4 Discussion
According to the recent progress in treatment of AS16, more accurate and adequate assessment is required for the severity of AS from the prognostic aspect.17,18 Echocardiography is the main tool for assessing AS to obtain Vmax, mPG, and AVA as the major indexes of the severity of AS non-invasively and repeatedly.3,4 However, one or two indexes of them may sometimes discord with each other, and then require alternative solution. One of the possible situations was in case of LFLG AS, which requires dobutamine stress echocardiography for correct diagnosis3–7, though we did not include such cases in this study. Another situation may be due to technical difficulty, not able to obtain good echocardiographic images, suboptimal Doppler signals with ambiguous envelope in the real world.
Here, the ratio of AT to ET of transaortic Doppler flow signal, i.e. AT/ET, is recently reported as a reliable index of AS severity8-12, and positively correlated with peak V, mPG, and negatively correlated with AVA or AVAi. This index has potential to be added as a useful information of AS severity in such cases.
Some investigators have already reported that the usefulness of AT/ET. Nakamura et al reported that AT/ET correlated well with peak V and catheter derived mPG in 1987.8 Zekry reported AT/ET was useful in assessing dysfunction of prosthetic valve in aortic valve position.13 AT/ET is an index of easy measurement, easy recognition by eyes, and more importantly it is independent from the transaortic flow rate. Therefore, AT/ET is useful in differentiating true severe LFLG AS and pseudo AS, as Kamimura et al11and Gamaza-Chulián12 reported. Griguer et al reported that prognosis was poor in patients with severe AS of AT/ET>0.36.19
However, these reports were based on the data acquired from no matter which acoustic window of the fastest transaortic velocity for a single patient with AS or they assumed as the identity of the AT/ET obtained from any acoustic windows. To obtain the true peak V of patients with AS, multiple acoustic windows approaching is recommended, while time interval measurement for AT/ET is able in apical or right parasternal acoustic window.20,21 However, the results of time interval measurement obtained from each acoustic window have not been compared so far, therefore, we investigated the effects of acoustic windows or approaching on AT/ET measurement by measuring AT, ET, and AT/ET on continuous Doppler transaortic flow waves acquired from apical (3C and 5C) and R acoustic windows in the same patients, and found the non-uniform results between different acoustic windows.
In this study, there was no significant difference in cET between apical (3C and 5C) and R approach in patients with AS. However, cAT was significantly greater in R approach than apical approach, then AT/ET was also greater in R approach than apical approach.
There are some possible reasons for greater AT/ET by R approach than apical approach. One is the difference in peak V caused by approaches. Thaden et al reported that the highest peak V of patients with severe AS was achieved by R approach in nearly half of their cases.22 Since severe AS was also more than half in our 81 patients in our study and the higher peak V was obtained by R approach rather than apical approach (3C or 5C), time interval difference from earlier suboptimal peak V to later true peak V might cause greater AT measurement value.
Another possibility was an error in time-interval measurement caused by R approach. A flow characteristic of R approach is turbulent signal in Valsalva sinus after severely restricted aortic valve as recognized by color Doppler image. We cannot exclude the possibility of the effect of the turbulent flow signal on the envelope of the continuous wave Doppler signal which may make difficult to recognize the exact timing of the peak velocity without clear envelope of Doppler signal. Because of this reason, the variability of the inter-observer measurement difference was greater in R approach than in apical approach. Therefore, we need to pay an attention to measure AT/ET by R approach, though it is superior in peak V detection by R approach to apical approach.
Additionally, we consider the effect of HR on ET or AT measurement since ET or AT is affected by HR. Zoghbi et al23 reported that ET shortened by 15 msec as 10 bpm increase in HR in patients with AS in HR range of 55-110 bpm. Also, Reant et al24reported that ET shortened by 15 msec per 10 bpm increase in HR in AS with reduced LV systolic function. However, the ratio AT/ET itself is reported to be independent of HR19, thus the effect of HR on AT/ET may be minor.
Finally, AT/ET is supposed to be a useful index of AS severity because this is easy to measure, easy to recognize, and applicable to LFLG AS.12 However, it needed to pay attention to the effect of acoustic windows on the decision of AS severity by AT/ET as revealed in the study.
Limitation of study: This study did not include AS patients with decreased LVEF so-called LFLG AS, or patients with atrial fibrillation, or patient with other significant valvular lesions. The exact cause of the AT/ET discrepancy has not been determined in the clinical situation though possible cause has mentioned in the text.