Highlights
– In the DISCHARGE prespecified secondary analysis, an initial CT-first strategy yielded no clinically meaningful advantage over an ICA-first strategy in patient-reported quality of life (QOL) or angina at a median 3.5 years of follow-up.
– Both groups experienced small but statistically significant improvements in several QOL domains; clinically important differences were generally absent.
– Female patients reported worse baseline and follow-up health status than male patients and demonstrated larger magnitude improvements over time, but neither testing strategy eliminated sex-related disparities.
Background
Patients presenting with stable chest pain represent a heterogeneous population in whom clinicians must balance diagnostic accuracy, procedural risk, downstream testing, and patient-centered outcomes such as symptoms and quality of life (QOL). Noninvasive coronary computed tomography angiography (CT) and invasive coronary angiography (ICA) offer complementary strengths: CT provides noninvasive anatomical visualization of coronary atherosclerosis, while ICA offers definitive luminal assessment with the option of immediate revascularization. Randomized trials comparing anatomical and functional testing (and different diagnostic pathways) have addressed clinical events and resource use, but less is known about how initial testing strategy influences patient-reported health status and angina over long-term follow-up. The DISCHARGE trial prespecified a secondary analysis to compare health status outcomes between CT-first and ICA-first strategies among patients with stable chest pain and intermediate pretest probability of coronary artery disease (CAD).
Study design
Population and setting
DISCHARGE was a large, pragmatic multi-center randomized clinical trial conducted in 26 European centers. Between October 2015 and April 2019, 3561 patients with stable chest pain and intermediate probability of CAD were randomized to an initial diagnostic strategy of CT or ICA and followed for a median of 3.5 years. The prespecified secondary analysis reported here focused on patient-reported outcomes collected longitudinally.
Interventions and endpoints
Participants were randomized 1:1 to CT-first or ICA-first diagnostic pathways. The prespecified primary QOL outcomes were the EuroQOL 5-dimensions 3-level visual analog scale (EQ-5D-3L-VAS) and the 12-item Short Form physical component score (SF-12-PCS). The primary chest pain outcome was angina (reported as event rates over follow-up). Additional patient-reported outcomes included EQ-5D-3L summary index (SI), SF-12 mental component summary (SF-12-MCS), and the Hospital Anxiety and Depression Scale subscales for anxiety (HADS-A) and depression (HADS-D). Health status assessments were completed at baseline and serially over follow-up; 96.0% of randomized patients in the CT arm and 95.3% in the ICA arm completed at least one health-status assessment.
Key findings
Overall QOL and symptom outcomes
Over a median 3.5 years of follow-up, the CT-first and ICA-first groups had similar trajectories in patient-reported health status across prespecified measures. Both strategies were associated with statistically significant improvements from baseline in several QOL measures, but the magnitude of change was small and comparable between arms.
Example effect sizes (3.5-year change from baseline):
- EQ-5D-3L-VAS: CT group mean change = +4.0 (95% CI, 3.1 to 4.9; P <.001); ICA group mean change = +4.6 (95% CI, 3.6 to 5.6; P = .002).
- SF-12-PCS: both groups showed small improvements (prespecified measure; specific numeric changes are reported in the primary publication).
Mood and psychological measures
Most psychological indices improved comparably between groups. HADS-D (depression subscale) showed a statistically significant improvement only in the CT group (mean change approximately -0.2; 95% CI, -0.4 to 0; P = .04), while the ICA group showed a similar point estimate that did not reach statistical significance (mean change approximately -0.2; 95% CI, -0.4 to 0; P = .12). The absolute magnitude of these changes was small and of uncertain clinical importance.
Angina and chest pain outcomes
Chest pain (angina) outcomes were similar at 3.5 years between CT and ICA strategies. One-year angina rates were comparable in the CT group relative to the ICA group, but sex-stratified analysis identified heterogeneity: within the ICA arm, 1-year angina rates were higher in female than male patients (10.2% vs 6.2%; P = .007). Overall, the initial testing strategy did not produce a clinically meaningful difference in angina burden by 3.5 years.
Sex differences in health status
Female patients reported worse health status at baseline and at follow-up compared with male patients. For example, the baseline difference in EQ-5D-3L-VAS between men and women was 5.2 points (95% CI, 4.0 to 6.3; P <.001); at 3.5 years the between-sex difference remained (3.1 points; 95% CI, 1.9 to 4.4; P <.001). Despite worse absolute scores, women experienced greater magnitude improvements than men on several scales: between-group difference in change favored women for EQ-5D-3L-VAS (between-group difference in change -1.9; 95% CI, -3.4 to -0.5; P = .009), SF-12-PCS (-1.4; 95% CI, -2.1 to -0.7; P <.001), and HADS-A (0.3; 95% CI, 0 to 0.7; P = .04). In plain terms, women started with worse health status and improved more over time, but residual sex disparities persisted at long-term follow-up.
Clinical significance of effect sizes
Although multiple outcomes achieved statistical significance, the absolute changes were modest. For instruments such as the EQ-5D VAS and SF-12, conventional thresholds for minimally important clinical differences are generally larger than the point estimates observed here. Thus, while the trial demonstrates equivalence of patient-reported outcomes between CT and ICA strategies, it does not support a meaningful advantage of one diagnostic-first approach over the other in terms of long-term QOL or angina burden in this population.
Expert commentary
The DISCHARGE health-status analysis provides clinically useful, patient-centered information to complement prior event-focused trials comparing diagnostic strategies for stable chest pain. Its large size, pragmatic design, high follow-up rate for patient-reported outcomes, and prespecified analysis strengthen confidence in the findings.
Clinical implications:
- Choice of initial anatomic (CT) versus invasive (ICA) testing in intermediate-risk stable chest pain patients can be guided largely by diagnostic goals, local expertise, patient preferences, procedural risk, and resource considerations rather than expectations of superior long-term QOL.
- CT retains the advantage of noninvasiveness and may change downstream management pathways, but those differences do not appear to translate into better long-term QOL compared with ICA-first approaches in this trial population.
- Persistent and clinically relevant sex differences in baseline health status and angina burden warrant attention: diagnostic strategy alone did not close these gaps, highlighting the need for sex-aware clinical evaluation, risk modification, and systematic follow-up tailored to women’s presentation and needs.
Limitations to consider:
- Secondary analysis: although prespecified, these are secondary outcomes and the trial was not powered primarily for small differences in QOL measures.
- Generalisability: the study enrolled patients with intermediate pretest probability in European centers; outcomes may differ in populations with different risk spectra or health systems.
- Unmeasured downstream care: the analysis does not fully disentangle how subsequent medical therapy, revascularization, or non-cardiac care influenced patient-reported outcomes over time.
Conclusion
In patients with stable chest pain and intermediate pretest probability of CAD, an initial diagnostic strategy of CT versus ICA resulted in similar patient-reported quality-of-life and angina outcomes at a median of 3.5 years in the DISCHARGE trial. Both strategies were associated with small improvements in QOL measures over baseline, but the magnitude was modest and unlikely to be clinically consequential. Female patients had worse baseline and follow-up health status than male patients and experienced greater improvements over time, yet sex-related disparities persisted irrespective of testing strategy. Clinicians should therefore individualize the choice of first test based on diagnostic yield, patient preferences, local resources, and procedural risk, while addressing persistent sex-based differences in symptom burden and health status through comprehensive care pathways.
Funding and trial registration
The DISCHARGE trial group and study funding details are reported in the primary publication. Trial registration: ClinicalTrials.gov Identifier: NCT02400229.
References
1. DISCHARGE Trial Group; Rieckmann N, Neumann K, Maurovich-Horvat P, et al. Health Status Outcomes After Computed Tomography or Invasive Coronary Angiography for Stable Chest Pain: A Prespecified Secondary Analysis of the DISCHARGE Randomized Clinical Trial. JAMA Cardiol. 2025 Jul 1;10(7):728-739. doi: 10.1001/jamacardio.2025.0992. PMID: 40366703; PMCID: PMC12079563.
2. 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. (PROMISE trial)
3. Knuuti J, Wijns W, Saraste A, et al. 2019 ESC Guidelines for the diagnosis and management of chronic coronary syndromes. Eur Heart J. 2020;41:407-477.
Suggested next research steps
Future work should explore whether targeted, sex-specific diagnostic and therapeutic pathways can reduce residual disparities in health status and angina and whether integrating patient preferences and symptom trajectories into testing algorithms improves long-term patient-centered outcomes. Longitudinal assessment of downstream therapies and their mediating role between diagnostic strategy and QOL would further clarify the optimal pathways for patients with stable chest pain.
