Angioplasty in the Diabetic Patient

ABSTRACT: Patients with diabetes have an increased risk of coronary artery disease, and are at an increased risk of mortality and morbidity with coronary revascularization procedures. This article provides a review of the currently available information on percutaneous coronary intervention (PCI) in the diabetic patient. The effectiveness of PCI in diabetes is discussed, and the factors that may influence outcomes are explored. Recent developments in PCI procedures, such as stents and drug-eluting stents, glycoprotein IIb/IIIa inhibitors and brachytherapy, are evaluated in terms of their ability to improve the prognosis in this patient group. This article is reprinted with permission from the Journal of Invasive Cardiology, January 2004;16:23-27. Patients with diabetes have an increased risk of coronary artery disease (CAD), and are at an increased risk of mortality and morbidity with coronary revascularization procedures. This is due to various modifications associated with diabetes such as endothelial dysfunction (an abnormal response of the cells of the vascular wall), and a greater risk of thrombotic events.^1-3 In 1995, the publication of results from the Bypass Angioplasty Revascularization Investigation (BARI) trial raised doubts about the use of angioplasty in diabetic patients.⁴ The data showed that the 5-year survival rate of diabetics with multivessel coronary lesions was better after coronary artery bypass graft (CABG) surgery than after angioplasty. This was supported by results from the Emory Angioplasty versus Surgery Trial (EAST)⁵ and the Coronary Angioplasty versus Bypass Revascularization Investigation (CABRI).⁶ More recently, new tools and approaches to percutaneous coronary intervention (PCI) have been developed, including stents and drug-eluting stents (DES), glycoprotein (GP) IIb/IIIa inhibitors and brachytherapy. These new techniques have led to an improvement in the prognosis for diabetic patients undergoing PCI. For example, the Evaluation of Platelet IIb/IIIa Inhibitor for Stenting (EPISTENT) trial showed that PCI incorporating stenting and abciximab gives similar outcomes in both diabetic and non-diabetic patients.⁷ Early studies comparing PCI and CABG in diabetics BARI. BARI investigated the impact of percutaneous transluminal coronary angioplasty (PTCA) and CABG on long-term mortality rates. Patients were recruited from 1988-1991 and followed for 5 years. A separate analysis of diabetic patients was performed in 1992.⁴ The data showed that the 5-year survival rate of diabetic patients was worse than in non-diabetic patients, regardless of the revascularization method, and suggested that the revascularization method can have an impact on survival (Figure 1). The five-year survival rate of diabetic patients treated by angioplasty (65.5%) was significantly lower than that of patients treated by surgery (80.6%). In addition, the benefit of surgery was greater in diabetics treated with insulin than in those treated with oral hypoglycemic drugs. CABRI. The CABRI study supports the results of BARI.⁶ Indeed, a subanalysis of the diabetic patients in CABRI found that the mortality rate of the diabetic patients was double that of non-diabetic patients, and there was a non-significant trend for a higher mortality rate at 4 years in diabetics randomized to receive PTCA compared to those randomized to receive CABG (22.6% versus 12.5%, respectively).⁶ The authors concluded that the higher diabetic mortality rate was most likely related to a more rapid progression of lesions in these patients than to greater post-revascularization disease. EAST. The EAST trial failed to show a survival advantage for CABG over PCI at 8-year follow-up.⁵ Death occurred in 7.3% of the CABG group and 10.7% of the PTCA group (p = 0.73). Although patients with proximal left anterior descending (LAD) stenosis (20.4% versus 14.4%, respectively) and those with diabetes (24.5% versus 39.9%, respectively) tended to have better late survival with surgical intervention, this did not reach statistical significance. Registry data. Other sources of information are available, particularly retrospective studies that are based on records involving several thousand patients.^8-10 These studies do not show any difference in mortality rates between diabetics treated by angioplasty versus those treated with bypass surgery. However, a more detailed analysis of the results suggests that the condition of patients approaching the interventional cardiologist is less serious than that of those approaching the surgeon; for example, the patients have more severe coronary lesions or higher ejection fractions (EF).⁸ If adjustments are made to these criteria, it is then possible to demonstrate that surgery is associated with a better long-term prognosis. ⁸ Retrospective studies consistently demonstrate that the medium-term (five-year) prognosis is worse for diabetic patients treated by angioplasty than for non-diabetic patients. This prognosis is linked to a greater frequency of cardiovascular events in diabetics, and is less favorable in those treated with insulin than in others. In diabetic patients with multisite lesions, surgery is associated with better long-term survival. PCI in diabetes: Possible explanations for poor prognosis Increased atherosclerosis. The greater frequency of events observed in diabetics does not seem to be related to a greater frequency of periprocedural complications; therefore, other explanations must be considered.^4,8 Rozenman et al. studied the coronary angiograms of 248 patients, including 55 diabetics, eight months after a conventional balloon angioplasty procedure.¹¹ New lesions were found more frequently in the diabetic patients (14.8%) than in the non-diabetic patients (9.4%), and were more frequently found in arteries treated in the initial procedure than in other arteries. This phenomenon, particularly if associated with a shift toward coronary occlusion, can explain some of the cardiovascular events observed in diabetics after revascularization, and the greater protective effect produced by mammary bypass. Restenosis. Other authors have studied the effects of diabetes on the risk of restenosis after conventional balloon angioplasty. When the results of the available studies are combined, it appears that the rate of restenosis is higher in diabetic patients (54%) than in non-diabetics (35%; for a review of available data, see reference 12). This may explain the higher reintervention rate observed in diabetics treated by angioplasty compared with non-diabetics treated using the same technique, and diabetics treated by bypass surgery.^4,13 It has been suggested that this increased risk of restenosis in diabetic patients is related to the combination of an increased production of growth factor, an accumulation of advanced glycosylated end products and an increased inflammatory response.^14,15 In 1997, we reported on the angiographic monitoring at six months of 300 patients, including 57 diabetics, treated with conventional balloon angioplasty.16 In this series, a particular form of restenosis, occlusive restenosis [defined as a Thrombolysis in Myocardial Infarction (TIMI) flow of 0 or 1], was identified in the diabetic patients on angiographic monitoring. Occlusive restenosis was found in 14% of the lesions in diabetic patients, versus only 3% of the non-diabetic patients (Figure 2). We have recently confirmed these results in a large series of more than 500 diabetic patients treated with balloon angioplasty.¹² The rate of restenosis per successfully treated lesion was 62%, with 49% non-occlusive restenoses and 13% occlusive restenoses. For each patient, the risk of evolution toward occlusion increased with the number of lesions treated. This figure was 11% for a patient dilated at a single site and > 25% for a patient dilated at 2 or more sites. In this study, evolution toward occlusion was associated with a significant deterioration in left ventricular function. Patients with occlusion of a dilated site experienced a 7% decrease in left ventricular ejection fraction (LVEF) over the 6 months following angioplasty (basal LVEF, 57% versus 50% at 6 months), while no significant change was observed in patients either without restenosis or with restenosis in its non-occlusive form.¹² Analysis of the long-term prognosis of these patients has shown that those presenting an evolution toward occlusive restenosis had a 10-year mortality rate that was twice as high as both those without restenosis and those with non-occlusive restenosis (60% versus 30%, respectively).¹⁷ Causes of death included acute MI, sudden death and cardiac heart failure. In our opinion, these results can partly account for the poor prognosis of multivessel diabetic patients after balloon angioplasty. In our scheme, a BARI type multisite angioplasty in a diabetic patient will result in more than 25% risk of coronary occlusion at 6 months. The considerable decrease in the EF associated with this contributes to the increased cardiovascular mortality observed (Figure 3). Recent developments in PCI The results discussed above are from a period where, in more than 90% of cases, angioplasty procedures were carried out with the balloon alone as the only tool. Since then, other tools and pharmacological and physical approaches have been introduced (e.g., stents and DES, GP IIb/IIIa inhibitors, brachytherapy). In addition, secondary preventive methods have been developed and refined. These new innovations should translate into an improved prognosis for diabetic patients. GP IIb/IIIa inhibitors. New and powerful antiplatelet agents have been developed (the leading of which is abciximab), that block the fibrinogen binding site on the platelet GP IIb/IIIa receptor, thus blocking the final common pathway of platelet activation. When combined with angioplasty, the administration of this type of drug produces a significant reduction in the number of cardiovascular events in the 6 months following the procedure.^7,18-20 GP IIb/IIIa inhibitors were not used in the BARI study. However, these molecules have since shown a greater benefit in diabetic patients than in non-diabetic patients. For example, in the Evaluation in Percutaneous Transluminal Coronary Angioplasty to Improve Long-term Outcome with Abciximab Glycoprotein IIb/IIIa Blockade (EPILOG) study, at 6 months the number of deaths and infarctions in the diabetic group was reduced by more than 50% through the use of abciximab with balloon angioplasty.¹⁹ GP IIb/IIIa inhibitors are also of benefit after stent implantation. Analysis of the diabetic group of patients included in the EPISTENT study showed that the combined use of stents and abciximab reduced the 1-year mortality rate.⁷ A meta-analysis of the Evaluation of IIb/IIIa Platelet Receptor Antagonist 7E3 in Preventing Ischemic Complications (EPIC), EPILOG and EPISTENT trials confirmed the benefit of abciximab on the long-term survival of diabetic patients following percutaneous revascularization (Figure 4).²¹ At 1 year, abciximab use decreased mortality in the diabetic patients (n = 1,462) from 4.5% to 2.5% (p = 0.031), and in the non-diabetic patients (n = 5,072) from 2.6% to 1.9% (p = 0.099). The authors concluded that abciximab decreases the mortality rate of diabetic patients to the level of placebo-treated, non-diabetic patients. Stents. The use of stents has improved the results of angioplasty in the medium term by decreasing the risk of restenosis. Several investigators have studied the effects of diabetes on the risk of restenosis after stent implantation.²² Although these studies indicate that diabetes remains a risk factor for restenosis with this method of revascularization, they also show that stent implantation in the diabetic patient reduces the risk of restenosis six months post-procedure (54% versus 34%). More importantly perhaps, the risk of occlusion of the dilated site at 6 months is also reduced by stent insertion, from approximately 15% to 12,16,22 The physical benefit observed on the angiographic status of the dilated vessel seems to translate into a clinical benefit. Indeed, we recently reported that diabetic patients treated with coronary stents had a lower rate of repeat revascularization and a lower combined rate of cardiovascular death and MI than diabetic patients treated with balloon angioplasty alone (Figure 5).²³ These results were confirmed by the recent report of the two-year follow-up results of the diabetic subgroup of the Arterial Revascularization Therapy Study (ARTS), which compared bypass surgery with coronary stenting in patients with multivessel CAD. In this study, the combined rate of death, MI and stroke at two years was similar for diabetics for both the stenting and surgery groups (16.1% versus 14.6%, respectively). However, the need for repeat revascularization was higher in diabetics treated with coronary stenting.²⁴ DES. The specific issue of the higher rates of repeat revascularization in diabetics after PCR might be largely solved by the recent development of stents impregnated with cytostatic molecules. The results of the diabetic subgroup (n = 279) of the Sirolimus-Coated Velocity Stent in Treatment of Patients with De Novo Coronary Artery Lesions (SIRIUS) trial are encouraging. In this study, the use of a sirolimus-coated stent was associated with a significant reduction in in-segment restenosis rates in this high-risk population (50.5% versus 17.6%).²⁵ The Randomized Study with the Sirolimus-Eluting Bx Velocity Balloon-Expandable Stent (RAVEL) trial randomized 238 patients with de novo coronary stenosis to receive either a standard stainless-steel stent or a rapamycin-coated Bx Velocity stent (Cordis Corporation, Miami, Florida).²⁶ At six months, the restenosis rate was 0% for the coated-stent group compared with 26% for the bare stent group (p Brachytherapy. The use of endocoronary brachytherapy has been evaluated in the treatment of patients with severe coronary restenosis. This strategy is highly efficient in the prevention of re-restenosis in the diabetic population, as demonstrated by Gruberg et al., who reported a significant reduction in the binary recurrent restenosis rate in this group of patients (63.8% versus 15.7%; p 27 Effects of managing risk factors. Although this is not the subject of this review, it is nevertheless important to remember that the benefit of secondary preventive measures using angiotensin-converting enzyme inhibitors and statins has been clearly demonstrated in the diabetic population in the Heart Outcomes Prevention Evaluation (HOPE)²⁸ and Heart Protection Study (HPS)²⁹ trials, respectively. In these two studies, the use of ramipril²⁸ or simvastatin²⁹ in diabetic patients with cardiovascular disease was associated with a significant reduction of cardiovascular events at long-term follow-up compared with placebo. The U.K. Prospective Diabetic Study (UKPDS) trial showed that the use of metformine was associated with a significant reduction in cardiovascular events and deaths in overweight diabetic patients without cardiovascular disease.³⁰ What method of coronary revascularization should be offered to a diabetic patient in 2002? The results discussed herein provide some indicators of the revascularization strategy to be offered to the diabetic patient. First, balloon angioplasty alone must be totally abandoned in favor of angioplasty in combination with stents (with or without DES) and GP IIb/IIIa inhibitors, as suggested by various studies including EPIC, EPILOG and EPISTENT.^7,22,24 Second, in the absence of contradictory data, angioplasty remains the method of choice for diabetic patients with a single lesion of the coronary tree. Third, surgery must be considered to be the method of choice in multivessel patients, on the condition that at least 1 arterial graft is inserted, as demonstrated in the BARI trial.⁴ In patients who are less than 60 years old with multivessel, non-diffuse lesions, it may be tempting to propose revascularization by modern angioplasty, using stent with or without DES, GP IIb/IIIa inhibitors and secondary prevention. This attitude is supported by the results of the EPISTENT study and data from records demonstrating that when the decision is left to the clinical judgement of the physician, his judgement seems to allow him to correct the theoretical handicap of angioplasty in this population (i.e., the more severe diabetic patients are not sent to angioplasty, as demonstrated by the analysis of the registry data). Another option in these patients may be combining the insertion of exclusive arterial grafts with revascularization by angioplasty and stent implantation at the other sites. This approach would avoid the use of venous grafts, which are associated with unsatisfactory results in this population.⁴ The Treat Angina with Aggrastat and Determine Cost of Therapy with an Invasive or Conservative Strategy (TACTICS) study recently demonstrated in diabetic patients with unstable angina that prescribing a GP IIb/IIIa inhibitor and providing early coronary angiography (leading, the the majority of cases, to revascularization by angioplasty) was preferable to a more conservative strategy that did not include early coronary angiography (and angioplasty).³¹ Future studies A new comparison of CABG and PCI in diabetic patients with multivessel disease has just received approval from the National Institute of Health. The Future Revascularization in Patients with Diabetic Mellitus Optimal Management of Multivessel Disease (FREEDOM) trial will include 1,500 diabetic patients and will begin in 2003. The diabetic patients in the PCI arm of the study will receive contemporary treatment with DES and GP IIb/IIIa inhibitors. Also of interest, the BARI 2 diabetes study will investigate whether initial treatment with angioplasty or bypass surgery is better than initial treatment with a medical program in patients with type 2 diabetes. Conclusion Studies show that the use of balloon angioplasty in the diabetic population is associated with poor results in the medium term (5-10 years). The BARI study has brought this problem to our attention and stimulated research that has already produced explanations for this poor success rate. Identification of the specific problems that these patients pose (rapid progression of lesions and the tendency of lesions to develop toward occlusion), the availability of tools to counteract these problems (stents, DES, antiplatelet agents, metformine, statins, enzyme-converting inhibitors) and the encouraging results of their use in the diabetic population mean that the rapid modification of treatment can be envisaged and, in turn, result in an improvement in the prognosis of these patients. Thus, the results of the FREEDOM study are eagerly awaited.