduced temperature also lowers the magnitude of the calcium rise elicited by each of these stimuli. After rewarming, when the arterial rings were again contracted with noradrenaline, the response was restored to values obtained prior to cooling. We would conclude that cooling potentially suppresses contraction to a range of stimuli. However, upon rewarming full vasoconstrictor responses are restored. Because gradual cooling might mask stimulation occurring during graft storage there is a role for topically applied prophylactic strategies, such as phenoxybenzamine, which would prevent catecholamine-mediated contraction [7] and other

OO ET AL TEMPERATURE AND RADIAL ARTERY CONTRACTION

131

strategies, which would have a more general effect against a variety of vasoconstrictors [29]. Since vasoconstriction induced by either a rapid temperature drop and rapid rewarming were unaffected by phenoxybenzamine. Phenoxybenzamine-treated grafts would require additional vasodilator treatments. As glyceryl trinitrate blocked both rewarming and cooling-induced contraction in the radial artery, it represents the best available treatment to prevent contraction caused by rapid changes in temperature. Radial arteries have a larger medial cross-sectional area and greater density and tighter organization of smooth muscle cells when compared with internal thoracic arteries [30, 31], resulting in a higher agonistinduced force of contraction [17]. This may make radial arteries particularly prone to contraction induced by rapid changes in temperature. Studies that demonstrated cooling-induced relaxation in human conduit arteries and veins, including the internal thoracic artery, used a protocol which slowly reduced the temperature over a 20 to 60 minute period [9, 12]. Therefore, other arterial grafts may demonstrate contraction when exposed to rapid changes in temperature, but because the excised radial artery is stored prior to anastomosis these problems are particularly relevant to this graft. While gradual cooling may give some apparent benefit in theatre by relaxing the vessel, the subsequent rapid increase in temperature experienced by the graft as blood flow is restored is likely to lead to contraction and a reduction in blood supply to the newly grafted area. Placing the graft in a receiver containing warmed buffer would allow cooling to be gradual and rewarming the graft slowly, perhaps by placing it within the body cavity prior to grafting would also be beneficial. Because the radial artery is a thicker walled vessel than the internal thoracic artery it may be prone to hypoxia [32]. An alternative strategy to keep the radial artery attached in situ with the side branches occluded until immediately prior to the anastomosis would limit the level of hypoxia experienced by the graft and maintain endothelial integrity [33]. Rapid temperature changes during the surgical preparation of radial artery grafts in theatre should be avoided where possible. Gradual cooling and rewarming should be applied to the graft and glyceryl trinitrate used in addition to other strategies.

This project was funded by the Garfield-Weston Trust and the Cardiothoracic Centre, Liverpool. We are grateful to surgeons and theatre staff for the provision of samples of artery and to Dr Susan Coker for the use of organ chambers and Lectromed tension recorders.

References 1. Lytle BW, Loop FD, Cosgrove DM, Ratliff NB, Easley K, Taylor PC. Long term (5 to 12 years) serial studies of internal mammary artery and saphenous vein coronary bypass grafts. J Thorac Cardiovasc Surg 1985;89:248 –58. 2. Acar C, Ramsheyi A, Pagny JY, et al. The radial artery for coronary artery bypass grafting: clinical and angiographic

CARDIOVASCULAR

Ann Thorac Surg 2007;83:126 –33

CARDIOVASCULAR

132

3. 4. 5. 6. 7.

8. 9.

10. 11. 12.

13.

14.

15. 16. 17.

OO ET AL TEMPERATURE AND RADIAL ARTERY CONTRACTION

results at five years. J Thorac Cardiovasc Surg 1998;116: 981– 8. Possati G, Gaudino M, Prati F, et al. Long-term results of the radial artery used for myocardial revascularization. Circulation 2003;108:1350 – 4. Tatoulis J, Royse AG, Buxton BF, et al. The radial artery in coronary surgery: a 5-year experience— clinical and angiographic results. Ann Thorac Surg 2002;73:143–7. Acar C, Jebara VA, Portoghese M, et al. Revival of the radial artery for coronary artery bypass grafting. Ann Thorac Surg 1992;54:652– 60. Rosenfeldt FL, He GW, Buxton BF, Angus JA. Pharmacology of coronary artery bypass grafts. Ann Thorac Surg 1999;67: 878 – 88. Taggart DP, Dipp M, Mussa S, Nye PCG. Phenoxybenzamine prevents spasm in radial artery conduits for coronary artery bypass grafting. J Thorac Cardiovasc Surg 2000; 120:815–7. Chong WCF, Ong PJL, Hayward C, Moat N, Collins P. Effects of storage solutions on in vitro vasoreactivity of radial artery conduits. J Thorac Cardiovasc Surg 2001;122:470 –5. Bodelsson M, ArnekloNobin B, Chester AH, Tadjkarimi S, Tornebrandt K, Yacoub M. Differential effect of hypothermia on the vascular tone and reactivity of the human coronary artery and graft vessels. J Cardiovasc Surg 1991;32:288 –94. Herrera B, Eisenberg G, Holberndt O, et al. Paradoxical effects of temperature on vascular tone. Cryobiology 2000; 41:43–50. Mustafa SMD, Thulesius O. Cooling is a potent vasodilator of deep vessels in the rat. Can J Physiol Pharmacol 2001;79: 899 –904. Simonet S, Bonhomme E, Fabiani JN, Verbeuren T. Temperature-dependent basal tone in isolated human saphenous veins: implication of TP-receptors. Fund Clin Pharmacol 2000;14:461–7. Suzuki S, Sugita M, Ono S, Song C, Kondo T, Fujimura S. Difference in the effects of low temperatures on the tension of human pulmonary artery and vein ring segments. Respiration 2000;67:189 –93. Sagher O, Huang DL, Webb RC. Induction of hypercontractility in human cerebral arteries by rewarming following hypothermia: a possible role for tyrosine kinase. J Neurosurg 1997;87:431–5. Wagerle LC, Kim SJ, Russo P. Protein tyrosine kinase signalling in cold-stimulated contraction of newborn lamb cerebral arteries. Am J Physiol 1996;39:H645–50. Dahdah NS, Russo P, Wagerle LC. Phosphorylation in coronary artery cold-induced contraction in the newborn lamb. Cryobiology 2001;42:40 – 8. He GW, Yang CQ. Radial artery has higher receptormediated contractility but similar endothelial function compared with mammary artery. Ann Thorac Surg 1997;63: 1346 –52.

Ann Thorac Surg 2007;83:126 –33

18. Chanda J, Brichkov I, Canver CC. Prevention of radial artery graft vasospasm after coronary bypass. Ann Thorac Surg 2000;70:2070 – 4. 19. Bond BR, Zellner JL, Dorman BH, et al. Differential effects of calcium channel antagonists in the amelioration of radial artery vasospasm. Ann Thorac Surg 2000;69:1035– 40. 20. He GW, Yang CQ. Comparative study on calcium channel antagonists in the human radial artery: clinical implications. J Thorac Cardiovasc Surg 2000;119:94 –100. 21. Conant AR, Oo AY, Dashwood MR, et al. Endothelin receptors in cultured and native human radial artery smooth muscle. J Cardiovasc Pharmacol 2002;39:130 – 41. 22. Simpson AWM. Fluorescent measurement of [Ca2⫹]c: basic practical considerations. In: Lambert DG, ed. Calcium signalling protocols. Totowa, NJ: Human Press Inc; 1998:3–30. 23. Wilcox RA, Primrose WU, Nahorski SR, Challiss RAJ. New developments in the molecular pharmacology of the myoinositol 1,4,5-trisphosphate receptor. Trends Pharmacol Sci 1998;19:467–75. 24. He GW, Yang CQ. Characteristics of adrenoceptors in the human radial artery: clinical implications. J Thorac Cardiovasc Surg 1998;115:1136 – 41. 25. Gomez B, Borbujo J, Garciavillalon AL, et al. Alpha1adrenergic and alpha2-adrenergic response in human isolated skin arteries during cooling. Gen Pharmacol 1991;22: 341– 6. 26. Harker CT, Ousley PJ, Harris EJ, Edwards JM, Taylor LM, Porter JM. The effects of cooling on human saphenous-vein reactivity to adrenergic agonists. J Vasc Surg 1990;12:45–9. 27. Wang R, Karpinski E, Pang PKT. Temperature-dependence of L-type calcium-channel currents in isolated smooth-muscle cells from the rat tail artery. J Thermal Biol 1991;16:83–7. 28. Burdyga TV, Wray S. On the mechanisms whereby temperature affects excitation-contraction coupling in smooth muscle. J Gen Physiol 2002;119:93–104. 29. Conant AR, Shackcloth MS, Oo AY, Chester MR, Simpson AWM, Dihmis WC. Phenoxybenzamine treatment is insufficient to prevent spasm in the radial artery: the effect of other vasodilators. J Thorac Cardiovasc Surg 2003;126: 448 –54. 30. Acar C, Jebara VA, Portoghese M, et al. Comparative anatomy and histology of the radial artery and the internal thoracic artery. Implication for coronary-artery bypass. Surg Radiol Anat 1991;13:283– 8. 31. Van Son JAM, Smedts F, Vincent JG, Vanlier HJJ, Kubat K. Comparative anatomic studies of various arterial conduits for myocardial revascularization. J Thorac Cardiovasc Surg 1990;99:703–7. 32. Chester AH, Amrani M, Borland JAA. Vascular biology of the radial artery. Curr Opin Cardiol 1998;13:447–52. 33. Emir M, Gol MK, Ozisik K, et al. Harvesting techniques affect the integrity of the radial artery: an electron microscopic evaluation. Ann Thorac Surg 2004;78:1319 –25.

INVITED COMMENTARY The article by Oo and colleagues [1] provides important information regarding the behavior of the radial arteries as bypass conduits when submitted to temperature changes. Clearly, a sudden fall in the temperature, caused by immersing the radial arteries in a solution kept at room temperature after harvesting, induces significant contraction mediated by a rise in intracellular calcium released from sarcoplasmic stores. This sudden effect may contribute to spasm of the graft. In contrast, gradual cooling, by keeping the graft in a warm solution that is allowed to drift slowly to room temperature, acts as a vasodilator. Nevertheless, rewarming during blood re© 2007 by The Society of Thoracic Surgeons Published by Elsevier Inc

flow increases graft tension and may cause sustained contraction. Thus the important lesson from this article is that classical stimuli (ie, trauma, stretch, diathermy) involved in the pathogenesis of the radial artery vasospasm and temperature changes must be closely controlled. From this article clinical recommendations regarding radial arterial harvesting can be drawn. One key maneuver to prevent radial graft spasm is to keep it at a constant body temperature as long as possible by preventing exposure to the cool operating room environment. Endoscopic minimally invasive harvesting easily facilitates this, produces minimal arterial wall trauma, 0003-4975/07/$32.00 doi:10.1016/j.athoracsur.2006.09.097