Association of physician variation in use of manual aspiration thrombectomy with outcomes following primary percutaneous coronary intervention for ST-Elevation Myocardial Infarction

Selected in JAMA Cardiology by D. Giacoppo

The National Cardiovascular Data Registry CathPCI Registry.

Reviewer

Daniele Giacoppo

@DanieleGiacoppo

Interventional cardiologist / Cardiologist

Catania, Italy

Latest contributions

PCR @ TCT 2025 - Get the international perspective! NEO-MINDSET: Early withdrawal of aspirin after percutaneous coronary intervention for acute coronary syndrome PCRonline @ ESC Congress 2025: the interventional cardiology perspective!

View full profile

My Comment

Why this study – the rationale/objective?

Large-scale contemporary randomised trials have challenged the clinical effectiveness and safety of routine manual aspiration thrombectomy (mAT) during primary percutaneous coronary intervention (pPCI), showing no difference in composite and individual endpoints and even a possible increase in the risk of post-procedural stroke.
Consequently, current guidelines have heavily downgraded the use of mAT, admitting a possible role only as a bailout strategy in patients with substantial thrombotic burden.
However, some operators continue to endorse the usefulness of mAT in selected patients and consider the conclusions of major clinical trials as a result of the dilution of the mAT benefits by unselected application. This report from the National Cardiovascular Data Registry (NCDR) Cath PCI sought to describe the temporal trends and clinical outcomes of mAT use during pPCI in a nation-wide cohort of patients.

How was it executed – the methodology?

The NCDR Cath PCI is a national program that collects in-hospital data on patients undergoing cardiac catheterisation and PCI. All pPCIs performed for ST-segment elevation myocardial infarction (STEMI) from July 1, 2009, through June 30, 2016 with complete information were analysed. The primary endpoints were in-hospital mortality and stroke.

The registry was generally linked to the longitudinal data of the Centers for Medicare and Medicaid Services (CMS) and the secondary endpoints included all-cause death and stroke at 30 and 180 days.
The assessment of mAT use in a broad, unselected patient population was performed by using instrumental variable analyses, a statistical method designed to control for hidden bias in observational data. Indeed, the decision to use mAT relies on factors not collected in the registry, such as the burden and characteristics of thrombus observed at coronary angiography. Operator preference for mAT was used as the instrumental variable.

What is the main result?

A total of 683,584 pPCIs were included in the study. Time trends of the proportion of pPCI with mAT showed a progressive increase from Q3 2009 (10.0%) through Q4 2011 (13.8%). Subsequently, there was a progressive decline in mAT use along with the publication of the results of the INFUSE-AMI, TASTE, and TOTAL trials and guidelines updates.

At the end of the study period (Q2 2016) mAT was used in only 4.7% of pPCIs. Median operator use of mAT was 3.5% [0%-13.2%] but wide inter-individual variations were observed among prespecified groups of operators, with 34.1% (n=2,738) of physicians who did not perform mAT during study period and 15.8% (n=1,268) who used this approach in 33.8% [21.6%-83.3%] of pPCIs.

Patients treated with mAT were younger than those not treated with mAT, had more frequently single-vessel disease, required shorter total stent lengths, and received more often glycoprotein IIb/IIIa inhibitors. All the other characteristics, including major cardiovascular risk factors, symptomatic angina, heart failure, cardiogenic shock, and cardiac arrest at presentation were balanced between groups. The instrumental variable application harmonised the characteristics of pPCIs across operators with different frequency of mAT use and improved balance in age, procedure medications, number of diseased vessels, and total length of stents.
Crude incidences of in-hospital death (4.51% vs. 5.76%) and stroke (0.53% vs. 0.56%) did not statistically differ between groups. After instrumental variable analysis, the result in terms of death remained unchanged (adjusted risk difference -0.18%, p=0.29), while the risk of stroke was higher in patients who underwent mAT (adjusted risk difference 0.14%, p=0.03).

At 30- and 180-day follow-up crude incidences of death in patients who received mAT were lower compared with patients who did not receive mAT (9.63% vs. 11.80%; 12.90% vs. 16.10%; log-rank p<0.001) but differences between groups disappeared after instrumental variable analysis application (p=0.86 and p=0.54). The cumulative incidences of stroke were similar between groups both at 30 and 180 days (1.31% vs. 1.26%; 2.20% vs 2.10%; log-rank p=0.56). After instrumental variable analysis results did not change (p=0.40 and p=0.92).

Critical reading and the relevance for clinical practice

The use of mAT has declined substantially in recent years with wide degrees of inter-operator variability. The change in physician behaviour closely followed trial data demonstrating a lack of benefit of mAT and even a possible increased risk stroke. However, mAT continues to be used as adjunctive therapy during pPCIs, presumably in selected cases with significant thrombotic burden.

The findings emerged from this large-scale retrospective analysis of unselected patients show that after adjustment mAT does not result in any short- and mid-term clinical benefit and is associated with a significant excess in in-hospital stroke.

The conclusions of this study need to be considered in light of the following limitations.
Firstly, the design of the study is observational and the decision on whether to use mAT was at the physician’s discretion.
Secondly, the study was not prospective and variables analysis was based on the standardised information collected in the NCDR Cath PCI registry. Relevant angiographic factors such as thrombus characteristics and burden were not described and, although recent guidelines recommend mAT as bailout strategy, there is no certainty that the technique was not applied also upfront and for the treatment of lesions without significant thrombus.
Thirdly, crucial outcomes, such as myocardial infarction and target lesion thrombosis, or surrogate endpoints, such as measurements of myocardial perfusion and microvascular obstruction or stent apposition and expansion as assessed by more advanced methods (i.e. cardiac magnetic resonance and optical coherence tomography respectively) were not available in the study, thus several important questions about mAT role in selected subsets still remain unanswered. Finally, the application of instrumental variable analysis cannot overcome differences in unmeasured characteristics among operators.

In conclusion, whether there is still room for an adequately powered randomised clinical of patients with large thrombotic lesions is uncertain and findings from high-quality, focused investigations or large real-world registries can help in filling the gap in accumulated evidence regarding the real effectiveness and safety of selective mAT use during pPCI.

In light of these results, will you continue to use mAT in selected scenarios or is there no longer justification for such adjunctive tool in contemporary pPCIs?