Several studies have highlighted the prognostic role of preprocedural Thrombolysis In Myocardial Infarction (TIMI) flow in the infarct-related artery (IRA) in patients with ST-segment elevation myocardial infarction (STEMI). However, the impact of preprocedural IRA occlusion in patients with diabetes with STEMI has been insufficiently studied. The aim of this study was to evaluate the effects of baseline IRA occlusion and diabetic status in patients with STEMI who underwent primary percutaneous coronary intervention by using data from a pooled analysis of randomized trials comparing intracoronary with intravenous abciximab bolus administration. A total of 3,046 patients with STEMI who underwent primary percutaneous coronary intervention were included. Diabetes was present in 578 patients (19%). The primary outcome was mortality after a median follow-up period of 375 days. Secondary end points were reinfarction and stent thrombosis. In patients without diabetes, IRA occlusion versus no occlusion was not associated with increased rates of mortality (4.3% vs 2.7%, p = 0.051) and reinfarction (3.3% vs 2.5%, p = 0.33). Patients with diabetes with IRA occlusion compared with those without occlusion showed higher rates of mortality (10.6% vs 4.6%, p = 0.01) and reinfarction (5.6% vs 2.1%, p = 0.03). Baseline IRA occlusion increased the rate of stent thrombosis in the nondiabetic (2.1% vs 1.0%, p = 0.04) and diabetic (3.2% vs 0.8%, p = 0.05) cohorts. Interaction analysis demonstrated that the risk for death and reinfarction was significantly increased when diabetes and IRA occlusion occurred concomitantly. In conclusion, patients with STEMI with diabetes and baseline IRA occlusion had disproportionately higher rates of death and reinfarction. Preprocedural IRA occlusion increased the risk for stent thrombosis, irrespective of diabetic status.
Highlights
- •
The investigators assessed the role of diabetes mellitus and baseline coronary occlusion in patients with acute myocardial infarction who underwent primary percutaneous coronary intervention.
- •
A biologic interaction was found, with increased risk for death and reinfarction in patients presenting with diabetes mellitus and preprocedural occlusion.
- •
In contrast, the risk for stent thrombosis was higher in patients presenting with coronary occlusion, irrespective of diabetes status.
- •
Thus, in patients who underwent primary percutaneous coronary intervention, diabetes and baseline IRA occlusion, when both present, were associated with disproportionately higher rates of death and reinfarction.
ST-segment elevation myocardial infarction (STEMI) is commonly caused by atherosclerotic plaque rupture or erosion with superimposed thrombus that leads to abrupt coronary vessel occlusion. When performed expeditiously by an experienced team, primary percutaneous coronary intervention (PCI) is the preferred treatment in patients with STEMI because of its superiority to fibrinolysis in reducing the risk for cardiovascular events, including death. In recent decades, the prevalence of diabetes mellitus (DM) in patients with STEMI has steadily increased, but these patients continue to have at least twofold greater risk for death compared with patients without DM. However, the mechanisms underlying the greater STEMI-related mortality in patients with DM remain unclear, because this excess of mortality is independent of co-morbidities, left ventricular dysfunction, and coronary patency after reperfusion therapy. Although multiple studies have shown the prognostic role of preprocedural Thrombolysis In Myocardial Infarction (TIMI) flow grade in the infarct-related artery (IRA) in patients with STEMI, few data are available regarding this association in patients with DM. Against this background, we sought to evaluate the effects of baseline IRA occlusion and DM status in patients with STEMI who underwent primary PCI by using data from a pooled analysis of randomized trials comparing intracoronary with intravenous abciximab bolus administration.
Methods
Detailed data from this pooled analysis have previously been described. Our population is represented by a total of 3,158 patients enrolled in 5 randomized trials. Briefly, all patients with STEMI were admitted <12 hours after symptom onset and received dual-antiplatelet therapy with aspirin and a clopidogrel (300 to 600 mg) or prasugrel (60 mg) loading dose. Periprocedural anticoagulation consisted of intravenous unfractionated heparin in all cases. Patients were randomized to receive intracoronary (n = 1,590 [50.43%]) or intravenous (n = 1,568 [49.7%]) bolus abciximab at the time of primary PCI. In patients randomized to the intracoronary route, abciximab bolus was administered through the guiding catheter. We found no significant effect of intracoronary abciximab on reperfusion and clinical outcomes. For the purpose of this analysis, only patients with available information on baseline TIMI flow grade in the IRA were included. They were stratified according to the presence or the absence of DM and preprocedural IRA occlusion, as defined by TIMI flow grade 0 (vs 1 to 3). DM was defined as known DM at admission. The definitions of the study end points have previously been reported. The primary end point of the present analysis was death from any cause. Secondary end points included reinfarction and the composite of definite or probable stent thrombosis according to Academic Research Consortium definitions. Reperfusion end points were postprocedural TIMI grade 3 flow, myocardial blush grade (MBG) 2 or 3, and complete (>70%) ST-segment resolution (STR) at 60 to 90 minutes.
All analyses were carried out using SPSS software version 20.0 (IBM, Armonk, New York). Continuous variables are presented as mean ± SD or as median (interquartile range) according to their distribution. Categorical variables are expressed as counts and percentages. The normality of distribution of continuous variables was evaluated by using the Kolmogorov-Smirnov goodness-of-fit test and consequently compared with independent-samples Student’s t tests or Mann-Whitney U tests. Categorical variables were compared with chi-square or Fisher’s exact tests as appropriate. Survival analyses were performed using the Mantel-Cox method, and survival curves are presented as simple, nonstratified Kaplan-Meier curves across all trials. Cox regression analysis was used to test the differences in the risk for end points across 4 subgroups after correction for baseline characteristics. Risk estimates are expressed as odds ratios with 95% confidence intervals. To explore the effect of the biologic interaction between DM and coronary occlusion on the risk for study end points, we evaluated the interaction as departure from additivity according to the method proposed by Andersson et al. Three interaction measures were calculated: (1) relative excess risk due to interaction, (2) attributable proportion due to interaction, and (3) synergy index. The relative excess risk due to interaction is a measure of the excess of risk due to interaction relative to the risk without exposure. The attributable proportion due to interaction refers to the attributable proportion of disease that is due to interaction in patients with the 2 factors. The synergy index is the excess risk from exposure (to the 2 factors) when there is interaction relative to the risk from exposure (to the 2 factors) without interaction. Relative excess risk due to interaction >0, attributable proportion due to interaction >0, and synergy index >1 indicate the presence of a biologic interaction.
Results
After the exclusion of 112 patients with incomplete preprocedural TIMI flow grade data, a total of 3,046 patients were included in this study. Baseline characteristics are listed in Table 1 . DM was present in 578 patients (19%). Preprocedural IRA occlusion occurred at similar rates in patients with DM and those without DM (58.7% vs 60.1%, respectively, p = 0.52). There were no significant differences in baseline characteristics between the 2 groups among patients with DM. In the non-DM cohort, a higher rate of previous myocardial infarction, longer ischemic times, a different distribution of IRA, and a lower rate of anterior myocardial infarction location were observed in patients with preprocedural IRA occlusion ( Table 1 ). Median clinical follow-up duration was 375 days (interquartile range 30 to 372). Supplementary Tables 1 and 2 list reperfusion and clinical end points according to diabetic status and baseline IRA occlusion, respectively. Postprocedural reperfusion end points are presented in Figure 1 . Patients with occluded IRA on initial angiography tended to have worse reperfusion, although there were differences based on their diabetic status. In patients without DM, IRA occlusion versus no occlusion was associated with lower rates of postprocedural TIMI grade 3 flow, MBG 2 or 3, and complete STR. In patients with DM, baseline IRA occlusion was associated with lower rates of postprocedural TIMI grade 3 flow, but there were no significant differences in terms of MBG 2 or 3 and complete STR. A total of 138 patients (4.5%) died during follow-up ( Figure 2 ). In patients without DM, IRA occlusion versus no occlusion on baseline angiography was associated with a trend toward higher mortality rates (4.3% vs 2.7%, respectively, p = 0.051), whereas IRA occlusion versus no occlusion was associated with higher mortality rates in patients with DM (10.6% vs 4.6%, respectively, p = 0.01). Furthermore, in patients with baseline IRA occlusion, the presence of DM was associated with increased mortality rates (p <0.001). As shown in Figure 3 , only patients with DM with IRA occlusion had a higher risk for death, even after correction for postprocedural TIMI grade 3 flow (adjusted odds ratio 2.78, 95% confidence interval 1.66 to 4.66). As reported in Figure 2 , there was evidence for a significant interaction between DM and IRA occlusion on the risk for death. Reinfarction occurred in 98 patients (3.2%) ( Figure 2 ). In patients without DM, reinfarction rates did not differ between those with occluded those without occluded IRAs (3.3% vs 2.5%, respectively, p = 0.33). In contrast, patients with DM with IRA occlusion had a higher reinfarction rate (5.6% vs 2.1%, IRA occlusion vs no occlusion, respectively; p = 0.03). In patients with baseline IRA occlusion, the presence of DM was associated with higher rates of reinfarction (p = 0.03). As reported in Figure 3 , patients with DM with IRA occlusion had a higher risk for reinfarction that persisted after correction for baseline characteristics. As shown in Supplementary Figure 1 , there was a significant interaction between DM and IRA occlusion on the risk for reinfarction. Definite or probable stent thrombosis was observed in 54 patients (1.8%) ( Figure 2 ). Patients with baseline IRA occlusion compared with those without occlusion had higher rates of stent thrombosis in the non-DM (2.1% vs 1.0%, respectively, p = 0.04) and DM (3.2% vs 0.8%, respectively, p = 0.055) cohorts. Compared with patients without DM and with patent IRAs, we found two- and threefold higher risk for stent thrombosis in those without and those with DM with IRA occlusion, respectively ( Figure 3 ). Consistently, no significant interaction between DM and IRA occlusion was found for stent thrombosis ( Supplementary Figure 1 ).
Variable | Diabetes Mellitus | p ∗ | p ∗∗ | |||||
---|---|---|---|---|---|---|---|---|
Yes (n = 578) | No (n = 2,468) | |||||||
Occluded IRA (n = 339) | Non-occluded IRA (n = 239) | p | Occluded IRA (n = 1,483) | Non-occluded IRA (n = 985) | p | |||
Age (years) | 69 (59–76) | 70 (59–75) | 0.44 | 61 (51–71) | 62 (52–71) | 0.76 | <0.001 | <0.001 |
Men | 228 (67.3%) | 155 (64.9%) | 0.55 | 1,150 (77.5%) | 779 (79.1%) | 0.36 | <0.001 | <0.001 |
Hypertension | 286 (84.4%) | 193 (80.8%) | 0.26 | 853 (57.5%) | 558 (56.6%) | 0.67 | <0.001 | <0.001 |
Dyslipidemia | 189 (55.8%) | 129 (54.0%) | 0.67 | 522 (35.2%) | 334 (33.9%) | 0.51 | <0.001 | <0.001 |
Current smoker | 110 (32.4%) | 75 (31.4%) | 0.79 | 699 (47.1%) | 463 (47.0%) | 0.95 | <0.001 | <0.001 |
Family history of coronary artery disease | 105 (31.0%) | 64 (26.8%) | 0.27 | 497 (33.5%) | 351 (35.6%) | 0.28 | 0.37 | 0.009 |
Previous myocardial infarction | 53 (15.6%) | 29 (12.1%) | 0.23 | 156 (10.5%) | 80 (8.1%) | 0.047 | 0.008 | 0.051 |
Previous revascularization | 59 (17.4%) | 36 (15.1%) | 0.45 | 163 (11.0%) | 85 (8.6%) | 0.06 | 0.001 | 0.003 |
Ischemic time (hours) | 3.7 (2.5–6.6) | 3.5 (2.5–5.8) | 0.26 | 3.5 (2.2–5.3) | 3 (2.2–4.5) | <0.001 | 0.009 | 0.001 |
Randomization to intracoronary abciximab | 171 (50.4%) | 128 (53.6%) | 0.46 | 740 (49.9%) | 499 (50.7%) | 0.71 | 0.86 | 0.42 |
Thrombectomy | 105 (31.0%) | 57 (23.8%) | 0.06 | 487 (32.8%) | 317 (32.2%) | 0.73 | 0.51 | 0.01 |
Anterior myocardial infarction | 155 (45.7%) | 121 (50.6%) | 0.24 | 653 (44.0%) | 492 (49.9%) | 0.004 | 0.57 | 0.85 |
No. narrowed coronary arteries | 0.09 | 0.68 | 0.007 | <0.001 | ||||
1 | 148 (43.7%) | 83 (34.7%) | 767 (51.7%) | 495 (50.4%) | ||||
2 | 99 (29.2%) | 79 (33.1%) | 417 (28.1%) | 276 (28.1%) | ||||
3 | 92 (27.1%) | 77 (32.2%) | 299 (20.2%) | 212 (21.6%) | ||||
Infarct-related vessel | 0.20 | <0.001 | 0.25 | 0.28 | ||||
No infarct-related artery | 0 | 0 | 1 (0.1%) | 1 (0.1%) | ||||
Left anterior descending | 148 (43.7%) | 104 (43.7%) | 641 (43.3%) | 472 (48.0%) | ||||
Left circumflex | 46 (13.6%) | 32 (13.4%) | 173 (11.7%) | 126 (12.8%) | ||||
Right | 142 (41.9%) | 94 (39.5%) | 663 (44.8%) | 371 (37.7%) | ||||
Left main | 1 (0.3%) | 6 (2.5%) | 2 (0.1%) | 11 (1.1%) | ||||
Saphenous-vein graft | 2 (0.6%) | 2 (0.8%) | 1 (0.1%) | 2 (0.1%) |