Current international guidelines for the evaluation and diagnosis of patients with stable chest pain (SCP) suspected of chronic coronary syndrome (CCS) recommended pretest probability (PTP) stratification before cardiac imaging testing (CIT), such as coronary computed tomographic angiography (CCTA) [1, 2]. This is true whether this be the guideline released by European Society of Cardiology (ESC) in 2019 [1] or American societies in 2021 (2021 GL) [2]. For the estimation of PTP, both guidelines adopted the ESC-PTP model based on age, sex and symptoms [3]. Although ESC-PTP model has been externally validated in different CCTA-based cohorts of SCP patients, the details were inconclusive for determination of the low-risk group in which further CIT should be deferred for patients [4, 5, 6, 7, 8, 9]. A recent study demonstrated that the impact of implementing 2021 GL, which assigned all patients with ESC-PTP $\le$15% to the low-risk group, remained to be elucidated for the modest improvement in efficiency and outcomes [7]. This concern is particularly crucial, because patients with ESC-PTP $\le$15% account for approximately more than half of the current SCP cohorts, and these patients may benefit from optimal medical therapy (OMT) and potentially revascularization, despite low rates of obstructive coronary artery disease (CAD) and major adverse cardiovascular event (MACE) [4, 5, 6, 7, 8, 9, 10, 11]. Consequently, further risk assessment for patients in the the low-risk group according to 2021 GL is warranted.

To address this issue, 2021 GL recognized coronary artery calcium score (CACS), a direct marker of calcified coronary atherosclerosis, as a quick, lower-radiation and relatively inexpensive tool for further risk assessment [2]. For patients with SCP and no known CAD categorized as low-risk, the 2021 GL adopted a Class 2A recommendation regarding CACS as a reasonable first-line test for excluding calcified plaque and identifying patients with a low likelihood of obstructive CAD [2]. This recommendation was supported by a meta-analysis of 79,903 patients with SCP which found the association between CACS = 0 and the low prevalence of CAD and MACE [12] and a cohort study of 33,552 patients without obstructive CAD which demonstrated that the absolute benefit of directly proportional with the CAD burden measured by CACS [13]. Our previous research from the CCTA Improves Clinical Management of Stable Chest Pain (CICM-SCP) registry also confirmed the strong potential of CACS to improve effectiveness of the diagnostic strategy [10, 11]. However, the clinical value of CACS for patients in the low-risk group according to 2021 GL still remains unclear. A recent study demonstrated a 5-year warranty period for a CACS of 0 in low-risk population [14]. Thus, the present study aims to comprehensively investigate the diagnostic and prognostic impact of CACS, as well as the association between CACS and subsequent utilization of invasive procedures, in these low-risk patients (ESC-PTP $\le$15%).

In this CCTA-based and long-term follow-up cohort study, consecutive patients with SCP suspected of CCS were classified into low and high-risk group according to the recommendations of 2021 GL. Although a percentage of patients in the low-risk group had different degrees of CAD on CCTA or suffered clinical events, higher CACS was associated with an increased likelihood of CAD (especially nonobstructive CAD), intensive utilization of invasive procedures and elevated risk of MACE with stepwise grades (CACS = 0, 0–100 and CACS $>$100) or presence (CACS $>$0) v.s. absence (CACS = 0).

Several studies have shown a low diagnostic and prognostic yield of CIT in routine testing [17, 18, 19, 20]. Hence, the evaluation of SCP suggestive of CCS remains a challenge for physicians with significantly increased costs related to these patients [21, 22]. The 2021 GL recommended risk assessment by ESC-PTP model and for patients in low-risk group (ESC-PTP $\le$15%), further CIT should be deferred [2]. Consistent with other studies [4, 5, 6, 7, 8, 9, 10, 11], the present study demonstrated that although the low-risk group had less risk burden and MACE than the high-risk group did, there was still a considerable proportion of patients in the low-risk group had obstructive CAD detected by CCTA. An increasing body of evidences has pointed to the fact that in the SCP population, approximately one-third of patients with CACS $>$0 had obstructive CAD [12], which was supported by the substantially high odds of having obstructive CAD for patients with CACS $>$0, especially $>$100, after controlling for confounders by a large range of prognostic variables for this study.

Interestingly, more than 60% patients had nonobstructive CAD among those with CACS $>$0, leading to dramatically increased OR of having any CAD across CACS strata. Recent literature suggests that most MACE occurred in patients with nonobstructive CAD detected by CCTA [23, 24]. In a large-scale trial, CCTA arm demonstrated a lower rate of MACE than the traditional care arm did during a long follow-up of 5 years, which was mainly attributed to greater intensity of OMT in response to visualize (mostly nonobstructive) CAD [25]. A recent study of 33,552 consecutive patients without obstructive CAD determined by CCTA found statin therapy post-CCTA was associated to a risk reduction of MACE in 5-year follow-up, with the number need to treat of 36, 24 and 13 in patients with CACS = 0–100, 100–400 and $>$400, respectively [13]. CACS may have the potential to provide the opportunity for the screening of subclinical atherosclerosis to improve preventive OMT. Despite the intensive utilization of invasive procedures associated with higher CACS, most MACE arose in the moderate and late stages of follow-up on Kaplan-Meyer curves, emphasizing the important role of OMT in low-risk patients without CACS = 0.

In terms of the clinical practice, not performing any CIT is a difficult concept to embrace even in the low-risk group according to 2021 GL [26]. Thus, the 2021 GL offered the option to pursue CACS as a quick, lower-radiation and relatively inexpensive tool for further risk assessment in the low-risk group, but only at a strength of recommendation with 2a and a level of evidence with B provides little guidance on the use of CACS [2]. This is the first CCTA-based and longitudinal study comprehensively investigating the clinical value of CACS in a real-world cohort of patients assigned to low-risk group by 2021 GL, leading to a potential CACS-based paradigm for specific risk assessment in these patients. For those with CACS $>$0, OMT based on recent primary and secondary prevention guidelines should be referred [1, 27, 28, 29]. For those with CACS $>$100, additional CIT, such as CCTA, may be considered. However, as shown in the present study and other research, CACS may perform worse in some subgroups [30, 31, 32], CACS of 0 is not a complete guarantee to de-escalate or defer subsequent OMT and necessary CIT.

Several other limitations of the present study merit discussion. First, this was a subgroup analysis of an observational and natural history registry. Indications for CCTA and post–CCTA management relied on the decision making of local physicians in a nonrandomized fashion. Follow-up data indicating favorable outcomes were derived from patients whose clinical care benefited from guidance by CCTA. The influence of potential selection bias could not be completely excluded, although we used multivariable adjustment to control for potential confounding by a large range. Second, our previous studies [10, 11], as well as other similar studies [4, 30, 31, 32, 33] have demonstrated that applying a CACS-based estimation of PTP to all SCP patients seemed to have been more potential to effectively identify patients with low-risk. Thus, multicentric and multiethnic randomized controlled trials are needed to assess whether incorporation of CACS as a gatekeeper in the low-risk group according to 2021 GL is noninferior to current safety and could lead to meaningful reductions in downstream CIT and health care expenditure. Third, CAD was documented using CCTA in this study. Previous studies have demonstrated that CCTA had a high negative predictive value compared with ICA [34, 35]. Thus CCTA offered robust assurance to exclude obstructive CAD [36]. Fourth, our study did not include patients with dyspnea, and the conclusions should not be extrapolated to patients with known CAD, acute chest pain, no chest pain or classified into high-risk group according to 2021 GL [37].

This is the first CCTA-based study to investigate the diagnostic and long-term prognostic value of CACS. As well we investigated the association between CACS and subsequent utilization of invasive procedures, in patients with SCP suspected of CCS and assigned to the low-risk group according to 2021 GL. Although there is still a percentage of these low-risk patients having different degrees of CAD on CCTA or suffering MACE, high CACS conveyed a significant probability of substantial stenoses and clinical endpoints, respectively. These findings support the potential role of CACS as a further risk assessment tool to improve clinical management in patients for whom subsequent CITs have been deferred based on recommendations of 2021 GL.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

JZ and GS designed the research study. JZ, CW and CL collected the patient data. CW, XZ, CL and CZ analyzed the data. CW, XZ and CL wrote the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

This study was conducted after the acquisition of written informed consent from the participating patients and upon the approval by the ethics committee of Tianjin Chest Hospital (2017-KY-004). The study protocol was approved by the local institutional review boards in accordance with the Declaration of Helsinki.

Not applicable.

This work was supported by National Natural Science Foundation of China (62206197), Applied and Basic Research by Multi-input Foundation of Tianjin (21JCYBJC00820), Tianjin Health Research Project (TJWJ2022QN067), Tianjin Key Laboratory of Cardiovascular Emergency and Critical Care certified by Tianjin Municipal Science and Technology Bureau, Tianjin Medical Discipline Construction Project, Tianjin Key Research Program of Traditional Chinese Medicine (2023006) and Committee on Science and Technology, Jinnan District, Tianjin (20220108).

The authors declare no conflict of interest.