Serial Intravascular Ultrasound Analysis of Stent Strut Distribution and Fracture: An Integrated Analysis of the Taxus IV, V, and VI Trials

Abstract: Aims. Non-uniform distribution of drug-eluting stent struts may cause uneven drug deposition associated with an adverse neointimal response and clinical events. This study assesses circumferential stent strut distribution in bare-metal (BMS) and paclitaxel-eluting (Taxus) stents post implantation and at 9-month follow-up, as well as its impact on intimal hyperplasia (IH). Methods and Results. In the current analysis, intravascular ultrasound (IVUS) substudies of the Taxus IV, V, and VI trials were combined. Among them, 242 stents (117 BMS and 125 Taxus) had paired IVUS images post procedure and at 9-month follow-up that were reassessed at 1 mm intervals. Post implantation, the maximum interstrut angle (71.5 ± 17.7° vs 70.0 ± 19.6°; P=.53) and minimum number of stent struts (7.1 ± 1.0 vs 7.2 ± 0.8; P=.32) were similar in Taxus vs BMS subgroups, respectively. At 9-month follow-up, the maximum angle increased (92.8 ± 22.1° and 81.7 ± 20.6°) and stent strut numbers decreased (6.1 ± 0.9 and 6.5 ± 1.0) for both Taxus and BMS, respectively, as compared to immediately post implantation (all P<.001). The increased stent angle was more pronounced for Taxus compared with BMS (P<.01). Non-uniform strut distribution did not affect IH pattern or clinical outcomes in either stent population. No complete stent fractures were identified. Conclusion. Stent strut distribution changed from implantation to follow-up with an increased interstrut angle and fewer visible stent struts. These changes were more pronounced for Taxus as compared to BMS; however, non-uniform strut distribution was unrelated to increased IH or clinical outcomes.

J INVASIVE CARDIOL 2014;26(10):505-511

Key words: bare-metal stent, drug-eluting stent, intravascular ultrasound, stent fracture

_________________________________________

Drug-eluting stents (DESs) have significantly improved the long-term efficacy of percutaneous coronary interventions due to a marked reduction in binary in-stent restenosis as well as the need for repeat target lesion revascularization.^1-7 However, well-defined mechanical factors, such as stent underexpansion, increase the incidence of DES restenosis and thrombosis.^6-10 Recently, stent fracture has been identified as another potential risk factor for DES failure, most often after sirolimus-eluting stent (SES) implantation.^11-23 A form of partial stent fracture, non-uniform SES strut distribution, has been suggested as another cause of increased neointimal proliferation²⁴ due to non-homogeneous drug distribution. To understand the normal distribution of stent struts following Taxus stent implantation, whether this distribution changes during follow-up suggesting chronic stent fracture, and its relationship to outcomes, the Taxus IV, V, and VI intravascular ultrasound (IVUS) subgroups were combined, and the circumferential distribution of bare-metal stent (BMS) and Taxus stent struts post implantation and at 9-month follow-up and its impact on intimal hyperplasia (IH) and clinical outcomes were assessed.

Methods

Study protocol and population. Taxus IV, V, and VI were prospective, multicenter, double-blind trials in which patients with a single de novo native coronary artery lesion were randomly assigned to treatment with a paclitaxel-eluting Taxus Express2 stent or an otherwise identical BMS (Boston Scientific Corporation). The Taxus IV and V studies used the Taxus Express slow-release formulation (commercially available), whereas Taxus VI used the Taxus Express moderate-release formulation (not commercially available). In all three trials, prespecified sites enrolled patients into the IVUS cohort. Only those patients with paired volumetric images (n=547) were eligible for this detailed retrospective analysis. In the current study, the final cohort included 117 BMS-treated patients and 125 Taxus-treated patients with high-quality paired IVUS studies suitable for detailed strut-by-strut analysis; patients in the IVUS substudy were excluded from the current analysis if (contrary to the protocol), IVUS imaging was not performed, the pullback was not consistent throughout the length of the stent, or the imaging quality was not adequate. 

Each of the three individual studies was reviewed and approved by the respective institutional ethics committees, and enrolled patients signed an informed consent form. All three Taxus trials shared the same definitions for major adverse cardiac event (MACE) including stent thrombosis. Events were adjudicated by the same independent Clinical Events Committee. The IVUS substudy data from these three trials were analyzed at a single core laboratory (MedStar Health Research Institute, Washington Hospital Center, Washington, DC). Clinical follow-up was performed at 9 months and annually up to 5 years for Taxus IV and VI and up to 3 years for Taxus V. 

IVUS protocol and analysis. IVUS imaging was performed immediately post stent implantation and at 9-month follow-up. Detailed description of the IVUS protocol has been previously published.²⁵ Computerized planimetry (Tapemeasure; Indec, Inc) was used to manually trace lumen, stent, and external elastic membrane (EEM) cross-sectional areas (CSAs) every millimeter within the stented segment and to calculate IH CSA (stent minus lumen) and %IH (IH CSA divided by stent CSA). IVUS restenosis was defined as IH >50%. Analyses were done by individuals who were unaware of the clinical or angiographic findings and who were blinded to the treatment arms. For the specific purpose of the current study, the total number of stent struts was counted each millimeter; the largest interstrut angle was measured at 1-mm intervals along the entire stent length (Figure 1). The interstrut angle was measured from the center of mass of the stent struts, independent of the IVUS probe position. For each stent, the cross-sections with smallest number of visible struts and the maximum interstrut angle were identified. In addition, the mean interstrut angle was also calculated as the average of all of the largest interstrut angles throughout the length of the stent. Non-uniform stent strut distribution was defined as a stent with at least one cross-section with the maximum interstrut angle >90º. Stent fracture was defined as absence of all stent struts in at least one cross-section.

Quantitative coronary angiographic analysis. The angiographic studies were analyzed at a single core laboratory (Brigham and Women’s Hospital, Boston, Massachusetts). Two or more angiographic projections of the stenosis after intracoronary nitroglycerin were acquired, with repetition of identical angiographic projections at the time of follow-up angiography. With the contrast-filled injection catheter as the calibration source, quantitative coronary angiographic analysis was performed using a validated automated edge-detection algorithm (Medis; CMS) by a technician who was unaware of the clinical or IVUS findings and who was blinded to the treatment arm. Binary angiographic in-stent restenosis was defined as a diameter stenosis ≥50% at qualifying angiographic follow-up. Late lumen loss was defined as post-stent minimum lumen diameter minus follow-up minimum lumen diameter.

Statistical analysis. Continuous variables were presented as mean ± standard deviation and compared using two-tailed paired and unpaired Student’s t-tests. Categorical variables were summarized as frequencies and percentages and compared using chi-square statistics or paired McNemar test. The Breslow-Day and Levene’s tests were used to test the homogeneity for odds ratio (OR) and variance across trials. Multivariable analysis was performed to determine predictors of non-uniform stent distribution. All covariates were modeled univariately for each outcome and multivariately using a stepwise procedure in the logistic regression model. Except for the minimum number of struts, maximum interstrut angle, and mean interstrut angle, only baseline and procedural characteristics were included. Statistical significance was set at P<.05. For each outcome, predictors are listed in ascending order of P-values. Coefficients with P-values >.05 were not listed. Kaplan-Meier product-limit method and log-rank test were used to assess time-to-event endpoints between analyzed groups. Differences were considered to be statistically significant when the P-value was <.05. 

Results

Patient, lesion, and procedural characteristics. As shown in Table 1, there were no baseline differences in demographics, angiographic lesion characteristics, or procedural details between the Taxus and BMS subgroups. 

IVUS findings. IVUS measurements post procedure and at 9-month follow-up are shown in Table 2. Immediately after stent implantation, both the minimum number of IVUS-identifiable struts and the maximum and mean interstrut angles were comparable between Taxus and BMS. At 9-month follow-up, lumen CSA was significantly larger and IH was significantly smaller in Taxus-treated patients compared with BMS-treated patients. The minimum number of stent struts in the Taxus group decreased by 1.0 ± 1.1, from 7.1 ± 1.0 post procedure to 6.1 ± 0.9 at follow-up (P<.001).  The maximum interstrut angle increased in the Taxus group by 21.4 ± 24.9º, from 73.7 ± 17.9º to 95.1 ± 23.6º (P<.001) (Figure 2). The BMS group also demonstrated a similar change: the minimum number of struts decreased by only 0.7 ± 1.2, from 7.2 ± 0.8 post procedure to 6.5 ± 1.0 (P<.001) and the maximum interstrut angle increased by 12.7 ± 26.0°, from 71.8 ± 19.6º to 84.5 ± 20.8° (P<.001) (Figure 2). However, while the decrease in the number of struts was similar for Taxus and BMS (P=.12), the increase in the maximum and mean interstrut angles in BMS was significantly less than for Taxus (P<.01 and P=.03, respectively). 

Stent fracture. There were no Taxus or BMS complete stent fractures (defined as absence of all stent struts in at least one cross-sectional image) as assessed by IVUS.  

Incidence of non-uniform stent distribution. Post procedure, non-uniform stent strut distribution was observed in 24 Taxus patients (19.2%) and 21 BMS patients (17.9%). This increased at 9-month follow-up to 61 patients (49%) for Taxus (P<.001) and 38 patients (33%) for BMS (P<.01). The increase in the number of stents with non-uniform stent strut distribution was significantly more pronounced in Taxus as compared to BMS (29.6% vs 14.5%; P<.01). The homogeneity test shows that the non-uniform stent strut distribution (P=.41) and maximum interstrut angle (P=.56) were similar across the three trials.

Predictors of non-uniform stent strut distribution. Potential predictors of non-uniform stent strut distribution are shown in Table 3. Using multivariate analysis, independent predictors of non-uniform stent strut distribution were patients with Taxus stents (OR, 2.11; 95% confidence interval [CI], 1.22-3.66; P<.01), lesion calcification (OR, 2.02; 95% CI, 1.09-3.75; P=.03), and lesion length (OR, 1.04; 95% CI, 1.00-1.08; P=.05). Independent predictors of non-uniform stent strut distribution in BMS were dilation pressure (OR, 1.35; 95% CI, 1.14-1.61; P<.001), stent length (OR, 1.06; 95% CI, 1.02-1.11; P<.01), baseline minimum lumen diameter (OR, 6.74; 95% CI, 1.54-29.48; P=.01), vessel tortuosity (OR, 6.63; 95% CI, 1.23-35.78; P<.03), and prior myocardial infarction (OR, 0.34; 95% CI, 0.12-0.98; P<.05). Non-uniform stent strut distribution in Taxus increased in patients with prior percutaneous coronary intervention (OR, 4.43; 95% CI, 1.85-10.65; P<.001) and larger vessels (OR, 3.42; 95% CI, 1.27-9.25; P=.02). 

Relation of non-uniform stent distribution to restenosis and clinical outcomes. Stents with non-uniform stent strut distribution had a similar degree of 9-month IH as compared to stents with uniform strut distribution (Table 4) and similar angiographic binary restenosis rates (8.3% vs 16.7% [P=.17] for Taxus; 43.2% vs 29.3% [P=.14] for BMS) and in-stent late lumen loss (0.43 ± 0.59% vs 0.48 ± 0.64% [P=.65] for Taxus; 1.03 ± 0.59% vs 0.92 ± 0.51% [P=.29] for BMS). Multivariate analysis did not show any relationship between non-uniform stent strut distribution and in-stent restenosis in the overall population or in any of the studied subgroups (P>.05 for all). Similarly, there was no impact of non-uniform stent strut distribution on MACE.  

Discussion

The present study: (1) describes the distribution of Taxus and equivalent BMS struts immediately post stent implantation; (2) shows that strut distribution changes during the 9-month follow-up period, with a shift toward larger interstrut angles, consistent with partial strut fracture; (3) demonstrates that this change is greater for Taxus as compared to BMS; and (4) shows that non-uniform stent strut distribution (defined as a maximum interstrut angle >90°) at 9 months was not linked to increased IH with either Taxus or BMS use. 

Few previous reports have investigated stent strut distribution and its relation to IH. Takebayashi et al showed that restenotic SES had fewer stent struts and larger maximum interstrut angles at the minimum lumen sites as compared to: (1) remote sites within the same stents; and (2) minimal lumen sites of stents without restenosis.²⁴ Similar results were reported by Sano et al, who found that well-expanded but restenotic SESs had significantly less uniform strut distribution at the minimum lumen site as compared to restenotic SESs with stent underexpansion.²⁶ However, different SES and Taxus stent designs, different polymers, and different drug and drug-release kinetics precluded extending these observations to Taxus stents. In a phantom model, Suzuki et al showed that Taxus stents presented significantly less uniform stent strut distribution as compared to SESs.²⁷ Similar results were reported by Hasegawa et al, who used optical coherence tomography to determine strut distribution in human coronaries and phantom models.²⁸ Finally, the incidence of stent fractures has been reported to be greater with SES use as compared to Taxus use.^11-23 Nevertheless, the current study, in which IVUS was used in a prospective manner both at baseline and at follow-up, showed that non-uniform stent strut distribution was not associated with increased IH formation in Taxus stents or equivalent BMS. The observed trend toward greater IH for Taxus stents with uniform strut distribution as compared to Taxus stents with non-uniform stent strut distribution may be attributed to greater baseline lumen, stent, and EEM areas in the latter subgroup. Multivariate analysis did not show any relationship between non-uniform stent strut distribution and in-stent restenosis in the overall population or in any of the studied subgroups (P>.05 for all). Similarly, there was no impact of non-uniform stent strut distribution on MACE.  

It is important to note that the number of stents with non-uniform stent strut distribution increased significantly from post procedure to follow-up, and that the increase in the interstrut angle and the decrease in the number of struts were significantly more pronounced for Taxus as compared to BMS. This may indicate partial stent strut fracture; however, it did not appear to impact IH accumulation. Interestingly, the predictors of non-uniform stent strut distribution in our study (lesion calcifications, stent/lesion length, dilatation pressure, and vessel tortuosity) were similar to predictors of stent fractures in other studies.^11-23 However, contrary to reports on stent fractures, a greater incidence of non-uniform stent strut distribution in curved vessel segments, the right coronary artery, or at overlapping stents regions was not present. 

Stents deployed into human coronary arteries are subject to various mechanical strains. Some of those are applied during stent implantation and result from stent delivery and positioning, balloon inflation pressure, and stent-vessel wall conformability. Longitudinal stent deformation — either compression or extension — is a recently reported complication of the newer (second- and third-generation) stent platforms.²⁹ The current analysis included only first-generation DES use. However, longitudinal stent deformation may present as non-uniform stent strut distributions. Other mechanical strains are more evident after stent implantation and are the result of vessel tortuosity, angulation, calcification, continuous torsion of the arteries, and overlapping stent segments. One possible explanation for the difference between Taxus and Express BMS is that the increased IH seen in BMS helps to stabilize stent strut position and integrity; studies of strut fracture comparing DES and equivalent BMS designs have all shown more strut fractures in the DES vs the otherwise identical BMS. 

No complete stent fractures were identified in the current cohort of patients. This corroborates a study by Popma et al in which only 13 fractures were found in more than 2500 Taxus trial patients.²³ All of the patients from the current analysis were included in this previous study. 

Study limitations. The current report was a retrospective analysis of prospectively-designed clinical trials. The results were limited to selected lesion and patient characteristics permitted by the study protocols. There were technical factors limiting the ability to include all patients in the IVUS substudy of Taxus. There were also technical limitations of the use of IVUS to calculate strut distribution. The changing angle between the IVUS probe and the center of the coronary lumen, as well as longitudinal and rotational movement of the IVUS catheter during pullback, may have influenced stent strut visibility. Frames with excessive amounts of calcium and non-uniform rotational distortion (NURD) were excluded from the analysis both post implantation and at follow-up. However, in some locations, isolated calcium may have mimicked a stent strut. Finally, although it has been implied that an increase in interstrut angles, especially when coupled with a decrease in stent strut number, may reflect partial strut fractures, this concept requires further validation. 

Conclusion

Stent struts are subjected to complex mechanical forces at placement and thereafter, which may lead to an alteration in the stent strut distribution over the first 9 months. This study shows that non-uniform stent strut distribution is a relatively common finding in Taxus BMS Express designs, but is unrelated to the development of IH or adverse clinical events.  

References

1. Stone GW, Ellis SG, Cox DA, et al. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. N Engl J Med. 2004;350(3):221-231.
2. Dawkins KD, Grube E, Guagliumi G, et al., on behalf of the TAXUS VI Investigators. Clinical efficacy of polymer-based paclitaxel-eluting stents in the treatment of complex, long coronary artery lesions from a multicenter, randomized trial: support for the use of drug-eluting stents in contemporary clinical practice. Circulation. 2005;112(21):3306-3313.
3. Stone GW, Ellis SG, Cannon L, et al; for the TAXUS V Investigators. Comparison of a polymer-based paclitaxel-eluting stent with a bare metal stent in patients with complex coronary artery disease: a randomized controlled trial. JAMA. 2005;294(10):1215-1223.
4. Colombo A, Drzewiecki J, Banning A, et al; for the TAXUS II Study Group. Randomized study to assess the effectiveness of slow- and moderate-release polymer-based paclitaxel-eluting stents for coronary artery lesions. Circulation. 2003;108(7):788-794.
5. Morice MC, Serruys PW, Sousa JE, et al; for the RAVEL Study Group. Randomized study with the sirolimus-coated Bx velocity balloon-expandable stent in the treatment of patients with de novo native coronary artery lesions. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med. 2002;346(23):1773-1780.
6. Moses JW, Leon MB, Popma JJ, et al; for the SIRIUS Investigators. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. N Engl J Med. 2003;349(14):1315-1323.
7. Mintz GS, Weissman NJ.  Intravascular ultrasound in the drug-eluting stent era. J Am Coll Cardiol. 2006;48(3):421-429. Epub 2006 Jul 12.
8. 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.
9. Sakurai R, Ako J, Morino Y, et al. Predictors of edge stenosis following sirolimus-eluting stent deployment: a quantitative intravascular ultrasound analysis from the SIRIUS trial. Am J Cardiol. 2005;96(9):1251-1253.
10. Sonoda S, Morino Y, Ako J, et al. Impact of final stent dimensions on long-term results following sirolimus-eluting stent implantation: serial intravascular ultrasound analysis from the SIRIUS trial. J Am Coll Cardiol. 2004;43(11):1959-1963.
11. Chung WS, Park CS, Seung KB, et al.  The incidence and clinical impact of stent strut fractures developed after drug-eluting stent implantation. Int J Cardiol. 2008;125(3):325-331.
12. Kim JS, Lee SY, Lee JM, et al. Significant association of coronary stent fracture with in-stent restenosis in sirolimus-eluting stents. Coron Artery Dis. 2009;20(1):59-63.
13. Shaikh F, Maddikunta R, Djelmami-Hani M, et al. Stent fracture, an incidental finding or a significant marker of clinical in-stent restenosis? Catheter Cardiovasc Interv. 2008;71(5):614-618.
14. Lee SH, Park JS, Shin DG, et al. Frequency of stent fracture as a cause of coronary restenosis after sirolimus-eluting stent implantation. Am J Cardiol. 2007;100(4):627-630. 
15. Yamada KP, Koizumi T, Yamaguchi H, et al. Serial angiographic and intravascular ultrasound analysis of late stent strut fracture of sirolimus-eluting stents in native coronary arteries. Int J Cardiol. 2008;130(2):255-259.
16. Umeda H, Gochi T, Iwase M, et al. Frequency, predictors and outcome of stent fracture after sirolimus-eluting stent implantation. Int J Cardiol. 2009;133(3):321-326.
17. Lee MS, Jurewitz D, Aragon J, et. al. Stent fracture associated with drug-eluting stents: clinical characteristics and implications. Catheter Cardiovasc Interv. 2007;69(3):387-394.
18. Aoki J, Nakazawa G, Tanabe K, et al. Incidence and clinical impact of coronary stent fracture after sirolimus-eluting stent implantation. Catheter Cardiovasc Interv. 2007;69(3):380-386.
19. Bilen E, Saatci Yasar A, Bilge M, et al.  Acute coronary syndrome due to complete bare metal stent fracture in the right coronary artery. Int J Cardiol. 2010;139(3):e44-e46.
20. Leong DP, Dundon BK, Puri R, Yeend RA. Very late stent fracture associated with a sirolimus-eluting stent. Heart Lung Circ. 2008;17(5):426-428.
21. Park JS, Shin DG, Kim YJ, et al. Acute myocardial infarction as a consequence of stent fracture and plaque rupture after sirolimus-eluting stent implantation. Int J Cardiol. 2009;134(2):e79-e81.
22. Kang WY, Kim W, Hwang SH, Kim W. Dark side of drug-eluting stent: multiple stent fractures and sudden death. Int J Cardiol. 2009;132(3):e125-e127.
23. Popma JJ, Almonacid A, Mandinov L, et al. Stent fracture and clinical outcomes in the TAXUS and ATLAS trials (Abstr). J Am Coll Cardiol. 2008;51(Suppl)B1-B16:2901-2906.
24. Takebayashi H, Mintz GS, Carlier SG, et al. Nonuniform strut distribution correlates with more neointimal hyperplasia after sirolimus-eluting stent implantation. Circulation. 2004;110(22):3430-3434.
25. Weissman NJ, Ellis SG, Grube E, et al. Effect of the polymer-based, paclitaxel-eluting TAXUS Express stent on vascular tissue responses: a volumetric intravascular ultrasound integrated analysis from the TAXUS IV, V, and VI trials.  Eur Heart J. 2007;28(13):1574-1582.
26. Sano K, Mintz GS, Carlier SG, et al. Volumetric intravascular ultrasound assessment of neointimal hyperplasia and nonuniform stent strut distribution in sirolimus-eluting stent restenosis. Am J Cardiol. 2006;98(12):1559-1562.
27. Suzuki Y, Ikeno F, Yeung AC. Drug-eluting stent strut distribution: a comparison between Cypher and Taxus by optical coherence tomography. J Invasive Cardiol. 2006;18(3):111-114.
28. Hasegawa T, Ako J, Ikeno F, et al.  Comparison of non-uniform strut distribution between two drug-eluting stent platforms. J Invasive Cardiol. 2007;19(6):244-246.
29. Finet G, Rioufol G. Coronary stent longitudinal deformation by compression: is this a new global stent failure, a specific failure of a particular stent design or simply an angiographic detection of an exceptional PCI complication? EuroIntervention. 2012;20;8(2):177-181. 

______________________________________

From ¹Coronary and Structural Heart Diseases Department, Institute of Cardiology, Warsaw, Poland; ²MedStar Health Research Institute, Washington Hospital Center, Washington, DC; ³Cardiovascular Research Foundation, New York, New York; ⁴General Cardiology and Interventional Angiology Department, Institute of Cardiology, Warsaw, Poland; ⁵Department of Cardiology, Brigham and Women’s Hospital, Boston, Massachusetts; ⁶Boston Scientific Corporation, Natick, Massachusetts; ⁷Department of Cardiology, Cleveland Clinic, Cleveland, Ohio; ⁸International Heart Center Essen, Elisabeth Hospital, Essen, Germany; ⁹Department of Cardiology, Columbia University Medical Center, New York, New York.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Mintz reports research grants, consulting fees/honoraria from Boston Scientific. Dr Popma reports research grants and scientific advisory board consultancy for Boston Scientific. Dr Mandinov and Mr Wang are Boston Scientific employees. Dr Ellis reports consulting fees and honoraria from Boston Scientific. Dr Grube is on the Boston Scientific advisory board and a member of the speaker’s bureau. Dr Dawkins is an employee and stockholder of Boston Scientific. Dr Stone is a consultant for Osprey, Reva, Boston Scientific, Abbott Vascular, Astra Zeneca, Eli Lilly - Daiichi Sankyo partnership, The Medicines Company, Gilead, InspireMD, TherOx, Atrium, Volcano Corporation, InfraRedx, Miracor, MPP group, Lutonix, Velomedix, CSI, Aga, Thoratec, and Medtronic; stock options with Biostar I, II, and III, Medfocus I, II, and Accelorator, Caliber, FlowCardia, Guided Delivery Systems, Arstasis, Micardia, Embrella, and VNT; and is on the Boston Scientific advisory board. Dr Weissman reports research grants from Boston Scientific. The remaining authors report no conflicts of interest regarding the content herein. 

Manuscript submitted August 20, 2013, provisional acceptance given October 9, 2013, final version accepted April 7, 2014.

Address for correspondence: Radoslaw Pracon, MD, PhD, National Institute of Cardiology, Coronary and Structural Heart Diseases Dept. 2nd Floor, 42 Alpejska St, 04-628 Warsaw, Poland. Email: radekpracon@yahoo.pl