Seeing Beyond the Coronaries: Pairing Echocardiography With Targeted Carotid Ultrasound After Percutaneous Coronary Intervention

While plaque morphology is typically the realm of the interventional cardiologist or cardiac computed tomography (CT) imager, ultrasound remains an important, often overlooked tool in the cardiac imager's armamentarium for characterizing the anatomy and physiology of vulnerable plaques [1, 2]. In this regard, Xue et al. report a single-center retrospective analysis of 475 patients undergoing percutaneous coronary intervention (PCI) in which they paired routine predischarge transthoracic echocardiography (TTE) with targeted carotid duplex imaging and followed patients for 12 months via query of health records and telephone interviews for the occurrence of major adverse cardiac events (MACE), defined as the composite of cardiac death, non-fatal myocardial infarction, target vessel revascularization, and recurrent angina [1]. In addition to standard TTE metrics, they evaluated patients’ plaque score, intima-media thickness (IMT), total plaque area (TPA), and grey scale median (GSM, an ultrasonic measure of plaque composition) [1]. This information on plaque thickness, area, echotexture, and even neovascularization [2] is readily available on a standard vascular ultrasound and has the potential to greatly inform prognostication. In particular, the period after an acute coronary syndrome, which may be marked by accelerated progression of atherosclerosis [3], is a particularly vulnerable time in which intensification of medical therapy may confer substantial health benefits. Though guideline-directed medical therapy is the standard of care for all patients with coronary artery disease [4], not all patients tolerate maximal doses of these therapies, and other competing risks may create equipoise about the optimal approach for a given patient. How hard should one push to treat, recognizing such treatments are not without risk? In this setting, carotid ultrasound represents a possible avenue to improve risk stratification and to impress upon the patient the risks associated with their individual atherosclerotic burden, thereby enhancing the teachability of this very “teachable moment” in their life [5, 6]. By characterizing the presence and severity of atherosclerotic plaque in other vascular beds, such as the carotid artery, vascular ultrasound can refine risk prediction while patients are convalescing in the hospital, ensuring therapy changes can be made and results are actionable at discharge. Moreover, such risk stratification could piggyback on existing routine diagnostics, such as echocardiography, which already provide important information on systolic and diastolic functional impairment, valvular heart disease, and possible complications of myocardial infarction or treatment sequelae [7]. Guidelines currently support the use of carotid ultrasound in the presence of recent neurologic symptoms, when carotid disease is known or suspected (e.g., via a bruit), or when imaging will inform revascularization decisions [8]. However, routine use of carotid duplex ultrasound is not commonplace despite offering a wealth of information on plaque morphology and potential future risk. Whether such information would add prognostic value over existing tools remains an important and promising area for investigation. In Xue et al. study, MACE occurred in 151 patients (31.8%) at 1 year. Those who experienced MACE were more likely to have conventional cardiovascular risk factors such as diabetes, prior myocardial infarction, and more comorbidities, but also had several abnormalities on ultrasound including increased wall thickness, lower left ventricular ejection fraction (LVEF), higher Tei index, greater IMT/TPA/plaque score, and lower GSM, with the combined ultrasound model achieving an AUC of 0.833. Perhaps more striking, the ultrasound model achieved such discrimination despite no differences between those with and without MACE in systolic blood pressure, hemoglobin, creatinine, eGFR, LDL-C, or troponin levels, coronary artery disease presentation, and SYNTAX score, or PRECISE-DAPT score. Among relevant predictors of MACE, ultrasonic features such as plaque score, Tei index, and carotid IMT, all had higher discriminatory power than clinical variables and even LVEF, potentially supporting the combination of TTE and carotid ultrasound for risk stratification. Although promising, this requires external validation and formal comparison against existing clinical scores before we can conclude that it truly outperforms conventional risk tools. Conceptually, this approach builds on dimensions lacking in traditional clinical and angiographic scores, which focus on clinical acuity or lesion complexity rather than extra-coronary plaque biology [9, 10]. Carotid plaque burden and echogenicity, measured with standardized acquisition and GSM normalization, enrich the information available on systemic atherosclerotic burden and are strongly associated with cardiovascular risk [6, 5, 11]. Moreover, use of targeted carotid imaging for risk stratification in individuals suspected of having elevated risk such as those with diabetes, prior myocardial infarction or transient ischemic attack, older age with multiple risk factors, a carotid bruit, a substantial inter-arm blood pressure difference, or poorly controlled LDL [12, 13], could be employed selectively in individuals already undergoing predischarge echocardiography to augment phenotypic characterization. At the same time, while suggesting possible value to this approach, caution should be exercised when interpreting these results. First, the generalizability of this approach is unclear, as the ultrasound model was tested in the same single center where it was derived, likely overestimating its performance in a broader population. Second, because only those who clinically received a carotid ultrasound demonstrating plaque were included, significant selection bias may have been introduced, thereby impairing the ability to evaluate the incremental value of including these carotid ultrasound variables in routine practice. Third, MACE was treated as a binary outcome at 12 months, despite occurring over the preceding 12 months. Thus, information about the timing of events is not available, and analyses do not account for individuals who experienced non-cardiac death early after PCI but who may have been prevented from experiencing MACE. Fourth, several standard TTE parameters that are useful for risk stratification were not considered, and one of the few that was used, the Tei index, is not routinely reported in many laboratories. Some predictors showed small absolute differences (e.g., heart rate) that nevertheless reached statistical significance but may not be clinically relevant. Fifth, inter-group comparisons were based on the occurrence of the outcome rather than predefined exposure strata, as in a case-control study, potentially leading to misspecification of absolute risks and bias. Sixth, important aspects of prediction, such as calibration and decision-curve analysis, are lacking but may be important for assessing the clinical utility of such an approach [14]. Lastly, these results should be seen as associative, not causal. For example, claims made by the authors, such as deeming beta-blocker use at discharge “protective” due to its association with lower event rates, imply causality. However, such claims cannot be made with the methods employed from retrospective data. At best, these associations are associative, and it is equally as probable that those unable to tolerate a beta-blocker at discharge are somehow sicker in non-quantified ways than those able to take the medication. These concerns mean the findings should be viewed as hypothesis-generating, not definitive evidence of clinical benefit. Moreover, many important safeguards are needed to ensure a high-quality ultrasound. For example, quality assurance should include fixed presets for GSM normalization to improve reproducibility and generalizability [6], clear instructions for IMT and plaque area measurement, periodic checks of intra-observer repeatability [6, 15], and clear reporting standards to facilitate integration with clinical practice. Such reporting should be linked to actionable steps for medication intensification, rehabilitation referral, and early follow-up to provide the maximal benefit to these patients. Nevertheless, the article by Xue et al. provide a clear direction for the potential value of combination testing. The evolving paradigm of combination ultrasound, such as carotid duplex, echocardiography, and lung ultrasound, has the potential to meaningfully change our point-of-care testing and risk stratification, but this approach requires standardization, rigorous testing, and careful quality assurance to ensure the right decisions are made. The insights gleaned from this article are critical to moving the needle towards a more personalized treatment approach, one that is tailored to those at greatest risk and delivers timely, impactful benefits to our patients. Dr. Strom reports research grants from the National Heart, Lung, and Blood Institute (1R01HL169517, 1R01AG063937). No additional funding supported the preparation of this commentary. Dr. Strom reports research grants from the National Heart, Lung, and Blood Institute (1R01HL169517, 1R01HL173998), Edwards Lifesciences, EchoIQ, Anumana, Philips Healthcare, Viz.ai, EVERSANA Lifesciences; consulting for Bracco Diagnostics, Edwards Lifesciences, Philips Healthcare, General Electric Healthcare, and EVERSANA Lifesciences; clinical endpoint committee for Pfizer and Agitated Solutions, and is a member of the scientific advisory boards for Ultromics, HeartSciences, Bristol Myers Squibb, Alnylam, Ultrasight, and EchoIQ. The authors declare no conflicts of interest.