Original ResearchFree Access

Association between Coronary Artery Disease Reporting and Data System–recommended Post–Coronary CT Angiography Management and Clinical Outcomes in Patients with Stable Chest Pain from a Chinese Registry

Published Online:https://doi.org/10.1148/radiol.222965

Abstract

Background

Coronary Artery Disease Reporting and Data System (CAD-RADS) was developed to standardize and optimize disease management in patients after coronary CT angiography (CCTA), but the impact of CAD-RADS management recommendations on clinical outcomes remains unclear.

Purpose

To retrospectively assess the association between the appropriateness of post-CCTA management according to CAD-RADS version 2.0 and clinical outcomes.

Materials and Methods

From January 2016 to January 2018, consecutive participants with stable chest pain referred for CCTA were prospectively included in a Chinese registry and followed for 4 years. Retrospectively, CAD-RADS 2.0 classification and the appropriateness of post-CCTA management were determined. Propensity score matching (PSM) was used to adjust for confounding variables. Hazard ratios (HRs) for a major adverse cardiovascular event (MACE), relative risks for invasive coronary angiography (ICA), and the corresponding number needed to treat were estimated.

Results

Of the 14 232 included participants (mean age, 61 years ± 13 [SD]; 8852 male), 2330, 2756, and 2614 were retrospectively categorized in CAD-RADS 1, 2, and 3, respectively. Only 26% of participants with CAD-RADS 1–2 disease and 20% with CAD-RADS 3 received appropriate post-CCTA management. After PSM, appropriate post-CCTA management was associated with lower risk of MACEs (HR, 0.34; 95% CI: 0.22, 0.51; P < .001), corresponding to a number needed to treat of 21 in CAD-RADS 1–2 but not CAD-RADS 3 (HR, 0.86; 95% CI: 0.49, 1.85; P = .42). Appropriate post-CCTA management was associated with decreased use of ICA in CAD-RADS 1–2 (relative risk, 0.40; 95% CI: 0.29, 0.55; P < .001) and 3 (relative risk, 0.33; 95% CI: 0.28, 0.39; P < .001), resulting in a number needed to treat of 14 and 2, respectively.

Conclusion

In this retrospective secondary analysis, appropriate disease management after CCTA according to CAD-RADS 2.0 was associated with lower risk of MACEs and more prudent use of ICA.

ClinicalTrials.gov registration no. NCT04691037

© RSNA, 2023

Supplemental material is available for this article.

See also the editorial by Leipsic and Tzimas in this issue.

Summary

Appropriate post–coronary CT angiography management based on the Coronary Artery Disease Reporting and Data System was associated with lower risk of major adverse cardiovascular events and less invasive procedures in a Chinese registry.

Key Results

  • ■ In this retrospective analysis of a prospective registry of 14 232 participants referred to coronary CT angiography (CCTA) for stable chest pain, 26% and 20% of those with Coronary Artery Disease Reporting and Data System (CAD-RADS) categories 1–2 and 3, respectively, received appropriate post-CCTA management.

  • ■ Appropriate post-CCTA management according to CAD-RADS version 2.0 was associated with lower risk of events in CAD-RADS 1–2 (hazard ratio, 0.34; P < .001) and decreased use of invasive coronary angiography in CAD-RADS 1–2 (relative risk, 0.40; P < .001) and 3 (relative risk, 0.33; P < .001).

Introduction

In recent randomized controlled trials (RCTs) (13), an initial coronary CT angiography (CCTA) strategy was associated with similar or better outcomes compared with either an initial strategy of invasive coronary angiography (ICA) or noninvasive functional testing. Accordingly, CCTA has been recommended as the first-line modality by current guidelines (4,5), and the clinical use of CCTA has increased dramatically in patients with stable chest pain (6,7).

For any imaging test, it is the link between identified abnormalities and alteration of management strategies that leads to improved outcomes, and not the test per se (79). However, numerous studies have reported variability in management after CCTA (1015), which might partly account for the suboptimal outcomes in the CCTA arms of recent randomized controlled trials (RCTs) (13,16). As CCTA is increasingly being used on a global scale, it is fundamental and critical to standardize and optimize post-CCTA management (7,14). For this purpose, the Coronary Artery Disease Reporting and Data System (CAD-RADS) was developed and has been regularly updated to establish consistent management after CCTA according to CAD-RADS classification, which considers the highest-grade coronary artery lesion and overall plaque burden detected at CCTA (17).

Although recent studies have supported the prognostic value of CAD-RADS, a high rate of ICA without optimal medical therapy was also identified, implying inadequate application of CAD-RADS recommendations in real-world clinical practice (11,12,18). Moreover, some studies have demonstrated the impact of CAD-RADS on therapeutic decision-making, such as escalation or initiation of statin medications, but the impact of CAD-RADS on clinical outcomes still remains unclear (19). Consequently, the aim of the present study was to retrospectively investigate the rates of appropriate post-CCTA management according to CAD-RADS categories in a clinical setting and assess the association between appropriateness of post-CCTA management and clinical outcomes in patients with stable chest pain.

Materials and Methods

Study Participants

The CCTA Improves Clinical Management of Stable Chest Pain registry is a prospective and ongoing cohort of patients who were referred for CCTA as first-line imaging testing for the assessment of stable chest pain (ClinicalTrials.gov registration no. NCT04691037). Details about the registry have been previously described (20,21) and are presented in Appendix S1. Briefly, from January 2016 to January 2018, 17 411 individuals referred for CCTA at Tianjin Chest Hospital and Beijing Chaoyang Hospital were considered (Fig 1). Patients were excluded for acute coronary syndrome, previous coronary artery disease, insufficient image quality, missing baseline data, non–sinus rhythm, structural heart disease, heart failure, or age greater than 90 years. Of the 14 232 participants finally enrolled, 4207 have been reported in previous studies (20,21). These prior articles dealt with development and comparison of different risk assessment models and strategies before CCTA, whereas in this article, the association between appropriateness of post-CCTA management according to CAD-RADS and resultant 4-year clinical outcomes are reported. The present study was a secondary analysis of prospectively collected data, conducted in accordance with the Declaration of Helsinki, approved by the Ethics Committees of local institutions, and all participants gave informed consent.

Flowchart shows inclusion of study participants, Coronary Artery                         Disease Reporting and Data System (CAD-RADS) categorizations, and the number                         of participants receiving appropriate and inappropriate management after                         coronary CT angiography (CCTA) according to CAD-RADS version 2.0                         recommendations. ACS = acute coronary syndrome, CAD = coronary artery                         disease, NYHA = New York Heart Association, PSM = propensity scoring                         matching.

Figure 1: Flowchart shows inclusion of study participants, Coronary Artery Disease Reporting and Data System (CAD-RADS) categorizations, and the number of participants receiving appropriate and inappropriate management after coronary CT angiography (CCTA) according to CAD-RADS version 2.0 recommendations. ACS = acute coronary syndrome, CAD = coronary artery disease, NYHA = New York Heart Association, PSM = propensity scoring matching.

Imaging Procedure, CAD-RADS Classification, and Post-CCTA Management

CCTA examinations were performed based on established guidelines (17) and local protocols, as previously described (20,21). All CCTA images were analyzed by three experienced local observers, two cardiovascular radiologists (H.Z. and S.L., with 25 and 20 years of cardiac imaging experience) and one cardiologist (Y.T., with 8 years of cardiac imaging experience). At image analysis, all segments greater than or equal to 2 mm in diameter were identified and analyzed using CAD-RADS (17). The percent luminal stenosis in every segment was categorized as 0% (no stenosis), 1%–24% (minimal stenosis), 25%–49% (mild stenosis), 50%–69% (moderate stenosis), 70%–99% (severe stenosis), and 100% (total occlusion). Interobserver disagreements were resolved by consensus. Initially, the CCTA outcome was classified according to local protocols, and further management, based on CCTA results, was left to the discretion of the referring physician.

Retrospectively, classification was determined using CAD-RADS version 2.0 recommendations (17). CAD-RADS classification was based on the highest-grade stenosis demonstrated with CCTA (17). Modifiers for graft (G), stent (S), nondiagnostic segment (N), vulnerable plaque (HRP), or ischemia (I) in CAD-RADS 2.0 (17) were not incorporated in the present study because we restricted analysis to participants without previous coronary artery disease, revascularization, or nondiagnostic images. Participants were not routinely referred for assessment of vulnerable plaque and noninvasive functional testing, such as fractional flow reserve derived from CCTA, myocardial CT perfusion, and stress testing.

Recommendations for management after CCTA according to CAD-RADS category were retrospectively based on the updated 2.0 guidelines (17) as follows. For CAD-RADS 0, no additional diagnostic testing or medical therapy was needed. For CAD-RADS 1 or 2, optimal medical therapy should be considered. For CAD-RADS 3, participants should be treated with optimal medical therapy and assessed using noninvasive functional testing, with subsequent ICA reserved for participants who had positive or inconclusive noninvasive functional testing results or remained symptomatically unresponsive to maximization of medical therapy. For CAD-RADS 4A, optimal medical therapy was needed, and ICA or noninvasive functional testing and options for revascularization based on ICA should be considered. For CAD-RADS 4B and 5, both medical therapy and options for revascularization based on ICA were needed. Further details of CAD-RADS classification are included in Appendix S1.

Clinical Data and Definitions at Baseline

Baseline characteristics that included age, sex, symptoms, body mass index, hypertension, hyperlipidemia, diabetes, smoking status, family history of premature coronary artery disease, renal insufficiency, abnormal electrocardiographic results, cerebrovascular disease, left ventricular ejection fraction, peripheral artery disease, and baseline medications were collected through contact with participants or review of electronic medical records. See Appendix S1 for details.

The present study was a secondary analysis of prospectively collected data, with initial post-CCTA management determined based on previously evaluated CCTA images and clinical practice at the local institution. Three cardiologists (C. Li, M.W., and E.Z., with 25, 15, and 8 years of cardiology experience, respectively), who were blinded to other clinical characteristics, retrospectively and independently adjudicated the appropriateness of post-CCTA management for each participant according to the following criteria: (a) no additional diagnostic testing or medical therapy for CAD-RADS 0; (b) optimal medical therapy (OMT) for CAD-RADS 1 and 2 based on plaque burden; (c) OMT and noninvasive functional testing (first) and ICA (only secondarily) after positive or inconclusive noninvasive functional testing results or severe symptoms refractory to maximization of OMT for CAD-RADS 3; (d) OMT, ICA, or noninvasive functional testing and options for revascularization for CAD-RADS 4A; and (e) OMT and revascularization based on ICA for CAD-RADS 4B or 5. Interobserver disagreements were resolved by consensus.

Follow-up for Clinical Outcomes

After undergoing CCTA, all participants were followed for 4 years. The primary outcome was a major adverse cardiovascular event (MACE), defined as a composite of all-cause death and nonfatal myocardial infarction. All-cause death was used rather than cardiac death to eliminate the need for possibly difficult adjudication of causes of death, especially given the relatively low mortality rate, but the causes of death in the registry were also investigated. The secondary outcome included ICA use, referral to revascularization, and unnecessary use of ICA. To measure the efficiency of ICA use, the diagnostic yield of ICA was defined as (1 minus the number of unnecessary ICA uses divided by the number of ICA uses) × 100%. Additional information is presented in Appendix S1.

Statistical Analysis

All statistical analyses were performed in R (version 4.0.3; The R Foundation), SAS (version 9.3; SAS Institute), SPSS (version 26; IBM), or MedCalc (version 15.2.2; MedCalc Software) by J. Zhou, H.W., Y.H., and P.Z. Two-tailed P < .05 was considered indicative of a statistically significant difference. The one-sample Kolmogorov-Smirnov test was used to check the assumption of normal distribution. The Student t test was used to compare normally distributed continuous data, and the Mann-Whitney U test was used to compare nonnormally distributed continuous data. Categorical variables were compared using the χ2 test or Fisher exact test, as appropriate.

Given the observational nature of this study, two statistical measures were employed to account for nonrandom allocation of participants to the appropriate or inappropriate post-CCTA management groups; these were multivariable adjustment and propensity score matching (PSM). The a priori analysis plan included selection of covariates based on previous similar studies (11) and clinical relevance. The main analysis was with PSM, and data reporting was based on the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist for cohort studies and special reporting guidelines for PSM. Appendix S1 gives additional details about the statistical analyses.

Kaplan-Meier curves for cumulative event-free survival were constructed from whichever of the following occurred first: end point of concern (eg, MACE, ICA), death, the end of the 4-year follow-up period, or loss to follow-up. The corresponding P values were estimated using the log-rank test. Cox proportional hazard regressions were used to calculate hazard ratios (HRs) and 95% CIs, which assessed the appropriateness of post-CCTA management to the time of the first MACE (or censoring) in total and for subgroups of participants. Interaction analyses were conducted in relative subgroups to assess whether the association between the appropriateness of post-CCTA management and the primary end point varied for participants with different risk profiles (detailed in Fig S2). Also, to address unmeasured confounding, E value analysis was conducted to assess the robustness of the association between appropriateness of post-CCTA management and clinical outcomes (22). All model assumptions were examined, including linearity, collinearity, additivity, and proportional hazards. HRs or relative risks of study end points with 95% CIs are illustrated in forest plots. The number needed to treat is reported as the inverse of the absolute risk reduction (23).

Results

Baseline Characteristics and MACEs according to CAD-RADS Category

As illustrated in Figure 1, after excluding participants with acute coronary syndrome (n = 1045), previous coronary artery disease (n = 875), insufficient image quality (n = 573), missing baseline data (n = 228), non–sinus rhythm (n = 187), structural heart disease (n = 122), heart failure (n = 97), or age greater than 90 years (n = 52), 14 232 participants were finally included in this study (mean age, 61 years ± 13 [SD]; 8852 male). Of the 14 232 participants, 3636 (26%) were categorized as CAD-RADS 0, 2330 (16%) as CAD-RADS 1, 2756 (19%) as CAD-RADS 2, 2614 (18%) as CAD-RADS 3, 1885 (13%) as CAD-RADS 4A, 582 (4%) as CAD-RADS 4B, and 429 (3%) as CAD-RADS 5. Except for cerebrovascular disease, all other baseline characteristics were significantly different across CAD-RADS categories (P < .05 for all comparisons) (Table 1). During the 4-year follow-up, 6% (852 of 14 232) of participants experienced a MACE; 1% (146 of 14 232) of participants died and 5% (706 of 14 232) experienced nonfatal myocardial infarction. Among the 146 deaths, 59 were due to fatal myocardial infarction or other cardiac causes. The rates of MACEs were significantly different across CAD-RADS categories (P < .001) (Table 1).

Table 1: Baseline Participant Characteristics Stratified according to CAD-RADS Category

Table 1:

Appropriateness of Post-CCTA Management according to CAD-RADS

Figure 1 shows the number of participants who received appropriate and inappropriate post-CCTA management according to CAD-RADS 2.0. Most participants with CAD-RADS 1 (81%, 1880 of 2330 [95% CI: 79, 82]), 2 (69%, 1900 of 2756 [95% CI: 67, 71]), and 3 (80%, 2086 of 2614 [95% CI: 78, 81]) disease received inappropriate post-CCTA management. Alternatively, the percentage of participants rated as receiving inappropriate post-CCTA management was low in CAD-RADS 0 (1%, 46 of 3636 [95% CI: 0.96, 1.72]), 4A (8%, 148 of 1885 [95% CI: 6.63, 9.10]), 4B (6%, 37 of 582 [95% CI: 4.38, 8.48]), and 5 (5%, 20 of 429 [95% CI: 2.82, 7.04]). Inappropriate post-CCTA management included immediate ICA after CCTA for CAD-RADS 3 (75%, 1950 of 2614 [95% CI: 73, 77]) and suboptimal medical therapy for CAD-RADS 1 (78%, 1825 of 2330 [95% CI: 76, 80]) and CAD-RADS 2 (66%, 1830 of 2756 [95% CI: 64, 68]). PSM analysis was conducted in two groups; (a) CAD-RADS 1 and 2 were merged as one group for the similar appropriate and inappropriate management criterion (1306 participants in the appropriate group for receiving optimal medical therapy vs 3780 participants in the inappropriate group for receiving suboptimal medical therapy) and (b) CAD-RADS 3 (528 participants in the appropriate group vs 2086 participants in the inappropriate group) (Fig 1).

Association between Appropriate Post-CCTA Management and Clinical Outcomes in CAD-RADS 1 and 2

Prior to PSM, we observed differences in baseline characteristics between appropriate and inappropriate post-CCTA management groups. The HRs for MACEs when comparing appropriate versus inappropriate post-CCTA management were 0.78 (95% CI: 0.58, 1.03; P = .25) and 0.39 (95% CI: 0.11, 0.67; P < .001), respectively, after multivariable adjustment. The E value for unmeasured confounding was 4.57. As shown in Table S1, 1229 participants categorized as CAD-RADS 1–2 were inpatients and 3857 were outpatients. When comparing inpatients to outpatients, the probability of receiving appropriate post-CCTA management was higher for inpatients (30% vs 24%, P < .001) and the rates of MACEs were similar (6% vs 4%, P = .06).

For PSM analysis in CAD-RADS 1–2, 1028 participant pairs were successfully matched (Table 2). The logistic regression model for predicting appropriateness of post-CCTA management revealed that participants who had a greater risk profile and higher burden of coronary artery disease were more likely to receive appropriate post-CCTA management (Table S2). Follow-up details before and after PSM are presented in Table S3. After PSM, appropriate post-CCTA management was associated with lower risk of MACEs than inappropriate management (3% vs 8%, log-rank P < .001) (Table S3, Fig 2A). The imbalance in diabetes after PSM was adjusted and the HR comparing MACEs in the appropriate versus inappropriate post-CCTA management groups was 0.34 (95% CI: 0.22, 0.51; P < .001) (Fig 3), corresponding to a number needed to treat of 21 and E value of 5.33. The results of the prespecified subgroup analyses are presented in Figure S1.

Table 2: Baseline Participant Characteristics for CAD-RADS Categories 1 and 2 Stratified according to Appropriate or Inappropriate Management before and after Propensity Score Matching

Table 2:
Kaplan-Meier curves of propensity score–matched participants                         for (A) Coronary Artery Disease Reporting and Data System (CAD-RADS) 1 and 2                         and (B) CAD-RADS 3 show cumulative event-free survival in the inappropriate                         (dashed lines) and appropriate (solid lines) post–coronary CT                         angiography (CCTA) management groups. Compared with inappropriate post-CCTA                         management, appropriate management was associated with lower risk of major                         adverse cardiovascular events (log-rank P < .001) in CAD-RADS                         1–2 (A) but similar risk in CAD-RADS 3 (B) (log-rank P =                         .23).

Figure 2: Kaplan-Meier curves of propensity score–matched participants for (A) Coronary Artery Disease Reporting and Data System (CAD-RADS) 1 and 2 and (B) CAD-RADS 3 show cumulative event-free survival in the inappropriate (dashed lines) and appropriate (solid lines) post–coronary CT angiography (CCTA) management groups. Compared with inappropriate post-CCTA management, appropriate management was associated with lower risk of major adverse cardiovascular events (log-rank P < .001) in CAD-RADS 1–2 (A) but similar risk in CAD-RADS 3 (B) (log-rank P = .23).

Forest plot shows the association between appropriateness of                         post–coronary CT angiography (CCTA) management and major adverse                         cardiovascular events (MACEs) in propensity score–matched                         participants. Hazard ratios (HRs) with 95% CIs are indicated by dots and                         corresponding horizontal lines, respectively. The HRs comparing MACEs for                         the appropriate versus inappropriate post-CCTA management groups were 0.34                         (95% CI: 0.22, 0.51; P < .001) and 0.86 (95% CI: 0.49, 1.85; P = .42)                         in Coronary Artery Disease Reporting and Data System (CAD-RADS) categories                         1–2 and 3, respectively. MI = myocardial infarction.

Figure 3: Forest plot shows the association between appropriateness of post–coronary CT angiography (CCTA) management and major adverse cardiovascular events (MACEs) in propensity score–matched participants. Hazard ratios (HRs) with 95% CIs are indicated by dots and corresponding horizontal lines, respectively. The HRs comparing MACEs for the appropriate versus inappropriate post-CCTA management groups were 0.34 (95% CI: 0.22, 0.51; P < .001) and 0.86 (95% CI: 0.49, 1.85; P = .42) in Coronary Artery Disease Reporting and Data System (CAD-RADS) categories 1–2 and 3, respectively. MI = myocardial infarction.

Compared with the inappropriate post-CCTA management group, the use of invasive procedures was higher in the appropriate management group. More specifically, the relative risk was 0.40 (95% CI: 0.29, 0.55; P < .001) for ICA, 0.46 (95% CI: 0.28, 0.75; P = .003) for revascularization, and 0.37 (95% CI: 0.24, 0.56; P = .001) for unnecessary use of ICA, corresponding with a number needed to treat of 14, 40, and 22, respectively (Fig 4). The diagnostic yield of ICA in the appropriate and inappropriate post-CCTA management groups was comparable (P = .78). Appropriate post-CCTA management appears to have reduced the use of ICA during the period from the 3rd to 30th month after CCTA (log-rank P < .001) (Fig S2). A representative case of inappropriate post-CCTA management and clinical outcome in CAD-RADS 2 is presented in Figure 5A and 5B.

Forest plot shows the association between the appropriateness of                         management after coronary CT angiography (CCTA) and invasive procedures in                         propensity score–matched participants. Relative risks with 95% CIs                         are indicated by dots and corresponding horizontal lines, respectively. The                         relative risk values comparing invasive coronary angiography (ICA) for the                         appropriate versus inappropriate post-CCTA management groups were 0.40 (95%                         CI: 0.29, 0.55; P < .001) and 0.33 (95% CI: 0.28, 0.39; P <                         .001) in Coronary Artery Disease Reporting and Data System (CAD-RADS)                         categories 1–2 and 3, respectively.

Figure 4: Forest plot shows the association between the appropriateness of management after coronary CT angiography (CCTA) and invasive procedures in propensity score–matched participants. Relative risks with 95% CIs are indicated by dots and corresponding horizontal lines, respectively. The relative risk values comparing invasive coronary angiography (ICA) for the appropriate versus inappropriate post-CCTA management groups were 0.40 (95% CI: 0.29, 0.55; P < .001) and 0.33 (95% CI: 0.28, 0.39; P < .001) in Coronary Artery Disease Reporting and Data System (CAD-RADS) categories 1–2 and 3, respectively.

Representative cases for inappropriate management after coronary CT                         angiography (CCTA) based on Coronary Artery Disease Reporting and Data                         System (CAD-RADS) version 2.0. (A, B) Coronary CT angiogram (A) in a                         48-year-old male participant with diabetes and hypertension who was referred                         for CCTA due to nonanginal chest pain shows a proximal lesion in the right                         coronary artery with mild stenosis (arrow), indicating CAD-RADS category 2.                         However, he was not prescribed aspirin or statins after the CCTA examination                         and he experienced myocardial infarction 28 months later. Invasive coronary                         angiogram (B) shows a total occlusion stenosis with thrombus at the site of                         the originally mild stenosis in the proximal segment of the right coronary                         artery (arrow). (C, D) Coronary CT angiogram (C) in a 71-year-old female                         participant with hypertension who was referred for CCTA due to atypical                         angina shows a proximal lesion in the left anterior descending artery with                         moderate stenosis (arrow), indicating CAD-RADS category 3. The participant                         strongly demanded an immediate invasive coronary angiography (ICA)                         examination after CCTA. Invasive coronary angiogram (D) shows a proximal                         lesion in the left anterior descending artery with luminal stenosis of 60%                         (arrow). After ICA, the participant received aspirin, statin, amlodipine,                         β-blocker, and nitrate medications and subsequently had no clinical                         events during the 4-year follow-up.

Figure 5: Representative cases for inappropriate management after coronary CT angiography (CCTA) based on Coronary Artery Disease Reporting and Data System (CAD-RADS) version 2.0. (A, B) Coronary CT angiogram (A) in a 48-year-old male participant with diabetes and hypertension who was referred for CCTA due to nonanginal chest pain shows a proximal lesion in the right coronary artery with mild stenosis (arrow), indicating CAD-RADS category 2. However, he was not prescribed aspirin or statins after the CCTA examination and he experienced myocardial infarction 28 months later. Invasive coronary angiogram (B) shows a total occlusion stenosis with thrombus at the site of the originally mild stenosis in the proximal segment of the right coronary artery (arrow). (C, D) Coronary CT angiogram (C) in a 71-year-old female participant with hypertension who was referred for CCTA due to atypical angina shows a proximal lesion in the left anterior descending artery with moderate stenosis (arrow), indicating CAD-RADS category 3. The participant strongly demanded an immediate invasive coronary angiography (ICA) examination after CCTA. Invasive coronary angiogram (D) shows a proximal lesion in the left anterior descending artery with luminal stenosis of 60% (arrow). After ICA, the participant received aspirin, statin, amlodipine, β-blocker, and nitrate medications and subsequently had no clinical events during the 4-year follow-up.

Association between Appropriate Post-CCTA Management and Clinical Outcomes in CAD-RADS 3

Except for medication use, baseline characteristics differed between the appropriate and inappropriate post-CCTA management groups in CAD-RADS 3 before PSM (Table 3). The HRs for MACEs when comparing appropriate with inappropriate post-CCTA management were 0.67 (95% CI: 0.34, 1.03; P = .09) and 0.79 (95% CI: 0.32, 1.27; P = .18), respectively, after multivariable adjustment. As shown in Table S1, 1038 participants were inpatients and 1576 were outpatients. When comparing inpatients to outpatients, the probability of receiving appropriate post-CCTA management was lower (16% vs 23%, P < .001) and the rates of MACEs were similar (4% vs 3%, P = .83).

Table 3: Baseline Participant Characteristics for CAD-RADS Category 3 Stratified according to Appropriate or Inappropriate Management before and after Propensity Score Matching

Table 3:

For PSM analysis in CAD-RADS 3, 340 participant pairs were successfully matched (Table 3). The logistic regression model for predicting appropriateness of post-CCTA management in CAD-RADS 3 revealed that participants who had a greater risk profile and higher burden of coronary artery disease were associated with inappropriate post-CCTA management (Table S4). Follow-up details before and after matching are presented in Table S5. After PSM, we did not observe a difference in the occurrence of MACEs in the appropriate versus inappropriate post-CCTA management groups (log-rank P = .23) (Fig 2B). The imbalance in typical and atypical anginal symptoms between groups after PSM was adjusted and the HR for MACEs was 0.86 (95% CI: 0.49, 1.85; P = .42) (Fig 3). The results of the prespecified subgroup analyses are presented in Figure S1.

Compared with the inappropriate post-CCTA management group, the use of invasive procedures was reduced in the appropriate group. Specifically, the relative risk was 0.33 (95% CI: 0.28, 0.39; P < .001) for ICA, 0.55 (95% CI: 0.38, 0.79; P = .03) for revascularization, and 0.27 (95% CI: 0.22, 0.34; P < .001) for unnecessary ICA, resulting in a number needed to treat of 2, 11, and 2, respectively (Fig 4). The diagnostic yield of ICA in the appropriate post-CCTA management group (36% [95% CI: 27, 46]) was also higher than that of the inappropriate management group (22% [95% CI: 17, 27], P < .001). A representative case of inappropriate post-CCTA management and clinical outcome in CAD-RADS 3 is presented in Figure 5C and 5D.

Discussion

In this retrospective analysis of a prospective registry, consecutive participants with stable chest pain were included and the appropriateness of management after coronary CT angiography (CCTA) based on Coronary Artery Disease Reporting and Data System (CAD-RADS) categorization was assessed. We found that suboptimal medical therapy for CAD-RADS 1–2 and immediate invasive coronary angiography (ICA) after CCTA for CAD-RADS 3 were main factors for suboptimal post-CCTA management according to CAD-RADS 2.0 recommendations. Appropriate post-CCTA management was associated with a reduction of major adverse cardiovascular events (MACEs) and subsequent invasive procedures in both CAD-RADS 1 and 2. Additionally, appropriate post-CCTA management was associated with decreased use of ICA and a 64% increase in diagnostic yield of ICA, without excess risk of a MACE, in CAD-RADS 3. Thus, compared with inappropriate post-CCTA management, management based on CAD-RADS 2.0 may yield greater clinical benefit in patients who undergo CCTA for the assessment of stable chest pain.

Several large-scale clinical studies found that CAD-RADS provided discriminatory value for prediction of cardiac events presenting as a “dose-response phenomenon” (11,12,18), which is supported by data in this study showing the stepwise increase in MACE rates across CAD-RADS categories. We also found that nearly 80% of participants with CAD-RADS 3 received inappropriate post-CCTA management, mainly due to immediate ICA after CCTA. This finding is in agreement with the frequent use of ICA for patients who were asymptomatic or not receiving optimal medical therapy in the CONFIRM (Coronary CT Angiography Evaluation for Clinical Outcomes: An International Multicenter Registry) study (11), and the relatively high rate of unnecessary use of ICA observed in our previous studies (20,21). The present study further demonstrated that appropriate post-CCTA management in CAD-RADS 3 may avoid unnecessary invasive procedures and led to improved diagnostic yield of ICA at no expense of excess MACEs. Moreover, the major discrepancy between the Kaplan-Meier curves for the appropriate and inappropriate post-CCTA management groups occurred in the first few months after CCTA, whereas the two curves ran parallel throughout the follow-up. This implies that post-CCTA management may reduce the risk of invasive procedure–related events, as demonstrated in recent RCTs that compared invasive and conservative strategies (1,24).

A literature review suggested that most MACEs occur in patients with normal noninvasive functional testing results and nonsignificant coronary artery disease detected with CCTA (25). In the SCOT-HEART (Scottish CT of the Heart) trial, the CCTA arm demonstrated a lower rate of MACEs than the traditional care arm, which was mainly attributed to greater intensity of optimal medical therapy in response to visualizing (mostly nonsignificant) coronary artery disease (10). Similarly, results from another large cohort of consecutive patients with nonobstructive coronary artery disease as determined with CCTA also found that statin therapy prescribed after CCTA was associated with a reduction of MACEs over a 5-year follow-up, leading to a number needed to treat between 13 and 36 (13). In the present study, the corresponding number needed to treat is 21 and, notably, use of cardiac catheterization was generally not based on CCTA results but rather was a clinical requirement decided during follow-up. This suggests that appropriate management after CCTA may effectively relieve symptoms and prevent progression of coronary artery disease.

Participants in the CCTA arm of several recent RCTs had similar rates of MACEs over a short-term follow-up and underwent immediate ICA examination more often compared with other modalities (13), resulting in some controversy regarding initial use of CCTA (7,9). Importantly, however, the decisions of management after the initial imaging test in these RCTs were left to the local physicians (13). Thus, the extent to which participants received appropriate post-CCTA management remains uncertain. The higher MACE rates and increase in redundant invasive procedures associated with inappropriate post-CCTA management in the present study may partly account for the suboptimal clinical outcomes observed in CCTA arms in previous RCTs (13,9,16,26).

Several limitations of the present study merit discussion. First, it is important to recognize that the current analysis is not a prospective study on the impact of CAD-RADS 2.0 categorization on management and outcomes, but a retrospective analysis assessing whether the management of CCTA results at the original time point correlates with CAD-RADS 2.0 recommendations and its relationship to outcomes. Second, confounding by indication was a major concern for interpretation of our results; participants with a greater risk profile and burden of coronary artery disease were more likely to receive aggressive interventions, resulting in a higher likelihood of appropriate post-CCTA management in CAD-RADS 1 and 2 and inappropriate management in CAD-RADS 3. We tried to control for this confounding with a large range of prognostic variables, but our findings could be biased by incomplete matching and unmeasured confounding variables. However, the results of the multivariable adjustment that included all participants remained consistent with the main analyses using PSM. Meanwhile, to overturn the effect of appropriate post-CCTA management, an unmeasured confounding variable would need to have an association with both appropriate post-CCTA management and MACE equivalent E values of 4.57–5.33 in CAD-RADS 1 and 2, after controlling for all measured confounding variables. Third, we restricted PSM to participants with CAD-RADS 1–3 categorization because the number of participants receiving inappropriate post-CCTA management was low in the other categories, which made the analyses clinically nonsignificant and statistically difficult. The absolute clinical benefit of an initial invasive strategy in low-risk subgroups of CAD-RADS 3–5, such as no left main disease (24) or hemodynamically significant stenoses (26), remains debatable and warrants further investigation (27). Fourth, this study only included participants from two centers located in the same region, and thus unmeasured confounding factors related to cohorts in different geographic regions may impact our results (28). Finally, this study only included participants with new-onset stable chest pain so that the conclusions should not be extrapolated to those with known coronary artery disease with acute or no chest pain.

In conclusion, the rates of appropriate management after coronary CT angiography (CCTA) according to Coronary Artery Disease Reporting and Data System (CAD-RADS) version 2.0 recommendations were low in participants with nonsignificant coronary artery disease (CAD-RADS categories 1–3). Appropriate post-CCTA management informed by CAD-RADS might reduce major adverse cardiovascular events and redundant or unnecessary use of invasive procedures in patients with stable chest pain. The actual influence of appropriate post-CCTA management needs to be comprehensively validated in rigorously designed pragmatic and cost-effectiveness trials.

Disclosures of conflicts of interest: J. Zhou No relevant relationships. C. Li No relevant relationships. H.Z. No relevant relationships. C. Liu No relevant relationships. J.Y. No relevant relationships. J. Zhao No relevant relationships. Y. Hou No relevant relationships. Y.T. No relevant relationships. H.W. No relevant relationships. Y.L. No relevant relationships. C.X. No relevant relationships. M.W. No relevant relationships. C.W. No relevant relationships. E.Z. No relevant relationships. S.W. No relevant relationships. P.Z. No relevant relationships. D.S. No relevant relationships. S.L. No relevant relationships. Y.G. No relevant relationships. Y. Huo No relevant relationships. H.C. No relevant relationships. Z.G. No relevant relationships. Y.C. No relevant relationships.

Author Contributions

Author contributions: Guarantors of integrity of entire study, J. Zhou, H.Z., S.L.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, J. Zhou, C. Li, H.Z., C. Liu, J.Y., J. Zhao, Y.T., H.W., Y.L., C.X., M.W., C.W., E.Z., S.W., P.Z., D.S., S.L., Y.G., H.C., Z.G., Y.C.; clinical studies, J. Zhou, C. Li, H.Z., C. Liu, J.Y., J. Zhao, Y.T., H.W., Y.L., C.X., M.W., C.W., E.Z., S.W., D.S., S.L., Y.G., Y. Huo, H.C., Z.G., Y.C.; experimental studies, J. Zhou, H.Z., Y. Hou, S.L.; statistical analysis, J. Zhou, H.Z., Y. Hou, H.W., P.Z., S.L.; and manuscript editing, J. Zhou, H.Z., S.L.

* J. Zhou and C. Li contributed equally to this work.

** Z.G. and Y.C. are co-senior authors.

This work was supported by the National Natural Science Foundation of China (62206197, 62106160), Applied and Basic Research by Multi-input Foundation of Tianjin (21JCYBJC00820), Tianjin Health Research Project (TJWJ2022QN067), National Key Research and Development Program of China (2016YFC1300300), Tianjin Key Laboratory of Cardiovascular Emergency and Critical Care certified by Tianjin Municipal Science and Technology Bureau, Tianjin Medical Discipline Construction Project, Tianjin Health Science and Technology Project (MS20015), and Tianjin Key Research Program of Traditional Chinese Medicine (2022001, 2023006).

Data sharing: Data generated or analyzed during the study are available from the corresponding author by request.

References

  • 1. DISCHARGE Trial Group; Maurovich-Horvat P, Bosserdt M, et al. CT or invasive coronary angiography in stable chest pain. N Engl J Med 2022;386(17):1591–1602. Crossref, MedlineGoogle Scholar
  • 2. SCOT-HEART Investigators; Newby DE, Adamson PD, et al. Coronary CT Angiography and 5-Year Risk of Myocardial Infarction. N Engl J Med 2018;379(10):924–933. Crossref, MedlineGoogle Scholar
  • 3. Douglas PS, Hoffmann U, Patel MR, et al. Outcomes of anatomical versus functional testing for coronary artery disease. N Engl J Med 2015;372(14):1291–1300. Crossref, MedlineGoogle Scholar
  • 4. Knuuti J, Wijns W, Saraste A, et al. 2019 ESC Guidelines for the diagnosis and management of chronic coronary syndromes. Eur Heart J 2020;41(3):407–477. Crossref, MedlineGoogle Scholar
  • 5. Writing Committee Members; Gulati M, Levy PD, et al. 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR Guideline for the Evaluation and Diagnosis of Chest Pain: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol 2021;78(22):e187–e285. Crossref, MedlineGoogle Scholar
  • 6. Dai YX, Ge JB. Precision medicine in coronary artery disease: Time for implementation into practice. Cardiol Plus 2021;6(1):1–3. CrossrefGoogle Scholar
  • 7. Genders TS, Ferket BS, Hunink MG. The Quantitative Science of Evaluating Imaging Evidence. JACC Cardiovasc Imaging 2017;10(3):264–275. Crossref, MedlineGoogle Scholar
  • 8. Hecht HS, Shaw L, Chandrashekhar YS, Bax JJ, Narula J. Should NICE guidelines be universally accepted for the evaluation of stable coronary disease? A debate. Eur Heart J 2019;40(18):1440–1453. Crossref, MedlineGoogle Scholar
  • 9. Shaw LJ, Phillips LM, Nagel E, Newby DE, Narula J, Douglas PS. Comparative Effectiveness Trials of Imaging-Guided Strategies in Stable Ischemic Heart Disease. JACC Cardiovasc Imaging 2017;10(3):321–334. Crossref, MedlineGoogle Scholar
  • 10. Adamson PD, Williams MC, Dweck MR, et al. Guiding Therapy by Coronary CT Angiography Improves Outcomes in Patients With Stable Chest Pain. J Am Coll Cardiol 2019;74(16):2058–2070. Crossref, MedlineGoogle Scholar
  • 11. Xie JX, Cury RC, Leipsic J, et al. The Coronary Artery Disease-Reporting and Data System (CAD-RADS): Prognostic and Clinical Implications Associated With Standardized Coronary Computed Tomography Angiography Reporting. JACC Cardiovasc Imaging 2018;11(1):78–89. Crossref, MedlineGoogle Scholar
  • 12. Bittner DO, Mayrhofer T, Budoff M, et al. Prognostic Value of Coronary CTA in Stable Chest Pain: CAD-RADS, CAC, and Cardiovascular Events in PROMISE. JACC Cardiovasc Imaging 2020;13(7):1534–1545. Crossref, MedlineGoogle Scholar
  • 13. Øvrehus KA, Diederichsen A, Grove EL, et al. Reduction of Myocardial Infarction and All-Cause Mortality Associated to Statins in Patients Without Obstructive CAD. JACC Cardiovasc Imaging 2021;14(12):2400–2410. Crossref, MedlineGoogle Scholar
  • 14. Honigberg MC, Lander BS, Baliyan V, et al. Preventive Management of Nonobstructive CAD After Coronary CT Angiography in the Emergency Department. JACC Cardiovasc Imaging 2020;13(2 Pt 1):437–448. Crossref, MedlineGoogle Scholar
  • 15. Chow BJ, Small G, Yam Y, et al. Prognostic and therapeutic implications of statin and aspirin therapy in individuals with nonobstructive coronary artery disease: results from the CONFIRM (COronary CT Angiography EvaluatioN For Clinical Outcomes: An InteRnational Multicenter registry) registry. Arterioscler Thromb Vasc Biol 2015;35(4):981–989. Crossref, MedlineGoogle Scholar
  • 16. Jørgensen ME, Andersson C, Nørgaard BL, et al. Functional Testing or Coronary Computed Tomography Angiography in Patients With Stable Coronary Artery Disease. J Am Coll Cardiol 2017;69(14):1761–1770. Crossref, MedlineGoogle Scholar
  • 17. Cury RC, Leipsic J, Abbara S, et al. CAD-RADS™ 2.0 - 2022 Coronary Artery Disease-Reporting and Data System: An Expert Consensus Document of the Society of Cardiovascular Computed Tomography (SCCT), the American College of Cardiology (ACC), the American College of Radiology (ACR), and the North America Society of Cardiovascular Imaging (NASCI). J Cardiovasc Comput Tomogr 2022;16(6):536–557. Crossref, MedlineGoogle Scholar
  • 18. Williams MC, Moss A, Dweck M, et al. Standardized reporting systems for computed tomography coronary angiography and calcium scoring: A real-world validation of CAD-RADS and CAC-DRS in patients with stable chest pain. J Cardiovasc Comput Tomogr 2020;14(1):3–11. Crossref, MedlineGoogle Scholar
  • 19. Hull RA, Berger JM, Boster JM, et al. Adoption of coronary artery disease - Reporting and Data System (CAD-RADS™) and observed impact on medical therapy and systolic blood pressure control. J Cardiovasc Comput Tomogr 2020;14(5):421–427. Crossref, MedlineGoogle Scholar
  • 20. Zhou J, Chen Y, Zhang Y, et al. Epicardial Fat Volume Improves the Prediction of Obstructive Coronary Artery Disease Above Traditional Risk Factors and Coronary Calcium Score. Circ Cardiovasc Imaging 2019;12(1):e008002. Crossref, MedlineGoogle Scholar
  • 21. Zhou J, Li C, Cong H, et al. Comparison of Different Investigation Strategies to Defer Cardiac Testing in Patients With Stable Chest Pain. JACC Cardiovasc Imaging 2022;15(1):91–104. Crossref, MedlineGoogle Scholar
  • 22. VanderWeele TJ, Ding P. Sensitivity Analysis in Observational Research: Introducing the E-Value. Ann Intern Med 2017;167(4):268–274. Crossref, MedlineGoogle Scholar
  • 23. Ludwig L, Darmon P, Guerci B. Computing and interpreting the Number Needed to Treat for Cardiovascular Outcomes Trials : Perspective on GLP-1 RA and SGLT-2i therapies. Cardiovasc Diabetol 2020;19(1):65. Crossref, MedlineGoogle Scholar
  • 24. Maron DJ, Hochman JS, Reynolds HR, et al. Initial Invasive or Conservative Strategy for Stable Coronary Disease. N Engl J Med 2020;382(15):1395–1407. Crossref, MedlineGoogle Scholar
  • 25. Arbab-Zadeh A, Fuster V. From Detecting the Vulnerable Plaque to Managing the Vulnerable Patient: JACC State-of-the-Art Review. J Am Coll Cardiol 2019;74(12):1582–1593. Crossref, MedlineGoogle Scholar
  • 26. Curzen N, Nicholas Z, Stuart B, et al. Fractional flow reserve derived from computed tomography coronary angiography in the assessment and management of stable chest pain: the FORECAST randomized trial. Eur Heart J 2021;42(37):3844–3852. Crossref, MedlineGoogle Scholar
  • 27. Lawton JS, Tamis-Holland JE, Bangalore S, et al. 2021 ACC/AHA/SCAI Guideline for Coronary Artery Revascularization: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation 2022;145(3):e18–e114. MedlineGoogle Scholar
  • 28. Hachamovitch R, Peña JM, Xie J, Shaw LJ, Min JK. Imaging Registries and Single-Center Series. JACC Cardiovasc Imaging 2017;10(3):276–285. Crossref, MedlineGoogle Scholar

Article History

Received: Nov 18 2022
Revision requested: Jan 25 2023
Revision received: Apr 11 2023
Accepted: Apr 20 2023
Published online: June 13 2023