Home > Discussion - Intravascular Diagnostic...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Raman G, Yu W, Ip S, et al. Intravascular Diagnostic Procedures and Imaging Techniques Versus Angiography Alone in Coronary Artery Stenting: Comparative Effectiveness Review [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2013 Feb. (Comparative Effectiveness Reviews, No. 104.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.


Key Findings and Strength of Evidence

Our review found that all the eligible studies addressed only two techniques, FFR and IVUS. Comparative data with respect to angiography alone were available on the use of FFR, which measures the physiological severity of coronary stenosis, to decide which coronary lesions require stenting (Key Question 1), and on the use of IVUS, which visualizes coronary anatomy, to optimize stent deployment (Key Question 2). There were insufficient data concerning the use of intravascular diagnostic techniques immediately after PCI to evaluate the success of stent placement as compared with angiography alone (Key Question 3), or for direct comparisons between intravascular diagnostic techniques (Key Question 4). Data were also available on the association (or lack thereof) between IVUS and factors such as left main disease, sex, diabetes mellitus status, and lesion length and reference diameter (Key Question 5).

Our review suggests that the use of FFR decide which coronary lesions require intervention would confer a lower risk of the combined endpoint of death and MI or of MACE in patients with intermediate coronary stenosis, as compared with stenting guided by angiography alone. Additionally, our review indicates that FFR-guided stenting would decrease costs of the procedure and would lead to fewer stents implanted, as compared with angiography alone. These findings may not hold for patients with more severe CAD. Specific ally, the evidence was derived from studies that focused on men with lower grade angina, and excluded patients with left main disease and acute MI. Therefore, the use of FFR in order to decide which lesions require stenting is most applicable in patients with stable multivessel disease and intermediate coronary stenosis, excluding left main disease and acute MI.

Based primarily on the FAME trial, we conclude that there is moderate evidence that the use of FFR during stenting confers a lower risk of the combined endpoint of death or MI or of MACE in patients with intermediate coronary lesions, excluding left main disease and acute MI.

Our review also indicates that the use of IVUS compared with angiography alone to guide PCI and stent deployment achieved some measureable, Our review also indicates that the use of IVUS compared with angiography alone to guide stent deployment achieved some measureable improvements in QCA outcomes, including MLD, percent diameter stenosis, and reference vessel diameter. However, the gains achieved in RCTs for intermediate outcomes with IVUS-guided stenting did not translate into significant differences in mortality or MI. Nevertheless, there were significant reductions in clinically-driven repeat revascularization and restenosis rates during medium-term (>30 days to 1 year) or long-term (>1 year) followup with IVUS-guided stenting versus angiography-guided stenting, with a reduction in repeat revascularization of about 30 percent (mostly observed in RCTs of modest sample size).

The lower repeat revascularization and restenosis rates reported with IVUS-guided stenting should be interpreted cautiously; as these studies were conducted using a previous generation of bare-metal stents, and their results may no longer be applicable to current clinical practice with a widespread use of drug-eluting stents.

In the reviewed studies, IVUS-guided stenting appears to be associated with longer procedural times, greater radiation exposure, and greater contrast use than angiography-guided stenting, all factors that may be associated with short- and long-term risks of complications. Discussion regarding the report and recommendations for future research follow.

Context of Findings

Our search identified three recently published systematic reviews comparing the effect of IVUS-guided PCI and non-IVUS-guided PCI. These reviews included a total of 21 nonoverlapping studies (9 trials in Casella 2003,15 15 in Berry 2000,60 and 7studies in Parise 201161), with followup durations that ranged from 5 months to 2.5 years. Both randomized and nonrandomized trials, as well as registries, were included in these reviews. The clinical endpoints evaluated were target lesion revascularization, target vessel revascularization, MACE, mortality, MI, CABG, and restenosis. Angiographic outcomes including restenosis rate, MLD, percent diameter stenosis, acute gain, late lumen loss, net gain, and resource utilization were also evaluated. The definition of MACE varied across the reviews.

All three reviews consistently reported a significant reduction in 6-month angiographic restenosis rate and target vessel revascularization with IVUS-guided PCI versus non–IVUS-guided PCI. Two of these reviews also found a significant decrease in MACE with the use of IVUS for guiding PCI over non–IVUS-guided PCI.15,61 No significant differences were observed between groups for the clinical outcomes of mortality or MI.

We reviewed 31 studies for comparisons of IVUS- and angiography-guided stent deployment, including two trials that were conducted in the era of drug-eluting stents. Our analyses revealed that only repeat revascularization was significantly lower in the IVUS-guided PCI group, as compared with the angiography-guided PCI group, during intermediate-term and long-term followup. Nonetheless, only six24,26,29,31-33 and four23,25,27,30 of the nine eligible RCTs had medium-term and long-term followup, respectively. Our review including recent literature did not find a significant decrease in MACE in the IVUS group compared with the angiography alone group. The disparity in our findings could be explained by the differences in eligibility criteria, in the number of included studies, or the methods of analyses. The first review searched until 1999 but found only two RCTs overlapping with our review, because of differences in eligibility criteria;15 the second review searched until 2001 and identified only five of the total nine RCTs included in our review;60 and the third review combined medium- and long-term data, found a statistical significant results for MACE.61

In this review, we examined both older studies (examining PCI with bare-metal stents) and more recent studies (examining PCI with drug-eluting stents). Our review also comprehensively evaluated nonrandomized comparative studies of intravascular diagnostic techniques. Our analyses evaluated both intermediate and clinical outcomes at various time points. Such extensive evaluations have not been carried out by prior reviews, which most often evaluated only the last reported time point. Also in contrast to prior reviews, we examined the impact of FFR in both RCTs and nonrandomized studies on short-, intermediate-, and long-term outcomes, and found consistent results. In addition, our review synthesized data and analyzed gaps in the literature on the use of intravascular diagnostic techniques at various stages of the stenting (before, during, and after), and evaluated the role of these techniques in therapeutic decisionmaking. In summary, our review comprehensively examined both IVUS and FFR data, and has identified a lack of comparative studies for all other emerging novel and hybrid techniques.


Reviewed studies were all conducted in tertiary care centers (with only one exception37), and were carried out mostly in Western Europe and North America. Studies included patients with various eligibility criteria for CAD undergoing PCI and stent placement at entry. Some studies included patients who had to be willing and be able to undergo followup angiography. The majority of the patients in these studies were men, and the reviewed studies specifically excluded individuals with left main disease or acute MI. Minorities were underrepresented, although a few studies reported baseline data by race or ethnicity. These eligibility criteria likely selected groups of patients with intermediate coronary stenosis, better functional status, and higher socioeconomic status (which is inversely associated with severity of CAD62), thus limiting applicability in patients with severe CAD. Thus, overall, there are several important groups of patients who have not been adequately represented in the available studies.

Two studies reported the effect of various patient or lesion characteristics on outcomes among those who had an IVUS-guided stent placement versus stenting guided by angiography alone. These included controls of age, sex, and left main disease. These subgroup analyses were limited by a lack of reporting for all subgroups, or statistical analyses for other intravascular diagnostic techniques. Thus, no overall conclusion could be drawn regarding the effect of patient characteristics on outcomes for FFR-guided stent placement versus stenting guided by angiography alone.

Drug-eluting stent deployment came into clinical use since 2000. Most IVUS trials (seven of nine RCTs) reviewed were performed before 2000. Interventional techniques and technology have evolved considerably since then, not only in terms of high-pressure balloon inflation, but also in stent design, composition, delivery systems, balloon technology, adjunctive pharmacotherapy, and other features. Current bare-metal stents are radically different than those used before 2000, and only two RCTs (both conducted in Eastern Europe) evaluated IVUS-guided stent placement in patients with a drug-eluting stent, and none evaluated second-generation drug-eluting stents or bioabsorbable stents. Thus, overall, there are several important groups of patients who have not been adequately represented in the available literature.

Clinical and Policy Decisionmaking Implications

There is moderate strength of evidence favoring FFR-guided PCI over angiography-guided PCI in patients with intermediate coronary lesions; these findings are supported by one large trial (FAME) and one nonrandomized study. Although the evidence was rated to be of moderate strength, there is the possibility that future studies will not support the favorable effect of FFR-guided stenting. The phenomenon of an initial effect eventually dissipating through subsequent studies has been well documented elsewhere.63Although some data exist for the role of FFR after intervention in side branches or after stent deployment, no randomized or direct comparative studies have evaluated FFR in these circumstances.64-66 It is also worth noting that the FAME trial included patients with intermediate stenosis and lower grades of angina. The intrinsic risk of a non-ischemic stenosis may be lower than the risk of stent implantation itself. Treating low-risk lesions could lead to additional invasive tests or treatments that could adversely impact long-term clinical outcomes. Therefore, the use of stents in treating low-risk lesions should be weighed against this consideration. These decisions are, of course, not always straightforward in clinical practice.

Currently, IVUS is extensively applied in certain clinical situations and specific lesion subsets (e.g., left main disease), without the backing of sufficient comparative data. Additionally IVUS is used to assess stent apposition and adequate stent expansion, lesion coverage, and edge dissections when the operator is in doubt and cannot angiographically determine with certainty whether a potentially life-threatening technical complication exists (i.e., one that could lead to stent thrombosis and potentially death). However, IVUS cannot fully assess the physiological significance of lesions (in deciding if a coronary lesion needs intervention), which depends not only on minimal lumen area, but also on numerous other factors including lesion length, reference vessel dimensions, and the amount of myocardium jeopardized by the lesion.

FFR and IVUS are often used as complementary modalities during an intervention to evaluate different aspects of coronary artery disease and decide its management. Therefore, head-to-head comparisons of these techniques may not be possible or meaningful. Our review did not find comparative data correlating findings of OCT, IVUS-virtual histology, and NIRS with subsequent outcomes and events, or on their relative impacts and resource utilization profiles. Further research is needed to evaluate the future use of hybrid and other novel intravascular diagnostic techniques. Intravascular diagnostic techniques are quickly evolving, and differences in their learning curves and the skill with which they are employed can potentially influence outcomes. Additional studies are necessary to determine the implications of these factors on clinical and policy decisionmaking.

CER Limitations

Intravascular diagnostic techniques are quickly evolving, which likely explains why we found few comparative studies except for two techniques, IVUS and FFR. There was insufficient evidence to answer two of the five review's Key Questions. Our review included only direct comparisons and only studies that had two distinct comparison groups (intravascular diagnostic technique and angiography vs. angiography alone). We excluded studies that lacked a distinct angiography-guided PCI group both at intervention and at followup. We also did not examine the impact of different thresholds for FFR, or the impact of either technology on treatment decisions besides stenting.

Other restrictions included the focus of Key Questions on the short time-frame around PCI, thereby excluding studies evaluating the intravascular diagnostic techniques during followup only (but not during PCI). The reporting of timing of intravascular diagnostic technique application in reviewed studies was often unclear (e.g., during PCI or immediately after).

Evidence Base Limitations

Outcome reporting (primarily with respect to patient-centered outcomes) was not complete in the included studies. There was also substantial heterogeneity in definitions of the composite outcome MACE. Less than one-quarter of the included populations were women, and studies often did not evaluate the use of intravascular diagnostic techniques in patients with acute MI and left main disease. Most of the IVUS studies enrolled and followed patients before 2000. None of the studies included in our review was itself sufficiently powered to address the effectiveness of IVUS to improve long-term outcomes, and few studies reported long-term outcome data. We were not able to conduct meaningful subgroup analyses stratifying older versus newer studies (studies conducted before 2000 vs. those conducted since 2000), owing to the small number of IVUS RCTs conducted since 2000.

Few studies evaluated the comparative effectiveness of these intravascular diagnostic techniques in patients undergoing drug-eluting stent implantation, specifically the latest generation of stents. And studies often did not evaluate the effect of training of operators and their variability in the application of these techniques on clinical outcomes. Studies did not report on the effect of evolution intravascular diagnostic technique during study period.

Ongoing Research

A search in the ClinicalTrials.gov registry yielded one active (recent, ongoing), one completed, and one recently terminated trial examining intravascular diagnostic techniques that are potentially relevant to the Key Questions in our report. None of the entries provided results. One RCT evaluated the effect of FFR-guided PCI. The remaining two RCTs compared IVUS-guided PCI with angiography-guided PCI.

The first RCT (DEFER-DES), conducted in South Korea, compared FFR-guided stenting with stent placement guided by angiography alone for the treatment of intermediate coronary lesions using drug-eluting stents, and has since been terminated owing to the slow enrollment (ClinicalTrials.gov number NCT00592228).

The second RCT (FAVOR) is an ongoing trial conducted in South Korea comparing the effectiveness of FFR-guided PCI with IVUS-guided PCI for the treatment of intermediate coronary lesions. The primary outcome of this trial is MACE; secondary outcomes are the individual components of MACE. Patients will be followed clinically for up to 2 years. This trial is expected to enroll 1,400 patients and will be completed by January 2014 (ClinicalTrials.gov number NCT01175863).

The third RCT (AVIO) is a completed study from Italy, comparing IVUS versus angiography alone in the optimization of drug-eluting stents (NCT00936169).

Evidence Gaps

Table 2 summarizes the evidence gaps with regards to the five Key Questions of this systematic review.

Table 2. Evidence gaps.

Table 2

Evidence gaps.

Future Research Needs

This review has identified substantial gaps in the intravascular diagnostic technique literature. Chief among them are the contemporary role of IVUS guidance in the placement of drug-eluting stents; the prognostic role of FFR, which should be confirmed in further trials; and evaluation of hybrid and novel techniques for comparative efficacy and safety. While early studies evaluating drug-eluting stents have used IVUS during stent placement, comparative studies, particularly RCTs of drug eluting stent placement guided by IVUS or angiography alone, are lacking. The potential advantage of IVUS guidance in drug-eluting stent or bioabsorbable stent placement requires further evaluation. IVUS continues to be used in stenting small vessels, complex lesions, and long lesions. It is important, then, that additional RCTs in these populations are conducted to assess the comparative effectiveness of IVUS in the drug-eluting stent era.

FFR and IVUS could be used for other decisions beyond guiding and optimizing stent deployment—for example, FFR could be used in other revascularization options (e.g., CABG), or to identify patients with stable CAD who may benefit from stenting (e.g., patients in the FAME II trial) but otherwise would be treated with current best medical therapy only.12 The role of FFR in high-risk patients with bifurcation lesions, left main coronary artery stenosis, ostial stenosis, acute coronary syndrome, or in side branches and other clinical situations, should be studied in future trials. The roles of FFR and IVUS in other vascular territories, outside of the coronary circulation, should also be better defined in future trials. Data correlating findings of high-resolution imaging techniques of OCT, IVUS-virtual histology, and NIRS with subsequent outcomes and events are not yet available. Although OCT is a very useful technology, particularly in stent research, its clinical role remains to be determined and will depend upon data demonstrating that OCT improves patient care and outcomes. The same applies for NIRS. Although the PROSPECT trial suggests that the addition of radiofrequency backscatter analysis to grayscale IVUS (IVUS-virtual histology) might provide incremental information in predicting the site of future coronary events, further studies are warranted to investigate this hypothesis, and at present, PCI of nonsignificant lesions on the basis of plaque composition alone is not justified.67,68 Catheters are currently deployed in combination with multiple imaging modalities (FFR, OCT, IVUS, or others) for more comprehensive assessment, with an aim towards improving the effectiveness and efficiency of interventions. But these hybrid systems could also potentially add to the time, risk, and resource utilization of catheterization procedures.

At present, the lack of available comparative data for hybrid and novel devices (as opposed to individual devices) limits evaluations of their effectiveness in routine clinical practice. Additionally, up and coming techniques require further evaluation, such as virtual FFR, which can quantify the FFR for each lesion from the data taken noninvasively with computer analysis of coronary computed tomography angiograms or magnetic resonance angiograms.

Future research is also needed to enrich our understanding of the comparative effectiveness of angiography and intravascular diagnostic techniques (both older and novel) in diverse populations (including by race/ethnicity and socioeconomic status), in women, and in patients with left main disease and acute MI, as published studies often excluded or recruited a small proportion of these populations while evaluating established techniques such as FFR. Furthermore, more studies with followup duration greater than 1 year are needed to enhance our understanding of the long-term impact of the use of intravascular diagnostic techniques as compared with angiography.

Investigators should attempt to achieve consensus in harmonizing outcomes assessment. Studies either reported data by patients or by lesions, thereby complicating synthesis across studies. Future research is also needed to assess the usefulness of how these procedural data are presented, for example, if data by patients are preferable over data by lesions. Until consensus is achieved, investigators should be encouraged to present data both by patients and by lesions.


There is a moderate strength of evidence that that the use of FFR to decide whether intermediate coronary lesions require stenting confers a lower risk of death and MI, decreases procedural costs, and leads to fewer stents implanted, as compared with stenting decisions based on angiography alone. However, these findings are based on a single RCT (the FAME trial 21); further trials are needed to confirm and expand upon these results. There is a moderate strength of evidence that the use of IVUS to guide stent optimization reduces repeat revascularization and restenosis but does not affect mortality or MI rates, as compared with angiography alone. However, most of the IVUS trials were performed before 2000. There are only two RCTs evaluating IVUS-guided drug-eluting stent placement, and none with second generation drug-eluting stents or bioabsorbable stents. These factors affect the present-day applicability of the existing data. Furthermore, the majority of the eligible studies focused on men with lower grade disease, and excluded patients with left main disease. Future studies (regardless of technology or the current intervention of interest) should include a more representative proportion of women and patients with more serious coronary artery diseases. Future work will also need to evaluate longer-term (on the order of years) patient outcomes to better appreciate the true impact of these techniques.

Cover of Intravascular Diagnostic Procedures and Imaging Techniques Versus Angiography Alone in Coronary Artery Stenting: Comparative Effectiveness Review
Intravascular Diagnostic Procedures and Imaging Techniques Versus Angiography Alone in Coronary Artery Stenting: Comparative Effectiveness Review [Internet].
Comparative Effectiveness Reviews, No. 104.
Raman G, Yu W, Ip S, et al.


AHRQ (US Agency for Healthcare Research and Quality)

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...