Almost 1.5 million coronary revascularization procedures, either coronary artery bypass grafting (CABG) or percutaneous coronary interventions (PCI), are performed each year in the United States, and approximately one quarter of them involve diabetic patients.41 The randomized data on diabetic patients are scarce and are mainly derived from subgroup analyses of revascularization trials of patients with multivessel disease initiated in the late 1980s and early 1990s. Overall, diabetic patients with multives-sel disease seem to have a better prognosis after CABG than after PCI. Although diabetic patients frequently have concurrent risk factors and comorbidities, diabetes has been identified as an independent predictor of CV events during and after revascularization, both percutaneous and surgical. Several pathophysiologic peculiarities of diabetic atherosclerosis previously discussed may negatively affect prognosis and response to coronary revascularization.
Table 2-5. Indications for Cardiac Testing in Diabetic Patients according to the American Diabetes Association
Testing for CAD is warranted in patients with >1 of the following characteristics:
Typical or atypical cardiac symptoms Resting ECG suggestive of ischemia or infarction Peripheral or carotid occlusive arterial disease Sedentary lifestyle, age >35 years, and plans to begin a vigorous exercise program
Testing is also warranted for patients with >2 of the following risk factors in addition to diabetes:
Total cholesterol >240 mg/dL (6.2 mmol/L), LDL cholesterol >160 mg/dL (4.1 mmol/L), or HDL cholesterol <35 mg/dL (0.9 mmol/L) Blood pressure >140/90 mm Hg Smoking
Family history of premature CAD Positive micro/macroalbuminuria test
Whereas in-hospital and 30-day outcomes after PCI in diabetic patients are comparable to those of non-diabetics, large-scale registries have identified diabetes as an independent predictor of long-term mortality and need for repeat revascularization. Putative mechanisms negatively affecting outcomes may include endothelial dysfunction, prothrombotic state, greater propensity for restenosis and adverse vascular remodeling after angioplasty, increased protein glycosyl-ation, and vascular matrix deposition (see Fig. 2-2).9 All of these processes appear to be exacerbated by hyperglycemia and hyperinsulinemia. In addition, diabetic patients more frequently have noncar-diac comorbidities that may also negatively affect outcomes.
Restenosis has been a main challenge in the percutaneous treatment of diabetic patients. Restenosis in the diabetic population is reviewed here, and details about vessel wall response after catheter-induced barotrauma can be found in Chapter 32. The restenotic process in diabetic patients is characterized by excessive proliferative response and increased vascular matrix deposition.9 Mechanisms that may play a role include the interaction of the RAGE and its ligand, the peroxisome proliferator-activated receptor (PPAR)-y, and thrombospondin-1 (TSP-1). RAGE is a cell surface molecule that is expressed at low levels in homeostasis but becomes upregulated at the site of vascular injury, particularly within the expanding neointima. In animal studies, blockade of RAGE resulted in significantly decreased neointimal proliferation, migration, and expression of extracellular matrix proteins. Finally, the link between insulin resistance and restenosis after coronary stenting was recently documented in humans. Among 120 patients undergoing coronary stenting who had an oral glucose tolerance test (OGTT), insulin sensitivity independently predicted the minimal lumen diameter at follow-up.42
In the diabetic population, the benefit of stenting was initially demonstrated in a single-center analysis comparing angiographic and clinical outcomes in 314 diabetic patients, matched for baseline characteristics, who underwent either stenting or percutaneous transluminal coronary angioplasty (PTCA).43 At 6 months, the restenosis rate was significantly lower in the stent group than in the PTCA group (27% versus 62%). At 4 years, the incidence of cardiac death or nonfatal MI was lower in the stent group (14.8% versus 26.0%; P = .02), as was the need for repeat revascularization (35.4% versus 52.1%; P = .001). On a broader scale, a dramatic benefit of stent-ing in diabetic patients was detected in a study comparing outcomes of the diabetic cohorts of the 1997-2001 National Heart, Lung, and Blood Institute (NHLBI) Dynamic Registry, which was characterized by a large use of stents (87%), and the 1985-1986
NHLBI PTCA registry, in which patients were treated with PTCA only. The most recently performed procedures were characterized by an impressive reduction in in-hospital complications such as abrupt vessel closure (0.9% versus 2.2%), MI (1.0% versus 7.4%), urgent CABG (0.8% versus 6.2%), and death (1.9% versus 4.3%).44 However, restenosis remains a limitation of stent-based PCI, particularly in diabetic patients. A meta-analysis of six stent trials identified diabetes as an independent predictor of restenosis (odds ratio [OR] = 1.3), and, among the 1166 diabetic patients included, the overall rate of restenosis was 37%.45 The most important conditions predisposing to in-stent restenosis among diabetic patients have been a small reference vessel diameter and long stented segments.
Within the Do Tirofiban And Reopro Give similar Efficacy outcomes Trial (TARGET), contemporary PCI (i.e., based on third-generation-stents and triple anti-platelet therapy) was associated with similar 30-day event rates among diabetic (n = 1117) and nondiabetic patients (n = 3692).46 In addition, no significant difference in major adverse cardiac events (MACE) at 6 months was observed, although diabetic patients had greater target vessel revascularization (TVR) than nondiabetic patients (10.3% versus 7.8%, P = .008). At 1 year, there was a trend toward increased mortality in the diabetic group (2.5% versus 1.6%, P = .056), but diabetes was not an independent predictor of mortality. These encouraging results were not replicated in another recent stent study that included a large diabetic population (n = 2694), the Prevention of REStenosis with Tranilast and its Outcomes (PRESTO) trial.47 Although no difference in in-hospi-tal events was observed between diabetic and non-diabetic patients, after adjustments for baseline characteristics, diabetes was identified as independent predictor of death (relative risk [RR] = 1.9) and of TVR (RR = 1.3) at 9 months.
Within registries, the outcomes of diabetic patients in recent years have remained unfavorable compared with those of nondiabetics, despite a broader use of stents. The NHLBI Dynamic Registry enrolled consecutive patients (1058 with diabetes and 3571 without diabetes) who underwent PCI from July 1997 to June 1999. At 1 year, diabetic patients had significantly higher adjusted risks of mortality (RR = 1.8) and of repeat revascularization (RR = 1.4).48 Similarly, an analysis of 100,253 PCI patients enrolled in the 1998-2000 American College of Cardiology-National Cardiovascular Data Registry (ACC-NCDR) showed that diabetes was an independent predictor even of in-hospital mortality (OR = 1.4).49
Drug-eluting stents (DES) have revolutionized the field of interventional cardiology by dramatically reducing the incidence of restenosis and, as a consequence, the need for TVR (see Chapter 15). The available information on DES in diabetic patients remains scarce and is derived from registries or from subgroup analyses of randomized trials comparing the siroli-mus-eluting stent (SES) Cypher (Cordis, Johnson and Johnson), or the paclitaxel-eluting stent (PES) Taxus (Boston Scientific), with bare metal stents (BMS). Among diabetic patients (n = 279) enrolled in the randomized Sirolimus-Eluting Bx-Velocity Balloon Expandable Stent in the Treatment of Patients with de novo Native Coronary Artery Lesions (SIRIUS) trial, SES implantation was associated with a significant reduction in restenosis compared with BMS implantation (6.9% versus 22.3%; P < .001).50 The relative risk reduction (RRR) was of the same magnitude in diabetic and nondiabetic patients. However, because of higher event rates in the diabetic population, the absolute benefit was greater than among nondiabetic patients (154 versus 111 restenoses prevented per 1000 patients treated). The incidence of MACE was also significantly reduced in diabetic patients, from 25% in those with BMS to 9.2% in those with SES.
With respect to the PES, among the 155 diabetic patients enrolled in the TAXUS IV trial (one third of them on insulin), a significant improvement in outcomes compared with BMS was detected.51 At 12 months, the TVR rate was reduced from 24.0% to 11.3%, and the MACE rate was reduced from 27.7% to 15.6%. The first study that addressed the efficacy of DES specifically in diabetics randomly assigned 160 patients to SES or BMS.52 At 9 months, the in-segment late lumen loss was significantly less in the SES group (0.06 mm) than in the BMS group (0.47 mm). As a result, the incidence of target lesion revascularization (7.3% versus 31.3%) and of MACE (11.3% versus 36.3%) were significantly lower in the SES group.
Despite these findings, several observations suggest that diabetic restenosis may be resilient even in the setting of DES. Accordingly, in the SIRIUS trial, diabetes remained an independent predictor of poor angiographic and clinical outcome among patients undergoing SES implantation.50 In addition, in the same trial, the restenosis rate among diabetic patients with lesions longer than 15 mm in vessels smaller than 2.5 mm was as high as 23.7%. Finally, in the small group of patients treated with insulin (n = 82), the benefit in terms of restenosis with drug elution was modest (35.0% versus 50.0%; P = .38). Similarly, in the e-Cypher registry including more than 15,000 patients undergoing SES implantation, both non-insulin-requiring and insulin-requiring diabetes were independent predictors of MACE at 12 months (OR = 1.4 and OR = 2.2, respectively).53
Although not of all the reports have identified diabetes as an independent predictor of poor DES-based PCI outcome, the results of a recent singlecenter experience on 260 consecutive diabetic patients undergoing multivessel DES implantation are sobering: the study detected a 9-month MACE rate of 25%.54 Particularly high were the event rates among patients taking insulin (adjusted OR = 2.7). With respect to subacute or late stent thrombosis, diabetes was not found to be a predictor of increased events in the BMS era. Conversely, in the e-Cypher registry, insulin treatment at baseline was found to be an independent predictor of late stent thrombosis (OR = 2.8)53 Similarly, in a prospective cohort of 2229 patients undergoing DES implantation, diabetes was found to be an independent predictor of late stent thrombosis (OR = 3.7).55 It remains to be determined whether late DES thrombosis will become a clinically relevant problem, particularly in the diabetic population. Potentially, the clopidogrel resistance associated with diabetes could be an underlying link.56 With respect to whether SES or PES may be the best choice for the diabetic patients, no conclusion can be made at this point, although a randomized angio-graphic study of 250 diabetic patients favored SES.57
Paralleling what was described for PCI, diabetes also negatively affects outcomes after CAbG. The impact of diabetes on morbidity and mortality in patients undergoing surgical coronary revascularization was addressed in a retrospective cohort study, based on the 1997 Society of Thoracic Surgery (STS) database, that included 41,663 diabetic patients among a total population of 146,786 patients.58 At 30 days, the mortality rate was significantly higher in the diabetes group (3.7% versus 2.7%). The unadjusted and adjusted OR for mortality in these diabetic patients were 1.4 and 1.2, respectively. With respect to diabetes treatments at presentation, the adjusted OR for mortality among patients taking oral hypoglycemic drugs was 1.1, and that for patients taking insulin was 1.4. In addition, the overall morbidity rate and the infection rate were significantly higher in the diabetic patients. With respect to long-term mortality after CABG, a prospective New England cohort study that included 11,186 consecutive diabetic patients and 25,455 nondiabetic patients undergoing CABG from 1992 to 2001 detected a significantly higher annual mortality rate among diabetic patients (5.5%) compared with nondiabetics (3.1%).59
In addition to increased periprocedural morbidity and mortality, as well as long-term mortality, diabetes is associated with an increased rate of repeat revascularization after CABG. A prospective analysis on 26,927 patients who were contacted every 5 years up to 25 years after CABG at a single institution in the United States identified diabetes as an independent predictor of subsequent coronary revasculariza-tion (Fig. 2-6).60 As part of the metabolic syndrome, diabetes is frequently associated with obesity, hypertension, and hypertriglyceridemia. The impact of these four factors (the "deadly quartet") on 8-year mortality after CABG was assessed in a single-center database that included 6428 patients.61 Compared with individuals who had no risk factors, the HR for mortality increased from 1.6 among those with one risk factor to 3.9 for those with four risk factors. The yearly mortality rate ranged from 1% in patients with
Figure 2-6. Predicted freedom from repeat coronary revascularization after coronary artery bypass surgery stratified by diabetes mellitus and its treatment. (From Sabik JF, Blackstone EH, Gillinov AM, et al: Occurrence and risk factors for reintervention after coronary artery bypass grafting. Circulation 2006;114:I454-460.)
100 90 80 70 60 50 40 30 20 10
100 90 80 70 60 50 40 30 20 10
Pharmacologically treated i—i—i—r i—i—i—r
no risk factors to 3.3% in patients with four risk factors. The prevalence and impact of undiagnosed diabetes were addressed in a German retrospective analysis of 7310 patients undergoing CABG between 1996 and 2003.62 In this cohort, 5.2% of the patients had undiagnosed diabetes, defined as a fasting plasma glucose level of 126 mg/dL or higher. The perioperative mortality rate was significantly higher in the group of patients with undiagnosed diabetes (2.4%), compared with nondiabetic individuals (0.9%) or with patients with known diabetes (1.4%).
The use of multiple arterial conduits, including bilateral internal mammary artery (IMA) grafts, has been shown to improve the long-term results of CABG and reduce the need for repeat revascularization.60 A recently published observational cohort showed improved 10-year survival and lower rates of recurrent MI and repeat CABG in diabetic patients with preserved left ventricular function who received bilateral IMA grafts.63 Additionally, and in contrast with previous reports, no significant difference in the incidence of sternal wound infections was detected. With respect to the impact of off-pump surgery, a retrospective analysis compared 346 diabetic patients undergoing off-pump CABG with control subjects and showed reduced complication rates but no survival advantage.64
Was this article helpful?
Diabetes is a disease that affects the way your body uses food. Normally, your body converts sugars, starches and other foods into a form of sugar called glucose. Your body uses glucose for fuel. The cells receive the glucose through the bloodstream. They then use insulin a hormone made by the pancreas to absorb the glucose, convert it into energy, and either use it or store it for later use. Learn more...