Background: In patients with severe aortic valve stenosis (AS), the onset of heart failure is associated with increased mortality and higher operative risk. Heart failure may result from either systolic, diastolic, or "overall" left ventricular dysfunction. The index "isovolumic contraction time and isovolumic relaxation time divided by ejection time" was shown to be a sensitive indicator of "overall" cardiac dysfunction in patients with dilated cardiomyopathy and cardiac amyloidosis. We sought to define the role of the Tei index in patients with severe AS and to validate this index against conventional measures of systolic and diastolic LV function.
Patients and methods: Fifty-three participants underwent left heart catheterization for invasive measurement of LV end-diastolic pressure as a marker of diastolic function: 10 AS patients (valve orifice 0.6 +/- 0.2 qcm) with depressed systolic LV function (defined by LV ejection fraction < or = 45% [mean 32% +/- 8%], 7 male/3 female, 72 +/- 10 years old, DAS group), 22 AS patients (valve orifice 0.7 +/- 0.2 qcm) with preserved systolic LV function (ejection fraction > 45% [mean 55% +/- 6%], 13 male/9 female, 71 +/- 11 years old, PAS group) and 21 asymptomatic control participants (ejection fraction > 45% [mean 62% +/- 8%], 14 male/7 female, 66 +/- 8 years old, CON group). Within 24 hours from catheterization, conventional 2-dimensional and Doppler echocardiographic examination including measurement of the Tei index was performed.
Results: LV end-diastolic pressure was elevated in the DAS and in the PAS group in comparison with control participants (32 +/- 6 mm Hg and 22 +/- 7 mm Hg vs 11 +/- 4 mm Hg, respectively, P <.01 for both comparisons). DAS patients were in a higher New York Heart Association functional class than PAS patients (3.2 +/- 0.4 vs 2.2 +/- 0.4, P <.001) The Tei index was easily and reproducibly obtained in all study participants. In the DAS group, isovolumic contraction time was prolonged and ejection time was shortened in comparison with the CON group (102 +/- 20 ms vs 52 +/- 15 ms, P <.01; and 235 +/- 44 ms vs 316 +/- 45 ms, P <.01), resulting in a significantly increased Tei index (0.78 +/- 0.28 vs 0.40 +/- 0.11, P <.01). In the PAS group, isovolumic relaxation time was shortened (62 +/- 18 ms vs 81 +/- 26 ms for the CON group, P <.01) and ejection time was prolonged (335 +/- 34 ms vs 316 +/- 45 ms for the CON group, P <.05), resulting in a decreased Tei index (0.29 +/- 0.12 vs 0.40 +/- 0.11, P <.05). Receiver operating characteristic curve analysis for the Tei index yielded an area under the curve of 0.98 +/- 0.03 for separating DAS and PAS patients. Using a Tei index greater than 0.42 as a cutoff, DAS patients were identified with a sensitivity of 100% and a specificity of 91%.
Conclusion: The Tei index is significantly increased in patients with severe AS and depressed overall cardiac LV function. In AS patients with predominant diastolic dysfunction, in whom systolic function is preserved, the index is decreased in comparison with control patients. The index differentiates between symptomatic AS patients with depressed and less symptomatic AS patients with preserved systolic LV function, and may thus provide relevant information in the work-up and care of such patients.