METABOLICALLY UNHEALTHY NORMAL WEIGHT (MUN) INDIVIDUALS ASSOCIATED WITH ELEVATED CARDIOVASCULAR DISEASE RISK SCORES DERIVED FROM RETINAL IMAGING

Therapeutic Area

ASCVD/CVD in Special Populations

Background

Traditional BMI-based classification of obesity may not adequately capture the variability in metabolic health among individuals with similar BMI values. Metabolically unhealthy normal weight (MUN) individuals have a normal BMI but exhibit unfavorable metabolic profiles, such as hypertension, dyslipidemia or insulin resistance, and may face elevated cardiovascular disease (CVD) risk. Emerging evidence shows that retinal imaging is a valuable non-invasive method to assess microvascular and metabolic health. This study quantified retinal-based CVD risk scores (Dr. Noon CVD) in MUN individuals to better understand the implications for CVD prevention and management.

Methods

In the data from the UK Biobank cohort (N=59,493), we identified MUN participants (n=12,605) based on their BMI (18.5 – <25 kg/m2) and presence of one or more of metabolic syndrome components. They were compared to metabolically healthy normal participants (MHN) (n=8,410) using multivariable linear regression. [In this study, metabolic syndrome components were defined as systolic BP>130 mmHg, on antihypertensive medication, diagnosed diabetic, waist-hip ratio >0.95 for women or >1.03 for men]. Dr. Noon CVD scores were derived from retinal images using a previously validated algorithm. The interaction effects of sex and ethnicity on Dr. Noon CVD scores in MUN and MHN were also evaluated. Lifestyle factors, including smoking status, physical activity levels and dietary habits were included to assess their impact on Dr. Noon CVD scores and account for possible confounding effects.

Results

MUN participants exhibited significantly higher (72.27% higher, p<0.0001) Dr. Noon CVD scores compared to MHN individuals, indicating increased CVD risk despite having a normal BMI. Lower scores were seen in female participants (adjusted effect: 0.027, p=0.039) and higher in white participants (adjusted effect: 0.098, p=0.041) in the MUN group. Additionally, lifestyle factors strongly influenced scores: smokers and physically inactive individuals demonstrated markedly higher CVD risk compared to their healthier counterparts. These findings highlight substantial heterogeneity in CVD risk among individuals with normal BMI based on metabolic health status.

Conclusions

In this study, we show that Dr. Noon CVD score was able to identify MUN individuals as a high-risk group. Dr. Noon could hence be used for early identification of metabolic abnormalities in normal-weight populations, allowing for timely management to mitigate CVD burden.