JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY VOL. 64, NO.

11, 2014

ª 2014 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION ISSN 0735-1097/$36.00

PUBLISHED BY ELSEVIER INC. http://dx.doi.org/10.1016/j.jacc.2014.05.064

Effect of Aspiration Thrombectomy

on Microvascular Obstruction
in NSTEMI Patients
The TATORT-NSTEMI Trial

Holger Thiele, MD,*y Suzanne de Waha, MD,*z Uwe Zeymer, MD,x Steffen Desch, MD,*y Bruno Scheller, MD,k
Bernward Lauer, MD,{ Tobias Geisler, MD,# Meinrad Gawaz, MD,# Oliver Gunkel, MD,** Leonhard Bruch, MD,yy
Norbert Klein, MD,zz Dietrich Pfeiffer, MD,zz Gerhard Schuler, MD,* Ingo Eitel, MD*

ABSTRACT

BACKGROUND Aspiration thrombectomy in ST-segment elevation myocardial infarction is recommended by current

guidelines based on several randomized trials. There are no trials assessing thrombectomy in non–ST-segment elevation
myocardial infarction (NSTEMI) patients.

OBJECTIVES The TATORT-NSTEMI (Thrombus Aspiration in Thrombus Containing Culprit Lesions in Non–ST-Elevation
Myocardial Infarction) trial sought to assess the effect of aspiration thrombectomy on microvascular injury in patients
with NSTEMI compared with standard percutaneous coronary intervention (PCI).

METHODS This prospective, controlled, multicenter study randomized 440 patients to adjunctive thrombectomy
(n ¼ 221) compared with conventional PCI (n ¼ 219) in NSTEMI patients with thrombus-containing lesions. The primary
endpoint of the extent of microvascular obstruction (MO) in the percentage of left ventricular mass (%LV) was assessed
by cardiac magnetic resonance imaging within 4 days. Secondary endpoints included infarct size, myocardial salvage
index, and angiographic parameters including myocardial blush grade and Thrombolysis In Myocardial Infarction flow
grade. The combined clinical endpoint consisted of death, reinfarction, target vessel revascularization, and new
congestive heart failure within 6 months.

RESULTS The primary endpoint of MO was not different between the thrombectomy and the standard PCI group with
2.0%LV (interquartile range [IQR]: 0.8 to 4.1) versus 1.4%LV (IQR: 0.7 to 2.6) (p ¼ 0.17). Similarly, no significant dif-
ferences were observed for infarct size (8.6%LV; IQR: 4.0 to 14.7 vs. 7.4%LV; IQR: 4.1 to 13.1; p ¼ 0.46), myocardial
salvage index (63.3; IQR: 35.4 to 87.2 vs. 65.6; IQR: 46.9 to 82.6; p ¼ 0.45), or angiographic parameters such as blush
grade (p ¼ 0.63) and Thrombolysis In Myocardial Infarction flow grade (p ¼ 0.66). Clinical follow-up at 6 months
revealed no differences in the combined clinical endpoints (p ¼ 0.22).

CONCLUSIONS Aspiration thrombectomy in conjunction with PCI in NSTEMI with a thrombus-containing lesion does
not lead to a reduction in MO. (Thrombus Aspiration in Thrombus Containing Culprit Lesions in Non-ST-Elevation
Myocardial Infarction [TATORT-NSTEMI]; NCT01612312) (J Am Coll Cardiol 2014;64:1117–24) © 2014 by the American
College of Cardiology Foundation.

From the *Department of Internal Medicine/Cardiology, University of Leipzig Heart Center, Leipzig, Germany; yMedical Clinic II,
University Hospital Schleswig-Holstein, University of Lübeck, Lübeck, Germany; zHeart Center Bad Segeberg, Bad Segeberg,
Germany; xInstitut für Herzinfarktforschung, Klinikum der Stadt Ludwigshafen, Ludwigshafen, Germany; kDepartment of
Internal Medicine III, University of Saarland, Homburg, Germany; {Department of Cardiology, Zentralklinik Bad Berka, Germany;
#Department of Cardiology/Cardiovascular Medicine, University of Tübingen, Tübingen, Germany; **Department of Internal
Medicine II, Klinikum Frankfurt/Oder, Frankfurt/Oder, Germany; yyDepartment of Internal Medicine, Unfallkrankenhaus Berlin,
Berlin, Germany; and the zzDepartment of Internal Medicine I, University of Leipzig, Leipzig, Germany. The TATORT-NSTEMI trial
is an investigator-initiated study that was supported by unrestricted grants from Terumo Europe, Leuven, Belgium; Daiichi-Sankyo,
Munich, Germany; and Lilly Germany, Bad Homburg, Germany. The authors are solely responsible for the design and conduct of the
study and all study analyses and the drafting and editing of the paper and its final contents. These companies had no influence
1118 Thiele et al.
Thrombus Aspiration in NSTEMI
T
ABBREVIATIONS he use of manual thrombectomy
AND ACRONYMS during percutaneous coronary inter-
vention (PCI) in patients with acute
CMR = cardiac magnetic
resonance
ST-segment elevation myocardial infarction
(STEMI) has been shown to improve perfu-
IQR = interquartile range
sion and in meta-analyses also to decrease
LV = left ventricular
cardiac death and repeat myocardial infarc-
%LV = percentage of
left ventricular mass
tion (1–3). Consequently, current European
and U.S. guidelines recommend manual
MO = microvascular
obstruction thrombectomy in patients with STEMI (4,5).
NSTEMI = non–ST-segment However, in non–ST-segment elevation
elevation myocardial infarction myocardial infarction (NSTEMI), w50% to
PCI = percutaneous coronary 70% of all patients also display relevant
intervention thrombus burden in the culprit vessel (6–8).
STEMI = ST-segment elevation Furthermore, with rates of angiographic no-
myocardial infarction
reflow ranging from 15% to 40% depending
TIMI = Thrombolysis In
on thrombus burden and intraprocedural
Myocardial Infarction
thrombotic events, in a significant portion of
patients with NSTEMI, only suboptimal reperfusion
success can be achieved (9–11). Thus, thrombectomy
in patients with NSTEMI may represent a useful
intervention with the potential to reduce myocardial
damage and improve patient prognosis. Due to a
lack of data, current guidelines do not give a
clear recommendation for thrombus aspiration in
NSTEMI (12).
SEE PAGE 1125
The lack of guideline recommendations underlines
the need for an adequately powered randomized
clinical trial to address the current role of thrombus
aspiration in patients with NSTEMI and relevant
thrombus. The TATORT-NSTEMI (Thrombus Aspira-
tion in Thrombus Containing Culprit Lesions in
Non-ST-Elevation Myocardial Infarction) trial was
designed to test the hypothesis that adjunctive
thrombectomy compared with conventional
alone leads to a reduction in microvascular obstruc-
tion (MO) assessed by cardiac magnetic resonance
(CMR) imaging.
METHODS
TATORT-NSTEMI is a prospective, controlled, multi-
center, randomized, open-label trial designed to
compare PCI with routine thrombus aspiration with PCI
without thrombus aspiration in NSTEMI patients and
on the study design, data collection, data analysis, and final drafting of this manuscript. Dr. Thiele has received grants from Eli
Lilly and Terumo. Dr. Zeymer has received consulting and research grants from The Medicines Company and Terumo. All other
authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Listen to this manuscript’s audio summary by JACC Editor-in-Chief Dr. Valentin Fuster.
You can also listen to this issue’s audio summary by JACC Editor-in-Chief Dr. Valentin Fuster.
Manuscript received March 3, 2014; revised manuscript received May 4, 2014, accepted May 5, 2014.
JACC VOL. 64, NO. 11, 2014
SEPTEMBER 16, 2014:1117–24
relevant thrombus burden undergoing early invasive
PCI. The detailed methods and design of this trial were
published previously (13). In brief, inclusion criteria
were the following: 1) ischemic symptoms lasting >20
min; 2) occurrence of last symptoms <72 h before
randomization; 3) cardiac troponin T levels above the
99th percentile; 4) culprit lesion containing thrombus
(Thrombolysis In Myocardial Infarction [TIMI] thrombus
grades 2 to 5 within the lesion); and 5) intended early PCI.
Exclusion criteria were cardiogenic shock, STEMI, no
identifiable culprit lesion, coronary morphology ineli-
gible for thrombectomy (e.g., very tortuous vessels, se-
vere calcification); indication for acute bypass surgery;
age younger than 18 and older than 90 years; contrain-
dications to treatment with heparin, aspirin, or thieno-
pyridines; pregnancy; current participation in another
clinical study; comorbidity with a limited life
expectancy <6 months; and contraindications to CMR at
study entry. All patients gave written informed consent
before randomization. The lead ethical committee at the
University of Leipzig and at all local ethical committees
of the participating sites approved the study.
PRIMARY AND SECONDARY ENDPOINTS. The pri-
mary study endpoint of the TATORT-NSTEMI trial
was defined as the extent of late MO assessed by CMR
at days 1 to 4 after randomization. Secondary end-
points included other CMR parameters of myocardial
damage, such as infarct size, myocardial salvage (14),
myocardial salvage index, and left ventricular (LV)
ejection fraction. Furthermore, angiographic markers
of reperfusion success such as the TIMI flow post-PCI
and myocardial blush grade were assessed.
For enzymatic infarct size determination, high-
sensitivity troponin T after 24 and 48 h was
evaluated. A combined clinical endpoint of death,
reinfarction, target vessel revascularization, and new
PCI
congestive heart failure was assessed within 6
months of follow-up. The detailed definitions for the
individual endpoints were published previously and
reflect standard definitions (13,15). Clinical outcome
at follow-up was assessed by a telephone interview.
Any clinical event was verified by hospital or general
practitioner records. All clinical and safety endpoints
were adjudicated based on data provided by the
clinical trial sites by a Clinical Endpoints Committee
blinded to treatment assignment.
JACC VOL. 64, NO. 11, 2014
SEPTEMBER 16, 2014:1117–24
Safety assessment included stroke and bleeding
until hospital discharge. Bleeding severity was
classified according to the GUSTO (Global Use of
Strategies to Open Occluded Coronary Arteries)
criteria as severe or life-threatening, moderate, or
mild bleeding.
CMR IMAGE ACQUISITION AND ANALYSIS. The pri-
mary endpoint of late MO and other secondary
endpoints, such as infarct size and myocardial
salvage, were assessed by CMR performed on a 1.5- or
3.0-T scanner on days 1 to 4 after randomization
using a standard scan protocol, as described previ-
ously (13). In brief, late MO and infarct size were
assessed using delayed-enhancement short-axis im-
ages covering the whole left ventricle w15 min after
injection of gadolinium chelate. An inversion recov-
ery turbo gradient echo sequence was used for
image acquisition. For determination of edema/area
at risk, short-axis slices covering the whole left
ventricle using a T2-weighted imaging triple in-
version recovery turbo spin echo sequence before
contrast administration were obtained. The myo-
cardial salvage index was calculated as edema (area
at risk) minus infarct size divided by the area at risk,
as described previously (14). Assessment of LV func-
tion and volumes was performed by a standard
steady-state free precession technique acquiring
short-axis slices from base to apex. LV ejection frac-
tion was calculated from the short-axis functional
views. The scan protocol and core laboratory image
analysis were standardized at all sites and have been
used effectively in previous multicenter studies
(16–19). The blinded observers used certified CMR
evaluation software (cmr42, Circle Cardiovascular Im-
aging Inc., Calgary, Alberta, Canada). Semiautomated
computer-aided threshold detection was applied to
identify regions of edema, hypointense cores, MO,
and infarcted myocardium, as previously described
(8–10). Infarct size, area at risk, and MO were ex-
pressed as the percentage of LV mass (%LV). The CMR
core laboratory has excellent reproducibility and
low interobserver and intraobserver variability for
infarct size and myocardial salvage assessment (12).
ASSESSMENT OF SECONDARY ANGIOGRAPHIC
ENDPOINTS. A central blinded analysis for TIMI
flow, blush grade, and thrombus burden was per-
formed at the angiographic core laboratory of the
University of Saarland. Evaluation of thrombus
burden was performed according to the TIMI
thrombus grade scale.
STATISTICAL ANALYSIS. The aim of this study was
to demonstrate superiority of manual thrombus
aspiration compared with standard PCI with respect
Thiele et al. 1119
Thrombus Aspiration in NSTEMI
460 NSTEMI patients with thrombus
Not randomized (n=20)
440 NSTEMI patients
221 assigned to thrombectomy 219 assigned to standard PCI
No CMR (n=40) No CMR (n=27)
Claustrophobia (n=11) Claustrophobia (n=5)
PM/ICD (n=2) PM/ICD (n=3)
Obesity (n=1) Obesity (n=1)
Death (n=3) Death (n=3)
Other (n=23) Renal insuff. (n=1)
Other (n=14)
Primary endpoint analysis MO (n=181) Primary endpoint analysis MO (n=192)
Secondary endpoint MBG (n=184) Secondary endpoint MBG (n=179)
Secondary endpoint TIMI-flow (n=221) Secondary endpoint TIMI-flow (n=219)
Clinical follow-up 6 months (n=220) Clinical follow-up 6 months (n=218)
F I G U R E 1 Study Flow Chart
Study flow chart indicating the number of randomized patients and patients with endpoint
assessment. CMR ¼ cardiac magnetic resonance; insuff. ¼ insufficiency; MBG ¼ myocardial
blush grade; MO ¼ microvascular obstruction; NSTEMI ¼ non–ST-segment elevation
myocardial infarction; PCI ¼ percutaneous coronary intervention; PM/ICD ¼ pacemaker/
implantable cardioverter-defibrillator; TIMI ¼ Thrombolysis In Myocardial Infarction.
to the primary endpoint of late MO. Initially, the
sample size calculation estimated an absolute dif-
ference in the primary endpoint of late MO of 1.0 
3.3%LV between the 2 treatment groups (20). Based
on a ¼ 5%, a 2-sided hypothesis test, and a statis-
tical power of 80%, 172 patients per group were
needed. During the course of the trial, 2 other studies
were published on MO and NSTEMI, reporting a lower
prevalence and extent of MO than expected (21,22).
Therefore, the Steering Committee decided to increase
the sample size to 204 patients per group, assuming
F I G U R E 2 Patient With MO After Thrombotic Occlusion of the Left Circumflex Artery
(A) Contrast-enhanced image showing transmural inferolateral necrosis with a core of late
microvascular obstruction (MO). (B) Contours for assessment of infarct size and the extent
of MO (red contour, endocardial; green, epicardial; dark blue, remote myocardium; pink,
infarcted myocardium; light blue, MO). (C) Computer-aided signal intensity analysis of
infarct size and MO normalized to normal, uninjured myocardium. The yellow overlay
(infarct size) indicates a signal intensity of >5 SDs above remote, uninjured myocardium.
The orange overlay within the infarct indicates the computer-detected area of MO.
1120 Thiele et al. JACC VOL. 64, NO. 11, 2014

Thrombus Aspiration in NSTEMI SEPTEMBER 16, 2014:1117–24

an absolute difference in MO of 0.5  1.8%LV between

T A B L E 1 Patient and Procedural Characteristics
the 2 treatment groups with a mean extent of 1.0%LV
Thrombectomy Standard PCI of late MO in the whole study cohort. This difference
(n ¼ 221) (n ¼ 219) p Value
was judged clinically relevant (20). On the basis of
Age, yrs 69 (58–75) 68 (58–74) 0.60
clinical experience and data on STEMI, it was further
Male 158 (72) 163 (74) 0.54
Cardiovascular risk factors expected that thrombectomy could not be performed
Current smoking 92 (42) 80 (37) 0.26 in 5% of all patients in the thrombectomy group and
Hypertension 180 (81) 166 (65) 0.12 that as many as 10% of all patients would have no
Hypercholesterolemia 81 (37) 85 (39) 0.67 CMR data (19,20). This resulted in a total sample size of
Diabetes mellitus 64 (29) 64 (29) 0.98 2  230 patients.
BMI, kg/m2 28 (25–31) 28 (25–31) 0.55
Pre-specified subgroup analyses by sex, the pres-
Previous infarction 16 (7) 26 (12) 0.10
ence of diabetes, TIMI thrombus burden (2 to 4 vs. 5),
Previous PCI 12 (6) 29 (13) <0.01
Previous CABG 7 (3) 10 (5) 0.45
TIMI flow pre-PCI (0 to 1 vs. 2 to 3), and additional
Systolic blood pressure, mm Hg 144 (125–160) 140 (121–158) 0.32 administration of glycoprotein IIb/IIIa inhibitors
Heart rate, beats/min 76 (68–86) 76 (67–85) 0.78 were performed. Interactions between treatment
GRACE score 145 (116–168) 137 (115–162) 0.27 groups were tested by a general linear model. The
Radial arterial access 30 (14) 29 (13) 0.90 pre-specified subgroup analysis based on symptom
Killip class on admission 0.88
duration could not be performed due to a large
1 196 (88) 194 (89)
number of missing variables as symptom onset could
2 22 (10) 22 (10)
not be clearly determined in w50% of patients.
3 3 (1) 3 (1)
4 1 (1) 0 A multivariable logistic regression analysis was
Diseased vessels 0.21 performed to assess predictors of MO. The variables
1 87 (39) 105 (48) included were sex, previous infarction, smoking, hy-
2 84 (39) 64 (29) pertension, hypercholesterolemia, diabetes, culprit
3 50 (22) 50 (23) lesion location, number of diseased vessels, and TIMI
Infarct-related artery 0.56
flow and TIMI thrombus grade before PCI.
Left anterior descending 78 (35) 73 (33)
All data were analyzed according to the intention-
Left circumflex 87 (39) 90 (41)
Right coronary 50 (23) 51 (23)
to-treat principle. The differences in primary and
Left main 1 (1) 0 secondary endpoints between treatment groups were
Bypass graft 5 (2) 5 (2) assessed for continuous numerical variables by the
Thrombectomy performed 217 (98) 0 <0.001 Wilcoxon rank sum test. Categorical variables were
Aspiration of macroscopic 162 (73) – analyzed by the chi-square or Fisher exact test. For
thrombus material
clinical events, the hazard ratio and 95% confidence
Primary stenting 131 (59) 96 (44) <0.01
Drug-eluting stent 154 (75) 153 (70) 0.56
interval are reported as well as Kaplan-Meier curves
Bare-metal stent 65 (29) 66 (30) 0.89 with log-rank comparison. A 2-tailed p value <0.05
Concomitant medications was considered statistically significant.
Bivalirudin 5 (2) 5 (2) 0.99
Unfractionated heparin 199 (90) 198 (90) 0.98 RESULTS
LMW heparin 17 (8) 16 (7) 0.87
Beta-blockers 207 (94) 207 (95) 0.66
Of the 460 patients initially scheduled for randomi-
ACE inhibitors/AT-1 185 (84) 187 (86) 0.61
antagonist zation, only 440 patients were ultimately enrolled.
Aspirin 219 (99) 217 (99) 0.62 Due to a technical error in the Internet-based ran-
Clopidogrel 57 (26) 57 (26) 0.95 domization system, 20 data slots were erroneously
Prasugrel 92 (41) 83 (38) 0.61 marked as randomized patients. This was only
Ticagrelor 72 (33) 79 (33) 0.66
discovered at the final database lock and analysis.
Glycoprotein IIb/IIIa inhibitors 13 (6) 14 (6) 0.83
Given the unambiguous study results, the Steering
Statins 212 (96) 211 (97) 0.77
Aldosterone antagonist 45 (20) 32 (15) 0.11
Committee decided to close the study at 440 patients
and not to randomize another 20 patients. Thus, 221
Values are median (interquartile range) or n (%). were randomly assigned to thrombectomy and 219 to
ACE ¼ angiotensin-converting enzyme; AT-1 ¼ angiotensin 1; BMI ¼ body mass index; CABG ¼ coronary artery
bypass graft; GRACE ¼ Global Registry for Acute Coronary Events; LMW ¼ low molecular weight; PCI ¼ primary
standard PCI (Figure 1).
percutaneous coronary intervention. All patients received the assigned treatment, and
thrombectomy was successfully performed in 217
patients (98%) in the thrombectomy group with
macroscopic thrombus aspiration in 73%. There was
JACC VOL. 64, NO. 11, 2014 Thiele et al. 1121
SEPTEMBER 16, 2014:1117–24 Thrombus Aspiration in NSTEMI

no crossover in the standard PCI group. The baseline

characteristics and medication were similar between Baseline variable No. patients p Value Mean difference in MO %LV
groups with the exception of a higher rate of previous for interaction (95% confidence interval)

PCIs in the standard PCI group (Table 1). Procedural All Patients 373
characteristics and in-hospital medication also were
Male sex 273 0.97
similar between groups. However, in the thrombec- Female sex 100
tomy group, significantly more patients underwent
Diabetes 99 0.73
primary stenting without pre-dilation of the lesion
No diabetes 274
(p < 0.001) (Table 1).
TIMI thrombus grade 2-4 246 0.74
CMR IMAGING. Results of CMR imaging were avail- TIMI thrombus grade 5 127

able in 181 (82%) of the thrombectomy group and

TIMI-flow pre PCI 0-1 199 0.82
in 192 (88%) of the standard PCI group (Figure 1). TIMI-flow pre PCI 2-3 174
Reasons for not undergoing CMR are listed in Figure 1.
Glycoprotein IIb/IIIa-inhibitor 24 0.25
The median time between randomization and CMR No glycoprotein IIb/IIIa-inhibitor 349
was 1 day (interquartile range [IQR]: 1 to 3 days) in the
Direct stenting 166 0.23
thrombectomy group and 1 day (IQR: 1 to 2 days) in
No direct stenting 207
the standard PCI group (p ¼ 0.49). Figure 2 shows a
-3 -2 -1 0 1 2 3
typical finding on CMR imaging in a patient with a Standard PCI Thrombectomy
lateral infarction with MO. There was no difference in better better

the primary endpoint of late MO or in any other

parameter derived from CMR imaging (Table 2).
There was also no effect of thrombectomy in the pre-
defined subgroups (Figure 3). Multivariable predictors
of MO presence were pre-PCI TIMI flow, diabetes, and
culprit lesion location in the left anterior descending
F I G U R E 3 Primary Endpoint MO and Subgroups
Mean difference and 95% confidence intervals in the primary endpoint of MO in the %LV
between the standard PCI group and the thrombectomy group for predefined and post-hoc
subgroups. %LV ¼ percentage of left ventricular mass; PCI ¼ percutaneous coronary
intervention; other abbreviations as in Figure 1.

and left circumflex arteries.

ANGIOGRAPHIC ENDPOINTS. Angiographic central

assessment of the TIMI flow grades was possible in (p ¼ 0.16). Thus, the composite major adverse cardiac
all patients, whereas the blush grade was only event rate at 6 months of follow-up was 7.3% after
assessable in 83% of the patients due to technical thrombectomy and 10.6% in the standard PCI group
reasons and/or image quality (Figure 1). There were (hazard ratio: 0.68; 95% confidence interval: 0.36 to
no differences in angiographic parameters at base- 1.28; p ¼ 0.22), as shown in Figure 4.
line as well as after PCI in the 2 randomized groups Safety events were rare and not significantly
(Table 3). different between both study groups. There were no
in-hospital strokes or severe/life-threatening bleed-
ENZYMATIC INFARCT SIZE. The enzymatic infarct
ings. Moderate bleeding occurred in 4 patients (2%) in
size assessed by high-sensitivity troponin T values each group (p ¼ 0.99).
was similar in the thrombectomy group and the
standard PCI group with 960 ng/l (IQR: 295 to
T A B L E 2 Cardiac Magnetic Resonance Imaging Results
2,077 ng/l) versus 739 ng/l (IQR: 360 to 1,549 ng/l) at
24 h (p ¼ 0.24) and also at 48 h (1,038 ng/l [IQR: 362 to Thrombectomy Standard PCI
(n ¼ 181) (n ¼ 192) p Value
2,010 ng/l] vs. 885 ng/l [IQR: 337 to 1,562 ng/l];
Area at risk (edema), %LV 21.0 (16.6–25.8) 20.3 (14.9–25.0) 0.11
p ¼ 0.20).
Infarct size, %LV 8.6 (4.0–14.7) 7.4 (4.1–13.1) 0.42
CLINICAL OUTCOME. At 6 months of follow-up, Myocardial salvage, %LV 11.2 (7.0–16.9) 11.9 (7.4–17.2) 0.78
there were 6 deaths (2.8%) in the thrombectomy Myocardial salvage index 63.3 (35.4–87.2) 65.6 (46.9–82.6) 0.45

group versus 10 deaths (4.6%) in the standard PCI MO present 58 (32.0) 58 (30.2) 0.79
MO, %LV 1.9 (0.8–4.1) 1.4 (0.7–2.6) 0.17
group (p ¼ 0.30). Nonfatal reinfarctions occurred in 4
LV ejection fraction, % 50 (44–56) 52 (44–59) 0.10
(1.9%) after thrombectomy and in 5 (2.4%) in the
LV end-diastolic volume, ml 140 (112–165) 136 (115–168) 0.80
standard PCI group (p ¼ 0.72); target vessel revascu-
LV end-systolic volume, ml 66 (52–90) 65 (51–86) 0.42
larization was observed in 2.8% versus 1.9%
(p ¼ 0.54). New congestive heart failure was regis- Values are median (interquartile range) or n (%).
LV ¼ left ventricular; %LV ¼ percentage of left ventricular mass; MO ¼ microvascular
tered in 4 patients (1.8%) in the thrombectomy group obstruction; PCI ¼ percutaneous coronary intervention.
and 9 patients (4.2%) in the standard PCI group
1122 Thiele et al. JACC VOL. 64, NO. 11, 2014

Thrombus Aspiration in NSTEMI SEPTEMBER 16, 2014:1117–24

T A B L E 3 Angiographic Results
100
Thrombectomy
Cumulative Freedom From Death, Reinfarction,

Thrombectomy Standard PCI

TVR, and New Congestive Heart Failure

(n ¼ 221) (n ¼ 219) p Value

Standard PCI
80 TIMI thrombus grade 0.91
P=0.22; log-rank before PCI
2 29/221 (13) 21/219 (10)
60
3 59/221 (27) 66/219 (30)
(%)

4 57/221 (26) 61/219 (28)

40 5 76/221 (34) 71/219 (32)
TIMI flow before PCI 0.61
0 86/221 (39) 81/219 (37)
20
I 20/221 (9) 13/219 (6)
II 58/221 (26) 70/219 (32)
0 III 57/221 (26) 55/219 (25)
0 30 60 90 120 150 180
TIMI flow post-PCI 0.66
Time After Randomization (Days)
Number at risk 0 4/221 (2) 9/219 (4)
Thrombectomy 221 210 209 209 208 203 195 I 1/221 (1) 4/219 (2)
Standard PCI 219 204 201 200 199 192 186
II 19/221 (9) 13/219 (6)
III 197/221 (89) 193/219 (88)
F I G U R E 4 Secondary Combined Clinical Endpoint
Myocardial blush grade 0.92
before PCI
Kaplan-Meier curve of the combined clinical endpoint of death, reinfarction, target vessel 0 80/184 (44) 72/179 (40)
revascularization (TVR), and new congestive heart failure at 6-month follow-up. PCI ¼ I 47/184 (26) 56/179 (31)
percutaneous coronary intervention. II 54/184 (29) 50/179 (28)
III 3/184 (2) 1/179 (1)
Myocardial blush grade 0.63
post-PCI
DISCUSSION 0 4/181 (2) 4/179 (2)
I 8/181 (4) 7/179 (4)

The main results of this first randomized multicenter
trial assessing aspiration thrombectomy in patients
with NSTEMI with thrombus-containing lesions in
the infarct-related artery can be summarized as fol-
lows. Aspiration thrombectomy led to aspiration of
macroscopic thrombus material in 75% of patients,
which resulted in a significantly higher rate of pri-
mary stent implantation without pre-dilation in the
thrombectomy group compared with the standard
PCI arm. However, aspiration thrombectomy did
not reduce the extent of MO compared with standard
PCI without thrombectomy, which is corroborated
by a lack of benefit in secondary endpoints.
In patients with STEMI, the pivotal
(Thrombus Aspiration During Percutaneous Coronary
Intervention in Acute Myocardial Infarction Study)
trial comparing a strategy of routine aspiration
thrombectomy with a strategy without thrombectomy
resulted in improved reperfusion indexes as the pri-
mary endpoint (2). Although not powered for clinical
outcome, cardiac mortality at 1 year also was reduced
(3). In contrast, in 1 recent randomized, controlled trial
involving high-risk anterior STEMI patients, throm-
bectomy in addition to standard PCI did not result in a
reduction of infarct size (23). Several meta-analyses in
STEMI patients demonstrated a mortality benefit after
thrombectomy compared with standard PCI alone
(1,24–27). However, it also has been shown that single
II 49/181 (27) 55/179 (31)
III 120/181 (66) 113/179 (63)
Values are n/N (%).
PCI ¼ percutaneous coronary intervention; TIMI ¼ Thrombolysis In Myocardial
Infarction.
or multicenter study design had a significant impact on
outcomes in these trials, and the major effect was
driven by the single-center TAPAS trial. Nevertheless,
based on these data, current American and European
guidelines suggest performing manual aspiration
thrombectomy in STEMI with a Class IIa Level of Evi-
TAPAS dence: B recommendation (4,5). These guideline rec-
ommendations were recently challenged by the largest
randomized trial so far, which did not show a mortality
benefit with routine aspiration thrombectomy (28).
Another large-scale trial assessing the effects of
routine thrombus aspiration on clinically important
endpoints is currently ongoing (29).
In the majority of both STEMI and NSTEMI patients,
the key pathophysiological substrate is rupture of a
vulnerable plaque with subsequent coronary throm-
bosis (Central Illustration). Whereas in STEMI, the
thrombus is mostly fibrin rich leading to total vessel
occlusion, the thrombus in many patients with
NSTEMI is predominantly platelet rich and unstable,
leading to only partial or intermittent occlusion of the
JACC VOL. 64, NO. 11, 2014 Thiele et al. 1123
SEPTEMBER 16, 2014:1117–24 Thrombus Aspiration in NSTEMI

coronary vessel (12). Despite these pathophysiological

considerations, the majority of patients with NSTEMI
demonstrate relevant thrombus burden or even
occluded coronary arteries and thus might benefit
from thrombectomy (6,8,11). Nevertheless, current
guidelines do not recommend thrombus aspiration in
NSTEMI patients (12,30). This is mainly due to the lack
of data in NSTEMI patients with only 1 published
prospective observational study so far (7). In this
observational study that included 70 NSTEMI pa-
tients, thrombus was detected in w50% on the initial
angiogram. Manual thrombus aspiration was associ-
ated with a significant reduction of the TIMI thrombus
score and an increase in the rate of TIMI flow grade 3.
After thrombus aspiration, direct stenting without
pre-dilation was possible in 56% of patients (7), which
is slightly lower than in our trial. Trials in STEMI pa-
tients also show heterogeneous direct stenting rates
ranging from 21% to 92% (26).
STUDY LIMITATIONS. A negative trial always raises a
question of power. Our initial sample size calculation C E N T R A L I L L U S T R A T I O N Theoretical Advantage of Thrombectomy in NSTEMI

was mainly based on assumptions due to a lack of data

on NSTEMI. Based on new data published during the
course of the trial, the sample size was increased.
However, 20 patients, in principle eligible for the trial,
did not undergo final randomization due to a technical
error. Thus, it may be argued that this trial is under-
Illustration of the theoretical pathophysiological advantage of thrombectomy before
percutaneous coronary intervention (PCI) in non–ST-segment elevation
myocardial infarction (NSTEMI). Thrombectomy may reduce the thrombus burden and
prevent distal embolization with subsequent reduction of reperfusion injury
(i.e., microvascular obstruction), if applied before PCI. This was the hypothesis of the
TATORT-NSTEMI trial.

powered. However, as there was numerically an even

greater extent of MO in the thrombectomy arm with no
CONCLUSIONS
effect in the secondary CMR, angiographic, and enzy-
matic endpoints, any benefit of thrombectomy is
Manual aspiration thrombectomy in patients with
highly questionable. There was a slightly higher rate of
thrombus-containing lesions did not reduce the
left circumflex artery lesions, and this may have
extent of no-reflow compared with standard PCI as
resulted in smaller infarcts than in other coronary ar-
measured by MO. These findings are supported by a
teries. However, thrombus-containing lesions are
lack of benefit in angiographic, enzymatic, and clin-
more often observed inferolaterally or poster-
ical secondary endpoints.
olaterally, and this increases the infarct size and af-
fects clinical outcome (6). MO is a surrogate endpoint,
REPRINT REQUESTS AND CORRESPONDENCE: Dr.
and evidently assessing clinical outcome as a primary
Holger Thiele, University Hospital Schleswig-Holstein,
endpoint would lead to more robust data. However,
University of Lübeck, Medical Clinic II, Ratzeburger
late MO has been shown to be a strong independent
Allee 160, 23538 Lübeck, Germany. E-mail: holger.thiele@
predictor of clinical outcome (20). Blinding was not
uksh.de.
possible as a result of the intervention used. Never-
theless, several methods to avoid bias were imple-
PERSPECTIVES
mented, such as blinded CMR imaging and
angiographic core laboratories, reducing any influence
COMPETENCY IN MEDICAL KNOWLEDGE: Thrombus is
of the open-label design of the trial. In clinical trials
frequently identified in the infarct-related artery in patients with
using CMR parameters as surrogate endpoints, the
NSTEMI undergoing coronary angiography.
time period of CMR image acquisition after the index
event/randomization is of importance for the final
TRANSLATIONAL OUTLOOK: Larger trials powered for clin-
quality and reproducibility of the results. Therefore, a
ical endpoints are needed to assess the impact of thrombectomy
narrow time period of 1 to 4 days after randomization
on clinical outcomes in patients with NSTEMI.
was defined, and there were no significant differences
in the timing of CMR between the treatment groups.
1124 Thiele et al.
Thrombus Aspiration in NSTEMI
REFERENCES
1. Burzotta F, De Vita M, Gu YL, et al. Clinical impact
of thrombectomy in acute ST-elevation myocardial
infarction: an individual patient-data pooled anal-
ysis of 11 trials. Eur Heart J 2009;30:2193–203.
2. Svilaas T, Vlaar PJ, van der Hoorst IC, et al.
Thrombus aspiration during primary percutaneous
coronary intervention. N Engl J Med 2008;358:
557–67.
3. Vlaar PJ, Svilaas T, van der Horst IC, et al.
Cardiac death and reinfarction after 1 year in the
Thrombus Aspiration during Percutaneous coro-
nary intervention in Acute myocardial infarction
Study (TAPAS): a 1-year follow-up study. Lancet
2009;371:1915–20.
4. O’Gara PT, Kushner FG, Ascheim DD, et al. 2013
ACCF/AHA guideline for the management of
ST-elevation myocardial infarction. A report of the
American College of Cardiology Foundation/
American Heart Association Task Force on Practice
Guidelines. J Am Coll Cardiol 2013;61:e78–140.
5. Steg PG, James SK, Atar D, et al. ESC Guidelines
for the management of acute myocardial infarc-
tion in patients presenting with ST-segment
elevation. Eur Heart J 2012;33:2569–619.
6. Bahrmann P, Rach J, Desch S, Schuler GC,
Thiele H. Incidence and distribution of occluded
culprit arteries and impact of coronary collaterals on
outcome in patients with non-ST-segment elevation
myocardial infarction and early invasive treatment
strategy. Clin Res Cardiol 2011;100:457–67.
7. Vlaar PJ, Diercks GF, Svilaas T, et al. The feasi-
bility and safety of routine thrombus aspiration in
patients with non-ST-elevation myocardial infarc-
tion. Catheter Cardiovasc Interv 2008;72:937–42.
8. Wang TY, Zhang M, Fu Y, et al. Incidence, dis-
tribution, and prognostic impact of occluded culprit
arteries among patients with non-ST-elevation
acute coronary syndromes undergoing diagnostic
angiography. Am Heart J 2009;157:716–23.
9. McEntegart MB, Kirtane AJ, Cristea E, et al.
Intraprocedural thrombotic events during percu-
taneous coronary intervention in patients with
non-ST-segment elevation acute coronary syn-
dromes are associated with adverse outcomes:
analysis from the ACUITY (Acute Catheterization
and Urgent Intervention Triage Strategy) trial.
J Am Coll Cardiol 2012;59:1745–51.
10. Hong YJ, Jeong MH, Choi YH, et al. Impact of
plaque components on no-reflow phenomenon
after stent deployment in patients with acute cor-
onary syndrome: a virtual histology-intravascular
ultrasound analysis. Eur Heart J 2011;32:2059–66.
11. Zhao X-Q, Theroux P, Snapinn SM, Sax FL.
Intracoronary thrombus and platelet glycoprotein
IIb/IIIa receptor blockade with tirofiban in unstable
angina or non-Q-wave myocardial infarction:
Angiographic results from the PRISM-PLUS Trial
(Platelet Receptor Inhibition for Ischemic Syn-
drome Management in Patients Limited by
Unstable Signs and Symptoms). Circulation 1999;
100:1609–15.
12. Hamm CW, Bassand J-P, Agewall S, et al. ESC
Guidelines for the management of acute coronary
syndromes in patients presenting without persis-
tent ST-segment elevation. Eur Heart J 2011;32:
2999–3054.
13. de Waha S, Eitel I, Desch S, et al. Thrombus
Aspiration in ThrOmbus containing culpRiT lesions
in Non-ST-Elevation Myocardial Infarction
(TATORT-NSTEMI): study protocol for a random-
ized controlled trial. Trials 2013;14:110.
14. Eitel I, Desch S, Fuernau G, et al. Prognostic
significance and determinants of myocardial
salvage assessed by cardiovascular magnetic
resonance in acute reperfused myocardial infarc-
tion. J Am Coll Cardiol 2010;55:2470–9.
15. Thygesen K, Alpert JS, Jaffe AS, et al. Third
universal definition of myocardial infarction. Eur
Heart J 2012;33:2551–67.
16. Eitel I, Wöhrle J, Suenkel H, et al. Intra-
coronary compared with intravenous bolus abcix-
imab application during primary percutaneous
coronary intervention in ST-segment elevation
myocardial infarction. Cardiac magnetic resonance
substudy of the AIDA STEMI trial. J Am Coll
Cardiol 2013;61:1447–54.
17. Patel MR, Smalling RW, Thiele H, et al. Intra-
aortic balloon counterpulsation and infarct size in
patients with acute anterior myocardial infarction
without shock: the CRISP AMI Randomized Trial.
JAMA 2011;306:1329–37.
18. Thiele H, Eitel I, Meinberg C, et al. Randomized
comparison of pre-hospital-initiated facilitated
percutaneous coronary intervention versus pri-
mary percutaneous coronary intervention in acute
myocardial infarction very early after symptom
onset. The LIPSIA-STEMI Trial (Leipzig Immediate
Prehospital Facilitated Angioplasty in ST-Segment
Myocardial Infarction). J Am Coll Cardiol Intv 2011;
4:605–14.
19. Thiele H, Hildebrand L, Schirdewahn C, et al.
Impact of high-dose N-acetylcysteine versus pla-
cebo on contrast-induced nephropathy and
myocardial reperfusion injury in unselected pa-
tients with ST-elevation myocardial infarction
undergoing primary percutaneous coronary inter-
vention. J Am Coll Cardiol 2010;55:2201–9.
20. de Waha S, Desch S, Eitel I, et al. Impact of
early vs. late microvascular obstruction assessed
by magnetic resonance imaging on long-term
outcome after ST-elevation myocardial infarction:
a comparison with traditional prognostic markers.
Eur Heart J 2010;31:2660–8.
21. Cochet A, Lalande A, Lorgis L, et al. Prognostic
value of microvascular damage determined by
cardiac magnetic resonance in non ST-segment
elevation myocardial infarction: comparison
JACC VOL. 64, NO. 11, 2014
SEPTEMBER 16, 2014:1117–24
between first-pass and late gadolinium-enhanced
images. Invest Radiol 2010;45:725–32.
22. Xu J, Song YB, Hahn JY, et al. Comparison of
magnetic resonance imaging findings in non-ST-
segment elevation versus ST-segment elevation
myocardial infarction patients undergoing early
invasive intervention. Int J Cardiovasc Imaging
2012;28:1487–97.
23. Stone GW, Maehara A, Witzenbichler B, et al.
Intracoronary abciximab and aspiration throm-
bectomy in patients with large anterior myocardial
infarction. JAMA 2012;307:1817–26.
24. Bavry AA, Kumbhani DJ, Bhatt DL. Role of
adjunctive thrombectomy and embolic protection
devices in acute myocardial infarction: a compre-
hensive meta-analysis of randomized trials. Eur
Heart J 2008;29:2989–3001.
25. De Luca G, Dudek D, Sardella G, Marino P,
Chevalier B, Zijlstra F. Adjunctive manual throm-
bectomy improves myocardial perfusion and
mortality in patients undergoing primary percu-
taneous coronary intervention for ST-elevation
myocardial infarction: a meta-analysis of ran-
domized trials. Eur Heart J 2008;29:3002–10.
26. Kumbhani DJ, Bavry AA, Desai MY,
Bangalore S, Bhatt DL. Role of aspiration and
mechanical thrombectomy in patients with acute
myocardial infarction undergoing primary angio-
plasty: an updated meta-analysis of randomized
trials. J Am Coll Cardiol 2013;62:1409–18.
27. Mongeon F-P, Belisle P, Joseph L,
Eisenberg MJ, Rinfret S. Adjunctive thrombectomy
for acute myocardial infarction: a bayesian meta-
analysis. Circ Cardiovasc Interv 2010;3:6–16.
28. Fröbert O, Lagerqvist B, Olivecrona GK, et al.
Thrombus aspiration during ST-segment elevation
myocardial infarction. N Engl J Med 2013;369:
1587–97.
29. Jolly SS, Cairns J, Yusuf S, et al. Design and
rationale of the TOTAL trial: a randomized trial of
routine aspiration ThrOmbecTomy with percuta-
neous coronary intervention (PCI) versus PCI ALone in
patients with ST-elevation myocardial infarction un-
dergoing primary PCI. Am Heart J 2014;167:315–21.
30. Jneid H, Anderson JL, Wright RS, et al. 2012
ACCF/AHA focused update of the guideline for the
management of patients with unstable angina/
non–ST-elevation myocardial infarction (updating
the 2007 guideline and replacing the 2011 focused
update). A report of the American College of
Cardiology Foundation/American Heart Associa-
tion Task Force on Practice Guidelines. J Am Coll
Cardiol 2012;60:645–81.
KEY WORDS cardiac magnetic resonance
imaging, microvascular obstruction,
non–ST-segment elevation myocardial
infarction, perfusion, thrombectomy