Sex differences in machine learning computed tomography-derived fractional flow reserve

Area of Science:

• Cardiology
• Medical Imaging
• Artificial Intelligence

Background:

• Coronary computed tomography angiography (CCTA) is crucial for diagnosing coronary artery disease.
• Machine learning fractional flow reserve (ML-FFR_CT) offers a non-invasive method to assess the hemodynamic significance of coronary stenoses.
• Understanding sex-based differences in cardiovascular risk assessment is vital for personalized medicine.

Purpose of the Study:

• To investigate potential sex differences in the association between ML-FFR_CT and incident cardiovascular outcomes.
• To compare ML-FFR_CT values and their prognostic implications in men and women with suspected coronary artery disease.

Main Methods:

• A retrospective cohort study included 471 patients undergoing CCTA and single photon emission computed tomography (SPECT).
• ML-FFR_CT was calculated using a machine learning algorithm, with ML-FFR_CT < 0.8 defining significant stenosis.
• The primary outcome was a composite of death or non-fatal myocardial infarction (D/MI), analyzed using multivariable adjusted models.

Main Results:

• Women had less obstructive stenosis by CCTA (9% vs. 18%) and a higher median ML-FFR_CT (0.76 vs. 0.71) compared to men.
• The prevalence of significant stenosis (ML-FFR_CT < 0.8) was similar between sexes (39% vs. 44%).
• Multivariable analysis revealed no significant association between ML-FFR_CT < 0.8 and D/MI, with no significant interaction detected for sex.

Conclusions:

• In symptomatic patients undergoing CCTA and SPECT, ML-FFR_CT values were higher in women than men.
• ML-FFR_CT did not show a significant association with incident death or myocardial infarction in this cohort.
• The prognostic value of ML-FFR_CT for cardiovascular events does not appear to differ significantly between men and women.