Cardio-oncology & otherRandomised controlled trial
Artificial intelligence guided screening for cardiomyopathies in an obstetric population: a pragmatic randomized clinical trial
- Randomised controlled trial
- open-label
- pragmatic
- 6 hospitals in Nigeria
Enrolment August 2022 to September 2023; follow-up through May 2024
Authors' conclusion
In pregnant and postpartum women, AI-guided screening using a digital stethoscope improved the diagnosis of pregnancy-related cardiomyopathy.
Study context
- Population
- Pregnant and postpartum women in Nigeria1,195 completed baseline visit
- Participants
- 1,232 participants
- Comparison
- Artificial intelligence (AI)-guided screening vs usual care
- Interventionn = 616
- Usual caren = 616reference
- Primary endpoint
- Identification of LVSD (intervention: positive AI screen confirmed by echocardiography; control: clinically recognised LVSD on echocardiography)
- Follow-up
- During the study period
Primary result
LVSD detected (AI-enabled digital stethoscope)
Odds ratio
2.12
95% CI 1.05–4.27
P = 0.032
Events by treatment group
Intervention vs Usual care
OR 2.12 (95% CI 1.05–4.27); P = 0.032
As reported in the abstract
Further findings
Left ventricular systolic dysfunction (LVSD) identification using the 12-lead AI-ECG model
- Detected with AI-ECG model
- 20 out of 587 (3.4%)
- Detected without AI screening
- 12 out of 608 (2.0%)
- OR
- 1.75 (95% CI 0.85-3.62)
- P
- P = 0.125
Prespecified subgroups
- Similar direction of effect in prespecified subgroup analysis
Safety
Serious adverse events related to study participation
- No serious adverse events related to study participation
Original publication
Artificial intelligence guided screening for cardiomyopathies in an obstetric population: a pragmatic randomized clinical trial. Nat Med. Published 2 September 2024. doi:10.1038/s41591-024-03243-9
Summarised from the published abstract; every number is checked against it