Machine-learning model helps predict short-term response to cardiac-resynchronization therapy

Reuters Health - 06/09/2021 - A machine-learning (ML) model can predict short-term response to cardiac-resynchronization therapy (CRT) and may aid care planning, researchers say.

Optimizing CRT delivery can improve long-term CRT outcomes but requires substantial staff resources, Dr. Stacey Howell of the Knight Cardiovascular Institute, Oregon Health and Science University, in Portland, and colleagues note in JACC: Clinical Electrophysiology.

In the SMART-AV trial, they applied ML to develop a model that could help predict short-term CRT response. Participants included 741 adults (mean age, 66; 33% women, all with NYHA heart-failure class III to IV and ejection fraction of 35% or less).

Baseline clinical, electrocardiographic, echocardiographic and biomarker characteristics, and left ventricular (LV) lead position (43 variables) were included in eight different ML models.

The primary endpoint was a composite of freedom from death and heart-failure hospitalization and a greater than 15% reduction in left ventricular end-systolic volume index at six months after CRT. A total of 337 patients (46%) met this primary endpoint.

The ML model that was the most accurate for six-month CRT response includes routinely available baseline clinical, ECG, and echocardiographic characteristics, some of which are potentially modifiable.

The model predicts CRT response with 70% accuracy, 70% sensitivity, and 70% specificity, the authors report. Patients in the fifth versus first quintile of the prediction model had 14-fold higher odds of composite CRT response, a significant increase.

"Before the procedure, our calculator tool could be potentially used for shared decision making and set-up of management goals (eg, target weight, target systolic blood pressure, 6-minute walk distance, biomarker levels). Such discussion with a CRT candidate might motivate compliance to diet, fluid restriction, and medication," Dr. Howell and colleagues say.

"Furthermore, our prediction model can be used to identify CRT recipients with a low probability of CRT response. They must be referred to a multidisciplinary CRT-HF clinic very early, immediately after CRT implantation," they note.

The model should be further validated in prospective studies, the team cautions.

The SMART-AV trial was sponsored by Boston Scientific. Two of the authors are employees of the company.

SOURCE: https://bit.ly/3mAX5Qr JACC: Clinical Electrophysiology, online August 25, 2021.

By Reuters Staff



Posted on 6 September 20219 September 2021