International Journal of Cardiology 353 (2022) 35–42

Contents lists available at ScienceDirect

International Journal of Cardiology

journal homepage: www.elsevier.com/locate/ijcard

Review

Intravascular ultrasound-guided versus coronary angiography-guided

percutaneous coronary intervention in patients with acute myocardial
infarction: A systematic review and meta-analysis
Frederik T.W. Groenland 1, Tara Neleman 1, Hala Kakar 1, Alessandra Scoccia 1,
Annemieke C. Ziedses des Plantes 1, Pascal R.D. Clephas 1, Sraman Chatterjee 1, Mahova Zhu 1,
Wijnand K. den Dekker 1, Roberto Diletti 1, Felix Zijlstra 1, Karim D. Mahmoud 1,
Nicolas M. Van Mieghem 1, Joost Daemen *, 1
Department of Cardiology, Thoraxcenter, Erasmus University Medical Center, Rotterdam, the Netherlands

A R T I C L E I N F O A B S T R A C T

Keywords: Background: Intravascular ultrasound (IVUS) can overcome the intrinsic limitations of coronary angiography for
Acute myocardial infarction lesion assessment and stenting. IVUS improves outcomes of patients presenting with stable or complex coronary
Coronary angiography artery disease, but dedicated data on the impact of IVUS-guided percutaneous coronary intervention (PCI) in
Intravascular ultrasound
patients with acute myocardial infarction (AMI) remains scarce.
Percutaneous coronary intervention
Methods: We systematically searched Embase, MEDLINE, Web of Science Core Collection, Cochrane Central
Register of Controlled Trials and Google Scholar for studies that compared clinical outcomes for IVUS- versus
angio-guided PCI in patients with AMI. The primary endpoint was all-cause mortality and the secondary endpoint
major adverse cardiovascular events (MACE). Mantel-Haenszel random-effects model was used to calculate
pooled risk ratios (RR) with 95% confidence intervals (CI).
Results: Nine studies (8 observational, 1 RCT) with a total of 838.902 patients (796.953 angio-guided PCI, 41.949
IVUS-guided PCI) were included. In patients with AMI, IVUS-guided PCI was associated with a significantly lower
risk of all-cause mortality (pooled RR: 0.70; 95% CI, 0.59–0.82; p < 0.01), MACE (pooled RR: 0.86; 95% CI,
0.74–0.99; p = 0.04) and target vessel revascularization (TVR) (pooled RR: 0.83; 95% CI, 0.73–0.95; p < 0.01). In
the subset of patients presenting with ST-segment elevation, IVUS-guided PCI remained associated with a
reduced risk for both all-cause mortality (pooled RR: 0.79; 95% CI, 0.66–0.95, p = 0.01) and MACE (pooled RR:
0.86; 95% CI, 0.74–0.99, p = 0.04).
Conclusions: This is the first systematic review and meta-analysis comparing IVUS- versus angio-guided PCI in
patients with AMI, showing a beneficial effect of IVUS-guided PCI on all-cause mortality, MACE and TVR. Results
of ongoing dedicated prospective studies are needed to confirm these findings.

1. Introduction and geographic miss, often remain unrecognized by angiography alone

[2,3].
Despite its well-known limitations, coronary angiography remains Intravascular ultrasound (IVUS) is an intracoronary imaging tech­
the mainstream diagnostic modality to guide percutaneous coronary nique that can overcome these limitations by allowing tailored lesion
intervention (PCI). Coronary angiography is hampered by the inability preparation, stent selection and stent optimization [3–6]. An increasing
to adequately assess lesion severity and visualize intracoronary plaque body of evidence, composed of both randomized and observational data,
characteristics [1]. Moreover, key reasons for stent failure, including demonstrates that IVUS-guided PCI reduces major adverse cardiovas­
underexpansion, stenting-related complications (e.g. edge dissections) cular events (MACE) and target vessel failure (TVF) as compared to

* Corresponding author at: Department of Cardiology, Thoraxcenter, Erasmus University Medical Center, Dr. Molewaterplein 40, Room Rg-628, 3015 GD Rot­
terdam, the Netherlands.
E-mail address: [email protected] (J. Daemen).
1
These authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation.

https://doi.org/10.1016/j.ijcard.2022.01.021
Received 29 November 2021; Received in revised form 29 December 2021; Accepted 10 January 2022
Available online 15 January 2022
0167-5273/© 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
F.T.W. Groenland et al. International Journal of Cardiology 353 (2022) 35–42

angio-guided PCI in a broad spectrum of patients [7–11]. Nevertheless, 2.4. Data extraction and quality assessment
patients with acute myocardial infarction (AMI) were vastly underrep­
resented in most studies. The latter is of particular interest given the fact Extraction of relevant study, baseline and procedural characteristics
that two dedicated studies comparing IVUS- versus angio-guided PCI in and outcome data was independently performed by both reviewers (FG
this subset of patients showed conflicting results for clinical outcomes and TN) with the use of a standardized data extraction form. Baseline
[12,13]. The use of IVUS in the acute setting was linked to higher rates of characteristics of interest were age, sex, and clinical presentation with
spasm and dissection, and increased balloon dilatations that could hy­ STEMI. Procedural characteristics of interest were procedure time and
pothetically lead to higher rates of distal embolization [12,14]. contrast use (endpoints) and the use of drug-eluting stents (DES).
Furthermore, it is unknown how the use of IVUS in patients with AMI Both reviewers (FG and TN) independently performed a systematic
impacts procedural characteristics, such as procedure time and contrast quality assessment of included studies. For observational studies the
use. methodological quality was assessed with the preferred ‘Risk Of Bias In
As the role of IVUS-guided PCI in the setting of AMI remains unclear, Non-randomized Studies - of Interventions (ROBINS-I)’ tool [17]. With
we performed a systematic review and meta-analysis of studies help of this tool, risk of bias in 7 different domains (confounding, se­
comparing clinical outcomes between IVUS-guided and angio-guided lection, intervention classification, deviation from intervention, missing
PCI in patients with AMI. data, measurement of outcome, selection of reported results) was clas­
sified as low, moderate, serious, or critical, resulting in an overall risk of
2. Methods bias judgement. For a randomized controlled trial (RCT) the preferred
‘revised Cochrane risk-of-bias tool for randomized trials (RoB-2)’ tool
The protocol of this systematic review and meta-analysis was regis­ was used to assess methodological quality, scoring risk of bias in 5
tered in the PROSPERO international prospective register for systematic different domains (randomization process, deviation from intended in­
reviews (CRD42021252142). The study was performed according to the terventions, missing outcome data, measurements of outcomes, selec­
Preferred Reporting Items for Systematic reviews and Meta-Analyses tion of reported results) as low, some concerns or high [18].
(PRISMA) extension for searching and the ‘PRISMA 2020 Checklist’ Disagreements were resolved in a consensus meeting, including the
was used (Supplementary Table 1) [15]. opinion of a third reviewer (JD).

2.1. Data sources and search strategy 2.5. Statistical analysis

The systematic search strategy as performed by our hospital’s med­ For the categorical endpoints a pooled risk ratio (RR) with corre­
ical library specialists was previously reported [16]. The following sponding 95% confidence intervals (CI) was calculated using the Mantel-
electronic databases were searched on May 5th, 2021: Embase, MED­ Haenszel random-effects model. For each study, outcome data at
LINE, Web of Science Core Collection, Cochrane Central Register of maximum follow-up time was used for the pooled analyses. If only event
Controlled Trials and Google Scholar. Key search terms included: percentages were reported in an included study, the absolute number of
“intravascular ultrasound” (and/or “intracoronary ultrasound”) and patients with an event was calculated (rounded down). Funnel plots for
“acute myocardial infarction” (and/or “acute heart infarction” and/or the primary and secondary endpoint were obtained to assess the po­
“ST(-segment) elevation myocardial infarction”). No language or pub­ tential of publication bias [19]. Presence of study heterogeneity was
lication date filters were applied. We searched for prospective and quantified with the Q and I2 statistic. To explore a potential cause for
retrospective observational studies and randomized controlled trials study heterogeneity, a subgroup analysis with only STEMI patients was
(RCTs). A complete overview of the search strategy for each database is performed. Moreover, to assess the assumption that studies with serious
provided separately (Supplementary Table 2). risk of bias did not impact the outcome in the main analysis, a sensitivity
analysis was performed.
Review Manager (Rev-Man, version 5.4.1., the Nordic Cochrane
2.2. Study selection process Centre, The Cochrane Collaboration, 2020) was used for statistical
analysis and to acquire forest plots. A p value <0.05 (two-sided) was
After removal of duplicates, two reviewers (FG and TN) indepen­ considered statistically significant.
dently screened all initial search records on title and abstract. Subse­
quently, independent full text evaluation for potentially eligible studies 3. Results
was performed. A study was included if the following entry criteria were
met: 1) Comparison of clinical outcomes between IVUS-guided and 3.1. Search results and study selection process
angio-guided PCI in a study population with AMI, 2) Differentiation
between IVUS-guided and optical coherence tomography (OCT)-guided The initial search strategy resulted in 3183 records. After full-text
PCI if both were compared to angio-guided PCI, 3) Full text availability evaluation of 39 potentially eligible records, 9 studies were included
and 4) No duplicate record (e.g. meeting abstract, studies with similar in the systematic review and meta-analysis (Fig. 1) [12,13,20–26]. One
study populations). An AMI study population was defined as follows: all eligible record was excluded because the study population was also part
patients presented with myocardial infarction, including at least 50% of of a larger included study by Ya’qoub et al. [25,27]
patients having ST-segment elevation myocardial infarction (STEMI) or
undergoing primary PCI. 3.2. Main characteristics and quality assessment of included studies
Disagreements were resolved in a consensus meeting, including the
opinion of a third reviewer (JD). An overview of included studies with main study, baseline and
procedural characteristics is presented in Table 1 and Supplementary
2.3. Outcome measures Table 3. Most studies were based on either prospective or retrospective
observational data except for 1 RCT. Dedicated data on all-cause mor­
The primary endpoint was all-cause mortality and the secondary tality was reported in 7 studies and a composite cardiovascular endpoint
endpoint was MACE (or a similar composite endpoint related to car­ was used in 8 studies. Moreover, data on cardiac death was reported in 4
diovascular disease). Other endpoints of interest included cardiac death, studies and data on TVR in 5 studies. Maximum follow-up time differed
target vessel revascularization (TVR) and procedural characteristics from in-hospital outcome up to 5 years.
(procedure time and contrast use). A total of 838.902 patients with AMI underwent PCI. Angio-guided

36
F.T.W. Groenland et al. International Journal of Cardiology 353 (2022) 35–42

Fig. 1. Flowchart of the study selection process according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement.
Legend: An overview of the study selection process.
AMI is acute myocardial infarction, IVUS is intravascular ultrasound, OCT is optical coherence tomography, PCI is percutaneous coronary intervention.

PCI was performed in 796.953 patients and IVUS-guided PCI in 41.949 all-cause mortality was considered as low (Supplementary Fig. 1).
patients (Table 1). Among studies, age differed from 53.7 to 70.0 years Conversely, the funnel plot for MACE was slightly asymmetric, as
and patients were male in 60.5 to 75.6%. Six studies included only smaller studies with larger standard errors only reported lower RR in
STEMI patients and in most studies drug-eluting stents (DES) were used favor of IVUS.
in the majority of cases. With respect to the other clinical outcomes, IVUS-guided PCI in pa­
Quality assessment was performed using the ROBINS-I tool in 8 tients with AMI was associated with a significantly reduced risk for TVR
studies and the RoB-2 tool in one study (Supplementary Table 4). (pooled RR: 0.83; 95% CI; 0.73–0.95; p < 0.01; I2 = 0%), but the
Overall risk of bias was scored as moderate in 7 observational studies. beneficial effect of IVUS on cardiac death did not reach statistical sig­
One observational study had serious risk of bias due to inappropriate nificance (pooled RR: 0.62, 95% CI, 0.29–1.33; p = 0.22; I2 = 72%)
adjustment for important confounding domains (shock and/or Killip (Fig. 2).
Class) and an unclear intervention definition (the IVUS group was solely In addition to the pooled risk ratios, the unadjusted and adjusted
identified through ICD codes). The included RCT was considered to have clinical event rates as reported by the included studies are provided
some concerns regarding the overall risk of bias. No studies were clas­ separately (Supplementary Table 5). After multivariate adjustment
sified to have a critical risk of bias. IVUS-guided PCI was associated with improved clinical outcomes in 3
studies, while in 3 other studies no significant associations were found.
The three remaining studies provided no (un)adjusted effect measures.
3.3. Pooled analyses for clinical outcomes

In patients with AMI undergoing PCI, the use of IVUS significantly 3.4. Sensitivity and subgroup analyses for clinical outcomes
reduced the risk for all-cause mortality (pooled RR: 0.70; 95% CI,
0.59–0.82; p < 0.01; I2 = 62%) and MACE (pooled RR: 0.86; 95% CI, In the main analyses study heterogeneity was moderate to substan­
0.74–0.99; p = 0.04; I2 = 61%) (Fig. 2). The risk of publication bias for tial (I2 > 50%) for all clinical outcomes, except for TVR (Fig. 2). For all-

37
F.T.W. Groenland et al. International Journal of Cardiology 353 (2022) 35–42

Table 1
Main characteristics of the included studies.
Study /first Year Design Primary Composite endpoint Maximum Patients Age Male STEMI DES
author endpoint Follow-up (n) (years) (%) (%) (%)

IVUS/ IVUS/ IVUS/ IVUS/ IVUS/

angio angio angio angio angio
a
Ahmed [20] 2011 Retrospective MACE, composite of all-cause 1 year 2127/ 61.3/ 75.6/ 58.8/ 89.0/
observational mortality, non-fatal MI, TVR 8235 63.8 71.3 58.9 76.2
a
Kim [21] 2020 Prospective POCE, composite of all-cause 1 year 2333/ NR/ NR/ NR/ NR/
observationalb mortality, any infarction, any 9072 64.0 73.9 53.6 92.2
revascularization
a
Maluenda [12] 2010 Prospective MACE, composite of all-cause 1 year 382/523 63.6/ 66.2/ 100.0/ 79.9/
observationalb mortality, Q-wave MI, TLR 61.1 68.6 100.0 72.3
Nakatsuma 2016 Retrospective TVR MACE, composite of all-cause 5 years 932/2096 65.9/ 74.8/ 100.0/ 39.5/
[22] observational mortality, MI, TVR 67.4 75.1 100.0 11.3
Okura [13] 2019 Prospective All-cause MACE, composite of all-cause In-hospital 1947/689 69.0/ 77.0/ 74.4/ 66.0/
observationalb mortality mortality, cardiac failure, VF/ 70.0 75.0 77.5 49.0
VT, bleeding
a
Wang [23] 2015 RCT MACE, composite of CD, MI, 1 year 38/42 56.4/ 60.5/ 100.0/ NR/NR
TVR, intractable myocardial 53.7 66.7 100.0
ischemia
Witzenbichler 2014 Prospective Definite or MACE, composite of CD, 1 year 421/392 NR/NR NR/NR 100.0/ 100.0/
[24] observationalb probable ST definite/probable ST, MI 100.0 100.0
Ya’qoub [25] 2021 Retrospective Readmission NR In-hospital 33,644/ 61.0/ 74.1/ 100.0/ NR/NR
observational 775,688 62.4 71.0 100.0
a
Youn [26] 2011 Prospective MACE, composite of all-cause 3 years 125/216 60.0/ 74.4/ 100.0/ 100.0/
observationalb mortality, MI, TVR, TLR 61.4 63.0 100.0 100.0

CD is cardiac death, DES is drug-eluting stent, IVUS is intravascular ultrasound, MACE is major adverse cardiovascular event, MI is myocardial infarction, NR is not
reported, POCE is patient orientated composite endpoint, RCT is randomized controlled trial, ST is stent thrombosis, STEMI is ST-segment elevation myocardial
infarction, TLR is target lesion revascularization, TVR is target vessel revascularization, VF is ventricular fibrillation, VT is ventricular tachycardia.
a
Primary endpoint is similar to composite endpoint.
b
Post-hoc analysis.

cause mortality, heterogeneity was mainly caused by conflicting data in TVR were 14% and 17% respectively. A similar numerical reduction was
the studies of Maluenda et al. and Okura et al. [12,13] The large registry observed for cardiac death, although this pooled analysis did not reach
of Ya’qoub et al. in favor of IVUS contributed most to the pooled analysis statistical significance. The discrepancy between all-cause mortality and
(29.4%) [25]. In a sensitivity analysis, in which this study with serious cardiac death could raise questions with respect to the plausibility of the
risk of bias was excluded, IVUS-guided PCI remained associated with a results. However, the sample size of the pooled population was signifi­
significantly lower risk for all-cause mortality (pooled RR: 0.68; 95% CI, cantly lower for cardiac death, since this endpoint was investigated by
0.52–0.88; p < 0.01; I2 = 67%) (Supplementary Fig. 2). For MACE, study only 4 studies, with 3 studies reporting event data for the pooled analysis
heterogeneity was mainly caused by the large study of Kim et al., fa­ (Fig. 3). This resulted in a lower statistical power. Furthermore, het­
voring IVUS with a lower RR as compared to other studies [21]. erogeneity was more pronounced for cardiac death, which might be
In subgroup analyses including studies with only STEMI patients, the explained by the fact that clear definitions were not available for all
pooled RR for all-cause mortality was 0.79 (95% CI, 0.66–0.95; p = 0.01; studies and thus could have differed. Of note, determining the cause of
I2 = 49%) whereas pooled RR for MACE was 0.86 (95% CI, 0.74–0.99; p death can be difficult and is more prone to bias in retrospective and
= 0.04; I2 = 11%) (Fig. 3). Tests for subgroup differences between AMI observational studies. Obviously, these potential pitfalls are less appli­
patients presenting with- or without ST segment elevation did not reach cable to all-cause mortality.
statistical significance. In addition to the present systematic review and meta-analysis, 3
recent observational studies (large registries) compared the use of
3.5. Procedural characteristics intravascular imaging with angio-guided PCI in patients with AMI and
showed similar results [28–30]. Intravascular imaging guidance was
Procedure time (1 study) and contrast use (0 studies) were largely associated with a reduction in all-cause mortality and a composite of
unreported and therefore not compared between both techniques. cardiac death, non-fatal myocardial infarction and stent thrombosis.
Although IVUS was the most frequently used intravascular imaging
4. Discussion modality, these studies were not included in the present study, because
no distinction was made between IVUS and OCT.
This is the first systematic review and meta-analysis assessing the Based on the present findings, it seems reasonable to conclude that
clinical impact of IVUS-guided PCI in patients with AMI. The main the beneficial effect of IVUS-guided PCI also applies to patients with
findings of this study can be summarized as follows: IVUS-guided PCI in AMI. This can be explained by both the general advantages of IVUS, as
patients with AMI is associated with a significantly lower risk for all- well as the specific potential benefits of IVUS in the acute setting. First,
cause mortality (pooled RR 0.70) and MACE (pooled RR 0.86) as pre-intervention IVUS allows accurate sizing of lesion length and lumen
compared to angio-guided PCI. These findings were consistent for AMI and vessel diameter, which enables selection of correct stent and balloon
patients with ST-segment elevation. sizes [1,4,5]. As a result, geographic miss and procedural complications
The results of the present study, specifically focusing on patients due to malsizing can be prevented. Second, IVUS can be used to assess
presenting with AMI, support the profound and growing body of evi­ different plaque types and disease extent which might improve proce­
dence on the use of IVUS to improve PCI outcome in stable and more dural planning and treatment strategies [1,4,5]. Third, post-intervention
complex populations [7–11]. In the pooled analysis for all-cause mor­ IVUS can be used to guide optimization strategies for relevant post-PCI
tality, IVUS-guided PCI was associated with a significant 30% relative findings, such as underexpansion, malapposition and stenting-related
risk reduction in all-cause mortality, while the reductions for MACE and complications (e.g. edge dissections), as well as residual focal lesions

38
F.T.W. Groenland et al. International Journal of Cardiology 353 (2022) 35–42

Fig. 2. Studies comparing intravascular ultrasound-guided versus angio-guided percutaneous coronary intervention in patients with acute myocardial infarction –
Main pooled analyses for clinical outcomes.
Legend: Pooled analyses for all-cause mortality, major adverse cardiovascular events, cardiac death and target vessel revascularization. Risk ratios are provided,
including statistical tests for heterogeneity and overall effect. The horizontal line is the 95% confidence interval.
CI is confidence interval, IVUS is intravascular ultrasound, M-H is Mantel-Haenszel.

or high plaque burden at stent edges [1–3,5,7]. More specifically for and hypothetically also the administration of peri-procedural pharma­
patients presenting with AMI, IVUS can be used to visualize specific cotherapy (e.g. glycoprotein IIb/IIIa receptor antagonist) [33,34].
culprit lesion plaque characteristics, such as plaque ruptures and Conversely, IVUS-guided optimization could lead to increased balloon
attenuation, which are associated with no-reflow [31–33]. Moreover, dilatations with more distal embolization, specifically in case of high
IVUS allows the assessment of thrombus (burden), thrombus protrusion thrombus burden. However, this does not seem to negatively impact
and vulnerable attenuated plaque, which might impact treatment stra­ clinical outcomes. Two included studies reported a higher percentage of
tegies (e.g. aspiration thrombectomy, atherectomy and filter protection) post-dilatation in the IVUS-guided PCI group, but differences in

39
F.T.W. Groenland et al. International Journal of Cardiology 353 (2022) 35–42

Fig. 3. Studies comparing intravascular ultrasound-guided versus angio-guided percutaneous coronary intervention in patients with acute myocardial infarction –
Subgroup analyses for all-cause mortality and major adverse cardiovascular events.
Legend: Subgroup analysis for all-cause mortality and major adverse cardiovascular events, comparing studies with only STEMI patients to studies with both STEMI
and NSTEMI patients. Risk ratios are provided, including statistical tests for the overall effect within the subgroup and difference between subgroups. The horizontal
line is the 95% confidence interval.
CI is confidence interval, IVUS is intravascular ultrasound, M-H is Mantel-Haenszel, NSTEMI is non-ST-segment elevation myocardial infarction, STEMI is ST-segment
elevation myocardial infarction.

40
F.T.W. Groenland et al. International Journal of Cardiology 353 (2022) 35–42

composite cardiovascular endpoints, as compared to the angio-guided present meta-analysis definitions for MACE were largely comparable.
PCI group, were not observed [12,26]. Fourth, differences in DES use, DES type (generation), and lesion
When interpreting the results of the pooled analyses, it is noteworthy complexity between studies and study groups, could potentially have
that intravascular imaging is widely utilized in modern Asian countries, impacted clinical outcomes. Meta-regression was not performed, since
whereas its usage in the United States and Europe still lags behind [35]. less than 10 studies were included and data of the described variables
As a result, the majority of the included studies was derived within Asian was largely missing, significantly decreasing the potential reliability of
populations. Differences in underlying epidemiology (incidence, risk adjusted results [19]. Finally, study heterogeneity was moderate to
factors) of cardiovascular disease and patient demographics, along with substantial for all clinical outcomes. However, Mantel-Haenszel
differences in plaque and lesion phenotype between Asian and Western random-effects model was used to account for this and subgroup ana­
populations, should therefore be considered [36–38]. Intracoronary lyses were performed to explore possible sources of study heterogeneity.
imaging studies have shown differences in plaque morphology between
both ethnicities, with Western patients having higher lipid indexes, 5. Conclusions
higher plaque burden, more calcification and longer lesion lengths as
compared to Asian populations [39,40]. Moreover, cultural differences This is the first systematic review and meta-analysis comparing
in operator and patient preference with respect to either the frequency IVUS- versus angio-guided PCI in patients with AMI, showing a benefi­
of IVUS use and preference for PCI over surgery, might preclude the cial effect of IVUS-guided PCI on all-cause mortality, MACE and TVR.
generalizability of our findings. Results of ongoing dedicated prospective studies are needed to confirm
Dedicated ongoing prospective studies, including the SPECTRUM these findings.
study (NCT05007535), the iSTEMI trial (NCT04775914), the IMPROVE Supplementary data to this article can be found online at https://doi.
trial (NCT04221815) and the IVUS-CHIP trial (NCT04854070), will org/10.1016/j.ijcard.2022.01.021.
provide more insight in the potential beneficial effect of IVUS-guided
PCI in Western populations. Whereas the IMPROVE and IVUS-CHIP Data availability statement
trial will focus on IVUS guidance for complex high-risk indicated pro­
cedures (including high-risk lesions in patients with non-STEMI), the Data in this systematic review and meta-analysis was extracted from
SPECTRUM study and iSTEMI trial investigate the use of IVUS during the included studies (full text, tables, figures, and supplementary files).
primary PCI. The data extraction forms are available upon reasonable request.
In the present meta-analysis, IVUS-guided PCI was also associated
with improved clinical outcomes in STEMI subgroup analyses. Although Declaration of Competing Interest
the forest plot for all-cause mortality indicated a less beneficial effect in
STEMI patients, subgroup differences were not statistically significant. Joost Daemen received institutional grant / research support from
The SPECTRUM study, iSTEMI trial and a small RCT from China Abbott Vascular, ACIST Medical, Astra Zeneca, Boston Scientific, Med­
(NCT04929158) will specifically assess the impact of IVUS guidance in tronic, Microport, Pie Medical, and ReCor Medical. Nicolas Van Mie­
STEMI. Moreover, these studies will provide more insight in relevant ghem received institutional research grant support from Abbott
procedural characteristics such as procedure time and contrast use, Vascular, Abiomed, Boston Scientific, Daiichi-Sankyo, Edward Life­
which were underreported in the included studies of this meta-analysis. sciences, Medtronic, and PulseCath. The remaining authors report to
We hypothesize that if IVUS-guided PCI is performed by an experienced have no disclosures.
team with a contemporary IVUS system, procedure times will not be
significantly longer. Of note, IVUS guidance can reduce contrast use, Acknowledgements
although data in STEMI patients is lacking [41].
Finally, a comparison between IVUS guidance in the acute setting The authors would like to thank Sabrina Meertens-Gunput, medical
versus other invasive imaging techniques or coronary physiology, was librarian, for her help during the search strategy process and for per­
beyond the scope of the present study. A recent meta-regression analysis forming the systematic search. The authors would also like to thank
showed a trend towards lower rates of subsequent myocardial infarction Sanne Hoeks, epidemiologist, for providing specific methodological
with IVUS as compared to fractional flow reserve, in patients with acute knowledge on how to conduct a systematic review and meta-analysis.
coronary syndrome [42]. Dedicated studies are needed to further This research did not receive any specific grant from funding
address the respective value of invasive imaging versus physiological agencies in the public, commercial, or not-for-profit sectors.
tools in acute patients.
References
4.1. Limitations
[1] G.S. Mintz, Clinical utility of intravascular imaging and physiology in coronary
First, mainly retrospective and prospective observational studies artery disease, J. Am. Coll. Cardiol. 64 (2014) 207–222.
[2] L.J.C. van Zandvoort, K. Masdjedi, K. Witberg, J. Ligthart, M.N. Tovar Forero,
based on large AMI registries were included in this systematic review R. Diletti, et al., Explanation of postprocedural fractional flow reserve below 0.85,
and meta-analysis. In general, data derived from registries is more prone Circ. Cardiovasc. Interv. 12 (2019), e007030.
to bias and might affect results. Quality assessment was performed to [3] E. Shlofmitz, Z.A. Ali, A. Maehara, G.S. Mintz, R. Shlofmitz, A. Jeremias,
Intravascular imaging-guided percutaneous coronary intervention: a universal
provide a complete overview of each study’s risk of bias per domain. approach for optimization of stent implantation, Circ. Cardiovasc. Interv. 13
Moreover, risk of publication bias was assessed by visual inspection of (2020), e008686.
funnel plots. The funnel plot for MACE showed a slightly asymmetric [4] G.S. Mintz, S.E. Nissen, W.D. Anderson, S.R. Bailey, R. Erbel, P.J. Fitzgerald, et al.,
American College of Cardiology Clinical Expert Consensus Document on standards
pattern, indicating that publication bias in favor of IVUS could not be for acquisition, measurement and reporting of intravascular ultrasound studies
completely excluded. However, visual inspection of funnel plots has (IVUS). A report of the American College of Cardiology Task Force on Clinical
been found a subjective tool for the assessment of publication bias [43]. Expert Consensus Documents, J. Am. Coll. Cardiol. 37 (2001) 1478–1492.
[5] A. Maehara, M. Matsumura, Z.A. Ali, G.S. Mintz, G.W. Stone, IVUS-guided versus
The aforementioned dedicated prospective studies are needed to
OCT-guided coronary stent implantation: a critical appraisal, JACC Cardiovasc.
confirm the potential positive impact of IVUS-guided PCI in patients Imaging 10 (2017) 1487–1503.
with AMI. Second, follow-up time differed among studies (in-hospital up [6] L.J.C. van Zandvoort, K. Masdjedi, M.N. Tovar Forero, M.J. Lenzen, J. Ligthart,
to 5 years), although maximum follow-up in most included studies was R. Diletti, et al., Fractional flow reserve guided percutaneous coronary intervention
optimization directed by high-definition intravascular ultrasound versus standard
one year. Third, inherent to using a composite endpoint in a meta- of care: rationale and study design of the prospective randomized FFR-REACT trial,
analysis is that definitions differ among included studies. In the Am. Heart J. 213 (2019) 66–72.

41
F.T.W. Groenland et al. International Journal of Cardiology 353 (2022) 35–42

[7] J. Zhang, X. Gao, J. Kan, Z. Ge, L. Han, S. Lu, et al., Intravascular ultrasound versus stent implantation in patients with ST-segment elevation myocardial infarction,
angiography-guided drug-eluting stent implantation: the ULTIMATE trial, J. Am. Clin. Cardiol. 34 (2011) 706–713.
Coll. Cardiol. 72 (2018) 3126–3137. [27] M. Khalid, N.K. Patel, B. Amgai, A. Bakhit, M. Khalid, P. Kafle, et al., In-hospital
[8] X.F. Gao, Z. Ge, X.Q. Kong, J. Kan, L. Han, S. Lu, et al., 3-Year outcomes of the outcomes of angiography versus intravascular ultrasound-guided percutaneous
ULTIMATE trial comparing intravascular ultrasound versus angiography-guided coronary intervention in ST-elevation myocardial infarction patients, J. Commun.
drug-eluting stent implantation, JACC Cardiovasc. Interv. 14 (2021) 247–257. Hosp. Intern. Med. Perspect. 10 (2020) 436–442.
[9] I.Y. Elgendy, A.N. Mahmoud, A.Y. Elgendy, A.A. Bavry, Outcomes with [28] M. Megaly, A. Pershad, M. Glogoza, A. Elbadawi, M. Omer, M. Saad, et al., Use of
intravascular ultrasound-guided stent implantation: a meta-analysis of randomized intravascular imaging in patients with ST-segment elevation acute myocardial
trials in the era of drug-eluting stents, Circ. Cardiovasc. Interv. 9 (2016), e003700. infarction, Cardiovasc. Revasc. Med. 30 (2021) 59–64.
[10] S. Buccheri, G. Franchina, S. Romano, S. Puglisi, G. Venuti, P. D’Arrigo, et al., [29] S. Vallabhajosyula, S.C. El Hajj, M.R. Bell, A. Prasad, A. Lerman, C.S. Rihal, et al.,
Clinical outcomes following intravascular imaging-guided versus coronary Intravascular ultrasound, optical coherence tomography, and fractional flow
angiography-guided percutaneous coronary intervention with stent implantation: a reserve use in acute myocardial infarction, Catheter. Cardiovasc. Interv. 96 (2020)
systematic review and bayesian network meta-analysis of 31 studies and 17,882 E59–E66.
patients, JACC Cardiovasc. Interv. 10 (2017) 2488–2498. [30] T. Yamashita, K. Sakamoto, N. Tabata, M. Ishii, R. Sato, S. Nagamatsu, et al.,
[11] F. Darmoch, M.C. Alraies, Y. Al-Khadra, H. Moussa Pacha, D.S. Pinto, E.A. Osborn, Imaging-guided PCI for event suppression in Japanese acute coronary syndrome
Intravascular ultrasound imaging-guided versus coronary angiography-guided patients: community-based observational cohort registry, Cardiovasc. Interv. Ther.
percutaneous coronary intervention: a systematic review and meta-analysis, J. Am. 36 (2021) 81–90.
Heart Assoc. 9 (2020), e013678. [31] M. Endo, K. Hibi, T. Shimizu, N. Komura, I. Kusama, F. Otsuka, et al., Impact of
[12] G. Maluenda, G. Lemesle, I. Ben-Dor, S.D. Collins, A.I. Syed, R. Torguson, et al., ultrasound attenuation and plaque rupture as detected by intravascular ultrasound
Impact of intravascular ultrasound guidance in patients with acute myocardial on the incidence of no-reflow phenomenon after percutaneous coronary
infarction undergoing percutaneous coronary intervention, Catheter. Cardiovasc. intervention in ST-segment elevation myocardial infarction, JACC Cardiovasc.
Interv. 75 (2010) 86–92. Interv. 3 (2010) 540–549.
[13] H. Okura, Y. Saito, T. Soeda, K. Nakao, Y. Ozaki, K. Kimura, et al., Frequency and [32] T. Higuma, T. Soeda, N. Abe, M. Yamada, H. Yokoyama, S. Shibutani, et al.,
prognostic impact of intravascular imaging-guided urgent percutaneous coronary A combined optical coherence tomography and intravascular ultrasound study on
intervention in patients with acute myocardial infarction: results from J-MINUET, plaque rupture, plaque erosion, and calcified nodule in patients with ST-segment
Heart Vessel. 34 (2019) 564–571. elevation myocardial infarction incidence, morphologic characteristics, and
[14] D. Hausmann, R. Erbel, M.J. Alibelli-Chemarin, W. Boksch, E. Caracciolo, J. outcomes after percutaneous coronary intervention, JACC Cardiovasc. Interv. 8
M. Cohn, et al., The safety of intracoronary ultrasound. A multicenter survey of (2015) 1166–1176.
2207 examinations, Circulation 91 (1995) 623–630. [33] H. Ohashi, H. Ando, H. Takashima, K. Waseda, M. Shimoda, M. Fujimoto, et al.,
[15] M.L. Rethlefsen, S. Kirtley, S. Waffenschmidt, A.P. Ayala, D. Moher, M.J. Page, et Diagnostic performance of high-resolution intravascular ultrasound for the
al., PRISMA-S: an extension to the PRISMA statement for reporting literature detection of plaque rupture in patients with acute coronary syndrome, Circ. J. 83
searches in systematic reviews, Syst. Rev. 10 (2021) 39. (2019) 2505–2511.
[16] W.M. Bramer, G.B. de Jonge, M.L. Rethlefsen, F. Mast, J. Kleijnen, A systematic [34] K. Hibi, K. Kozuma, S. Sonoda, T. Endo, H. Tanaka, H. Kyono, et al., A randomized
approach to searching: an efficient and complete method to develop literature study of distal filter protection versus conventional treatment during percutaneous
searches, J. Med. Libr. Assoc. 106 (2018) 531–541. coronary intervention in patients with attenuated plaque identified by
[17] J.A. Sterne, M.A. Hernán, B.C. Reeves, J. Savović, N.D. Berkman, M. Viswanathan, intravascular ultrasound, JACC Cardiovasc. Interv. 11 (2018) 1545–1555.
et al., ROBINS-I: a tool for assessing risk of bias in non-randomised studies of [35] K.C. Koskinas, M. Nakamura, L. Räber, R. Colleran, K. Kadota, D. Capodanno, et al.,
interventions, BMJ 355 (2016), i4919. Current use of intracoronary imaging in interventional practice - results of a
[18] J.A.C. Sterne, J. Savović, M.J. Page, R.G. Elbers, N.S. Blencowe, I. Boutron, et al., European Association of Percutaneous Cardiovascular Interventions (EAPCI) and
RoB 2: a revised tool for assessing risk of bias in randomised trials, BMJ 366 Japanese Association of Cardiovascular Interventions and Therapeutics (CVIT)
(2019), l4898. Clinical Practice Survey, EuroIntervention 14 (2018) e475–e484.
[19] J.P.T. Higgins, J. Thomas, J. Chandler, M. Cumpston, T. Li, M.J. Page, et al., [36] H. Ueshima, A. Sekikawa, K. Miura, T.C. Turin, N. Takashima, Y. Kita, et al.,
Cochrane handbook for systematic reviews of interventions version 6.2 (updated Cardiovascular disease and risk factors in Asia: a selected review, Circulation 118
February 2021), Cochrane (2021). Available from, www.training.cochrane.org (2008) 2702–2709.
/handbook. [37] J. Hata, Y. Kiyohara, Epidemiology of stroke and coronary artery disease in Asia,
[20] K. Ahmed, M.H. Jeong, R. Chakraborty, Y. Ahn, D.S. Sim, K. Park, et al., Role of Circ. J. 77 (2013) 1923–1932.
intravascular ultrasound in patients with acute myocardial infarction undergoing [38] S. Kohsaka, T. Kimura, M. Goto, V.V. Lee, M. Elayda, Y. Furukawa, et al.,
percutaneous coronary intervention, Am. J. Cardiol. 108 (2011) 8–14. Difference in patient profiles and outcomes in Japanese versus American patients
[21] N. Kim, J.H. Lee, S.Y. Jang, M.H. Bae, D.H. Yang, H.S. Park, et al., Intravascular undergoing coronary revascularization (collaborative study by CREDO-Kyoto and
modality-guided versus angiography-guided percutaneous coronary intervention in the Texas Heart Institute Research Database), Am. J. Cardiol. 105 (2010)
acute myocardial infarction, Catheter. Cardiovasc. Interv. 95 (2020) 696–703. 1698–1704.
[22] K. Nakatsuma, H. Shiomi, T. Morimoto, K. Ando, K. Kadota, H. Watanabe, et al., [39] K.L. Bryniarski, E. Yamamoto, T. Sugiyama, L. Xing, H. Lee, I.K. Jang, Differences
Intravascular ultrasound guidance vs. angiographic guidance in primary in coronary plaque morphology between East Asian and Western White patients: an
percutaneous coronary intervention for st-segment elevation myocardial infarction optical coherence tomography study, Coron. Artery Dis. 29 (2018) 597–602.
- long-term clinical outcomes from the CREDO-Kyoto AMI registry, Circ. J. 80 [40] R.P. Rusinova, G.S. Mintz, S.Y. Choi, H. Araki, D. Hakim, E. Sanidas, et al.,
(2016) 477–484. Intravascular ultrasound comparison of left main coronary artery disease between
[23] H.X. Wang, P.S. Dong, Z.J. Li, H.L. Wang, K. Wang, X.Y. Liu, Application of white and Asian patients, Am. J. Cardiol. 111 (2013) 979–984.
intravascular ultrasound in the emergency diagnosis and treatment of patients with [41] J. Mariani Jr., C. Guedes, P. Soares, S. Zalc, C.M. Campos, A.C. Lopes, et al.,
st-segment elevation myocardial infarction, Echocardiography 32 (2015) Intravascular ultrasound guidance to minimize the use of iodine contrast in
1003–1008. percutaneous coronary intervention: the MOZART (Minimizing cOntrast
[24] B. Witzenbichler, A. Maehara, G. Weisz, F.J. Neumann, M.J. Rinaldi, D.C. Metzger, utiliZation With IVUS Guidance in coRonary angioplasTy) randomized controlled
et al., Relationship between intravascular ultrasound guidance and clinical trial, JACC Cardiovasc. Interv. 7 (2014) 1287–1293.
outcomes after drug-eluting stents: the assessment of dual antiplatelet therapy with [42] M. Iannaccone, M. Abdirashid, U. Annone, G. Saint-Hilary, P. Meier, A. Chieffo, et
drug-eluting stents (ADAPT-DES) study, Circulation 129 (2014) 463–470. al., Comparison between functional and intravascular imaging approaches guiding
[25] L. Ya’qoub, M. Gad, A.M. Saad, I.Y. Elgendy, A.N. Mahmoud, National trends of percutaneous coronary intervention: a network meta-analysis of randomized and
utilization and readmission rates with intravascular ultrasound use for ST- propensity matching studies, Catheter. Cardiovasc. Interv. 95 (2020) 1259–1266.
elevation myocardial infarction, Catheter. Cardiovasc. Interv. 98 (2021) 1–9. [43] N. Terrin, C.H. Schmid, J. Lau, In an empirical evaluation of the funnel plot,
[26] Y.J. Youn, J. Yoon, J.W. Lee, S.G. Ahn, M.S. Ahn, J.Y. Kim, et al., Intravascular researchers could not visually identify publication bias, J. Clin. Epidemiol. 58
ultrasound-guided primary percutaneous coronary intervention with drug-eluting (2005) 894–901.

42