Comparison of Systolic Time Ratio from Electrical Cardiometry Versus Echocardiography for Evaluation of Left Ventricular Systolic Function in Patients Undergoing Cardiac Surgery: A Prospective Observational Study

INTRODUCTION

In patients undergoing cardiac surgery, left ventricular (LV) systolic function is an important predictor of cardiovascular outcomes.^[1] Left ventricular ejection fraction (LVEF) is a commonly used measure of LV systolic function. Patients with low pre-operative LVEF represent a group with higher surgical risk. These patients have an increased risk for post-operative complications and mortality after cardiac surgery due to reduced cardiovascular reserve.^[2] Therefore, early recognition of such patients is of paramount importance.

LVEF can be easily assessed by echocardiography (ECHO) in the intraoperative or intensive care setting. However, the reproducibility of LVEF has been a matter of controversy. Recently, new techniques such as tissue Doppler imaging, 3D evaluation, and speckle tracking ECHO have been proposed to quantify LV systolic function more precisely, but these methods are technically more complex, time-consuming, and user dependent.^[3]

Historically, systolic time ratio (STR) was a measure in evaluating LV performance but is no longer frequently used.^[4-8] STR is defined as the ratio of the pre-ejection period (the time from the onset of the QRS complex on the electrocardiogram to the opening of the aortic valve) and the LV ejection time (time from the aortic valve opening to the aortic valve closing).

STR can be measured by the ICON monitor (Osypka Medical, Berlin, Germany), which uses the principle of electrical cardiometry (EC) and can also be measured by pulse Doppler ECHO.^[9] Thoracic electrical bioimpedance (TEB)-based EC measures cardiac output (CO) and other hemodynamic parameters noninvasively.^[10] For this purpose, four surface electrocardiogram electrodes are attached to the left side of the neck and the lower thorax (approximately at the level of the xiphoid process). An alternating electrical current (AC) of constant amplitude is applied through the pair of outer electrodes to the thorax. The change in TEB is related to the aortic flow pattern and is influenced by the alignment of red blood cells in the aorta. When aortic flow ceases and the aortic valve closes, red blood cells become randomly oriented and interfere with electrical conduction. The ejection of blood forces red blood cells to align in parallel with the flow as the left ventricle contracts and the aortic valve opens, resulting in decreased impedance and higher conductivity.^[11] The rate of change in thoracic electrical impedance is used in calculating hemodynamic measures such as stroke volume (SV), CO, pre-ejection period, LV ejection time, and STR by mathematical algorithms.^[12-15]

Studies have shown that STR can distinguish intact (>50%) from impaired ejection fraction (EF) (≤50%).^[16] Using contrast ventriculography to measure EF, Garrard et al. demonstrated a very high correlation between STR and EF (R = 0.90) in 68 patients with various cardiac diseases.^[17] In a study of 453 consecutive patients without valvular heart disease, Boudoulas et al. found agreement between normal EF and normal STR (defined as 0.42) in 85% of cases.^[18] Studies have also demonstrated the ease of obtaining an accurate LV systolic performance evaluation using STR measurement by Doppler ECHO. STR correlated well with LV systolic performance indices such as LVEF, dP/dt max (maximum rate of LV pressure during isovolumetric contraction), SV, LV output, and global longitudinal strain (GLS). This method was mainly found helpful in the case of poor-quality windows and for detecting LVEF <35%.^[19]

There is no study in the literature regarding the comparison between STR measured from EC and ECHO perioperatively to the best of our knowledge. Therefore, we planned to conduct a comparative study in patients undergoing elective cardiac surgery to study the correlation between STR measured with the EC device ICON (Osypka Medical, Berlin, Germany) and ECHO. The objective is to find a reliable noninvasive technique for estimating LV systolic function during the perioperative period.

MATERIAL AND METHODS

After obtaining Institutional Ethics Committee approval and written informed consent, we included 120 patients of either gender, belonging to the American Society of Anesthesiologists (ASA) grade II–III, age >5 years, undergoing elective cardiac surgery using cardiopulmonary bypass (CPB).

Method of STR monitoring

1. The ICON (Osypka Medical, Berlin, Germany) electrical cardiometry device was connected, and the patient’s demographic and anthropometric data (age, weight, and height) were entered. Four skin electrodes were applied on the neck and thorax per manufacturer recommendations [Figure 1] to calculate STR automatically in the designated time points (before anesthesia induction [T0 baseline], after induction [T1], at the end of the surgery [T2], and after extubation [T3])

2. Patients underwent transthoracic echocardiographic evaluation using transthoracic probe S5-1 connected to an ECHO machine (IE 33, Philips; Bothell, USA). Based on pulsed Doppler aortic acquisitions, STR was obtained as the ratio of the aortic pre-ejection period (PEP: delay from Q wave of QRS to aortic valve opening, ms) to LV ejection time (LVET, ms). All the recordings were taken at the same time points as that of ICON monitor recording.

Close

Figure 1:

Placement sites for electrodes (specific to each letter) as per recommendations from manufacturer.

Export to PPT

RESULTS

One hundred and twenty patients were recruited for the study (67 males and 53 females) undergoing elective cardiac surgery. The mean age was 48 ± 17 years. Demographic details of the patients and the procedures performed are given in [Tables 1 and 2].

STR value obtained using EC at T0, T1, T2, and T3 were significantly different from those obtained from ECHO, P < 0.001 [Table 3].

Bland–Altman analysis showed that EC measured STR compared with ECHO at T0 [Figure 2], with a mean bias of 0.108 and (with limits of agreement −0.19, 0.41) at T1 [Figure 3] is 0.092 (with limits of agreement −0.22, 0.40), at T2 [Figure 4] is 0.11 (with limits of agreement between −0.05 and 0.28), and at T3 [Figure 5] is 0.085 (with limits of agreement between −0.101 and 0.271).

Close

Figure 2:

Bland–Altman analysis of systolic time ratio (STR) by electrical cardiometry (EC) and echocardiography (ECHO) at T0. X-axis: Average of STR from EC and ECHO, Y-Axis: Difference of STR by EC and ECHO. Mean bias of 0.108 and (with limits of agreement −0.19, 0.14).

Export to PPT

Close

Figure 3:

Bland–Altman analysis of systolic time ratio (STR) by electrical cardiometry (EC) and echocardiography (ECHO) at T1. X-axis: Average of STR from EC and ECHO, Y-Axis: Difference of STR by EC and ECHO. Mean bias is 0.092 (with limits of agreement between −0.218 and 0.401).

Export to PPT

Close

Figure 4:

Bland–Altman analysis of systolic time ratio (STR) by electrical cardiometry (EC) and echocardiography (ECHO) at T2. X-axis: Average STR from EC and ECHO, Y-Axis: Difference of STR by EC and ECHO. Mean bias of 0.11 (with limits of agreement between −0.04 and 0.28).

Export to PPT

Close

Figure 5:

Bland–Altman analysis of systolic time ratio (STR) by electrical cardiometry EC and echocardiography (ECHO) at T3. X-axis: Average STR from EC and ECHO, Y-Axis: Difference of STR by EC and ECHO. Mean bias of 0.085 (with limits of agreement between −0.101 and 0.27).

Export to PPT

DISCUSSION

This prospective, randomized, and observational study was conducted to evaluate the correlation between the STR measured from an EC monitor versus ECHO.

LVEF is a main indicator for the evaluation of LV systolic function and can be easily obtained by ECHO. However, this equipment is costly in clinical practice and requires trained personnel, which also limits its application. Several generations of TEB devices using bioimpedance algorithms based on cylinder or cone models of the human thorax have been commercially developed. Various problems with earlier models now have been overcome with the up-gradation of computer technology and refinement of algorithms to calculate CO. The technique now is referred to as “electrical cardiometry.” EC is a relatively affordable and simple technique that can monitor hemodynamics noninvasively in real time.

Reant et al.^[9] studied 134 consecutive heart failure (HF) patients and 43 control subjects to compare STR echocardiographic measurements with LV function parameters, including myocardial strain by speckle tracking. In patients with increasingly altered LVEF or GLS, PEP significantly increased, whereas LVET decreased, resulting in a significantly increased PEP/LVET ratio. Thompson et al.^[16] conducted a retrospective study on 52 patients to assess the relationship between EF by ECHO or gated nuclear ventriculography and STR by impedance cardiography (ICG) in outpatients with chronic HF. There was an inverse correlation between STR and EF (P < 0.001). An EF ≤ 50% and STR ≥ 0.50 demonstrated 93% sensitivity and 85% specificity. STR was able to distinguish intact (>50%) from impaired EF (≤50%). Hemodynamic parameters measured by EC compared with the other gold standard techniques have mixed results.

Malik V,^[20] Rajput R^[21] and Narula J et al.^[10] found a good correlation between the hemodynamic parameters measured by electrical cardiometry and pulmonary artery catheter. Our results agree with the work done by Cox PBW et al.^[22] Elgebaly AS et al. found a strong positive correlation between echocardiography and non invasive cardiometry while measuring cardiac output and stroke volume in patients undergoing lung surgery.^[23] A prospective observational study evaluated the accuracy, precision, and trending ability of cardiac index (CI) measurements using the Aesculon™ bioimpedance electrical cardiometry to the continuous PAC technique before, during, and after cardiac surgery. CI obtained by continuous PAC and Aesculon bioimpedance were not interchangeable in patients undergoing cardiac surgery. The percentage error between the two techniques was above 30% every time.^[24]

Feng SM et al.^[25] conducted a prospective observational study to assess the accuracy and interchangeability of stroke volume and cardiac output measurements during caesarean delivery using electrical velocimetry and transthoracic echocardiography during cesarean delivery in 20 parturient patients and found that electrical velocimetry monitoring has limited accuracy, precision, and interchangeability with transthoracic echocardiography.

The present study showed a clinically unacceptable difference between STR values measured by ICON monitor using the principle of electrical cardiometry and by pulse doppler echocardiography before induction (T0 baseline), after induction (T1), and at the end of the surgery (T2) and after extubation (T3). Whether the device over or underestimated the actual value of STR remains speculative.

Surgery in the area of the diaphragm or affecting the chest can decrease the accuracy of thoracic electrical bioimpedance, as the technique shows a poor correlation after cardiac surgery or major abdominal surgery.^[22] Critchley LA and Peng ZY et al. demonstrated (in dogs and humans) that vasodilation/ vasoconstriction status, lung fluid balance, and acute lung injury can all reduce ICG accuracy when compared to the thermodilution technique.^[26-27] The application of positive airway pressure can also be a confounding factor that changes the thoracic volume and influences impedance measurements.^[28]

The error may be attributed to flaws in the EC device’s algorithm or in performing pulse Doppler ECHO. Considering these unfavorable reports and our findings, the value of STR by EC for evaluating LVEF remains controversial.

CONCLUSION

No association between STR as measured by EC and ECHO was found in our study. We cannot categorically attribute all the errors to EC, and we acknowledge that ECHO may have some inherent errors. However, given the poor agreement between STR measured by EC and pulse Doppler ECHO, this clinically accepted tool, we cannot endorse the use of STR for evaluating LVEF and thus LV systolic function. Our results do not conclusively prohibit using EC in the perioperative period to evaluate LV systolic function. Further work must be done to establish the role of STR as a surrogate marker of LV systolic function.