Automated Abdominal Aortic Calcification Assessment and Major Adverse Cardiovascular Events in People Undergoing Osteoporosis Screening
Authors:
Date:
Abstract:
Background
Automated assessment of abdominal aortic calcification using machine learning (ML-AAC24), a marker of asymptomatic cardiovascular disease (CVD), can be opportunistically undertaken on bone density machine-derived vertebral fracture assessment (VFA). This study examined the association between ML-AAC24 extent (low, moderate, high) and subsequent Major Adverse Cardiovascular Events (MACE) in people undergoing osteoporosis screening in Manitoba, Canada.
Methods:
10,250 individuals (mean age 75.5 years, 94% female) without previous myocardial infarction or ischemic stroke underwent VFA during routine osteoporosis screening. ML-AAC24 scores were grouped according to established cut-points: low (≤1), moderate (2-5), and high (6-24). MACE (all-cause mortality, hospitalisation for acute myocardial infarction, or non-haemorrhagic cerebrovascular disease) rates were calculated by ML-AAC24 groups.
Results:
Over a mean follow-up of 3.9 years, MACE were recorded in 1,265 people (12.3%). Among those with low, moderate, and high ML-AAC24, MACE per 1,000 person-years were 18.4 (95% CI 16.4–20.5), 34.1 (95% CI 30.9–37.4), and 55.6 (95% CI 50.8–60.1), respectively; with a similar gradient observed after stratifying by age and sex. In those most likely to benefit from pharmaceutical intervention (aged <80 years, not currently on statins), MACE rates by ML-AAC24 groups were 13.5 (95% CI 11.5–15.8), 25.9 (95% CI 22.1–30.3) and 44.1 (95% CI 37.0–52.0), respectively. Conclusions: In people undergoing routine osteoporosis screening, individuals with moderate and high ML-AAC24 had greater MACE rates compared to low ML-AAC24. Automated AAC assessment during osteoporosis screening could help to identify people at higher risk of future MACE, particularly in older women who are under-screened and undertreated for CVD.