Percutaneous coronary intervention within bypass grafts accounts for a significant percentage of total interventions. Bypass graft interventions are associated with an increased risk for stent thrombosis (ST), a condition that leads to significant morbidity and mortality. Despite this, the procedural characteristics and long-term outcomes of patients with bypass-graft ST have not been reported. The aim of the present study was to evaluate the procedural success and long-term outcomes of patients presenting with ST of coronary bypass grafts. Clinical and procedural characteristics of 205 ST cases at 5 academic hospitals were reviewed. Long-term mortality and major adverse cardiovascular events (stroke, reinfarction, and revascularization) were ascertained through review of medical records and the Social Security Death Index. Kaplan-Meier analysis was used to determine the association between ST in a bypass graft and long-term outcomes. Thirteen patients (6%) in the cohort presented with ST of a coronary bypass graft. Patients with bypass-graft ST had less severe presentations with a lower proportion of ST-segment elevation myocardial infarction (23% vs 69%, p <0.001). Despite this, ST of a bypass graft was associated with a trend toward reduced postprocedural Thrombolysis In Myocardial Infarction (TIMI) flow grade (p = 0.09), leading to lower angiographic (58% vs 92%, p <0.001) and procedural (62% vs 92%, p <0.001) success. After multivariate adjustment, bypass-graft ST was associated with increased long-term mortality (hazard ratio 3.3, 95% confidence interval 1.0 to 10.7) and major adverse cardiovascular events (hazard ratio 2.7, 95% confidence interval 1.1 to 6.9). In conclusion, ST in coronary bypass grafts is associated with reduced angiographic and procedural success as well as increased long-term major adverse cardiovascular events compared to ST in native coronary vessels.
Stent thrombosis (ST) is a complication of percutaneous coronary intervention (PCI) characterized by the development of intraluminal thrombus that results in obstruction of the stented vessel. Although rare, ST is associated with significant morbidity and mortality, with death occurring in 5% to 10% of patients. Previous research has identified a number of factors that increase the risk for ST, including previous intervention within a bypass graft. The inherent complexity of PCI within bypass grafts and differences in techniques used for the treatment of these vessels may contribute to this phenomenon. Despite this, the procedural characteristics and long-term outcomes of patients treated for ST within a bypass graft have not been reported. With this in mind, we evaluated the procedural success and long-term outcomes of patients presenting with ST of coronary artery bypass grafts.
Methods
The University of California ST registry contains consecutive cases of angiographically determined definite ST at 5 academic medical centers (the University of California, Davis; the University of California, San Diego; the University of California, San Francisco; San Francisco Veteran’s Affairs Administration Hospital; and San Francisco General Hospital) from 2005 to 2012. Cases were initially identified using cardiac catheterization laboratory records (2005 to 2008) and subsequently included prospective enrollment of subjects (2008 to 2012). After the identification of a potential patient with ST, each case was reviewed for clinical and angiographic characteristics by 2 independent interventional cardiologists. Only cases of angiographically determined definite ST as defined by the Academic Research Consortium were included in the registry. The study was reviewed and approved by the institutional review board at each participating site and has been registered with the unique trial identifier NCT00931502 .
Definitive ST cases were reviewed for demographic, procedural, and in-hospital outcomes as previously described. Briefly, trained physicians reviewed each medical record for the details related to the clinical presentation, medication compliance, and in-hospital outcomes (death, stroke, or repeat myocardial infarction) of patients with ST. An interventional cardiologist blinded to other clinical factors adjudicated the location of ST and the resulting Thrombolysis In Myocardial Infarction (TIMI) flow grade. Thrombus grading was also performed on a 5-point scale as previously described. Interventional therapies used were recorded by the individual sites and adjudicated by a blinded interventional cardiologist.
Long-term mortality was assessed using in-hospital records and the Social Security Death Index. Prospectively enrolled patients were also included in a follow-up phone interview to assess for recurrent major adverse cardiovascular events (MACEs), including repeat myocardial infarction, repeat revascularization, stroke, or death.
Univariate analysis was used to identify differences between subjects with ST of bypass grafts compared to native coronary arteries. Continuous variables were compared using the Kruskal-Wallis test, while categorical values were compared using the chi-square or Fisher’s exact test. Long-term mortality was analyzed using Kaplan-Meier survival analysis and log-rank tests. A Cox proportional-hazards model was developed to explore the relation between bypass-graft ST and risk for mortality and MACEs. Known risk factors for mortality (age, diabetes, and gender) were automatically included. Second, a list of possible confounders was generated using a directed acyclic graph. Confounders from this second group were retained if they were found to be associated with the outcome, using a p value <0.10 as a cutoff for inclusion. After backward stepwise selection, the final variables in the model included age, gender, diabetes, history of drug abuse, insurance status, and preadmission use of aspirin or clopidogrel. The proportional-hazards assumption was verified using log-log plots. All analyses were performed using Stata version 11 (StataCorp LP, College Station, Texas). A p value <0.05 was considered statistically significant.
Results
Among a total cohort of 205 patients with ST, 12 (5%) had ST in saphenous vein grafts and 1 (1%) in an arterial graft, for a total of 13 cases (6%). The demographic data for the patient population are listed in Table 1 . As shown, patients with ST of a bypass graft were less likely to present with ST-segment elevation myocardial infarction compared to those with ST of a native vessel (p <0.001). As listed in Table 2 , the percentage of patients with angiographic collateral vessels was similar between the 2 groups (p = 0.46). Likewise, the initial TIMI flow (p = 0.87) and thrombus grade (p = 0.55) were comparable between those with ST in a bypass graft or a native coronary artery. The procedural data for the patient population are listed in Table 3 . Glycoprotein IIb/IIIa inhibitors were used less frequently in patients with ST of a bypass graft (p <0.01). There was a nonsignificant trend toward decreased postprocedural TIMI flow grade in patients who had bypass-graft ST (p = 0.09) leading to lower rates of angiographic (p <0.001) and procedural (p <0.001) success. A summary of the angiographic and procedural characteristics of each patient with bypass-graft ST is listed in Table 4 .
| Demographic Characteristic | Bypass Graft (n = 13) | Non–Bypass Graft (n = 192) | p Value |
|---|---|---|---|
| Age (yrs) | 66 ± 12 | 61 ± 12 | 0.14 |
| Men | 13 | 165 (86%) | 0.15 |
| Previous coronary artery bypass grafting | 13 | 14 (7%) | <0.001 |
| Diabetes | 4 | 74 (39%) | 0.58 |
| Hypertension | 10 | 149 (78%) | 0.90 |
| Previous myocardial infarction | 10 | 136 (71%) | 0.64 |
| Active drug abuse | 0 | 18 (9%) | 0.2 |
| Medical noncompliance | 1 | 35 (18%) | 0.3 |
| Uninsured | 1 | 55 (29%) | 0.1 |
| Chronic kidney disease | 3 | 27 (14%) | 0.37 |
| Ejection fraction (%) | 53 ± 9 | 47 ± 14 | 0.17 |
| History of drug abuse | 0 | 18 (9%) | 0.30 |
| Insured | 12 | 137 (71%) | 0.10 |
| Taking aspirin | 11 | 142 (74%) | 0.39 |
| Taking thienopyridine | 9 | 86 (45%) | 0.09 |
| Presentation | <0.001 | ||
| ST-segment elevation myocardial infarction | 3 | 132 (69%) | |
| Non–ST-segment elevation myocardial infarction | 5 | 46 (24%) | |
| Unstable angina | 5 | 14 (7%) | |
| ST timing | 0.71 | ||
| Early | 4 | 57 (30%) | |
| Late | 3 | 29 (15%) | |
| Very late | 6 | 106 (55%) | |
| Indication for index stenting | 0.03 | ||
| Stable angina | 5 | 20 (15%) | |
| Acute coronary syndromes | 8 | 163 (85%) | |
| Previous stent | 0.10 | ||
| Drug-eluting stent | 5 | 102 (53%) | |
| Bare-metal stent | 6 | 40 (21%) | |
| Unknown | 2 | 50 (26%) | |
| Cardiogenic shock | 2 | 38 (20%) | 0.70 |
| Cardiopulmonary resuscitation | 1 | 10 (5%) | 0.70 |
| Intubation | 1 | 20 (10%) | 0.75 |
| Initial systolic blood pressure (mm Hg) | 138 ± 43 | 127 ± 38 | 0.34 |
| Angiographic Characteristic | Bypass Graft (n = 13) | Non–Bypass Graft (n = 192) | p Value |
|---|---|---|---|
| Target coronary artery | <0.01 | ||
| Left anterior descending | 1 | 94 (49%) | |
| Diagonal | 7 | 9 (5%) | |
| Circumflex/obtuse marginal | 2 | 38 (20%) | |
| Right | 3 | 50 (26%) | |
| Target vessel location | 0.65 | ||
| Ostial or proximal | 5 | 84 (45%) | |
| Mid-distal | 8 | 103 (55%) | |
| Angiographic collateral vessels | 2 | 47 (24%) | 0.46 |
| ST at bifurcation | 0 | 29 (15%) | 0.10 |
| Proximal luminal diameter (mm) | 3.5 ± 0.7 | 3.1 ± 0.6 | 0.03 |
| Distal luminal diameter (mm) | 3.1 ± 0.6 | 2.6 ± 0.5 | 0.02 |
| Initial TIMI flow grade <3 | 11 | 174 (91%) | 0.87 |
| Thrombus grade | 0.55 | ||
| 1–3 | 1 | 26 (14%) | |
| 4 or 5 | 12 | 166 (87%) | |
| Thrombus location in stent | 0.70 | ||
| Proximal | 10 | 119 (66%) | |
| Mid-distal | 3 | 61 (34%) |
| Procedural Characteristic | Bypass Graft (n = 13) | Non–Bypass Graft (n = 192) | p Value |
|---|---|---|---|
| Balloon angioplasty | 7 | 72 (38%) | 0.24 |
| Embolic protection device | 3 | 0% (0) | 0.01 |
| Intravascular ultrasound | 0 | 56 (29%) | 0.02 |
| Glycoprotein IIb/IIIa Inhibitor | 6 | 149 (78%) | 0.01 |
| Postprocedural TIMI flow grade | 0.09 | ||
| <3 | 3 | 17 (9%) | |
| 3 | 10 | 175 (92%) | |
| Angiographic success | 7 | 173 (92%) | <0.001 |
| Procedural success | 8 | 175 (92%) | <0.001 |
| Patient | Timing of ST | Time to ST | Age (yrs) | Graft Type | Target Vessel | Presentation | Type of Stent | Preprocedural TIMI Flow Grade | Thrombus Grade ∗ | EPD | Treatment of ST |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Early | 6 d | 61 | SVG | Diagonal | UAP | DES | 0 | 5 (4) | Y | Aspiration and angioplasty |
| 2 | Early | 15 d | 73 | SVG | OM | UAP | DES | 3 | 4 | N | Angioplasty and BMS |
| 3 | Early | 25 d | 84 | SVG | RCA | UAP | BMS | 2 | 3 | N | Aspiration and DES |
| 4 | Early | 29 d | 75 | SVG | RCA | UAP | BMS | 0 | 5 (4) | N | Aspiration and POBA |
| 5 | Late | 86 d | 77 | SVG | Diagonal | STEMI | DES | 0 | 5 (5) | Y | Unsuccessful |
| 6 | Late | 145 d | 70 | SVG | Diagonal | NSTEMI | DES | 0 | 5 (4) | N | Aspiration and POBA |
| 7 | Late | 158 d | 44 | SVG | Diagonal | NSTEMI | BMS | 0 | 5 (5) | N | Unsuccessful |
| 8 | Very late | 1.1 yrs | 64 | SVG | Diagonal | NSTEMI | BMS | 3 | 4 | N | Unsuccessful |
| 9 | Very late | 1.3 yrs | 60 | LIMA | LAD | STEMI | Unknown | 1 | 5 (2) | N | Angioplasty and DES |
| 10 | Very late | 2.3 yrs | 53 | SVG | Diagonal | UAP | DES | 0 | 5 (5) | N | Unsuccessful |
| 11 | Very late | 3.2 yrs | 53 | SVG | OM | NSTEMI | DES | 0 | 5 (3) | N | Angioplasty and DES |
| 12 | Very late | 3.7 yrs | 63 | SVG | RCA | NSTEMI | BMS | 0 | 5 (1) | N | Native RCA stented with DES |
| 13 | Very late | 4.6 yrs | 77 | SVG | LAD | STEMI | BMS | 0 | 5 (2) | Y | Aspiration and DES |
∗ Thrombus grades in parentheses represent TIMI thrombus grades after initial wire crossing in cases of complete stent occlusion from thrombus.
The clinical outcomes of patients with bypass-graft ST are listed in Table 5 . The rates of in-hospital reinfarction, stroke, or mortality were similar between the 2 groups. The long-term mortality and MACE rates in patients with ST are shown in Figures 1 and 2 , respectively. Patients with bypass-graft ST had a trend toward increased mortality (hazard ratio [HR] 2.0, 95% confidence interval [CI] 0.7-5.8) and MACEs (HR 1.9, 95% CI 0.8 to 4.4) during a median follow-up period of 2.4 years. After multivariate adjustment, bypass-graft ST was associated with significantly higher rates of death (HR 3.3, 95% CI 1.0 to 10.7) and MACEs (HR 2.7, 95% CI 1.1 to 6.9). In a sensitivity analysis limited only to patients with histories of bypass surgery (n = 27), patients with ST of a bypass graft had a similar estimated risk for higher long-term mortality (HR 2.0, 95% CI 0.7 to 5.8, p = 0.20) and MACEs (HR 2.0, 95% CI 0.5 to 8.0, p = 0.30) compared to bypass-grafted patients with ST of a native vessel.
