We assessed the impact of aspiration thrombectomy (AT) in patients with ST elevation myocardial infarction undergoing primary percutaneous coronary intervention (PPCI) on major adverse cardiac events at 30 days and 1-year mortality in 517 consecutive patients who were included in the prospective, nationwide, multicenter, observational Acute Coronary Syndrome Israeli Survey in 2010. Two hundred seventeen patients (42%) underwent AT (AT-PPCI) and 300 patients conventional (C) PPCI. Both groups had similar infarct-related artery distribution and ostial or proximal culprit lesion. Patients in AT-PPCI versus C-PPCI had lower systolic blood pressure and worse Killip class on admission, more frequent Thrombolysis In Myocardial Infarction flow 0 or 1 before PPCI (80% vs 56%), less frequent restoration of flow after indwelling a guidewire in the infarct-related artery (32% vs 52%), and more use of IIb/IIIa glycoprotein inhibitors (69% vs 49%), respectively (p ≤0.05 for all comparisons). Thirty-day major adverse cardiac events was similar in the AT-PPCI and C-PPCI groups, 10.6% versus 9.7%, p = 0.73; adjusted odds ratio 0.97, 95% confidence interval 0.45 to 2.10, p = 0.95. One-year mortality was lower in the AT-PPCI versus C-PPCI group, 3.7% versus 6.7%, p = 0.13; adjusted hazard ratio 0.31, 95% confidence interval 0.10 to 0.96, p = 0.042. In conclusion, this study of consecutive patients with ST elevation myocardial infarction undergoing PPCI demonstrates that AT was an independent predictor of reduced 1-year mortality.
Coronary reperfusion with primary percutaneous coronary intervention (PPCI) improves outcome of patients with ST elevation myocardial infarction (STEMI). Achievement of Thrombolysis In Myocardial Infarction (TIMI) 3 flow correlates with better clinical outcome and is attained in most of the patients undergoing PPCI. Aspiration thrombectomy (AT) has been used to improve coronary blood flow, ST-segment elevation resolution, and outcome. However, the recently published Thrombus Aspiration during ST-Segment Elevation Myocardial Infarction (TASTE) trial, a multicenter, prospective, open-label, randomized controlled trial, revealed that routine thrombus aspiration before PCI compared with PCI alone did not reduce 30-day mortality in patients with STEMI. The present study was conducted to assess the impact of AT before stent implantation in patients with STEMI undergoing PPCI on 30-day major adverse cardiac event (MACE) and on 1-year all-cause mortality in the “real-world practice” of a contemporary national survey.
Methods
Consecutive 517 patients with STEMI hospitalized in 23 intensive coronary care units equipped with an on-site catheterization laboratory were included in the prospective, multicenter, observational, nationwide, biannual Acute Coronary Syndrome Israeli Survey (ACSIS) conducted from March 2010 to April 2010. All patients presented with symptoms of acute myocardial ischemia lasting >30 minutes but <12 hours. Electrocardiography revealed ST elevation ≥0.1 mm in ≥2 adjacent leads.
All patients underwent PPCI; however, adjunctive medical therapies, the decision on performing AT, and stents type were at the discretion of each participating center and operating physician.
Different AT devices were used in different centers; however, only manual aspiration devices (Pronto Extraction Catheter, Vascular Solutions, Inc., Minneapolis, Minnesota; Export Aspiration Catheter, Medtronic Inc., Minneapolis, Minnesota; Eliminate Aspiration Catheter, Terumo Corporation, Leuven, Belgium; QuickCat Extraction Catheter, Spectranetics Corporation, Colorado Springs, Colorado; QXT Extraction Catheter, Vascular Solutions, Inc., Minneapolis, Minnesota; and ASAP Aspiration Catheter kit, Merit Medical Systems Inc., South Jordan, Utah) were used.
The ACSIS was designed as a prospective observational case series of all consecutive patients with acute coronary syndrome without age limit in all 26 intensive coronary care units in Israel. Demographic, historical and clinical data, admission and discharge electrocardiograms, medical therapies, angiographic and PPCI characteristics including TIMI flow and myocardial blush grade before and at the end of the intervention, ST-segment elevation resolution ≥70% at the electrocardiography performed within 1 hour after PPCI, complications, and follow-up data were recorded on electronic case report forms by dedicated physicians. All angiographic characteristics were analyzed by 2 angiographers at each site. The assessment of thrombus burden was not collected.
Thirty-day MACE included death, myocardial infarction, hospitalization for unstable angina pectoris, stent thrombosis, urgent revascularization, and stroke. Thirty-day and 1-year all-cause mortality rates were determined from hospital charts or by matching the identification numbers of the patients with the “National Population Register of the State of Israel”. At 1 year, 1 patient was lost to follow-up (0.2%).
The organization, data acquisition, management, and follow-up were performed at the Israeli Society for the Prevention of Heart Attacks national coordinating center. Data checks for completeness and consistency were based on computerized data queries issued at the coordinating center and completed from the medical reports attached to each patient form. The survey received the local ethics committee approval in each hospital.
All statistical analyses were performed using SAS, version 8.2, statistical software (SAS Institute, Cary, North Carolina). To determine the significance of the differences between proportions, chi-square test was used. Results of continuous variables are reported as median (twenty-fifth to seventy-fifth percentiles) and were compared by the Kruksal-Wallis test. Logistic regression analyses were conducted to compare 30-day MACE in AT-PPCI and C-PPCI adjusting for pertinent variables. Covariate-adjusted mortality rates using Cox proportional hazards regression models were conducted to compare 1-year mortality in both groups. The consistency of the results of the multivariate models was further assessed by incorporating a propensity score for AT in the proportional hazards models. The propensity score was calculated by running a saturated model based on covariates thought to be associated with AT including age, gender, diabetes, PPCI performed during week or weekend days, day or night-time, systolic blood pressure, heart rate and Killip class on arrival, TIMI flow 0 or 1 before PPCI, culprit lesion at ostial or proximal vessel, anterior myocardial infarction, use of IIb/IIIa glycoprotein (GP) inhibitors, and centers performing AT in <50% or ≥50% of the PPCI cases. One-year Kaplan-Meier survival curves in both groups were estimated and compared using the log-rank test.
Results
Of 517 patients with STEMI who had PPCI, 217 (42%) underwent AT-PPCI and 300 patients (58%) C-PPCI ( Table 1 ). AT was performed in 21 of the 23 centers and its use varied significantly among them (ranged from 14% to 82% of PPCI cases). Use of AT in ≥50% of PPCI cases in a center was noted in 8 of the 23 centers. The characteristics of patients in both groups were similar; however, patients in AT-PPCI group had lower systolic blood pressure on admission and more frequently presented with Killip class II to IV compared with the C-PPCI group. There was no difference in the frequency of using AT during day or nighttime and during week or weekend days.
| Variable | PPCI | p Value | |
|---|---|---|---|
| With AT (n = 217) | Without AT (n = 300) | ||
| Mean age (yrs) | 60.7 ± 11.8 | 61.2 ± 11.8 | 0.21 |
| Women | (44/217), 20 | (51/300), 17 | 0.34 |
| Diabetes mellitus | (54/217), 25 | (95/300), 32 | 0.11 |
| Hypertension | (116/215), 54 | (154/298), 52 | 0.61 |
| Hyperlipidemia | (153/216), 71 | (208/299), 70 | 0.76 |
| Current smoker | (107/216), 50 | (139/294), 47 | 0.61 |
| Previous angina pectoris | (57/216), 26 | (60/299), 20 | 0.09 |
| Previous MI | (42/216), 19 | (69/299), 23 | 0.32 |
| Previous PCI | (49/215), 23 | (65/298), 22 | 0.79 |
| Previous CHF | (6/215), 3 | (11/298), 4 | 0.57 |
| Arrival characteristics | |||
| Admission heart rate (beats/min) | 77 ± 19 | 78 ± 19 | 0.55 |
| Admission systolic blood pressure (mm Hg) | 134 ± 30 | 139 ± 30 | 0.05 |
| Killip II–IV on arrival | (29/217), 13 | (24/300), 8 | 0.05 |
| Arrival at night | (72/187), 39 | (117/257), 46 | 0.14 |
| Arrival at weekend | (49/217), 23 | (76/298), 26 | 0.44 |
Angiographic characteristics were similar in both groups ( Table 2 ). There was a significantly higher frequency of TIMI flow 0 and 1 before PPCI, less frequent restoration of flow after introducing the guidewire in the infarct-related artery, and significantly more frequent use of IIb/IIIa GP inhibitors in AT-PPCI group compared with the C-PPCI group (p <0.0001, for all comparisons). Variables independently associated with the use of AT are depicted in Table 3 . TIMI flow 3 and myocardial blush grade 3 at the end of PPCI were comparable between the 2 groups; however, ST-segment elevation resolution was achieved more frequently in AT-PPCI compared with C-PPCI group ( Table 2 ). AT-PPCI group received slightly more frequently angiotensin-converting enzyme inhibitors or angiotensin receptor blockers.
| Variable | PPCI | p Value | |
|---|---|---|---|
| With AT (n = 217) | Without AT (n = 300) | ||
| Radial access | (59/216), 27 | (83/299), 28 | 0.91 |
| Narrowed coronary arteries | 0.81 | ||
| 0 | (2/216), 1 | (5/300), 1.7 | |
| 1 | (81/216), 37 | (114/300), 38 | |
| 2 | (79/216), 37 | (101/300), 34 | |
| 3 | (54/216), 25 | (80/300), 27 | |
| Infarct-related coronary artery | |||
| Left anterior descending | (104/216), 48 | (137/299), 46 | 0.25 |
| Left circumflex | (24/216), 11 | (46/299), 15 | |
| Right | (81/216), 38 | (109/299), 36 | |
| LM/SVG/LIMA | (7/217), 3 | (5/300), 2 | |
| Culprit lesion ostial-proximal vessel | (101/213), 47 | (125/295), 42 | 0.26 |
| Door-to-balloon time (minutes) | 73 (41–123) | 60 (39–94) | 0.038 |
| Time from symptom onset to balloon (minutes) | 195 (130–317) | 188 (131–330) | 0.78 |
| Time from symptom onset to balloon ≤180 minutes | (107/217), 49 | (154/300), 51 | 0.65 |
| TIMI flow 0–1 before PPCI | (172/216), 80 | (166/298), 56 | <0.0001 |
| Restoration of flow after guidewire | (63/197), 32 | (119/231), 52 | <0.0001 |
| Use of IIb/IIIa GP inhibitors | (150/217), 69 | (147/300), 49 | <0.0001 |
| Patients in centers performing AT ≥50 of PPCI | (110/217), 51 | (70/300), 23 | <0.0001 |
| TIMI flow 3, end of PPCI | (196/217), 90 | (271/293), 92 | 0.38 |
| MBG 3, end of PCI | (118/197), 60 | (170/265), 64 | 0.35 |
| STR at first ECG after PCI | (152/194), 78 | (155/237), 65 | 0.003 |
| Stent diameter (mm) | 3.19 ± 0.66 | 3.02 ± 0.53 | 0.002 |
| Number of stents | 1.07 ± 0.51 | 1.1 ± 0.55 | 0.57 |
| Drug-eluting stents | (32/195), 16 | (61/273), 22 | 0.11 |
| Use of clopidogrel/prasugrel before PPCI | (182/215), 85 | (243/292), 83 | 0.66 |
| In-hospital medications | |||
| Aspirin | (216/217), 99 | (295/300), 98 | 0.41 |
| Clopidogrel/prasugrel | (215/217), 99 | (295/300), 98 | 0.71 |
| UF heparin | (123/217), 57 | (182/300), 61 | 0.36 |
| LMWH | (55/217), 25 | (83/300), 28 | 0.55 |
| Fondaparinux | (8/217), 4 | (7/300), 2 | 0.37 |
| Bivalirudin | (18/217), 8 | (21/300), 7 | 0.58 |
| β Blockers | (179/217), 83 | (250/299), 84 | 0.74 |
| ACE-I/ARB | (193/217), 89 | (242/300), 81 | 0.01 |
| Statin | (211/217), 97 | (292/299), 98 | 0.76 |
| In-hospital complications | |||
| No reflow | (10/214), 5 | (10/296), 3 | 0.46 |
| Major bleeding | (4/217), 1.8 | (8/300), 2.7 | 0.53 |
| Acute renal failure | (12/217), 5.5 | (15/298), 5 | 0.80 |
| TIA/stroke | (0/217), 0 | (4/300), 1.3 | 0.037 |
| Variable | OR | 95% CI | p Value |
|---|---|---|---|
| Center | 4.38 | 2.89–6.45 | <0.0001 |
| TIMI 0 or 1 | 3.93 | 2.48–6.22 | <0.0001 |
| Use of IIa/IIIb GP inhibitors | 1.76 | 1.16–2.68 | 0.008 |
| Diabetes | 0.58 | 0.36–0.92 | 0.03 |
In-hospital complication rates were similarly low in both groups ( Table 2 ), whereas transient ischemic attack and stroke were slightly more frequent in C-PPCI group.
Thirty-day MACE ( Table 4 ) was comparable between the AT-PPCI and the C-PPCI groups, 10.6% versus 9.7%, respectively, p = 0.73 (unadjusted odds ratio 1.11, 95% confidence interval [CI] 0.62 to 1.98); 30-day mortality 3.2% versus 3.7%, respectively, p = 0.64 (unadjusted odds ratio 0.80, 95% CI 0.31 to 2.06). Logistic regression analyses ( Table 5 ) adjusting for pertinent variables revealed that Killip class ≥II on arrival, PPCI performed during weekend days, and older age were independently associated with worse 30-day MACE. The use of AT was not associated with a better outcome. Similar results were obtained in a model adjusting for the propensity score alone. The use of IIb/IIIa GP inhibitors was not associated with 30-day MACE, nor was the interaction between the use of thrombus aspiration and IIb/IIIa GP inhibitors.
