Evaluation of stable chest pain following emergency department presentation: Impact of first‐line cardiac computed tomography diagnostic strategy in an Australian setting

International guidelines provide increasing support for computed tomography coronary angiography (CTCA) in investigating chest pain. A pathway utilising CTCA first‐line for outpatient stable chest pain evaluation was implemented in an Australian ED.

• A stable chest pain pathway with first-line outpatient CTCA reduced downstream investigations, without increasing costs or ICA in low-intermediate-risk patients after ED presentation.• Compared with functional testing, outpatient CTCA led to greater changes in preventive pharmacotherapies and greater yield of significant disease at ICA proceeding to revascularisation.• CTCA may be an appropriate first-line outpatient test in Australian stable chest pain pathways for select low-/intermediate-risk patients.

Introduction
Chest pain is one of the most common complaints in adults presenting to the ED. 1 Although the cause of chest pain is often not life-threatening, serious causes such as coronary artery disease (CAD) need to be considered. 1 Following exclusion of acute coronary syndrome, patients commonly undergo early outpatient evaluation for CAD with non-invasive testing if appropriate. 2 Diagnostic algorithms have traditionally relied upon pre-test probability scores for CAD and functional testing. 3][6] The CTCA is a well-established non-invasive modality that allows for a comprehensive three-dimensional assessment of coronary artery anatomy, plaque composition and burden, and severity of luminal stenosis. 7A strength of CTCA is its high sensitivity and negative predictive value, providing clinicians with diagnostic confidence in ruling out the presence of CAD. 8 CTCA also identifies both nonobstructive and obstructive CAD with high accuracy, thereby assisting in risk stratification and guiding the use of preventive therapies, such as statins and aspirin. 9The SCOT-HEART trial demonstrated that compared to functional testing, CTCA in stable chest pain significantly reduced the 5-year risk of death from coronary heart disease and non-fatal myocardial infarction (MI), likely due to more appropriate use of preventive therapies. 10Advances in technology have reduced radiation exposure and can also allow assessment of vulnerable plaque, perivascular inflammation and fractional flow reserve where available. 7iven the increased recommendations for CTCA by international guidelines, the impact of implementing CTCA as a first-line diagnostic test for chest pain on downstream investigations, cost, use of preventive therapies, and invasive coronary angiography (ICA) requires further evaluation in real-world settings, particularly in the Australian context. 11In addition, translating evidence-based recommendations into practice helps improve patient outcomes.Registry data have demonstrated that many patients undergoing ICA after functional testing have non-obstructive CAD. 12 Furthermore, studies have shown that patients undergoing chest pain investigation are typically low/intermediate risk, suggesting that CTCA may be a reasonable first-line test for ruling-out obstructive CAD in the majority of cases. 13The present study aimed to implement and evaluate the clinical impact of a novel chest pain pathway utilising an outpatient CTCA first-line strategy in an Australian tertiary hospital ED.

Study design
This observational study complies with the Declaration of Helsinki and was performed at a Western Australian adult tertiary hospital following approval (FSFHG GEKO 42094).A stable chest pain diagnostic pathway (Fig. 1) was developed by a multidisciplinary team of cardiologists, radiologists, and emergency physicians.Following education in the ED and cardiology departments, the pathway was implemented into the ED and cardiology rapid access chest pain clinic (RACPC) as standard of care in August 2021.The use of the pathway was at the discretion of ED clinicians, where the cardiology team were available for consultation.
In patients with suspected stable angina, the pathway recommends CTCA as a first-line outpatient test with follow up in the cardiology RACPC.Further investigation and management are recommended based on the severity of luminal stenosis demonstrated on CTCA (Fig. 1).In patients with non-diagnostic CTCA or CAD of uncertain haemodynamic significance (e.g.moderate or 50-69% stenosis), consideration of functional testing is recommended.Criteria are provided recommending non-invasive functional testing, rather than CTCA, in select patients, such as those aged ≥75 years, who have known CAD or previous percutaneous coronary intervention, or who have contraindications to CTCA such as renal impairment (estimated glomerular filtration rate <30 mL/min/1.73m 2 ), inability to control heart rate or contrast hypersensitivity.
In a pre-post design, the impact of the pathway was prospectively assessed over 6 months (August 2021 to January 2022) compared to an immediate 6-month pre-implementation group (February 2021 to July 2021).Patients were included if they were discharged from the ED after an acute coronary syndrome had been ruled out with electrocardiography (ECG) and high-sensitivity cardiac troponin, and had outpatient non-invasive testing organised for stable chest pain with follow up in the RACPC.In our institution's ED, the HEART score (Table S1) and 3-h troponin are used to risk-stratify patients with suspected cardiac chest pain. 14Patients who did not undergo non-invasive testing, proceeded directly to ICA, were followed up externally to the institution, or did not attend the post-ED follow up in the RACPC were excluded.

Non-invasive cardiac testing
CTCA was performed using prospective ECG-gating on a 64-slice dual-source high-pitch helical scanner (High-definition Flash; Siemens, Forchheim, Germany) following administration of intravenous contrast and in accordance with guideline recommendations. 15Patients received sublingual nitroglycerin and beta-blockers were administered to achieve a pre-scan heart rate of <65/min.Scans were reported by cardiac radiologists and stenosis severity was defined according to the coronary artery disease-reporting and data system (CAD-RADS): 0 (no stenosis), 1 (minimal or 1-24%), 2 (mild or 25-49%), 3 (moderate or 50-69%), 4 (severe 70-99% or >50% if left main coronary artery) and 5 (occluded or 100%), with the most severe stenosis defining the patient's score. 15Locally available non-invasive functional tests included stress ECG, stress echocardiography (exercise or pharmacological) and myocardial (sestamibi) perfusion scan.These tests were reported by cardiologists or in the case of myocardial perfusion scans, nuclear medicine physicians or radiologists.ICA is performed only following referral by a cardiologist.By study design, patients undergoing ICA in the present study would have undergone non-invasive testing prior.

Outcome measures
The primary aim was to assess whether the implementation of the pathway influenced the choice of first-line outpatient investigation and downstream investigations.The secondary aims were to assess the impact of CTCA on preventive medications compared with functional testing.Additionally, in patients who underwent ICA, we assessed rates of false positive non-invasive testing, and ICA resource utilisation and efficiency, defined as the ICA to revascularisation ratio (i.e. percentage of patients undergoing ICA proceeding to revascularisation). 16Expenditure following pathway implementation was evaluated using costs from the Medicare Benefits Schedule and hospitallevel data, as presented in Table S2.Data were obtained from digital medical records.

Statistical analysis
Statistical analyses were performed using SAS (version 9.4).Results are presented as mean AE standard deviation, number (per cent) or median (quartiles 1-3).Differences in continuous variables were compared using Student's t test or Mann-Whitney U test.Differences in categorical variables were compared using Pearson's chi-squared test or Fisher's exact test (if cell count <5).Statistical significance was defined as P < 0.05.

Results
Overall, 315 patients were included, with 143 pre-implementation and 172 post-implementation.Baseline characteristics at ED presentation are presented in Table
Burden of CAD on CTCA and functional testing according to the study period are presented in Table S3.There was no significant difference in the proportions with normal, minimal/mild, moderate and severe disease on CTCA between the study periods (P = 0.834) and there was no significant difference in ischaemic burden for functional testing (P = 0.547).In all patients undergoing CTCA, 76.9% had CAD-RADS 0-2 disease (40.7% normal coronary arteries and 36.2%minimal or mild disease).Functional testing or ICA following CTCA according to CAD-RADS is presented in Table S4.

Medication changes
There were no significant differences between study periods for bidirectional medication changes (aspirin: 22.4% vs 25.0%, P = 0.695 and statin: 20.3% vs 19.2%, P = 0.896), as presented in Table S5.In the entire cohort, bidirectional medication changes occurred more frequently following CTCA compared with functional testing (aspirin: 30.1% vs 14.3%, P = 0.005 and statin: 27.5% vs 7.9%, P < 0.001), as presented in Table 3. CTCA was associated with a greater proportion of stopping aspirin, as well as a greater proportion of stopping and starting statin (Table 3).Changes to aspirin and statin were observed more frequently after CTCA at all categories of CAD-RADS compared to both normal and positive functional tests, as presented in Figure 2.

Cost analysis
The total cost of non-invasive testing was lower post-implementation at $139 331.25 (mean $810.07 per patient) compared to pre-implementation at $126 236.30(mean $882.77per patient, P = 0.011 for comparison).The total cost of elective ICA preimplementation was $66 904.32, of which 42.1% ($28 170.24) was associated with ICA not proceeding to revascularisation.The total cost of elective ICA post-implementation was $56 340.48, and the proportion of costs associated with ICA not proceeding to ICA was lower at 18.8% ($10 563.84).

Discussion
Implementation of a stable chest pain assessment pathway in the ED, recommending CTCA as a first-line  outpatient test, successfully increased the use of CTCA.Most patients were low/intermediate risk, with a mean HEART score of four.Pathwayguided management was associated with fewer patients undergoing two non-invasive cardiac tests, thus reducing investigation costs, without increasing the rate of ICA.This highlights that a CTCA-based strategy can provide diagnostic reassurance and negates the need for further investigation in many low-/intermediate-risk patients.In patients referred for CTCA, approximately 75% had no, minimal or mild CAD in both study periods and thus likely required no further testing.CTCA was also associated with more bidirectional changes to preventive therapies compared with functional testing.In patients referred for ICA, CTCA demonstrated lower time to ICA and a greater yield of severe CAD compared to functional testing, with higher revascularisation to ICA ratio suggesting more efficient resource utilisation.
This contemporary Australian real-world study provides further support to the accumulating evidence for CTCA as a first-line test for low-/intermediate-risk stable chest pain patients after ED presentation.
8][19] Few studies have evaluated the utility of CTCA in chest pain pathways in the Australian context, highlighting the novelty of the current implementation study. 20,21In the current study, CTCA was not used as a same-day test due to resource availability and logistical limitations, rather CTCA was recommended as a first-line early outpatient test prior to RACPC follow up, in line with current practice.
The ideal first-line test for CAD should have high diagnostic accuracy, be able to risk-stratify patients and influence decision-making regarding the need for preventive therapies or further investigations. 11In the contemporary era, many patients undergoing outpatient investigation of chest pain are low/intermediate risk of obstructive CAD (such as the current study); thus, CTCA with its high sensitivity and diagnostic accuracy would be advantageous in ruling out CAD. 22 Importantly, non-obstructive CAD, which is not detected by functional tests, is associated with increased cardiovascular risk. 23TCA better detects non-obstructive and obstructive CAD, thereby providing the opportunity to risk-stratify patients and to commence, or withdraw preventive therapies across all stages of CAD. 9,24There was no significant change in medications post-implementation, which may reflect the relatively high CTCA use pre-implementation.
However, changes in aspirin and statin occurred significantly more following CTCA compared to functional testing in the overall cohort, suggesting more appropriate preventive therapy use.
Prior studies have raised concerns that CTCA use may increase rates of ICA. 25,26In the first 12 months of SCOT-HEART, rates of ICA were greater in those undergoing CTCA compared to functional testing. 10owever, by 5 years, no significant difference in rates of ICA was observed, suggesting that patients with obstructive CAD were referred earlier for ICA. 10 In the current study, it is notable that although pathwayguided management increased the use of CTCA as expected, this did not increase rates of ICA, which may be due to the relatively low risk of the patients studied.Pathway-guided management negated the need for multiple investigations thereby reducing healthcare costs, likely due to the high prevalence of non-obstructive CAD coupled with the increased diagnostic certainty of CTCA.In line with this, CTCA also reduced the time from ED to ICA.Reduced functional testing has also been shown in other studies of CTCA in outpatients with chest pain. 27,28 has been argued that CTCA could be the 'gatekeeper' to ICA, as referrals for the invasive procedure should ideally be reserved for patients who may require revascularisation. 29In patients referred for ICA following a functional test in the current study, only 36% had severe CAD.Referral for ICA following CTCA was associated with a greater yield of significant CAD at ICA compared to functional testing, with a greater proportion proceeding to revascularisation.This suggests that CTCA may better select patients requiring ICA and revascularisation, leading to more efficient ICA resource utilisation and potentially reduced healthcare expenditure.In defence of functional testing, identification of myocardial ischaemia remains important in guiding revascularisation in patients with known CAD. 4,5 It is important to recognise that functional tests can be positive in the setting of coronary microvascular dysfunction, which also portends worse outcomes but is not detected by the standard CTCA. 30Additionally, functional testing can be used for assessment of viability, exercise tolerance, heart rate response, myocardial scar and lesions of uncertain haemodynamic significance (such as moderate lesions on CTCA). 4,5Furthermore, the limitations of CTCA should be considered, such as exposure to ionising radiation, requirement for iodinated contrast and the need for adequate heart rate control at time of acquisition to optimise image quality. 11imitations of the present study include its relatively modest sample size and pre-post design.Data was collected from digital medical records and patients who were not followed up within the institution were excluded.The use of the pathway was at the discretion of the ED clinician and data were not reported on ED clinician compliance with the pathway.In addition, the study was limited to a single Australian tertiary centre located in a metropolitan area, thus the results may not be generalisable to other settings with differences in care delivery and it cannot be determined whether there was a general temporal trend to increased CTCA use across the whole healthcare system.CTCA use was already relatively high pre-implementation (almost one in two patients) which may have attenuated the study results.Wait-time for CTCA did not change post-implementation, as resource availability was not increased.The present study did not evaluate newer CTCA technologies for assessment of high-risk plaque, perivascular inflammation or fractional flow reserve, which could further strengthen the role of CTCA. 7 larger study with longer follow up would be required to provide a more comprehensive cost-analyses and to evaluate late cardiovascular outcomes in this low-/intermediate-risk group, although the prognostic value of CTCA in stable chest pain evaluation has been shown by the SCOT-HEART trial.10 In conclusion, implementation of a stable chest pain assessment pathway in a tertiary hospital ED increased outpatient CTCA utilisation and was associated with reduced downstream investigations, without increasing rates of ICA.When compared to functional testing in the overall cohort, CTCA resulted in a greater number of changes to preventive therapies and a greater yield of significant CAD on ICA proceeding to coronary revascularisation.The findings of observational study support the notion that CTCA is a robust and potential 'gate-keeper' to ICA in low-/intermediate-risk patients with chest pain.Larger studies evaluating the real-world impact of stable chest pain assessment pathways with CTCA as a first-line test are needed in other populations.

Figure 2 .
Figure 2. Medication changes following computed tomography coronary angiography according to coronary artery disease severity and following a normal functional test.CAD-RADS, coronary artery disease-reporting and data system.*Includes four equivocal functional test results.

TABLE 2 .
Outpatient diagnostic tests for coronary artery disease according to study period

TABLE 1 .
Characteristics at time of ED presentation according to study period

TABLE 3 .
Preventive medication changes following last non-invasive test in the entire study cohort