Abstract
Diabetic patients have been associated with poor procedural and long term outcome if they were treated either with percutaneous coronary interventions or coronary artery bypass surgery.
Recently several randomized clinical trials (RCT) in this subset of patients have been published showing a greater incidence of major adverse cardiovascular events, death/myocardial infarction/stroke, if they were treated with first generation drug eluting stents (DES) which was not observed previously in the bare metal stent era.
However, almost simultaneously with this data, several RCT demonstrated better safety profile with new generation DES including biocompatible polymers, biodegradable polymers and lately complete absorbable DES, all of them showed reduction in adverse cardiac events compared to 1st generation DES in patients with diabetes.
In this editorial we review the old and new randomized data in diabetic patients and conclude that there are many unresolved issues to make a definitive statement regarding which is the best revascularization preference in diabetic patients and the measured final efficacy of PCI and CABG will not be reached until the arrival of RCT using next generation DES, including complete absorbable scaffolds.
Highlights
- •
Diabetic patients have been associated with poor procedural and long term outcome if they were treated either with percutaneous coronary interventions or coronary artery bypass surgery. Recently several randomized clinical trials (RCT) in this subset of patients have been published showing a greater incidence of major adverse cardiovascular events, death/myocardial infarction/stroke, if they were treated with first generation drug eluting stents (DES) which was not observed previously in the bare metal stent era.
- •
In recent years, several RCT demonstrated better safety profile with new generation DES including biocompatible polymers, biodegradable polymers and lately complete absorbable DES, all of them showed reduction in adverse cardiac events compared to 1st generation DES in patients with diabetes.
- •
First generation DES have been largely replaced by the safer and newer generations of DES, therefore, measured final efficacy of PCI and CABG will not be reached until the arrival of RCT using next generation DES, including complete absorbable DES.
Percutaneous coronary intervention (PCI) in diabetic patients has been associated with poorer outcomes compared to coronary artery by-pass graft (CABG) since the beginning of this therapeutic option for the treatment of coronary artery disease (CAD). This fact was not modified by the introduction of 1st generation drug eluting stents (DES) .
Furthermore, the largest trial ever performed in this cohort of patients, FREEDOM Trial (comparison of two treatments for multivessel coronary artery disease in individuals with diabetes), showed a significant lower incidence of the primary endpoint—the composite of death, myocardial infarction (MI) and cerebrovascular accident (CVA)—at 4-year follow-up if patients were treated with CABG .
In an accompanying editorial in the journal where the trial was published, the author stated that all diabetic patients had to be treated with CABG and all patients were to be aware of the benefits of the surgical procedure, signing an informed consent after a coronary angiography .
We recognize that diabetic patients with coronary artery disease were and are a therapeutic challenge to PCI. However, in spite of negative long term results from large randomized clinical trials (RCT), we consider that we still do not have enough data to close the door on PCI; furthermore, there are certain clinical situations where PCI with stent implantation should be a revascularization option.
Briefly, we will expound the data in which we base our assumptions and contextualize when and on whom PCI could be performed:
1
Results on diabetics in the youngest tertile are similar between PCI and CABG
A meta-analysis from 10 RCT showed that, over a median follow-up of 5.9 years, the effect of CABG vs. PCI on mortality varied according to age with an adjusted CABG:PCI hazard ratios of 1.23 (0.95–1.59) in the youngest tertile, 0.89 (0.73–1.10) in the middle tertile, and 0.79 (0.67–0.94) in the oldest tertile, with a CABG:PCI HR < 1 (favor to CABG) for patients aged 59 and older. In such analysis, younger diabetic patients had similar 5.9-year survival independently to revascularization treatment, 19% with CABG and PCI: HR. 0.89 (0.55–1.46) p = 0.65 unadjusted and 0.93 (0.55–1.55) p = 0.77 adjusted.
2
Meta-analysis of RCT comparing bare metal stents (BMS) vs. CABG—such as ARTS, ERACI II, MASS II and SoS—showed similar findings at 5 years, even in diabetic patients
In this meta-analysis performed by Daemen et al , for the composite end point of death, MI and CVA, no heterogeneity in treatment was found between patients with diabetes and those without diabetes (p for interaction = 0.65). In diabetics, the cumulative incidence of death was not significantly higher in the PCI group, 12.4% compared to 7.9% in the CABG group (p = 0.09), and the cumulative incidence of death, MI or CVA, the primary end point, was similar after PCI with BMS vs. CABG (21.4% vs. 20.9%, respectively; p = 0.9). The only significant difference was a threefold incidence of repeat revascularization in the PCI group; HR increase for repeat revascularization was 0.18 (95% CI, 0.11 to 0.29) in the PCI group . Additionally, if we compare results from this meta-analysis with those obtained in RCT between 1st generation DES vs. CABG, such as SYNTAX and FREEDOM, an indirect comparison suggested that 1st generation DES had a greater number of death/MI rate than BMS in spite of a significant reduction of repeat revascularization procedures. Moreover, this lack of benefit with 1st generation DES was seen in diabetic and non-diabetic population, in fact, the observed incidence in the composite of death, MI and CVA in FREEDOM was 20% higher than previously seen in diabetic patients randomized in BMS/CABG trials .
These findings imply that if we used a wrong stent device to treat diabetic patients in the above mentioned trials it would be, perhaps, the main limitation to search potential benefits of PCI in these trials.
2
Meta-analysis of RCT comparing bare metal stents (BMS) vs. CABG—such as ARTS, ERACI II, MASS II and SoS—showed similar findings at 5 years, even in diabetic patients
In this meta-analysis performed by Daemen et al , for the composite end point of death, MI and CVA, no heterogeneity in treatment was found between patients with diabetes and those without diabetes (p for interaction = 0.65). In diabetics, the cumulative incidence of death was not significantly higher in the PCI group, 12.4% compared to 7.9% in the CABG group (p = 0.09), and the cumulative incidence of death, MI or CVA, the primary end point, was similar after PCI with BMS vs. CABG (21.4% vs. 20.9%, respectively; p = 0.9). The only significant difference was a threefold incidence of repeat revascularization in the PCI group; HR increase for repeat revascularization was 0.18 (95% CI, 0.11 to 0.29) in the PCI group . Additionally, if we compare results from this meta-analysis with those obtained in RCT between 1st generation DES vs. CABG, such as SYNTAX and FREEDOM, an indirect comparison suggested that 1st generation DES had a greater number of death/MI rate than BMS in spite of a significant reduction of repeat revascularization procedures. Moreover, this lack of benefit with 1st generation DES was seen in diabetic and non-diabetic population, in fact, the observed incidence in the composite of death, MI and CVA in FREEDOM was 20% higher than previously seen in diabetic patients randomized in BMS/CABG trials .
These findings imply that if we used a wrong stent device to treat diabetic patients in the above mentioned trials it would be, perhaps, the main limitation to search potential benefits of PCI in these trials.
3
Threefold increased incidence of CVA with CABG
CABG was associated with higher incidence of CVA at 5-year follow-up in all randomized trials, RR 1.72 (1.18–2.53). This finding was particularly observed in diabetics .
In the FREEDOM trial, diabetics had a significant greater incidence of CVA:RR 2.16 (1.27–3.69). Furthermore, in FREEDOM, the 5-year increased CVA rate following CABG compared to PCI was independent of diabetic status 7.5% vs. 3.7% for CABG and PCI in insulin treated diabetics (ITDM) and 4.3% vs. 1.7% for CABG and PCI in non-insulin treated diabetic patients (non-ITDM) . Similarly, at 5 years, the SYNTAX trial non-ITDM patients had a 5.2% non-fatal stroke rate which was higher than the 1.6% incidence seen in the DES arm (p = 0.09), although it is not significant due to the small sample size .
Therefore, due to the reduced sample size of ITDM patients in both FREEDOM and SYNTAX, it would be impossible to determine significant differences between the outcomes of PCI and CABG in those patients; in FREEDOM composite of death/MI/CVA, PCI vs. CABG HR was 1.21 [0.87–1.69] .
4
FREEDOM showed significant differences in favor to CABG only in USA and Canada sites
This is a crucial point: not all centers performed either PCI or CABG in the same fashion. This is also true in other RCT such as SYNTAX where there were also large differences among sites .
In FREEDOM, primary end point was reached in favor of CABG only in USA and Canada sites, death/MI/CVA was 16% and 28% with DES and CABG respectively, from a population of 770 patients. Conversely, non-USA and Canada sites, from a population of 1130 patients, primary end point was 25% with DES and 21% with CABG, not statistically significant, with p = 0.05 for interaction between North American and outside North American sites .
In the past, certain RCT showed lower incidence of death with PCI at one year compared to CABG, and this difference was less significant but remained in favor of PCI at 5-year follow-up .
In fact, two studies from South America—ERACI II and MASS II—showed similar mortality with both revascularization strategies in diabetic patients treated either with BMS or CABG [RR 1 (0.27–3.72) and 0.95 (0.41–2.22) in ERACI II and MASS II, respectively] . In contrast, other trials conducted in Great Britain and Europe showed significant survival advantage with CABG at 6 years: RR 0.13 (0.02–1.04) , and the advantage in the latter was driven for an incidence of in-hospital mortality with CABG less than 1%.
Thus, old—ERACI, MASS and SoS—and new RCT—FREEDOM and SYNTAX—are consistent to show geographic disparities in the results between PCI and CABG.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
