Relationship between tei index and left ventricular geometric patterns in a hypertensive population: a cross-sectional study

The relationship between Tei Index (TI) and left ventricular (LV) geometric patterns has not been previously well described. The present study therefore set out to describe the nature of this relationship if any, and to also assess whether a relationship exists between the geometric patterns and LV ejection fraction (LVEF) so as to establish a basis for comparison.

The study was carried out in the echocardiography laboratory of Aminu Kano Teaching Hospital (AKTH) in Kano, North-Western Nigeria. The study was cross-sectional in design. Hypertensive subjects referred for echocardiography to AKTH were serially recruited from October 2008 to September 2009. TI was defined as the sum of isovolumic contraction and relaxation times divided by the ejection time, and values of LV TI < 0.40 were considered normal, while higher values were considered abnormal. Four patterns of LV geometry (normal, concentric remodelling, concentric LV hypertrophy and eccentric LV hypertrophy) were determined from the LV mass index and LV relative wall thickness as previously described. Binary logistic regression models and Pearson's Correlation (r) Coefficient were used to analyse the associations between TI or LVEF and a number of variables.

A total of 142 subjects were recruited into the study. The prevalence of abnormal TI (26.8%; 38 persons) in the total population was lower than that of reduced LV ejection fraction (< 50%) (38.0%; 54 persons) (p = 0.335). There was no association between any LV geometric pattern and abnormal TI. However, there was significant relationship between the geometric patterns and low LVEF (< 50%); tested in a binary logistic regression model. HR was a significant predictor of TI with regression coefficient of -0.218, 95% confidence interval (CI) of -0.005 - < -0.001 and p-value of 0.011. Similarly, HR was the only variable that significantly predicted abnormal TI in a binary logistic regression model with an odds ratio of 1.058 (95% CI = 1.002-1.118; p = 0.044), and also the only variable that correlated with TI significantly (r = -0.212; p-value = 0.014).

This study has found that LV geometric patterns and LVEF were not associated with TI in hypertensives, but there was strong association between LV geometric patterns and LVEF. TI was found to be dependent on HR.