| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2004;26:988-994
© 2004 Elsevier Science NL
a Department of Cardiothoracic Surgery, John Radcliffe Hospital, Headley Way, Headington, Oxford OX3 9DU, UK
b Department of Cardiovascular Medicine, University of Oxford, Oxford, UK
c Nuffield Department of Surgery, University if Oxford, Oxford, UK
Received 9 December 2003; received in revised form 9 June 2004; accepted 16 June 2004.
* Corresponding author. Tel.: +44-1865-221-121; fax: +44-1865-220-244. (E-mail: david.taggart{at}orh.nhs.uk).
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
were measured at 0, 2, 6 and 12h post-incubation. In parallel, 36 murine infrarenal aortic interposition grafts (2mm) were performed. Twelve grafts were pre-treated (15min) with phenoxybenzamine, 12 with verapamil/nitroglycerin and 12 remained untreated (controls). Isometric tension responses to the same agonists were measured in grafts harvested 2, 6, 13 and 23h after surgery. Results: Phenoxybenzamine prevented -adrenergic vasoconstriction for up to 16h in vivo (grafts), and 12h in vitro (ungrafted segments). Verapamil/nitroglycerin was effective for at least 2h in vitro, but did not prevent vasoconstriction after 2h in vivo. Conclusions: The mouse model appears to be a useful technique for assessing the pharmacological properties of antispasmodic agents in vivo. Phenoxybenzamine has an extended action in arterial grafts in vivo. Verapamil/nitroglycerin is short-lived in vivo but lasts longer in vitro. Measurements of antispasmodic duration of action in vitro should be interpreted with caution. |
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
Verapamil and nitroglycerin solution (VG) [7], and phenoxybenzamine [8–10] have been advocated as topical RA antispasmodic agents and are in clinical use. A number of in vitro studies have evaluated the duration of action of these agents following a single topical application. The effects of phenoxybenzamine can persist for 18h in human RA in vitro [11] and 48h in canine RA in vitro [12], but this has not been confirmed in the in vivo setting. Studies of VG solution are inconsistent, suggesting a duration of action between 4h [10] and 24h [7]. However, a common and major limitation of all of these in vitro studies is the absence of in vivo factors that may have potentially major effects on duration of action of topically administered antispasmodic agents, including luminal flow and resultant endothelial shear, distending pressure, circulating vasoactive mediators and the autonomic nervous system. Therefore, in vitro studies may tend to overestimate the duration of action of these agents compared with the clinical setting. Measurement of the duration of action in vivo would provide additional insights to enable rational selection of an appropriate agent for clinical use.
Accordingly, we compared the duration of action of phenoxybenzamine and VG in vivo, using the mouse aortic interposition graft as a model system for a free arterial graft. Furthermore, to evaluate possible discrepancies between in vivo and in vitro duration of action, we compared the in vivo findings with the in vitro duration of action of the two agents in ungrafted segments of mouse aorta.
|
2. Materials and methods |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
To determine the duration of action of phenoxybenzamine and VG solution in preventing vasoconstriction in mouse aortas in vitro (n=3–6), we incubated three contiguous 2mm ungrafted aortic segments from each aorta with phenoxybenzamine (10µM), VG (30µM each) or buffer (controls), respectively, for 15min in the organ chambers. Cumulative dose responses to phenylephrine (1nM to 10µM, Sigma) and prostaglandin F2 (1nM to 30µM, Sigma) were measured serially at 0, 2, 6 and 12h after exposure to the antispasmodic agent for each segment. Numbers of washes between vasoconstrictors were standardised between experiments.
2.2. Animal model
All procedures were conducted in accordance with the Animals (Scientific Procedures) Act 1986 (Home Office, UK). Male C57/BL6 mice (9–18 weeks old) were used to perform abdominal aortic interposition grafting by modification of a previously described method [13]. Briefly, mice were anaesthetised with a subcuticular injection of 0.3ml Hypnoval® (500µg/ml midazolam, Antigen Pharmaceuticals) and 0.3ml Hypnorm® (10µg/ml fentanyl and 330µg/ml fluanisone, Janssen Animal Health). Donor aortas were harvested from age matched littermates. Following anaesthesia, the abdominal and thoracic cavities of the donor animal were opened widely. Systemic heparinisation (100U via inferior vena cava) was followed by exsanguination and immediate harvest of the descending thoracic aorta. A midline abdominal incision was made in the anaesthetised recipient animal, and the abdominal aorta exposed and mobilised by dissection of the periaortic fat. The aorta was transected between two vascular clips (World Precision Instruments, Stevenage, UK). Blood was rinsed from the cut ends with heparinised saline (500U/ml). A 2mm donor aortic segment was positioned by insertion of 11-0 monofilament nylon proximal and distal interrupted stay sutures (Ethilon, Johnson and Johnson Ltd, UK). The distal and proximal anastomoses were completed with 8–10 further interrupted sutures. The vascular clips were removed and haemostasis achieved. The viscera were replaced and the abdomen closed in two layers with a continuous 4-0 absorbable suture (4-0 Softgut Chromic, Sherwood, Davis, and Geck). Animals were recovered in a warming cabinet until fully mobile, and given diet and water ad libitum.
To evaluate the duration of action of phenoxybenzamine and VG solution after topical application to vascular grafts in vivo, we incubated donor aortic segments with topical antispasmodic agents (phenoxybenzamine 10µM, or verapamil/nitroglycerin 30µM each) or buffer alone (controls) for 15min, immediately prior to grafting. Twelve mice received grafts pre-treated with phenoxybenzamine, 12 received grafts pre-treated with VG solution and 12 received control grafts (n=3 animals per treatment group per time point).
The mice were sacrificed at approximately 2, 6, 13 and 23h after surgery, measured from removal of the vascular clamps and resumption of circulation. The grafts were harvested immediately by reopening the abdomen, exposing the graft, placing vascular clamps proximally and distally to the graft, and cutting through the suture lines, leaving the majority of the graft undamaged and suitable for analysis. The grafts were placed in ice-cold buffer and transferred to the laboratory for vasomotor studies to measure contractile responses to K+, phenylephrine and prostaglandin F2 (see above). Grafts from mice that died during the procedure or in the post-operative period did not undergo vasomotor studies and were excluded from the analysis.
2.3. Statistical analysis
Contractile responses, shown as mean±SEM, are expressed as a percentage of the response to modified KHB (60mM K+) in individual rings. Dose response curves were compared to controls using a general linear model for repeated measures (SPSS for Windows 11.0). Single factor ANOVA was used to compare receptor-independent contractile responses to 60mM KCl in grafts. Two tailed Students t-tests were used to compare maximal vasoconstrictor-induced responses at each time point to their respective controls. A probability value of less than 0.05 (P<0.05) was considered statistically significant.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
-adrenergic agonist phenylephrine persisted until 12h post-treatment in ungrafted aortic segments in vitro (Fig. 1). VG solution prevented contractile responses to phenylephrine at 0 and 2h post-treatment, but its effects were lost by 6h (Fig. 1). In control vessels, contractile responses to prostaglandin F2 were of greater magnitude than responses to phenylephrine (Fig. 3). However, neither phenoxybenzamine nor VG prevented vasoconstriction in response to prostaglandin F2 (Fig. 3b).
|
|
-adrenoceptor-mediated vasoconstriction in mouse aorta, but neither has substantial effects on prostanoid-mediated vasoconstriction. More importantly, these data indicate that phenoxybenzamine has a longer duration of action than VG in vitro. Notably, there was a trend for the contractile responses of control segments to diminish with time in response to serial administration of agonists (Figs. 1 and 3), but this failed to reach statistical significance. This suggests that the condition of mouse aortic segments may deteriorate with prolonged incubation in organ baths.
3.2. In vivo duration of action of antispasmodics
Next, we evaluated the duration of action of phenoxybenzamine and VG solution after topical application to vascular grafts in vivo. We performed 76 abdominal aortic interposition grafts in total, of which 47 survived to the planned time point for graft harvest (21 operative deaths, 8 animals sacrificed after graft thrombosis). Eleven grafts were not viable after harvest and were thus excluded from the analysis.
Absolute tension responses to K+ in previously grafted aortic segments were as follows: 2.4±0.04mN at 2h after surgery (mean time 2.1h, range 2–2.25h), 3.6±0.4mN at 6h (6.2, 5.8–7.0), 4.6±0.9mN at 13h (13.3, 11.1–13.8), and 3.1±0.5mN at 23h (23.4, 20.1–24.9). The contractile responses were not significantly different at any of the four time points after surgery (P=0.09, single factor ANOVA), suggesting that there was no change in contractile function of grafted segments over the time course of the study.
Phenoxybenzamine abolished contractile responses to phenylephrine up to 16h following surgery. However, by 24h the effect of phenoxybenzamine had diminished by approximately 50% (Fig. 2). In contrast, pre-treatment with VG solution failed to prevent contractile responses to phenylephrine within 2h after completion of surgery. Thus, the marked early effect of VG observed in the in vitro aortic segments was lost in functioning arterial grafts in vivo, even at the earliest time point (Fig. 2). Neither agent prevented prostanoid-mediated vasoconstriction at any time point after surgery (Fig. 3b). These data indicate that topically applied phenoxybenzamine remains effective in vivo for a prolonged period, at least 16h, whereas the effects of VG solution are more rapidly lost in vivo than they are during in vitro studies.
|
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
-adrenergic vasoconstriction for up to 16h in vivo. Verapamil/nitroglycerin solution also inhibits -adrenergic vasoconstriction for at least 2h in mouse aortas in vitro, but this effect is lost within 2h in functioning vascular grafts in vivo. Phenoxybenzamine has a longer duration of action than verapamil/nitroglycerin, both in vivo and in vitro. Finally, by comparison of parallel studies performed in vitro and in vivo, we conclude that in vitro studies may overestimate the true in vivo duration of action of antispasmodic agents, as exemplified by verapamil/nitroglycerin.
4.1. Interpretation
This study used the mouse aortic interposition graft as a model of a free arterial graft to specifically address the comparative duration of action of phenoxybenzamine and verapamil/nitroglycerin solution in vivo. Direct comparisons between the duration of action of each agent in vivo and in vitro enabled assessment of potential differences between the duration of action of antispasmodic agents in clinical versus experimental settings. Our findings provide important insights comparing the pharmacology and clinical utility of agents used to reduce RA graft spasm.
Phenoxybenzamine is a non-selective -adrenoceptor antagonist, with a long duration of action resulting from alkylation of the target receptors. It is a particularly useful antispasmodic agent in the clinical setting as the levels of endogenous catecholamines are raised 20-fold after CABG and remain elevated for at least a day after surgery [14]. Infusions of adrenaline or noradrenaline may be administered after CABG to ameliorate low cardiac output states and systemic vasodilatation, respectively.
The efficacy of phenoxybenzamine against -adrenergic vasoconstriction is becoming increasingly established. Our original observation that phenoxybenzamine prevents -adrenergic vasoconstriction in the RA [8] has been confirmed by recent studies [9–12]. Dipp [9] demonstrated that incubation with phenoxybenzamine caused less endothelial injury than papaverine to RA conduits. Most recently, we demonstrated the efficacy of phenoxybenzamine against -adrenergic stimuli, but not angiotensin II or prostaglandin F2 and a duration of action of at least 6h in an in vitro RA preparation [10].
A potential strength of phenoxybenzamine as an antispasmodic agent is its prolonged duration of action compared to other agents. Harrison and colleagues [11] reported that pre-incubation of RA rings with phenoxybenzamine for 20min inhibited contractile responses to noradrenaline for up to 18h. However, in that study, arterial rings remained suspended in organ chambers between vasoconstrictor challenges, and thus the factors that exist in functioning vascular grafts were not adequately modelled. Subsequently, Velez and colleagues [12] reported that pre-incubation of canine RA rings with phenoxybenzamine for 30min, followed by storage in culture medium for up to 48h, resulted in complete attenuation of the response to phenylephrine and noradrenaline compared to control rings. Similarly, that study was limited by the exclusive use of in vitro techniques, and did not accurately reflect the clinical situation.
The current study demonstrates that phenoxybenzamine, at a clinical concentration, prevents -adrenergic vasoconstriction in mouse aortas for at least 16h in vivo, and for at least 12h in vitro. Notably, the effect is specific to -adrenergic vasoconstriction, since phenoxybenzamine-treated grafts exhibited strong responses to prostaglandin F2. Mouse aorta behaved similarly to human RA in this respect [10]. Our observations in vivo are consistent with the known pharmacology of phenoxybenzamine, in particular, the requirement for -adrenoceptor turnover for return of responsiveness to adrenergic agonists [15]. The in vivo findings from this study both confirm and extend the current data, by demonstrating that phenoxybenzamine can act for prolonged periods in functioning vascular grafts.
VG solution, a mixture of verapamil (a highly selective voltage-dependent calcium channel antagonist with a plasma half-life of up to 4.8h [16]) and nitroglycerin (a nitric oxide donor with a short half-life) has been used clinically to prevent RA vasospasm since its effects were first described by He [7] in 1996. The same author demonstrated vasodilatation in response to VG solution applied to RA rings precontracted with potassium chloride and reported that the effect persisted for at least 24h in RA segments stored for 24h at 4°C. In the current study, we have demonstrated that VG solution is capable of preventing -adrenergic constriction in mouse aorta for at least 2h if the vascular segment remains in vitro, consistent with our previous findings in human RA [10]. However, a striking finding was the rapid and complete loss of VG-mediated inhibition of vasoconstriction within 2h in functioning vascular grafts.
The practical implication of these findings is that VG solution may be less useful clinically to prevent post-operative graft vasospasm than previous in vitro data have suggested. Although a substantial body of data clearly shows that VG has potent actions against a broad range of contractile agonists, its activity in a functioning vascular graft would remain limited to the intra-operative period only and may not extend into the immediate post-operative period, during which patients receiving RA grafts are at the greatest risk of graft vasospasm. In contrast, phenoxybenzamine remains effective for at least 16h following topical application to vascular grafts, and should inhibit -adrenergic vasoconstriction for a prolonged period in the post-operative patient, of particular importance for those patients requiring inotropic support after CABG. While phenoxybenzamine does have an extended duration of action, the ideal agent should provide protection against vasospasm for at least 24–48h after surgery.
4.2. Strengths of this study
A major strength of this work is the systematic comparison of the duration of action of two antispasmodic agents in current use, at clinically relevant doses, using both in vitro and in vivo models. Although classical organ bath methodology allows the study of pharmacological effects on smooth muscle contraction, the in vitro approach cannot model the influences of critical factors that affect vascular function in vivo, including blood flow and resultant endothelial shear, distending pressure, circulating vasoactive mediators and the autonomic nervous system [17]. There is also concern that prolonged incubation of vascular segments in organ baths is non-physiological, and indeed the responses of the in vitro control rings in this study deteriorated with time, whereas the responses of grafted (in vivo) control segments were maintained.
The murine aortic interposition graft model has yielded useful data in the field of transplant arteriosclerosis [18,19] and atherosclerotic lesion regression [20], but this is the first report of this model being used to study the pharmacological properties of topical agents used to reduce RA graft vasospasm. The experimental design mimics operating room practice of RA use, by exposing the conduit to the antispasmodic agent for 15min, constructing proximal and distal anastomoses and then allowing blood to flow freely through the grafted conduit.
4.3. Limitations of this study
A limitation of our work was the use of a murine aortic interposition graft to model a RA coronary bypass graft. Larger animals such as the pig, sheep or dog may have provided more clinically relevant models with RA conduits of similar anatomical characteristics to the human RA. However, it does not necessarily follow that mouse aorta is not a valid model, especially as the pharmacological data achieved by using mouse aorta in vitro are strikingly similar to the behaviour of human RA segments incubated with the antispasmodic agents phenoxybenzamine and VG solution [10].
An additional limitation of this study was the use of organ chambers to evaluate vasomotor responses in both grafted and ungrafted aortic segments. Therefore, the study does not represent an entirely in vivo assessment of duration of action but does aim to address some of the shortcomings associated with previous approaches of exclusively in vitro methods for assessing the more dynamic situation that occurs in the post-operative patient.
4.4. Conclusion
Phenoxybenzamine has a longer duration of action than verapamil/nitroglycerin both in vivo and in vitro. Verapamil/nitroglycerin has a limited duration of action in vivo, despite previous in vitro data suggesting an extended duration of action. Thus, the exclusive use of in vitro techniques to study topical agents used to reduce vasospasm in RA conduits may overestimate the possible duration of action of these agents in the clinical setting. The use of an animal model of a free vascular graft enables useful comparisons between agents in a dynamic situation, more reflective of the clinical setting.
|
Acknowledgments |
|---|
The authors would like to thank Janssen Cilag for their kind gift of 11-0 Ethilon sutures and the staff of the Biomedical Services Unit at the John Radcliffe Hospital site for their expert care of the animals used in this study.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstract1. Introduction2. Materials and methods3. Results4. DiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Kulik, F. D. Rubens, D. Gunning, M. E. Bourke, T. G. Mesana, and M. Ruel Radial Artery Graft Treatment With Phenoxybenzamine is Clinically Safe and May Reduce Perioperative Myocardial Injury Ann. Thorac. Surg., February 1, 2007; 83(2): 502 - 509. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
) and verapamil/nitroglycerin solution (VG,
) were used to measure cumulative dose responses to phenylephrine at 0, 2, 6 and 12h post-incubation. Untreated rings were controls (
). Data points represent mean±SEM contractile responses expressed relative to the response to 60mM K+ in individual segments prior to incubation with antispasmodic agents (n=3–6 aortic segments at each time point). *P<0.05 compared to controls.
