ABSTRACT
Aim:
The present study was conducted to validate cardiac output (CO) and cardiac index (CI) obtained from electrical cardiometry (EC) ICON ® with transthoracic echocardiography (TTE) in postoperative pediatric cardiac surgical patients.
Materials and Methods:
A prospective observational study was conducted in 25 pediatric patients with age < 10 years who underwent elective cardiac surgery.
Data Analysis:
BlandAltman plot was constructed for interchangeability and Polar plot was constructed to know trending ability.
Results:
A total of 250 datasets were analyzed. Spearman’s correlation coefficient for CO between ICON ® and TTE showed good positive correlation (r = 0.850, 95% confidence interval 0.81 to 0.881, P <.0001). Moderate positive correlation was observed between ICON ® and TTE for CI (r = 0.60, 95% confidence interval 0.515 to 0.674, P <.0001).
Linear regression equations for CO and CI between ICON ® and TTE were: y = 0.5230 + 0.8078 X (R2 = 0.6597, P <.001) and y = 1.8350 + 0.5869 X (R2 = 0.3985, P <.001) [y- ICON ®; X - TTE], respectively.
BlandAltman plot for CO between ICON ® and TTE showed a bias of 0.3012 with limits of agreement (LOA) being -0.69 to 1.3 and for CI bias was 0.6939 with LOA-2.1 to 3.5.
Polar plot analysis showed an angular bias of 8.1750, with radial LOA being −13.74° to 30.08° for CO and angular bias of 6.6931, with radial LOA being −15.69° to 29.07° for CI.
Conclusion:
ICON ® monitor-derived parameters are not interchangeable with the values derived from TTE. However, the ICON ® monitor demonstrated a good trending ability for both CO and CI.
Keywords: Cardiac index, cardiac output, electrical cardiometry, transthoracic echocardiography
INTRODUCTION
The ability to determine patient’s hemodynamic status is critical in intensive care. Low cardiac output syndrome (LCOS) is a common complication following cardiac surgery, leading to postoperative morbidity and mortality.[1,2] Identification of LCOS is essential to ensure prompt intervention, maximizing oxygen delivery (DO2) and for maintaining tissue perfusion and organ function.[3] The importance of monitoring cardiac performance in pediatric cardiac intensive care unit is due to the fact that low cardiac output (CO) is associated with higher mortality.
Commonly used methods to assess systemic perfusion in critically ill children are indirect parameters like lactates, base excess, mixed venous oxygen saturation, and diuresis.[4] These parameters do not allow continuous monitoring. As surrogate variables, they may be affected by several confounding factors like liver clearance of lactates, metabolic acidosis, intracardiac shunts, and diuresis. Fick’s technique, pulse contour analysis, and thermodilution by pulmonary artery catheter are acceptable methods of CO measurement.[5] However, owing to the invasiveness of catheter placement, feasibility is limited in pediatric population.
Transthoracic echocardiography (TTE) allows identification of LCOS, evaluation of systolic and diastolic function, assessment of volume status, residual lesions, shunts, and effusions. The CO obtained by TTE is noncontinuous and hence clinician has to settle for intermittent point in time of CO measurement.[6] The reliability and reproducibility of echocardiographic measurements may vary based on the patient’s size and echocardiographic windows and operator characteristics.
In recent times, there has been burgeoning interest on noninvasive CO monitors like bioimpedance, bioreactance, applanation tonometry, partial carbon dioxide (CO2) rebreathing, pulse wave transit time, and ultrasonic methods. There have been varied results among these noninvasive CO monitors.
Electrical cardiometry (EC) measures the CO continuously based on thoracic electrical bioimpedance (TEB).
There is limited literature on EC-derived hemodynamic parameters and its validation with TTE in paediatric cardiac surgery. Hence, the present study was conducted to validate CO and cardiac index (CI) by EC (ICON®) with TTE in postoperative pediatric cardiac surgical patients.
MATERIALS AND METHODS
This was a prospective observational study conducted at a tertiary care hospital. After obtaining institutional ethical committee approval, the present study was conducted in the immediate postoperative period with a written informed consent obtained from parents/guardians.
Pediatric patients in the pediatric postoperative cardiac surgical unit with age < 10 years who had undergone elective cardiac surgery were included in the study. Patients not consenting, with open chest, abnormal cardiac rhythm, with residual shunt, and with spontaneous breathing were excluded from the study.
All patients were electively ventilated on Synchronized Intermittent Mandatory Ventilation (SIMV) volume control or pressure control mode. Sedation protocol was standardized for all patients: IV infusion Inj. Dexmedetomidine 0.25-0.5 mcg/kg/hr and IV Fentanyl 1 mcg/kg bolus SOS, duration based on treating physician.
Ten simultaneous CO (l/min) and CI (l/min/m2) values were recorded from EC (ICON® monitor, Markus Osypka, Osypka Medical Inc, Germany) and TTE (HD7 Diagnostic Ultrasound System, Philips Ultrasound, Bothell, Washington, USA) over first 24 hours postoperatively.
Demographic data including age, sex, height, and weight were noted. Patients’ heart rate and blood pressure were also recorded.
TTE was performed with HD7 Diagnostic Ultrasound System (Philips Ultrasound, Bothell, Washington, USA) using S4 and S8 MHz transducers. LVOT VTI (velocity time integral) was recorded in apical 5 chamber view using Pulsed Wave Doppler and LVOT diameter measured just below the aortic valve from parasternal long-axis view. Stroke volume (SV) was calculated using the formula SV = LVOT VTI * [3.14* (LVOT diameter/2) 2]. CO was calculated using the formula CO = Heart rate * SV. CI was calculated using the formula CI = CO/Body surface area. TTE values were taken by the an experienced echocardiographer who was blinded to the study. An average of three values were considered.
EC (ICON® monitor, Markus Osypka, Osypka Medical Inc., Germany) [Figure 1] was used to calculate CO and CI. It is a form of TEB that relates changes in electrical conductivity to aortic blood flow. A biomedical engineer verified and calibrated all equipment before initiation of the study. Four ICON® sensors were applied to the left side of the neck and thorax of the patient (one pair [sensor A and B] located side by side in the vertical direction on the left side of the neck and the other pair [sensor C and D] on the lower thorax along the left mid-axillary line, at the level of the xiphoid process) [Figure 2]. Alternatively, sensor A can be placed on forehead and sensor D at the left thigh. Two sensors introduced low amplitude, high frequency electrical current and the remaining two sensors measured impedance offered by the thorax. Based on the impedance, the monitor determined the CO and CI using a complex algorithm. Due to availability of only a single ICON® monitor, data were captured and recorded periodically over 24 hours.
Figure 1.
ICON® monitor, Markus Osypka, Osypka Medical Inc., Germany
Figure 2.
Placement of Electrodes for ICON®
Statistical analysis
The results were presented as mean ± standard deviation. CO and CI values were analyzed using Spearman’s correlation to determine the strength of relationship between ICON® and TTE. Correlation coefficient values range from being negatively correlated (-1) to uncorrelated (0) to positively correlated (+1) (0.0 is no association, +0.2 is weakly positive, +0.5 is moderately positive, +0.8 is strongly positive, +1.0 is perfectly positive).
Linear regression analysis was used to create a graphic representation of the relationship with the formula of the “best fit” line allowing the CO and CI measurements of ICON® to be calculated from TTE. The coefficient of determination (R2) is the proportion of variation in the dependent variable (ICON®) can be explained by a linear regression model using the independent variable (TTE).
BlandAltman limits of agreement (LOA) plots were constructed for these data. LOA plots visually represent the bias (mean difference) and variability (95% LOA) between two methods of measurement. Ninety five percent LOA were determined by 1.96*standard deviation (SD) of the mean difference of CO and CI values between ICON® and TTE. Polar plot was also been constructed to know trending ability between the two monitors.
A P value <.05 was considered statistically significant. Statistical analysis was performed using MedCalc software, free trial version 20.210 (Ostend, Belgium).
RESULTS
Demographic details of the patients are given in Table 1. A total of 25 patients were enrolled in the study with mean age = 28.82 ± 32 months. A total of 250 datasets were been analyzed. Spearman’s correlation coefficient of CO between ICON® and TTE showed a good positive correlation (r = 0.850, 95% confidence interval [CI] 0.81 to 0.881, P <.0001). A moderate positive correlation was also observed between ICON® and TTE for CI (r = 0.60, 95% CI 0.515 to 0.674, P <.0001).
Table 1.
Demographic characteristics.
| No of cases | |
|---|---|
| Ventricular Septal Defect | 10 |
| Patent Ductus Arteriosus | 3 |
| Total Anomolous Pulmonary Venous Connection | 6 |
| Tetralogy of Fallot | 4 |
| Aortopulmonary Window | 1 |
| Atrioventricular canal defect | 1 |
| Males | 12 |
| Females | 13 |
| Age (months) | 28.82±32 |
Linear regression equations for CO and CI between ICON® and TTE were: y = 0.5230 + 0.8078 X (R2 = 0.6597, P <.001) and y = 1.8350 + 0.5869 X (R2 = 0.3985, P <.001) [y- ICON®; X- TTE], respectively.
BlandAltman plot for CO between ICON® and TTE showed a bias of 0.3012 with LOA being -0.69 to 1.3 [Figure 3]. Whereas for CI bias was 0.6939 with LOA: -2.1 to 3.5 [Figure 4].
Figure 3.
Bland Altman plot to compare CO between ICON® and TTE
Figure 4.
Bland Altman plot to compare CI between ICON® and TTE
Polar plot analysis showed an angular bias of 8.175°, SD = 11.18° with radial LOA being -13.74° to 30.08° for CO [Figure 5a] and angular bias of 6.6931°, SD = 11.42° with radial LOA being -15.69° to 29.07° for CI [Figure 5b].
Figure 5.
(a) Polar plot analysis for cardiac output (b) Polar plot analysis for cardiac index
DISCUSSION
The most commonly used methods to assess systemic perfusion in critically ill children are predominately noncontinuous parameters. Invasive monitors are continuous but with associated complications and are limited in pediatric population.
Numerous investigators have validated TTE as an accurate method of performing hemodynamic measurements.[7-9] Despite being validated an accurate method of hemodynamic monitoring, its reliability and reproducibility may vary based on multiple factors.
EC is a form of TEB that relates changes in electrical conductivity to aortic blood flow. The principle is related to changes in thoracic electrical impedance, which is mainly influenced by the degree of red blood cell (RBC) alignment in the aorta throughout the cardiac cycle. During diastole when aortic flow stops, RBCs are randomly orientated and interfere with electrical conduction. In contrast, during systole when left ventricle contracts, RBCs are forced to align in parallel with aortic flow, and the electrical current in the aorta passes with less impedance. This results in decreased impedance and increased conductivity. Four ICON® sensors are applied to the patient as shown in Figure 2. Two sensors introduced low amplitude, high frequency electrical current, and the remaining two sensors measured impedance offered by the thorax. Based on the impedance change generated, the monitor determines the CO and CI using a complex algorithm. Increased blood volume, flow velocity, and alignment of RBCs during systole reduce impedance.[10] Thoracic impedance is also affected by thoracic fluid content. Thoracic impedance is inversely proportional to the baseline thoracic fluid content.[11]
The present study demonstrates a good positive correlation (r = 0.850) for CO with statistical significance and moderate positive correlation for CI with statistical significant coefficient of determination (r = 0.6) which shows a good trending ability between EC ICON® and TTE (P value <.0001). However, the two monitors are not interchangeable as bias for CO and CI were 0.3012 and 0.6939, respectively, with LOA -0.69 to 1.3 and -2.1 to 3.5, respectively.
Noori S et al.[12] conducted a prospective observational study to validate electrical velocimetry using Aesculon versus echocardiography (echo) by measuring left ventricle output (LVO) in healthy term neonates during the first two postnatal days. LVOev and LVOecho were similar (534 ± 105 vs. 538 ± 105 mL/min, P =0.7). They concluded electrical velocimetry is as accurate in measuring LVO as echocardiography. The present study showed a bias of CO of 300 mL/min.
Boet A et al.[13] did a prospective observational blinded study on 79 stable preterm neonates to compare measurement of SV and CO by EC versus echocardiography. A good correlation was found for SV (r = 0.743; P <.0001) and CO (r = 0.7; P <.0001) measured by EC and echocardiography. Mean biases (and variabilities) were -1.1 (from 0.7 to -2.9) mL and -0.21 (from 0.15 to -0.55) l min −.1 for SV and CO, respectively. The present study demonstrated a good corelation for CO (r = 0.850) similar to Boet A et al. However, the present study has wide LOA and so the two monitors are not interchangeable.
Rauch R et al.[14] conducted a study to evaluate the reliability and accuracy of EC versus TTE using ICON device for the noninvasive determination of CO in 64 obese children and adolescents of age group 8-18 years. They found a significant correlation between the COEC and COEcho measurements (P<0.0001, r = 0.91). The mean difference between the two methods (COEC - COEcho) was 0.015 l min-1. By Bland and Altman method, the upper and lower LOA were + 1.21 and -0.91 l min-1, respectively. They concluded that EC provides accurate and reliable CO measurements in obese children and adolescents. However, in the present study, the two monitors are not interchangeable with wide LOA. Rauch R et al. found good correlation and acceptable LOA probably because adolescent patient population in their study.
Hsu KH et al.[15] compared EC (Aesculon) with echocardiography to monitor CO in 36 preterm infants with a hemodynamically significant PDA (hsPDA). A total of 105 measurements were taken. Mean CO EC and CO echo were 252 ± 32 and 258 ± 45 mL/kg/min, respectively. There was moderate correlation and no significant difference. Bland-Altman analysis showed a bias, LOA, and error percentage of -5.3 mL/kg/min, -78.3 to 67.7 mL/kg/min, and 28.6%, respectively. The present study had wide LOA probably because of 250 datasets estimation versus 105 datasets estimation in the study done by Hsu KH et al.
Song R et al.[16] presented 70-paired measurements in 38 preterm infants with hsPDA. They found a good correlation of CO measured by EC and echo and this finding was independent of the presence of the PDA. The results were similar to the present study.
Critchley LA et al.[17] has demonstrated that polar plot analysis showing an angular bias of < 5o and radial LOA ± 30° have good trending ability of the monitor. In the present study, polar plot analysis showed an angular bias of 8.1750° with radial LOA being -13.74° to 30.08° for CO and angular bias of 6.6931° with radial LOA being -15.69° to 29.07° for CI.
Study done by Yaseen KA et al.[18] concluded that the type of surgical procedure affects the performance of EC CO monitoring. In the present study, a subgroup analysis was not being performed.
Limitations
We used echocardiography to validate CO and CI obtained from EC. Echocardiography is not the gold standard and has its own limitations. The most accurate method is thermodilution. However, it is invasive and associated with complications. Another limitation is a subgroup analysis was not being performed as number of patients in each subgroup were lesser. Another limitation was thoracic impedance is inversely proportional to the baseline thoracic fluid content. This could not be negated.
CONCLUSION
EC has the advantage of being noninvasive, operator-independent, and continuous. ICON® monitor-derived parameters are not interchangeable with the values derived from TTE. However, the ICON® monitor demonstrated a good trending ability for both CO and CI. Hence, we conclude EC can be used for monitoring trends and guide in clinical decision-making. Our future plan is to continue the study for subgroup analysis with a large sample volume.
Financial support and sponsorship
Nil.
Conflicts of interest
Nil.
Acknowledgments
The authors thank Markus Osypka, Osypka Medical Inc., Germany for providing the technical equipment ICON® monitor needed for this study.
REFERENCES
- 1.Maganti MD, Rao V, Borger MA, Ivanov J, David TE. Predictors of low cardiac output syndrome after isolated aortic valve surgery. Circulation. 2005;112(9 Suppl):I448–52. doi: 10.1161/CIRCULATIONAHA.104.526087. [DOI] [PubMed] [Google Scholar]
- 2.Maganti M, Badiwala M, Sheikh A, Scully H, Feindel C, David TE, et al. Predictors of low cardiac output syndrome after isolated mitral valve surgery. J Thorac Cardiovasc Surg. 2010;140:790–6. doi: 10.1016/j.jtcvs.2009.11.022. [DOI] [PubMed] [Google Scholar]
- 3.Lomivorotov VV, Efremov SM, Kirov MY, Fominskiy EV, Karaskov AM. Low-cardiac-output syndrome after cardiac surgery. J Cardiothorac Vasc Anesth. 2017;31:291–308. doi: 10.1053/j.jvca.2016.05.029. [DOI] [PubMed] [Google Scholar]
- 4.Ricci Z, Cecconi G, Lillo R, Di Chiara L, Toscano A, Iacobelli R. Cardiac output measurement with echocardiography and pressure recording analytical method in pediatric patients admitted to the cardiac intensive care unit: A retrospective assessment of bias between the two methods. J Cardiothorac Vasc Anesth. 2021;35:1351–7. doi: 10.1053/j.jvca.2020.11.065. [DOI] [PubMed] [Google Scholar]
- 5.Pugsley J, Lerner AB. Cardiac output monitoring: Is there a gold standard and how do the newer technologies compare? Semin Cardiothorac Vasc Anesth. 2010;14:274–82. doi: 10.1177/1089253210386386. [DOI] [PubMed] [Google Scholar]
- 6.Scolletta S, Romano SM, Biagioli B, Capannini G, Giomarelli P. Pressure recording analytical method (PRAM) for measurement of cardiac output during various haemodynamic states. Br J Anaesth. 2005;95:159–65. doi: 10.1093/bja/aei154. [DOI] [PubMed] [Google Scholar]
- 7.Evangelista A, Garcia-Dorado D, Garcia del Castillo H, Gonzalez-Alujas T, Soler-Soler J. Cardiac index quantification by Doppler ultrasound in patients without left ventricular outflow tract abnormalities. J Am Coll Cardiol. 1995;25:710–6. doi: 10.1016/0735-1097(94)00456-z. [DOI] [PubMed] [Google Scholar]
- 8.Coats AJ. Doppler ultrasonic measurement of cardiac output: Reproducibility and validation. Eur Heart J. 1990;11(Suppl I):49–61. doi: 10.1093/eurheartj/11.suppl_i.49. [DOI] [PubMed] [Google Scholar]
- 9.Huntsman LL, Stewart DK, Barnes SR, Franklin SB, Colocousis JS, Hessel EA. Noninvasive Doppler determination of cardiac output in man. Clinical validation. Circulation. 1983;67:593–602. doi: 10.1161/01.cir.67.3.593. [DOI] [PubMed] [Google Scholar]
- 10.Mathews L, Singh RK. Cardiac output monitoring. Ann Card Anaesth. 2008;11:56–68. doi: 10.4103/0971-9784.38455. [DOI] [PubMed] [Google Scholar]
- 11.Gho K, Woo SH, Lee SM, Park KC, Park GN, Kim J, et al. Predictive and prognostic roles of electrical cardiometry in noninvasive assessments of community-acquired pneumonia patients with dyspnoea. Hong Kong J Emerg Med. 2021;28:205–14. [Google Scholar]
- 12.Noori S, Drabu B, Soleymani S, Seri I. Continuous non-invasive cardiac output measurements in the neonate by electrical velocimetry: A comparison with echocardiography. Arch Dis Child Fetal Neonatal Ed. 2012;97:F340–3. doi: 10.1136/fetalneonatal-2011-301090. [DOI] [PubMed] [Google Scholar]
- 13.Boet A, Jourdain G, Demontoux S, De Luca D. Stroke volume and cardiac output evaluation by electrical cardiometry: Accuracy and reference nomograms in hemodynamically stable preterm neonates. J Perinatol. 2016;36:748–52. doi: 10.1038/jp.2016.65. [DOI] [PubMed] [Google Scholar]
- 14.Rauch R, Welisch E, Lansdell N, Burrill E, Jones J, Robinson T, et al. Non-invasive measurement of cardiac output in obese children and adolescents: Comparison of electrical cardiometry and transthoracic Doppler echocardiography. J Clin Monit Comput. 2013;27:187–93. doi: 10.1007/s10877-012-9412-7. [DOI] [PubMed] [Google Scholar]
- 15.Hsu KH, Wu TW, Wu IH, Lai MY, Hsu SY, Huang HW, et al. Electrical cardiometry to monitor cardiac output in preterm infants with patent ductus arteriosus: A comparison with echocardiography. Neonatology. 2017;112:231–7. doi: 10.1159/000475774. [DOI] [PubMed] [Google Scholar]
- 16.Song R, Rich W, Kim JH, Finer NN, Katheria AC. The use of electrical cardiometry for continuous cardiac output monitoring in preterm neonates: A validation study. Am J Perinatol. 2014;31:1105–10. doi: 10.1055/s-0034-1371707. [DOI] [PubMed] [Google Scholar]
- 17.Critchley LA, Lee A, Ho AM. A critical review of the ability of continuous cardiac output monitors to measure trends in cardiac output. Anaesth Analg. 2010;111:118092. doi: 10.1213/ANE.0b013e3181f08a5b. [DOI] [PubMed] [Google Scholar]
- 18.Yassen KA, Ghadeer ZA, AlHejji F, Alothman F, Alethan H, AlAli D, et al. The Role of electrical cardiometry in paediatrics and neonatal anaesthesia and intensive care: A narrative review. Anesthesia and Critical care. 2022;4:73–83. [Google Scholar]