Diabetes mellitus is a major risk factor for coronary artery disease (CAD) and for diffuse and progressive atherosclerosis. We evaluated the outcomes of drug-eluting stent (DES) placement and coronary artery bypass grafting (CABG) in 891 diabetic patients (489 for DES implantation and 402 for CABG) and 2,151 nondiabetic patients (1,058 for DES implantation and 1,093 for CABG) with multivessel CAD treated from January 2003 through December 2005 and followed up for a median 5.6 years. Outcomes of interest included death; the composite outcome of death, myocardial infarction (MI), or stroke; and repeat revascularization. In diabetic patients, after adjusting for baseline covariates, 5-year risk of death (hazard ratio 1.01, 95% confidence interval 0.77 to 1.33, p = 0.96) and the composite of death, MI, or stroke (hazard ratio 1.03, 95% confidence interval 0.80 to 1.31, p = 0.91) were similar in patients undergoing DES or CABG. However, rate of repeat revascularization was significantly higher in the DES group (hazard ratio 3.69, 95% confidence interval 2.64 to 5.17, p <0.001). These trends were consistent in nondiabetic patients (hazard ratio 0.80, 95% confidence interval 0.55 to 1.16, p = 0.23 for death; hazard ratio 0.77, 95% confidence interval 0.56 to 1.05, p = 0.10 for composite of death, MI, or stroke; hazard ratio 2.77, 95% CI 1.95 to 3.91, p <0.001 for repeat revascularization). There was no significant interaction between diabetic status and treatment strategy on clinical outcomes (p for interaction = 0.36 for death; 0.20 for the composite of death, MI, or stroke; and 0.40 for repeat revascularization). In conclusion, there was no significant prognostic influence of diabetes on long-term treatment with DES or CABG in patients with multivessel CAD.
Diabetes mellitus (DM) is a major risk factor for coronary artery disease (CAD), making patients prone to a diffuse, multiple, and rapidly progressive form of CAD. About 25% of patients with significant CAD who undergo percutaneous coronary intervention or coronary artery bypass grafting (CABG) have DM, and DM is significantly associated with higher rates of ischemic complications and recurrent revascularization in these patients. CABG has been shown to be superior to percutaneous coronary intervention in diabetic patients with multivessel CAD, indicating that DM is a major consideration in selecting an optimal revascularization strategy. However, these studies were conducted before the introduction of drug-eluting stents (DESs), which have markedly decreased the incidence of angiographic restenosis and repeat revascularization compared to bare-metal stents. We therefore compared the long-term effects of treatment with DESs and CABG in diabetic and nondiabetic patients with multivessel CAD and evaluated the interaction between diabetic status and treatment procedure in these patients.
Methods
This study is a subgroup analysis of patients in the Asan Multivessel Registry with and without medically treated DM. The Asan Multivessel Registry is a single-center prospective study designed to evaluate the effects of percutaneous coronary intervention with DESs and CABG on patients with multivessel CAD in clinical practice. Briefly, this registry included consecutive patients with multivessel CAD who received percutaneous coronary intervention with DESs, with or without other devices, or underwent isolated CABG at Asan Medical Center (Seoul, Korea) from January 1, 2003 through December 31, 2005. Patients who had previous CABG or underwent concomitant valvular or aortic surgery, had an acute myocardial infarction (MI) within 24 hours before revascularization, or presented with cardiogenic shock were excluded.
The decision to perform percutaneous coronary intervention or CABG depended on a physician’s choice, considering available clinical and anatomic factors, and/or a patient’s preference. During the study period, coronary stenting was performed exclusively with DESs. Percutaneous coronary intervention was performed according to current practice guidelines. Choice of a specific type of DES (i.e., sirolimus-eluting stent [CYPHER and CYPHER SELECT, Cordis, Johnson and Johnson, Bridgewater, New Jersey] or paclitaxel-eluting stent [TAXUS Express and TAXUS Liberté, Boston Scientific, Natick, Massachusetts]) was at the discretion of the operator. At this time, second-generation DESs (i.e., zotarolimus-, everolimus-, and biolimus-eluting stents) were not available to the treating physicians. Antiplatelet therapy and periprocedural anticoagulation followed standard regimens. After the procedure, patients were prescribed aspirin indefinitely and clopidogrel for ≥6 months regardless of DES type. Treatment beyond this time was at the discretion of each physician. Surgical revascularization was performed using standard bypass techniques; whenever possible, the internal thoracic artery was used for revascularization of the left anterior descending coronary artery. When possible, complete revascularization was performed using arterial conduits or saphenous vein grafts. This study was approved by our local institutional review board.
End points of the study were death; composite of death, MI, or stroke; and repeat revascularization. Death was defined as death from any cause. A diagnosis of acute MI was defined as complications at index admission (defined as new pathologic Q waves after index treatment) or follow-up MI requiring subsequent hospitalizations (i.e., emergency admission with a principal diagnosis of MI), as described previously. Stroke, as indicated by neurologic deficits, was confirmed by a neurologist based on imaging studies. Repeat revascularization included target vessel revascularization regardless of whether the procedure was clinically or angiographically driven and nontarget-vessel revascularization. In the DES group, stent thrombosis was defined as definite or probable events according to the Academic Research Consortium classification. All outcomes of interest were carefully verified and adjudicated by independent clinicians. The diabetic subgroup was defined as all patients actively receiving treatment with oral hypoglycemic agents or insulin.
The registry prospectively contains information on patient demographics, coexisting clinical conditions, hemodynamic status, left ventricular function, extent of disease, details of procedures, and in-hospital and follow-up outcomes by independent research personnel. Patients were clinically followed 1 month and 6, and 12 months after the procedure and annually thereafter by office visit or telephone contact. The follow-up period was through January 31, 2010 to ensure that all patients had an opportunity for ≥4 years and up to approximately 7 years of follow-up. For validation of complete follow-up data on mortality, information about vital status was obtained through January 31, 2010 from the National Population Registry of the Korea National Statistical Office using a unique personal identification number.
Treatment-related differences in long-term outcomes between the 2 procedures were analyzed separately in patients with and without medically treated DM. Prevalence rates of risk factors and characteristics of the patients in the 2 treatment groups were compared using t test or Wilcoxon rank-sum test for continuous variables and with chi-square statistics or Fisher’s exact test for categorical variables. Survival curves were constructed using the Kaplan–Meier method and compared using log-rank test.
Differences in risk-adjusted long-term rates of study outcomes between patients in the DES and CABG groups were assessed using multivariable Cox proportional hazards regression. Adjusted covariates included patient age and gender, presence or absence of different clinical and coexisting conditions, left ventricular function, and number and extent of diseased vessels. The proportional hazards assumption was confirmed by examination of log(–log [survival]) curves and by testing of partial (Schoenfeld) residuals, and no relevant violations were found. To decrease the impact of treatment selection bias and potential confounding in an observational study, we also performed rigorous adjustment for baseline differences using weighted Cox proportional hazards regression models with inverse probability-of-treatment weighting. Weights for patients undergoing CABG were the inverse of (1 – propensity score), and weights for patients undergoing stenting were the inverse of the propensity score. Propensity scores were estimated without regard to outcomes using multiple logistic regression analysis. We developed a full nonparsimonious model that included all variables listed in Table 1 . Model discrimination was assessed with c-statistics, and model calibration was assessed with Hosmer–Lemeshow statistics. Treatment effects were evaluated separately in diabetic and nondiabetic patients. Then, interaction terms in the multivariate Cox model and weighted Cox model using the inverse probability-of-treatment weighting method were used to test for the statistical significance of the effects of the 2 treatment strategies according to diabetic status on clinical outcomes.
| Variable | Diabetic Patients (n = 891) | Nondiabetic Patients (n = 2,151) | ||||
|---|---|---|---|---|---|---|
| DES | CABG | p Value | DES | CABG | p Value | |
| (n = 489) | (n = 402) | (n = 1,058) | (n = 1,093) | |||
| Age (years) | 63.5 ± 9.3 | 62.8 ± 8.1 | 0.19 | 61.3 ± 10.5 | 61.4 ± 8.7 | 0.76 |
| Men | 304 (62.2%) | 275 (68.4%) | 0.06 | 769 (72.7%) | 821 (75.1%) | 0.20 |
| Body mass index (kg/m 2 ) | 25.3 ± 3.1 | 24.6 ± 3.0 | 0.001 | 24.9 ± 2.9 | 24.8 ± 3.0 | 0.43 |
| Medically treated diabetes | 0.66 | |||||
| Oral hypoglycemic agent | 403 (82.4%) | 326 (81.1%) | — | — | — | |
| Requiring insulin | 86 (17.6%) | 76 (18.9%) | — | — | — | |
| Years with diabetes | 9.9 ± 8.2 | 10.9 ± 8.4 | 0.08 | — | — | — |
| Hypertension ⁎ | 313 (64.0%) | 247 (61.4%) | 0.44 | 570 (53.9%) | 320 (29.3%) | <0.001 |
| Current smoker | 113 (23.1%) | 66 (16.4%) | 0.02 | 344 (32.5%) | 138 (12.6%) | <0.001 |
| Hyperlipidemia † | 105 (21.5%) | 202 (50.2%) | <0.001 | 268 (25.3%) | 272 (24.9%) | 0.84 |
| Previous coronary angioplasty | 88 (18.0%) | 59 (14.7%) | 0.2 | 182 (17.2%) | 90 (8.2%) | <0.001 |
| Previous congestive heart failure | 11 (2.2%) | 28 (7.0%) | 0.001 | 11 (1.0%) | 40 (3.7%) | <0.001 |
| Moderate or severe chronic obstructive pulmonary disease | 6 (1.2%) | 8 (2.0%) | 0.42 | 10 (0.9%) | 20 (1.83%) | 0.08 |
| Cerebrovascular or carotid artery disease | 36 (7.4%) | 69 (17.2%) | <0.001 | 49 (4.6%) | 98 (9.0%) | <0.001 |
| Peripheral vascular disease | 17 (3.5%) | 43 (10.7%) | <0.001 | 17 (1.6%) | 71 (6.5%) | <0.001 |
| Renal failure | 26 (5.3%) | 43 (10.7%) | 0.004 | 16 (1.5%) | 44 (4.0%) | <0.001 |
| EuroSCORE | .7 ± 2.5 | 4.5 ± 2.7 | <0.001 | 3.1 ± 2.3 | 3.7 ± 2.4 | <0.001 |
| Previous myocardial infarction | <0.001 | <0.001 | ||||
| 1–7 days before treatment | 29 (5.9%) | 21 (5.2%) | 109 (10.3%) | 63 (5.8%) | ||
| ≥8 days before treatment | 6 (1.2%) | 88 (21.9%) | 12 (1.1%) | 122 (11.2%) | ||
| No previous myocardial infarction | 454 (92.8%) | 293 (72.9%) | 937 (88.6%) | 908 (83.1%) | ||
| Electrocardiographic findings | 0.82 | 0.01 | ||||
| Sinus rhythm | 458 (93.7%) | 381 (94.8%) | 1,000 (94.5%) | 1,059 (96.9%) | ||
| Atrial fibrillation | 17 (3.5%) | 12 (3.0%) | 31 (2.9%) | 14 (1.3%) | ||
| Others | 14 (2.9%) | 9 (2.2%) | 27 (2.6%) | 20 (1.8%) | ||
| Ejection fraction (%) | <0.001 | <0.001 | ||||
| <30% | 9 (1.9%) | 20 (5.1%) | 5 (0.5%) | 29 (2.7%) | ||
| 30%–40% | 8 (1.7%) | 40 (10.1%) | 25 (2.4%) | 57 (5.3%) | ||
| 40%–50% | 58 (12.0%) | 42 (10.6%) | 86 (8.4%) | 138 (12.8%) | ||
| ≥50% | 407 (84.4%) | 293 (74.2%) | 910 (88.7%) | 854 (79.2%) | ||
| Data missing | 7 (1.4%) | 7 (1.7%) | 0.79 | 32 (3.0%) | 15 (1.4%) | 0.01 |
| Mean ejection fraction (%) | 58.3 ± 9.2% | 54.7 ± 12.2% | <0.001 | 59.0 ± 8.3% | 56.8 ± 10.5% | <0.001 |
| 2-Vessel disease | 258 (52.8%) | 66 (16.4%) | <0.001 | 610 (57.7%) | 277 (25.3%) | <0.001 |
| With proximal left anterior descending coronary artery disease | 89 (18.2%) | 39 (9.7%) | <0.001 | 230 (21.7%) | 107 (9.8%) | <0.001 |
| Without proximal left anterior descending coronary artery disease | 169 (34.6%) | 27 (6.7%) | <0.001 | 380 (35.9%) | 170 (15.6%) | <0.001 |
| 3-Vessel disease | 231 (47.2%) | 336 (83.6%) | <0.001 | 448 (42.3%) | 816 (74.7%) | <0.001 |
| With proximal left anterior descending coronary artery disease | 106 (21.7%) | 216 (53.7%) | <0.001 | 176 (16.6%) | 442 (40.4%) | <0.001 |
| Without proximal left anterior descending coronary artery disease | 125 (25.6%) | 120 (29.9%) | <0.001 | 272 (25.7%) | 374 (34.2%) | <0.001 |
| Left main coronary artery disease | 46 (9.4%) | 103 (25.6%) | <0.001 | 132 (12.5%) | 269 (24.6%) | <0.001 |
| Total occlusion | 29 (5.9%) | 189 (47.0%) | <0.001 | 81 (7.7%) | 467 (42.7%) | <0.001 |
| Discharge medications | ||||||
| Aspirin | 484 (99.0%) | 385 (95.8%) | 0.004 | 1,050 (99.2%) | 1,061 (97.1%) | <0.001 |
| Clopidogrel | 481 (98.4%) | 316 (78.6%) | <0.001 | 1,048 (99.1%) | 716 (65.5%) | <0.001 |
| Aspirin and clopidogrel | 481 (98.4%) | 316 (78.6%) | <0.001 | 1,048 (99.1%) | 716 (65.5%) | <0.001 |
| Statin | 331 (67.7%) | 230 (57.2%) | 0.001 | 692 (65.4%) | 506 (46.3%) | <0.001 |
| Angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers | 214 (43.8%) | 139 (34.6%) | 0.006 | 347 (32.8%) | 279 (25.5%) | <0.001 |
| β Blockers | 426 (87.1%) | 112 (27.9%) | <0.001 | 925 (87.4%) | 245 (22.4%) | <0.001 |
| Calcium channel blockers | 430 (87.9%) | 321 (79.9%) | 0.001 | 913 (86.3%) | 894 (81.8%) | 0.005 |
| Nitrates | 374 (76.5%) | 369 (91.8%) | <0.001 | 809 (76.5%) | 923 (84.4%) | <0.001 |
| SYNTAX score in available cohort (n = 1,914) | ||||||
| Number of patients | 443 | 217 | 957 | 297 | ||
| Mean SYNTAX score | 18.3 ± 7.9 | 30.4 ± 10.7 | <0.001 | 17.0 ± 7.7 | 29.5 ± 10.5 | <0.001 |
| SYNTAX score category | <0.001 | <0.001 | ||||
| Low (≤22) | 324 (73.1%) | 53 (24.4%) | 744 (77.7%) | 76 (25.6%) | ||
| Intermediate (23–32) | 95 (21.4%) | 79 (36.4%) | 179 (18.7%) | 105 (35.4%) | ||
| High (≥33) | 24 (5.4%) | 85 (39.2%) | 34 (3.6%) | 116 (39.1%) | ||
⁎ Defined as systolic blood pressure ≥140 mm Hg, or diastolic blood pressure ≥90 mm Hg, or receiving antihypertensive treatment.
† Defined as total cholesterol >200 mg/dl or receiving antilipidemic treatment.
In addition, outcomes were analyzed based on extent of diseased vessels (2- or 3-vessel disease) and Synergy between Percutaneous Coronary Intervention with TAXUS and Cardiac Surgery (SYNTAX) score.
All reported p values are 2-sided, and p values <0.05 were considered statistically significant. No adjustments were performed for multiple testing in several subgroups. SAS 9.1 (SAS Institute, Cary, North Carolina) and the R programming language (R Foundation for Statistical Computing, Vienna, Austria) were used for statistical analyses.
Results
From January 2003 through December 2005, 891 diabetic patients and 2,151 nondiabetic patients with multivessel CAD underwent percutaneous coronary intervention with DES implantation or CABG. Baseline characteristics of study patients are listed in Table 1 . Compared to patients undergoing DES, diabetic and nondiabetic patients undergoing CABG had higher-risk profiles for clinical and angiographic characteristics. Of diabetic patients who underwent percutaneous coronary intervention, 77.1% received sirolimus-eluting stents and 22.9% received paclitaxel-eluting stents; in nondiabetic patients, 80.5% received sirolimus-eluting stents and 19.5% received paclitaxel-eluting stents. Median follow-up in the overall population was 5.6 years (interquartile range 4.6 to 6.3), with 97.4% undergoing complete follow-up for major clinical events including 97.7% of patients in the DES group and 97.0% in the CABG group (p = 0.20). Predictors of choice of revascularization strategy are listed in Table 2 .
| Predictor | Diabetic Patients | Nondiabetic Patients | ||||
|---|---|---|---|---|---|---|
| OR (95% CI) | Chi-Square | p Value | OR (95% CI) | Chi-Square | p Value | |
| Age | 1.00 (0.97–1.03) | 1.66 | 0.95 | 1.01 (0.99–1.03) | 0.18 | 0.20 |
| Male gender | 0.87 (0.56–1.37) | 3.04 | 0.55 | 0.82 (0.61–1.09) | 1.94 | 0.18 |
| Body mass index (kg/m 2 ) | 1.04 (0.98–1.11) | 12.07 | 0.24 | 1.02 (0.98–1.06) | 0.33 | 0.32 |
| Hypertension ⁎ | 1.21 (0.80–1.81) | 0.02 | 0.37 | 3.32 (2.60–4.25) | 137.53 | <0.0001 |
| Current smoker | 1.77 (1.06–2.96) | 13.61 | 0.03 | 4.81 (3.51–6.58) | 127.88 | <0.0001 |
| Hyperlipidemia † | 0.24 (0.16–0.36) | 84.95 | <0.0001 | 0.88 (0.67–1.15) | 8.86 | 0.34 |
| Insulin requiring diabetes | 1.04 (0.60–1.80) | 0.02 | 0.89 | — | — | — |
| Years with diabetes | 1.00 (0.97–1.03) | 0 | 0.96 | — | — | — |
| Previous coronary angioplasty | 0.85 (0.51–1.40) | 1.18 | 0.52 | 2.41 (1.67–3.49) | 31.82 | <0.0001 |
| Previous congestive heart failure | 1.14 (0.39–3.33) | 5.50 | 0.82 | 0.50 (0.20–1.25) | 9.71 | 0.14 |
| Moderate or severe chronic obstructive pulmonary disease | 0.45 (0.11–1.87) | 2.08 | 0.27 | 0.61 (0.22–1.66) | 1.74 | 0.33 |
| Cerebrovascular or carotid artery disease | 0.34 (0.15–0.80) | 13.53 | 0.01 | 0.45 (0.25–0.80) | 21.69 | 0.007 |
| Peripheral vascular disease | 0.49 (0.20–1.17) | 10.51 | 0.11 | 0.23 (0.11–0.47) | 33.51 | <0.0001 |
| Renal failure | 0.46 (0.20–1.07) | 2.22 | 0.07 | 0.52 (0.24–1.14) | 6.06 | 0.10 |
| EuroSCORE | 1.09 (0.93–1.27) | 7.51 | 0.31 | 1.03 (0.94–1.13) | 10.47 | 0.58 |
| Previous myocardial infarction | 0.28 (0.15–0.52) | 39.54 | <0.0001 | 0.58 (0.40–0.84) | 16.32 | 0.004 |
| Ejection fraction (%) | 1.00 (0.98–1.02) | 1.90 | 0.92 | 1.01 (0.99–1.02) | 5.95 | 0.48 |
| 2-Vessel disease | 3.20 (2.11–4.86) | 37.16 | <0.0001 | 2.26 (1.78–2.87) | 49.16 | <0.0001 |
| Proximal left anterior descending coronary artery disease | 0.60 (0.41–0.89) | 6.53 | 0.01 | 1.34 (1.05–1.71) | 5.44 | 0.02 |
| Left main disease | 0.74 (0.44–1.27) | 24.11 | 0.28 | 0.56 (0.41–0.76) | 53.79 | <0.0001 |
| Total occlusion | 0.12 (0.07–0.21) | 71.00 | <0.0001 | 0.21 (0.15–0.29) | 97.80 | <0.0001 |
| SYNTAX score | 0.91 (0.89–0.93) | 194.10 | <0.0001 | 0.90 (0.89–0.92) | 524.77 | <0.0001 |
⁎ Defined as systolic blood pressure ≥140 mm Hg, or diastolic blood pressure ≥90 mm Hg, or receiving antihypertensive treatment.
† Defined as total cholesterol >200 mg/dl or receiving antilipidemic treatment.
Unadjusted event-free survival curves are shown in Figure 1 and crude and adjusted risks according to treatment approach and diabetic status are presented in Table 3 . After adjustment for differences in baseline risk factors between treatment procedures using multivariable Cox regression analysis and weighted Cox regression using inverse probability-of-treatment weighting methods, adjusted treatment-related risks of death and the composite of death, MI, or stroke did not differ significantly in diabetic and nondiabetic patients. Adjusted risk of repeat revascularization was consistently higher in the DES than in the CABG group. When we tested the interaction between diabetic status and treatment strategy on clinical outcomes, we found no statistically significant interactions after adjustment for possible confounders ( Table 3 ).
| Outcomes | Total Number of Events/Number of Patients | Unadjusted | Multivariable Adjusted † | Adjusted by IPTW | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| DES | CABG | HR (95% CI) | p Value | HR (95% CI) | p Value | Interaction p Value for Diabetic Status | HR (95% CI) | p Value | Interaction p Value for Diabetic Status | |
| Death | 0.36 | |||||||||
| Diabetic patients | 57/489 | 60/402 | 0.82 (0.57–1.17) | 0.27 | 1.37 (0.86–2.17) | 0.19 | 0.32 | 1.01 (0.77–1.33) | 0.96 | |
| Nondiabetic patients | 72/1,058 | 115/1,093 | 0.68 (0.51–0.91) | 0.01 | 0.85 (0.63–1.15) | 0.30 | 0.80 (0.55–1.16) | 0.23 | ||
| Composite outcome (death, myocardial infarction, or stroke) | 0.12 | 0.20 | ||||||||
| Diabetic patients | 72/489 | 76/402 | 0.80 (0.58–1.10) | 0.16 | 1.38 (0.92–2.08) | 0.12 | 1.03 (0.80–1.31) | 0.91 | ||
| Nondiabetic patients | 99/1,058 | 158/1,093 | 0.67 (0.52–0.86) | 0.002 | 0.79 (0.61–1.02) | 0.07 | 0.77 (0.56–1.05) | 0.10 | ||
| Repeat revascularization | 0.46 | 0.40 | ||||||||
| Diabetic patients | 91/489 | 22/402 | 3.88 (2.43–6.20) | <0.001 | 3.61 (2.25–5.77) | <0.001 | 3.69 (2.64–5.17) | <0.001 | ||
| Nondiabetic patients | 168/1,058 | 65/1,093 | 3.12 (2.33–4.16) | <0.001 | 3.12 (2.34–4.17) | <0.001 | 2.77 (1.95–3.91) | <0.001 | ||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
