Patients with diabetes mellitus have worse outcomes after percutaneous coronary intervention than patients without diabetes mellitus. We compared the risk of stent thrombosis, myocardial infarction, death, and target lesion revascularization in diabetic and nondiabetic patients after implantation of drug-eluting stents or bare metal stents. In the Western Denmark Heart Registry, 12,347 consecutive patients (1,575 with and 10,772 without diabetes) were identified and followed up for 2 years. The 2-year risk of definite stent thrombosis was 0.52% in patients with diabetes mellitus and 0.71% in nondiabetic patients (adjusted relative risk [RR] 0.74, 95% confidence interval [CI] 0.41 to 1.34, p = 0.321). The 2-year risk of myocardial infarction was greater in the diabetic patients (6.9%) than in the nondiabetic patients (3.6%; adjusted RR 1.96, 95% CI 1.58 to 2.43; p <0.001). The all-cause 2-year mortality rate was almost twice as great for the diabetic patients compared to the nondiabetic patients (12.4% vs 6.7%; adjusted RR 1.91, 95% CI 1.63 to 2.23; p <0.001). The 2-year risk of target lesion revascularization was 8.5% in the diabetic patients and 6.8% in the nondiabetic patients (adjusted RR 1.28, 95% CI 1.10 to 1.49; p <0.001). In conclusion, 2 years after drug-eluting stent or bare metal stent implantation, diabetic patients had a greater risk than nondiabetic patients of myocardial infarction and death. Drug-eluting stent treatment reduced the risk of target lesion revascularization compared to bare metal stent treatment, regardless of diabetes status.
Patients with diabetes mellitus have tended to have worse outcomes after percutaneous coronary intervention (PCI) than nondiabetic patients. Angiographic and intravascular ultrasound studies have suggested increased late lumen loss and intimal hyperplasia as potential mechanisms underlying the greater restenosis risk. A more diffuse and accelerated form of atherosclerosis in diabetic patients, accompanied by small vessel size, long lesions, and greater plaque burden, might contribute to their well-documented increased risk of restenosis after stent implantation. Drug-eluting stents (DESs) have shown promising results in patients with and without diabetes mellitus. Although the initial use of DESs was not associated with safety concerns, recent studies have reported increased risks of stent thrombosis, myocardial infarction (MI), and death associated with the use of DESs compared to bare metal stent (BMS) use. The present study examined the influence of diabetes mellitus on the long-term outcomes of patients treated with PCI using DESs or BMSs, with a recommended 12-month duration of dual antiplatelet therapy.
Methods
We conducted the present follow-up study using Western Denmark’s healthcare databases, covering the region’s entire population (approximately 3.0 million inhabitants; 55% of the Danish population). All patients were followed up for 24 months. A detailed description of the databases has been previously reported. In brief, the Danish National Health Service provides universal tax-supported healthcare, guaranteeing patients free access to general practitioners and hospitals. Our data came from the Danish Civil Registration System, which has kept electronic records on gender, date of birth, residence, date of emigration, and changes in vital status since 1968. The information on vital status is updated daily. The records include a unique 10-digit civil registration number that is assigned at birth and is used in all public registries, allowing accurate record linkage. The National Registry of Causes of Deaths and National Patient Registry was used to obtain the causes of death and the diagnoses assigned by the treating physician during hospitalization and coded according to the International Classification of Diseases , eighth revision until the end of 1993 and the tenth revision thereafter.
We used the Western Denmark Heart Registry to identify all PCIs recorded from January 1, 2002 to June 30, 2005. The Western Denmark Heart Registry collects detailed patient and procedure data for all interventions performed in western Denmark’s 3 coronary intervention centers (Odense University Hospital, Aarhus University Hospital Skejby, and Aarhus University Hospital Aalborg). For each patient, we included only the first PCI procedure performed during the study period (the index procedure). The DES used was either a Cypher stent or a Taxus Express stent. We excluded patients treated with balloon angioplasty only or a combination of BMSs and DESs (n = 645, 4.9%). The post-PCI antiplatelet regimens included lifelong acetylsalicylic acid (75 to 150 mg/day) and clopidogrel (loading dose of 300 mg followed by 75 mg/day). Since November 2002, the recommended duration of clopidogrel treatment has been 12 months for both stent types. Patients were considered to have diabetes if their Western Denmark Heart Registry records indicated receipt of dietary treatment, oral antidiabetic medication, or insulin.
The study end points were the interval to stent thrombosis (classified as definite, probable, or possible), MI, all-cause mortality, cardiac death, and target lesion revascularization (TLR). The end point events were ascertained from the Western Denmark Heart Registry, the Danish National Patient Registry, which tracks all hospitalizations in Denmark, and the Danish Registry of Causes of Death.
We defined the types of stent thrombosis according to the Academic Research Consortium definition, with a modification for probable stent thrombosis. Probable stent thrombosis was assumed for any unexplained death within the first 30 days after intracoronary stenting.
We defined a new MI as hospitalization for MI occurring >28 days after the index PCI. We ascertained the admissions and readmissions for MI ( International Classification of Diseases , tenth revision, codes I21 to I21.9) from the National Patient Registry and deaths from the Civil Registration System. We validated the recorded cause of death using the original death certificates obtained from the National Registry of Causes of Death and classified deaths according to their underlying cause.
From the Western Denmark Heart Registry we ascertained TLR, defined as repeat PCI of the index lesion or coronary artery bypass grafting. For all cases of stent thrombosis, we retrieved the relevant medical records and catheterization films.
From the Western Denmark Heart Registry, we also retrieved data on other potential predictors of subsequent cardiovascular events. For each patient, we also obtained data on all hospital diagnoses from the National Patient Registry and computed the co-morbidity scores using the Charlson Co-morbidity Index, which covers 19 major disease categories, including diabetes mellitus, heart failure, cerebrovascular diseases, and cancer. The index value is a weighted summary of the diagnoses, such that the weight is based on the 1-year mortality associated with each disease in the original Charlson data set. The data for all key patient and procedure characteristics were >95% complete, and the ascertainment of end points (stent thrombosis, death, MI, and TLR) was 100% complete.
The distributions of continuous variables in the 2 groups (with or without diabetes) were compared using either the 2-sample t test or the Mann-Whitney U test, depending on whether the data followed the normal distribution. We compared the distributions of categorical variables using the chi-square test.
We counted the end point events that occurred during the follow-up period and compared their rates for the 2 cohorts of patients, with and without diabetes mellitus. Follow-up began on the date of the index PCI procedure. In the analyses with stent thrombosis, MI, or death as the outcome, the follow-up period continued until the date of the respective event, death, emigration, or 24 months after implantation, whichever came first. We constructed Kaplan-Meier curves for patients with and without diabetes, stratified by lesion type treated. We used the life-table method to compute the 2-year risk of each end point (proportion of the population at risk with the outcome of interest). We used Cox proportional hazards regression analysis to estimate the relative risk (RR) for each end point. Because the hazards were not proportional throughout the follow-up period, we estimated the RRs within the periods during which the proportionality assumption held. The RR in these analyses reflected the risk among patients alive and at risk of a specific end point at the start of each period (eg, after 30 days or 1 year of follow-up). In all regression analyses, we included the age, gender, diabetes mellitus status, clinical indication, procedure duration, number of stents, and co-morbidities (and stent length and size of the reference vessel in the lesion-specific analyses). To maximize the precision of the RR estimates, we used the change-in-estimate method, which entailed retaining only those variables that changed the RR estimates for an outcome by >10%. The number of variables included in the final regression models varied from 0 to 3. We used Statistical Analysis Systems software, version 9.2 (SAS Institute, Cary, North Carolina).
Results
We included 12,347 consecutive patients with a total of 17,147 lesions. Of these 12,347 patients, 1,575 (12.8%; total of 2,301 lesions) had diabetes mellitus. The median patient age was 64.3 years (range 56 to 72). The prevalence of patients >75 years old was similar for those with and without diabetes (18.9% vs 18.6%, p = 0.756). The indications for PCI were ST-segment elevation MI (30.0%), non-ST-segment elevation MI/unstable angina (30.4%), stable angina (36.9%), and other (2.7%). The indication for PCI differed between those with and without diabetes. The indication for the diabetic patients was stable angina in 44.3%, non–ST-segment elevation MI in 31.7%, ST-segment elevation MI in 20.4%, and other in 3.4%. The indication for the nondiabetic patients was stable angina in 35.8%, non–ST-segment elevation MI in 30.2%, ST-segment elevation MI in 31.4%, and other in 2.6% (p <0.0001). The baseline patient, procedure, and lesion characteristics differed substantially between those with and without diabetes, and this difference was also seen after stratifying by treatment with DES versus BMS. Patients with diabetes were more likely to have a history of revascularization, smaller vessels (3.2 ± 0.6 vs 3.3 ± 0.6 mm; p <0.001) and a greater rate of DES treatment (42.4% vs 29.9%; p <0.001). They were also slightly older (median age 65.0 years, range 57 to 72 vs 64.0 years, range 56.0 to 72.0; p = 0.01) and included more women (32.4% vs 27.2%; p <0.01; Tables 1 and 2 ).
| Variable | Diabetic Patients | Nondiabetic Patients | ||||
|---|---|---|---|---|---|---|
| DES (n = 593) | BMS (n = 982) | p Value | DES (n = 2,906) | BMS (n = 7,866) | p Value | |
| Men | 416 (70.2%) | 648 (66.0%) | 0.087 | 2,117 (72.8%) | 5,727 (72.8%) | 0.993 |
| Age (years) | <0.01 | <0.001 | ||||
| Median | 62.0 | 66.0 | 62.0 | 64.0 | ||
| Interquartile range | 56.0–70.0 | 58.0–73.0 | 54.0–70.0 | 56.0–73.0 | ||
| Family history | 234 (45.5%) | 331 (40.6%) | 0.075 | 1,167 (46.5%) | 2,860 (42.9%) | 0.002 |
| Smoker | 170 (30.2%) | 255 (31.4%) | 0.495 | 1,009 (40.6%) | 2,809 (42.3%) | 0.155 |
| Hypertension | 345 (64.7%) | 490 (57.2%) | 0.006 | 1,056 (41.7%) | 2,573 (37.6%) | <0.001 |
| Previous coronary bypass | 57 (10.6%) | 83 (9.6%) | 0.510 | 180 (7.1%) | 400 (5.7%) | 0.012 |
| Previous percutaneous coronary intervention | 69 (13.5%) | 101 (12.1%) | 0.472 | 271 (11.2%) | 638 (9.5%) | 0.021 |
| Previous myocardial infarction | 187 (36.7%) | 303 (36.2%) | 0.863 | 604 (24.8%) | 2,035 (30.3%) | <0.001 |
| Lipid-lowering therapy | 364 (67.8%) | 473 (55.0%) | <0.001 | 1,367 (53.7%) | 2,849 (41.6%) | <0.001 |
| Procedure time (minutes) | 31.0 ± 19.9 | 28.3 ± 19.4 | 0.002 | 29.8 ± 20.2 | 27.2 ± 19.3 | <0.001 |
| Flouroscopy time (minutes) | 11.1 ± 8.5 | 10.9 ± 9.7 | 0.070 | 10.6 ± 8.4 | 10.1 ± 9.9 | <0.001 |
| Contrast (ml) | 163.2 ± 106.1 | 159.3 ± 110.0 | 0.270 | 157.2 ± 97.5 | 155.1 ± 104.0 | <0.001 |
| No. of treated lesions | 1.6 ± 0.9 | 1.4 ± 0.7 | <0.001 | 1.4 ± 0.7 | 1.3 ± 0.6 | 0.002 |
| Indication for percutaneous coronary intervention | <0.001 | <0.001 | ||||
| Stable angina pectoris | 314 (53.0%) | 385 (39.2%) | 1,271 (43.7%) | 2,589 (32.9%) | ||
| Non–ST-segment elevation myocardial infarction/unstable angina pectoris | 173 (29.2%) | 327 (33.3%) | 869 (29.9%) | 2,383 (30.3%) | ||
| ST-segment elevation myocardial infarction | 84 (14.2%) | 237 (24.1%) | 675 (23.2%) | 2,709 (34.4%) | ||
| Other | 22 (3.7%) | 33 (3.4%) | 92 (3.2%) | 184 (2.3%) | ||
| Co-morbidity index score | 0.190 | 0.561 | ||||
| 0 | 295 (49.7%) | 535 (54.5%) | 1,947 (67.0%) | 5,234 (66.5%) | ||
| 1–2 | 244 (41.1%) | 366 (37.3%) | 805 (27.7%) | 2,169 (27.6%) | ||
| 3+ | 54 (9.1%) | 81 (8.2%) | 155 (5.3%) | 462 (5.9%) | ||
| Variable | Diabetic Patients | Nondiabetic Patients | ||||
|---|---|---|---|---|---|---|
| DES | BMS | p Value | DES | BMS | p Value | |
| No. of lesions | 978 | 1,323 | 4,439 | 10,396 | ||
| Vessel | <0.001 | <0.001 | ||||
| Right | 173 (29.2%) | 327 (33.3%) | 1,078 (24.3%) | 4,180 (40.2%) | ||
| Left anterior descending | 314 (53.0%) | 385 (39.2%) | 2,359 (53.1%) | 3,787 (26.4%) | ||
| Left circumflex | 84 (14.2%) | 237 (24.1%) | 846 (19.1%) | 2,346 (22.6%) | ||
| Left main | 22 (3.7%) | 33 (3.4%) | 156 (3.5%) | 83 (0.8%) | ||
| Saphenous vein graft | 17 (1.3%) | 17 (1.7%) | 0.463 | 39 (0.9%) | 93 (0.9%) | 0.927 |
| Lesion length (mm) | 16.0 ± 9.6 | 13.5 ± 7.4 | <0.001 | 16.0 ± 9.5 | 13.5 ± 7.2 | <0.001 |
| Lesion type | <0.001 | <0.001 | ||||
| A | 217 (22.2%) | 273 (20.6%) | 932 (21.0%) | 2,281 (21.9%) | ||
| B1 | 378 (38.7%) | 793 (20.6%) | 1,789 (40.3%) | 6,115 (58.8%) | ||
| B2 | 79 (8.1%) | 155 (11.7%) | 437 (9.8%) | 1,157 (11.1%) | ||
| C | 304 (31.1%) | 102 (7.7%) | 1,281 (28.9%) | 846 (8.1%) | ||
| Restenotic lesion | 11 (0.8%) | 21 (2.2%) | 0.018 | 107 (1.0%) | 108 (2.4%) | <0.001 |
| Stent length (mm) | 19.3 ± 8.8 | 16.6 ± 7.4 | <0.001 | 19.2 ± 8.6 | 16.9 ± 7.1 | <0.001 |
| Stents used (n) | 1.2 ± 0.6 | 1.2 ± 0.5 | 0.005 | 1.2 ± 0.6 | 1.2 ± 0.5 | <0.001 |
| Maximum balloon pressure (atm) | 16.3 ± 4.1 | 15.3 ± 3.6 | <0.001 | 15.9 ± 4.1 | 15.2 ± 3.5 | <0.001 |
| Maximum balloon diameter (mm) | 3.3 ± 0.5 | 3.4 ± 0.6 | <0.001 | 3.3 ± 0.5 | 3.5 ± 0.6 | <0.001 |
| Reference segment (mm) | 3.2 ± 0.5 | 3.3 ± 0.6 | 0.001 | 3.2 ± 0.5 | 3.3 ± 0.6 | <0.001 |
| Minimum lumen diameter (mm) | 0.46 ± 0.44 | 0.45 ± 0.53 | 0.02 | 0.41 ± 0.47 | 0.43 ± 0.53 | 0.499 |
| Stenosis (% luminal diameter) | 86.0 ± 12.5 | 87.1 ± 12.4 | 0.002 | 87.8 ± 12.4 | 87.8 ± 12.7 | 0.124 |
Comparing the patients with and without diabetes, we found no significant difference in the incidence of definite stent thrombosis, with an occurrence in 12 lesions in 12 diabetic patients (2-year risk 0.52%) and in 106 lesions in 106 patients without diabetes (2-year risk 0.71%, adjusted RR 0.74, 95% CI 0.41 to 1.34). The risk of acute, subacute, and late definite stent thrombosis was also similar between the 2 groups ( Figure 1 ). Very late definite stent thrombosis occurred in 1 lesion in 1 patient with diabetes mellitus (2-year risk 0.04%) and in 17 lesions in 17 patients without diabetes (2-year risk 0.11%). None of the 118 cases of definite stent thrombosis occurred in saphenous vein grafts.
Of the 118 patients who developed definite stent thrombosis, 92 (78%) were receiving dual antiplatelet therapy (aspirin and clopidogrel) at the time of the thrombotic event. Of the 18 patients with very late stent thrombosis, 3 (17%) were receiving dual antiplatelet therapy, 11 (61%) were receiving aspirin only, and 4 (22%) had discontinued both aspirin and clopidogrel.
Definite, probable, or possible stent thrombosis was found in 62 patients with diabetes mellitus (2-year risk 3.9%) and in 237 patients without diabetes mellitus (2-year risk 2.2%, RR 1.83, 95% CI 1.38 to 2.42; Figure 1 ). Controlling for covariates did not attenuate this estimate (adjusted RR 1.59, 95% CI 1.20 to 2.10; Table 3 ). Definite, probable, or possible stent thrombosis did not vary according to stent type.
| Variable | Diabetic Patients | Nondiabetic Patients | Adjusted RR (95% CI) for Diabetes vs No Diabetes | Adjusted RR (95% CI) for DES vs BMS | |||
|---|---|---|---|---|---|---|---|
| DES | BMS | DES | BMS | Diabetic Patients | Nondiabetic Patients | ||
| All-cause death | 65 (11.0%) | 131 (13.3%) | 160 (5.5%) | 564 (7.2%) | 1.91 (1.63–2.23) | 0.80 (0.59–1.08) | 0.90 (0.76–1.07) |
| ≤12 months | 40 (6.7%) | 96 (9.8%) | 100 (3.4%) | 407 (5.2%) | 1.87 (1.55–2.26) | 0.67 (0.47–0.98) | 0.78 (0.63–0.97) |
| >12 months | 25 (4.5%) | 35 (4.0%) | 60 (2.1%) | 157 (2.1%) | 1.98 (1.49–2.63) | 1.43 (0.85–2.41) | 1.03 (0.76–1.38) |
| Cardiac death | 28 (4.7%) | 81 (8.2%) | 78 (2.7%) | 305 (3.9) | 1.99 (1.61–2.46) | 0.63 (0.41–0.98) | 0.83 (0.64–1.06) |
| Noncardiac death | 30 (5.1%) | 44 (4.5%) | 66 (2.3%) | 217 (2.8) | 1.69 (1.31–2.18) | 1.09 (0.69–1.74) | 0.81 (0.61–1.07) |
| Myocardial infarction | 43 (7.4%) | 61 (6.5%) | 112 (3.9%) | 268 (3.5) | 1.96 (1.58–2.43) | 1.14 (0.77–1.68) | 0.83 (0.64–1.06) |
| 28 days to 12 months | 28 (4.8%) | 45 (4.8%) | 60 (2.1%) | 193 (2.5%) | 2.01 (1.55–2.61) | 0.97 (0.60–1.56) | 0.78 (0.63–0.97) |
| ≥12 months | 15 (2.8%) | 16 (1.9%) | 52 (1.9%) | 75 (1.0%) | 1.85 (1.26–2.73) | 1.62 (0.81–3.24) | 1.84 (1.29–2.63) |
| Definite stent thrombosis | 6 (0.6%) | 6 (0.5%) | 37 (0.8%) | 69 (0.7%) | 0.74 (0.41–1.34) | 1.75 (0.55–5.53) | 1.25 (0.84–1.86) |
| Overall stent thrombosis | 22 (3.7%) | 40 (4.1%) | 59 (2.0%) | 178 (2.3%) | 1.59 (1.20–2.10) | 1.01 (0.60–1.71) | 0.89 (0.66–1.19) |
| <30 days | 5 (0.8%) | 18 (1.5%) | 27 (0.9%) | 102 (1.3%) | 1.02 (0.65–1.59) | 0.71 (0.25–2.00) | 0.81 (0.53–1.24) |
| 30 days to 24 months | 17 (3.0%) | 22 (2.4%) | 32 (1.1%) | 76 (1.1%) | 2.56 (1.78–3.69) | 1.24 (0.66–2.34) | 1.12 (0.74–1.70) |
| Probable stent thrombosis | 1 (0.2%) | 15 (1.5%) | 10 (0.3%) | 51 (0.6%) | 2.17 (1.25–3.77) | 0.16 (0.02–1.27) | 0.75 (0.38–1.48) |
| Possible stent thrombosis | 15 (2.5%) | 20 (2.0%) | 17 (0.6%) | 61 (0.8%) | 3.19 (2.14–4.75) | 1.20 (0.61–2.35) | 0.74 (0.43–1.27) |
| Target lesion revascularization | 64 (6.5%) | 132 (10.0%) | 224 (5.0%) | 789 (7.8%) | 1.28 (1.10–1.49) | 0.63 (0.47–0.85) | 0.65 (0.56–0.75) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
