‘FREEDOM’ trial: Have the interventional cardiologists lost their freedom?

The recent publication of the results of “FREEDOM” trial has cast considerable doubt in the minds of physicians about the utility of multi-vessel angioplasty even while using drug eluting stents.1 We need to examine the results of this landmark trial in detail in addition to the earlier data available on PCI among diabetics before incorporating the results of the FREEDOM trial into our daily practice. According to the Diabetes Atlas 2006 published by the International Diabetes Federation, the number of people with diabetes in India which is currently around 40.9 million is expected to rise to 69.9 million by 2025.2 A national survey of diabetes conducted in six major cities in India in the year 2000 showed that the prevalence of diabetes in urban adults was 12.1%. Prevalence of impaired glucose tolerance (IGT) was also high at 14.0%.3 With this high degree prevalence, it is understandable that 30–40% of patients in India who come for PCI are diabetics.4 Therefore, the results of FREEDOM trial are of particular relevance to us. The National Cholesterol Education Program report from the United States and guidelines from Europe consider type 2 diabetes to be a CHD equivalent, thereby elevating it to the highest risk category.5 This classification was based in part upon the observation that patients with type 2 diabetes without a prior MI (mean age 58) were at the same risk for MI (20 versus 19 percent) and coronary mortality (15 versus 16 percent) as patients without diabetes who had a prior MI (mean age 56) [Fig. 1].6 Fig. 1 Kaplan–Meier estimates of the probability of death from coronary heart disease in 1059 subjects6 [reproduced with permission]. Historically patients with diabetes who undergo PCI have been a difficult-to-treat patient population. They tend to have smaller and often tortuous arteries, longer lesions, diffused disease and a higher rate of treatment failures including relatively high rates of repeat procedures and stent thrombosis.7 The FREEDOM trial was sponsored by the National Heart, Lung, and Blood Institute, USA. It was a well planned and executed study with 1900 patients of DM (Type I & Type II) randomized to PCI or CABG. The inclusion criteria were that they had two or three vessel disease with at least one 70% lesion in the vessel treated. Patients with LMCA disease were excluded. About 32% of patients were on insulin in both the groups. The HbA1c at the outset was around 7.8% in both the groups. Majority of patients had three vessel diseases (82%). About 70% were men and the average age was 63 years.1,8 First generation sirolimus-eluting and paclitaxel-eluting stents were used exclusively in 51% and 43% of patients respectively, in the PCI group. Number of lesions stented was 3.5 ± 1.4. DAPT with aspirin and clopidogrel was recommended for a minimum period of 1 year to PCI group. The mean SYNTAX score was 26.2 ± 8.6 and mean Euro SCORE 2.75 ± 2.45 which shows majority patients were of intermediate SYNTAX score. Surgery was off-pump in 18.5% of cases. LIMA was used in 94% of patients.1,8 The primary composite endpoint which was a combination of death from any cause, myocardial infarction and stroke was significantly lower in the CABG-treated patients, as was the five-year rate of all-cause mortality and myocardial infarction. Following revascularization, the five-year rate of all-cause mortality, myocardial infarction, and stroke was 26.6% in the PCI-treated patients and 18.7% among patients who underwent CABG surgery, a statistically significant relative risk reduction of 30%. Stroke rates were significantly higher in the CABG arm, mainly because of the excess of stroke in the first 30 days. Of the strokes, most (87%) were ischemic strokes and 13% were hemorrhagic. The results are shown in Table 1. Table 1 Kaplan–Meier estimates of key outcomes at 2 year and 5 year after randomization in the FREEDOM trial [reproduced with permission].1 The trial despite excluding the need for repeat revascularization from the primary composite endpoint clearly shows that surgery is better in the long run for this group of patients. There is a mortality benefit as well as continuing benefit in terms of reduced risk of myocardial infarction among patients who underwent CABG. Although stroke is more in the surgical group, most cases of stroke occurred in the early postoperative period. Even the most diehard interventionist cannot escape the conclusion that surgery proved better in the FREEDOM trial. Importantly, surgery proved better in all subgroups including different subsets of SYNTAX score. There are some points to remember. Modern antiplatelet agents like prasugrel or ticagrelor were not used in FREEDOM trial. We are now aware that prasugrel has a particular advantage for diabetics in that it reduces stent thrombosis. Compared to the clopidogrel group in the TRITON TIMI 38 study, the prasugrel group had a 52% relative reduction for stent thrombosis.9 With regard to ticagrelor, the pivotal PLATO trial has shown that at 12 months, the primary endpoint—a composite of death from vascular cause, myocardial infarction, or stroke—had occurred in 9.8% of patients receiving ticagrelor as compared with 11.7% in those receiving clopidogrel. Myocardial infarction has also been shown to be reduced in the ticagrelor arm (5.8% in the ticagrelor group vs. 6.9% in the clopidogrel group, p = 0.005).10 Thus the use of modern generation P2 Y12 inhibitors if used in FREEDOM trial might have made a difference to the outcomes. In addition, we need to examine the outcomes of newer generation drug eluting stents in diabetic patients. The new Resolute Integrity Zotarolimus-eluting coronary stent (Medtronic) has been approved by the Food and Drug Administration (FDA) specifically for use in diabetics. It is the first drug eluting stent approved by the FDA for use in patients with diabetes. This was done on the basis of the results of the RESOLUTE Global Clinical Program, which included a total of five trials with identical inclusion and exclusion criteria [the randomized RESOLUTE All Comers (R-AC, N = 1140), plus RESOLUTE First-in-Man (N = 139), RESOLUTE US (N = 1402), RESOLUTE International (N = 2349), and RESOLUTE Japan (N = 100)]. All 5 Resolute trials were designed with similar endpoints and statistical methodologies, and all required the same regimen of dual antiplatelet therapy. An ad hoc analysis has pooled outcomes for the 5130 recipients of the Resolute ZES in these 5 clinical trials.11 For the total population of DM patients, 2-year outcomes were compared with those for 3595 non-DM patients from the pooled cohort. Expectedly, rates of major adverse cardiac events were significantly higher in DM patients than in non-DM patients. However, the 2-year incidence of stent thrombosis was very low regardless of the presence or severity of diabetes (0.82% for non-DM patients, 0.93% for non-insulin-treated DM patients, and 1.79% for insulin-treated DM patients). When outcomes were analyzed by treatment with insulin, the incidence of target lesion failure was nearly the same for non-insulin-treated DM patients versus non-DM patients but was significantly higher for insulin-treated DM patients (8.4% for non-diabetics, 8.9% for non-insulin-treated DM patients, and 16.7% for insulin-treated DM patients).12 With respect to everolimus-eluting stent – Xience (Abbott), the SPIRIT V registry showed a low rate of death, myocardial infarction, and TVR at 1 year (6.8%) in 2700 patients and the outcomes of the DM subgroup (30% of the population) did not significantly differ from the overall study results.13 The overall one year cumulative stent thrombosis rate was 0.66% as shown in Fig. 2. Fig. 2 Definite and probable stent thrombosis rates at 1 year in the SPIRIT V registry. We made an attempt to compare the 2 year results of FREEDOM surgical arm with 2 year results of everolimus and zotarolimus eluting stents as shown in Table 2. Although these are different trials with different endpoints, one can draw some idea from this table and it appears that the results are reasonably comparable. Is it possible that if one of these modern drug eluting stents were used in FREEDOM trial the results could have been different? Table 2 Comparison of FREEDOM surgical arm with outcomes of Xience V trials and RESOLUTE All Comers trial. One should also remember that even surgery is not without its disadvantages among diabetics. In a review of 9920 patients with diabetes and 2278 patients without diabetes from a single center over 15 years, Calafiore et al revealed lower survival rates in diabetic patients versus non-diabetic patients at five-year (78% versus 88%) and 10-year (50% versus 71%) follow-up.14 Diabetes increases the risk of stroke, rates of sternal wound infection, mediastinitis and saphenous vein harvest site infections, renal failure, postoperative myocardial dysfunction leading to postoperative low cardiac output syndrome with CABG and valve surgery, leading to prolonged postoperative intensive care unit stays. In the follow-up period, they have increased risk for readmission. Thus we need to understand that a diabetic has certain inherent disadvantages even while under going CABG.7 Conclusion: based on the data from FREEDOM trial, CABG appears to be the treatment of choice for multi-vessel disease in diabetics. However as antiplatelet therapy and especially modern drug eluting stents evolve, the difference in outcomes between PCI and CABG could narrow down.