Percutaneous coronary intervention (PCI) of calcified coronary lesions has been associated with increased rates of adverse ischemic events. However, the potential association between the presence and severity of calcific deposits and bleeding complications has yet to be investigated. Data from 6,855 patients with non–ST-segment elevation acute coronary syndrome (NSTEACS) or ST-segment elevation myocardial infarction (STEMI) treated with PCI were pooled from 2 large-scale randomized controlled trials—Acute Catheterization and Urgent Intervention Triage Strategy and Harmonizing Outcomes with Revascularization and Stents in Acute Myocardial Infarction. Patients were stratified into 3 groups according the grade of target PCI lesion calcium (none to mild, moderate, and severe) as assessed by an independent angiographic core laboratory. Thirty-day bleeding event rates were assessed and compared among groups. In the total cohort undergoing PCI, none-to-mild target lesion calcium was found in 4,665 patients (68.1%), moderate target lesion calcium in 1,788 patients (26.1%), and severe target lesion calcium in 402 patients (5.9%). The 30-day rates of non–coronary artery bypass graft surgery major bleeding increased significantly with each degree of coronary calcium (none to mild = 5.9%, moderate = 7.2%, and severe = 11.2%, p = 0.0003). By multivariable analysis, presence of severe calcium was an independent predictor of non–coronary artery bypass graft major bleeding after PCI (hazard ratio 1.54, 95% confidence interval 1.09 to 2.17, p = 0.01). In conclusion, in patients undergoing PCI for non–ST-segment elevation acute coronary syndrome and ST-segment elevation myocardial infarction, target lesion coronary calcium was an independent predictor of major bleeding events. Further studies are needed to elucidate mechanisms underlying this finding and to optimize treatment of this high-risk population.
Percutaneous coronary intervention (PCI) in heavily calcified coronary lesions results in suboptimal acute results with increased rates of recurrent ischemic events and mortality. Coronary calcium is often associated with systemic vascular calcific deposits. The presence of femoral arterial calcium may predispose to bleeding by alteration of mechanical properties and homeostatic capacity. Moreover, patients with coronary calcium may have other co-morbidities that increase the likelihood of bleeding. We therefore hypothesized that the presence and severity of coronary calcium may be associated with an increase in bleeding events after PCI. To examine this issue, we analyzed the pooled databases from the large-scale Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) and the Harmonizing Outcomes with Revascularization and Stents in Acute Myocardial Infarction (HORIZONS-AMI) trials.
Methods
The present study represents a patient-level pooled analysis of 2 large-scale, prospective, randomized trials, which evaluated bivalirudin as an anticoagulant in acute coronary syndrome (ACS): the ACUITY trial in patients with non–ST-segment elevation acute coronary syndrome (NSTEACS) and the HORIZONS-AMI trial in patients with ST-segment elevation myocardial infarction (STEMI). The study designs and primary results have previously been described in detail.
In brief, ACUITY was a multicenter, prospective, randomized trial of 13,819 patients with moderate- and high-risk NSTEACS who were treated with an early invasive strategy. Before coronary angiography, patients were randomly assigned to heparin (unfractionated or low molecular weight) plus a glycoprotein IIb/IIIa inhibitor (GPI), bivalirudin plus a GPI, or bivalirudin monotherapy. Angiography was performed within 72 hours of randomization, and depending on coronary anatomy, patients were triaged to PCI, coronary artery bypass graft (CABG) surgery, or medical therapy. In patients undergoing PCI, stent choice (bare metal or drug eluting) was per operator discretion. Dual antiplatelet therapy with aspirin and clopidogrel was recommended for at least 1 year. In HORIZONS-AMI, 3,602 patients with ST-segment elevation myocardial infarction presenting within the first 12 hours of symptom onset, in whom primary PCI was planned, were randomly assigned to treatment with unfractionated heparin plus GPI or bivalirudin monotherapy. Patients were randomly assigned to either TAXUS (Boston Scientific, Natick, Massachusetts) paclitaxel-eluting stents or bare-metal stents in a 3:1 ratio. Aspirin and clopidogrel were administered before catheterization and were continued for at least 1 year. An independent clinical events committee, blinded to treatment assignment, adjudicated all major adverse events in these 2 trials, including major bleeding.
Our primary objective was to assess the relation between target lesion coronary calcium and the risk of major bleeding after PCI for ACS at 30 days. The databases of the ACUITY and HORIZONS-AMI trials were combined, and patients were stratified in 3 groups based on the presence and severity of target lesion coronary artery calcium (none to mild, moderate, and severe). For the present study, we included only patients from the PCI subgroup in whom quantitative coronary angiography was performed by an independent core laboratory, blinded to randomization and clinical outcomes (Cardiovascular Research Foundation, New York, New York). This comprised all patients from HORIZONS-AMI and a prespecified subset of 6,921 patients enrolled in the formal ACUITY angiographic substudy. Moderate lesion calcium was defined in the core laboratory as radiopaque densities noted during the cardiac cycle involving only 1 side of the vascular wall, and severe lesion calcium was defined as radiopaque densities noted without cardiac motion before contrast injection generally involving both sides of the arterial wall.
As per the ACUITY and HORIZONS protocols, major bleeding was defined as intracranial or intraocular bleeding, access site hemorrhage requiring intervention, hematoma ≥5 cm in diameter, reduction in hemoglobin levels ≥4 g/dl without or ≥3 g/dl with an overt source, reoperation for bleeding, or any blood product transfusion. Bleeding according to the Thrombolysis In Myocardial Infarction major and/or minor scale was determined in both studies as well. We classified major bleeding according to whether it arose from the vascular access site. Major access site bleeding was defined as bleeding from the access vessel that resulted in interventional or surgical correction, hematoma ≥5 cm, retroperitoneal bleeding, or hemoglobin drop ≥3 g/dl with prolonged access site bleeding (>30 minutes), or ecchymosis or hematoma. We also assessed for major bleeds not limited to the access site only (i.e., non–access site only, both non–access site and access site, and indeterminate source). Non–access site was defined as intracranial, intraocular, or a reduction in hemoglobin levels ≥3 g/dl with an overt non–access site source (e.g., gastrointestinal or genitourinary).
The Mehran bleeding score was determined for each patient using the following clinical variables and compared among groups of calcium severity: gender, age, serum creatinine (mg/dl), white blood cell count (10 9 /L), anemia, ACS clinical presentation, and antithrombotic regimen (heparin vs bivalirudin).
All analyses were done by intention to treat. Continuous data are presented as mean ± SD and were compared using analysis of variance. Thirty-day event rates were estimated using Kaplan-Meier methodology and compared using the log-rank test. Multivariable regression analysis using Cox proportional hazard models for 30-day time points was performed to assess the association between coronary calcium and major bleeding with variable entry/stay criteria of 0.1/0.1. Multivariable analysis included stepwise adjustment for the following pool of variables based on clinical importance and those that differed significantly between calcium severity groups: severe target lesion calcium (vs none-to-mild calcium), moderate target calcium (vs none-to-mild calcium), male gender, age, randomization to heparin plus GPI versus bivalirudin with or without GPI, diabetes, hypertension, baseline creatinine clearance <60 ml/min, baseline hemoglobin, and baseline platelet count. We also performed a separate multivariable analysis with moderate and severe target lesion calcium grouped together (vs none-to-mild calcium) with the same covariates listed previously. p Values <0.05 were considered to be statistically significant. Statistical analyses were performed using SAS, version 9.2 (SAS Institute, Cary, North Carolina).
Results
A total of 6,855 patients were treated by PCI and analyzed by quantitative coronary angiography analysis (3,587 with NSTEACS from ACUITY and 3,268 with STEMI from HORIZONS-AMI), representing the study population. In them, 4,665 patients (68.1%) had none-to-mild target lesion calcium, 1,788 (26.1%) had moderate target lesion calcium, and 402 (5.9%) had severe target lesion calcium.
Baseline clinical, angiographic, and procedural characteristics, stratified by the degree of target lesion calcium, are presented in Tables 1 and 2 . Patients with more severely calcified lesions were older, more likely to have renal insufficiency, lower left ventricular ejection fraction, and higher baseline white blood cell counts compared with those with none-to-mild calcium. They were also more likely to have STEMI than non-STEMI or unstable angina at initial clinical presentation. Not surprisingly, severe target lesion calcium was also associated with more complex lesion characteristics as summarized in Table 2 . Patients with greater degrees of coronary calcium were slightly less likely to have received bivalirudin plus a GPI as procedural anticoagulation, although rates of bivalirudin monotherapy and other anticoagulants were similar among the groups, as was use of discharge and 30-day antiplatelet agents. The calculated Mehran bleeding score increased with the extent of calcium.
| Variable | Quantity of Calcium | Overall p Value | ||
|---|---|---|---|---|
| None/Mild (n = 4,665) | Moderate (n = 1,788) | Severe (n = 402) | ||
| Age (yrs) | 59.2 (51.7–69.0) | 63.1 (55.0–72.0) | 65.3 (55.8–74.0) | <0.0001 |
| Men | 3,385/4,665 (72.6) | 1,342/1,788 (75.1) | 292/402 (72.6) | 0.12 |
| Diabetes mellitus | 1,143/4,651 (24.6) | 418/1,784 (23.4) | 103/402 (25.6) | 0.52 |
| Insulin treated | 324/4,651 (7.0) | 114/1,784 (6.4) | 35/402 (8.7) | 0.25 |
| Hypertension ∗ | 2,849/4,652 (61.2) | 1,112/1,784 (62.3) | 249/402 (61.9) | 0.71 |
| Hyperlipidemia † | 2,452/4,630 (53.0) | 934/1,780 (52.5) | 201/399 (50.4) | 0.60 |
| Current smoker | 1,842/4,658 (39.5) | 667/1,779 (37.5) | 141/401 (35.2) | 0.10 |
| Previous myocardial infarction | 1,076/4,598 (23.4) | 406/1,765 (23.0) | 72/398 (18.1) | 0.054 |
| Previous PCI | 1,440/4,652 (31.0) | 493/1,782 (27.7) | 93/401 (23.2) | 0.0005 |
| Previous coronary bypass | 579/4,658 (12.4) | 243/1,786 (13.6) | 58/402 (14.4) | 0.28 |
| Previous stroke | 165/2,609 (6.3) | 54/804 (6.7) | 11/144 (7.6) | 0.78 |
| Renal insufficiency ‡ | 642/4,362 (14.7) | 326/1,657 (19.7) | 107/374 (28.6) | <0.0001 |
| Left ventricular ejection fraction (%) | 62.6 (53.8–70.8) | 61.3 (52.1–69.0) | 59.6 (51.7–67.8) | <0.0001 |
| White blood cell count (10 3 /L) | 9.1 (7.1–11.6) | 9.4 (7.4–11.8) | 10.0 (7.8–12.2) | <0.0001 |
| Hemoglobin (g/dl) | 14.3 (13.2–15.3) | 14.2 (13.2–15.3) | 14.1 (13.1–15.3) | 0.33 |
| Platelet count (10 3 /mm 3 ) | 236 (197–282) | 237 (199–279) | 238 (195–286) | 0.84 |
| Biomarker positive | 1,389/2,457 (56.5) | 441/742 (59.4) | 81/131 (61.8) | 0.22 |
| Unstable angina pectoris | 1,233/4,665 (26.4) | 368/1,788 (20.8) | 60/402 (14.9) | <0.0001 |
| Non–ST-segment elevation myocardial infarction | 1,396/4,665 (29.9) | 444/1,788 (24.8) | 86/402 (21.4) | <0.0001 |
| STEMI | 2,036/4,665 (43.6) | 976/1,788 (54.6) | 256/402 (63.7) | <0.0001 |
| Mehran bleeding score | 11.86 ± 7.19 | 13.24 ± 7.26 | 14.69 ± 7.65 | <0.0001 |
∗ Hypertension is defined as taking antihypertensive medication on admission.
† Hyperlipidemia is defined as taking lipid-lowering medication on admission.
‡ Renal insufficiency is defined as creatinine clearance rate <60 ml/min by the Cockcroft-Gault equation.
| Variable | Quantity of Calcium | Overall p Value | ||
|---|---|---|---|---|
| None/Mild (n = 4,665) | Moderate (n = 1,788) | Severe (n = 402) | ||
| Antithrombotic medication | ||||
| Bivalirudin + GPI | 901/4,665 (19.3) | 284/1,788 (15.9) | 51/402 (12.7) | <0.0001 |
| Bivalirudin monotherapy | 1,914/4,665 (41.0) | 735/1,788 (41.1) | 173/402 (43.0) | 0.73 |
| Heparin + GPI | 1,850/4,665 (39.7) | 769/1,788 (43.0) | 178/402 (44.3) | 0.02 |
| Intraprocedural GPI | 2,751/4,665 (59.0) | 1,053/1,788 (58.9) | 229/402 (57.0) | 0.73 |
| Number of vessels treated | 1.11 ± 0.33 | 1.14 ± 0.37 | 1.12 ± 0.35 | 0.003 |
| Number of lesions treated | 1.31 ± 0.67 | 1.40 ± 0.75 | 1.38 ± 0.86 | <0.0001 |
| Number of stents | 1.31 ± 0.74 | 1.50 ± 0.89 | 1.53 ± 0.96 | <0.0001 |
| Drug-eluting stent | 3,660/4,609 (79.5) | 1,376/1,764 (78.0) | 303/380 (79.7) | 0.44 |
| Bare-metal stent | 929/4,609 (20.2) | 402/1,764 (22.8) | 84/380 (22.1) | 0.06 |
| Baseline QCA | ||||
| Lesion length (mm) | 13.0 (10.0–19.0) | 14.4 (10.1–21.0) | 14.9 (10.0–20.0) | <0.0001 |
| RVD (mm) | 2.79 (2.41–3.15) | 2.80 (2.45–3.19) | 2.84 (2.43–3.19) | 0.08 |
| Diameter stenosis (%) | 80.2 (66.3–98.0) | 81.0 (66.6–100.0) | 84.3 (67.2–100.0) | 0.0004 |
| Total occlusion | 1,444/5,824 (24.8) | 720/2,428 (29.7) | 211/550 (38.4) | <0.0001 |
| Thrombus | 2,144/5,813 (36.9) | 1,073/2,422 (44.3) | 299/549 (54.5) | <0.0001 |
| Bifurcation present | 2,218/5,829 (38.1) | 1,121/2,430 (46.1) | 264/550 (48.0) | <0.0001 |
| ACC/AHA classification | ||||
| A/B1 | 2,155/5,818 (37.0) | 308/2,424 (12.7) | 57/549 (10.4) | <0.0001 |
| B2 | 985/5,818 (16.9) | 459/2,424 (18.9) | 89/549 (16.2) | 0.07 |
| C | 2,678/5,818 (46.0) | 1,657/2,424 (68.4) | 403/549 (73.4) | <0.0001 |
| Discharge medications | ||||
| Aspirin | 4,287/4,605 (93.1) | 1,635/1,742 (93.9) | 370/391 (94.6) | 0.33 |
| Thienopyridine | 4,211/4,607 (91.4) | 1,604/1,742 (92.1) | 363/391 (92.8) | 0.47 |
| Aspirin + thienopyridine | 4,074/4,605 (91.4) | 1,555/1,742 (89.3) | 354/391 (90.5) | 0.36 |
| 30-Day medications | ||||
| Aspirin | 4,396/4,497 (97.8) | 1,639/1,694 (96.8) | 373/384 (97.1) | 0.08 |
| Thienopyridine | 4,317/4,499 (96.0) | 1,617/1,698 (95.2) | 367/384 (95.6) | 0.44 |
| Aspirin + thienopyridine | 4,469/4,499 (99.3) | 1,684/1,698 (99.2) | 383/384 (99.7) | 0.47 |
Rates of 30-day major bleeding end points stratified by degree of coronary calcium are presented in Table 3 and Figure 1 . Major bleeding not related to CABG developed in 444 patients (6.5%), and was strongly associated with target lesion calcium. Thrombolysis In Myocardial Infarction major and minor bleeding were also associated with the degree of target lesion calcium. Patients in the severely calcified lesion group required transfusions more frequently than those in the moderate and none-to-mild calcium groups (6.5% vs 3.5% vs 2.9% respectively, p = 0.003). Specifically, however, the relation between major bleeding and target lesion calcium was explained by an increase in bleeds not limited to the access site and non–access site bleeding but not access site–related bleeding with more severe coronary calcium ( Table 3 ).
| Variable | (A) None/Mild (n = 4,665) | (B) Moderate (n = 1,788) | (C) Severe (n = 402) | Overall p Value | (A vs B) p Value | (A vs C) p Value | (B vs C) p Value |
|---|---|---|---|---|---|---|---|
| Major bleeding (non-CABG) | 273 (5.9) | 128 (7.2) | 43 (11.2) | 0.0003 | 0.05 | 0.0001 | 0.02 |
| Retroperitoneal bleed | 34 (0.7) | 14 (0.8) | 1 (0.7) | 0.51 | 0.81 | 0.27 | 0.25 |
| Access site hemorrhage | 19 (0.4) | 5 (0.3) | 3 (0.8) | 0.39 | 0.46 | 0.32 | 0.16 |
| Puncture site hematoma ≥5 cm | 86 (1.8) | 37 (2.1) | 7 (1.8) | 0.80 | 0.54 | 0.88 | 0.67 |
| Hemoglobin drop ≥4 g/dl without overt source | 82 (1.8) | 50 (2.8) | 16 (4.0) | 0.001 | 0.008 | 0.002 | 0.21 |
| Hemoglobin drop ≥3 g/dl with overt source | 88 (1.9) | 41 (2.3) | 16 (4.0) | 0.02 | 0.28 | 0.004 | 0.06 |
| Any blood product transfusion | 136 (2.9) | 62 (3.5) | 24 (6.5) | 0.003 | 0.23 | 0.0007 | 0.02 |
| Major access site bleeding only | 121 (2.6) | 50 (2.8) | 10 (2.9) | 0.87 | 0.63 | 0.90 | 0.73 |
| Major bleeding not limited to access site | 194 (4.2) | 96 (5.4) | 35 (9.2) | <0.0001 | 0.03 | <0.001 | 0.01 |
| Major non–access site bleeding only | 11 (0.2) | 5 (0.3) | 4 (1.0) | 0.03 | 0.75 | 0.007 | 0.04 |
| TIMI major/minor bleeding | 280 (6.0) | 135 (7.6) | 40 (10.0) | 0.002 | 0.02 | 0.002 | 0.12 |
| TIMI major bleeding | 84 (1.8) | 50 (2.8) | 13 (3.3) | 0.01 | 0.01 | 0.04 | 0.65 |
| TIMI minor bleeding | 233 (5.0) | 101 (5.7) | 34 (8.5) | 0.01 | 0.27 | 0.003 | 0.04 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
