Prognostic impact of high-risk plaque morphology and impaired physiology in untreated non-culprit coronary lesions: a systematic review and meta-analysis

Graphical abstract

Highlights

• •High-risk plaque morphology predicts adverse outcomes in untreated non-culprit lesions.
• •Impaired physiology and vulnerable plaque show a comparable impact on TVF risk.
• •Integrated coronary morpho-functional assessment improves TVF risk stratification.

Abstract

Background

The management of angiographically intermediate coronary lesions (AICLs) remains challenging. Although current guidelines recommend physiological assessment to guide treatment decisions, residual atherothrombotic risk related to vulnerable plaque morphology may persist in deferred non-culprit lesions (NCLs). We performed a systematic review and meta-analysis to evaluate the prognostic impact of high-risk plaque morphology and impaired physiology in untreated NCLs.

Methods

A systematic search of PubMed, Embase, and Web of Science identified 10 studies that included both morphological (optical coherence tomography [OCT], intravascular ultrasound [IVUS], or near infrared spectroscopy) and physiological (fractional flow reserve [FFR], instantaneous wave-free ratio, quantitative flow ratio [QFR], Murray fractal law-based QFR, optical flow reserve, OCT-FFR, or IVUS-FFR) assessment in the same untreated AICLs. Overall, 4757 patients with 7662 untreated NCLs were stratified into four groups: reference, morphology-positive, functional-positive, and morpho-functional-positive. The primary endpoint was target vessel failure (TVF), defined as a composite of cardiac death, target-vessel myocardial infarction, and target-vessel revascularization. Hazard ratios (HRs) were pooled using random-effects models.

Results

Compared with the reference group, TVF risk was increased in morphology-positive lesions (HR 2.94, 95% confidence interval [CI] 2.22–3.90) and functional-positive lesions (HR 3.59, 95% CI 1.56–8.24). The highest risk was observed in the morpho-functional-positive group. Indirect comparison showed no significant difference between morphology-positive and functional-positive lesions.

Conclusions

High-risk plaque morphology and impaired physiology were independently associated with adverse outcomes in untreated NCLs, while their coexistence identified lesions at highest risk. An integrated morpho-functional assessment may improve risk stratification in multivessel coronary artery disease.

1. Introduction

Angiographically intermediate coronary lesions (AICLs) cover a wide range of stenoses diameters, typically ranging from 40% to 90% by visual assessment.[1] Given the discrepancies between the angiographic and hemodynamic relevance of moderate coronary lesions, international guidelines recommend a functional severity assessment of angiographically intermediate stenoses to guide the clinical decision to revascularize or defer. [2], [3] The limitation of the ischemia-guided approach lies in its dichotomous nature, which ignores the identification of vulnerable plaques prone to acute destabilization despite non-flow-limiting stenoses. [4] Coronary computed tomography angiography analyses demonstrated that coronary atheroma burden provides better prognostic value than ischemic coronary stenoses, and that “high-risk plaque” features predict future coronary events in deferred vessels with non-significant functional evaluation, thus supporting the role of quantitative and qualitative morphological assessment for cardiovascular risk stratification in untreated lesions. [5], [6] To date, thin-cap fibroatheroma (TCFA), lipid-rich plaque (LRP), and plaque burden (PB) are considered major criteria for the definition of vulnerable plaque, but their role in guiding the treatment strategy is limited by their low positive predictive value. [7], [8] The validation of angio-derived (e.g., quantitative flow ratio [QFR]) or imaging-derived (e.g., optical flow ratio [OFR], optical coherence tomography-derived fractional flow reserve [OCT-FFR], and intravascular ultrasound-derived FFR [IVUS-FFR]) functional indexes has enabled the integration of physiology during intravascular imaging (IVI), providing a better understanding of the dynamic nature of atherosclerotic coronary artery disease. In this background, we performed a meta-analysis to compare the long-term prognostic impact of morphological high-risk plaques and impaired physiology in untreated non-culprit coronary vessels and to evaluate their integrated effect in target vessel failure (TVF) prediction.

2. Methods

This systematic review and meta-analysis were conducted to compare the prognostic role of morphological and functional assessment of untreated non-culprit coronary plaques in patients with acute or chronic coronary syndromes undergoing intracoronary evaluation. The protocol was registered in the PROSPERO database (registration number: CRD420251128824).

We systematically searched PubMed, Embase, and Web of Science for keywords related to non-culprit lesions (NCLs), OCT, IVUS, FFR, QFR, high-risk vulnerable plaque from inception to 26 July 2025. The full search strategy is available in Sup. Table S1. We included observational studies that reported: (I) both morphological (e.g., OCT or IVUS or near infrared spectroscopy [NIRS]) and physiological (e.g., FFR or instantaneous wave-free ratio [iFR] or QFR or Murray fractal law-based QFR [μQFR] or OFR or OCT-FFR or IVUS-FFR) assessment on the same angiographically intermediate untreated coronary plaque in patients with acute myocardial infarction, unstable or stable angina pectoris; (II) a reference group with both morphological and functional negative assessment; (III) outcome data as hazard ratios (HRs) with 95% confidence intervals (CI). We applied no restrictions on study language, follow-up duration or publication date. Two authors (A.C. and V.A.) independently screened titles and abstracts, assessed full texts for inclusion, and extracted data on study design, patients’ baseline characteristics, imaging and physiology technique used, length of follow-up, and outcomes with relative HR. Discrepancies were resolved by a third reviewer (P.C.). For baseline characteristics, COMBINE (OCT-FFR) [10] (Optical Coherence Tomography Morphologic and Fractional Flow Reserve Assessment in Diabetes Mellitus Patients) and PECTUS-obs [10] (Identification of Risk Factors for Acute Coronary Events by OCT After STEMI and NSTEMI Patients With Residual Non-Flow Limiting Lesions) were reported as a pooled dataset (Sup. Table S2), since the original publication provided only aggregated information from the individual patient-data analysis. Conversely, for study-level morphological and functional characteristics and outcome analyses, data were extracted and analyzed separately for each study, as reported in their original manuscripts (Sup. Table S3).

From HR and CI, we calculated log(HR), standard error (SE) and logSE comparing four predefined groups: (I) reference group (NCLs with both morphological and physiological negative assessment); (II) morphology-positive group (NCLs with high-risk plaque morphology and physiology negative assessment); (III) functional-positive group (NCLs with impaired physiological assessment without high-risk plaque morphology); morpho-functional-positive group (NCLs with both high-risk plaque morphology and physiology positive assessment). The outcome was the HR for TVF across each group compared with the reference group. TVF was commonly defined as a composite of cardiac death, target-vessel myocardial infarction (TV-MI) and target-vessel revascularisation (TVR).

In line with the study objective, heterogeneity in physiological and morphological criteria was intentionally accepted, as all thresholds were previously validated and reflected different expressions of functional significance or plaque vulnerability; accordingly, the aim was to compare physiology- versus morphology-based risk stratification rather than the performance of individual cut-off values. A summary of the study-specific thresholds used to define high-risk plaque morphology and impaired physiology is provided in Sup. Table S3.

For quality assessment, we used the Newcastle-Ottawa Scale for observational studies (Sup.Table S4). Meta-analyses were conducted using a random-effects model (REML estimator) with log(HR) as the effect size and SE as the precision. Heterogeneity was assessed with the I² statistic and τ². Subgroup analyses were performed stratifying studies by imaging technique (OCT vs IVUS). For the indirect comparison between the morphology-positive and functional-positive groups, we used the HRs reported within each study versus its own reference category, calculated the difference between their log(HR) values, and pooled these within-study log(HR) contrasts using a random-effects model. All analyses were performed using R (CRAN® 3.3.4) with the meta package.

3. Results

After full-text review, we identified ten observational studies (see PRISMA diagram [9], Supplementary Fig. S2), including five prospective studies (CLIMA [13] [Relationship Between Coronary Plaque Morphology of Left Anterior Descending Artery and Long Term Clinical Outcome], COMBINE OCT-FFR [10], PECTUS-obs [10], PROSPECT [11] [Providing Regional Observations to Study Predictors of Events in the Coronary Tree], PROSPECT II [12] [Providing Regional Observations to Study Predictors of Events in the Coronary Tree II]) and five retrospective studies (Osumi et al. [14], Xu et al. [15], Hong et al. [16], Kazikazi et al. [17], Cho et al. [18]). A total of 4757 patients and 7662 untreated NCLs with available data on both morphological and physiological plaque evaluation were included in the analysis (Sup. Tables S2–3).

The mean patient age ranged from 57.2 to 69 years across the studies. The rate of females and diabetic patients was 23.02 – 29.3% and 17.5 – 44.4%, respectively. The follow-up duration varied between 1 and 5 years. Eight studies included only acute coronary syndrome patients, while in the other two studies, the proportion of patients presenting with chronic coronary syndrome ranged from 46.6 to 64.1% (Table 1). The type of functional assessment used was FFR in four studies, OFR in two studies, while QFR, μQFR, IVUS-FFR, OCT-FFR, and iFR were each used in one study. The IVI device used was OCT in seven studies, IVUS in two studies, and IVUS-NIRS in one study.

Table 1.

Key features of the included studies.

Study (year)	Sample size		Untreated NCLs definition	Clinical presentation (%)		Invasive assessment technique		Follow-up (years)
	Patients	NCLs		ACS	CCS	Physio	Morpho
CLIMA (2025)[13]	983	983	untreated coronary plaque in the proximal LAD	53.4	46.6	OFR	OCT (TCFA)	1
Osumi et al. (2024)[14]	298	298	most stenotic untreated segment in the culprit vessel	100		QFR	OCT (TCFA)	2.7
Xu et al. (2025)[15]	580	1275	untreated coronary segments with luminal narrowing in non-IRA	100		μQFR	OCT (TCFA)	4.1
Hong et al. (2022)[16]	604	604	untreated non-culprit vessel	100		OFR	OCT (TCFA)	2
Kazikazi et al. (2022)[17]	278	278	untreated coronary segment	100		OCT-FFR	OCT (TCFA)	3
COMBINE (OCT-FFR) (2025)[10]	390	445	untreated non culprit plaque lesion	100		FFR	OCT (TCFA, LRP, PR/TR)	2
PECTUS-obs (2025)[10]	420	494	untreated non culprit plaque lesion	100		FFR	OCT (TCFA, LRP, PR/TR)	2
Cho et al. (2020)[18]	459	552	untreated intermediate lesion	35.9	64.1	FFR	IVUS (PB ≥70%)	5
PROSPECT (2022)[11]	660	2626	untreated non culprit lesion	100		IVUS-FFR	IVUS (PB ≤70%)	3.4
PROSPECT II (2025)[12]	162	184	untreated non culprit lesion	100		FFR/iFR	IVUS+NIR (PB≥70%+MaxLCBI4mm2≥324.7)	4

NCLs = non-culprit lesions; LAD = left anterior descending artery; ACS = acute coronary syndromes; CCS = chronic coronary syndromes; Physio = physiology; Morpho = morphology; OFR = optical flow ratio; OCT = optical coherence tomography; TCFA = thin-cap fibroatheroma; QFR = quantitative flow ratio; IRA = infarct related artery; μQFR = Murray fractal law-based QFR; OCT-FFR= OCT–derived fractional flow reserve; LRP = lipid-rich plaque; PR = plaque rupture; TR = thrombus; IVUS = intravascular ultrasound; PB = plaque burden; IVUS-FFR = IVUS–derived fractional flow reserve; iFR = instantaneous wave-free ratio; NIRS = near-infrared spectroscopy; MaxLCBImm² = maximum lipid-core burden index.

The sample size differed across the predefined groups and included eight studies in the morphology-positive group, five studies in the functional-positive group, and four studies in the morpho-functional-positive group (Sup. Table S3). The morphology-positive group was composed by six studies with OCT assessment, [10], [13], [14], [15], [17] one study with IVUS analysis [18] and one study with combined IVUS-NIRS evaluation. [12] TCFA was the high-risk plaque feature of interest in the four OCT studies, [13], [14], [15], [17] while PB ≥ 70% was the high-risk plaque feature of interest in the IVUS study. [18] In COMBINE (OCT-FFR) and PECTUS-obs studies, the morphology-positive group was defined by the presence of at least two OCT-defined high-risk plaque criteria between TCFA, lipid arc ≥ 90°, and the presence of either plaque rupture (PR) or thrombus; [10] while the PROSPECT II identified the morphology-positive group by using a combined IVUS-NIRS criteria, namely PB ≥ 70% and Max LCBI 4 mm2 ≥ 324,7. [12].

Morphological plaque features of untreated NCLs according to functional assessment were available for four studies (Table 2). At IVI analysis, hemodynamically significant lesions were longer (in 3 of 4 reporting studies) and more frequently exhibited high-risk characteristics, including TCFA (in 2 of 2 reporting studies), PB ≥ 70% (in 1 of 2 reporting studies), macrophages infiltration (in 1 of 2 reporting studies), and LRP (in 2 of 2 reporting studies). The definition of untreated NCLs was heterogeneous across studies but always included the identification of untreated AICLs. TVF was generally defined as a lesion-level composite of cardiac death, TV-MI, and TVR. While cardiac death and TV-MI were consistently included across studies, the definition of TVR varied, encompassing ischemia-driven TVR, unplanned coronary revascularization, or rehospitalization for unstable or progressive angina (Sup. Table S2).

Table 2.

Morphological plaque features in lesions with versus without functionally significant lesions.

Study	IVI device	IVI Morphological Plaque Features			P value
			Physio – group	Physio + group
CLIMA[13]	OCT	TCFA, % (n)	18.7 (148)	28.3 (32)	0.023
		Lipid arc > 180°, % (n)	42.4 (336)	60.2 (68)	0.001
		MLA<3.5 mm2, % (n)	36 (311)	61.7 (74)	< 0.001
		Macrophages, % (n)	60.7 (481)	73.5 (83)	0.009
		Lesion length, mm	12.9 [6.4–20.2]	15.8 [8.4–23.4]	0.011
Xu et al.[15]	OCT	TCFA, % (n)	18.8 (216)	26.2 (45)	0.025
		Macrophages, % (n)	90.8 (1042)	95.3 (164)	0.064
		Lesion length, mm	16.7 ± 8.4	22.1 ± 11.7	< 0.001
PROSPECT[11]	IVUS	PB ≥ 70%, % (n)	2.1 (56)	38.2 (232)	< 0.001
		MLA< 4 mm2,% (n)	13.4 (350)	47.3 (287)	< 0.001
		Lesion length, mm	11.9 [11.5–12.3]	32.8 [31.5–34.2]	< 0.001
PROSPECT II[12]	IVUS and NIRS	PB ≥ 70%, % (n)	67.8 (118)	73.3 (11)	0.91
		MLA< 4 mm2, % (n)	82.8 (144)	100 (15)	-
		MaxLCBI 4 mm2 ≥ 324.7, % (n)	40.8 (69)	60 (9)	0.26
		Lesion length, mm	28.0 [22.4–33.6]	28.5 [24.3–32.7]	0.88

Values are expressed as mean (± standard deviation) or % (n). IVI = intravascular imaging; Physio − group = functional-negative group; Physio + group = functional-positive group; OCT = optical coherence tomography; TCFA = thin-cap fibroatheroma; LRP = lipid-rich plaque; MLA = minimum lumen area; IVUS = intravascular ultrasound; PB = plaque burden; NIRS = near-infrared spectroscopy; MaxLCBI 4 mm² = maximum lipid-core burden index.

Compared to the reference group, the functional-positive group was associated with a significantly increased risk of TVF (random effects HR 3.59, 95% CI 1.56–––8.24, p = 0.002) (Fig. 1).The morphology-positive group was also associated with increased risk of subsequent TVF (random effects HR 2.94, 95% CI 2.22–––3.90, p < 0.001)(Fig. 2). The highest risk of long-term clinical outcomes was observed in the morpho-functional-positive group, with a pooled random effects HR of 8.75 (95% CI 4.93 – 15.55, p < 0.001) (Fig. 3). In the indirect comparison, the pooled random effects HR for the morphology-positive group versus the functional-positive group was 1.27 (95% CI 0.63–––2.53, p = 0.503) (Fig. 4), indicating no statistically significant difference in the associated risk of TVF. Nevertheless, this finding should be interpreted with caution and does not imply equivalence between morphology-based and physiology-based strategies, given the limited number of included studies and the wide CIs of the indirect comparison. Analysis stratified by IVI technique revealed consistent findings across studies using OCT and those using IVUS, with no significant interaction observed between imaging modality and effect estimates. Heterogeneity was low-to-moderate across most analyses (Supplementary Figs. S2–4). Subgroup analyses for TVF components (cardiac death, TV-MI, and TVR) were not feasible due to the limited number of studies reporting separate HRs. Nevertheless, exploratory evidence showed heterogeneous findings regarding the prognostic impact of the morphology-positive group of untreated NCLs on hard versus soft endpoints. In the CLIMA study, the increased risk of TVF of non-flow-limiting AICLs with TCFA was mainly related to hard outcomes such as cardiac death (HR 3.44, 95% CI 1.45–––8.17, p = 0.005) and TV-MI (HR 9.42, 95% CI 2.36–37.7, p = 0.002), rather than TVR (HR 2.52, 95% CI 0.93–6.8, p = 0.069). In the pooled COMBINE (OCT-FFR) and PECTUS-obs analysis, both TV-MI (HR 9.42, 95% CI 2.01–44.1, p = 0.004) and TVR (HR 3.45, 95% CI 1.74–6.84, p < 0.001) significantly contributed to the risk of TVF. Conversely, in Cho et al., the composite endpoint of non-ischemic high-risk NCLs was predominantly driven by ischemia-driven TVR (HR 11.74, 95% CI 2.53–54.46, p = 0.002), with no significant association with cardiac death (HR 0.57, 95% CI 0.18–1.77, p = 0.331). The risk of functionally significant NCLs-related clinical events appeared to be mainly driven by TVR in both PROSPECT (HR 12.78, 95% CI 6.81–24.0, p < 0.001) and CLIMA (HR 2.87, 95% CI 1.19–6.9, p = 0.014), with a weak and non-significant association with TV-MI (HR 8.46, 95% CI 0.77–––93.27, p = 0.08)(HR 3.15, 95% CI 0.97–––10.21, p = 0.057). Consistently, Xu et al. revealed that most events associated with NCLs were caused by TVR (32 unplanned coronary revascularizations vs 11 non-fatal MI) at 5-year follow-up, and multivariate analysis founded an independent association with TVF only for OCT-defined TCFA (adjusted HR 4.46, 95% CI 2.33–8.56, p < 0.001) and not for μQFR ≤ 0.80 (adjusted HR 1.46, 95% CI 0.71–3.01, p = 0.304).

Fig. 1.

Forest plot showing target vessel failure (TVF) in the functional positive group.

Fig. 2.

Forest plot showing target vessel failure (TVF) in the morphology positive group.

Fig. 3.

Forest plot showing target vessel failure (TVF) in the morpho-functional positive group.

Fig. 4.

Indirect comparison of the morphology positive versus the physiology positive group.

4. Discussion

Our meta-analysis, which included over 7500 AICLs, offers important insights on the prognostic evaluation of untreated NCLs: (I) high-risk plaque morphology provides a significant incremental prognostic impact, irrespective of the functional severity of the NCLs, and improves the identification abilities of patients with subsequent TVF; (II) flow-limiting AICLs were associated with morphological features of vulnerability (III) high-risk morphology and impaired physiology similarly predicted the risk of TVF, and their coexistence defined NCLs at the highest risk of long-term clinical events. These findings underscore the critical relevance of an integrated morpho-functional assessment of untreated NCLs for better risk stratification. However, the HRs reported in this meta-analysis are unadjusted and should therefore be interpreted as prognostic associations rather than causal estimates, as adjusted, lesion-level estimates were inconsistently reported or unavailable in the included studies. Consequently, the present findings should be regarded as hypothesis-generating rather than practice-changing.

Our meta-analysis demonstrated that the presence of high-risk plaque characteristics in deferred non-ischemic vessels was associated with an approximately 3-fold increased risk of TVF. Notably, the clinical utility of this finding is underscored by the fact that four OCT studies and one IVUS study identified the morphology-positive group by evaluating the presence of a single vulnerable plaque marker, TCFA, and PB ≥ 70%, respectively. TCFA is considered the hallmark of rupture-prone plaque, while large PB represents the strongest quantitative feature determining the risk of symptomatic atherothrombotic event rather than silent plaque disruption. [8] The clinical impact of PB was underlined in the PROSPECT study, which found that the detection of PB ≥ 70% in NCLs with radiofrequency IVUS-defined TCFA almost tripled the risk of major adverse cardiovascular events at a mean follow-up of 3.4 years (HR shifted from 3.90 in NCLs with TCFA to 10.83 in NCLs with TCFA and PB ≥ 70%). [19] Moreover, the extended follow-up of the CLIMA study showed that the PPV of the four combined high-risk OCT criteria (e.g., TCFA with < 75 µm minimal fibrous cap thickness, lipid arc > 180°, minimum lumen area < 3.5 mm², and macrophages accumulation) for adverse outcomes increased from 19.4% at 1 year to 27.8% at 5-year follow-up regardless high-intensity lipid-lowering therapy, and that the single presence of TCFA was 5-fold as prevalent (18,3% vs 3,6%) and similarly predictive of cardiac death and/or target segment-MI (adjusted HR 3.04, 95% CI 1.86 – 4.96, p < 0.001 vs adjusted HR 4.33, 95% CI 2.01 – 9.33, p < 0.001).[20] These data suggest that TCFA, irrespective of other OCT-defined high-risk features, represents an independent prognostic marker of hard coronary events. Collectively, these results suggest that TCFA and PB should be considered as indicators of the overall atherosclerotic disease burden and instability rather than punctual predictors of symptomatic coronary events, since the propensity of a patient to a clinically manifest plaque destabilization is mainly explained by a stochastic model in which the more advanced coronary artery disease, the greater number of high-risk coronary lesions, the higher probability that one of them will disrupt causing an atherothrombotic event. [8], [21].

To date, the best treatment strategy for vulnerable non-flow-limiting AICLs is uncertain. While current guideline-supported strategies for NCLs remain focused on intensive risk factor modification and physiology-guided revascularization, [1], [22] interventional approaches are exploring preemptive PCI with drug-eluting stents or drug-coated balloons to seal and passivate morphologically vulnerable but non-ischemic plaques, aiming to mitigate residual risk after functionally complete revascularization. The recent PREVENT (Preventive Percutaneous Coronary Intervention Versus Optimal Medical Therapy Alone For The Treatment Of Vulnerable Atherosclerotic Coronary Plaques) study was the first randomized controlled trial to demonstrate that preventive stenting of FFR-negative vulnerable plaques with at least two high-risk features reduced the primary endpoint of TVF at 2-year follow-up when compared to optimal medical therapy alone.[23] However, the event rate was lower than expected in both arms since the trial enrolled mainly stable patients, and the results were driven by soft endpoints like ischemia-driven TVR or hospitalization for unstable or progressive angina. In our meta-analysis, the risk of TVF also appeared to be mainly driven by TVR rather than by cardiac death or TV-MI. Although only a minority of studies reported HRs for the individual components of TVF, the available data extracted consistently suggested that morphology-positive AICLs were more strongly associated with subsequent TVR than with fatal or non-fatal ischemic hard events. From a clinical standpoint, the predominance of TVR as the main driver of TVF requires cautious interpretation, as repeat revascularization may reflect physician-driven decision-making and symptom recurrence rather than spontaneous plaque-related events. In the PROSPECT study, the higher risk associated with vulnerable NCLs was mainly attributable to subsequent revascularization, while the absolute incidence of cardiac death or MI remained relatively low. [19] This dissociation should be acknowledged when interpreting the clinical significance of TVF-driven outcomes. Several large-scale ongoing trials like the INTERCLIMA (Interventional Strategy for Non-Culprit Lesions With Major Vulnerability Criteria Identified by OCT in Patients With ACS; NCT05027984) study, the COMBINE-INTERVENE (Combined Ischemia and Vulnerable Plaque Percutaneous Intervention to Reduce Cardiovascular Events; NCT05333068) study, and the VULNERABLE (Treatment of Functionally Non-Significant Vulnerable Plaques in Patients With Multivessel ST-Elevation Myocardial Infarction; NCT05599061) study will provide additional information on the impact of preventive PCI of vulnerable plaques on hard endpoints like cardiac death or non-fatal-MI.

Our work also analyses the integrated role of plaque phenotype and coronary physiology in non-revascularized NCLs to provide a better understanding of the nature of coronary events. We demonstrated that high-risk morphological features and impaired physiology portend a similar predictive value for adverse cardiovascular events, and that the combined presence of plaque vulnerability and functionally severe stenoses in untreated NCLs was associated with an 8-fold higher risk of TVF, compared to AICLs with morpho-functional-negative assessment. Coronary artery disease progression and instability rely on the constant reciprocal interplay between local physiological factors and lesion-specific vulnerable plaque characteristics. [4] The intricate interactions between coronary hemodynamic and vessel composition was investigated in a sub-analysis of the P3 study which demonstrated that among functionally significant coronary artery disease, the PPG index can distinguish plaque phenotype into focal coronary artery disease, characterized by larger PB and mostly by LRP with TCFA, and diffuse coronary artery disease, generally featured by a higher calcium burden.[24] The relationship of impaired local hemodynamic with an increase in plaque quantity and a conversion into high-risk lesions prone to future coronary events is supported by multiple non-invasive and invasive imaging studies.[25], [26] By analyzing over 1,000 vessels from the coronary computed tomography angiography −FFR registry, Yang et al. founded an almost linear correlation between flow-limiting lesions and high-risk plaque prevalence: the lower the FFR values, the higher the proportion of qualitative and quantitative high-risk plaque characteristics. [27] In addition, Usui et al. showed that the physiological severity defined by the FFR was significantly associated with OCT-defined TCFA and the lipid volume index in intermediate-to-obstructive stable coronary lesions. [28] The evidence that ischemia is likely a marker of plaque vulnerability is further confirmed by our meta-analysis where major criteria of high-risk plaque, namely OCT-defined TCFA, IVUS-defined PB ≥ 70%, NIRS-defined LRP and macrophages infiltration were more frequently localized in ischemic coronary plaques.

Emerging hybrid approaches like IVI-derived physiological evaluation methods or the pressure pullback gradient index emphasized the usefulness of simultaneously assessing atherosclerotic plaque composition and functional significance of NCLs, using a single device (IVI catheter or pressure wire) with a single pullback. The findings of the present analysis strongly support the use of IVI-based FFR computational tools, which extract the hemodynamic severity of vessel stenoses from IVI-derived reconstruction of the coronary lumen geometry, without requiring hyperemic agents or pressure wires. [29], [30], [31] This simplification may lower the barrier for widespread adoption of morpho-functional assessment of AICLs, particularly in multivessel coronary artery disease or in patients undergoing complete coronary revascularization planning. The rationale for incorporating morpho-functional plaque assessment during coronary angiography is particularly relevant in the setting of acute coronary syndromes, where the hybrid approach enables both the combination of IVI- and physiology-guided stent optimization in the culprit lesion and the simultaneous risk stratification of TVF associated with NCLs. A recent ILUMIEN-IV (Optical Coherence Tomography [OCT] Guided Coronary Stent Implantation Compared with Angiography: A Multicenter Randomized Trial in PCI) post-hoc analysis showed that lower post-percutaneous coronary intervention virtual flow reserve (VFR), an OCT-derived physiological assessment, was an independent predictor of TVF at 2-year follow-up; further supporting the integration of IVI-derived physiology during IVI-guided percutaneous coronary intervention.[32] Moreover, in the VISION (The Virtual Flow Reserve versus Angiography‐derived FFR for Stent Implantation) study, the application of OCT-VFR significantly modified percutaneous coronary intervention planning in a substantial proportion of cases.[33] By redefining coronary artery disease pattern and stent length, OCT-VFR reduced the risk of overtreatment associated with OCT alone and avoided undertreatment linked to angiography-derived physiology, thus supporting the role of this hybrid tool in refining percutaneous coronary intervention strategies. [33].

From a clinical perspective, these findings emphasize the prognostic value of an integrated anatomical and physiological assessment for the identification of “significant” NCLs at risk of future events. Rather than relying solely on approximative angiographic visual estimation, the ability to simultaneously characterize plaque, and quantify flow may offer the most accurate and effective risk stratification.

5. Limitations

The findings of this analysis should be interpreted in consideration of several limitations. First, the definition of untreated NCLs was not fully uniform across the included studies, reflecting differences in study design, clinical setting, and lesion- vs. vessel-level analyses. This variability may have introduced heterogeneity in event attribution and outcome assessment. Nevertheless, all studies consistently focused on AICLs that were deferred from revascularization, allowing a meaningful comparison of prognostic implications across different assessment strategies. Second, the HRs reported in each included study were unadjusted, as adjusted estimates were available for only a minority of studies; therefore, we prioritized a broader and more inclusive analysis over one restricted to adjusted data. Third, since only two studies enrolled patients with chronic coronary syndrome, the findings of this meta-analysis are mainly applicable to the acute coronary syndrome population and cannot be generalized to all individuals with coronary artery disease. Fourth, follow-up duration varied across studies, which may have influenced event rates and comparability of outcomes. Fifth, the moderate risk of bias observed on the Newcastle–Ottawa Scale reflects the substantial methodological heterogeneity across the included studies, including differences in patient selection, imaging protocols, TVF definition − especially regarding TVR − and the thresholds used to define high-risk plaque morphology or functionally significant AICLs. Moreover, physiological data were in some studies derived retrospectively from angiographic or IVI reconstructions rather than from wire-based measurements, introducing potential measurement variability. However, the objective of this meta-analysis was not to compare individual techniques or specific cut-offs, but rather to classify NCLs according to their functional significance, irrespective of the modality used, in order to compare the prognostic value of a morphology-based assessment versus a conventional physiology-based approach.

6. Conclusions

This meta-analysis showed that high-risk plaque morphology and impaired physiology in untreated NCLs were both independently associated with an increased risk of TVF at long-term follow-up. Functional-positive NCLs were associated with high-risk plaque composition, and the coexistence of vulnerable morphological features and hemodynamic significance identified AICLs at the highest risk of future events. These findings support the importance of an integrated morpho-functional approach to improve risk stratification, enhance the identification of coronary segments prone to TVF, refine revascularization decisions, and potentially improve long-term patient outcomes.

CRediT authorship contribution statement

Vincenzo Acerbo: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Arturo Cesaro: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Flavio Giuseppe Biccirè: Writing – review & editing, Visualization, Supervision, Formal analysis, Conceptualization. Vincenzo De Sio: Writing – review & editing, Visualization. Antonio Capolongo: Writing – review & editing, Visualization. Maria Grazia Monaco: Visualization. Elisabetta Moscarella: Writing – review & editing, Visualization. Felice Gragnano: Writing – review & editing, Writing – original draft, Visualization, Supervision, Methodology, Conceptualization. Francesco Pelliccia: Visualization, Supervision, Conceptualization. Francesco Prati: Visualization, Supervision, Conceptualization. Dobromir Dobrev: Visualization, Supervision. Paolo Calabrò: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: