Contemporary Clinical Trials 38 (2014) 51–58
Contents lists available at ScienceDirect
Contemporary Clinical Trials journal homepage: www.elsevier.com/locate/conclintrial
Percutaneous coronary intervention versus coronary artery bypass graft for stable angina: Meta-regression of randomized trials☆,☆☆ Fabrizio D'Ascenzo a,g,⁎, Umberto Barbero a, Claudio Moretti a,g, Tullio Palmerini b, Diego Della Riva b, Andrea Mariani b, Pierluigi Omedè a, James J. DiNicolantonio e, Giuseppe Biondi-Zoccai c,d,f, Fiorenzo Gaita a a
Division of Cardiology, University of Turin, Italy Division of Cardiology, University of Bologna, Italy Department of Medico-Surgical Sciences and Biotechnologies, Italy d Sapienza University of Rome, Latina, Italy e Mid America Heart Institute at Saint Luke's Hospital, Kansas City, MO, United States f Meta-analysis and Evidence based medicine Training in Cardiology (METCARDIO), Italy b c
a r t i c l e
i n f o
Article history: Received 9 November 2013 Received in revised form 10 March 2014 Accepted 12 March 2014 Available online 21 March 2014 Keywords: PCI CABG Stroke Meta-regression
a b s t r a c t Aims: Percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG) perform similarly in terms of lowering mortality and myocardial infarction rates in patients with stable angina, except in subjects with high-risk lesions. PCI is burdened from higher rates of revascularization, but offers a reduction in stroke. To date, the impact of clinical variables on the risk-benefit assessment has not been established. Methods and results: Using event rates as a dependent variable, meta-regression was performed to test whether an interaction existed between baseline clinical features (age, gender, diabetes mellitus, previous myocardial infarction and ejection fraction) and choice of revascularization, focusing on death, myocardial infarction, repeat revascularization and stroke. 20 randomized clinical trials (RCT) including 12,844 patients with stable angina were included. Compared to CABG, PCI significantly reduced the risk of stroke, both at 30 days (odds ratio [OR] 0.36 [95% confidence interval: 0.20–0.62]) and at follow up (median = 12 months, OR = 0.57 [0.41–0.80]). This reduction in stroke was significantly higher in females (B = −0.12, p = 0.03). For repeat revascularization, PCI performed worse than CABG, both in the overall population and in patients with multivessel disease (OR = 4.71 [3.17–7.01]) and (OR = 7.18 [4.32–11.93]). Women (B = 3.4, p = 0.01) and those with diabetes mellitus (B = 1.8, p = 0.002) were at increased risk of subsequent revascularization after PCI. Conclusion: PCI significantly reduces the risk of stroke compared to CABG particularly in female patients: however the risk of revascularization is increased with PCI, especially in women and in those with diabetes. © 2014 Elsevier Inc. All rights reserved.
Abbreviations: CAD, coronary artery disease; DES, drug eluting stent; CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention; RCT, randomized controlled trial. ☆ Funding: None. ☆☆ Conflicts of interest: None. ⁎ Corresponding author at: Division of Cardiology, University of Turin, S. Giovanni Battista “Molinette” Hospital, Corso Bramante 88–90, 10126 Turin, Italy. E-mail address: [email protected] (F. D'Ascenzo).URL: http://www.cardiogroup.org (F. D'Ascenzo).
http://dx.doi.org/10.1016/j.cct.2014.03.002 1551-7144/© 2014 Elsevier Inc. All rights reserved.
52
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
1. Introduction
2.4. Internal validity and quality appraisal
Coronary artery disease (CAD) represents the most important cause of death in the world [1] with a detrimental impact on both patients' survival and quality of life. Surgical and percutaneous revascularizations are the two main options for these patients. Apart from high-risk lesions [2–6], PCI (Percutaneous coronary intervention) offers comparable results to CABG (Coronary Artery Bypass Graft) in terms of reducing the risk of death and myocardial infarction. On the contrary, CABG provides lower rates of subsequent revascularization, at the cost of a higher frequency of strokes [7], both in the peri-operative period and during long-term follow up. The choice of two different revascularization strategies (PCI and CABG) offers physicians the opportunity/challenge of balancing stroke with repeat revascularization [7,8]. In this setting, clinical features of patients help to drive decisions. Age, gender, presence of diabetes mellitus, previous ischemic events (such as myocardial infarction) and ejection fraction represent the most important predictors of prognosis in patients with CAD [9–16], and are exploited in most of the available surgical risk scores [17,18]. Moreover, in some subgroups of patients, these clinical features appear to influence outcomes according to revascularization choice [16], but to the best of our knowledge, their possible interaction both on risk (stroke) and on benefit (reducing death, myocardial infarction and revascularization) has not been previously assessed.
Unblinded, independent reviewers (GBZ, FDA) evaluated the quality of included studies on pre-specified forms, according to Cochrane's indications, reporting attrition, detection, and performance and selection bias [21].
2. Methods Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) and amendment to the Quality of Reporting of Meta-analyses (QUOROM) statement [19–21] were followed during elaboration of the manuscript.
2.1. Search strategy and study selection Two reviewers (TP; LA) independently searched Biomed Central, CENTRAL, and Medline/PubMed using the following keywords “coronary angioplasty, coronary artery bypass, singlevessel coronary artery disease, multivessel coronary artery disease, and left main coronary artery disease”. Studies were included according to the following criteria: (a) randomized design and (b) comparing percutaneous with surgical revascularization. Exclusion criteria were as follows: (a) duplicate publication and (b) not reporting clinical events.
2.2. Data extraction We appraised age, gender, extent of CAD (single vessel or multivessel or left main disease), options of surgical revascularization (on vs. off pump, rates of arterial graft use), diabetes, ejection fraction, previous myocardial infarction and stroke.
2.3. Outcome selection Selected end points were death, myocardial infarction, repeated revascularization and stroke.
2.5. Data analysis and synthesis Continuous variables were reported as mean (standard deviation) or median (range) and categorical variables as n (%). Funnel plot analysis was used to evaluate potential publication bias, and Cochran Q2 tests and I2 to investigate heterogeneity. Using rates of event as dependent variable, a meta-regression was performed to test whether an interaction between baseline clinical features (age, gender, diabetes mellitus, previous myocardial infarction and ejection fraction) and revascularization choice was present, appraising death, myocardial infarction, repeat revascularization and stroke as outcomes. Statistical analyses were performed with Comprehensive Meta-analysis [22] and Review Manager 4.2.4 [23]. 3. Results 2.189 potentially relevant articles were initially selected and 20 randomized controlled trials (RCTs) with 13.549 patients (including 6.749 assigned to CABG and 6.800 assigned to PCI) were included in the final meta-analysis (Fig. 1, Table 1; see online references.). The main features of patients in the included trials are reported in Tables 1 and 2. Nine trials enrolled patients with multivessel disease and four trials enrolled patients with unprotected left main disease, all presenting with stable angina. PCI significantly reduced the risk of stroke, both at 30 days (OR = 0.36; 95% CI: 0.20–0.62) and at follow up (OR = 0.57; 95% CI: 0.41–0.80) (see Fig. A, web appendix only), also when evaluating patients with multivessel disease (OR = 0.34, 95% CI: 0.19–0.61) and (OR = 0.49, 95% CI: 0.25–0.97). At follow up (median 12.1 months) a significant interaction between female gender and benefit offered from PCI was found (B = −0.12; p = 0.03; see Figs. 2 and 3). No significant difference was found for death, both for the overall population and for studies enrolling only those with multivessel disease ((OR = 0.99, 95% CI: 0.77–1.27) and (OR = 1.11, 95% CI: 0.87–1.43) respectively), and no significant interaction was noted (see Fig. B, web appendix, and Fig. 4). Neither PCI nor CABG performed superior in reducing acute myocardial infarction (OR = 1.03, 95% CI: 0.77–1.37). Similar findings were found for patients with multivessel disease (OR = 1.00; 95% CI: 0.72–1.39) (See Fig. C, Appendix, web only and Fig. 5). For repeat revascularization, PCI performed worse than CABG, both in the overall population and in patients with multivessel disease (OR = 4.71; 95% CI: 3.17–7.01) and (OR = 7.18, 95% CI: 4.32–11.93). Significant interactions between female gender (B = 3.4; p = 0.01) and diabetes mellitus (B = 1.8; p = 0.002; See Fig. D, Appendix, web only and Fig. 6) were found.
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
53
Fig. 1. Included and excluded studies.
4. Discussion The main results of the present study are 1) no significant differences regarding death and myocardial infarction are reported between PCI and CABG, and they are not affected by baseline features, 2) PCI reduces the risk of subsequent stroke, especially in women and 3) PCI is associated with a higher probability of a new revascularization procedure, especially in women and patients with diabetes.
Excluding high-risk lesions, PCI has achieved comparable results in terms of death and myocardial infarction to those of surgical revascularization. This was due to the improvements in procedural technique [24–26] (with increased use of intravascular ultrasound and fractional flow reserve, and new imaging techniques such as optimal coherence tomography), as well as the extensive use of drug eluting stents (DES), of the second and third generations. Furthermore, the prognostic benefit on survival of revascularization in patients with multi-vessel or left
Table 1 Randomized trials comparing coronary artery bypass graft surgery versus percutaneous coronary intervention. Study
Year of publication
N. of patients (CABG/PCI)
Extent of CAD
Stent use
Type of CABG
Arterial graft use
ARTS I (1) AWESOME (2) BARI (3) CARDIa (4) EAST (5) ERACI II (6) GABI (7) MASS II (8) SoS (9) SYNTAX (10) SYNTAX LMa (11) Budriot (12) Le Mans (13) PRECOMBAT (14) RITA (15) Groningen (16) Leipzig (17) Seoul (18) SIMA (19) FREEDOM (20)
2001 2001 1996 2010 1994 2001 1994 2004 2002 2009 2010 2011 2008 2011 1993 2002 2002 2005 2000 2012
1205 (605/600) 454 (232/222) 1829 (914/915) 510 (256/254) 392 (194/198) 450 (225/225) 359 (177/182) 408 (203/205) 988 (500/488) 1800 (897/903) 705 (348/357) 201 (101/100) 105 (53/52) 600 (300/300) 1011 (501/510) 102 (51/51) 220 (110/110) 189 (70/119) 121 (59/62) 1900 (953/947)
MVCAD 82% MVCAD MVCAD MVCAD MVCAD MVCAD MVCAD MVCAD MVCAD MVCAD ULMCAD ULMCAD ULMCAD ULMCAD 55% MVCAD SVCAD SVCAD SVCAD SVCAD MVCAD
100% 54% 0 100% 0 100% 0 68% 100% 100% 100% 100% 100% 100% 0 100% 100% 100% 100% 100%
On-pump On-pump On-pump 31% off-pump On-pump On-pump On-pump On-pump On-pump 15% off-pump NA 46% off-pump 2% off-pump 64% off pump On-pump Off-pump Off-pump Off-pump On-pump
93% 70% 82% 94% 86% 89% 37% 92% 93% 97% NA 99% 72% 94% 74% 100% 100% 100% 93% 94%
CABG denotes coronary artery bypass graft surgery; PCI denotes percutaneous coronary intervention, CAD denotes coronary artery disease, MVCAD denotes multivessel coronary artery disease, ULMCAD denotes unprotected left main coronary artery disease. a The SYNTAX left main (LM) trial was considered only in the analysis stratified by the extent of coronary artery disease.
54
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
Table 2 Baseline features (all variables are reported CABG/PCI). Study
Age (years)
Female gender (%)
Diabetes mellitus (%)
Prior myocardial infarction (%)
Ejection fraction (%)
Previous stroke/TIA for CABG/PCI (%)
Renal insufficiency for CABG/PCI (%)
ARTS I (1) AWESOME (2) BARI (3) CARDIa (4) EAST (5) ERACI II (6) GABI (7) MASS II (8) SoS (9) SYNTAX (10) SYNTAX LM* (11) Budriot (12) Le Mans (13) PRECOMBAT (14) RITA (15) Groningen (16) Leipzig (17) Seoul (18) SIMA (19) FREEDOM (20)
61 ± 9/61 ± 10 67/67 61.1/61.8 63.6 ± 9.1/64.3 ± 8.5 61.4 ± 10.0/61.8 ± 10.1 61.4 ± 10.1/62.5 ± 11.5 – 60 ± 9/60 ± 9 62 ± 9,5/61 ± 9,2 65.0 ± 9.8/65.2 ± 9.7 65.6 ± 10.1/65.4 ± 9.8 69/66 61.3 ± 8.4/60.6 ± 10.5 62.7 ± 9.5/61.8 ± 10.0 – 60 ± 1.6/61 ± 1.3 61.6 ± 10/62.5 ± 10.2 61.4 ± 9.9/60.5 ± 9.6 60/59 63.2 ± 8.9/63.1 ± 9.2
24/23 – 26/27 22.1/29.3 27.3/25.3 18.6/22.7 20/21 28/33 22/20 27.1/23.6 24.4/28 22%28 27/40 23/24 21.3/17.3 22/25 23/28 35.7/36.1 17/24 27/31
16/19 34/29 25/24 100/100 21.2/24.7 17.3/17.3 15/10 29/23 15/14 24.6/25.6 25.6/23.8 33/40 17/19 30/34 – 8/18 25/34 48.6/37 13/11 100
42/44 71/70 55/54 – 40.7/40.9 27.7/28.5 47/46 41/52 47/44 33.8/31.9 25.4/28.5 14/19 32/36 6.7/4.3 41.9/42.5 24/18 45/45 22.9/21.8 2/2 26/25
60 ± 13/61 ± 12 44/47 57.6/57.1 60.0 ± 12.7/59.1 ± 14.4 62.0 ± 11.8/60.8 ± 11.6 – – 67 ± 9/67 ± 8 57/57 – – 65.0/65.0 53.7 ± 6.7/53.5 ± 10.7 60.6 ± 8.5/61.7 ± 8.3 – – 63 ± 11/62 ± 15 – 67/67 65.7 ± 12.1/66.6 ± 10.5
– 14/9 – 5.6/3.5 – – 2/5 – 3/1 9.9/8.2 4.1/4.5 6/3 – – – – – – – –
– – – 4/5.5 – – – – – – 2.3/1.4 – – 0.3/1.3 – – – – – –
main disease is definitively assessed, regardless of the method used. This result is well established and demonstrated by many trials and meta-analyses [27–38], even after follow-up of more than 10 years [39]. In our study, no baseline clinical features influenced the risk of death and myocardial infarction. Previous evidence has shown that CABG increases the risk of stroke compared to PCI [7,39]. Strokes are associated with a
considerable increase in in-hospital and long-term mortality [39–43], and furthermore, are perceived as worse than or equivalent to death [42]. After surgical revascularization, postoperative respiratory failure, which is frequently reported in patients with small body surface area and heart failure, increases the risk of stroke due to prolonged mechanical ventilation, leading to decreased cerebral perfusion and oxygenation [44].
Fig. 2. Meta-regression for risk of stroke at 30 days (median 12.1 months). X axis represents the percentages of patients with each risk factors. Beta (B) is meta-regression coefficient, and p value (p) for interaction. Age (B = −0.02; p = 0.51). Female gender (B = −0.009; p = 0.91). Diabetes mellitus (B = −0.02; p = 0.81). Ejection fraction (B = −0.02; p = 0.24). Prior myocardial infarction (B = 0.02; p = 0.31).
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
55
Fig. 3. Meta-regression for risk of stroke at follow up (median 12.1 months). X axis represents the percentages of patients with each risk factors. Beta (B) is meta-regression coefficient, and p value (p) for interaction. Age (B = −0.03; p = 0.48). Female gender (B = −0.02; p = 0.025). Diabetes mellitus (B = −0.023; p = 0.69). Ejection fraction (B = 0.02; p = 0.59). Prior myocardial infarction (B = 0.01; p = 0.23).
Fig. 4. Meta-regression for risk of death at follow up (median 12.1 months). X axis represents the percentages of patients with each risk factors. Beta (B) is meta-regression coefficient, and p value (p) for interaction. Age (B = −0.03; p = 0.45). Female gender (B = 0.03; p = 0.65). Diabetes mellitus (B = −0.001; p = 0.49). Ejection fraction (B = 0.45; p = 0.12). Prior myocardial infarction (B = 0.45; p = 0.43).
56
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
Fig. 5. Meta-regression for risk of AMI at follow up (median 12.1 months). X axis represents the percentages of patients with each risk factors. Beta (B) is metaregression coefficient, and p value (p) for interaction. Age (B = 0.61; p = 0.45). Female gender (B = 0.03; p = 0.45). Diabetes mellitus (B = 2.3; p = 0.56). Ejection fraction (B = 0.35; p = 0.24). Prior myocardial infarction (B = 0.09; p = 0.7).
Fig. 6. Meta-regression for risk of revascularization at follow up (median 12.1 months). X axis represents the percentages of patients with each risk factors. Beta (B) is meta-regression coefficient, and p value (p) for interaction. Age (B = 0.08; p = 0.61). Female gender (B = −0.02; p = 0.68). Diabetes mellitus (B = 1.8; p = 0.02). Ejection fraction (B = 0.35; p = 0.24). Prior myocardial infarction (B = 0.02; p = 0.61).
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
Other possible reasons are represented by cerebral embolization, iatrogenic, mobilization of atherosclerotic plaques, air/fat embolism, hemodynamic fluctuations, and cerebral hyperthermia. Pertaining to PCI, thrombus formation on devices, displacement of intracoronary thrombus, air embolization, and a higher likelihood of periprocedural hypotension resulting in cerebral hypoperfusion [45–48] represent possible causes for strokes. In the present study, the reduction of stroke at follow up obtained with PCI is enhanced in female patients. Female gender has been widely described as an independent predictor for stroke after CABG [47,48]. As previously assessed, small body area and heart failure, which are typically more frequent in women, are the most important causes for postoperative respiratory failure. PCI is associated with a higher probability of repeat revascularization compared to CABG, especially in women and diabetic patients. Diabetes mellitus is a known, potent predictor of restenosis after PCI, conferring a 50% increase in risk for repeat revascularization [49]. The latest guidelines on myocardial revascularization warn about performing CABG for patients with multivessel coronary disease, generally seen in patients with diabetes [50]. The need for subsequent revascularization after PCI is enhanced in women. Gender-based differences in outcomes, with higher rates of in-hospital mortality and more frequent recurrent angina for women during long-term follow-up were described from the early stages of coronary angioplasty [51], with a progressive improvement over the advances in technology leading to smaller guiding and balloon catheters, the appearance of bare metal stents BMS and then of DES. Unfortunately, the data are limited by the fact that most of these large randomized trials have low numbers of women enrolled, frequently 30% or less. Contemporary studies however show that revascularization with PCI for stable CAD in women yields outcomes similar to men, particularly in regard to the risk of target vessel revascularization [51,52]. The present work has limitations. First, although deriving data from a large number of trials, meta-regression is considered by most authors as a hypothesis-generating tool and should not be taken at face value as a means to definitely test scientific hypotheses. Moreover, it has been reported that study characteristics are often strongly associated with each other, leading to collinearity [53], even if in the present work, collinearity was clinically present between two features, which are myocardial infarction and ejection fraction. Finally, as reported in Fig. f (web appendix only) the risk of bias was generally low for selection and performance bias, while it was often unclear for attrition bias. In conclusion, our study shows that diabetic patients may benefit from CABG rather than PCI due to a reduced risk of repeat revascularization. However, women face a lower risk of stroke with PCI at the cost of an increased risk of repeat revascularization. Well-designed clinical trials are needed, particularly in women and diabetics, to test if these interactions hold true. Appendix A. Supplementary data Web Appendix only a Online list of reference for included studies b Risk of stroke at 30 days (above) and at follow up (below)
c d e f
57
Risk of death Risk of myocardial infarction Risk or revascularization Risk of bias
Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.cct.2014.03.002.
References [1] De Bruyne B, Pijls NH, Kalesan B, Barbato E, Tonino PA, Piroth Z, et al. Fractional flow reserve-guided PCI versus medical therapy in stable coronary disease. N Engl J Med 2012;367:991–1001. [2] [No authors listed]Myocardial infarction and mortality in the coronary artery surgery study (CASS) randomized trial. N Engl J Med 1984:310750–8. [3] Morrison DA, Sethi G, Sacks J, Grover F, Sedlis S, Esposito R, et al. A multicenter, randomized trial of percutaneous coronary intervention versus bypass surgery in high-risk unstable angina patients. The AWESOME (Veterans Affairs Cooperative Study #385, angina with extremely serious operative mortality evaluation) investigators from the Cooperative Studies Program of the Department of Veterans Affairs. Control Clin Trials 1999;20:601–19. [4] Capodanno D, Stone GW, Morice MC, Bass TA, Tamburino C. Percutaneous coronary intervention versus coronary artery bypass graft surgery in left main coronary artery disease: a meta-analysis of randomized clinical data. J Am Coll Cardiol 2011;58:1426–32. [5] Kappetein AP, Feldman TE, Mack MJ, Morice MC, Holmes DR, Ståhle E, et al. Comparison of coronary bypass surgery with drug-eluting stenting for the treatment of left main and/or three-vessel disease: 3-year followup of the SYNTAX trial. Eur Heart J 2011;32:2125–34. [6] Onuma Y, Wykrzykowska JJ, Garg S, Vranckx P, Serruys PW, ARTS I and II Investigators. 5-Year follow-up of coronary revascularization in diabetic patients with multivessel coronary artery disease: insights from ARTS (arterial revascularization therapy study)-II and ARTS-I trials. JACC Cardiovasc Interv 2011;4:317–23. [7] Palmerini T, Biondi-Zoccai G, Reggiani LB, Sangiorgi D, Alessi L, De Servi S, et al. Risk of stroke with coronary artery bypass graft surgery compared with percutaneous coronary intervention. J Am Coll Cardiol 2012;60:798–805. [8] http://www.ema.europa.eu/docs/en_GB/document_library/Report/2011/ 07/WC500109477.pdf. [9] Steg PG, Greenlaw N, Tardif JC, Tendera M, Ford I, Kääb S, et al. Women and men with stable coronary artery disease have similar clinical outcomes: insights from the international prospective CLARIFY registry. Eur Heart J 2012;26. [10] Zhang Z, Weintraub WS, Mahoney EM, Spertus JA, Booth J, Nugara F, et al. Relative benefit of coronary artery bypass grafting versus stentassisted percutaneous coronary intervention for angina pectoris and multivessel coronary disease in women versus men (one-year results from the Stent or Surgery trial). Am J Cardiol 2004;93:404–9. [11] Sato H, Kasai T, Miyauchi K, Kubota N, Kajimoto K, Miyazaki T, et al. Long-term outcomes of women with coronary artery disease following complete coronary revascularization. J Cardiol 2011;58:158–64. [12] D'Ascenzo F, Gonella A, Quadri G, Longo G, Biondi-Zoccai G, Moretti C, et al. Comparison of mortality rates in women versus men presenting with ST-segment elevation myocardial infarction. Am J Cardiol 2011;107:651–4. [13] Weintraub WS, Grau-Sepulveda MV, Weiss JM, O'Brien SM, Peterson ED, Kolm P, et al. Comparative effectiveness of revascularization strategies. N Engl J Med 2012;19(366):1467–76. [14] Kirschbaum SW, Springeling T, Boersma E, Moelker A, van der Giessen WJ, Serruys PW, et al. Complete percutaneous revascularization for multivessel disease in patients with impaired left ventricular function: pre- and post-procedural evaluation by cardiac magnetic resonance imaging. JACC Cardiovasc Interv 2010;3:392–400. [15] Hee L, Mussap CJ, Yang L, Dignan R, Kadappu KK, Juergens CP, et al. Outcomes of coronary revascularization (percutaneous or bypass) in patients with diabetes mellitus and multivessel coronary disease. Am J Cardiol 2012;110:643–8. [16] Hlatky MA, Boothroyd DB, Bravata DM, Boersma E, Booth J, Brooks MM, et al. Coronary artery bypass surgery compared with percutaneous coronary interventions for multivessel disease: a collaborative analysis of individual patient data from ten randomised trials. Lancet 2009;4(373):1190–7. [17] http://riskcalc.sts.org/STSWebRiskCalc273/. [18] http://www.euroscore.org/.
58
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
[19] Moher D, Liberati A, Tetzlaff J, PRISMA Group, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009;21:339–41. [20] Moher D, Cook DJ, Eastwood S, et al. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of Reporting of Meta-analyses. Lancet 1999;354:1896–900. [21] Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.2 [updated September 2009]. In: Higgins JPT, Green S, editors. The Cochrane Collaboration; 2009 [Available from www.cochrane-handbook.org]. [22] Wallace Byron C, Schmid Christopher H, Lau Joseph, Trikalinos Thomas A. Meta-analyst: software for meta-analysis of binary, continuous and diagnostic data. BMC Med Res Methodol 2009;9(1):80. [23] The Cochrane Collaboration: RevMan 4.2 User Guide. Oxford: The Cochrane Collaboration; 2005 [http://www.cc-ims.net/RevMan/documentation.htm]. [24] Singh M, Rihal CS, Gersh BJ, et al. Twenty-five-year trends in in-hospital and long-term outcome after percutaneous coronary intervention: a single-institution experience. Circulation 2007;115:2835–41. [25] Singh M, Rihal CS, Gersh BJ, et al. Mortality differences between men and women after percutaneous coronary interventions. A 25-year, single-center experience. J Am Coll Cardiol 2008;51:2313–20. [26] Hilliard AA, From AM, Lennon RJ, et al. Percutaneous revascularization for stable coronary artery disease temporal trends and impact of drugeluting stents. J Am Coll Cardiol Intv 2010;3:172–9. [27] Bakhai A, Hill RA, Dundar Y, et al. Percutaneous coronary intervention with stents versus coronary artery bypass grafting for people with stable angina or acute coronary syndromes. Cochrane Database Syst Rev 2005: CD004588. [28] Coronary angioplasty versus coronary artery bypass surgery: the Randomized Intervention Treatment of Angina (RITA) trial. Lancet 1993;341:573–80. [29] Hamm CW, Reimers J, Ischinger T, et al. A randomized study of coronary angioplasty compared with bypass surgery in patients with symptomatic multivessel coronary disease. German Angioplasty Bypass Surgery Investigation (GABI). N Engl J Med 1994;331:1037–43. [30] King III SB, Lembo NJ, Weintraub WS, et al. A randomized trial comparing coronary angioplasty with coronary bypass surgery. Emory Angioplasty versus Surgery Trial (EAST). N Engl J Med 1994;331:1044–50. [31] The Bypass Angioplasty Revascularization Investigation (BARI) Investigators. Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. N Engl J Med 1996;335:217–25. [32] Serruys PW, Unger F, Sousa JE, et al. Comparison of coronary-artery bypass surgery and stenting for the treatment of multivessel disease. N Engl J Med 2001;344:1117–24. [33] Rosenberg MS. The file-drawer problem revisited: a general weighted method for calculating fail-safe numbers in meta-analysis. Int J Org Evol 2005;59:464–8. [34] Hueb WA, Bellotti G, de Oliveira SA, et al. The Medicine, Angioplasty or Surgery Study (MASS): a prospective, randomized trial of medical therapy, balloon angioplasty or bypass surgery for single proximal left anterior descending artery stenoses. J Am Coll Cardiol 1995;26:1600–5. [35] Hueb W, Lopes NH, Gersh BJ, et al. Five-year follow-up of the Medicine, Angioplasty, or Surgery Study (MASS II): a randomized controlled clinical trial of 3 therapeutic strategies for multivessel coronary artery disease. Circulation 2007;115:1082–9.
[36] Yousef ZR, Redwood SR, Bucknall CA, et al. Late intervention after anterior myocardial infarction: effects on left ventricular size, function, quality of life, and exercise tolerance: results of the Open Artery Trial (TOAT Study). J Am Coll Cardiol 2002;40:869–76. [37] Steg PG, Thuaire C, Himbert D, et al. DECOPI (DEsobstruction COronaire en Post-Infarctus): a randomized multi-centre trial of occluded artery angioplasty after acute myocardial infarction. Eur Heart J 2004;25:2187–94. [38] Bravata DM, Gienger AL, McDonald KM, et al. Systematic review: the comparative effectiveness of percutaneous coronary interventions and coronary artery bypass surgery. Ann Intern Med 2007;147:703–16. [39] Roach GW, Kanchuger M, Mangano CM, for the Multicenter Study of Perioperative Ischemia Research Group and the Ischemia Research and Education Foundation Investigators, et al. Adverse cerebral outcomes after coronary bypass surgery. N Engl J Med 1996;335:1857–63. [40] Stamou SC, Hill PC, Dangas G, et al. Stroke after coronary artery bypass: incidence, predictors, and clinical outcome. Stroke 2001;32:1508–13. [41] Hoffman SJ, Routledge HC, Lennon RJ, et al. Procedural factors associated with percutaneous coronary intervention-related ischemic stroke. J Am Coll Cardiol Interv 2012;5:200–6. [42] Solomon NA, Glick HA, Russo CJ, Lee J, Schulman KA. Patient preferences for stroke outcomes. Stroke 1994;25:1721–5. [43] D'Ancona G, de Ibarra JI Saez, Baillot R, Mathieu P, Doyle D, Metras J, et al. Determinants of stroke after coronary artery bypass grafting. Eur J Cardiothorac Surg 2003;2:552–6. [44] Charlesworth DC, Likosky DS, Marrin CA, Maloney CT, Quinton HB, Morton JR, et al. Development and validation of a prediction model for strokes after coronary artery bypass grafting. Ann Thorac Surg 2003;76:436–43 [20]. [45] John R, Choudhri AF, Weinberg AD, Ting W, Rose EA, Smith CR, et al. Multicenter review of preoperative risk factors for stroke after coronary artery bypass grafting. Ann Thorac Surg 2000;69:30–5. [46] Hogue CW, Murphy SF, Schechtman KB, vila-Roman VG. Risk factors for early or delayed stroke after cardiac surgery. Circulation 1999;100:642–7. [47] Alam M, et al. Association of gender with morbidity and mortality after isolated coronary artery bypass grafting. A propensity score matched analysis. Int J Cardiol 2012;167(1):180–4. [48] Eifert S, Kilian E, Beiras-Fernandez A, et al. Early and mid term mortality after coronary artery bypass grafting in women depends on the surgical protocol: retrospective analysis of 3441 on- and off-pump coronary artery bypass grafting procedures. J Cardiothorac Surg 2010;5:90. [49] Ben-Gal Y, Moshkovitz Y, Nesher N, Uretzky G, Braunstein R, Hendler A, et al. Drug-eluting stents versus coronary artery bypass grafting in patients with diabetes mellitus. Ann Thorac Surg 2006;82:1692–7. [50] Wijns W, Kolh P, Danchin N, for the The Task Force on Myocardial Revascularization of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS), et al. Guidelines on myocardial revascularization. Eur Heart J 2010;31:2501–55. [51] Stefanini GG, Kalesan B, Pilgrim T, Raber L, Onuma Y, Silber S, et al. Impact of gender on clinical and angiographic outcomes among patients undergoing revascularization with drug-eluting stents. JACC Cardiovasc Interv 2012;5(3):301–10. [52] Quadri G, D'ascenzo F, Bollati M, Moretti C, Omedé P, Sciuto F, et al. Diffuse coronary disease: short- and long-term outcome after percutaneous coronary intervention. Acta Cardiol Apr 2013;68(2):151–60. [53] Berlin JA, Antman EM. Advantages and limitations of metaanalytic regressions of clinical trials data. Online J Curr Clin Trials Jun 4 1994:1–3 [Doc No 134].
Contents lists available at ScienceDirect
Contemporary Clinical Trials journal homepage: www.elsevier.com/locate/conclintrial
Percutaneous coronary intervention versus coronary artery bypass graft for stable angina: Meta-regression of randomized trials☆,☆☆ Fabrizio D'Ascenzo a,g,⁎, Umberto Barbero a, Claudio Moretti a,g, Tullio Palmerini b, Diego Della Riva b, Andrea Mariani b, Pierluigi Omedè a, James J. DiNicolantonio e, Giuseppe Biondi-Zoccai c,d,f, Fiorenzo Gaita a a
Division of Cardiology, University of Turin, Italy Division of Cardiology, University of Bologna, Italy Department of Medico-Surgical Sciences and Biotechnologies, Italy d Sapienza University of Rome, Latina, Italy e Mid America Heart Institute at Saint Luke's Hospital, Kansas City, MO, United States f Meta-analysis and Evidence based medicine Training in Cardiology (METCARDIO), Italy b c
a r t i c l e
i n f o
Article history: Received 9 November 2013 Received in revised form 10 March 2014 Accepted 12 March 2014 Available online 21 March 2014 Keywords: PCI CABG Stroke Meta-regression
a b s t r a c t Aims: Percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG) perform similarly in terms of lowering mortality and myocardial infarction rates in patients with stable angina, except in subjects with high-risk lesions. PCI is burdened from higher rates of revascularization, but offers a reduction in stroke. To date, the impact of clinical variables on the risk-benefit assessment has not been established. Methods and results: Using event rates as a dependent variable, meta-regression was performed to test whether an interaction existed between baseline clinical features (age, gender, diabetes mellitus, previous myocardial infarction and ejection fraction) and choice of revascularization, focusing on death, myocardial infarction, repeat revascularization and stroke. 20 randomized clinical trials (RCT) including 12,844 patients with stable angina were included. Compared to CABG, PCI significantly reduced the risk of stroke, both at 30 days (odds ratio [OR] 0.36 [95% confidence interval: 0.20–0.62]) and at follow up (median = 12 months, OR = 0.57 [0.41–0.80]). This reduction in stroke was significantly higher in females (B = −0.12, p = 0.03). For repeat revascularization, PCI performed worse than CABG, both in the overall population and in patients with multivessel disease (OR = 4.71 [3.17–7.01]) and (OR = 7.18 [4.32–11.93]). Women (B = 3.4, p = 0.01) and those with diabetes mellitus (B = 1.8, p = 0.002) were at increased risk of subsequent revascularization after PCI. Conclusion: PCI significantly reduces the risk of stroke compared to CABG particularly in female patients: however the risk of revascularization is increased with PCI, especially in women and in those with diabetes. © 2014 Elsevier Inc. All rights reserved.
Abbreviations: CAD, coronary artery disease; DES, drug eluting stent; CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention; RCT, randomized controlled trial. ☆ Funding: None. ☆☆ Conflicts of interest: None. ⁎ Corresponding author at: Division of Cardiology, University of Turin, S. Giovanni Battista “Molinette” Hospital, Corso Bramante 88–90, 10126 Turin, Italy. E-mail address: [email protected] (F. D'Ascenzo).URL: http://www.cardiogroup.org (F. D'Ascenzo).
http://dx.doi.org/10.1016/j.cct.2014.03.002 1551-7144/© 2014 Elsevier Inc. All rights reserved.
52
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
1. Introduction
2.4. Internal validity and quality appraisal
Coronary artery disease (CAD) represents the most important cause of death in the world [1] with a detrimental impact on both patients' survival and quality of life. Surgical and percutaneous revascularizations are the two main options for these patients. Apart from high-risk lesions [2–6], PCI (Percutaneous coronary intervention) offers comparable results to CABG (Coronary Artery Bypass Graft) in terms of reducing the risk of death and myocardial infarction. On the contrary, CABG provides lower rates of subsequent revascularization, at the cost of a higher frequency of strokes [7], both in the peri-operative period and during long-term follow up. The choice of two different revascularization strategies (PCI and CABG) offers physicians the opportunity/challenge of balancing stroke with repeat revascularization [7,8]. In this setting, clinical features of patients help to drive decisions. Age, gender, presence of diabetes mellitus, previous ischemic events (such as myocardial infarction) and ejection fraction represent the most important predictors of prognosis in patients with CAD [9–16], and are exploited in most of the available surgical risk scores [17,18]. Moreover, in some subgroups of patients, these clinical features appear to influence outcomes according to revascularization choice [16], but to the best of our knowledge, their possible interaction both on risk (stroke) and on benefit (reducing death, myocardial infarction and revascularization) has not been previously assessed.
Unblinded, independent reviewers (GBZ, FDA) evaluated the quality of included studies on pre-specified forms, according to Cochrane's indications, reporting attrition, detection, and performance and selection bias [21].
2. Methods Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) and amendment to the Quality of Reporting of Meta-analyses (QUOROM) statement [19–21] were followed during elaboration of the manuscript.
2.1. Search strategy and study selection Two reviewers (TP; LA) independently searched Biomed Central, CENTRAL, and Medline/PubMed using the following keywords “coronary angioplasty, coronary artery bypass, singlevessel coronary artery disease, multivessel coronary artery disease, and left main coronary artery disease”. Studies were included according to the following criteria: (a) randomized design and (b) comparing percutaneous with surgical revascularization. Exclusion criteria were as follows: (a) duplicate publication and (b) not reporting clinical events.
2.2. Data extraction We appraised age, gender, extent of CAD (single vessel or multivessel or left main disease), options of surgical revascularization (on vs. off pump, rates of arterial graft use), diabetes, ejection fraction, previous myocardial infarction and stroke.
2.3. Outcome selection Selected end points were death, myocardial infarction, repeated revascularization and stroke.
2.5. Data analysis and synthesis Continuous variables were reported as mean (standard deviation) or median (range) and categorical variables as n (%). Funnel plot analysis was used to evaluate potential publication bias, and Cochran Q2 tests and I2 to investigate heterogeneity. Using rates of event as dependent variable, a meta-regression was performed to test whether an interaction between baseline clinical features (age, gender, diabetes mellitus, previous myocardial infarction and ejection fraction) and revascularization choice was present, appraising death, myocardial infarction, repeat revascularization and stroke as outcomes. Statistical analyses were performed with Comprehensive Meta-analysis [22] and Review Manager 4.2.4 [23]. 3. Results 2.189 potentially relevant articles were initially selected and 20 randomized controlled trials (RCTs) with 13.549 patients (including 6.749 assigned to CABG and 6.800 assigned to PCI) were included in the final meta-analysis (Fig. 1, Table 1; see online references.). The main features of patients in the included trials are reported in Tables 1 and 2. Nine trials enrolled patients with multivessel disease and four trials enrolled patients with unprotected left main disease, all presenting with stable angina. PCI significantly reduced the risk of stroke, both at 30 days (OR = 0.36; 95% CI: 0.20–0.62) and at follow up (OR = 0.57; 95% CI: 0.41–0.80) (see Fig. A, web appendix only), also when evaluating patients with multivessel disease (OR = 0.34, 95% CI: 0.19–0.61) and (OR = 0.49, 95% CI: 0.25–0.97). At follow up (median 12.1 months) a significant interaction between female gender and benefit offered from PCI was found (B = −0.12; p = 0.03; see Figs. 2 and 3). No significant difference was found for death, both for the overall population and for studies enrolling only those with multivessel disease ((OR = 0.99, 95% CI: 0.77–1.27) and (OR = 1.11, 95% CI: 0.87–1.43) respectively), and no significant interaction was noted (see Fig. B, web appendix, and Fig. 4). Neither PCI nor CABG performed superior in reducing acute myocardial infarction (OR = 1.03, 95% CI: 0.77–1.37). Similar findings were found for patients with multivessel disease (OR = 1.00; 95% CI: 0.72–1.39) (See Fig. C, Appendix, web only and Fig. 5). For repeat revascularization, PCI performed worse than CABG, both in the overall population and in patients with multivessel disease (OR = 4.71; 95% CI: 3.17–7.01) and (OR = 7.18, 95% CI: 4.32–11.93). Significant interactions between female gender (B = 3.4; p = 0.01) and diabetes mellitus (B = 1.8; p = 0.002; See Fig. D, Appendix, web only and Fig. 6) were found.
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
53
Fig. 1. Included and excluded studies.
4. Discussion The main results of the present study are 1) no significant differences regarding death and myocardial infarction are reported between PCI and CABG, and they are not affected by baseline features, 2) PCI reduces the risk of subsequent stroke, especially in women and 3) PCI is associated with a higher probability of a new revascularization procedure, especially in women and patients with diabetes.
Excluding high-risk lesions, PCI has achieved comparable results in terms of death and myocardial infarction to those of surgical revascularization. This was due to the improvements in procedural technique [24–26] (with increased use of intravascular ultrasound and fractional flow reserve, and new imaging techniques such as optimal coherence tomography), as well as the extensive use of drug eluting stents (DES), of the second and third generations. Furthermore, the prognostic benefit on survival of revascularization in patients with multi-vessel or left
Table 1 Randomized trials comparing coronary artery bypass graft surgery versus percutaneous coronary intervention. Study
Year of publication
N. of patients (CABG/PCI)
Extent of CAD
Stent use
Type of CABG
Arterial graft use
ARTS I (1) AWESOME (2) BARI (3) CARDIa (4) EAST (5) ERACI II (6) GABI (7) MASS II (8) SoS (9) SYNTAX (10) SYNTAX LMa (11) Budriot (12) Le Mans (13) PRECOMBAT (14) RITA (15) Groningen (16) Leipzig (17) Seoul (18) SIMA (19) FREEDOM (20)
2001 2001 1996 2010 1994 2001 1994 2004 2002 2009 2010 2011 2008 2011 1993 2002 2002 2005 2000 2012
1205 (605/600) 454 (232/222) 1829 (914/915) 510 (256/254) 392 (194/198) 450 (225/225) 359 (177/182) 408 (203/205) 988 (500/488) 1800 (897/903) 705 (348/357) 201 (101/100) 105 (53/52) 600 (300/300) 1011 (501/510) 102 (51/51) 220 (110/110) 189 (70/119) 121 (59/62) 1900 (953/947)
MVCAD 82% MVCAD MVCAD MVCAD MVCAD MVCAD MVCAD MVCAD MVCAD MVCAD ULMCAD ULMCAD ULMCAD ULMCAD 55% MVCAD SVCAD SVCAD SVCAD SVCAD MVCAD
100% 54% 0 100% 0 100% 0 68% 100% 100% 100% 100% 100% 100% 0 100% 100% 100% 100% 100%
On-pump On-pump On-pump 31% off-pump On-pump On-pump On-pump On-pump On-pump 15% off-pump NA 46% off-pump 2% off-pump 64% off pump On-pump Off-pump Off-pump Off-pump On-pump
93% 70% 82% 94% 86% 89% 37% 92% 93% 97% NA 99% 72% 94% 74% 100% 100% 100% 93% 94%
CABG denotes coronary artery bypass graft surgery; PCI denotes percutaneous coronary intervention, CAD denotes coronary artery disease, MVCAD denotes multivessel coronary artery disease, ULMCAD denotes unprotected left main coronary artery disease. a The SYNTAX left main (LM) trial was considered only in the analysis stratified by the extent of coronary artery disease.
54
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
Table 2 Baseline features (all variables are reported CABG/PCI). Study
Age (years)
Female gender (%)
Diabetes mellitus (%)
Prior myocardial infarction (%)
Ejection fraction (%)
Previous stroke/TIA for CABG/PCI (%)
Renal insufficiency for CABG/PCI (%)
ARTS I (1) AWESOME (2) BARI (3) CARDIa (4) EAST (5) ERACI II (6) GABI (7) MASS II (8) SoS (9) SYNTAX (10) SYNTAX LM* (11) Budriot (12) Le Mans (13) PRECOMBAT (14) RITA (15) Groningen (16) Leipzig (17) Seoul (18) SIMA (19) FREEDOM (20)
61 ± 9/61 ± 10 67/67 61.1/61.8 63.6 ± 9.1/64.3 ± 8.5 61.4 ± 10.0/61.8 ± 10.1 61.4 ± 10.1/62.5 ± 11.5 – 60 ± 9/60 ± 9 62 ± 9,5/61 ± 9,2 65.0 ± 9.8/65.2 ± 9.7 65.6 ± 10.1/65.4 ± 9.8 69/66 61.3 ± 8.4/60.6 ± 10.5 62.7 ± 9.5/61.8 ± 10.0 – 60 ± 1.6/61 ± 1.3 61.6 ± 10/62.5 ± 10.2 61.4 ± 9.9/60.5 ± 9.6 60/59 63.2 ± 8.9/63.1 ± 9.2
24/23 – 26/27 22.1/29.3 27.3/25.3 18.6/22.7 20/21 28/33 22/20 27.1/23.6 24.4/28 22%28 27/40 23/24 21.3/17.3 22/25 23/28 35.7/36.1 17/24 27/31
16/19 34/29 25/24 100/100 21.2/24.7 17.3/17.3 15/10 29/23 15/14 24.6/25.6 25.6/23.8 33/40 17/19 30/34 – 8/18 25/34 48.6/37 13/11 100
42/44 71/70 55/54 – 40.7/40.9 27.7/28.5 47/46 41/52 47/44 33.8/31.9 25.4/28.5 14/19 32/36 6.7/4.3 41.9/42.5 24/18 45/45 22.9/21.8 2/2 26/25
60 ± 13/61 ± 12 44/47 57.6/57.1 60.0 ± 12.7/59.1 ± 14.4 62.0 ± 11.8/60.8 ± 11.6 – – 67 ± 9/67 ± 8 57/57 – – 65.0/65.0 53.7 ± 6.7/53.5 ± 10.7 60.6 ± 8.5/61.7 ± 8.3 – – 63 ± 11/62 ± 15 – 67/67 65.7 ± 12.1/66.6 ± 10.5
– 14/9 – 5.6/3.5 – – 2/5 – 3/1 9.9/8.2 4.1/4.5 6/3 – – – – – – – –
– – – 4/5.5 – – – – – – 2.3/1.4 – – 0.3/1.3 – – – – – –
main disease is definitively assessed, regardless of the method used. This result is well established and demonstrated by many trials and meta-analyses [27–38], even after follow-up of more than 10 years [39]. In our study, no baseline clinical features influenced the risk of death and myocardial infarction. Previous evidence has shown that CABG increases the risk of stroke compared to PCI [7,39]. Strokes are associated with a
considerable increase in in-hospital and long-term mortality [39–43], and furthermore, are perceived as worse than or equivalent to death [42]. After surgical revascularization, postoperative respiratory failure, which is frequently reported in patients with small body surface area and heart failure, increases the risk of stroke due to prolonged mechanical ventilation, leading to decreased cerebral perfusion and oxygenation [44].
Fig. 2. Meta-regression for risk of stroke at 30 days (median 12.1 months). X axis represents the percentages of patients with each risk factors. Beta (B) is meta-regression coefficient, and p value (p) for interaction. Age (B = −0.02; p = 0.51). Female gender (B = −0.009; p = 0.91). Diabetes mellitus (B = −0.02; p = 0.81). Ejection fraction (B = −0.02; p = 0.24). Prior myocardial infarction (B = 0.02; p = 0.31).
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
55
Fig. 3. Meta-regression for risk of stroke at follow up (median 12.1 months). X axis represents the percentages of patients with each risk factors. Beta (B) is meta-regression coefficient, and p value (p) for interaction. Age (B = −0.03; p = 0.48). Female gender (B = −0.02; p = 0.025). Diabetes mellitus (B = −0.023; p = 0.69). Ejection fraction (B = 0.02; p = 0.59). Prior myocardial infarction (B = 0.01; p = 0.23).
Fig. 4. Meta-regression for risk of death at follow up (median 12.1 months). X axis represents the percentages of patients with each risk factors. Beta (B) is meta-regression coefficient, and p value (p) for interaction. Age (B = −0.03; p = 0.45). Female gender (B = 0.03; p = 0.65). Diabetes mellitus (B = −0.001; p = 0.49). Ejection fraction (B = 0.45; p = 0.12). Prior myocardial infarction (B = 0.45; p = 0.43).
56
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
Fig. 5. Meta-regression for risk of AMI at follow up (median 12.1 months). X axis represents the percentages of patients with each risk factors. Beta (B) is metaregression coefficient, and p value (p) for interaction. Age (B = 0.61; p = 0.45). Female gender (B = 0.03; p = 0.45). Diabetes mellitus (B = 2.3; p = 0.56). Ejection fraction (B = 0.35; p = 0.24). Prior myocardial infarction (B = 0.09; p = 0.7).
Fig. 6. Meta-regression for risk of revascularization at follow up (median 12.1 months). X axis represents the percentages of patients with each risk factors. Beta (B) is meta-regression coefficient, and p value (p) for interaction. Age (B = 0.08; p = 0.61). Female gender (B = −0.02; p = 0.68). Diabetes mellitus (B = 1.8; p = 0.02). Ejection fraction (B = 0.35; p = 0.24). Prior myocardial infarction (B = 0.02; p = 0.61).
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
Other possible reasons are represented by cerebral embolization, iatrogenic, mobilization of atherosclerotic plaques, air/fat embolism, hemodynamic fluctuations, and cerebral hyperthermia. Pertaining to PCI, thrombus formation on devices, displacement of intracoronary thrombus, air embolization, and a higher likelihood of periprocedural hypotension resulting in cerebral hypoperfusion [45–48] represent possible causes for strokes. In the present study, the reduction of stroke at follow up obtained with PCI is enhanced in female patients. Female gender has been widely described as an independent predictor for stroke after CABG [47,48]. As previously assessed, small body area and heart failure, which are typically more frequent in women, are the most important causes for postoperative respiratory failure. PCI is associated with a higher probability of repeat revascularization compared to CABG, especially in women and diabetic patients. Diabetes mellitus is a known, potent predictor of restenosis after PCI, conferring a 50% increase in risk for repeat revascularization [49]. The latest guidelines on myocardial revascularization warn about performing CABG for patients with multivessel coronary disease, generally seen in patients with diabetes [50]. The need for subsequent revascularization after PCI is enhanced in women. Gender-based differences in outcomes, with higher rates of in-hospital mortality and more frequent recurrent angina for women during long-term follow-up were described from the early stages of coronary angioplasty [51], with a progressive improvement over the advances in technology leading to smaller guiding and balloon catheters, the appearance of bare metal stents BMS and then of DES. Unfortunately, the data are limited by the fact that most of these large randomized trials have low numbers of women enrolled, frequently 30% or less. Contemporary studies however show that revascularization with PCI for stable CAD in women yields outcomes similar to men, particularly in regard to the risk of target vessel revascularization [51,52]. The present work has limitations. First, although deriving data from a large number of trials, meta-regression is considered by most authors as a hypothesis-generating tool and should not be taken at face value as a means to definitely test scientific hypotheses. Moreover, it has been reported that study characteristics are often strongly associated with each other, leading to collinearity [53], even if in the present work, collinearity was clinically present between two features, which are myocardial infarction and ejection fraction. Finally, as reported in Fig. f (web appendix only) the risk of bias was generally low for selection and performance bias, while it was often unclear for attrition bias. In conclusion, our study shows that diabetic patients may benefit from CABG rather than PCI due to a reduced risk of repeat revascularization. However, women face a lower risk of stroke with PCI at the cost of an increased risk of repeat revascularization. Well-designed clinical trials are needed, particularly in women and diabetics, to test if these interactions hold true. Appendix A. Supplementary data Web Appendix only a Online list of reference for included studies b Risk of stroke at 30 days (above) and at follow up (below)
c d e f
57
Risk of death Risk of myocardial infarction Risk or revascularization Risk of bias
Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.cct.2014.03.002.
References [1] De Bruyne B, Pijls NH, Kalesan B, Barbato E, Tonino PA, Piroth Z, et al. Fractional flow reserve-guided PCI versus medical therapy in stable coronary disease. N Engl J Med 2012;367:991–1001. [2] [No authors listed]Myocardial infarction and mortality in the coronary artery surgery study (CASS) randomized trial. N Engl J Med 1984:310750–8. [3] Morrison DA, Sethi G, Sacks J, Grover F, Sedlis S, Esposito R, et al. A multicenter, randomized trial of percutaneous coronary intervention versus bypass surgery in high-risk unstable angina patients. The AWESOME (Veterans Affairs Cooperative Study #385, angina with extremely serious operative mortality evaluation) investigators from the Cooperative Studies Program of the Department of Veterans Affairs. Control Clin Trials 1999;20:601–19. [4] Capodanno D, Stone GW, Morice MC, Bass TA, Tamburino C. Percutaneous coronary intervention versus coronary artery bypass graft surgery in left main coronary artery disease: a meta-analysis of randomized clinical data. J Am Coll Cardiol 2011;58:1426–32. [5] Kappetein AP, Feldman TE, Mack MJ, Morice MC, Holmes DR, Ståhle E, et al. Comparison of coronary bypass surgery with drug-eluting stenting for the treatment of left main and/or three-vessel disease: 3-year followup of the SYNTAX trial. Eur Heart J 2011;32:2125–34. [6] Onuma Y, Wykrzykowska JJ, Garg S, Vranckx P, Serruys PW, ARTS I and II Investigators. 5-Year follow-up of coronary revascularization in diabetic patients with multivessel coronary artery disease: insights from ARTS (arterial revascularization therapy study)-II and ARTS-I trials. JACC Cardiovasc Interv 2011;4:317–23. [7] Palmerini T, Biondi-Zoccai G, Reggiani LB, Sangiorgi D, Alessi L, De Servi S, et al. Risk of stroke with coronary artery bypass graft surgery compared with percutaneous coronary intervention. J Am Coll Cardiol 2012;60:798–805. [8] http://www.ema.europa.eu/docs/en_GB/document_library/Report/2011/ 07/WC500109477.pdf. [9] Steg PG, Greenlaw N, Tardif JC, Tendera M, Ford I, Kääb S, et al. Women and men with stable coronary artery disease have similar clinical outcomes: insights from the international prospective CLARIFY registry. Eur Heart J 2012;26. [10] Zhang Z, Weintraub WS, Mahoney EM, Spertus JA, Booth J, Nugara F, et al. Relative benefit of coronary artery bypass grafting versus stentassisted percutaneous coronary intervention for angina pectoris and multivessel coronary disease in women versus men (one-year results from the Stent or Surgery trial). Am J Cardiol 2004;93:404–9. [11] Sato H, Kasai T, Miyauchi K, Kubota N, Kajimoto K, Miyazaki T, et al. Long-term outcomes of women with coronary artery disease following complete coronary revascularization. J Cardiol 2011;58:158–64. [12] D'Ascenzo F, Gonella A, Quadri G, Longo G, Biondi-Zoccai G, Moretti C, et al. Comparison of mortality rates in women versus men presenting with ST-segment elevation myocardial infarction. Am J Cardiol 2011;107:651–4. [13] Weintraub WS, Grau-Sepulveda MV, Weiss JM, O'Brien SM, Peterson ED, Kolm P, et al. Comparative effectiveness of revascularization strategies. N Engl J Med 2012;19(366):1467–76. [14] Kirschbaum SW, Springeling T, Boersma E, Moelker A, van der Giessen WJ, Serruys PW, et al. Complete percutaneous revascularization for multivessel disease in patients with impaired left ventricular function: pre- and post-procedural evaluation by cardiac magnetic resonance imaging. JACC Cardiovasc Interv 2010;3:392–400. [15] Hee L, Mussap CJ, Yang L, Dignan R, Kadappu KK, Juergens CP, et al. Outcomes of coronary revascularization (percutaneous or bypass) in patients with diabetes mellitus and multivessel coronary disease. Am J Cardiol 2012;110:643–8. [16] Hlatky MA, Boothroyd DB, Bravata DM, Boersma E, Booth J, Brooks MM, et al. Coronary artery bypass surgery compared with percutaneous coronary interventions for multivessel disease: a collaborative analysis of individual patient data from ten randomised trials. Lancet 2009;4(373):1190–7. [17] http://riskcalc.sts.org/STSWebRiskCalc273/. [18] http://www.euroscore.org/.
58
F. D'Ascenzo et al. / Contemporary Clinical Trials 38 (2014) 51–58
[19] Moher D, Liberati A, Tetzlaff J, PRISMA Group, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009;21:339–41. [20] Moher D, Cook DJ, Eastwood S, et al. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of Reporting of Meta-analyses. Lancet 1999;354:1896–900. [21] Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.2 [updated September 2009]. In: Higgins JPT, Green S, editors. The Cochrane Collaboration; 2009 [Available from www.cochrane-handbook.org]. [22] Wallace Byron C, Schmid Christopher H, Lau Joseph, Trikalinos Thomas A. Meta-analyst: software for meta-analysis of binary, continuous and diagnostic data. BMC Med Res Methodol 2009;9(1):80. [23] The Cochrane Collaboration: RevMan 4.2 User Guide. Oxford: The Cochrane Collaboration; 2005 [http://www.cc-ims.net/RevMan/documentation.htm]. [24] Singh M, Rihal CS, Gersh BJ, et al. Twenty-five-year trends in in-hospital and long-term outcome after percutaneous coronary intervention: a single-institution experience. Circulation 2007;115:2835–41. [25] Singh M, Rihal CS, Gersh BJ, et al. Mortality differences between men and women after percutaneous coronary interventions. A 25-year, single-center experience. J Am Coll Cardiol 2008;51:2313–20. [26] Hilliard AA, From AM, Lennon RJ, et al. Percutaneous revascularization for stable coronary artery disease temporal trends and impact of drugeluting stents. J Am Coll Cardiol Intv 2010;3:172–9. [27] Bakhai A, Hill RA, Dundar Y, et al. Percutaneous coronary intervention with stents versus coronary artery bypass grafting for people with stable angina or acute coronary syndromes. Cochrane Database Syst Rev 2005: CD004588. [28] Coronary angioplasty versus coronary artery bypass surgery: the Randomized Intervention Treatment of Angina (RITA) trial. Lancet 1993;341:573–80. [29] Hamm CW, Reimers J, Ischinger T, et al. A randomized study of coronary angioplasty compared with bypass surgery in patients with symptomatic multivessel coronary disease. German Angioplasty Bypass Surgery Investigation (GABI). N Engl J Med 1994;331:1037–43. [30] King III SB, Lembo NJ, Weintraub WS, et al. A randomized trial comparing coronary angioplasty with coronary bypass surgery. Emory Angioplasty versus Surgery Trial (EAST). N Engl J Med 1994;331:1044–50. [31] The Bypass Angioplasty Revascularization Investigation (BARI) Investigators. Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. N Engl J Med 1996;335:217–25. [32] Serruys PW, Unger F, Sousa JE, et al. Comparison of coronary-artery bypass surgery and stenting for the treatment of multivessel disease. N Engl J Med 2001;344:1117–24. [33] Rosenberg MS. The file-drawer problem revisited: a general weighted method for calculating fail-safe numbers in meta-analysis. Int J Org Evol 2005;59:464–8. [34] Hueb WA, Bellotti G, de Oliveira SA, et al. The Medicine, Angioplasty or Surgery Study (MASS): a prospective, randomized trial of medical therapy, balloon angioplasty or bypass surgery for single proximal left anterior descending artery stenoses. J Am Coll Cardiol 1995;26:1600–5. [35] Hueb W, Lopes NH, Gersh BJ, et al. Five-year follow-up of the Medicine, Angioplasty, or Surgery Study (MASS II): a randomized controlled clinical trial of 3 therapeutic strategies for multivessel coronary artery disease. Circulation 2007;115:1082–9.
[36] Yousef ZR, Redwood SR, Bucknall CA, et al. Late intervention after anterior myocardial infarction: effects on left ventricular size, function, quality of life, and exercise tolerance: results of the Open Artery Trial (TOAT Study). J Am Coll Cardiol 2002;40:869–76. [37] Steg PG, Thuaire C, Himbert D, et al. DECOPI (DEsobstruction COronaire en Post-Infarctus): a randomized multi-centre trial of occluded artery angioplasty after acute myocardial infarction. Eur Heart J 2004;25:2187–94. [38] Bravata DM, Gienger AL, McDonald KM, et al. Systematic review: the comparative effectiveness of percutaneous coronary interventions and coronary artery bypass surgery. Ann Intern Med 2007;147:703–16. [39] Roach GW, Kanchuger M, Mangano CM, for the Multicenter Study of Perioperative Ischemia Research Group and the Ischemia Research and Education Foundation Investigators, et al. Adverse cerebral outcomes after coronary bypass surgery. N Engl J Med 1996;335:1857–63. [40] Stamou SC, Hill PC, Dangas G, et al. Stroke after coronary artery bypass: incidence, predictors, and clinical outcome. Stroke 2001;32:1508–13. [41] Hoffman SJ, Routledge HC, Lennon RJ, et al. Procedural factors associated with percutaneous coronary intervention-related ischemic stroke. J Am Coll Cardiol Interv 2012;5:200–6. [42] Solomon NA, Glick HA, Russo CJ, Lee J, Schulman KA. Patient preferences for stroke outcomes. Stroke 1994;25:1721–5. [43] D'Ancona G, de Ibarra JI Saez, Baillot R, Mathieu P, Doyle D, Metras J, et al. Determinants of stroke after coronary artery bypass grafting. Eur J Cardiothorac Surg 2003;2:552–6. [44] Charlesworth DC, Likosky DS, Marrin CA, Maloney CT, Quinton HB, Morton JR, et al. Development and validation of a prediction model for strokes after coronary artery bypass grafting. Ann Thorac Surg 2003;76:436–43 [20]. [45] John R, Choudhri AF, Weinberg AD, Ting W, Rose EA, Smith CR, et al. Multicenter review of preoperative risk factors for stroke after coronary artery bypass grafting. Ann Thorac Surg 2000;69:30–5. [46] Hogue CW, Murphy SF, Schechtman KB, vila-Roman VG. Risk factors for early or delayed stroke after cardiac surgery. Circulation 1999;100:642–7. [47] Alam M, et al. Association of gender with morbidity and mortality after isolated coronary artery bypass grafting. A propensity score matched analysis. Int J Cardiol 2012;167(1):180–4. [48] Eifert S, Kilian E, Beiras-Fernandez A, et al. Early and mid term mortality after coronary artery bypass grafting in women depends on the surgical protocol: retrospective analysis of 3441 on- and off-pump coronary artery bypass grafting procedures. J Cardiothorac Surg 2010;5:90. [49] Ben-Gal Y, Moshkovitz Y, Nesher N, Uretzky G, Braunstein R, Hendler A, et al. Drug-eluting stents versus coronary artery bypass grafting in patients with diabetes mellitus. Ann Thorac Surg 2006;82:1692–7. [50] Wijns W, Kolh P, Danchin N, for the The Task Force on Myocardial Revascularization of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS), et al. Guidelines on myocardial revascularization. Eur Heart J 2010;31:2501–55. [51] Stefanini GG, Kalesan B, Pilgrim T, Raber L, Onuma Y, Silber S, et al. Impact of gender on clinical and angiographic outcomes among patients undergoing revascularization with drug-eluting stents. JACC Cardiovasc Interv 2012;5(3):301–10. [52] Quadri G, D'ascenzo F, Bollati M, Moretti C, Omedé P, Sciuto F, et al. Diffuse coronary disease: short- and long-term outcome after percutaneous coronary intervention. Acta Cardiol Apr 2013;68(2):151–60. [53] Berlin JA, Antman EM. Advantages and limitations of metaanalytic regressions of clinical trials data. Online J Curr Clin Trials Jun 4 1994:1–3 [Doc No 134].
