Preventive treatment of thin-cap fibroatheroma in patients with diabetes mellitus: please, mind the gap!

This editorial refers to ‘Impact of non-culprit lesion TCFA on adverse cardiovascular events in patients with diabetes mellitus who present with acute myocardial infarction’, by Y. Chen et al., https://doi.org/10.1093/ehjci/jeaf219. Vulnerable coronary plaques (VP) are defined as those at high-risk of developing atherothrombotic complications.1–4 For decades, identifying the VP has been the never-ending search for the Holly Grail in cardiology with the hope of devising therapeutic strategies to prevent the occurrence of acute myocardial infarction (MI). It is well known that most MI originate from plaques that do not severely impinge into the lumen but, nevertheless, are large and hold a characteristic anatomic substrate.1–4 The unsurpassed spatial resolution provided by optical coherence tomography (OCT) (15 μm) together with its unprecedented ability to differentiate tissue plaque components, provides a unique opportunity to identify VP. Histopathological correlations have confirmed the value of OCT to differentiate lipidic from fibrotic and calcified plaques. Likewise, OCT may unravel areas of macrophages and cholesterol crystals accumulation and microvessels. Furthermore, in patients suffering an acute coronary syndrome (ACS) the presence of a fibrin- or platelet-rich intraluminal thrombus on OCT represents the hallmark of a complicated, ‘culprit’ plaque. These acute thrombi usually overlie ruptured or eroded plaques and less often eruptive calcified nodules. More importantly, however, OCT is the only technique currently available able to measure the thickness of the fibrous cap.1–4 This is crucial as thin cap fibroatheromas (TCFA) represent the most frequent underlying substrate for ACS. Some studies have demonstrated that OCT-detected TCFA confers an independent risk for future cardiovascular events.5,6 Therefore, the idea of preventive focal sealing of VP emerged as an attractive possibility in the cathlab to passivate these high-risk plaques. Nevertheless, the positive predictive value of OCT-detected TCFA for future cardiovascular events remains very low and, currently, unable to counterbalance the low but inherent acute and long-term risk associated with any coronary intervention.5,6 Only a minority of TCFA eventually lead to a ruptured plaque causing a clinically detectable atherothrombotic complication. Most TCFA remain uneventful during follow-up and some even experience progressive phenotypic changes (i.e. cap thickening, fibrosis or calcification) consistent with plaque stabilization.1–6 Moreover, some MI originate from plaque erosion, a substrate impossible to identify before the complication occurs. However, the potential benefit of VP treatment may be more evident when this anatomic substrate is encountered in patients with an enhanced inflammatory and thrombotic risk. Aggressive systemic therapeutic strategies (intense lipid-lowering, anti-inflammatory drugs) might offer novel therapeutic venues in these high-risk patients. Patients with diabetes mellitus (DM) constitute the classical paradigm of this perfect storm scenario, as they have an enhanced inflammatory and prothrombotic milieu and an increased risk of atherothrombotic complications.6 However, the natural history of TCFA in patients with DM remains unsettled. Is a preventive treatment of TCFA in patients with DM justified by the currently available evidence? In this issue of the European Journal of Cardiac Imaging Cheng et al.7 present an interesting observational study including 1312 AMI patients who underwent 3-vessel OCT to assess the prognostic implications of TCFA in non-culprit lesions (NCL) according to the DM status. TCFA was defined as a lipid-rich plaque with a thinnest fibrous cap thickness <65 μm and maximum lipid arc >180°. DM patients (n = 311, 23.7%) had more NCL and a smaller vessel diameter compared with those without DM. Importantly, DM patients showed a significantly higher prevalence of NCL TCFA (43.7% vs. 28.9%, P < 0.001). During a median follow-up of 4.1 years, patients with TCFA had more NCL-related major adverse cardiac events (MACE) (including cardiac death, non-fatal MI and unplanned revascularization). MACE occurred also more frequently in DM patients (10.2% vs. 4.2%; HR: 2.10; 95% CI: 1.23–3.57), primarily driven by non-fatal MI (3.7% vs. 1.0%; HR: 3.33). Notably, DM was independently associated with the presence of TCFA, but not with NCL-MACE after adjusting for TCFA. Of note, insulin-dependent DM conferred an even poorer prognosis and was an independent predictor of both TCFA and NCL-MACE. This is an important study confirming the prognostic importance of OCT-detected TCFA and suggesting that DM patients with MI have a higher incidence of NCL-MACE, primarily due to a higher prevalence of NCL TCFA. Some issues, however, deserve further attention. First, the methodology of this large single-centre study is unique. Indeed, in the dynamic and unstable setting of a MI obtaining 3-vessel OCT imaging after the culprit lesion have been treated is indeed very challenging. Accordingly, these investigators should be commended for this meritorious research endeavour that sheds new light on this unique scenario. As expected, only patients deemed candidates for OCT examination in all three main epicardial coronary arteries were included. This would represent a selection bias. No information was provided on the number and characteristics of the excluded MI patients, so caution is required when considering the generalizability of current findings. Second, OCT of the culprit lesions was not obtained before treatment although the MI-related vessel was also interrogated. Information on the nature of the culprit lesions (ruptured plaque, erosion or eruptive calcified nodule) would have provided interesting additional insights with respect to the correlations with the pathological substrate of NCL, including the presence of TCFA. Third, up to 50% of MI patients have significant multivessel disease. Whether in this study all significant NCL were treated according to an angiographic vs. physiological assessment, remains unclear. It appears that OCT imaging was restricted to non-severe NCL. This should be considered because it has been suggested that the prevalence of TCFA increases in relation to lesion severity.8,9 Moreover, the risk of TCFA is higher in severe lesions. Therefore, it is likely that the presence of NCL TCFA would have been even higher if all NCL, including even those requiring intervention, would have been interrogated with OCT. Fourth, up to 18.6% of patients with DM had a thrombus identified by OCT in NCL. Whether these findings indicate silent, clinically unrecognized, plaque complications or just difficulties in identifying the real culprit lesion (i.e. in non-ST segment elevation MI) remains unclear. Moreover, whether these thrombus laden NCL, were TCFA or indeed responsible of the increased risk, was not defined. Fifth, the potential incremental predictive value provided by surrogate markers of inflammation or prothrombotic activity was not examined. This precludes gaining insights on the additional risk of TCFA associated with a high-risk blood or in high-risk patients. Sixth, in the multivariate regression model angiographic diameter stenosis >50%, rather than minimal lumen area (MLA) by OCT, was included. This is controversial considering the major predictive value of absolute MLA in most previous studies. In the PROSPECT natural history study, both MLA and plaque burden emerged as independently associated with an adverse outcome.10 Unfortunately, OCT is unable to visualize and measure the total plaque area at the lesion site and thus the potential prognostic implications of plaque burden, could not be analyzed. Furthermore, in the multivariate model, even after adjusting for TCFA, the presence of DM remained closely predictive (P = 0.059) of MACE. This would suggest that not only TCFA but also the DM status per se hold a major prognostic value. Last but not least, the authors suggested that their results are consistent with the growing recognition that targeting VP, even in the absence of significant stenosis, can improve patient outcomes and reduce future cardiovascular risk. We, respectfully disagree with this conclusion because, the predictive value of TCFA, alone associated with other adverse morphologic features, for NCL-MACE remains still too low to justify a coronary intervention.1–6 In the present study up to 85% of the patients with DM and NCL TCFA were free of MACE at 5 years. This suggests that major attention should be paid to ‘mind the gap’ until further evidence is obtained to better identify high-risk patients. There is robust scientific evidence demonstrating the prognostic value of intracoronary imaging to identify plaque features conferring a higher risk for adverse events.1–6 In PROSPECT the presence of TCFA -as detected by virtual histology- and MLA and plaque burden -by intravascular ultrasound- were identified as major determinants of adverse events during follow-up.10 In PROSPECT II lesions with a large lipid-core and high-plaque burden on NIRS-IVUS were at increased risk of future events.11 The only randomized study suggesting the value of preventive treatment of non-flow-limiting VP to reduce cardiac events was PREVENT.12 In this study, however, adverse plaque characteristics were mainly based on MLA and plaque burden on intravascular ultrasound whereas OCT findings played a marginal role. Data regarding the prognostic value of OCT features of VP have been recently provided by three large prospective studies. In CLIMA 1003 patients underwent OCT interrogation of the left anterior descending coronary artery to ascertain the relationship between plaque morphology and clinical outcomes.5 The study demonstrated the superior predictive value of combined multiple OCT high-risk plaque features in the same lesion (MLA, cap thickness <75 µm, lipid circumference, and macrophages), as an independent predictor of 1-year MACE (HR 7.54, 95% CI: 3.1–18.6). In the extended 5-year clinical follow-up, the simultaneous presence of these 4 OCT high-risk features remained independently associated with adverse events.13 Interestingly, however, the presence of TCFA was five-fold as prevalent and similarly predictive of 5-year MACE. A sub-analysis of CLIMA suggested the additional importance of associated inflammatory markers. In that study, large superficial macrophage accumulations were independently associated with MACE in patients with high C-reactive protein levels.14 COMBINE was a natural history study assessing the importance of plaque phenotype by OCT precisely in patients with DM.6 Among 550 DM patients, 390 (81%) patients had ≥1 FFR-negative lesions [98 (25%) TCFA and 292 (75%) without TCFA]. TCFA was the strongest independent predictor of MACE (HR 5.12; 95% CI: 2.12–12.34; P < 0.001). A sub-analysis of COMBINE suggested that 73% of NCL were lipidic but only 1/3 of them were TCFA.15 Only plaques with TCFA were associated with adverse events, whereas thick-cap lipidic plaques did not portended an increased risk. Another sub-analysis of COMBINE demonstrated that plaques with TCFA, a high-plaque burden, reduced MLA, and a complicated appearance, were predictors of adverse events. However, the presence of multiple OCT-determined vulnerability features drastically increased the likelihood of 5-year MACE.16 In PECTUS 438 post-MI patients with FFR-negative NCL were studied with OCT. High-risk plaque phenotype (including a fibrous cap thickness <65 μm, a lipid arc >90°, and plaque rupture or thrombus presence) was associated with a two-fold risk of adverse events mainly driven by unplanned revascularizations.17 Finally, a recent pooled analysis of COMBINE and PECTUS, including 810 patients with FFR-negative NCL, demonstrated that high-risk plaques were associated with MACE and that the risk increased with the presence of multiple high-risk features.18 There is overwhelming evidence supporting the predictive clinical value of OCT-detected VP features.1–6 Among those the TCFA phenotype is associated with the highest risk. The presence of multiple adverse morphologic traits is not so prevalent but exponentially increases the risk. Nevertheless, we should remain skeptical on the value of preventive focal sealing of VP until more evidence is accrued. The results of several ongoing randomized trials (COMBINE-INTERVENE, INTERCLIMA, VULNERABLE)19 should be awaited to support this decisive step that should be grounded on robust clinical evidence. Whether plaque analysis using artificial intelligence might provide a plaque staging classification system with superior predictive value to inform medical treatment escalation or even justify a coronary intervention, remains to be proven. At present, we should just mind the (evidence) gap! Fernando Alfonso (Writing—original draft [MD]), David Del Val (Writing—original draft), and Elvin Kedhi (Writing—original draft) None declared. Data sharing is not applicable as this is an editorial.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.