AI-Enabled Electrocardiogram Alert Intervention Reduces All-Cause Mortality

Chin Lin, PhD

Credit: National Defense Medical Center

A recent pragmatic randomized clinical trial provided insight into the abilities of an artificial intelligence (AI)-enabled electrocardiography (ECG) alert intervention to identify hospitalized patients with a raised mortality risk.¹

Achieving its primary endpoint, the study demonstrated implementation of the AI-ECG alert across nearly 16,000 patients was associated with a significant reduction in all-cause mortality within 90 days.

“While the precise means by which implementation of the AI-ECG alert led to decreased mortality are to be fully elucidated, these results indicate that such implementation assists in the detection of high-risk patients, prompting timely clinical care and reducing mortality,” wrote the investigative team, led by Chin Lin, PhD, associate professor, school of medicine, National Defense Medical Center.

Early identification of vulnerable patients may improve hospital outcomes but could prove challenging for implementation into current clinical practice. This multi-site randomized controlled trial sought to assess the application of an AI-enable ECG system to identify hospitalized patients with a high mortality risk across 39 physicians and 15,965 patients.

Patients randomized to the AI-ECG alert intervention were provided the screening tool, which included an AI report and warning messages to specify patients at a high mortality risk.² Once the AI-ECG warned of a high mortality risk, a message was immediately sent to the attending physician.

Notifications appeared in the recipient’s smartphone message system for prompt attention. The message relayed to the physician that “An ECG was received for patient X. An ECG indicates a high risk of mortality. Please intensively attend to the patient’s conditions.” Physicians were allowed to click on a link to connect the ECG and the result of AI-ECG prediction to identify the ECG further.²

Warning messages were actively sent for high-risk cases identified by the AI and the AI-ECG report of low-risk cases also presented the degree of risk. Physicians could check the relative severity by accessing the electronic health record (EHR) for patients in the intervention group.

The control group comprised no intervention and patients remained on routine clinical practice. Primary outcomes for the analysis included all-cause mortality within 90 days, with tracking by the AI-ECG.¹ The secondary outcomes included cardiovascular cause mortality, arrhythmia medication, electrolyte examination, and cardiac examination.

Upon analysis, implementation of the AI-ECG alert was linked with a significant reduction in all-cause mortality within 90 days. Specifically, 3.6% of patients in the intervention cohort died within 90 days, compared with 4.3% in the control group (hazard ratio [HR], 0.83; 95% CI, 0.70 - 0.99).

According to a pre-specified analysis, the reduction in all-cause mortality associated with the AI-ECG alert was identified primarily among those with high-risk ECGs (HR, 0.69; 95% CI, 0.53 - 0.90).

Regarding the secondary outcome analyses, Lin and colleagues found those in the intervention group with high-risk ECGs received increased levels of intensive care compared with the control group.

Among the high-risk ECG cohort, implementation of the AI-ECG alert was associated with a significant reduction in the risk of cardiac death (0.2% in the intervention arm vs. 2.4% in control; HR, 0.07; 95% CI, 0.01 - 0.56).

References

1. _{Lin, CS., Liu, WT., Tsai, DJ. et al. AI-enabled electrocardiography alert intervention and all-cause mortality: a pragmatic randomized clinical trial. Nat Med (2024). https://doi.org/10.1038/s41591-024-02961-4}
2. _{Applying an artificial intelligence-enabled electrocardiographic system for reducing mortality - full text view. ClinicalTrials.gov. November 11, 2021. Accessed April 30, 2024. https://classic.clinicaltrials.gov/ct2/show/NCT05118035.}