Sürücü index and others

Abstract

Background

In this study, we have aimed to evaluate the correlation between echocardiographic parameters that test systolic and diastolic function together.

Method

The study population was divided into two groups according to Vp. Group-1 (n = 103) represented the control group (Vp > 50 cm/s) and group-2 (n = 86) represented patients with systolic and diastolic dysfunctions together (Vp ≤ 50 cm/s). The echocardiographic parameters that evaluate systolic and diastolic function together, such as the Tei and the Sürücü indices, were compared between the groups.

Results

In group-2, the Tei index was higher (p = 0.001) and the Sürücü index was lower (p < 0.001). We also showed that the Tei and Sürücü indices were significantly and negatively correlated. That is, as the Sürücü index decreases, the Tei index increases (p = 0.001).

Conclusion

Vp is an index more affected by diastolic parameters but rarely by systolic parameters because it is measured at diastolic period. The Tei index, on the other hand, is affected by preload variables and needs two different heart cycles for calculation. The Modified Tei index, however, has limited diagnostic value because of high inter-observer variability. In this study, the usability of the Sürücü index is shown in comparison with other indices used for this purpose. Considering that it is less affected by preload variables, can be calculated over a single heart cycle, and has the ability to test variables of both systolic and diastolic periods unlike Vp. We postulate that the Sürücü index is more usable and reliable.

1. Introduction

Investigation of left ventricular (LV) function abnormalities separately as systolic and diastolic has been the conventional approach since the beginning of use of echocardiography (ECHO). However today, as a result of developments in ECHO methods and diversity of software (Color M-mode flow propagation velocity (Vp), tissue Doppler, strain rate, etc.), it no longer possible to separate precisely LV dysfunctions as systolic or diastolic. Previously, we reported that subtle systolic dysfunction may be associated even in patients with preserved ejection fraction (defined as LV diastolic dysfunction in the past) who subsequently developed diastolic heart failure.¹ Likewise in the cases of systolic heart failure and systolic left ventricular dysfunction (ejection fraction (EF) < 45%),² LV diastolic dysfunction can co-exist.^1,3,4 These findings show that in clinical studies, the use of ECHO parameters that test systolic and diastolic functions together will be inevitable in the near future. For this purpose, different ECHO parameters that evaluate both systolic and diastolic functions together, such as Vp, Tei index, and Modified Tei index, are now utilized. However, almost all of these indices have conditions that constrain the diagnostic value. It has been acknowledged that Vp tests diastolic dysfunctions conventionally and is affected relatively less by systolic dysfunctions.^5,6 The Tei index was defined in 1995 for the first time by Tei et al.⁷ By definition, it is calculated by adding isovolumetric contraction time to isovolumetric relaxation time and dividing by ejection time (Fig. 1). As seen clearly in the figure, two different times and two different heart cycles are required. Consequently, it is affected by many parameters such as heart rate, time between two different heart cycles, afterload and preload variables measured during the latter heart cycle, and inter-observer variability. In traces obtained by pulsed wave tissue Doppler imaging (pw-TDI), the ratio obtained by using the same formula is named as the Modified Tei index. Modified Tei index, obtained by this method, is calculated over the same heart cycle, which can be considered as an advantage. Unfortunately, the starting and ending times of isovolumetric contraction, and the isovolumetric relaxation and ejection times, being intertwined, may not be interpreted properly. Therefore, in studies using Modified Tei index, the inter-observer variability was even more compared to Tei index.⁸ For this purpose, considering the unique limitations of Vp,^5,6 Tei index⁹ and Modified Tei index,⁸ we have defined a novel index (Ea-Aa/Sa) using the pw-TDI parameters taken from LV basal-lateral annulus level.¹⁰ In this study, we aim to investigate the correlation between the Sürücü index and others. We also aim to demonstrate the superiority of the Sürücü index compared to the others.

Fig. 1.

Tei index calculation.

2. Method

The present study was conducted prospectively in Private Avcılar Anadolu Hospital, Private Beylikdüzü Kolan Hospital and Private Avcılar Hospital between February 2010 and May 2014. Exclusion criteria were as follows: high levels of urea, creatinine and fasting blood sugar, history of diabetes mellitus, detection of rhythm types other than sinus rhythm and complete bundle branch block in electrocardiography (ECG), detection of chronic obstruction lung disease history or findings, detection of moderate to severe valve dysfunction (regurgitation and/or stenosis) in ECHO. Thus, we aimed to collect a homogenous patient population that did not have any sign of systemic illness (exception of systemic hypertension). Subsequently, 189 cases were included in the study.

Height, weight, systolic and diastolic blood pressure (BP) and resting heart rate were measured. Body mass index (BMI) was calculated. ECHO examination was conducted in left lateral decubitus position using EnVisor C (Philips Medical Systems, Andover MA, USA), Vivid-3/expert Vingmed technology (GE medical system) and Vivid-S5/expert Vingmed technology (GE medical system) with a phased-array transducer of 3.5 MHz. An average of six to eight beats was collected for evaluation by avoiding images of pre and post extra-systolic beats. The collected images were recorded by the system. On the recorded images, at least three consecutive measurements were made and their mean was calculated.

In the parasternal long axis images, aortic valve diameter, left atrium (LA) diameter, LV end-diastolic diameter, LV end-systolic diameter were obtained. Ejection fraction (EF) was measured by Teichholz method.¹¹ Using the formula defined by Devereux et al,¹² the LV mass index (LVMI) was measured on the parasternal short axis images.¹³ Using the apical four and five chamber images transmitral E, transmitral A velocities, E-velocity deceleration time (E-dt), aortic velocity and aortic velocity time integral were obtained. From the same apical four-chamber view, the color Doppler sector map of mitral inflow was adjusted to obtain the longest column of color flow from the mitral annulus and apex. An M-mode cursor was placed through the center of this flow, avoiding boundary regions. The Vp was measured as the slope of the first aliasing velocity during early filling (Fig. 2), from the mitral valve plane to 4 cm away into LV cavity.¹⁴

Fig. 2.

Color M-mode flow propagation (Vp) velocity.

For pw-TDI images, the minimal output set up (which could avoid ultrasonographic background noise of blood flow velocities) was used by lowering the filter level down to 50 Hz. The velocity limits of Doppler trace were kept at the lowest possible level, often between 20 and +20 cm/s. Simultaneous ECG record was obtained at 50 mm/s. The pw-TDI parameters were recorded at basal-lateral mitral annulus level of LV on apical four chamber images. Among the pw-TDI systolic parameters, systolic velocity (Sa), Q-Sa interval which represents the time period between the Q wave in simultaneous ECG and the peak Sa velocity were measured (Fig. 3). Lastly, the Sürücü index was calculated by using the pw-TDI parameters taken from LV basal-lateral annulus level, by subtracting Aa velocity from Ea velocity and dividing it by Sa velocity (Ea-Aa/Sa).¹⁰

Fig. 3.

Tissue Doppler imaging velocities that taken basal-lateral mitral annulus.

In general practice, it is known that the Vp value is more than 50 cm/s in healthy subjects,¹⁵ we preferred to investigate our study population by dividing it into two groups based on the Vp value. Group-1 (n = 103) represented the control group (Vp > 50 cm/s) and group-2 (n = 86) represented patients with systolic and diastolic dysfunctions together (Vp ≤ 50). Data are expressed as mean ± standard deviation. Independent sample T-test was used to compare the groups (SPSS for Windows version 10.0.1 manufactured by SPSS, Chicago, IL, USA). The inter-observer and intra-observer variability of the first fifty patients were calculated by Pearson's correlation test using the data generated by two cardiologists (HS and CB) for the inter-observer variability, by one cardiologist (HS) for intra-observer variability. A p-value less than 0.050 was considered significant.

3. Results

The general demographic features of the groups are presented in Table 1. The patients' sex, BMI, waist circumference, heart rate, systolic and diastolic BP's were not different among the groups. On the other hand, patient's age was older in group-2 (p < 0.001). We also evaluated medical treatment regimens among the groups. Twenty-two (12%) of the cases that were included in the study were not under medical treatment. Among the cases who were under medical treatment, 82 (43%) were taking β-blocker, 50 (26%) were taking ACE inhibitor, 34 (18%) were taking ACE inhibitor and diuretic, 28 (15%) were taking calcium channel blocker, 15 (8%) were taking AII receptor blocker and 32 (17%) were taking AII receptor blocker and diuretic. There was no difference between the groups for medical treatment strategy.

Table 1.

General characteristics of the patients.

Parameters	Group 1 (n = 103)	Group 2 (n = 86)	p value
Age	54 ± 9	59 ± 10	p < 0.001
Female/Male ratio	64/39	57/29	NS
BMI	31 ± 5	30 ± 6	NS
Waist circumference (cm)	105 ± 12	104 ± 13	NS
Systolic BP (mmHg)	141 ± 23	139 ± 20	NS
Diastolic BP (mmHg)	84 ± 10	85 ± 10	NS
Heart rate	72 ± 11	73 ± 12	NS

BMI: body mass index, BP: blood pressure, LVMI: left ventricular mass index.

The left heart echocardiographic parameters (Doppler, M-mode and pw-TDI) in the groups are given in Table 2. Among the geometric parameters of left heart (EF, RWT, LWMI, LA and LV dimensions), only the LV diastolic dimension was prominent in group-2 (p = 0.034). In group-2, transmitral E velocity was lower and e-dt was longer (respective p value: <0.001, 0.038), but transmitral A velocity was prominent (p = 0.006). The most striking finding was however difference between the Tei and the Sürücü indices. In group-2, Tei index was higher (p = 0.001) and Sürücü index was lower (p < 0.001). We also looked for the correlations between the Tei and the Sürücü indices (Fig. 4). As the Sürücü index decreased, the Tei index increased (p = 0.001). Furthermore, we investigated correlations between all index and patient's variability (such as age, sex, BMI and LVMI). These correlations are given Table 3. We found that the Sürücü index was negatively correlated with age and LVMI (respective p value: p < 0.001, 0.001). Finally, persons who had a normal coronary artery and had not taken any medication (clinically normal population) were investigated. In this population (n = 26), the Sürücü index was found to be 0.20 ± 0.51.

Table 2.

The value of left ventricular standard echocardiographic parameters.

Parameters	Group 1 (n = 51)	Group 2 (n = 87)	p value
Aortic valve diameter (mm)	30.5 ± 3.3	29.7 ± 4.1	NS
Left atrium (mm)	34.2 ± 5.4	34.3 ± 5.6	NS
LV end-diastolic diameter (mm)	48.0 ± 4.8	48.8 ± 6.2	NS
LV end-systolic diameter (mm)	29.5 ± 4.7	29.6 ± 5.6	NS
Ejection fraction (%)	68.2 ± 7.1	69.2 ± 8.4	NS
Relative wall thickness (%)	44.1 ± 7.1	43.4 ± 7.0	NS
LVMI (gr/m2)	119.9 ± 36.2	130.1 ± 38.4	NS
Transmitral E velocity (m/sc)	0.71 ± 0.16	0.72 ± 0.16	NS
Transmitral A velocity (m/sc)	0.74 ± 0.20	0.83 ± 0.23	p = 0.014
E-velocity deceleration time (msc)	210 ± 48	213 ± 51	NS
LV stroke volume (ml)	42.7 ± 9.0	43.6 ± 13.44	NS
MAPSE (mm)	18.5 ± 3.8	17.6 ± 3.5	NS
MAPSE-acceleration (cm/sc)	7.14 ± 1.7	6.11 ± 1.5	p < 0.001
Basal-lateral systolic velocity (cm/sc)	13.49 ± 2.04	10.38 ± 1.61	p < 0.001
Q-Sa interval (msn)	183 ± 41	248 ± 51	p < 0.001

LV: left ventricular, LVMI: left ventricular mass index, MAPSE: mitral annular plane systolic excursion, Q-Sa interval: represents the time period between the Q wave in simultaneous ECG and the peak Sa velocity.

Fig. 4.

Correlation between Tei index and Sürücü index.

Table 3.

Correlation between all index and patient's variability (such as age, sex, BMI and LVMI).

	Age	Sex	BMI	LVMI
Sürücü index	p < 0.001 (negatively)	NS	NS	p = 0.001 (negatively)
Vp	p < 0.001 (negatively)	NS	NS	NS
Tei index	NS	NS	NS	NS

BMI: body mass index, LVMI: left ventricular mass index, Vp: color M-mode flow propagation velocity.

The inter-observers variability for TDI was 5.3%, r = 0.98, when the time period was taken into account and 6.1%, r = 0.97, when the peak velocity was taken into account. The intra-observers variability was %5, 3, r = 0, 98, when the time period was taken into account, and 4.7%, r = 0.99, when the peak velocity was taken into account.

4. Discussion

Standard pulsed-wave Doppler ECHO provides the temporal distribution of blood flow velocities at a specific location, whereas Vp provides the spatiotemporal distribution of these velocities across a vertical line. Thus, the information displayed in a color M-mode recording of LV inflow is comparable to that given by multiple simultaneous pulse Doppler tracings obtained at different levels from the mitral orifice to the LV apex. The Vp value that describes the speed of propagation (slope of the black to red transition at the leading edge of the transmitral E wave as shown that Fig. 2) is enhanced with rapid relaxation and LV suction.¹⁶ Unfortunately, in the limited literature, it has been shown that Vp value is decreases in LV systolic dysfunction.^5,6 Brun et al¹⁷ demonstrated a significant negative correlation between Vp and time constant of relaxation (τ) in various groups of patients with coronary disease and cardiomyopathy with systolic dysfunction, suggesting that a rapid ventricular relaxation (short τ) promotes a faster propagation of blood into the ventricle. These findings suggest that Vp represents a useful noninvasive index for assessing LV relaxation and also systolic function. Using the ratio of transmitral E velocity (influenced by preload) to Vp value (not influenced by preload), Garcia et al⁵ determined that with 87% sensitivity, this ratio is positively correlated with pulmonary capillary wedge pressure (PCWP), measured by an invasive procedure. This study also demonstrated that Vp was effected by systolic dysfunctions. In our study, a strong correlation has been ascertained between Vp and Tei/Sürücü indices that test systolic and diastolic dysfunctions together. With Vp value less than 50 cm/s, it was observed that the Sürücü index decreased and the Tei index increased specifically. Therefore, it is obvious that the newly defined Sürücü index has a meaningful correlation both with Tei index and Vp value. As per definition the Sürücü index tests both systolic and diastolic functions; however, considering the constrained diagnostic value of Vp in testing systolic functions, it could be said that the Sürücü index is more valuable, easy to use and not affected by instantaneous hemodynamic differences (due to it being calculated by the same heart cycle).

Other indices that test both systolic and diastolic functions are the Tei and the Modified Tei index. The use of Modified Tei index is limited due to the high ratio of inter-observer variability. In our study, where the newly defined index is used, inter-observer variability ratio was 5.3% compared to 21.5% in a study which modified Tei index was used.⁸ Therefore, it could be said that the diagnostic value of Sürücü index is higher than that of the Modified Tei index. On the other hand, Tei index was known to be simple, reproducible and independent of age, heart rate, mean arterial pressure and the degree of mitral regurgitation.² Unfortunately, the Tei index is also significantly related to preload alterations.¹⁸ The Tei index is a measure of combined systolic and diastolic LV function.^7,19–21 Tei et al¹⁹ demonstrated that LV positive and negative dp/dt can be noninvasively evaluated by the Tei index. It has already been applied clinically in patients with dilated cardiomyopathy,² cardiac amyloidosis,²² primary pulmonary hypertension,²³ acute myocardial infarction^24,25 and congestive heart failure.¹⁸ Thus, it is obvious that the Tei index is beneficial.^2,7,18–25 However, by definition (Fig. 1), heart cycles at two different times are needed to be calculated. Hemodynamic variabilities between two heart cycles constrain the diagnostic value of the method. Besides, Tei index parameters are affected specifically by preload variabilities,¹⁸ and it is known that parameters (pw-TDI) used in Sürücü index are less affected by them.^26,27 On the other hand Sürücü index is simpler and easier to calculate compared to the Tei index, besides being less affected by variabilities (such as preload), which could be an important advantage.

Our first survey for the newly defined Sürücü index is based on the study published in 2009.¹ In this study, we identified first decrease in Ea velocity, then a little increase in Aa velocity and decrease in Sa velocity in advanced PCWP values along with PCWP increase. We have now defined a new index by combining these findings which would yield similar or even superior results as compared to the Tei index. In our first study with this consideration, we found a strong correlation between LV diastolic dysfunction and Sürücü index.¹⁰ In this study, we have aimed to compare the newly defined Sürücü index with the other predefined indices (Vp, Tei index and Modified Tei index) used for this purpose. We have found that when Sürücü index is compared to Vp value, evaluating both systolic and diastolic parameters together is an advantage that the Sürücü index has. However, Vp value is a parameter exclusively measured at the diastolic period. Because of this, in the literature, the number of studies showing it being affected by systolic dysfunctions are limited.^5,6

Sürücü index has many similarities when compared with the Modified Tei index. The most important similarity is that both use pw-TDI parameters and are calculated over the same heart cycle. However, the use of time intervals of pw-TDI traces in modified Tei index and the time intervals being intertwined decreases the sensitivity of this method and causes the inter-observer variability value to be very high. In our study, on the contrary, inter-observer variability values are within acceptable limits because wave amplitudes are used instead of time intervals. Evaluated from this point of view, it could be said that diagnostic value of Sürücü index is more acceptable than that of the Modified Tei index. While comparing Tei index and Sürücü index, Sürücü index is less affected by preload variabilities and calculated over a single heart cycle, which is an advantage.

5. Conclusion

Considering that it is less affected by preload variables, can be calculated over a single heart cycle, and has the ability to test variables of both systolic and diastolic periods. It can be said that the Sürücü index is more usable and reliable.

6. Study limitations

Our study had two limitations. First one is the normal and pathological values of the newly defined Sürücü index are not apparent. In this study, Sürücü index was 0.20 ± 0.51 in the populations who do not use any medication and had normal coronary arteries. Therefore, the value 0.20 and over could be the normal value of the Sürücü index. Considering Fig. 4, 0.10/0.20 could be classified as subtle dysfunction, 0.00/0.10 as silent dysfunction, −0.10/0.00 as moderate dysfunction, −0.20/−0.10 as prominent dysfunction, −0.20 and over as severe dysfunction.

The second limitation of our study is its use on patients with atrial fibrillation. Because in such a case, Aa velocity on the pw-TDI trace will disappear. Therefore, the diagnostic value of the Sürücü index for these patients will be evaluated in a new study. For this purpose, a new study has been started on patients with non-valvular atrial fibrillation, and we hope to publish the results soon.

Conflicts of interest

The authors have none to declare.