Canadian Journal of Cardiology



      Heart failure with preserved ejection fraction (HFpEF) predominantly affects women, although the basis for this female predominance is poorly understood. HFpEF is characterized by increased arterial stiffness and hemodynamic load, and recently coronary microvascular dysfunction and rarefaction have been described as novel arterial abnormalities in this syndrome. However, the role of arterial load on coronary microvascular dysfunction in men and women remains unknown. Thus, we sought to evaluate sex-specific associations of hemodynamic load with coronary microvascular function in subjects without heart failure.

      Methods and Results

      Subjects with a cardiac 82Rb positron-emission tomography between 2010 and 2013, ejection fraction ≥ 50%, no history of heart failure, dyspnea or coronary artery disease, and no regional perfusion defects were eligible. Left ventricular microvascular function was assessed by hyperemic stress/rest myocardial flow reserve (MFR). Steady and pulsatile components of arterial load were estimated by systemic vascular resistance index [SVRi = (80*mean arterial pressure/ cardiac output)*BSA] and indexed arterial compliance [ACi=(stroke volume/pulse pressure)/BSA], respectively. “Low MFR” was defined as the lowest sex-specific MFR quartile. Multivariable linear and logistic regression analyses adjusted for age, creatinine, heart rate, hypertension, diabetes, dyslipidemia, smoking and use of aspirin, statins and anti-hypertensives evaluated associations of SVRi and AoCi with MFR and “Low MFR”, respectively. Only variables with univariate P≤0.20 were included in final models. Interaction terms for sex and arterial load measures were included, and if significant (P≤0.10), sex-specific regression analyses were performed. 297 subjects (61% women, age: 61.4±11.0 years) were included. SVRi was higher [5,348±1676 vs. 4,616±1515 (dyne*s/cm5)*m2] and AoCi was lower [0.39±0.16 vs. 0.52±0.28 (mL/mmHg)/m2] in women (P<0.0001 each). Unadjusted associations of different combinations of ACi and SVRi with MFR are shown in the Figure. The interaction term sex*ACi was significant in predicting MFR (P=0.02) and “Low MFR” (P=0.06). Results of sex-specific models are shown in the Table. Associations of SVRi and AoCi with MFR and “Low MFR” were present in women (P≤0.02 for each) but not in men (P≥0.21 for each). Findings persisted after additional adjustment for resting myocardial flow, left ventricular mass and diastolic function.


      In subjects at highest risk for HFpEF (older women), lower steady and higher pulsatile arterial load were associated with alterations in MFR. Since coronary perfusion is related directly to pressure pulsatility, a combination of lower peripheral resistance and higher aortic stiffness (lower AoCi) in women may adversely affect the coronary microvasculature, a mechanism that could predispose to HFpEF.
      Figure thumbnail fx1
      Figure thumbnail fx2