Authors' objectives: Background The aim of this health technology assessment (HTA) is to evaluate the safety, effectiveness, cost, cost-effectiveness and budget impact of revascularisation with percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG) or both (i.e. PCI or CABG) in comparison to optimal medical therapy (OMT) in patients with chronic coronary syndrome (CCS). Ethical, legal, social and organisational issues related to the interventions are also explored.

Authors' results and conclusions: Clinical evaluation RCT results Seven RCTs were included; 3 compared PCI plus OMT with OMT alone, 1 compared CABG plus OMT with OMT alone, 2 compared revascularisation (either PCI plus OMT or CABG plus OMT) with OMT alone, and 1 compared PCI plus OMT, CABG plus OMT and OMT alone. In total, 10,924 patients (range 396 to 5,179) were included. Risk of bias was ‘high’ in 2 trials and of ‘some concern’ in 5 trials. CABG plus OMT demonstrated significant favourable results across several outcomes: MACE, hospitalisation (9 fewer hospitalisations per 100 patients; high certainty of evidence) and MI (3 fewer MIs per 100 patients; high certainty of evidence). The benefits of CABG plus OMT were only apparent for all-cause mortality at 120 months follow-up and subsequent revascularisation at 60 months followup. The initially favourable HRQoL scores for CABG plus OMT diminished at longer-term follow-up with no additional benefit compared to OMT alone. PCI plus OMT reported significant favourable results for HRQoL (general and cardiac-specific), anginal frequency and subsequent revascularisation during the short-term (≤ 24 months). Anginal stability at 12 months was not significantly different, and there were no reported clinical benefits above OMT alone for all-cause mortality, hospitalisation, MI and stroke. MACE only favoured PCI plus OMT at 24 months follow-up, whilst 12- and 60-month follow-up were not significant. Subsequent revascularisation favoured PCI plus OMT up to 24 months, but was insignificant at 60 months. Revascularisation (i.e. PCI or CABG) plus OMT demonstrated significant favourable results for anginal frequency and general HRQoL at 12 months, treatment satisfaction up to 24 months, and subsequent revascularisation at 60 months (17 fewer subsequent revascularisations per 100; moderate certainty of evidence). Hospitalisation due to HF was significantly lower in the comparison group receiving OMT alone. Other outcomes reported no significant differences, or were not reported. NRSI results Eight NRSIs were included; 5 compared PCI plus OMT with OMT alone, 2 compared revascularisation plus OMT with OMT alone, and 1 compared PCI plus OMT, CABG plus OMT, and OMT alone. In total 25,803 (range 83 to 9,676) participants were included. Study quality was acceptable in 7 of the studies. For CABG plus OMT, all-cause mortality, subsequent revascularisation and MI were not significantly different compared with OMT alone; no data was available on hospitalisation or stroke. For PCI plus OMT, all-cause mortality was significantly lower (at 120 months), but there were no clear benefits for MACE or subsequent revascularisation, and no differences for hospitalisation or MI. Stent thrombosis and target vessel revascularisation were significantly lower in participants receiving OMT alone. For revascularisation (i.e. PCI or CABG), all-cause mortality and need for subsequent revascularisation were significantly lower in participants receiving revascularisation plus OMT compared to OMT alone at 60 months follow-up. Hospitalisation and MI were not significantly different. Economic evaluation The de novo economic evaluation demonstrated unfavourable outcomes (i.e. high incremental costeffectiveness ratio [ICER] or dominance) for PCI, CABG and revascularisation (i.e. PCI or CABG) under the base case. Economic outcomes improved when baseline event rates were sourced from more targeted population groups, including patient cohorts with chronic kidney disease or left ventricular ejection fraction ≤35%. For all comparisons, the relative effect of revascularisation with respect to all-cause mortality was a major driver of cost-effectiveness. The cost of inpatient PCI and CABG procedures, as well as the baseline annual mortality transition, were also important parameters. Under current policy conditions, CABG and PCI procedures for CCS were estimated to be responsible for anticipated costs of Swiss francs (CHF) 59.7 million and CHF146.1 million in 2023, respectively. Considering observed trends in the use of revascularisation procedures over the period 2016 to 2019 (i.e. reducing utilisation of CABG, increasing utilisation of inpatient and outpatient PCI), anticipated CABG costs were projected to decrease to CHF51.2 million by 2027 while anticipated PCI costs were projected to increase to CHF184.3 million. Scenario analyses using alternative data sources and/or assumptions for PCI procedure numbers found costs for PCI ranging from CHF135.9 million to CHF142.3 million in 2023, and CHF114.1 million to CHF156.2 million in 2027. Ethical, legal, social, organisational evaluation Ten publications related to ethical, social and organisational issues were identified; none were identified relating to legal considerations. Regarding ethical issues, treating physicians should ensure patients have a comprehensive understanding of the risks associated with PCI compared to OMT alone, so informed consent can be elicited. Social issues identified that cultural distrust of healthcare providers, patients' perception of their illness, and assumptions by healthcare providers based on patients’ social characteristics all impact the care a patient received. However, these social issues are based on patient groups in the USA and may not be applicable to the Swiss healthcare system. Shared decision-making between patients and clinicians can ameliorate these challenges. Conclusions The RCT evidence demonstrated generally favourable outcomes for the use of CABG plus OMT compared with OMT alone. For PCI plus OMT, the evidence of a benefit was less clear and was largely limited to short-term outcomes (≤ 24 months). The NRSI evidence was more difficult to interpret due to selection bias, residual confounding factors, unknown participant dropout between treatment groups, limited outcome reporting, and results reporting contradictory findings when reported as hazard ratio or risk ratio. As such, the RCT evidence was deemed more reliable, and was used as the basis for the economic evaluation. The economic evidence demonstrated unfavourable outcomes (i.e. high ICERs or dominance) for all interventions under base case baseline event rate estimates. For all comparisons, the relative effect of revascularisation with respect to all-cause mortality was a major driver of cost-effectiveness. Economic outcomes improved when baseline event rates were sourced from more targeted population groups, including patient cohorts with chronic kidney disease or left ventricular ejection fraction ≤35%.

Authors' methods: Systematic searches were conducted in MEDLINE, Embase, the Cochrane Library, the INAHTA database, and EconLit up to 29 June 2022. Systematic reviews, randomised controlled trials (RCTs) and nonrandomised studies of interventions (NRSI) published after 1 January 2000 were eligible for inclusion. Relevant outcomes included major adverse cardiac events (MACE), all-cause mortality, hospitalisation, subsequent revascularisation, health-related quality of life (HRQoL), stent thrombosis, myocardial infarction (MI), and stroke. Risk of bias was assessed using study design-specific tools, and the overall strength of evidence for important outcomes was assessed using the Grading of Recommendation, Assessment, Development and Evaluations (GRADE) approach. Separate random-effects meta-analyses were conducted for RCTs and NRSIs using R. Heterogeneity was evaluated qualitatively (i.e. forest plots), and quantitatively (i.e. I2, Chi2 and Tau2 ). A systematic review of economic studies was undertaken in the same databases. Data were extracted from retrieved cost and cost-effectiveness studies, and the results described narratively. The applicability of the retrieved cost-effectiveness studies to the HTA context was assessed. None of the existing evidence was directly applicable to the Swiss context, thus a de novo Markov model was built.

Project Status: Completed

URL for project: https://www.bag.admin.ch/bag/en/home/versicherungen/krankenversicherung/krankenversicherung-leistungen-tarife/hta/hta-projekte/invasiveeingriffe.html

URL for protocol: https://www.bag.admin.ch/dam/bag/en/dokumente/kuv-leistungen/leistungen-und-tarife/hta/berichte/h0065iekh-hta-protocol.pdf.download.pdf/h0065iekh-hta-protocol.pdf

Year Published: 2023

URL for published report: https://www.bag.admin.ch/dam/bag/en/dokumente/kuv-leistungen/leistungen-und-tarife/hta/berichte/h0065iekh-hta-report.pdf.download.pdf/h0065iekh-hta-report.pdf

English language abstract: An English language summary is available

Publication Type: Full HTA

Country: Switzerland

Organisation Name: Swiss Federal Office of Public Health (FOPH)

Contact Address: Federal Office of Public Health, Schwarzenburgstrasse 157, CH-3003 Berne, Switzerland

Contact Name: Stephanie Vollenweider

Contact Email: hta@bag.admin.ch

Copyright: Swiss Federal Office of Public Health