Editor comment: Focus on cardiovascular risk assessment

Traditional cardiovascular risk scoring systems are a cornerstone in the prediction of adverse cardiovascular events as they play a significant role to accurately predict and stratify a patient’s cardiovascular disease (CVD) risk and thus can be instrumental in implementing treatment guidelines.1 In this issue, the performance of traditional and novel risk scoring systems and the potential for their refinement are discussed. Risk prediction models (RPMs), estimating lifetime coronary artery disease (CAD) risk are recommended tools to guide cardiovascular risk management by identifying individuals expected to benefit most from cardiovascular risk management. However, their implementation in clinical practice is limited by their abundance, lack of external validation, and lack of head-to-head comparison. Here, 28 CAD RPMs were analysed and externally validated in the three cohorts. No single best-performing RPM could be identified across all cohorts and outcomes, with comparable moderate-to-good discrimination and calibration for most RPMs, indicating that no single best-performing RPM could be distinguished.2 Risk prediction model performance was found to depend on endpoint incidence in the original development and validation cohorts, underlining the importance of external validation in multiple independent cohorts to evaluate RPM robustness and generalizability. Cardiorespiratory fitness (CRF) is an established health marker that is strongly associated with mortality outcomes, but its use in clinical practice is limited because it requires considerable time and expense, including laboratory equipment, qualified personnel to administer it, and medical supervision. Non-exercise estimated cardiorespiratory fitness (NEE-CRF) has been increasingly proposed to determine CRF using validated prediction equations,3 but methodological limitations of these studies have limited a broad application of NEE-CRF. Here, CRF was estimated using a previously developed and validated equation by Matthews et al. and evaluated in a large prospective cohort of men and women: higher NEE-CRF was independently associated with lower all-cause, CVD, and cancer mortality risk.4 However, as is the case in all epidemiological studies the findings are observational and do not suggest cause and effect relationship. All patients with symptomatic atherosclerotic disease are at very high 10-year risk of cardiovascular events. However, large variation remains in the residual risk of recurrent CVD. Identification of the patient who benefits most from further risk factor lowering may help to effectively reduce residual risk. As age is one of the most important factors in CVD risk, treatment decisions solely based on 10-year risk can lead to more intensive treatment of the elderly. Younger patients on the other hand who may have a high lifetime risk may not be identified for intensive preventive treatment as their 10-year risks are low. To deal with these shortcomings, two large studies are here presented. The first one is dealing with the possibility to predict CVD-free life expectancy rather than 10-year risk: it is based on the idea that lifetime benefit based treatment may reduce overtreatment of older patients and provide more effective intervention in younger ones.5 In patients with symptomatic atherosclerotic disease in the UCC-SMART cohort (1996–2018; n = 7697), residual risk reduction guided by lifetime benefit estimation resulted in more CVD-free life years and more CVD events avoided compared to the conventional risk factor-based strategy. However, the longer treatment durations lead to increased costs and potential side effects. The second study tested the SMART RPM, which estimates 10-year risk of myocardial infarction, stroke, or cardiovascular death and it is one of only two existing models for the general secondary prevention population. When tested in predicting the 10-year risk of cardiovascular events in a secondary prevention UK routine primary care cohort, the SMART risk prediction tool showed to be reliable, although the model performance was weaker among the cohort with newly diagnosed CVD.6 Although traditional scoring systems are widely used method for estimating CVD risk in the general population, it is well established that they underestimate the actual CVD risk in patients with chronic inflammatory disease states such as psoriasis especially since a high proportion of patients with psoriasis experience adverse cardiovascular events earlier in their life. According to recent European Society of Cardiology (ESC) guidelines, patients with documented high-grade atherosclerotic plaques on imaging have the same level of CVD risk as patients with documented clinical CVD. Here, it is presented the first study that compares the performance of risk scores (SCORE and ASCVD) to identify high CVD risk as defined by the presence of imaging-documented atherosclerotic plaques in vascular territories in psoriasis.7 It demonstrates that assessing lifetime CVD risk in patients with moderate to severe psoriasis does not adequately identify patients with subclinical atherosclerosis (coronary, femoral, and carotid atheroma plaques), which is the highest cardiovascular risk category in the ESC guidelines. Also applying a 1.5 multiplication factor to traditional risk score models, did not significantly improve their ability to predict the presence of subclinical atherosclerosis. The limited study population as well the lack of hard endpoints limit the generalizability of the current findings. The performance of cardiovascular risk scores developed for the general population in patients living with HIV (PLWH) is debated and it remains unclear which score is appropriate in clinical practice. Here it is presented the first study sought to compare the predictive performance of SCORE2, Pooled Cohort Equations (PCE) and the Data-Collection on Adverse Effects of Anti-HIV Drugs (D:A:D) scores in two cohorts including PLWH and individuals from the general population. As expected PLWH present a two-fold higher rate of incident CVD compared to individuals from the general population and the three scores presented comparable discriminative capacities.8 CHA2DS2-VASc score is widely used to guide anticoagulation therapy in non-valvular atrial fibrillation (NVAF), but its utility is debated, primarily because it is a vascular scoring system, which does not incorporate atrial fibrillation related parameters.9,10 Here, a meta-analysis to estimate its discrimination ability in predicting ischaemic stroke overall is modest and in both subgroups of patients with or without NVAF.11 This finding raises an important question: does NVAF have an independent impact on stroke risk which is unrelated to CHA2DS2-VASc parameters, or does the risk of stroke stem from common vascular risk factors and structural cardiac and vascular abnormalities that accompany NVAF? Thus, more accurate risk assessment tools are required for better patient management as well as more reliable research within the atrial fibrillation domain. The identification of individuals at increased risk of sudden cardiac death (SCD) is a cornerstone of clinical management in childhood hypertrophic cardiomyopathy (HCM). In 2019, an RPM for SCD in childhood HCM (HCM Risk-Kids) was developed, using data from 1024 patients.12 Here, the external validation of this model was performed in 421 patients. HCM Risk-Kids predictor variables included unexplained syncope, non-sustained ventricular tachycardia, maximal left ventricular wall thickness, left atrial diameter, and left ventricular outflow tract gradient. A 5-year predicted risk of ≥6% identified over 70% of SCD, confirming that HCM Risk-Kids can provide a method for individualized risk predictions in children with HCM. Another study, the risk prediction of different electrocardiogram (ECG) phenotypes was investigated in a population of 356 childhood HCM (mean age 10 years): the majority (97%) presented (97%) ECG abnormalities including repolarization abnormalities (78%), left ventricular hypertrophy (68%), abnormal QRS axis (35%); or QT prolongation (37%). Over a median follow-up of 3.9 years, 7% had an arrhythmic event: no ECG characteristic was associated with 5-year SCDrisk, suggesting a limited role of baseline ECG in risk stratification in childhood HCM.13 Work is beneficial for health, but many individuals develop CVD during their working lives. Occupational cardiology is an emerging field that combines traditional cardiology sub-specialisms with prevention and risk management unique to specific employment characteristics and conditions. In some occupational settings incapacitation through CVD has the potential to be catastrophic due to the nature of work and/or the working environment. These are often termed ‘hazardous’ or ‘high-hazard’ occupations. Consequently, many organizations that employ individuals in high-hazard roles undertake pre-employment medicals and periodic medical examinations to screen for CVD. The identification of CVD that exceeds predefined employer (or regulatory body) risk thresholds can result in occupational restriction, or disqualification, which may be temporary or permanent. This article will review the evidence related to occupational cardiology for several high-hazard occupations related to aviation and space, diving, high altitude, emergency workers, commercial transportation, and the military.14 Despite atherosclerotic plaque developing over years, many patients will have had no warning symptoms prior to a heart attack or sudden cardiac death: up to 27% patients present with these life-threatening events despite the absence of standard modifiable risk factors. Here, new potential tools in CVD risk stratification are overviewed and commented. Over the last 5 years, polygenic risk scores (PRSs), combining data based on millions of single-nucleotide polymorphisms (SNPs), have emerged as predictors of the risk of CVD and may have application, in a complementary manner, to existing risk prediction algorithms. However, there are minimal prospective trial data demonstrating the effectiveness of prevention strategies in this setting, and pathways for implementation are lacking. Here overview and commentary on the important advances in deriving and validating PRSs are presented: coordinated efforts are required to ensure that groups from different ethnic backgrounds have a dedicated PRS that the added clinical value of PRS is rigorously demonstrated in both clinical trials and registries of ‘real-life’ implementation.15 Available computed tomography (CT) technology allows the quantification of plaque burden, identification of high-risk plaques, the functional assessment of coronary lesions for ischaemia detection and revascularization for refractory angina symptoms, and more recently the assessment of perivascular Fat Attenuation Index (FAI), which is a non-invasive metric of coronary inflammation, allows for the first time the direct quantification of the residual vascular inflammatory burden. Machine learning and radiomic features’ extraction and spectral CT for tissue characterization are also expected to improve the diagnostic and prognostic yield of CT.16 Conflict of interest: none declared.