Metabolic health, obesity, and coronary microvascular dysfunction: insights from a nationwide coronary microvascular dysfunction registry
H Tanaka, Y Matsuura, K Kaikita, Y Suiko, K Moribayashi, T Kakuta, K Tsujita, M Takeda, H ShimokawaAbstract
Background
The prevalence of coronary microvascular dysfunction (CMD) and its endotype distribution across metabolic–obesity phenotypes remain unclear.
Purpose
To examine associations of metabolic–obesity phenotypes with invasively assessed CMD and to compare CMD endotypes across phenotypes.
Methods
We analyzed 1,289 patients from a nationwide coronary microvascular dysfunction registry. Obesity was defined using an Asian-specific cutoff as body mass index ≥23 kg/m². Metabolic unhealthiness was defined as ≥1 of the following: triglycerides ≥150 mg/dL, high-density lipoprotein cholesterol <40 mg/dL, fasting glucose ≥110 mg/dL, or hypertension. Patients were classified as metabolically healthy non-obesity (MHNO), metabolically unhealthy non-obesity (MUNO), metabolically healthy obesity (MHO), or metabolically unhealthy obesity (MUO). CMD was defined as an index of microcirculatory resistance (IMR) ≥25 and/or coronary flow reserve (CFR) <2.0, and CMD endotypes were categorized as isolated high IMR (CFR ≥2.0 and IMR ≥25), functional CMD (CFR <2.0 and IMR <25), or structural CMD (CFR <2.0 and IMR ≥25).
Results
The cohort comprised MHNO (n=188, 14.6%), MUNO (n=375, 29.1%), MHO (n=112, 8.7%), and MUO (n=614, 47.6%). Metabolic unhealthiness was more common in obesity than in non-obesity (84.6% vs 66.6%, p<0.001). Overall, CMD was present in 35.3% and was more prevalent in metabolically unhealthy than in metabolically healthy phenotypes regardless of obesity status (MUNO 39.7% and MUO 38.8% vs MHNO 22.9% and MHO 22.3%; p<0.001). In multivariable logistic regression analyses using MHNO as the reference group, MUNO and MUO were independently associated with higher odds of CMD (MUNO: adjusted odds ratio [aOR] 1.81, 95% confidence interval [CI] 1.20–2.74; p=0.005; MUO: aOR 1.98, 95% CI 1.33–2.94; p<0.001), whereas MHO was not associated with CMD (aOR 1.02, 95% CI 0.58–1.82; p=0.928). CMD endotype patterns also differed by phenotype, such that within obesity MUO had higher proportions of structural CMD (10.9% vs 2.7%, p=0.005) and a trend toward higher functional CMD (6.2% vs 1.8%, p=0.070) than MHO, despite similar isolated high IMR prevalence (21.7% vs 17.9%, p=0.450), whereas among metabolically healthy patients, obesity was associated with a trend toward more isolated high IMR (17.9% vs 10.1%, p=0.075) and significantly less functional CMD (1.8% vs 8.5%, p=0.022) than non-obesity.
Conclusion
In this nationwide registry, CMD was associated with metabolic health status irrespective of obesity status, and CMD endotype patterns within obesity differed by metabolic health status, supporting metabolic health–based obesity phenotyping for CMD risk assessment.