- Email: [email protected]
Stent restenosis in a Chinese population
International Journal of Cardiology 102 (2005) 137 – 141 www.elsevier.com/locate/ijcard
Stent restenosis in a Chinese population Kare H. Tang, Wilson W.M. Chan, Roman C.W. Chiu, Kin K. Tse Philip H.C. Wong, John E. Sanderson* Department of Medicine and Therapeutics, Division of Cardiology, The Chinese University of Hong Kong, 9/F, Clinical Sciences Building, Prince of Wales Hospital, Shatin, NT, Hong Kong SAR, China Received 13 March 2002; received in revised form 6 June 2004; accepted 19 June 2004 Available online 1 August 2004
Abstract Background: Stents are now widely used in Hong Kong and China and there is a clinical impression that restenosis is less common because of the lower prevalence of coronary artery disease and associated risk factors in the Chinese. However, there are no published data on angiographic stent restenosis rates in Chinese patients. Method: In a prospective study of 114 consecutive Chinese patients who underwent coronary stenting, quantitative coronary analyses were made at the time of stent implantation and subsequently at 6 months post-stenting (n=97). Results: At 6 months, restenosis (z50% diameter stenosis in the dilated segment) was present in 42 (43.3%) of the 97 patients and 54 (33.5%) of the total 161 lesions stented. Vessel reference diameter (VRD) of b3 mm and stented length of z18 mm were associated with higher restenosis rates (36% and 38%). Compared to those without, those with restenosis had a greater residual stenosis of 16.53F11.54% and smaller final minimal luminal diameter (MLD) of 2.41F0.49 mm, (pb0.01 and pb0.008 respectively). Standard coronary risk factors were not associated with a higher rate of restenosis. Lesion morphology was significantly associated with restenosis. Conclusion: Coronary stenting in Hong Kong Chinese patients is associated with a restenosis rate comparable to that demonstrated in previously published trials from populations in the West. D 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Stents; Restenosis; Asia; Chinese
1. Introduction Percutaneous transluminal coronary stenting (PTCS) is performed frequently in Hong Kong and now occurs in up to 80–90% of coronary angioplasty procedures. Their widespread use was stimulated by the landmark trials STRESS and Benestent [1,2]. The principal indication is to reduce restenosis rates compared to plain balloon angioplasty and to improve acute angiographic outcome. The effectiveness has been best shown in larger coronary vessels with minimum luminal diameter (MLD) of N3 m in Caucasians [1,2]. Few trials have documented benefit in smaller vessels (MLD b3 mm) [3]. This may be particularly * Corresponding author. Tel.: +852 2632 2064; fax: +852 2637 3852. E-mail address: [email protected] (J.E. Sanderson). 0167-5273/$ - see front matter D 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2004.06.001
important in a Chinese population who generally are of smaller stature and hence may have smaller coronary arterial diameter. In addition, the risk factor profile in Chinese and Asian populations is slightly different and there is a less procoagulable predisposition reflected by the lower incidence of thromboembolic diseases but high prevalence of diabetes [4,5]. In the West, the angiographic restenosis rate postcoronary arterial stenting has been demonstrated to be around 20–30% in large vessels, with some studies even reporting a restenosis rate of 15% or lower [6,7]. However, this could be a gross underestimation of the restenosis rate in real practice. Furthermore, there is a commonly held view in the local angiographic community that restenosis is uncommon in Chinese patients. However, there are no published reports prospectively assessing the angiographic restenosis rate in a Chinese population.
138
K.H. Tang et al. / International Journal of Cardiology 102 (2005) 137–141
In this study, we report the incidence of angiographic restenosis postcoronary stenting from a prospective study. This involved single and multiple vessels, involving both large and small coronary arteries in order to determine the restenosis rate in typical clinical practice of Chinese patients.
2. Methods 2.1. Study population The patients were selected from the coronary angioplasty registry of the Prince of Wales Hospital between the period of 20/04/98 and 05/10/98 and consisted of 114 consecutive patients who were subjected to coronary stenting during angioplasty. Patients who had only plain balloon angioplasty were excluded. The stented patients underwent repeated coronary angiogram, prospectively, approximately 6 months post percutaneous transluminal coronary stenting (PTCS) to assess the incidence of in-stent restenosis. The patients’ demographic data were also recorded as were the presence of known coronary risk factors. Direct angioplasty for acute myocardial infarction (MI) was not included. Twenty-four (20%) of these subjects had a recent history of unstable angina prior to the procedure. 2.2. Stent techniques Coronary stenting was performed by two principal operators, and indications were according to the operator deemed appropriate at the time. There was no strict protocol on how or which interventionalist should perform the procedure, and there was no restriction on the choice or the number of stents deployed. Routine use of IVUS was deemed unnecessary unless there was a specific indication. Adjunct rotablation was allowed according to lesion morphology. High-pressure deployment (N10 atmosphere) was utilised to optimise the stent result. All lesions were predilated and the stent deployment is optimised by poststenting dilatation. All stented patients were treated with aspirin 160 mg OD and ticlodipine 250 mg OD for 4 weeks followed by long-term aspirin unless specifically contraindicated. Abciximab was used in only two patients. 2.3. Angiographic analysis The cineangiogram on all eligible patients were reviewed and analysed by two independent experienced interventional cardiologists. Quantitative coronary angiographic (QCA) analysis was performed on all patient pre-PTCS and immediate post-PTCS as well as angiographic follow-up at approximately 6 months. This was performed by a commercially available automated edge detection quantitative angiography system QCA-CMS (Quantitative Coronary and Left Ventricular Angiography analytical software pack-
ages, running on the Cardiovascular Measurement System). Pre- and post-procedural angiograms were obtained after the administration of intracoronary nitroglycerine (200 Ag) in a minimum of two orthogonal views best documenting the target lesion. The same views in the exact angulation were used in the 6-months restenosis study. Images of distal empty catheter were also obtained in each view for calibration purposes. The vessel reference diameter (VRD) was taken as the mean of the vessel diameter proximal and distal to the stenotic lesion. The lesion length (LL) was measured as the stent length on the principle that the stent coverage was optimal with minimal coverage of normal segments. The minimal luminal diameter was recorded as the smallest diameter in the stented segment. The lesion morphology was noted and classified in accordance with the American Heart Association/American College of Cardiology (AHA/ACC) classification [8]. Restenosis was defined as the presence of z50% diameter stenosis in the dilated segment.
3. Statistics The statistical calculations were undertaken using the commercial statistical software (SPSS, TN, USA; release 9.0). Descriptive group data are presented as meanF standard deviation unless otherwise stated. The Chi square test and the Student’s t-test were used to compare qualitative and continuous variables, respectively, and P values of 0.05 or lower on two-tailed testing were considered to indicate statistical significance.
4. Results Between the period of study, a total of 141 patients had percutaneous coronary intervention procedures. Of these, 114 (81%) patients had coronary stenting with 100% success rate. Ninety-seven (85%) of these patients had a follow-up angiography. The residual 17 patients were all accounted for: 4 patients had medical contraindications to further angiography, 6 patients refused further investigations and 7 died (3 cancer, 1 stroke, 1 chronic renal failure, 1 unrelated post operation, 1 sudden death 6 weeks postcoronary intervention). Three patients had symptoms suggestive of mild angina. There was a predominance of males (76.3%). The average age was 62.5 (F9) years with a range of 40–79. There was a weak but statistically significant relationship between age and presence of significant restenosis. Those with restenosis had a greater mean age (64.6F8.6 years) as compared with the group without restenosis (mean age of 61.4F9.1 years, Pb0.031). Out of the total of 97 patients, 42 patients (43.3%) had restenosis in at least one vessel. A total of 161 lesions were stented, and of these, 54 (33.5%) developed in-stent
K.H. Tang et al. / International Journal of Cardiology 102 (2005) 137–141
restenosis. Eight total occlusions were demonstrated at the 6 months restudy angiogram. 4.1. Risk factors Diabetes, hyperlipidaemia, smoking history, history of myocardial infarction (MI) and hypertension were not statistical significantly correlated with the presence or absence of restenosis in our study population (Tables 1 and 2). 4.2. Vessel size Of the 161 lesions stented, 88 (54.7%) had a vessel reference diameter of b3 mm and 73 (45.3%) with reference diameter of z3 mm. Sixty-six (41.0%) stented lesion were b18 mm in length and 95 (59%) stented lesion was z18 mm in length. There was no statistical significance in the correlation between the restenosis rate with vessel reference diameter or lesion/stent length in our study (Table 3). An odds ratio of 1.63 was demonstrated between short (b18 mm) and long (z18 mm) stent lengths for restenosis, and an odds ratio of 0.7 for small (b3 mm) and large (z3 mm) vessel diameter for the development of restenosis. Grouping the stented lesions according to vessel size and stented length gave a 27.0% restenosis rate in lesions with vessel diameter b3 mm and lesion length b18 mm. However, there was a restenosis rate of 43.1% in lesions with vessel diameter b3 mm and lesion length z18 mm. Intermediate restenosis rates were found in lesions with vessel diameter z3 mm and lesion length of b18 mm (27.6%) and in lesions with vessel diameter z3 mm and lesion length of z18 mm (31.8%; Table 3). The difference in restenosis rate between these groups did not reach statistical significance.
Table 2 Comparison of groups with and without restenosis Number of lesions Number of patients Age (years) Diabetes mellitus Hyperlipidaemia Smoking Myocardial infarction Hypertension Residual stenosis (%) Final Minimal Luminal Diameter (mm) Pressure (atm) Single-vessel disease Two-vessels disease Three-vessels disease Lesion morphology (AHA) Type A Type B1 Type B2 Type C
No restenosis
Restenosis
p-Value
107 55 61.4F9.1 17 (15.1%) 21 (19.6%) 20 (18.7%) 33 (30.8%) 29 (27.1%) 11.8F10.7* 2.7F0.6*
54 42 64.6F8.6 8 (14.8%) 17 (31.5%) 18 (33.3%) 19 (35.2%) 17 (31.5%) 16.5F11.5 2.4F0.5
b0.03 NS NS NS NS NS b0.01 b0.008
13.8F2.3 36 (69.2%) 13 (46.4%) 6 (35.3%)
16.5F11.1 16 (30.8%) 15 (53.6%) 11 (64.7%)z
NS NS NS NS
3 (60%) 36 (80%) 50 (64.9%) 18 (52.9%)
2 (40%) 9 (20%)y 27 (35.1%) 16 (47.1%)
NS NS NS NS
*
p-values obtained from independent sample t-test, whereas p-values for other variables are obtained from Pearson’s Chi square test. y The percentage of restenosis in lesion B1 is significantly smaller than that in lesion C (odds ratio: 0.28, 95% CI: 0.1, 0.75, p=0.01). z The percentage of restenosis in patients with three-vessels disease is significantly larger than that in patients with only single-vessel disease (odds ratio: 3.67, 95% CI: 1.17, 11.51, p=0.026).
luminal diameter of 2.67 mm (F0.62) post-stenting as compared to those with restenosis with a residual stenosis of 16.53% (F11.54) and final minimal luminal diameter of 2.41 mm (F0.49). These differences reached statistical significance with Pb0.01 and b0.008, respectively (Table 2).
Table 3 Restenosis rates (161 lesions)
4.3. Effect of minimal luminal diameter The group without restenosis had a lower residual stenosis of 11.8% (F10.7) and larger final minimal Table 1 Patient demographics and risk factors (n=97) Age (years) Male Diabetes mellitus Hyperlipidaemia Smoking Previous myocardial infarction Hypertension Single-vessel disease Two-vessels disease Three-vessels disease Lesions/morphology A B1 B2 C
139
62.2F8.5 74 (76.3%) 25 (25.8%) 38 (39.2%) 38 (39.2%) 52 (53.6%) 46 (47.4%) 52 (53.6%) 28 (28.9%) 17 (17.5%) 5 (3.1%) 45 (28.0%) 77 (47.8%) 34 (21.1%)
Restenosis
No restenosis
Total
VRD (mm) b3 z3 p-Value
32 (36%) 22 (30%) NS
56 (64%) 51 (70%) NS
88 73
LL (mm) b18 z18 p-Value
18 (28%) 36 (37%) NS
47 (72%) 60 (63%) NS
65 96
VRD and LL (mm) b3 and b18 b3 and z18 z3 and b18 z3 and z18 p-Value
10 (27.0%) 22 (43.1%) 8 (27.6%) 14 (31.8%) NS
27 (73.0%) 29 (56.9%) 21 (72.4%) 30 (68.2%) NS
37 51 29 44
VRD, vessel reference diameter; LL, lesion length. Odds ratio of 1.63 and 0.7 between short (b18 mm) and long (18z mm) stented lesion lengths and for small (b3 mm)-vessel and large (z3 mm)-vessel diameter for the development of restenosis, respectively.
140
K.H. Tang et al. / International Journal of Cardiology 102 (2005) 137–141
The stented lesions were separated into four final MLD sizes (b2.40, 2.41–2.80, 2.81–3.20 and N3.20 mm) and into short (b18 mm) and long (z18 mm) stented lengths (Table 4). Lesions with a MLD V2.8 mm had the highest restenosis rate of 48–53% in long stented segments, with no restenosis in the stented segment with a MLD N3.2 mm, but in short segments, restenosis rates were 28–30% which were not significantly worst than lesions with larger MLD (23–29%).
Table 5 Vessel profile post-stenting (161 lesions) Final MLD (mm) VRD (mm) Residual stenosis (%) Inflation pressure (atm) Restenotic percentage (%)
2.59F0.59 3.00F0.49 13.37F11.06 13.65F4.44 43.61F23.22
atm, atmosphere.
diameter and the other in a z3-mm diameter coronary artery both of which were in long stented segments (z18 mm).
4.4. Other factors The mean deployment pressure for the whole group was 13.6F4.4 atm (Table 5). There was no statistically significance in the final stent deployment pressure between those with and without restenosis (16.5F11.1 and 13.8F2.3 atm, respectively). There was 52 (53.6%) single-vessel disease, 28 (28.9%) two-vessel disease 17 (17.5%) three- or more-vessels disease. The restenosis rates were 30.7%, 53.6% and 64.7%, respectively. The percentage of restenosis occurs in patients with three-vessels disease is significantly larger than that in patients with only single-vessel disease (odds ratio: 3.67, 95% CI: 1.17, 11.51). The most common stents used were AVE GFX, Seaquence, Multilink, NIR and Crossflex. There were a total of 14 different models of stent used within these 161 lesions. Examining the lesion morphology, there were 5 type A, 45 type B1, 77 type B2 and 34 type C lesions (Table 1), with respective restenosis rate of 40%, 20%, 25.1% and 47.1%. The percentage of restenosis in lesion B1 was significantly smaller than that in lesion C (odds ratio: 0.28, CI: 0.1, 0.75; Table 2). Only two cases required adjunct rotablation, both in calcified coronary stenosis, which subsequently developed restenosis. One was in a b3-mm
Table 4 Correlations between final minimum luminal diameter (MLD) and stent length with restenosis Final MLD (mm)
Length (mm)
No restenosis
Restenosisa
Total (65)
b2.40 2.4–2.80 2.81–3.20 N3.20
b18 b18 b18 b18
18 16 10 5
7 6 3 2
25 20 13 7
Final MLD (mm)
Length (mm)
No restenosis
Restenosisb
Total (96)
b2.40 2.41–2.80 2.81–3.20 N3.20
z18 z18 z18 z18
17 14 17 12
19 (52.8%) 13 (48.1%) 4 (19.0%) 0 (0%)
36 27 21 12
(28.0%) (30.0%) (23.1%) (28.6%)
Groups are divided using length of 18 mm as a cutoff. Only length N18 mm was significant. a Chi square, p=0.983. b Chi square, p=0.003.
5. Discussion In our Chinese patients, we found that the stent restenosis rate was 33.5% in a total of 161 stented lesions and this is comparable to the STRESS trial [1]. However, the STRESS study did not discriminate between longer lesions and smaller vessels. The final luminal diameter has been reported by various investigators to be an independent predictor of restenosis [9,10] and this is confirmed in our study. A higher residual stenosis is also statistically significantly associated with the likelihood to developed restenosis, as this correlates with the minimal luminal diameter. The restenosis rate is also reportedly lower in shorter stent/lesion lengths but in our study the smaller vessel diameter was associated with higher restenosis in only those with longer stents/lesions. This supports the results reported by Savage et al. [11] who suggested that coronary stenting is also beneficial in reducing restenosis in smaller diameter vessels. The stent restenosis rate per patient was particularly high with more than 40% angiographic restenosis in our study group of 97 patients. This is considerably higher than previous large studies as we did not discriminate between single- or multivessel disease. The angiographic restenosis rate is around 30% in single-vessel disease as compared with nearly 65% in patients who had multivessel stenting. There were eight (5%) late total occlusions in this study which is higher than that reported by Bauters et al. [9]. Four of the total occlusions occurred in two patients who received multivessel stenting. The exact mechanisms for late occlusions are unknown. Six of the eight late total occlusions had a residual stenosis post-stenting of greater than 18%. In this study, residual stenosis was significantly associated with restenosis. There was one sudden death that occurred 6 weeks after the coronary intervention that may be due to acute coronary occlusion although subacute thrombosis at such a period postintervention is less likely. All patients received the proven regimen of dual antiplatelet therapy with no obvious complications. We employed the regimen of ticlodipine, 250 mg, once daily rather than twice daily, as this was the local practise at the time of the study. There was no obvious evidence of subacute thrombosis and this confirms the
K.H. Tang et al. / International Journal of Cardiology 102 (2005) 137–141
effectiveness of this lower dose ticlodipine regime. One of the deaths occurred postsurgery (noncardiac) but there was no evidence to suggest an acute coronary event. Acute stenting success was excellent with no failure. The standard coronary risk factors did not show any significant correlation with restenosis and this is most likely due to two reasons. Firstly, the patient sample size was not powered to show any significance and secondly, stent restenosis is due to intimal hyperplasia and not due progression of atheroma. Diabetes is associated with higher restenosis rate compared to non-diabetics in plain balloon angioplasty, and this had also been reported after coronary stenting although there had been reports that suggest that this was not the case [12], as illustrated in our study. The distribution of lesion morphology in this study showed that the majority of stented lesions are type B or C. Only 3% of the lesions are type A. This is in agreement with current practice with the indication for stenting for more complex lesions which have a poorer angiographic results with simple balloon angioplasty. Our study, confirms that the more complex the lesion, the higher the chance of restenosis with restenosis rate of 20% in AHA type B1 morphology lesions as compared with 47.1% in type C morphology lesions. This is consistent with the study by Kastrati et al. [8]. The high incidence of restenosis in type A lesions in our study was probably due to the small number, with less than 3% of total lesions. It may be considered that a limitation of this study was the use of a large variety of coronary stents (14 in total) each with its own properties. A recent study has implicated the design of the stents could affect the restenosis rate. However, the aim of this study was not to compare stents and to reflect usual clinical practice. If only one stent has been used then the results would not be widely applicable. In conclusion, in this study, we have demonstrated for the first time that the incidence of coronary restenosis in Hong Kong Chinese postcoronary stenting is high. The results are comparable to the western Caucasian data and dispel the commonly held perception that restenosis is less common in Chinese due to the lower thromboembolic predisposition and other, perhaps genetic, factors. This study demonstrated similar factors shown in previous Western studies that increases the restenosis rate including small vessel size,
141
increased stent length, high-grade lesion morphology, multivessel stenting and increased residual stenosis. By being more selective in choosing which lesion to stent or angioplasty, a currently acceptable result could be achieved including in smaller vessels with short stenosis. Recent developments such as brachytherapy and rapamicin-coated stents may further reduce restenosis. References [1] Fischman DL, Leon MB, Baim DS, et al. A randomised comparison of coronary-stent placement and balloon angioplasty in patients in the treatment of coronary artery disease. N Engl J Med 1994;331: 496 – 501. [2] Serruys PW, De Jaegere P, Kiemeneij F, et al. A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med 1994;331: 489 – 95. [3] Konig R, Chan C, Eltchaninoff H, et al. Primary stenting of de novo lesions in small coronary arteries: a prospective, pilot study. Catheter Cardiovasc Diagn 1998;45:235 – 8. [4] Woo KS, Chook P, Young RP, Sanderson JE. New risk factors for coronary heart disease in Asia. Int J Cardiol 1997;62(Suppl 1): S39–42. [5] Woo KS, Tse LKK, Tse CY, Metreweli C, Vallance-Owen J. The prevalence and pattern of pulmonary thromboembolism in the Chinese in Hong Kong. Int J Cardiol 1988;20:373 – 80. [6] Serruys PW, Vandergiessen W, Garcia E, et al. Clinical and angiographic results with the Multi-Link stent implanted under intravascular ultrasound guidance (West-2 study). J Invasive Cardiol 1998;10:B20–7. [7] Escaned J, Goicolea J, Alfonso F, et al. Propensity and mechanism of restenosis in different coronary stent designs. J Am Coll Cardiol 1999;34:1490 – 7. [8] Kastrati A, Schomig A, Elezi S, et al. Prognostic value of the modified American College of Cardiology/American Heart Association stenosis morphology classification, for long-term angiographic and clinical outcome after coronary stent placement. Circulation 1999;100:1285 – 90. [9] Bauters C, Hulbert E, Prat A, et al. Predictors of restenosis after coronary stent implantation. J Am Coll Cardiol 1998;31:1291 – 8. [10] Braunucci E, Angoli L, Merlini PA, et al. Adjunctive stent implantation following directional coronary atherectomy in patients with coronary artery disease. J Am Coll Cardiol 1998;32:1855 – 60. [11] Savage M, Fischman DL, Pake R, et al. Efficacy of coronary stenting versus balloon angioplasty in small coronary arteries. J Am Coll Cardiol 1998;31:307 – 11. [12] Elezi S, Kalvati A, Pache J, et al. Diabetes mellitus and the clinical and angiographic outcome after stent placement. J Am Coll Cardiol 1998;32:1866 – 73.
Stent restenosis in a Chinese population Kare H. Tang, Wilson W.M. Chan, Roman C.W. Chiu, Kin K. Tse Philip H.C. Wong, John E. Sanderson* Department of Medicine and Therapeutics, Division of Cardiology, The Chinese University of Hong Kong, 9/F, Clinical Sciences Building, Prince of Wales Hospital, Shatin, NT, Hong Kong SAR, China Received 13 March 2002; received in revised form 6 June 2004; accepted 19 June 2004 Available online 1 August 2004
Abstract Background: Stents are now widely used in Hong Kong and China and there is a clinical impression that restenosis is less common because of the lower prevalence of coronary artery disease and associated risk factors in the Chinese. However, there are no published data on angiographic stent restenosis rates in Chinese patients. Method: In a prospective study of 114 consecutive Chinese patients who underwent coronary stenting, quantitative coronary analyses were made at the time of stent implantation and subsequently at 6 months post-stenting (n=97). Results: At 6 months, restenosis (z50% diameter stenosis in the dilated segment) was present in 42 (43.3%) of the 97 patients and 54 (33.5%) of the total 161 lesions stented. Vessel reference diameter (VRD) of b3 mm and stented length of z18 mm were associated with higher restenosis rates (36% and 38%). Compared to those without, those with restenosis had a greater residual stenosis of 16.53F11.54% and smaller final minimal luminal diameter (MLD) of 2.41F0.49 mm, (pb0.01 and pb0.008 respectively). Standard coronary risk factors were not associated with a higher rate of restenosis. Lesion morphology was significantly associated with restenosis. Conclusion: Coronary stenting in Hong Kong Chinese patients is associated with a restenosis rate comparable to that demonstrated in previously published trials from populations in the West. D 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Stents; Restenosis; Asia; Chinese
1. Introduction Percutaneous transluminal coronary stenting (PTCS) is performed frequently in Hong Kong and now occurs in up to 80–90% of coronary angioplasty procedures. Their widespread use was stimulated by the landmark trials STRESS and Benestent [1,2]. The principal indication is to reduce restenosis rates compared to plain balloon angioplasty and to improve acute angiographic outcome. The effectiveness has been best shown in larger coronary vessels with minimum luminal diameter (MLD) of N3 m in Caucasians [1,2]. Few trials have documented benefit in smaller vessels (MLD b3 mm) [3]. This may be particularly * Corresponding author. Tel.: +852 2632 2064; fax: +852 2637 3852. E-mail address: [email protected] (J.E. Sanderson). 0167-5273/$ - see front matter D 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijcard.2004.06.001
important in a Chinese population who generally are of smaller stature and hence may have smaller coronary arterial diameter. In addition, the risk factor profile in Chinese and Asian populations is slightly different and there is a less procoagulable predisposition reflected by the lower incidence of thromboembolic diseases but high prevalence of diabetes [4,5]. In the West, the angiographic restenosis rate postcoronary arterial stenting has been demonstrated to be around 20–30% in large vessels, with some studies even reporting a restenosis rate of 15% or lower [6,7]. However, this could be a gross underestimation of the restenosis rate in real practice. Furthermore, there is a commonly held view in the local angiographic community that restenosis is uncommon in Chinese patients. However, there are no published reports prospectively assessing the angiographic restenosis rate in a Chinese population.
138
K.H. Tang et al. / International Journal of Cardiology 102 (2005) 137–141
In this study, we report the incidence of angiographic restenosis postcoronary stenting from a prospective study. This involved single and multiple vessels, involving both large and small coronary arteries in order to determine the restenosis rate in typical clinical practice of Chinese patients.
2. Methods 2.1. Study population The patients were selected from the coronary angioplasty registry of the Prince of Wales Hospital between the period of 20/04/98 and 05/10/98 and consisted of 114 consecutive patients who were subjected to coronary stenting during angioplasty. Patients who had only plain balloon angioplasty were excluded. The stented patients underwent repeated coronary angiogram, prospectively, approximately 6 months post percutaneous transluminal coronary stenting (PTCS) to assess the incidence of in-stent restenosis. The patients’ demographic data were also recorded as were the presence of known coronary risk factors. Direct angioplasty for acute myocardial infarction (MI) was not included. Twenty-four (20%) of these subjects had a recent history of unstable angina prior to the procedure. 2.2. Stent techniques Coronary stenting was performed by two principal operators, and indications were according to the operator deemed appropriate at the time. There was no strict protocol on how or which interventionalist should perform the procedure, and there was no restriction on the choice or the number of stents deployed. Routine use of IVUS was deemed unnecessary unless there was a specific indication. Adjunct rotablation was allowed according to lesion morphology. High-pressure deployment (N10 atmosphere) was utilised to optimise the stent result. All lesions were predilated and the stent deployment is optimised by poststenting dilatation. All stented patients were treated with aspirin 160 mg OD and ticlodipine 250 mg OD for 4 weeks followed by long-term aspirin unless specifically contraindicated. Abciximab was used in only two patients. 2.3. Angiographic analysis The cineangiogram on all eligible patients were reviewed and analysed by two independent experienced interventional cardiologists. Quantitative coronary angiographic (QCA) analysis was performed on all patient pre-PTCS and immediate post-PTCS as well as angiographic follow-up at approximately 6 months. This was performed by a commercially available automated edge detection quantitative angiography system QCA-CMS (Quantitative Coronary and Left Ventricular Angiography analytical software pack-
ages, running on the Cardiovascular Measurement System). Pre- and post-procedural angiograms were obtained after the administration of intracoronary nitroglycerine (200 Ag) in a minimum of two orthogonal views best documenting the target lesion. The same views in the exact angulation were used in the 6-months restenosis study. Images of distal empty catheter were also obtained in each view for calibration purposes. The vessel reference diameter (VRD) was taken as the mean of the vessel diameter proximal and distal to the stenotic lesion. The lesion length (LL) was measured as the stent length on the principle that the stent coverage was optimal with minimal coverage of normal segments. The minimal luminal diameter was recorded as the smallest diameter in the stented segment. The lesion morphology was noted and classified in accordance with the American Heart Association/American College of Cardiology (AHA/ACC) classification [8]. Restenosis was defined as the presence of z50% diameter stenosis in the dilated segment.
3. Statistics The statistical calculations were undertaken using the commercial statistical software (SPSS, TN, USA; release 9.0). Descriptive group data are presented as meanF standard deviation unless otherwise stated. The Chi square test and the Student’s t-test were used to compare qualitative and continuous variables, respectively, and P values of 0.05 or lower on two-tailed testing were considered to indicate statistical significance.
4. Results Between the period of study, a total of 141 patients had percutaneous coronary intervention procedures. Of these, 114 (81%) patients had coronary stenting with 100% success rate. Ninety-seven (85%) of these patients had a follow-up angiography. The residual 17 patients were all accounted for: 4 patients had medical contraindications to further angiography, 6 patients refused further investigations and 7 died (3 cancer, 1 stroke, 1 chronic renal failure, 1 unrelated post operation, 1 sudden death 6 weeks postcoronary intervention). Three patients had symptoms suggestive of mild angina. There was a predominance of males (76.3%). The average age was 62.5 (F9) years with a range of 40–79. There was a weak but statistically significant relationship between age and presence of significant restenosis. Those with restenosis had a greater mean age (64.6F8.6 years) as compared with the group without restenosis (mean age of 61.4F9.1 years, Pb0.031). Out of the total of 97 patients, 42 patients (43.3%) had restenosis in at least one vessel. A total of 161 lesions were stented, and of these, 54 (33.5%) developed in-stent
K.H. Tang et al. / International Journal of Cardiology 102 (2005) 137–141
restenosis. Eight total occlusions were demonstrated at the 6 months restudy angiogram. 4.1. Risk factors Diabetes, hyperlipidaemia, smoking history, history of myocardial infarction (MI) and hypertension were not statistical significantly correlated with the presence or absence of restenosis in our study population (Tables 1 and 2). 4.2. Vessel size Of the 161 lesions stented, 88 (54.7%) had a vessel reference diameter of b3 mm and 73 (45.3%) with reference diameter of z3 mm. Sixty-six (41.0%) stented lesion were b18 mm in length and 95 (59%) stented lesion was z18 mm in length. There was no statistical significance in the correlation between the restenosis rate with vessel reference diameter or lesion/stent length in our study (Table 3). An odds ratio of 1.63 was demonstrated between short (b18 mm) and long (z18 mm) stent lengths for restenosis, and an odds ratio of 0.7 for small (b3 mm) and large (z3 mm) vessel diameter for the development of restenosis. Grouping the stented lesions according to vessel size and stented length gave a 27.0% restenosis rate in lesions with vessel diameter b3 mm and lesion length b18 mm. However, there was a restenosis rate of 43.1% in lesions with vessel diameter b3 mm and lesion length z18 mm. Intermediate restenosis rates were found in lesions with vessel diameter z3 mm and lesion length of b18 mm (27.6%) and in lesions with vessel diameter z3 mm and lesion length of z18 mm (31.8%; Table 3). The difference in restenosis rate between these groups did not reach statistical significance.
Table 2 Comparison of groups with and without restenosis Number of lesions Number of patients Age (years) Diabetes mellitus Hyperlipidaemia Smoking Myocardial infarction Hypertension Residual stenosis (%) Final Minimal Luminal Diameter (mm) Pressure (atm) Single-vessel disease Two-vessels disease Three-vessels disease Lesion morphology (AHA) Type A Type B1 Type B2 Type C
No restenosis
Restenosis
p-Value
107 55 61.4F9.1 17 (15.1%) 21 (19.6%) 20 (18.7%) 33 (30.8%) 29 (27.1%) 11.8F10.7* 2.7F0.6*
54 42 64.6F8.6 8 (14.8%) 17 (31.5%) 18 (33.3%) 19 (35.2%) 17 (31.5%) 16.5F11.5 2.4F0.5
b0.03 NS NS NS NS NS b0.01 b0.008
13.8F2.3 36 (69.2%) 13 (46.4%) 6 (35.3%)
16.5F11.1 16 (30.8%) 15 (53.6%) 11 (64.7%)z
NS NS NS NS
3 (60%) 36 (80%) 50 (64.9%) 18 (52.9%)
2 (40%) 9 (20%)y 27 (35.1%) 16 (47.1%)
NS NS NS NS
*
p-values obtained from independent sample t-test, whereas p-values for other variables are obtained from Pearson’s Chi square test. y The percentage of restenosis in lesion B1 is significantly smaller than that in lesion C (odds ratio: 0.28, 95% CI: 0.1, 0.75, p=0.01). z The percentage of restenosis in patients with three-vessels disease is significantly larger than that in patients with only single-vessel disease (odds ratio: 3.67, 95% CI: 1.17, 11.51, p=0.026).
luminal diameter of 2.67 mm (F0.62) post-stenting as compared to those with restenosis with a residual stenosis of 16.53% (F11.54) and final minimal luminal diameter of 2.41 mm (F0.49). These differences reached statistical significance with Pb0.01 and b0.008, respectively (Table 2).
Table 3 Restenosis rates (161 lesions)
4.3. Effect of minimal luminal diameter The group without restenosis had a lower residual stenosis of 11.8% (F10.7) and larger final minimal Table 1 Patient demographics and risk factors (n=97) Age (years) Male Diabetes mellitus Hyperlipidaemia Smoking Previous myocardial infarction Hypertension Single-vessel disease Two-vessels disease Three-vessels disease Lesions/morphology A B1 B2 C
139
62.2F8.5 74 (76.3%) 25 (25.8%) 38 (39.2%) 38 (39.2%) 52 (53.6%) 46 (47.4%) 52 (53.6%) 28 (28.9%) 17 (17.5%) 5 (3.1%) 45 (28.0%) 77 (47.8%) 34 (21.1%)
Restenosis
No restenosis
Total
VRD (mm) b3 z3 p-Value
32 (36%) 22 (30%) NS
56 (64%) 51 (70%) NS
88 73
LL (mm) b18 z18 p-Value
18 (28%) 36 (37%) NS
47 (72%) 60 (63%) NS
65 96
VRD and LL (mm) b3 and b18 b3 and z18 z3 and b18 z3 and z18 p-Value
10 (27.0%) 22 (43.1%) 8 (27.6%) 14 (31.8%) NS
27 (73.0%) 29 (56.9%) 21 (72.4%) 30 (68.2%) NS
37 51 29 44
VRD, vessel reference diameter; LL, lesion length. Odds ratio of 1.63 and 0.7 between short (b18 mm) and long (18z mm) stented lesion lengths and for small (b3 mm)-vessel and large (z3 mm)-vessel diameter for the development of restenosis, respectively.
140
K.H. Tang et al. / International Journal of Cardiology 102 (2005) 137–141
The stented lesions were separated into four final MLD sizes (b2.40, 2.41–2.80, 2.81–3.20 and N3.20 mm) and into short (b18 mm) and long (z18 mm) stented lengths (Table 4). Lesions with a MLD V2.8 mm had the highest restenosis rate of 48–53% in long stented segments, with no restenosis in the stented segment with a MLD N3.2 mm, but in short segments, restenosis rates were 28–30% which were not significantly worst than lesions with larger MLD (23–29%).
Table 5 Vessel profile post-stenting (161 lesions) Final MLD (mm) VRD (mm) Residual stenosis (%) Inflation pressure (atm) Restenotic percentage (%)
2.59F0.59 3.00F0.49 13.37F11.06 13.65F4.44 43.61F23.22
atm, atmosphere.
diameter and the other in a z3-mm diameter coronary artery both of which were in long stented segments (z18 mm).
4.4. Other factors The mean deployment pressure for the whole group was 13.6F4.4 atm (Table 5). There was no statistically significance in the final stent deployment pressure between those with and without restenosis (16.5F11.1 and 13.8F2.3 atm, respectively). There was 52 (53.6%) single-vessel disease, 28 (28.9%) two-vessel disease 17 (17.5%) three- or more-vessels disease. The restenosis rates were 30.7%, 53.6% and 64.7%, respectively. The percentage of restenosis occurs in patients with three-vessels disease is significantly larger than that in patients with only single-vessel disease (odds ratio: 3.67, 95% CI: 1.17, 11.51). The most common stents used were AVE GFX, Seaquence, Multilink, NIR and Crossflex. There were a total of 14 different models of stent used within these 161 lesions. Examining the lesion morphology, there were 5 type A, 45 type B1, 77 type B2 and 34 type C lesions (Table 1), with respective restenosis rate of 40%, 20%, 25.1% and 47.1%. The percentage of restenosis in lesion B1 was significantly smaller than that in lesion C (odds ratio: 0.28, CI: 0.1, 0.75; Table 2). Only two cases required adjunct rotablation, both in calcified coronary stenosis, which subsequently developed restenosis. One was in a b3-mm
Table 4 Correlations between final minimum luminal diameter (MLD) and stent length with restenosis Final MLD (mm)
Length (mm)
No restenosis
Restenosisa
Total (65)
b2.40 2.4–2.80 2.81–3.20 N3.20
b18 b18 b18 b18
18 16 10 5
7 6 3 2
25 20 13 7
Final MLD (mm)
Length (mm)
No restenosis
Restenosisb
Total (96)
b2.40 2.41–2.80 2.81–3.20 N3.20
z18 z18 z18 z18
17 14 17 12
19 (52.8%) 13 (48.1%) 4 (19.0%) 0 (0%)
36 27 21 12
(28.0%) (30.0%) (23.1%) (28.6%)
Groups are divided using length of 18 mm as a cutoff. Only length N18 mm was significant. a Chi square, p=0.983. b Chi square, p=0.003.
5. Discussion In our Chinese patients, we found that the stent restenosis rate was 33.5% in a total of 161 stented lesions and this is comparable to the STRESS trial [1]. However, the STRESS study did not discriminate between longer lesions and smaller vessels. The final luminal diameter has been reported by various investigators to be an independent predictor of restenosis [9,10] and this is confirmed in our study. A higher residual stenosis is also statistically significantly associated with the likelihood to developed restenosis, as this correlates with the minimal luminal diameter. The restenosis rate is also reportedly lower in shorter stent/lesion lengths but in our study the smaller vessel diameter was associated with higher restenosis in only those with longer stents/lesions. This supports the results reported by Savage et al. [11] who suggested that coronary stenting is also beneficial in reducing restenosis in smaller diameter vessels. The stent restenosis rate per patient was particularly high with more than 40% angiographic restenosis in our study group of 97 patients. This is considerably higher than previous large studies as we did not discriminate between single- or multivessel disease. The angiographic restenosis rate is around 30% in single-vessel disease as compared with nearly 65% in patients who had multivessel stenting. There were eight (5%) late total occlusions in this study which is higher than that reported by Bauters et al. [9]. Four of the total occlusions occurred in two patients who received multivessel stenting. The exact mechanisms for late occlusions are unknown. Six of the eight late total occlusions had a residual stenosis post-stenting of greater than 18%. In this study, residual stenosis was significantly associated with restenosis. There was one sudden death that occurred 6 weeks after the coronary intervention that may be due to acute coronary occlusion although subacute thrombosis at such a period postintervention is less likely. All patients received the proven regimen of dual antiplatelet therapy with no obvious complications. We employed the regimen of ticlodipine, 250 mg, once daily rather than twice daily, as this was the local practise at the time of the study. There was no obvious evidence of subacute thrombosis and this confirms the
K.H. Tang et al. / International Journal of Cardiology 102 (2005) 137–141
effectiveness of this lower dose ticlodipine regime. One of the deaths occurred postsurgery (noncardiac) but there was no evidence to suggest an acute coronary event. Acute stenting success was excellent with no failure. The standard coronary risk factors did not show any significant correlation with restenosis and this is most likely due to two reasons. Firstly, the patient sample size was not powered to show any significance and secondly, stent restenosis is due to intimal hyperplasia and not due progression of atheroma. Diabetes is associated with higher restenosis rate compared to non-diabetics in plain balloon angioplasty, and this had also been reported after coronary stenting although there had been reports that suggest that this was not the case [12], as illustrated in our study. The distribution of lesion morphology in this study showed that the majority of stented lesions are type B or C. Only 3% of the lesions are type A. This is in agreement with current practice with the indication for stenting for more complex lesions which have a poorer angiographic results with simple balloon angioplasty. Our study, confirms that the more complex the lesion, the higher the chance of restenosis with restenosis rate of 20% in AHA type B1 morphology lesions as compared with 47.1% in type C morphology lesions. This is consistent with the study by Kastrati et al. [8]. The high incidence of restenosis in type A lesions in our study was probably due to the small number, with less than 3% of total lesions. It may be considered that a limitation of this study was the use of a large variety of coronary stents (14 in total) each with its own properties. A recent study has implicated the design of the stents could affect the restenosis rate. However, the aim of this study was not to compare stents and to reflect usual clinical practice. If only one stent has been used then the results would not be widely applicable. In conclusion, in this study, we have demonstrated for the first time that the incidence of coronary restenosis in Hong Kong Chinese postcoronary stenting is high. The results are comparable to the western Caucasian data and dispel the commonly held perception that restenosis is less common in Chinese due to the lower thromboembolic predisposition and other, perhaps genetic, factors. This study demonstrated similar factors shown in previous Western studies that increases the restenosis rate including small vessel size,
141
increased stent length, high-grade lesion morphology, multivessel stenting and increased residual stenosis. By being more selective in choosing which lesion to stent or angioplasty, a currently acceptable result could be achieved including in smaller vessels with short stenosis. Recent developments such as brachytherapy and rapamicin-coated stents may further reduce restenosis. References [1] Fischman DL, Leon MB, Baim DS, et al. A randomised comparison of coronary-stent placement and balloon angioplasty in patients in the treatment of coronary artery disease. N Engl J Med 1994;331: 496 – 501. [2] Serruys PW, De Jaegere P, Kiemeneij F, et al. A comparison of balloon-expandable-stent implantation with balloon angioplasty in patients with coronary artery disease. N Engl J Med 1994;331: 489 – 95. [3] Konig R, Chan C, Eltchaninoff H, et al. Primary stenting of de novo lesions in small coronary arteries: a prospective, pilot study. Catheter Cardiovasc Diagn 1998;45:235 – 8. [4] Woo KS, Chook P, Young RP, Sanderson JE. New risk factors for coronary heart disease in Asia. Int J Cardiol 1997;62(Suppl 1): S39–42. [5] Woo KS, Tse LKK, Tse CY, Metreweli C, Vallance-Owen J. The prevalence and pattern of pulmonary thromboembolism in the Chinese in Hong Kong. Int J Cardiol 1988;20:373 – 80. [6] Serruys PW, Vandergiessen W, Garcia E, et al. Clinical and angiographic results with the Multi-Link stent implanted under intravascular ultrasound guidance (West-2 study). J Invasive Cardiol 1998;10:B20–7. [7] Escaned J, Goicolea J, Alfonso F, et al. Propensity and mechanism of restenosis in different coronary stent designs. J Am Coll Cardiol 1999;34:1490 – 7. [8] Kastrati A, Schomig A, Elezi S, et al. Prognostic value of the modified American College of Cardiology/American Heart Association stenosis morphology classification, for long-term angiographic and clinical outcome after coronary stent placement. Circulation 1999;100:1285 – 90. [9] Bauters C, Hulbert E, Prat A, et al. Predictors of restenosis after coronary stent implantation. J Am Coll Cardiol 1998;31:1291 – 8. [10] Braunucci E, Angoli L, Merlini PA, et al. Adjunctive stent implantation following directional coronary atherectomy in patients with coronary artery disease. J Am Coll Cardiol 1998;32:1855 – 60. [11] Savage M, Fischman DL, Pake R, et al. Efficacy of coronary stenting versus balloon angioplasty in small coronary arteries. J Am Coll Cardiol 1998;31:307 – 11. [12] Elezi S, Kalvati A, Pache J, et al. Diabetes mellitus and the clinical and angiographic outcome after stent placement. J Am Coll Cardiol 1998;32:1866 – 73.