Hemodynamic and glucometabolic factors in the prediction of left ventricular filling pressures

OBJECTIVE: To explore possible hemodynamic and glucometabolic determinants of left ventricular filling pressures as assessed by the non-invasive surrogate marker, averaged E/é, in otherwise healthy, middle-aged male survivors from a random population sample.

DESIGN AND METHODS: Prospective population-based cohort study examining associations between hemodynamic factors [systolic blood pressure (SBP), heart rate (HR)), glucometabolic factors (fasting blood glucose, fasting plasma insulin, Homeostatic Model Assessment (HOMA) derived indices of beta-cell function (HOMA-2B) and insulin sensitivity (HOMA-2S)], other traditional cardiovascular risk factors [age, smoking status, body mass index (BMI), total serum cholesterol, serum creatinine] assessed at baseline, and values of E/é assessed at follow-up examination, using multivariable linear regression analysis (significance level 0.05, p-stay 0.20 on multivariable analysis). Subjects with prevalent cardiovascular disease and/or diabetes mellitus were excluded. E/é was positively skewed and, therefore, naturally log-transformed, as was fasting plasma insulin. HOMA-indices were assessed as continuous variables, both non-transformed and after natural log-transformation, as well as categorically, using quartiles. Study subjects were included 1974-1992, whilst the follow-up with echocardiography was performed 2002-2006.

RESULTS: The final study population comprised 246 men with a median (IQR) age of 47 (47-48) years. Median (IQR) follow-up time was 28 (27-28) years, and median (IQR) E/é was 10 (8-12). In univariable analyses, E/é was associated positively with higher age, BMI, and serum creatinine, and negatively with shorter follow-up time. The multivariable model (adjusted r = 0.15) included all of these variables, i.e. age (beta = 0.016 per year [95% confidence interval (CI), 0.006 to 0.027]; p = 0.002), BMI (beta = 0.03 per kg/m [95% CI, 0.02 to 0.04]; p < 0.0001), serum creatinine (beta = 0.002 per micromole/l [95% CI, -0.001 to 0.005]; p = 0.18), and time elapsed between baseline examination and echocardiography (beta = -0.03 per year [-0.06 to -0.01]; p = 0.01). We did not find any significant interactions in the prediction of E/é.

CONCLUSION: In a prospective population-based cohort study including apparently healthy, middle-aged male subjects, higher age, BMI, and creatinine, but not SBP or HR, were significantly associated with higher left ventricular filling pressures as assessed by averaged E/é.