Arrhythmias & electrophysiologyCohort study
Artificial Intelligence Electrocardiography to Predict Atrial Fibrillation in Patients With Chronic Lymphocytic Leukemia
- Retrospective cohort study
- single-centre database (Mayo Clinic chronic lymphocytic leukemia (CLL) database)
- validation cohort
Authors' conclusion
An AI-ECG algorithm, in conjunction with the Mayo CLL AF risk score, can predict the risk of AF in patients with newly diagnosed CLL.
Study context
- Population
- Newly diagnosed CLL, Mayo Clinic CLL databasenewly diagnosed CLL
- Participants
- 754 participants
- Exposure
- Artificial intelligence electrocardiography (AI-ECG) score ≥0.1 vs <0.1
- Outcome
- Incident AF
- Follow-up
- Median 5.8 years
Main finding
As reported in the abstract
AF
AI-ECG ≥0.1 vs AI-ECG <0.1Hazard ratio (95% CI)P value
unadjusted
3.9(2.6–5.7)
P < 0.001adjusted for Mayo CLL AF risk score, HF, CKD, CLL therapy
2.5(1.6–3.9)
P < 0.001- 10-year cumulative risk of atrial fibrillation (AF) (all patients)
- 26.1%
- Median baseline AI-ECG score (range 0-0.93)
- 0.02
Observational data: associations, not effects.
Further findings
| Outcome | HR (95% CI) | P value | |
|---|---|---|---|
| AF in second cohort treated with BTK inhibitor (pretreatment AI-ECG ≥0.1) · n = 220; nonsignificant | 1.7 (0.8–3.6) | P = 0.19 | |
| 0.20.5125Hazard ratio (log scale) |
Original publication
Artificial Intelligence Electrocardiography to Predict Atrial Fibrillation in Patients With Chronic Lymphocytic Leukemia. JACC CardioOncol. Published 16 April 2024. doi:10.1016/j.jaccao.2024.02.006
Summarised from the published abstract; every number is checked against it