The FORZA (FFR or OCT Guidance to RevasculariZe Intermediate Coronary Stenosis Using Angioplasty; NCT01824030) trial is an open-label, single-center, prospective, randomized trial comparing clinical outcomes in patients with at least 1 AICL managed with OCT or FFR guidance. The study design and main results at 1-month, 13-month, 3-year, and 5-year follow-up have been previously published.6,8–13 The study was approved by the ethics committee of our institution (code 6261/13), and all patients signed a dedicated informed consent form.

Patients with acute or chronic coronary syndrome and at least 1 AICL were screened for enrollment. An AICL was defined as a coronary lesion with a visually estimated percentage diameter stenosis ranging between 30% and 80% in the nondistal segment of a major epicardial vessel. Key exclusion criteria included age <18 years, left ventricular ejection fraction ≤ 30%, recent (< 7 days) ST-segment elevation myocardial infarction (STEMI), recent (< 48hours) non-STEMI, prior STEMI in the territory supplied by the vessel with the AICL under investigation, severe valvular heart disease, multivessel disease with 1 or more untreated angiographically critical stenoses, and multivessel disease requiring coronary artery bypass graft intervention. The full list of inclusion and exclusion criteria is reported in table 1 of the supplementary data.

Eligible patients were randomly assigned in a 1:1 ratio to OCT or FFR guidance for both PCI performance and, in the case of revascularization, optimization of PCI outcomes. In patients with multiple AICLs in different vessels, all AICLs were managed according to the randomized arm. Operators were asked to strictly adhere to the guidance provided by the assigned tool, without utilizing the alternative modality.

Complex lesions were defined as those with at least 1 of the following angiographic features: lesion length >38mm, severe calcification, or bifurcation. Lesion length was assessed according to angiography. Severe calcification was defined according to angiography as fluoroscopic radiopacities involving both sides of the arterial wall in more than 1 location and with a total length of at least 15mm.14 Bifurcation lesions were defined as lesions adjacent to, and/or involving, the origin of a side branch of at least 2.5mm.15 “Complex patients” were defined as those with at least 1 vessel with a complex lesion.

OCT images were acquired at the site of the AICL using commercially available systems. The following measurements were collected: minimal luminal area (MLA), proximal reference luminal area (RLA), distal RLA, and mean RLA. Based on these parameters, the percentage of area stenosis (AS) was calculated using the following formula: (mean RLA−MLA)/mean RLA×100. Plaque rupture was defined as a recess in the plaque beginning at the luminal-intimal border.16

PCI was performed when at least 1 of the following criteria was met: a) AS ≥75%; b) AS between 50% and 75% and MLA <2.5 mm2; and c) AS between 50% and 75% and plaque rupture.

In patients who underwent revascularization, OCT was also used to guide PCI and optimize results. Additional interventions were performed in the presence of major stent malapposition (defined as a distance between the strut and vessel wall >350μm or <350 and >200μm for a length >600μm), major underexpansion (defined as an in-stent minimal cross-sectional area <75% of the RLA), or major edge dissection (defined as a dissection starting at the level of the stent edge and propagating for a length >600μm). The absence of any of these abnormalities was deemed an “optimal OCT result”. No specific PCI optimization algorithm for complex lesions was set. The details of the OCT study protocol have been previously reported in the study design.8

FFR was obtained using a 0.014-inch pressure monitoring guidewire and hyperemia induction through intravenous administration of 140mg/kg/min adenosine. PCI was not performed when FFR was >0.80, while it was performed when FFR was ≤0.80. Poststenting FFR was assessed, and when it was <0.90, additional stent postdilation could be performed. If FFR remained <0.90, a guidewire pullback was performed to identify potential pressure drops at least 5mm away from the initial stent, potentially requiring further stent implantation according to physician preference.17 The final achievement of an FFR ≥0.90 was defined as an “optimal FFR result”. The details of the FFR study protocol have been previously reported in the study design.8

The primary outcome was major adverse cardiac events (MACE), defined as a composite of all-cause death, myocardial infarction (MI), or target vessel revascularization. Key secondary outcomes included individual components of the primary outcome, the composite of cardiac death or MI, and cardiac death.

All analyses were performed according to the intention-to-treat principle and on a per-vessel basis. Categorical variables are reported as total numbers and percentages, while continuous variables are reported as mean±standard deviation or median [interquartile range (IQR)], according to their distribution assessed by the Shapiro-Wilk test. Comparisons among groups for categorical variables were performed with the chi-square or Fisher exact test, while those for continuous variables were performed with the Student t test or Mann-Whitney U test, as appropriate. The Kaplan–Meier method was used to characterize the time until the first event for the primary outcome. Hazard ratios (HR) and 95% confidence intervals (95%CI) were computed for primary and secondary outcomes with Cox proportional hazards time-to-event analyses. Subgroup analyses for the primary outcome were performed in complex and noncomplex cohorts according to deferred or performed PCI and in the complex lesion cohort according to the components of complex lesions and the number of complex lesion criteria. Statistical significance was defined as a P value <.05. Statistical analyses were performed using Stata version 18 (StataCorp).

A total of 350 patients (with 420 vessels with AICLs) were randomly assigned to OCT (174 patients, 200 vessels) or FFR (176 patients, 220 vessels). When stratifying the study population based on lesion complexity, 199 patients (212 vessels) had at least 1 complex lesion, while 151 patients (208 vessels) had only noncomplex lesions.

The baseline clinical characteristics of the study population are reported in table 1. All baseline clinical features were well-balanced between patients with complex and noncomplex lesions, except for a higher rate of diabetes mellitus (35.2% vs 25.2%; P=.045) and a lower rate of previous PCI (37.7% vs 48.3%; P=.046) in patients with complex lesions (table 1). Patient characteristics were well-balanced between the treatment arms for both complex and noncomplex lesions (table 2 of the supplementary data).

The baseline vessel characteristics are reported in table 2. Vessels with complex lesions were more frequently located in the left anterior descending artery (75.9% vs 55.8%; P<.001) and were more likely to undergo PCI (48.1% vs 29.3%; P<.001) than vessels with noncomplex lesions. Among vessels with complex lesions, 55 (25.9%) had a long lesion (of which 37 [67.3%] were treated with PCI), 53 (25.0%) had a severely calcified lesion (of which 20 [37.7%] were treated with PCI), and 169 (79.7%) had a bifurcation lesion (of which 85 [50.3%] were treated with PCI) (table 2 and table 3 of the supplementary data). The OCT graphical representation of the most common complexity criterion, along with the most frequent OCT criterion for performing PCI, is shown in figure 1.

Figure 1.

OCT representation of the most common angiographic complexity and PCI criteria. Bifurcation was the most common angiographic complexity criterion, while an AS of 50%-75% with an MLA <2.5 mm2 was the most frequent criterion for PCI in the OCT arm. The illustrated case shows a bifurcation of the LAD-D1, with a lesion mainly affecting the proximal main vessel, an AS of 62%, and an MLA of 2.3 mm2. AS, area stenosis; D1, first diagonal branch; LAD, left anterior descending artery; MLA, minimal lumen area; MV, main vessel; OCT, optical coherence tomography; PCI, percutaneous coronary intervention.

(0.57MB).

Vessels with complex and noncomplex lesions had similar rates of poststenting optimal results (54.5% vs 63.8%; P=.309), PCI optimization (33.8% vs 27.7%; P=.477), optimization by further balloon dilation (31.2% vs 25.5%; P=.502), optimization by additional stent implantation (7.8% vs 8.5%; P=.887), and final optimal results (85.5% vs 89.1%; P=.572).

Vessels in the OCT group were more likely to undergo PCI than those in the FFR group, both in complex (58.8% vs 38.2%; P=.003) and noncomplex lesions (39.8% vs 20.0%; P=.002) (table 4 of the supplementary data). The rates of OCT criteria for PCI were similar between vessels with complex or noncomplex lesions, including AS ≥ 75% (18.8% vs 29.2%; P=.236), AS 50% to 75% and MLA <2.5 mm2 (34.7% vs 29.2%; P=.357), and AS 50% to 75% with plaque ulceration (5% vs 3.1%; P=.526).

Finally, optimal results were more frequent in the OCT group than in the FFR group, both at the poststenting assessment (complex: 41.2% vs 19.2%, P=.054; noncomplex: 36.4% vs 7.1%, P=.041) and the final assessment (complex vessels: 100.0% vs 58.3%, P<.001; noncomplex vessels: 100.0% vs 64.3%, P<.001).

Clinical outcomes according to angiographic lesion complexity and randomization are presented in table 5 of the supplementary data. Compared with vessels with noncomplex lesions, those with complex lesions had numerically higher rates of MACE (20.8% vs 13.9%; HR, 1.52; 95%CI, 0.95-2.44; P=.078), cardiac death or MI (4.7% vs 3.4%; HR, 1.42; 95%CI, 0.54-3.73; P=.476), all-cause death (12.7% vs 8.2%; HR, 1.57; 95%CI, 0.86-2.88; P=.145), cardiac death (2.8% vs 2.4%; HR, 1.19; 95%CI, 0.36-3.90; P=.775), MI (2.4% vs 1.0%; HR, 2.48; 95%CI, 0.48-12.76; P=.279), and target vessel revascularization (8.0% vs 5.8%; HR, 1.40; 95%CI, 0.67-2.94; P=.368).

Clinical outcomes according to angiographic lesion complexity and randomization are presented in table 3.

Compared with FFR, OCT was associated with a lower risk of MACE in vessels with complex lesions (14.7% vs 26.4%; HR, 0.53; 95%CI, 0.28-0.98; P=.044) but a higher risk of MACE in vessels with noncomplex lesions (19.4% vs 9.1%; HR, 2.23; 95%CI, 1.04-4.81; P=.040), with a significant treatment-by-subgroup interaction for lesion complexity (Pinteraction=.004) (figure 2). Similar findings were found for the outcomes of cardiac death or MI (vessels with complex lesions: HR, 0.12; 0.01-0.91; P=.040; vessels with noncomplex lesions: HR, 1.59; 95%CI, 0.35-7.09; P=.546; Pinteraction=.044) and all-cause death (vessels with complex lesions: HR, 0.36; 95%CI, 0.15-0.85; P=.020; vessels with noncomplex lesions: HR, 2.86; 95%CI, 1.01-8.11; P=.049; Pinteraction=.003). Similar trends, although not statistically significant, were found for the outcomes of cardiac death (Pinteraction=1.000), MI (Pinteraction=.422), and target vessel revascularization (Pinteraction=.790) (table 3).

Figure 2.

Cumulative incidence of major adverse cardiac events at 5 years in complex and noncomplex groups. FFR, fractional flow reserve; OCT, optical coherence tomography.

(0.18MB).

Results for the primary outcome were consistent among the subgroups of deferred and performed PCI in both complex (Pinteraction=.451) and noncomplex (Pinteraction=.614) cohorts (table 6 of the supplementary data).

Results for the primary outcome in the complex lesion cohort revealed a consistent trend among the individual components of the definition of complex lesions (figure 3) with a significant risk reduction with OCT in the subgroup of severely calcified lesions (OCT vs FFR: HR, 0.20; 95%CI, 0.07-0.60; P=.004) and a nonsignificant reduction in the subgroups of lesions longer than 38mm (OCT vs FFR: HR, 0.48; 95%CI, 0.12-1.94; P=.306) and bifurcation lesions (OCT vs FFR: HR, 0.59; 95%CI, 0.29-1.20; P=.147).

Figure 3.

Subgroup analyses for major adverse cardiac events in the complex lesion cohort. 95%CI, 95% confidence interval; FFR, fractional flow reserve; OCT, optical coherence tomography.

(0.17MB).

Finally, when assessing the impact of the number of criteria for complex lesions (figure 3), OCT was associated with a significant reduction in the subgroup with 2 or more criteria for complex lesions (OCT vs FFR: HR, 0.20; 95%CI, 0.06-0.69; P=.011) and a nonsignificant reduction in the subgroup with a single criterion for complex lesions (OCT vs FFR: HR, 0.81; 95%CI, 0.38-1.71; P=.583).

Our study has some limitations. First, it is a post-hoc subgroup analysis of a small, single-center trial, which limits the generalizability of the results and may introduce selection bias due to the lack of randomization based on angiographic lesion complexity. Additionally, the study lacked sufficient statistical power to detect significant differences in individual outcomes, with few events such as MI or vessel revascularization. Second, our findings should be applied specifically to the target population included in the study, which mainly consisted of patients with chronic coronary syndrome and preserved left ventricular function. Third, the FORZA trial was an open-label study, with both physicians and patients being aware of the trial-group assignments. Fourth, no correction for multiple comparisons was applied. Fifth, some vessels lacked post-stenting and/or final OCT or FFR assessment, limiting outcome comparisons based on optimal results. Sixth, noncardiac deaths accounted for a large proportion of overall mortality. However, their distribution was balanced between the randomized groups, suggesting they did not influence the overall findings. Seventh, the inclusion of angiographic intermediate lesions inherently involved plaques with a limited burden, and consequently, the definition of “complex lesion” should be interpreted within this specific context. Eighth, in the OCT arm, there was no predefined protocol for PCI optimization. Ninth, the absence of a trial screening log limited our ability to thoroughly analyze and characterize the clinical or procedural criteria for patient exclusion. Finally, the potential impact of sex and gender was not evaluated in this analysis. Given these limitations, our findings should be interpreted with caution and considered hypothesis-generating.