Noninvasive Monitoring of Cardiac Output: A Useful Tool Yet?

BACKGROUND AND AIM: End-stage heart failure (HF) patients are at high risk for mortality and morbidity. We aimed to study the role of cardiac output (CO) assessed by Doppler as a noninvasive tool, to predict mortality, rehospitalization rate, and left ventricular assist device (LVAD) implantation at 6 months. METHODS AND
RESULTS: We retrospectively analyzed the data of 60 patients with end-stage HF from different etiologies and an LV ejection fraction ≤20%. Patients were divided into two groups, according to the cardiac index (CI) measured by cardiac ultrasound and Doppler - Group 1: low CO or CI <2 l/min/m2 and Group 2: preserved CO or CI ≥2 l/min/m2. Group 1 included 30 patients with mean CI of 1.52 ± l/min/m2. Group 2 included 30 patients with mean CI of 2.27 ± l/min/m2. At 6 months of follow-up, mortality was significantly higher in Group 1 compared to Group 2 (16% vs. 5%, P = 0.0001). Patients in Group 1 had higher hospitalization rate at 6 months compared to patients in Group 2 (3.5 ± 0.9 vs. 1.9 ± 0.6, P = 0.013). There was significantly more LVAD implantation in 16 patients (26%) in Group 1 versus seven patients (12%) in Group 2 (P = 0.04).
CONCLUSION: CI measured noninvasively by cardiac ultrasound is a simple and useful tool in HF patients' risk stratification and in predicting prognosis and outcome. Copyright:

INTRODUCTION

End-stage heart failure (HF) is a critical entity, which has the highest mortality worldwide, and a very high morbidity. Therapeutic options at this stage usually include heart transplantation and left ventricular assist device (LVAD) implantation as a bridge to transplantation or as destination therapy. The REMATCH study showed that treating end-stage HF patients with LVAD significantly reduced the risk of mortality at 1 year.[1] The recent Momentum 3 Trial showed reduction in rehospitalizations and improvement of quality of life with HeartMate III.[2] The timing for LVAD implantation is crucial to enable better outcome and to decrease mortality.[3]

Different prognostic invasive and noninvasive factors were studied in HF. Assessment of cardiac output (CO) by Doppler is a simple tool that seems to be forgotten in the daily cardiac echography practice and that does not appear to show in any of the risk scores for HF patients.

We aimed to study the role of CO assessed by Doppler as a noninvasive tool, to predict mortality, rehospitalization rate, and LVAD implantation at 6 months.

METHODS

We retrospectively reviewed the data of end-stage HF patients. Inclusion criteria were patients with chronic irreversible end-stage HF with LV ejection fraction (EF) ≤20%, patients aged between 16 and 65 years, patients with good acoustic window, and patients with simultaneous invasive assessment of hemodynamic parameters by the insertion of pulmonary artery catheter (Swan Ganz). Exclusion criteria were (1) patients with nonsinus rhythm, (2) patients with aortic regurgitation Grade 2 or more, and (3) patients with obstructive cardiomyopathy with intra-aortic gradient.

Hemodynamic data collected from Swan Ganz study were cardiac index (CI) and systolic pulmonary artery pressure (SPAP) as well as central venous pressure (CVP).

Two-dimensional transthoracic echocardiography (TTE) was performed in a standard manner by three well-trained cardiologists. We collected the data of the cardiac ultrasound performed within 24 h of the Swan Ganz study. Echocardiographic CO was measured using the following formula: LV outflow tract (LVOT) area × LVOT velocity time integral (VTI) × heart rate (HR). The echocardiographic pulmonary artery pressure was assessed through the tricuspid regurgitation flow. Echocardiographic data were compared to invasive data.

Patients were divided into two groups: group 1 included patients with CI <2 l/min/m2, while Group 2 included patients with CI ≥2 l/min/m2. Hospitalization rate, LVAD implantation rate, and mortality at 6 months were analyzed.

Quantitative data were expressed as mean ± standard deviation. Qualitative values were expressed as percentages. Comparison between qualitative variables was analyzed by the test of Fisher, with a significant P < 0.05. Comparison between quantitative variables was performed via the Welch two-sample t-test with a significant P < 0.05.

RESULTS

Sixty patients were included. Thirty patients were included in Group 1 and the other 30 were included in Group 2. The mean age was 45 ± 7.1 years. The etiology of HF was mainly ischemic cardiomyopathy (ICM) (54%), followed by dilated non-ICM (38%). One patient (2%) had fulminant myocarditis, one (2%) had arrhythmogenic right ventricle (RV) cardiomyopathy, and two patients (4%) had postpartum cardiomyopathy. The etiology in each group is shown in Graph 1. Baseline characteristics of the patients are shown in Table 1.

Graph 1

Heart failure etiology: 27% of patients in Group 2 had dilated cardiomyopathy, 67% had ischemic cardiomyopathy, 3% had postpartum cardiomyopathy, and 3% had arrhythmogenic right ventricle cardiomyopathy. 50% of patients in Group 1 had dilated cardiomyopathy, 44% had ischemic cardiomyopathy, 3% had postpartum cardiomyopathy, and 3% had myocarditis. CO = Cardiac output, ICM = Ischemic cardiomyopathy, ARVD = Arrhythmogenic right ventricle dysplasia

Table 1

Baseline characteristics of patients

	Male (%)	Age (years)	Smoking (%)	Diabetes (%)	Dyslipidemia (%)	HTN (%)	BMI	LVEF	NYHA	CRTD (%)	Creatinine
Normal CO	25 (83)	49.4±12	7 (23.3)	7 (23.3)	5 (16.7)	6 (20)	24.4±3.3	18.4±3.4	2.93±0.8	10 (33)	1.3±0.7
Low CO	25 (83)	43.2±9	5 (16.7)	6 (20)	3 (10)	3 (10)	24.9±4.4	16.5±3.6	3.3±0.5	12 (40)	1.5±0.9
P	1	0.09	0.07	0.2	0.1	0.04	0.5	0.1	0.03	0.5	0.2

HTN=Hypertension, BMI=Body mass index, LVEF=Left ventricular ejection fraction, NYHA=New York Heart Association, CO=Cardiac output, CRT-D=Cardiac resynchronization therapy-defibrillator

Thirteen patients died at 6 months (21%). Mortality rate was significantly higher in Group 1 compared to Group 2 (16% vs. 5%, P = 0.0001) [Graph 2].

Graph 2

Mortality rate and left ventricular assist device implantation in the two groups: 13 patients (21%) of all died at 6 months, mortality rate was significantly higher in Group 1, 10 patients (16%) compared to 3 patients in Group 2 (5%) (P = 0.001). Left ventricular assist device implantation rate was significantly higher in 16 patients (26%) in Group 1 versus 7 patients (12%) in Group 2 (P = 0.04). CO = Cardiac output, LVAD = Left ventricular assist device

Twenty-three patients underwent LVAD implantation at 6 months (38%) as a bridge to transplantation. Thirteen were implanted by HeartMate II assist device (Thoratec, Pleasanton, CA, USA), and 10 were implanted by heartware ventricular assist device (HVAD) assisted device (HeartWare, Boston, MA, USA). LVAD implantation rate was significantly higher in 16 patients (26%) in Group 1 versus seven patients (12%) in Group 2 (P = 0.04) [Graph 2].

Mean CO by echography assessment was 1.52 ± 0.4 ml/min/m2 in Group 1 compared to 2.27 ± 0.3 l/min/m2 in Group 2. There was a correlation between CO assessed by echo and by Swan Ganz [Table 2]. SPAP and CVP are also presented in Table 2.

Table 2

Cardiac output and PA pressure are comparable between Swan Ganz and echography

	CO echo l/min/m2	CO l/min/m2 Swan Ganz	P	SPAP mmHg Swan Ganz	CVP mmHg	SPAP echo mmHg
Normal CO	2.27±0.3	2.38±0.4	0.8	46.7±3.3	8±3.2	53±6.1
Low CO	1.52±0.4	1.62±0.3	0.7	45.13±4.1	10±5.1	49.16±4.0
P				0.7	0.06	0.1

PA=Pulmonary artery, CO=Cardiac output, SPAP=Systolic pulmonary artery pressure, CVP=Central venous pressure

Hospitalization rate was significantly higher in Group 1 compared to Group 2 (3.5 ± 0.9 vs. 1.9 ± 0.6, P = 0.013) [Table 3].

Table 3

Hospitalization rate was significantly lower in patients with normal cardiac output

	Normal CO	Low CO	P
Hospitalization	1.9±0.6	3.5±0.9	0.013

CO=Cardiac output

DISCUSSION

The findings of the current study encourage a systematic use of this noninvasive tool. The CO assessment using Doppler is a simple, easy, and noninvasive tool that should be measured in a standard cardiac ultrasound procedure, especially for HF patients. It is noninferior to the invasive assessment and can potentially have a prognostic value in the prediction of mortality and HF hospitalization. Another important issue could be raised: isn’t the cardiac ultrasound able to replace the invasive Swan Ganz study in some critically ill patients who cannot undergo invasive assessment?

Furthermore, since all trials showed better outcome for LVAD patients when implanted earlier, the CO could be considered as an early prognostic factor that might change the timing of the LVAD implantation.

Measurement of LVOT diameter and VTI is a valid and reliable tool to assess CO in critically ill patients.[4]

The cross-sectional area may be measured by echocardiography. Assuming a circular area of the LVOT, CO may be obtained using the equation CO = VTI × A × HR, where VTI is the VTI, A the area, and HR the heart rate. This was validated by comparison with thermodilution and by Fick method.[5]

Earlier findings revealed that cardiac stroke volume was measured simultaneously by Doppler echocardiography and thermodilution. The results suggested that Doppler echocardiography can reliably detect changes in stroke volume. This assessment is valid only in sinus rhythm patients, without significant aortic regurgitation and without obstructive cardiomyopathy.[6]

Prognostic factors in HF are ample and include biomarkers, hemodynamic invasive parameters, and echocardiographic parameters.

There are emerging HF biomarkers. Multimarker panel biomarkers were compared to the gold standard N-terminal pro-brain natriuretic peptide (NT-proBNP), including angiogenesis biomarkers (endostatin, IBP-4, IBP-7, sFlt-1, and PLGF), myocyte stress (GDF-15), extracellular matrix remodeling (galectin-3, mimecan, and TIMP-1), inflammation (galectin-3), and myocyte injury (hs-TnT). All markers were positively correlated with the NT-proBNP (each P < 0.05).[7] A recent study has proved that urinary excretion of C-type natriuretic peptide is increased in acute decompensated HF and predicts worse outcome.[8]

An article by Ouwerkerk et al. assessed 55 papers revealed that the strongest predictors for mortality in HF patients in these models were blood urea nitrogen and sodium.[9]

In another trial, all-cause 90-day mortality post-LVAD was correlated with preoperative CVP and higher age, while the CO as monitored by Swan Ganz was not correlated with mortality.[10]

Furthermore, in another study, a simple echocardiographic risk score of mortality in HF patients was developed. Five baseline TTE variables (end-systolic volume index, left atrial [LA] volume index, mitral E-wave deceleration time, tricuspid annular peak systolic excursion, and pulmonary artery systolic pressure) remained independent predictors of mortality.[11]

A recent study indicated that the ratios of peak power output to LV mass (peak power/mass) and of peak LV mass to power output (peak mass/power) are indices of LV performance. Peak power/mass added prognostic value to a model that included age, New York Heart Association (NYHA) class, etiology, EF, and diastolic dysfunction.[12]

Different RV function parameters were yet assessed as prognostic factors: (i) fractional area, (ii) tricuspid annular plane systolic excursion, (iii) integral of the systolic wave (tissue Doppler imaging [TDI]), and (iv) peak systolic velocity (PSV [TDI]). Only PSV (TDI) at a threshold value of 9.5 cm/s from the RV systolic parameters was found to be independent predictor of outcome: urgent transplantation, urgent ventricular assist device implantation, or an acute HF episode.[13]

LA area is another powerful predictor of outcome among HF patients with predominantly impaired systolic function.[14] LA area was associated with prognosis independently of age, NYHA class, LVEF, and restrictive filling pattern; this underscores the need to include this simple measurement in the echocardiographic evaluation of these patients. Echocardiographic assessment of systolic and diastolic ventricular function is often used to stratify the risk of patients with HF. A restrictive Doppler mitral inflow pattern is more clearly related to the severity of HF in terms of symptoms and exercise tolerance.[14]

The presence of either a relatively small left ventricle (<63 mm) or early systolic equalization of RV and right atrial pressure demonstrated by echocardiography was associated with increased 30-day morbidity and mortality post-LVAD implantation.[15]

Restrictive mitral flow and TDI annular velocities were also univariate predictors of death or HF hospitalization in patients with chronic systolic HF. E-wave deceleration time <140 ms and Em <8 cm/s were independently associated with the combined endpoint of death or HF hospitalization.[16]

Thus, prediction of early adverse outcomes by echocardiographic parameters is additive to laboratory and hemodynamic variables, and among all previously mentioned parameters, CO assessed by ultrasound is the easiest and less complex tool.

CONCLUSION

CI measured noninvasively by cardiac ultrasound and Doppler technique is a simple and underused tool in the daily practice that seems to be helpful in HF patients’ risk stratification. Patients with CI <2 l/min/m2 have significantly higher mortality, higher LVAD implantation rate, and hospitalization rate at 6 months according to our results.

Limitations

Our major limitation was the relatively small number of participants as well as the retrospective analysis. We might have some disparities in the groups, such as the difference in etiology between the two groups. Unfortunately, medical therapy that might influence the outcome could not be collected accurately. Some important echocardiographic parameters such as RV function were not collected. The major cause of death was HF, but the cause was not reported in all cases.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.