Coronary angiography (CAG) is the traditional and most widely used invasive imaging modality for guiding percutaneous coronary intervention (PCI). However, the two-dimensional projection of CAG cannot depict the structure of complex 3-dimensional arterial vessel wall, and thus evaluate the vessel dimensions and plaque characteristics, nor directly assess the result of stent implantation. Instead, intracoronary imaging through intravascular ultrasound (IVUS) and optical coherence tomography (OCT) can provide valuable incremental information that can be used clinically to optimize the stent implantation and minimize the stent-related problems (1–3). Fractional flow reserve (FFR) is a lesion-specific physiological index to evaluate the functional significance of coronary stenosis, and its benefit in guiding PCI has been proven by many clinical studies (4–6).

Although numerous meta-analyses and randomized trials have been published to compare the clinical outcomes between CAG and IVUS (1, 7, 8), CAG and OCT (9, 10), CAG and FFR (11, 12), yet just a few network meta-analyses are designed to compare the effects of all available modalities [CAG, IVUS, OCT/optical frequency domain imaging (OFDI), and FFR] for the guidance of PCI within a single analytical framework (13, 14). Moreover, randomized trials performed in the era of bare-metal stents (BMS) were also included in the aforementioned network meta-analyses, which may not be applicable to current clinical practice where drug-eluting stents (DES) have been widely used (15, 16). Pharmacological therapeutics have undergone great changes from BMS to DES era, especially the development of proprotein convertase subtilisin-kexin type 9 (PCSK-9) inhibitors, which can reduce levels of low-density lipoprotein cholesterol (LDL-C) by 50–70% when added to statins (17). Additionally, the previous network meta-analysis may be influenced by including observational studies. As more randomized trials and modalities have become available on PCI guidance, an updated comprehensive network meta-analysis of randomized trials is needed to evaluate the clinical outcomes associated with hemodynamic parameter (FFR or FFR related) or intravascular imaging (IVUS, OCT, or OFDI related)-guided PCI compared with CAG-guided PCI in the era of DES.

This network meta-analysis was conducted according to the PRISMA (preferred reporting items for systematic reviews and meta-analyses) network meta-analysis extension statement (18). The summary data were obtained from the published randomized trials with approval from the respective institutional review committees. Therefore, no further sanction was required for our network meta-analysis. This meta-analysis has been registered at the PROSPERO international prospective register of systematic reviews (CRD42021291442).

We conducted a systematic search of the literature on October 12, 2021. The databases included Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, and Web of Science. We also searched TCTMD, ClinicalTrials.gov, and major congress proceedings to identify potential studies. The medical subject headings or keywords included the following: coronary angiography, CAG; intravascular ultrasound, IVUS; optical coherence tomography, OCT; optical frequency domain imaging, OFDI; fractional flow reserve, FFR; instantaneous wave-free ratio, iFR; quantitative flow ratio, QFR; percutaneous coronary intervention, PCI; randomized controlled trial, RCT. The research syntax has been provided in Supplementary Table 1. Moreover, relevant randomized trials from reference lists of identified systematic reviews, meta-analyses, and relevant reviews were additionally hand searched to supplement the search of the electronic databases.

We included all randomized trials that compared any combination of the four category modalities: hemodynamic parameter-guided PCI [FFR, instantaneous wave-free ratio (iFR), quantitative flow ratio (QFR)], IVUS-guided PCI, OCT/OFDI-guided PCI, and CAG-guided PCI with DES implantation. Randomized trials without reporting our interested clinical outcomes were excluded. When multiple publications from the same randomized trial existed, we included the publication with the longest follow-up duration. The selection and data extraction processes were carried out in duplicate by two independent reviewers (HU MJ and GAO XJ), and any disagreement was resolved by consensus with a third-party reviewer (YANG JG).

The risk of bias for all included studies was assessed using the Cochrane risk of bias assessment tool (19). Publication bias was investigated with comparison adjusted funnel plots.

Our primary outcomes were major adverse cardiovascular events (MACE), cardiovascular death, myocardial infarction (MI), and target vessel/lesion revascularization (TVR/TLR) as reported by the trial authors. Secondary outcomes included all-cause death, stroke, stent thrombosis, and any revascularization. The definition of clinical outcomes was prescribed according to each randomized trial and can be found in Supplementary Table 2.

The Bayesian network meta-analysis was performed with a random effects model. Outcomes were reported as odds ratio (OR) with 95% credible interval (CrI) for all outcomes of interest. Four Markov chains were run simultaneously with 100,000 simulated draws after a burn-in of 10,000 iterations. We evaluated consistency with a node-splitting technique that compared the direct and indirect estimates for each comparison. The surface under the cumulative ranking curve (SUCRA) metric was used to compare the hierarchy of clinical outcomes of different PCI guidance modalities. SUCRA values vary between 0 and 100%, the higher the value, the higher the likelihood that a modality is in the top rank or highly effective (20). All analyses were conducted using R software (version 3.4.3) equipped with the “gemtc” package.

In our network meta-analysis, which included 28 randomized trials and 11,860 patients, we analyzed the clinical outcomes of four category PCI guidance modalities (CAG, IVUS, OCT/OFDI, and FFR/QFR) in the era of DES, and the findings can be summarized as follows. Firstly, IVUS led to lower risks of MACE than CAG, which was mainly due to lower risks of cardiovascular death and TVR/TLR. A trend toward decreased risk of stent thrombosis was also observed with IVUS. Secondly, hemodynamic parameter (FFR/QFR)-guided PCI could significantly reduce stroke compared with CAG, IVUS, and OCT/OFDI. Thirdly, MI, all-cause death, stent thrombosis, and any revascularization presented similar risks for the four category PCI guidance modalities.

Similar with our findings, a meta-analysis including seven trials with 3192 patients in the era of DES also revealed that IVUS-guided PCI was associated with a reduction in the risk of MACE (6.5 versus 10.3%; OR: 0.60; 95% CI: 0.46–0.77), which was mainly because of reduction in the risk of TLR (4.1 versus 6.6%; OR: 0.60; 95% CI: 0.43–0.84). The risk of stent thrombosis (0.6 versus 1.3%; OR: 0.49; 95% CI: 0.24–0.99) also appeared to be lower in the IVUS-guided group, and there was a trend toward lower risk of cardiovascular mortality (0.5 versus 1.2%; OR: 0.46; 95% CI: 0.21–1.00). After all, IVUS allows for easier visualization of the entire vessel structure, particularly when extensive circumferential calcification or attenuated plaques are not encountered (14). Moreover, in the Assessment of Dual Antiplatelet Therapy with Drug-Eluting Stents (ADAPT-DES) study which enrolled an all-comers population, IVUS-guided PCI was also associated with lower rates of stent thrombosis, MI, and TVR/TLR compared with CAG-guided PCI. Compared with CAG-guided PCI, a larger stent or balloon and/or higher inflation pressures (that would minimize under-expansion) were used in approximately 60% of IVUS-guided procedures, and additional stents (that would mitigate inflow/out-flow issues) were used in about 20% of patients. These strategies are most likely responsible for the lower rates of stent thrombosis and TVR/TLR observed in the IVUS-guided cohort in the present study (40). The aforementioned studies also confirmed that IVUS-guided PCI was superior to CAG-guided PCI not only in selected patients from randomized trials but also in all-comers from real-world scenarios.

An expert consensus document of the European Association of Percutaneous Cardiovascular Interventions claimed that IVUS and OCT are equivalent (and superior to CAG) in guiding and optimizing most PCI procedures (41). In our meta-analysis, it is indeed that no significant differences were observed between IVUS and OCT/OFDI. However, IVUS could decrease the risks of MACE, cardiovascular death, and TVR/TLR compared with CAG, where OCT/OFDI could not. Due to lower tissue penetration, especially in lipid-rich tissue, OCT is limited in assessing plaque burden and detecting vessel size in the presence of diffuse disease, whereas IVUS is an approach used to guide stenting sizing. Moreover, OCT is frequently unable to visualize the ostium as proper blood clearance is probably a challenge. Also, blood clearance needed for image acquisition in OCT increases the radio-contrast burden, which is particularly detrimental in patients with renal disease, whereas IVUS can minimize the use of iodine contrast in PCI procedure (24, 42). All of the aforementioned characteristics may contribute to the positive prognosis associated with IVUS and negative prognosis associated with OCT/OFDI. Meanwhile, it is noteworthy that compared with IVUS with numerous randomized trials comparing IVUS-guided versus CAG-guided PCI, there is limited research evidence on OCT-guided versus CAG-guided PCI with respect to clinical outcomes and no RCT is powered for clinical outcomes. Therefore, the lack of significant difference between OCT-guided versus CAG-guided PCI may be a result of limited number of patients and underpowered for the outcomes of interest. In addition, IVUS has been used clinically for almost three decades and extensive clinical experience has been gained, which may translate into positive prognosis.

In our meta-analysis, decreased risk of stroke associated with hemodynamic parameter (FFR/QFR)-guided PCI was also observed. As revealed in the British Heart Foundation FAMOUS–NSTEMI randomized trial, the proportion of patients treated initially by medical therapy was higher in the FFR-guided group than in the CAG-guided group [40 (22.7%) versus 23 (13.2%), difference 95% CI: 1.4–17.7%, p = 0.022], whereas the aggressive coronary revascularization was higher in the CAG-guided PCI (86.8 versus 77.3%). In a propensity score matching study including a total of 1,299 patients with left ventricular ejection fraction ≤ 50% (433 FFR-guided PCI, 866 CAG-guided PCI), FFR-guided PCI was associated with a lower risk of stroke compared with CAG-guided PCI (0 versus 2%; HR: 0.84; 95% CI: 0.62–0.96) during 1-year follow-up, whereas the differences disappeared after 5-years follow-up (3 versus 3%; HR: 0.68; 95% CI: 0.37– 1.65) (43). Therefore, it seems that CAG-associated stroke was mainly confined during peri-procedural and short-term follow-up, possibly due to more aggressive treatments in the CAG-guided group. However, we have to admit the fact that in our meta-analysis, just six trials (4,214 patients) in total reported 17 (0.40%) stroke events, which was too small in scale and maybe the reason for wide CrI. Considering the limited number of randomized trials focusing on the issue of stroke, further randomized trials are warranted to validate the rationality of different modalities in guiding PCI in the era of DES. The currently enrolling ILUMIEN III trial (NCT03507777) will randomize between 2,490 and 3,656 patients with high-risk clinical characteristics (diabetes) and/or complex angiographic lesions to compare the clinical outcomes between OCT-guided versus CAG-guided PCI. The principal results are expected to be published in 2022, which will provide significant evidence on the role of OCT in the guidance of PCI (44).

Despite the better prognosis associated with IVUS, yet CAG is still the mostly used modality in clinical practice. Moreover, the drawbacks associated with DES should be acknowledged. For example, although the healing response was similar and neoatherosclerosis was low in patients receiving durable- or biodegradable-polymer (45), histopathology, and intravascular imaging have detected neoatherosclerosis earlier and more frequently with DES compared with BMS (45). However, with the advancement in medical management (46) and techniques (47), it is promising that the prognosis associated with cardiovascular disease will improve greatly.

The present study should be interpreted with caution in light of some limitations. First, this is a study-level meta-analysis providing average treatment effects. The absence of patient-level data prevents us from assessing the effect of baseline clinical characteristics in PCI guidance modalities which might affect clinical outcomes. Second, subgroup analysis based on stable or acute coronary symptom is impossible because both stable and acute coronary symptom patients were included in the same trial. However, the ADAPT-DES study revealed that IVUS-guided PCI was superior to CAG-guided PCI in both stable and acute coronary symptom patients (40). Third, just six trials (4,214 patients) in total reported 17 (0.40%) stroke events, which was too small in scale and maybe the reason for wide CrI. Therefore, more randomized trials are warranted to validate the rationality of different modalities in guiding PCI in the era of DES. Fourth, IVUS has been used clinically for almost three decades and extensive clinical experience has been gained. However, the same scenario was not obtained for other PCI guidance modalities (OCT, OFDI, FFR, and QFR). Considering the fact that a long learning curve is required to commend a new PCI guidance modality, therefore, unfamiliar with the newly developed PCI guidance modality may negatively affect prognosis.

Our comprehensive network meta-analysis provides evidence that IVUS-guided PCI resulted in less MACE, cardiovascular death, and TVR/TLR. FFR/QFR-guided PCI resulted in decreased risk of stroke in the DES era. Further studies are still required to validate the rationality of different modalities in guiding PCI in the era of DES.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

M-JH: conceptualization, investigation, methodology, resources, validation, visualization, writing – original draft, and writing – review and editing. J-ST: methodology, resources, validation, visualization, and writing – original draft. LY, Y-YZ, J-GY, and X-JG: data curation, formal analysis, methodology, software, and writing – original draft. Y-JY: formal analysis, funding acquisition, investigation, methodology, project administration, resources, supervision, validation, visualization, and writing – review and editing. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.