Association of ambulatory blood pressure with coronary microvascular and cardiac dysfunction in asymptomatic type 2 diabetes

Abstract

Background: Type 2 diabetes (T2D) and hypertension commonly coexist and are associated with subclinical myocardial structural and functional changes. We sought to determine the association between blood pressure (BP) and left ventricular (LV) remodeling, systolic/diastolic function, and coronary microvascular function, among individuals with T2D without prevalent cardiovascular disease.

Methods: Participants with T2D and age-, sex-, and ethnicity-matched controls underwent comprehensive cardiovascular phenotyping including fasting bloods, transthoracic echocardiography, cardiovascular magnetic resonance imaging with quantitative adenosine stress/rest perfusion, and office and 24-h ambulatory BP monitoring. Multivariable linear regression was performed to determine independent associations between BP and imaging markers of remodeling and function in T2D.

Results: Individuals with T2D (n = 205, mean age 63 ± 7 years) and controls (n = 40, mean age 61 ± 8 years) were recruited. Mean 24-h systolic BP, but not office BP, was significantly greater among those with T2D compared to controls (128.8 ± 11.7 vs 123.0 ± 13.1 mmHg, p = 0.006). Those with T2D had concentric LV remodeling (mass/volume 0.91 ± 0.15 vs 0.82 ± 0.11 g/mL, p < 0.001), decreased myocardial perfusion reserve (2.82 ± 0.83 vs 3.18 ± 0.82, p = 0.020), systolic dysfunction (global longitudinal strain 16.0 ± 2.3 vs 17.2 ± 2.1%, p = 0.004) and diastolic dysfunction (E/e' 9.30 ± 2.43 vs 8.47 ± 1.53, p = 0.044) compared to controls. In multivariable regression models adjusted for 14 clinical variables, mean 24-h systolic BP was independently associated with concentric LV remodeling (β = 0.165, p = 0.031), diastolic dysfunction (β = 0.273, p < 0.001) and myocardial perfusion reserve (β = - 0.218, p = 0.016). Mean 24-h diastolic BP was associated with LV concentric remodeling (β = 0.201, p = 0.016).

Conclusion: 24-h ambulatory systolic BP, but not office BP, is independently associated with cardiac remodeling, coronary microvascular dysfunction, and diastolic dysfunction among asymptomatic individuals with T2D. (Clinical trial registration. URL: https://clinicaltrials.gov/ct2/show/NCT03132129 Unique identifier: NCT03132129).

Keywords: Ambulatory blood pressure; Blood pressure; Diabetic cardiomyopathy; Myocardial perfusion reserve; Type 2 diabetes.

Fig. 1

Study flowchart. LV indicates left ventricle; MRI: magnetic resonance imaging; T2D: type 2 diabetes

Fig. 2

Left ventricular (LV) mass/height ratio (A), LV ejection fraction (B), LV mass/volume ratio (LVM/V) (C), myocardial perfusion reserve (MPR) (D), global longitudinal strain (GLS) (E), and diastolic function (E/e') (F) in type 2 diabetes (T2D) compared to controls. NS indicates not significant

Fig. 3

MRI images displaying the 4-chamber view (A) and mid- ventricular short-axis slice (B) during diastole, global longitudinal strain assessment (GLS) (C), and stress myocardial perfusion map in mid-ventricular short-axis slice (D). Top row images were from a 59-year-old male with type 2 diabetes (T2D), BMI of 40 kg/m²; left ventricular mass (LVM) 146 g, left ventricular end-diastolic volume (LVEDV) 122 mL, left ventricular mass to volume ratio (LVM/V) 1.2 g/mL, GLS 13.0%, stress myocardial blood flow (MBF) 1.2 mL/min/g, rest MBF 0.6 mL/min/g, and myocardial perfusion reserve (MPR) 2.0. The bottom row images were from a 64-year-old male non-diabetic control, BMI of 32 kg/m²; LVM 179 g, LVEDV 223 mL, LVM/V 0.8 g/mL, GLS 15.3%, stress MBF 2.2 mL/min/g, rest MBF 0.6 mL/min/g, and MPR 3.7. Note for GLS with darker blue indicates greater shortening and higher strain

Fig. 4

Correlation between ambulatory 24-h systolic blood pressure (SBP) with left ventricular mass/volume ratio (LVM/V) (A), myocardial perfusion reserve (MPR) (B), global longitudinal strain (GLS) (C), diastolic function (E/e’) (D), and myocardial extracellular volume fraction (ECV) (E) in people with type 2 diabetes. p values displayed are adjusted for age, male sex, white ethnicity, never smoker, body mass index, HbA1c, number of anti-hypertensives, insulin, sodium-glucose co-transporter-2 inhibitor, and glucagon-like peptide-1 receptor use