OUP user menu

★ Editor's choice ★

Impact of thrombus aspiration during primary percutaneous coronary intervention on mortality in ST-segment elevation myocardial infarction

Awsan Noman, Mohaned Egred, Alan Bagnall, Ioakim Spyridopoulos, Sheila Jamieson, Javed Ahmed
DOI: http://dx.doi.org/10.1093/eurheartj/ehs309 3054-3061 First published online: 17 September 2012


Aims To assess the impact of thrombus aspiration during primary percutaneous coronary intervention (PPCI) on the mortality of patients with ST-elevation myocardial infarction (STEMI) patients.

Methods and results Retrospective analysis of prospectively collected data on 2567 consecutive PPCI-treated STEMI patients between 2008 and 2011. Cox proportional hazard models and multiple logistic regression analysis were used to adjust for known covariates. Thrombectomy was performed in 1095 patients (42.7%). Post-PPCI thrombolysis in myocardial infarction 3 flow was more frequently achieved in the thrombectomy group [adjusted odds ratio (OR); 1.92, 95% confidence interval (CI): 1.34–2.76, P = 0.0004]. Overall in-hospital and longer term (mean follow-up 9.9 months) mortality rates were 4.5 and 9.0%, respectively. Thrombectomy was associated with a significant reduction in in-hospital (adjusted OR: 0.51, 95% CI: 0.29–0.93, P = 0.027) and longer term mortality [adjusted hazard ratio (HR): 0.69, 95% CI: 0.48–0.96, P = 0.028]. With propensity weighting, the adjusted HR for longer term mortality for thrombectomy was 0.43 (95% CI: 0.19–0.97; P = 0.042). The association between thrombectomy and reduced longer term mortality was only significant in those with a total ischaemic time ≤180min (P = 0.001) but not in patients with a total ischaemic time >180min (P = 0.99).

Conclusion This study of real-world, unselected STEMI patients demonstrates that thrombus aspiration during PPCI is associated with a significant reduction in mortality, especially in those with a short total ischaemic time. These findings support the use of thrombectomy during PPCI in this group of patients.

  • Myocardial infarction
  • Angioplasty
  • Thrombectomy
  • Mortality
  • Total ischaemic time

See page 3005 for the editorial comment on this article (doi:10.1093/eurheartj/ehs326)


Coronary artery plaque rupture and the subsequent formation of an occlusive thrombus are central to the pathophysiology of the majority of ST-segment elevation myocardial infarction (STEMI) cases. Mechanical reperfusion with primary percutaneous coronary intervention (PPCI) is now recognized as the preferred strategy to achieve the immediate restoration of blood flow and limit myocardial infarction (MI).13

During PPCI, distal embolization of thrombus and plaque fragments may lead to microvascular obstruction, no-reflow, and failure to achieve normal flow in the infarct-related artery.4 Manual thrombus aspiration has been shown to improve coronary perfusion as assessed by time to ST-segment resolution and myocardial blush grade (MBG).59 However, the evidence supporting the benefit of thrombus aspiration on clinical outcomes is limited and inconsistent.1012

The primary aim of this observational study was to examine the effect of thrombectomy during PPCI on in-hospital and 1-year mortality of real-world, consecutive STEMI patients. The secondary endpoint was to assess this in relation to the total ischaemic time and examine the impact of thrombectomy on post-PPCI thrombolysis in MI (TIMI) 3 flow.


Study population

All consecutive STEMI patients undergoing PPCI between March 2008 and June 2011 at the Freeman Hospital, UK were included. The Freeman Hospital is a large tertiary centre performing over 800 PPCI procedures per year.

The diagnosis of STEMI was based on the presence of chest pain suggestive of myocardial ischaemia lasting longer than 30min accompanied by ST-segment elevation or new left bundle branch block on the electrocardiogram (ECG). Patients were considered for PPCI if they presented within 12h of symptom onset.1,2

All patients with suspected STEMI underwent 12-lead ECG at the first point of contact with the medical profession. This was typically performed by ambulance crews in the community or in the emergency department of referring hospitals. All ECGs were telemetered to the Freeman Coronary Care Unit where the decision to perform PPCI was taken. STEMI patients were given 300 mg of aspirin and were transferred directly to our cardiac catheterization laboratory. On arrival either 600 mg of clopidogrel or 60 mg of prasugrel were administered along with standard doses of heparin or bivalirudin according to the established guidelines. Glycoprotein (GP) IIb/IIIa inhibitors were administered to the majority of the patients in the absence of any contra-indication.

Coronary angiography was performed via the radial or femoral artery. The culprit lesion was identified and crossed with an angioplasty guidewire. Manual thrombus aspiration was performed at the discretion of the operator as per the standard protocol followed by conventional percutaneous coronary intervention to the culprit vessel.

Study design

This was a retrospective observational cohort study. The primary source of data was our local coronary artery disease (CAD) database (Dendrite) which holds information on each PCI procedure performed at our hospital. Baseline demographics, clinical presentation, procedure details and procedural complications were prospectively collected at the end of each procedure by the performing physician. Post-procedural complications, clinical data and discharge medications were updated on discharge.

Thrombus aspiration procedure

Patients were classified into thrombus aspiration or non-thrombus aspiration groups. The thrombus aspiration group included all patients in whom thrombus aspiration was attempted. All thrombus aspiration was performed manually using the following devices: Export (Medtronic; 93%), Eliminate (Terumo; 1.7%), Pronto (Vascular Solutions; 1.6%), and others including Hunter (IHT), QuickCat (Spectranetics), and Rescue (Boston Scientific) in 3.7%.

Outcome measures

The primary outcome measures were mortality during the index hospital admission (in-hospital mortality) and mortality up to 1-year follow-up (longer term mortality). Mortality data were provided by the Office of National Statistics, which records all mortality in the UK. This information was linked to our database using a National Health Service (NHS) patient-unique identification number (NHS number) and confirmed by the patient's date of birth and home address. Mortality was assessed up to the 31 July 2011, and the patient follow-up was censored when they died.

The secondary outcome measure was post-PPCI TIMI flow grade and the impact of thrombectomy on longer term mortality in relation to the total ischaemic time. The total ischaemic time was defined as the time from the onset of symptoms to the time of the first intracoronary reperfusion instrument. This was a thrombus aspiration catheter, a balloon or a stent. TIMI flow and procedure timings were obtained from our PCI database. These data were prospectively entered into the database by the interventional cardiologist at the end of each PPCI procedure.

Statistical analysis

Data are presented as percentages for categorical variables and as means ± standard deviation (SD) or medians and interquartile ranges (IQR: 25th to 75th) for continuous variables. Comparisons between groups were performed using the χ2 test for categorical variables and the independent t-test or Mann–Whitney U test for continuous variables as appropriate. Multiple logistic regression analysis was used to test for the effect of thrombus aspiration on post-procedure TIMI flow and in-hospital mortality. For longitudinal analysis, Kaplan–Meier survival curves were generated and the log-rank test was used to assess the differences between the survival distributions. The effects of known prognostic variables13,14 on the mortality of STEMI patients were examined using Cox proportional hazards regression analysis. In order to further minimize the confounding influences, the effect of thrombectomy on longer term survival was examined with propensity-weighted Cox proportional hazards regression with adjustment for other covariates not accounted for in the propensity model. The propensity score (a conditional probability of exposure to a treatment given observed covariates) was determined using a logistic regression model. Propensity scoring was performed for age, gender, previous angina or MI, peripheral vascular disease, admission haemoglobin, admission serum creatinine, admission serum glucose, PPCI access site, pre-PCI TIMI 2 or 3 flow, culprit vessel diameter, culprit lesion length, stent usage, GP IIb/IIIa inhibitor use, multi-vessel CAD, year of PPCI, angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers on discharge, and total ischaemic time. Missing data for these variables were rare (0–3% for each variable) except for admission serum glucose (6.9%). We tested for interaction between thrombectomy and the following covariates in the above Cox proportional hazards model: the total ischaemic time as a categorical variable at a cut-off of 180min, TIMI 2/3 pre-PPCI, gender, age at a cut-off of 70 years, anterior MI site, and diabetes.5 A two-sided P < 0.05 was considered statistically significant. All analysis was performed using SPSS (version 16, SPSS Inc., Chicago, USA).


Patient and procedural characteristics

A total of 2567 STEMI patients (mean age 63.2 ± 13.4 years, 70.3% male) were studied. 1095 patients (42.7%) underwent thrombectomy. Baseline demographics, clinical and procedural characteristics are shown in Table 1.

View this table:
Table 1

Demographics, clinical and procedural characteristics of thrombectomy and non-thrombectomy groups

Non-thrombectomy (n = 1472)Thrombectomy (n = 1095)P
Age (years)65.1 ± 13.560.7 ± 12.9<0.0001
Male (%)6972.10.10
 Haemoglobin (g/dL)13.7 ± 1.913.8 ± 1.70.046
 Creatinine (µmol/L)102.7 ± 53.194.0 ± 39.8<0.0001
 Glucose (mmol/L)8.4 ± 3.48.8 ± 3.50.006
 Cholesterol (mmol/L)4.9 ± 1.45.0 ± 1.30.013
Risk factors
 Hypertension (%)43.538.80.017
 Hypercholesterolaemia (%)30.436.20.003
 Family history (%)45.745.60.99
 Diabetes mellitus (%
 Current smoking (%)41.749.40.0002
 BMI (kg/m2)27.1 ± 5.227.9 ± 4.9<0.0001
Past history
 Angina (%)24.517.2<0.0001
 MI (%)17.911.6<0.0001
 Previous CABG (%)
 Previous PCI (%)
 CVA/TIA (%)
 PVD (%)
 Airways disease (%)13.311.90.30
 Heart rate (b.p.m.)76 ± 1975 ± 200.20
 Systolic BP (mmHg)132 ± 30127 ± 29<0.0001
 Cardiogenic shock (%)
 Ventilated (%)
 Anterior MI (%)38.738.20.97
 Radial (%)64.173.6<0.0001
 Multi-vessel CAD (%)35.631.20.02
 GP IIb/IIIa (%)74.689.6<0.0001
 Vessel diameter [mm, median (IQR)]3.5 (3.0–3.5)3.5 (3.25–4.0)<0.0001
 Lesion length [mm, median (IQR)]20 (15–28)24 (18–33)<0.0001
 Stent usage (%)
 Vessel ≥2.5 mm94.396.10.07
 Door to first devicea [minutes, median (IQR)]27 (20–38)23 (18–31)<0.0001
 Total ischaemic time [minutes, median (IQR)]177 (123–285)159 (114–256)0.006
 Total ischaemic time up to 180min (%)50.757.9<0.0001
 Out-of hours procedure (%)
 Night-time procedure (%)21.823.30.20
 Impaired LVSFb (%)63.566.80.23
Drugs on discharge
 Aspirin (%)90.790.90.89
 ACEi or ARB (%)84.988.10.021
 Beta blocker (%)82.985.80.058
 Clopidogrel (%)
 Statin (%)89.390.20.50
  • Data are presented as the mean ± SD unless indicated otherwise.

  • BMI, body mass index; MI, myocardial infarction; CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention; CVA/TIA, cerebrovascular accident/transient ischaemic attack; PVD, peripheral vascular disease; b.p.m., beat per minute; BP, blood pressure; CAD, coronary artery disease; GP IIb/IIIa, glycoprotein IIb/IIIa inhibitor; IQR, interquartile range; LVSF, left ventricular systolic function; ACEi, angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker.

  • aEither balloon, thrombectomy device or stent.

  • bLVSF data were available in only 1076 patients (41.8%).

The overall rate of in-hospital complications (excluding in-hospital death) was 13.4% in the thrombectomy group compared with that of 12.7% in the non-thrombectomy group (P = 0.59). In-hospital complications included procedural cardiac complications (coronary dissection, perforation, side branch occlusion, no reflow and tamponade), arterial access complications or any other in-hospital complications (CVA, bleeding, re-intervention and renal failure). None of these complication subsets significantly differed between the groups.

Predictors for thrombectomy use

Thrombectomy use steadily increased during the course of the study. The percentages of STEMI patients undergoing thrombus aspiration every year between 2008 and 2011 were 13.6, 47.6 and 61.7% (P < 0.0001). We found a significant variation in thrombectomy use between different operators, Figure 1. The predictors of thrombectomy use in multiple logistic regression analysis are shown in Table 2.

View this table:
Table 2

Multiple logistic regression model for predictors of thrombectomy use

Odds ratio (95% CI)P
Age (per 10 years increase)0.78 (0.73–0.85)<0.0001
Female gender1.04 (0.83–1.30)0.77
Anterior MI location0.97 (0.79–1.19)0.79
Non-routine-working hour PPCI1.02 (0.83–1.25)0.84
Year of PPCIa3.37 (2.95–3.86)<0.0001
Interventional cardiologistb2.0 (1.63–2.44)<0.0001
TIMI 2/3 pre-PPCI0.21 (0.16–0.27)<0.0001
Vessel diameter >2.5 mm2.82 (1.87–4.25)<0.0001
Lesion length >15 mm1.59 (1.23–2.05)<0.0001
Total ischaemic time >180 min0.83 (0.68–1.02)0.071
  • See Table 1 for abbreviations.

  • aSecond half of study cohort vs. first half.

  • bHigh (≥40%) vs. low volume (<40%) operators.

Figure 1

Thrombectomy use with different operators.

In-hospital mortality

A total of 30 patients (2.7%) died in-hospital in the thrombectomy group compared with 85 (5.8%) in the non-thrombectomy group (unadjusted OR: 0.46, 95% CI: 0.30–0.70, P = 0.0003). In multiple logistic regression model adjusted for age, gender, admission heart rate and systolic blood pressure, multi-vessel CAD, diabetes mellitus, previous MI, anterior MI location, pre-PCI TIMI 2 or 3 flow, the total ischaemic time, admission haemoglobin, and creatinine, thrombectomy was associated with a significant reduction in in-hospital all-cause mortality (adjusted OR: 0.51, 95% CI: 0.29–0.93, P = 0.027).

Longer term mortality

During a mean follow-up period of 9.9 ± 3.8 months (9.6 ± 3.6 months in the thrombectomy group vs. 10.2 ± 3.8 months in the non-thrombectomy group), 232 patients (9.0%) died. Of these, 170 patients were from the non-thrombectomy group (11.6%) and 62 from the thrombectomy group (5.7%) (unadjusted HR: 0.49, 95% CI: 0.37–0.66, P < 0.0001). Kaplan–Meier cumulative mortality curves are shown in Figure 2. In the Cox multivariate proportional hazards model, thrombus aspiration was associated with significantly lower longer term mortality even after correcting for confounders (adjusted HR: 0.69, 95% CI: 0.48–0.96, P = 0.028). All covariates in this model and their hazard ratios (HRs) are shown in Table 3.

View this table:
Table 3

Predictors of 1-year mortality in Cox proportional hazard model

Hazard ratio (95% CI)P
Thrombus aspiration0.69 (0.48–0.96)0.028
Age (per 10-year increase)1.67 (1.45–1.93)<0.0001
Female gender1.03 (0.73–1.44)0.88
Diabetes mellitus1.19 (0.78–1.80)0.42
Heart rate (per 10 b.p.m. increase)1.16 (1.10–1.23)<0.0001
Systolic BP (per 10 mmHg increase)0.87 (0.82–0.92)<0.0001
Previous MI1.46 (1.01–2.12)0.048
Anterior MI1.39 (1.01–1.917)0.046
Previous CABG2.53 (1.24–5.15)0.010
Haemoglobin (per 1 g/dL increase)0.82 (0.75–0.90)<0.0001
Creatinine (per 10 µmol/L increase)1.03 (1.02–1.04)<0.0001
Multi-vessel CAD1.73 (1.26–2.38)0.001
TIMI 2/3 pre-PPCI0.70 (0.48–1.02)0.06
Total ischaemic time (per 20min increase)1.006 (1.001–1.010)0.014
  • See Table 1 for abbreviations.

Figure 2

Kaplan–Meier survival curves for adjusted cumulative mortality in thrombectomy and non-thrombectomy groups. PPCI, primary percutaneous coronary intervention.

In the subgroup of patients in whom left ventricular systolic function (LVSF) was documented on discharge (n = 1076, 41.8%), thrombus aspiration remained an independent predictor of reduced 1-year mortality even after correcting for LVSF (adjusted HR: 0.37, 95% CI: 0.19–0.72, P = 0.003).

Sufficient data to calculate the propensity score for thrombectomy use were available in 2160 patients (1168 in the thrombectomy group and 992 in the non-thrombectomy group) (C-statistic for the propensity score of 0.78). With adjustment for propensity score and other covariates included in the above Cox proportional model, the HR in the thrombus aspiration group for longer term mortality was 0.43 (95% CI: 0.19–0.97, P = 0.042).

The above Cox proportional hazard model was repeated with the total ischaemic time entered into the model as a categorical variable at a cut-off point of 180min. A significant interaction was noted between thrombectomy and the total ischaemic time (Pinteraction = 0.024). The analysis was then performed separately for the two subgroups according to their total ischaemic time (up to or >180min). The association between thrombus aspiration and reduced longer term mortality was only significant in patients with the total ischaemic time up to 180min (adjusted HR: 0.41, 95% CI: 0.23–0.70, P = 0.001) but not in patients with the total ischaemic time >180min (adjusted HR: 1.0, 95% CI: 0.63–1.58, P = 0.99). With propensity scoring, the adjusted HR for thrombus aspiration in patients with the total ischaemic time up to 180min was 0.22 (95% CI: 0.07–0.77, P = 0.018) compared with 0.93 (95% CI: 0.41–2.10, P = 0.86) in patients with the total ischaemic time >180min. Kaplan–Meier curves for adjusted cumulative mortality in the thrombectomy and non-thrombectomy groups according to the total ischaemic time are shown in Figure 3.

Figure 3

Kaplan–Meier survival curves for adjusted cumulative mortality in the thrombectomy and non-thrombectomy groups according to the total ischaemic time. PPCI, primary percutaneous coronary intervention.

No significant interaction was seen between thrombectomy and the following factors: pre-PPCI TIMI 2/3 flow grade (Pinteraction = 0.16), age >70 years (P = 0.72), female gender (P = 0.60) anterior MI (P = 0.55), or diabetes (P = 0.22).

TIMI flow grade in the infarct-related artery

Patients in the thrombectomy group were less likely to have TIMI 2 or 3 flow pre-PCI compared with the non-thrombectomy group (11.9 vs. 33.0% respectively, P < 0.0001). Despite this, the rate of post-PPCI TIMI 3 flow grade was significantly higher in the thrombectomy group (94.6 vs. 88.7%, unadjusted OR: 2.23, 95% CI: 1.63–3.06, P < 0.0001), Figure 4. After adjustment for age, gender, cardiogenic shock, total ischaemic time, stent use, pre-PPCI TIMI flow grade, and GP IIb/IIIa, thrombectomy use was associated with a significantly increased rate of post-PPCI TIMI 3 flow (adjusted OR: 1.92, 95% CI: 1.34–2.76, P = 0.0004).

Figure 4

Thrombolysis in myocardial infarction flow grade pre- and post-primary percutaneous coronary intervention in thrombectomy and non-thrombectomy groups. PCI, percutaneous coronary intervention.

Longer term mortality was significantly lower in patients with post-PPCI TIMI 3 flow grade compared with those with TIMI 0/1/2 flow grade (6.9 vs. 26.4%, adjusted HR: 0.41, 95% CI: 0.27–0.62, P < 0.0001).


This study of unselected, real-world STEMI patients demonstrates that thrombus aspiration during PPCI is associated with an increase in the rate of post-PPCI TIMI 3 flow grade and a significant reduction in in-hospital and longer term mortality. This association was evident especially in patients with a total ischaemic time of up to 180min. To our knowledge, this is the largest cohort of PPCI-treated STEMI patients to have demonstrated an association between thrombus aspiration and reduced mortality.

Our overall in-hospital and longer term mortality rates15,16 and aspiration catheter usage17 were consistent with recent PPCI studies. Thrombectomy use in our cohort had steadily increased since the publication of the Thrombus Aspiration during Percutaneous coronary intervention in Acute myocardial infarction Study (TAPAS) in 2008.5 We found that thrombus aspiration was less likely to be used in older patients and in those with culprit lesions in smaller diameter arteries. Features including greater tortuosity, diffuse disease and calcification reduce the likelihood of successful delivery of the thrombectomy catheter18 and their greater prevalence in elderly patients could explain the reduced usage in this subgroup. In addition, we found patients with pre-PPCI TIMI 2 or 3 flow were less likely to undergo thrombectomy. This is likely to be due to a lower visible thrombus burden in these patients compared with those with pre-PPCI TIMI 0 or 1 flow as well as the belief held by some interventional cardiologist of the lack of thrombectomy benefit in patients with pre-PPCI TIMI 2 or 3 flow. Such unfounded belief also explains why patients with pre-PPCI TIMI 2 or 3 flow were excluded from some thrombus aspiration trials. TAPAS, however, did not demonstrate a significant difference in the effect of thrombectomy on MBG between patients with different pre-PPCI TIMI flow.5 Similarly, we did not find a significant interaction between pre-PPCI TIMI flow and the impact of thrombectomy on mortality.

Despite the lower prevalence of pre-PPCI TIMI 2 or 3 flow in the thrombectomy group, post-PPCI TIMI 3 flow grade was more prevalent in this group compared with that in the non-thrombectomy group. In fact, the rate of post-PPCI TIMI 3 flow grade increased by 90% with thrombectomy use even after adjusting for several potential confounders. Achieving post-PPCI TIMI 3 flow grade is a strong predictor of clinical outcomes in PPCI-treated STEMI patients.14 Mortality rates in our study were significantly lower in this group compared with those with post-PPCI TIMI 0/1/2 flow. These findings of improved post-PPCI TIMI 3 flow with thrombectomy give some insights into the potential mechanism by which thrombectomy may impact on the survival of STEMI patients.

Several randomized control trials have confirmed the beneficial effect of thrombus aspiration during PPCI on coronary flow and myocardial perfusion.5,79,19 To date, TAPAS remains the largest of these studies.5 Although TAPAS failed to demonstrate a significant benefit for thrombus aspiration on the secondary endpoints of post-PCI TIMI flow grade or 30-day mortality, there was a significant reduction in all-cause and cardiac 1-year mortality with thrombus aspiration.10 One-year mortality, however, was not a pre-specified endpoint in TAPAS and mortality rates in this trial were higher compared with other recent PPCI studies.20 The late mortality benefit seen in TAPAS may be explained by the possible beneficial effect of thrombectomy on LV remodelling and infarct size.21

Five subsequent meta-analyses of mortality outcomes with thrombus aspiration during PPCI have been published; two demonstrating a reduction in mortality with thrombectomy11,22 and the other three reporting neutral outcomes.12,23,24 These conflicting results might be due to differences in the type of thrombectomy device used as well as follow-up durations. Improved survival was seen with manual thrombectomy and longer follow-up durations, whereas no mortality benefits were found with mechanical thrombectomy and shorter follow-up durations. Furthermore, a recent large observational study found an increase in all-cause mortality with thrombus aspiration.17 This study, however, was limited by the failure to control for many potential confounders such as cardiogenic shock, which was more frequent in the thrombus aspiration group, and the lack of information on clinical presentation and type of thrombus aspiration device used. Finally, the recently published INFUSE-AMI trial has also added to the controversy in this field.25 Despite its highly selected population, thrombectomy did not impact on the markers of reperfusion or infarct size at 30 days. However, the effect of thrombectomy on 30-day infarct size was a secondary endpoint and the trial was not powered to assess the clinical hard endpoints such as mortality outcomes.

In TAPAS,5 a strong but non-significant trend for interaction (P = 0.09) was noted for the total ischaemic time at a cut-off point of 180min and the effect of thrombectomy on MBG. In our study, the observed improvement in longer term mortality outcomes with thrombectomy appeared to be dependent on the total ischaemic time. Thrombectomy use was associated with mortality reduction only in patients with the total ischaemic time of up to 180min. To our knowledge, this is the first report in the literature to demonstrate heterogeneity in the effect of thrombectomy with regard to any clinical outcome. Such observation points out that thrombectomy use may have to be limited to specific subgroups of STEMI patients and is important to be taken into consideration when designing future randomized trials in this field.

Current STEMI guidelines recommend the use of thrombus aspiration as an adjunct therapy during PPCI.1,2 This recommendation is mainly based on the result of TAPAS, despite its limitations. The result of the ongoing Thrombus Aspiration in ST-Elevation myocardial infarction in Scandinavia trial,26 which aims to recruit 5000 STEMI patients, is expected in 2013. Until this time, our findings further support the use of thrombectomy during PPCI and identify a subgroup in which thrombus aspiration is associated with better survival.


This is a retrospective observational study and as such it was not possible to account for all confounding influences. For example, we did not have complete data on culprit lesion characteristics such as target vessel tortuousity, visible thrombus burden, or details of post-PCI blush grade or time to ST resolution. Such data would likely have improved both our understanding of case selection for thrombectomy and the mechanism of improved survival. In addition, our data lack information on whether or not the aspiration catheter crossed the culprit lesion and the amount of thrombus material removed. Another limitation is this being a single-centre study. However, our hospital provides PPCI to a population of ∼ 2 million and serves seven satellite hospitals.


This large observational study of consecutive, unselected STEMI patients demonstrates that manual thrombus aspiration during PPCI is associated with a significant increase in post-procedure TIMI 3 flow grade and a reduction in mortality, particularly in patients with a short total ischaemic time. These findings support the use of thrombus aspiration in this group of patients though require the confirmation in adequately powered randomized control trials.



Conflict of interest: none declared.


We are grateful to our colleagues at Freeman Hospital, Drs R. Edwards, A. Zaman, I. Purcell, R. Das and V. Kunadian, and Professor B. Keavney for their help in collecting data. We are also grateful to Dr Simon Ogston (Senior Statistician, University of Dundee, UK) for his advice and help with statistical analysis.


View Abstract