Hot Line 12: PROTEUS, RAPID xAI and WESTCOR-POC
02 Sep 2024
ESC Congress 2024’s Late-Breaking Science presentations came to a close with three trials investigating the use of cutting-edge technology to help with the diagnosis of severe coronary artery disease (CAD) and myocardial infarction (MI).
As discussed by Doctor Ross Upton (University of Oxford - Oxford, UK), the PROTEUS trial evaluated the use of EchoGo Pro, which provides automated interpretation of stress echocardiography (SE) based on artificial intelligence (AI) image analysis. More than 2,000 patients were randomised to standard clinical decision-making (control) or AI-augmented decision-making, during which clinicians received an EchoGo Pro AI report indicating the likelihood of severe CAD. The primary analysis evaluated the appropriateness of standard clinical decision-making vs. AI-augmented decision-making when selecting patients for invasive coronary angiograms and related acute coronary events within 6 months.
Overall, the analyses found that AI-assisted decision-making did not demonstrate non-inferiority vs. clinical decision-making for correctly selecting patients for coronary angiography. Of those sent for angiography, 27 out of 36 referrals were correct in the control arm and 34 out of 49 referrals were correct in the AI arm. Of the patients that should have been sent for an angiogram and who subsequently experienced an event, 22 were in the control group and 19 in the AI group, but the difference was not statistically significant. However, further analyses found that AI performed better than current practice in female patients, those with pre-existing CAD and low-volume SE centres. “It is well reported that clinician performance in interpreting SEs ranges widely according to the experience level of the operator,” explained Dr. Upton. “These results suggest that AI can bring all operators, regardless of experience, up to the same level of accuracy and may be a useful training aid. While the PROTEUS trial did not demonstrate non-inferiority in all-comers, the AI diagnostic may benefit specific groups of patients in whom decision-making is known to be more complex.”
The next presentation, by Professor Derek Chew (Victorian Heart Hospital - Melbourne, Australia), described the use of AI to aid clinical decision-making in identifying and managing MI, based on the 4th Universal Definition, in patients presenting to the emergency department (ED). The RAPIDx AI cluster-randomised trial enrolled 14,131 patients from 12 centres across South Australia: 6 were randomised to the intervention arm (i.e. implementation of AI-based clinical decision support) and 6 to the control arm (i.e. unchanged standard of practice).
In the intention-to-treat analysis, there was no difference between the groups, with 26.0% in the intervention group and 26.4% in the control group experiencing a CV death, MI or unplanned CV readmission within 6 months. Importantly, among patients not classified as type 1 MI by AI-driven decision support, invasive coronary angiography was 47% less likely to be undertaken in the intervention group vs. the control group (5% vs. 9.4%). Additionally, where patients were classified as having a type 1 MI by the AI-based decision support, they were more likely to receive recommended medical therapies. Patients directly discharged from EDs with the decision support were less likely to die or have an MI within 30 days than those who received usual care (0.86% vs. 1.1%; non-inferiority p<0.001).
“We found no increase in early hazards or negative impacts on ED discharge decisions, establishing the safety of AI-based clinical decision support,” said Prof. Chew. “Our next steps include exploring approaches to enhancing trust and adoption of AI-based clinical decision support in the clinical community, investigating new models of care within which such AI-based decision support tools could be integrated to drive health system effectiveness and efficiency, and evaluating AI-based decision support for other acute cardiac conditions where early recognition represents a key challenge and driver in optimising outcomes.”
Last, but not least, Doctor Viola Thulin (Haukeland University Hospital - Bergen, Norway) presented the WESTCOR-POC trial that compared a 0-hour and 1-hour novel point-of-care (POC) high sensitivity cardiac troponin (hs-cTn) test (Atellica VTLi, Siemens Healthineers) with conventional 0-hour and 1-hour central laboratory hs-cTn testing. In total, 1,494 consecutive patients with symptoms suggestive of acute coronary syndrome presenting to the ED were randomised to the novel POC test with a turn-around time of 8 minutes or standard central lab testing. The median length of stay in the ED was 174 minutes for the POC testing group compared with 180 minutes in the standard testing group. However, among patients who were seen more quickly by a physician (within 60 minutes), POC testing reduced the length of ED stay by 15 minutes (147 vs. 162 minutes). Notably, POC testing provided the most benefit for patients diagnosed with non-ST-elevation MI, shortening their ED stay by 43 minutes compared with standard testing (median 137 vs. 180 minutes), with high-risk patients being admitted to the cardiac ward faster. Rates of combined deaths, MIs and acute revascularisations within 30 days were similar with POC and standard testing (11.4% vs. 9.4%, respectively). “POC troponin assays hold great promise to improve patient care. But our findings underscore the need for a process to map out and address obstacles to efficient patient flow, such as lack of relevant staff or lack of efficient discharge procedures, to realise the full potential of POC tests to manage chest pain patients in the ED,” said lead author, Doctor Kristin Aakre (Haukeland University Hospital - Bergen, Norway).
