The Impact of Routine and Intravascular Ultrasound-Guided High-Pressure Postdilatation After Drug-Eluting Stent Deployment: The STent OPtimization (STOP) Study

Abstract: Objectives. Drug-eluting stent (DES) implantations with low final cross-sectional area (CSA) are associated with adverse clinical outcomes. However, there is no guidance to facilitate optimal stent deployment (SD). The stent optimization (STOP) study was performed to assess DES routine postdilatation (PD) following implantation with intravascular ultrasound (IVUS) guidance. Methods. Forty-eight patients were included in this single-center prospective study. All DESs were deployed at 16 atm for 20 seconds and underwent routine non-compliant balloon PD (minimum 20 atm for 10 seconds). IVUS performed after SD (blinded) and PD (unblinded) measured CSA at 4 stent reference points. Optimal deployment was defined as distal and proximal stent CSA ≥60% distal and proximal reference CSA; mid and minimum stent CSA ≥70% of distal reference CSA. All per-protocol criteria were required to define optimal SD. Suboptimally deployed DESs underwent further PD with IVUS guidance (IVPD). Results. Fifty-two lesions were treated in 48 patients. CSA increased by 20% following PD. STOP criteria were only achieved in 21% of DESs after SD compared to 54% after PD. IVPD was performed in 20 DESs, which increased CSA by a further 21%. STOP criteria were eventually attained in 81% cases (P<.001 for all comparisons). Conclusion. DES deployment leads to suboptimal deployment, which can be optimized by routine PD. IVUS identifies DES implantations that benefit from further PD. Optimizing final DES-CSA may have long-term clinical benefits, although a randomized study is required. 

J INVASIVE CARDIOL 2014;26(12):640-646

Key words: stenting, DES, drug-eluting stent, postdilatation

___________________________________________________

The discovery of bare-metal stent (BMS) technology transformed the practice of percutaneous coronary intervention (PCI). With experience, the importance of optimal stent deployment (SD) was recognized to reduce rates of in-stent restenosis (ISR), stent thrombosis (ST), target lesion revascularization (TLR), and acute myocardial infarction (AMI).¹ Furthermore, the advent of intravascular ultrasound (IVUS) helped further optimize SD with improved acute and long-term clinical outcomes.² 

The introduction of drug-eluting stent (DES) implantation a decade later further reduced TLR compared to BMS, permitting an expansion of PCI indications and a new role in treating complex coronary disease.^3,4 However, despite these advancements, TLR rates are not negligible and remain multifactorial. In addition, there is a paucity of data to guide interventional cardiologists in achieving optimal SD. Perhaps in part due to an increased incidence of suboptimally expanded stents at implantation, the specter of late DES thrombosis appeared to be higher than for BMS devices and concerns were raised over DES use in “real-world” patients outside of randomized clinical trials.^5,6 For example, the emergence of late restenosis was adjudicated to be secondary to the treatment of more complex lesions.^7-10 Importantly, the stent design of the first-generation DES iterations may also have contributed to TLR rates. Indeed, a contemporary meta-analysis suggests that cobalt-chromium everolimus-eluting DES designs have a lower rate of ST when compared to either paclitaxel-eluting DES or BMS.¹¹ Improvements with everolimus-eluting DESs over first-generation stents are likely to be the result of a polymer more biocompatible rather than the stent platform of cobalt-chromium. However, despite improving stent technology with second- and third-generation DESs, neither ST nor TLR rates have been completely abolished.^12-14 

IVUS-guided PCI has failed to demonstrate superiority to angiography-guided PCI alone in randomized trials. However, there is a strong body of evidence to suggest that suboptimal DES deployment, defined by a reduced cross-sectional area (CSA) of the expanded stent, is associated with higher TLR and ST rates.^15-18 However, despite optimal DES deployment appearing important and intuitive, there remains a paucity of guidance as to how this should be achieved in practice and which technical aspects of DES deployment are important. In addition, contemporary registries have demonstrated that only around one-third of PCI procedures culminating in DES implantation are IVUS guided.¹⁹ Furthermore, there are no guidelines to recommend deployment pressures at which stents could be optimally implanted and whether postdilatation (PD) was to be practiced routinely.

Therefore, the aim of the Stent Optimization (STOP) study was to demonstrate the variability of CSA achieved following routine SD. In addition, we sought to demonstrate the usefulness of IVUS to optimize CSA with PD using a step-wise approach in non-calcified arteries. This study did not examine the role of IVUS in stent selection in accordance with the vessel size, as this has been investigated extensively. 

Methods

Patient selection. Forty-eight patients were included in this single-center, prospective, single-cohort study. The study had local ethical approval.

Inclusion criteria. All patients above the age of 18 years who could provide written informed consent undergoing elective or urgent PCI were eligible to participate if the coronary lesion could be treated with a single DES. Patients with more than one lesion (either in the same or another vessel) requiring non-overlapping stents could also be included. All patients were pretreated with 300 mg of aspirin and 600 mg of clopidogrel, in line with local policy. Consecutive patients who were felt suitable for the study were approached and consented to participate.

Exclusion criteria. Patients with heavily calcified vessels or a high thrombus burden were not included in the study. Furthermore, bifurcation lesions including left main stem stenoses or a coronary lesion mandating the use of IVUS prior to stent deployment were not included. Patients requiring primary PCI, within 24 hours of an acute myocardial infarction (MI), or those in cardiogenic shock were excluded. In addition, any patient with planned major surgery within 3 months of PCI, a history of bleeding diathesis, hypersensitivity to aspirin or clopidogrel, unexplained anemia, or a platelet count  of <100,000/µL was excluded from the study. 

Stent deployment. The access route, choice of anticoagulation, coronary guide catheter, and guidewire were left to the operator’s discretion. The lesion could be either directly stented or predilated as required. DES choice was determined by estimating the vessel diameter based on angiographic assessment after the administration of intracoronary glyceryl trinitrate (GTN; ≥200 µg). 

All stents were deployed at 16 atm with an inflation time of 20 seconds. Following SD and administration of a further ≥200 µg of intracoronary GTN, IVUS was performed using the Volcano Eagle Eye catheter (Volcano Therapeutics, Inc) from at least 5 mm distal to the distal end of the stent with an automated pullback (R100 device for all IVUS assessments in all patients), to at least 5 mm proximal to the proximal stent edge. The PCI operator was blinded to this IVUS run and was not permitted to visualize the stent appearance. 

IVUS. The initial IVUS run and all subsequent IVUS runs were evaluated by a trained IVUS operator working independently from the PCI operator and blinded to the angiographic data. 

CSA was measured at the following four sites within the stent (Figure 1): 

1. At the distal edge of the stent (dist CSA);
2. Mid-point of the stent (mid CSA);
3. The smallest CSA anywhere within the stent (min CSA);
4. At the proximal edge of the stent (prox CSA).

The presence of malapposition was also determined. In addition, the first IVUS run was used to determine the CSA of the reference vessel at the distal stent edge (dist ref CSA) and the proximal stent edge (prox ref CSA; Figure 2). These indices were required to assess for the STOP criteria that we considered to be essential in order to achieve an optimally deployed stent. Thus, the STOP primary endpoint was reached when all four of the following criteria were met: 

1. Dist stent CSA ≥60% dist ref CSA.
2. Mid stent CSA ≥70% dist ref CSA. 
3. Min stent CSA ≥70% dist ref CSA.
4. Prox stent CSA ≥60% prox ref CSA.

Postdilatation. Mandated by protocol, all stents had routine high-pressure PD with non-compliant balloons with a diameter at least equal to the deployed stent. Importantly, the operator was permitted to use multiple larger diameter balloons to optimize SD based on angiographic appearance. All balloons were inflated to a minimum pressure of 20 atm for at least 10 seconds.  

Following the administration of ≥200 µg of GTN, once the operator was satisfied that angiographically optimal SD had been achieved, a further IVUS was performed using the aforementioned standard IVUS protocol. The second IVUS run was unblinded and the images were made available to the operator. If the stent was adequately deployed according to the STOP criteria, the study was complete. 

Intravascular-ultrasound guided postdilatation (IVPD). If STOP criteria were not achieved, additional IVPD was recommended. This was performed with either larger diameter balloons and/or at higher inflation pressures. Furthermore, open IVUS use was allowed in this phase to assist optimization of SD. When the operator considered the final angiographic result optimal after a further 200 µg of intracoronary GTN, a final automated IVUS pullback was performed.

Follow-up. The secondary endpoints included 30-day and 12-month rates of MI, target vessel revascularization (TVR), TLR, ST, and cardiac death.

Statistical analysis. The sample size was estimated on the basis of being able to achieve significant difference in CSA with PD compared to CSA after SD. It was estimated that 50 lesions would provide 85% power to detect a significant increase in CSA with a significance level of .05. All data are shown as mean ± standard deviation or median (interquartile range) for normally and non-normally distributed variables, respectively. The CSA after PD was compared with SD-CSA using the non-parametric Wilcoxon matched-pairs test. The CSA achieved after IVPD was compared with PD-CSA using Mann-Whitney test, as not all patients required IVPD. The proportion of lesions after each treatment stage achieving STOP criteria were compared using a McNemar’s c² test.

Results

Fifty-two lesions in 48 patients were treated. The clinical and procedural characteristics are shown in Table 1. Biomatrix DES (Biosensors International) was used in 29 lesions (56%) and Promus Element DES (Boston Scientific) was used in 23 lesions (44%). Mean stent diameter and stent length per patient were 3.2 ± 0.4 mm and 23.7 ± 7.2 mm, respectively.  

Postdilatation. Optimal PD of the stent was achieved using one balloon in 19 stents (37%), two balloons in 25 stents (48%) and three balloons in 8 stents (15%). The largest diameter balloon used was equal to the stent in 5 cases (9%), 0.25 mm larger in 19 cases (37%), and ≥0.5 mm in 28 cases (54%). Comparing the CSA after routine high pressure PD to initial SD, we found statistically significant gain at all four sites: 7% dist, 19% mid, 20% min and 8% prox (P<.0001 for all comparisons as shown in Table 2). 

Intravascular-ultrasound guided postdilatation. IVUS evaluation after PD revealed 28 stents (54%) were adequately deployed using STOP criteria, while 24 stents (46%) were not. Out of these, 20 stents (38%) required further IVPD, which was undertaken using a larger balloon than originally used in 15 lesions (0.25 mm larger in 6 lesions and ≥0.5 mm in 9 lesions), and the same size in 5 lesions but at a higher pressure (6-10 atm higher). Following the comparison of the CSA after IVPD to routine PD, it was noted that there was a statistically significant gain at all four sites: 7% dist, 6% mid, 14% min, and 21% prox (Table 3). IVPD was not undertaken in 4 stents because of operator discretion. IVUS analysis confirmed well-deployed stents. 

STOP criteria. The percentage of lesions achieving STOP criteria at each site and in all four sites for each treatment stage is shown in Table 4. Optimal deployment was only observed in 11 stents (21%) following initial SD. PD significantly improved this to 28 stents (54%), while IVPD provided further significant improvement to 42 stents (81%). There was no evidence of any malapposed stents on IVUS for each treatment stage.

Secondary endpoints. No patients experienced MI, ST, TVR, TLR, or cardiac death within the first 30 days following PCI. The rates of MI and cardiac death within 12 months of the index procedure were 1 each (2.1%). In contrast, no patients had experienced ST, TLR, or TVR at 12 months. 

Discussion

This study has demonstrated that in the context of treating angiographic lesions, suboptimal stent deployment occurred in nearly 80% of cases following stent balloon inflation only. The site most commonly exhibiting the minimum predicted CSA occurred in the body of the stent adjacent to the highest density of plaque in comparison to the proximal stent. As the stents were sized in accordance with the distal reference lumen diameter, the distal end of the stents achieved STOP criteria more often in comparison to the proximal end. 

Our data suggest that reliance only on the stent balloon for SD even at relatively high pressure (16 atm) and long duration (20 seconds) resulted in stent underexpansion in the majority of cases. It is likely that initial inflation was unable to overcome resistance exerted through the lesion plaque and vessel wall. Postdilatation with high-pressure non-compliant balloons was therefore a necessary requirement and often more than one balloon was required with a larger diameter than the stent balloon.

In this study, PD required two or more balloons in 63% of cases and the maximum-sized balloon was greater in diameter than the deployed stent in 91% cases. However despite this and achieving a good angiographic result, just 54% of stents achieved the study criteria. This was not due to stent underexpansion, but rather due to the reference vessel being larger on IVUS than predicted from angiographic assessment. Under these circumstances, the stent can be expanded further with larger PD balloons or using higher PD pressures. Indeed, IVPD increased the number of stents achieving STOP criteria to 81%. 

IVUS is currently the most widely studied and utilized adjunctive coronary imaging tool available to assess many of the features that define optimal stent implantation, including appropriate sizing of the stent in relation to reference vessels, and detecting underexpansion, malapposition, geographical miss, and edge dissections.²⁰ Follow-up of first-generation DES implantation suggested that suboptimal stent deployment could translate into adverse clinical events. Underexpansion of deployed stents with a CSA <5-5.5 mm² was one of the two procedural factors associated with restenosis, while a smaller area of 4.2-4.65 mm² was associated with ST.^15-18 Furthermore, DESs with complete neointimal coverage have a larger minimum CSA compared to those with incomplete coverage.²¹  

More importantly, IVUS can also detect geographical miss, which can lead to an increased frequency of TLR and acute MI.²² Furthermore, late incomplete stent apposition is associated with a higher rate of acute MI and very late ST during long-term follow-up.²³ This is mainly acquired due to adverse vessel remodeling with time, but some contribution may come from acutely malapposed stents, which IVUS assessment would help identify.

Therefore, there is a good body of evidence to suggest that IVUS-guided DES implantation should improve clinical outcomes. However, there are no large randomized studies with hard clinical endpoints to demonstrate this. Furthermore, the advantages of IVUS assessment are several-fold. For example, IVUS helps the operator to assess the reference vessel diameter for stent selection, diagnose any edge dissections following stent deployment, and finally, to confirm optimal PD. A recent meta-analysis of 11 observational studies suggested that IVUS guidance reduces death, major adverse cardiovascular events, and ST compared to angiography-guided DES implantation.²⁴ Despite these convincing data, the routine use of PD following stent implantation may be as low as 18% in angiography-guided stent optimization in comparison to almost 44% in IVUS-guided stent optimization.^25,26 This confirms the low use of PD in routine practice, but also the PD in the IVUS groups was much lower than we would have anticipated necessary from our study findings. This could potentially disadvantage the clinical benefits of IVUS.²⁰ 

There have been only two randomized studies comparing IVUS-guided stent optimization to angiography-guided stent optimization in DES.^27,28 Individually, they were both underpowered, including 210 and 284 patients, respectively, and neither showed any advantage with IVUS guidance. However, in both studies, a high proportion of patients in the angiography-only group underwent high-pressure PD (100% and 68%). Based on our findings, we hypothesize that a significant number of these stents would be well apposed and it is possible that IVUS may not have offered any advantage over routine high-pressure PD.²⁹ 

Study limitations. We acknowledge that this is a relatively small, single-center study. However, we undertook this proof-of-concept study to demonstrate that PD is desirable to optimize stent deployment even with the current state-of-the-art DES iterations. Furthermore, only lesions requiring a single non-overlapping stent in non-heavily calcified vessels were considered. Therefore, our findings cannot be applied to bifurcation lesions or heavily calcified or chronically occluded vessels.  

The sizing of the initial stent and balloons to postdilate were chosen at the operator’s discretion, based on angiographic appearance alone. The use of IVUS to size the stent may have influenced the results by the selection of a larger diameter stent up front. However, the aim of this study was to assess stent deployment based on angiography-guided stent selection with the impact of routine non-compliant balloon PD. Subsequent further IVPD was only to be used when STOP criteria were not reached after a satisfactory angiographic result. IVUS-guided stent selection was not the focus of our study. The interventionists involved in this study were high-volume operators and were very familiar with IVUS. PCI operators with less IVUS experience might not be comfortable or find it difficult to optimally size their stents/balloons using IVUS guidance. 

STOP criteria were chosen to give a compromise between an achievable outcome and avoiding the risks of stent overexpansion. For example, in the recently published AVIO trial, the investigators aimed to achieve 80% and 70% of the nominal balloon area in the mid-stent segment and near the stent edges, respectively. However, this was only achievable in 48% of the treated lesions.²⁷ Furthermore, STOP criteria could not be applied to overlapping stents, as it would not take into account any varying vessel size over long segments, as other IVUS criteria have been suggested in such circumstances.¹⁹ Importantly, the authors believe that the proposed STOP criteria are valuable for short stents, but may not be applicable in evaluating long stents. Finally, IVUS may not be the best imaging modality for assessing malapposed stents, and alternative imaging modalities such as optical coherence tomography have far greater resolution.³⁰ 

Conclusion 

We have shown that standard stent balloon deployment will lead to a suboptimal result and that DES optimization can be maximized with high-pressure PD with a non-compliant balloon. Furthermore, IVUS use could be considered, as it facilitates stent PD and optimization. However, the clinical benefit of this strategy needs to be confirmed in an appropriately powered randomized clinical trial with hard clinical endpoints. 

References

1. Cutlip DE, Leon MB, Ho KK, et al. Acute and nine-month clinical outcomes after “suboptimal” coronary stenting: results from the Stent Anti-thrombotic Regimen Study (STARS) registry. J Am Coll Cardiol. 1999;34(3):698-706.
2. Parise H, Maehara A, Stone GW, Leon MB, Mintz GS. Meta-analysis of randomized studies comparing intravascular ultrasound versus angiographic guidance of percutaneous coronary intervention in pre-drug-eluting stent era. Am J Cardiol. 2011;107(3):374-382.
3. Stone GW, Ellis SG, Cox DA, et al; TAXUS IV Investigators. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med. 2004;350(3):2212-2231.
4. Moses JW, Leon MB, Popma JJ, et al. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med. 2003;349(14):1315-1323.
5. Uren NG, Schwarzacher SP, Metz JA, et al; POST Registry Investigators. Predictors and outcomes of stent thrombosis: an intravascular ultrasound registry. Eur Heart J. 2002;23(2):124-132.
6. Iakovou I, Schmidt T, Bonizzoni E, et al. Incidence, predictors, and outcome of thrombosis after successful implantation of drug-eluting stents. JAMA. 2005;293(17):2126-2130.
7. Morice MC, Colombo A, Meier B, et al. Sirolimus- vs paclitaxel-eluting stents in de novo coronary artery lesions: the REALITY trial: a randomized controlled trial. JAMA. 2006;295(8):895-904.
8. Zahn R, Hamm CW, Schneider S, et al; German Cypher Stent Registry. Incidence and predictors of target vessel revascularization and clinical event rates of the sirolimus eluting coronary stent (results from the prospective multicenter German Cypher Stent Registry). Am J Cardiol. 2005;95(11):1302-1308.
9. Mauri L, Silbaugh TS, Wolf RE, et al. Long-term clinical outcomes after drug-eluting and bare-metal stenting in Massachusetts. Circulation. 2008;118(18):1817-1827.
10. Palmerini T, Biondi-Zoccai G, Della Riva D, et al. Stent thrombosis with drug-eluting and bare-metal stents: evidence from a comprehensive network meta-analysis. Lancet. 2012;379(9824):1393-1402.
11. Stone GW, Rizvi A, Sudhir K, et al; SPIRIT IV Investigators. Randomized comparison of everolimus- and paclitaxel-eluting stents. 2-year follow-up from the SPIRIT (Clinical Evaluation of the XIENCE V Everolimus Eluting Coronary Stent System) IV trial. J Am Coll Cardiol. 2011;58(1):19-25.
12. Kedhi E, Joesoef KS, McFadden E, et al. Second-generation everolimus-eluting and paclitaxel-eluting stents in real-life practice (COMPARE): a randomized trial. Lancet. 2010;375(9710):201-209.
13. Stefanini GG, Byrne RA, Serruys PW, et al. Biodegradable polymer drug-eluting stents reduce the risk of stent thrombosis at 4 years in patients undergoing percutaneous coronary intervention: a pooled analysis of individual patient data from the ISAR-TEST 3, ISAR-TEST 4, and LEADERS randomized trials. Eur Heart J. 2012;33(10):1214-1222.
14. Takebayashi H, Kobayashi Y, Mintz GS, et al. Intravascular ultrasound assessment of lesions with target vessel failure after sirolimus-eluting stent implantation. Am J Cardiol. 2005;95(4):498-502.
15. Hong MK, Mintz GS, Lee CW, et al. Intravascular ultrasound predictors of angiographic restenosis after sirolimus-eluting stent implantation. Eur Heart J. 2006;27(11):1305-1310.
16. Fujii K, Carlier SG, Mintz GS, et al. Stent underexpansion and residual reference segment stenosis are related to stent thrombosis after sirolimus-eluting stent implantation: an intravascular ultrasound study. J Am Coll Cardiol. 2005;45(7):995-998.
17. Okabe T, Mintz GS, Buch AN, et al. Intravascular ultrasound parameters associated with stent thrombosis after drug-eluting stent deployment. Am J Cardiol. 2007;100(4):615-620.
18. Rogacka R, Latib A, Columbo A. IVUS-guided stent implantation to improve outcome: a promise waiting to be fulfilled. Curr Cardiol Rev. 2009;5(2):78-86.
19. Mintz GS. Features and parameters of drug-eluting stent deployment discoverable by intravascular ultrasound. Am J Cardiol. 2007;100(8B):26M-35M.
20. Witzenbichler B, Maehara A, Weisz G, et al. Relationship between intravascular ultrasound guidance and clinical outcomes after drug-eluting stents: The ADAPT-DES study. Circulation. 2014;129(4):463-470.
21. Sera F, Awata M, Uematsu M, Kotani J, Nanto S, Nagata S. Optimal stent-sizing with intravascular ultrasound contributes to complete neointimal coverage after sirolimus-eluting  stent implantation assessed by angioscopy. JACC Cardiovasc Interv. 2009;2(10):989-994.
22. Costa MA, Angiolillo DJ, Tannenbaum M, et al; STLLR Investigators. Impact of stent deployment procedural factors on long-term effectiveness and safety of sirolimus-eluting stents (final results of the multicenter prospective STLLR trial). Am J Cardiol. 2008;101(12):1704-1711.
23. Cook S, Eshtehardi P, Kalesan B, et al. Impact of incomplete stent apposition on long-term clinical outcome after drug-eluting stent implantation. Eur Heart J. 2012;33(11):1334-1343.
24. Zhang Y, Farooq V, Garcia-Garcia HM, et al. Comparison of intravascular ultrasound versus angiography-guided drug-eluting stent implantation: a meta-analysis of one randomised trial and ten observational studies involving 19,619 patients. EuroIntervention. 2012;8(7):855-865.
25. Roy P, Steinberg DH, Sushinsky SJ, et al. The potential clinical utility of intravascular ultrasound guidance in patients undergoing percutaneous coronary intervention with drug-eluting stents. Eur Heart J. 2008;29(15):1851-1857. Epub 2008 Jun 11.
26. Claessen BE, Mehran R, Mintz GS, et al. Impact of intravascular ultrasound imaging on early and late clinical outcomes following percutaneous coronary intervention with drug-eluting stents. JACC Cardiovasc Interv. 2011;4(9):974-981.
27. Jakabcin J, Spacek R, Bystron M, et al. Long-term health outcome and mortality evaluation after invasive coronary treatment using drug-eluting stents with or without the IVUS guidance. Randomized control trial. HOME DES IVUS. Catheter Cardiovasc Interv. 2010;75(4):578-583.
28. Chieffo A, Latib A, Caussin C, et al. A prospective, randomized trial of intravascular-ultrasound guided compared to angiography guided stent implantation in complex coronary lesions: the AVIO trial. Am Heart J. 2013;165(1):65-72. Epub 2012 Nov 20.
29. Chen HY, Hermiller J, Sinha AK, Sturek M, Zhu L, Kassab GS. Effects of stent  sizing on endothelial and vessel wall stress: potential mechanisms for in-stent restenosis. J Appl Physiol. 2009;106(5):1686-1691.
30. Kawamori H, Shite J, Shinke T, et al. The ability of optical coherence tomography to monitor percutaneous coronary intervention: detailed comparison with intravascular ultrasound. J Invasive Cardiol. 2010;22(11):541-545.

__________________________________________

From the Dorset Heart Centre, Royal Bournemouth Hospital, Bournemouth, Dorset, United Kingdom.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.

Manuscript submitted April 7, 2014, provisional acceptance given April 25, 2014, final version accepted June 25, 2014.

Address for correspondence: Dr Terry Levy, Consultant Interventional Cardiologist, Dorset Heart Centre, Royal Bournemouth Hospital, Castle Lane East, Bournemouth, BH7 7DW, United Kingdom. Email: terry.levy@rbch.nhs.uk