PCI for 3-vessel disease

Background

Coronary artery disease (CAD) is one of the leading causes of death in Europe. Various survival indices have demonstrated that patients with CAD have significantly impaired survival. One of the easiest risk scores which predict survival in CAD is based on the number of diseased vessels. The CASS registry has demonstrated that patients with a normal coronary morphology have an estimated 12 year survival rate of 91% in contrast to 74% in patients with 1 vessel disease (VD), 59% in 2-VD, and 50% in 3-VD (1). Besides consequent lifestyle modification, two accepted strategies for revascularization in symptomatic CAD are currently available:  1) percutaneous coronary intervention (PCI) and 2) coronary artery bypass grafting (CABG). Here we will discuss the current literature evaluating advantages and disadvantages of a medical treatment, PCI, or CABG in patients with 3-VD. 

1 - Percutanous coronary intervention versus medical treatment

Various studies in recent years have compared PCI with medical treatment. Due to fast medical and technical progress in the last couple of years (introduction of drug eluting stents/ballons, effective dual platelet inhibition, stent platform improvements etc.) only a few studies are available comparing “state of the art” PCI with “state of the art” medical treatment. The recently published COURAGE study investigated in 2287 patients with stable CAD whether optimal medical treatment with PCI vs. medical treatment alone is comparable or even better (2). In this study, the conservative and the PCI arm showed similar numbers of coronary events (death and myocardial infarction) which was used to argue against PCI and pro medical treatment in patients with stable 3-VD. However, the number of asymptomatic patients after 1 and 3 years was significantly higher in the PCI group compared to medical treatment alone. Interestingly, nearly 33% of initially medically treated patients needed PCI within the observational time frame of the study (7 years).

Furthermore, the patient selection was very rigorous (and mainly restricted to Veterans Affairs Hospitals) with a large number of patients screened but not included into the study which may imply a selection bias at least comparing to “real world” conditions. One of the main therapeutic goals in treating multi-vessel disease on top of improving prognosis is to get patients symptom-free. The results of the quality of life analysis of the COURAGE study demonstrated that the severity and frequency of angina significantly improved within the first 36 months in the PCI arm compared to patients with medical treatment alone (3). Patients with frequent, weekly episodes of angina especially showed symptomatic improvement.

2 - Angioplasty vs. coronary artery bypass grafting

The vast majority of studies published until today have addressed single angioplasty without stent implantation with CABG (e.g. BARI, CAPRI, RITA, EAST, GABI, ERACI) (fig.1). This approach is no longer the gold standard in PCI. Therefore, these studies will not be further discussed. Systematic meta-analyses point towards the fact that neither angioplasty nor CABG is associated with a survival benefit4.

3 - Percutanous coronary intervention with bare metal and drug eluting stents versus CABG

Four randomised trials have compared PCI with bare metal stents (BMS) vs. CABG (ARTS I, SOS, ERACI II, MASS II, AWESOME) (5). In these studies as well as in meta-analyses, a significant improvement in severity and frequency of angina and lower rates of target vessel revascularisation were observed in the CABG treated patients compared to PCI treated patients. However, the rate of strokes was significantly increased in CABG patients. Overall, no survival benefit was observed in patients treated with CABG compared to PCI (4). Interestingly, the recently published metaanalysis of the ARTS I, SOS, ERACI II and MASS II study which was the first to include long-term follow-up data, did not show any significant differences in the rates of death, cardiovascular death, myocardial infarction, cerebrovascular events and angina (fig. 1). However, again, higher rates of target lesion revascularizations were observed in the PCI group. Two randomized studies (ARTS II and ERACI III6) determined the efficacy of DES compared to CABG in the setting of stable multi-vessel disease. Both studies showed comparable levels of target vessel revascularization in the PCI but also in the CABG group. Major cardiovascular and cerebrovascular events (MACCE) were 22.7% in the PCI group and not significantly different to CABG treated patients (ERACI III6).

In this line of evidence, published registry data points towards comparable mortality and myocardial infarction rates in both percutanous coronary intervention with bare metal and drug eluting stents or CABG

Registry data are observational and imply a higher probability of treatment bias. On the other hand, registries provide important real-world data and often reflect clinical every day life. A recently published prospective registry from 8 hospitals in the US including 4336 patients showed higher revascularization rates in the PCI group compared to CABG with comparable mortality after 18 months (7). Surprisingly, no difference was observed between BMS and DES and CABG patients treated conventionally and off-pump which questions the validity of registry data.

Deeper insights as to whether to treat patients with stable 3-VD with CABG or PCI was given in the SYNTAX trial, a non-inferiority comparison of PCI vs. CABG which was presented at the European Society of Cardiology (ESC) Congress in Munich 2008 and published in the New England Journal of Medicine in February 2009.
In this trial, 1800 patients with 3-VD or left main coronary artery disease were randomly assigned to undergo CABG or PCI with DES after a team of cardiac surgeons and interventional cardiologists had determined that equivalent complete revascularization could be achieved with both procedures. The primary end point was a major adverse cardiac or cerebrovascular event (death from any cause, stroke, myocardial infarction, or revascularization) during the 12-month period after randomisation. In order to mimic real-world conditions, the study design allowed the follow-up of additional patients for whom only one of the two treatment options due to anatomical reasons or clinical conditions were suitable. These patients were followed in a parallel, nested CABG or PCI registry.

The SYNTAX trial confirmed the results of previous studies and showed similar rates of death and myocardial infarction in CABG and PCI treated patients (fig. 2). This is of pivotal interest since this is the first randomized trial in a large number of patients demonstrating that outcome is comparable after treating 3-VD with PCI or CABG. However, when looking at the primary endpoint, rates of MACCE at 12 months were significantly higher in the PCI group (17.8%, vs. 12.4% for CABG; P=0.002) and therefore the criterion for non-inferiority was not met. Detailed analysis clearly showed that the primary endpoint was not met due to an increased rate of repeated revascularization (13.5% vs. 5.9%, P<0.001) in the PCI group (fig. 2). This observation reflects the current knowledge that PCI is associated with an increased need for revascularization compared to CABG. Apparently, this is also the fact when using DES.

On the other hand, previous data concerning increased rates of stroke in CABG patients were confirmed in the SYNTAX trial. CABG patients had a significantly higher rate of stroke compared to PCI-treated patients (2.2%, vs. 0.6% with PCI; P=0.003). The authors conclude that CABG remains the standard of care for patients with 3-VD or left main coronary artery disease since the primary endpoint was not met at 1 year. 

It will be of pivotal interest to evaluate subgroup analyses of the SYNTAX trial in terms of patient morbidity and perioperative risk as well as analyses concerning lesion characteristics (e.g. restenosis rate in patients with short circumscripte lesions). These data will hopefully allow a better therapeutic stratification in patients with 3VD.

Conclusion:

What does the patient with stable CAD but frequent episodes of angina during garden work or cycling expect from his treating physician? Firstly, the patient wants to get rid of his angina and this problem should be solved in an easy and comfortable and less invasive way. Lifestyle modifications and state of the art medical therapy is the first pivotal step towards this goal. However, daily practice teaches us that in a large number of patients (>30% in the highly selected group of patients in the COURAGE trial (2) we cannot achieve this goal. At this point, PCI comes into the focus of interest in patients suitable for a complete revascularization. PCI improves symptom-free survival and quality of life with minimal invasiveness.

However, the patient needs to be informed about higher rates of target vessel revascularization compared to CABG with comparable outcome and on the other hand higher rates of stroke after CABG. Given the identical outcome, PCI with the drawback of a significantly higher rate of revascularization procedures (which is in most cases a repeated PCI) but less invasiveness will be the patient’s therapeutic first choice in most cases. From a scientific point of view, continuous research is needed to further reduce restenosis after PCI and stroke incidence after CABG. Future studies need to carefully evaluate the concomitant morbidity and impairment of quality of life associated with stroke after CABG.  It is assumed that stroke associated sequelae are generally more severe compared to repeated PCI. On the basis of the published results the treating physician will still need to evaluate individual comorbidities of each patient with 3VD in order to reach the best therapy for the patient.