Association of Thrombus Aspiration With Time and Mortality Among Patients With ST-Segment Elevation Myocardial Infarction

Key Points Question Is thrombus aspiration beneficial for patients who present early with ST-segment elevation myocardial infarction (STEMI)? Findings In this post hoc analysis of a randomized clinical trial, thrombus aspiration was associated with a greater reduction in distal embolization in patients who presented early with STEMI; however, it was not associated with improved clinical outcomes irrespective of time. Total ischemic time and door-to-device time were also important factors associated with mortality among patients with STEMI. Meaning This analysis suggests that thrombus aspiration does not appear to be associated with a benefit irrespective of ischemic time among patients with STEMI.


Introduction
In ST-segment elevation myocardial infarction (STEMI), time to reperfusion has been identified as one of the factors most strongly associated with outcome. [1][2][3][4] Ischemic time may be associated with the amount of myocardial recovery after primary percutaneous coronary intervention (PCI).
Observational analyses have found manual thrombus aspiration to be less effective for patients with longer ischemic times, so there is a hypothesis that thrombus aspiration is beneficial for patients who present early. 5 Others have argued that patients with longer ischemic times have an organized thrombus and thus have greater need for and benefit from thrombus aspiration. The TOTAL (Thrombectomy With PCI vs PCI Alone in Patients With STEMI) trial randomized 10 732 patients with STEMI to undergo either upfront thrombus aspiration with PCI or PCI alone and so was poised to answer this question. 6,7 Overall, in the TOTAL trial, manual thrombectomy followed by PCI compared with PCI alone did not reduce the risk of the primary outcome of cardiovascular (CV) death, myocardial infarction (MI), cardiogenic shock, or heart failure and was in fact associated with an increased risk of stroke. 6,8 We sought to assess whether there was a benefit associated with thrombectomy for patients with STEMI based on total ischemic time and first medical contact (FMC)-to-device time. In addition, we examined the association between time (total ischemic time and FMC-to-device time) with outcomes in this large contemporary STEMI trial.

TOTAL Study Design
The design of the TOTAL trial has been previously published and described in detail. 7 In brief, this was an international multicenter trial that randomized 10 732 patients presenting with STEMI to 1 of 2 strategies: routine thrombectomy followed by PCI vs PCI alone with only bailout thrombectomy.
Patients were randomized in a 1:1 ratio, and all outcomes were adjudicated blinded to treatment assignment. The TOTAL trial included patients with STEMI who presented within 12 hours of symptom onset and who were referred for primary PCI and provided informed consent. The TOTAL trial excluded patients who had received a previous coronary artery bypass graft and those who received fibrinolytic therapy. 7 Eligible patients were randomly assigned to 1 of the 2 strategies and received routine thrombectomy followed by PCI or PCI alone. Randomization was performed via online randomization at the Population Health Research Institute, Hamilton, Ontario, Canada, with random block sizes. Randomization was performed by study staff and clinicians. Detailed guidance was provided to instructors for the optimal technique for thrombus aspiration, including continued suction during removal of the device. 7 All participating sites had local ethics board approval for the trial. All participants provided written informed consent. No additional ethics board approval was required for this analysis as no new data were collected. This study followed the Consolidated

PCI Alone
The patients randomized to undergo PCI alone had the procedure performed under the operators' standard technique without thrombectomy. Predilation vs direct stenting was at the discretion of the operator. Bailout thrombectomy was allowed if the initial PCI attempt failed and the operator was unable to regain flow in the infarcted artery or if there was evidence of a residual large thrombus after predilation ballooning or stenting.

TOTAL Study Outcomes
The primary outcome was defined as the composite of the first occurrence of CV mortality, recurrent MI, cardiogenic shock, or New York Heart Association (NYHA) class IV heart failure within 180 days.
The primary safety outcome was stroke within 30 days, previously described in detail. 7 A formal clinical events committee blinded to the study groups adjudicated all primary and safety outcome events, major bleeding, transient ischemic attacks, target vessel revascularization, and stent thromboses.

Statistical Analysis
A modified intent-to-treat analysis was used to include only patients who underwent PCI for index STEMI because randomization occurred prior to angiography. All P values were from 2-sided tests and results were deemed statistically significant at P < .05. Patients were followed up for 1 year from randomization.

Subgroup Analysis
Data analysis was performed from February 22, 2019, and January 5, 2021. In the TOTAL trial, analyses were prespecified for the subgroups of time of symptom onset to FMC, defined as less than 6 hours vs 6 to 12 hours. 6 For this analysis, we performed a more detailed post hoc exploratory subgroup analysis using a Cox proportional hazards regression model within the following 2 subgroups: (1) total ischemic time (<2 hours, 2-6 hours, or >6 hours) and (2) FMC-to-device time (<60 minutes, 60-90 minutes, >90-120 minutes, or >120 minutes). Total ischemic time was defined as the symptom-onset-to-device time. We collected and analyzed the results for the individual components of the composite primary outcome (CV mortality, recurrent MI, cardiogenic shock, and NYHA class IV heart failure) as well as CV mortality within 1 year, in addition to distal embolization, no reflow, and final Thrombolysis In Myocardial Infarction (TIMI) 3 flow. We repeated the same subgroup analyses for the primary safety outcome of stroke at 30 days.

Observational Analysis: Time and Outcomes
In terms of observational analysis, we examined the association of time with CV mortality using a multivariable Cox proportional hazards regression model to assess the independent association of time with CV mortality among patients with STEMI for both total ischemic time and FMC-to-device time. Known factors associated with outcome among patients with STEMI were entered into the model and included age, sex, type 1 or 2 diabetes, prior PCI, TIMI thrombus grade, heart rate and systolic blood pressure at presentation, Killip class 2 or greater heart failure at presentation, anterior MI, peripheral arterial disease, and preprocedure TIMI flow. Well-established threshold values for total ischemic time (<2 hours and >6 hours) and FMC-to-device times (<90 minutes and >120 minutes) were used in multivariable analysis. Time was not used as a continuous variable because the association was nonlinear. 7737 of the 9986 patients (77.5%) who underwent primary PCI and had time data available were male. 6 Figure 1A). The TIMI 3 flow rates were not improved, and the no reflow phenomenon was not reduced with thrombus aspiration for patients with both early and late total ischemic times and FMC-to-device times ( Figure 1B and C).
For the primary composite outcome of CV mortality, recurrent MI, cardiogenic shock, or NYHA class IV heart failure, there was no benefit with thrombus aspiration irrespective of time, with no significant interaction between thrombus aspiration time defined by (1)    We plotted FMC-to-device and symptom-onset-to-FMC times by CV mortality and found that the lowest CV mortality at 1 year (12 of 634 [1.9%]) was among patients who sought medical attention within 2 hours of symptom onset and had a door-to-device time of less than 60 minutes.

FMC-to-Device Time and Outcome
The highest CV mortality (18 of 251 [7.2%]) was found among patients who took the longest to seek medical attention, with symptom-onset-to-FMC times of more than 6 hours and door-to-device times of more than 120 minutes.

Discussion
In this large international randomized clinical trial, despite thrombus aspiration being associated with a reduction in distal embolization in those patients with short ischemic times, there was not a corresponding reduction in clinical outcomes. Furthermore, we found a stepwise increase in CV mortality, as well as increasing rates of heart failure with increasing total ischemic time and FMC-todevice time. Time continues to be an important factor associated with outcomes in the contemporary era of primary PCI for patients with STEMI.
It was our hypothesis that thrombus aspiration would be associated with a greater benefit among patients presenting early owing to a larger amount of viable myocardium at risk and the thrombus being more amenable to removal with manual thrombectomy. Although there were greater reductions in distal embolization in those with shorter ischemic times, this did not translate into reductions in the no reflow and clinical outcomes. This is an important finding because   improvements in surrogate angiographic outcomes may not always translate into improvements in clinical outcomes.
Several large trials have examined the effects of aspiration thrombectomy on CV mortality and stroke among patients with STEMI undergoing primary PCI. 6,[9][10][11][12][13] An individual patient meta-analysis of 3 major studies on thrombus aspiration in STEMI (TOTAL, TASTE [Thrombus Aspiration in

ST-Elevation Myocardial Infarction in Scandinavia], and TAPAS [Thrombus Aspiration During
Percutaneous Coronary Intervention in Acute Myocardial Infarction]), with a combined total of 18 306 patients, showed no benefit of routine thrombus aspiration associated with CV mortality. 14 In a subgroup analysis by symptom onset less than 6 hours, 6 to 12 hours, and more than 12 hours, thrombus aspiration was not beneficial for CV mortality in any of these subgroups, which is consistent with our findings.
This analysis from the TOTAL trial showed that both total ischemic time and FMC-to-device times were independent factors associated with CV mortality. These findings are consistent with the  published literature both in the fibrinolytic era and in the era of primary PCI. 3,[15][16][17] De Luca et al 1 reported that time delay to treatment was definitely associated with 1-year mortality.
With improvement in systems of care and prehospital activation, FMC-to-device times have decreased significantly. 2 Current guidelines recommend a FMC-to-device time of 90 minutes or less to 120 minutes for primary PCI, which is consistent with our study that demonstrated increased mortality when FMC-to-device time is delayed. [18][19][20] We found that patients with longer ischemic times were more likely to be older and female and have diabetes. Further studies are needed to assess the reasons for this finding and how this disparity gap for age and sex, in particular, can be improved.
In this modern primary PCI trial, we found a stepwise increase in mortality with increasing total ischemic time and FMC-to-device time. Not only was CV mortality increased, but heart failure was also increased. Thrombus burden was not different between patients with short ischemic times and those with longer ischemic times. Given the high mortality rate, new therapies are needed to target the high-risk population of patients with longer ischemic times.

Limitations
This study has some limitations. First, this analysis is from a randomized clinical trial that required consent. As a result, there is inevitably a selection bias that excludes patients who are very sick and unable to provide consent. Second, patients with symptom onset of up to 12 hours were included in the trial. We thus have no data on patients presenting beyond 12 hours of symptom onset. A third limitation is that the subgroup analyses were not prespecified and should only be considered hypothesis generating.