Quantification of Pulmonary Regurgitation After Surgical Repair of Tetralogy of Fallot Using Vector Flow Mapping

Abstract

Background: Pulmonary valvular regurgitation in postoperative patients with repaired tetralogy of Fallot (rTOF) significantly impairs exercise capacity and causes right heart failure. Quantitative evaluation of the pulmonary valvular regurgitation fraction (PRF) by cardiac magnetic resonance (CMR) is commonly used to determine the indication for surgical or catheter interventions, but less commonly using echocardiography.

Methods and Results: We retrospectively investigated the feasibility and validation of vector flow mapping (VFM) for the quantification of PRF (VFM-PRF) in 34 pediatric patients with rTOF, comparing it to CMR-derived PRF (CMR-PRF) and other qualitative or semiquantitative echocardiographic indices. Each predictive value for CMR-PRF ≥40% was assessed using receiver operating characteristic curves. VFM-PRF and CMR-PRF showed good agreement, with a correlation coefficient of 0.90 and the highest predictive value for CMR-PRF ≥40%, resulting in an area under the curve of 0.93. Other conventional echocardiographic parameters demonstrated poor predictive accuracy.

Conclusions: This is the first report to demonstrate the accurate quantification of PRF by echocardiography using VFM in pediatric patients with rTOF, showing good agreement with CMR results. Particularly in children, VFM may be clinically useful in determining the indication for reintervention for pulmonary valve replacement, offering a possible alternative to CMR, which often requires deep sedation and general anesthesia.

Pulmonary valvular regurgitation (PR) after surgical repair of tetralogy of Fallot (rTOF) through right ventricular outflow tract reconstruction is a significant complication that may lead to right ventricular failure due to volume overload, resulting in ventricular dysfunction, arrhythmias, and sudden death.¹ Importantly, progressive right ventricular dysfunction significantly worsens patient survival.² Thus, in postoperative TOF patients with severe PR, surgical or transcatheter pulmonary valve replacement (PVR) should be performed at the optimal time to preserve right ventricular function before irreversible remodeling occurs.^3–7

To guide the timing of PVR, numerous guidelines and studies have identified cardiovascular magnetic resonance (CMR) measurements as the gold standard for PVR indications.^3,8–12 The recommended cutoff values of CMR parameters include a PR fraction (PRF) ≥25–40%, indexed right ventricular end-diastolic volume (RVEDVI) ≥150–170 mL/m², indexed right ventricular endsystolic volume (RVESVI) ≥80 mL/m², right ventricular ejection fraction (RVEF) <47%, and combinations of these criteria.^3,8–11,13 Furthermore, serial and consecutive monitoring of these parameters is essential to determine the optimal timing for PVR before irreversible remodeling occurs.^6,14 Based on these findings, surgical or catheter-based PVR is often recommended.^15–17

Although considered clinically effective as a golden standard, routine CMR monitoring presents several challenges, particularly in the pediatric population, because CMR examinations are time-consuming, resource-intensive, and often require deep sedation or anesthesia in younger patients. These factors limit its feasibility in routine clinical practice. In contrast, echocardiography is a widely used procedure in both pediatric and adult cardiovascular outpatient clinics, offering advantages such as shorter examination times, no need for deep sedation, and the ability to be easily repeated as often as necessary. However, a significant issue with echocardiography is that all previously reported echocardiographic indicators for PR assessment are qualitative or semiquantitative, and currently there is not an echocardiographic parameter that can accurately assess PR quantitatively.^13,18–21

Recently, vector flow mapping (VFM) was developed as a new echocardiographic imaging modality that can quantitatively measure blood flow based on color Doppler information.²² It quantitatively visualizes blood flow streamlines and vortices by calculating flow dynamics based on color Doppler velocity information. As a result, by setting a measurement line in the area where blood flows, the time-flow curve of the blood flow passing through the set measurement line can be measured quantitatively.^22–25 Using VFM, quantitative evaluation of aortic valvular regurgitation (AR) has been reported.^23,24 Li et al.²³ measured the time-flow integral of blood flow in both the antegrade and retrograde direction at the aortic valve position using VFM, calculating the regurgitant fraction from the obtained ratio. They demonstrated a strong correlation between AR as assessed by VFM and other semiquantitative echocardiographic measures, such as Doppler-derived AR fraction, effective regurgitant orifice area (EROA), and vena contracta width (VC). Similarly, Cai et al. found that the VFM time-flow curve profile of antegrade and retrograde flow at the level of the descending aorta strongly correlated with echocardiographic measures, including VC, AR jet width/annulus diameters, and EROA in patients with chronic AR.²⁴ Based on these findings, To et al.²⁵ explored the use of VFM for quantifying PRF in patients after TOF surgery. They quantified the ratio of systolic antegrade flow to diastolic regurgitation at the right ventricular outflow using a time-flow curve with VFM to derive PRF values. The authors reported a good correlation between VFM-derived and CMR-derived measures of PRF. However, most of their cases were adult patients, and the number of pediatric cases was limited.

Given these gaps in knowledge, the purpose of our study was to review the measurement method for the quantitative evaluation of PRF by VFM in children and adolescents with rTOF, then to compare and validate VFM-derived and CMR-derived PRF, and finally to compare VFM-derived PRF with conventional echocardiographic indices for predicting the severity of PRF ≥40%.

Methods

Study Population

We retrospectively enrolled a cohort of 37 consecutive patients with surgical repair of TOF or double-outlet right ventricle with pulmonary stenosis, all of whom had undergone both CMR imaging and echocardiography, including VFM within 6 months between April 2014 and June 2017. The exclusion criteria included: poor B-mode and color Doppler images; significant residual pulmonary stenosis, indicated by a peak systolic velocity >3.0 m/s, which would produce aliasing affecting VFM measurement;²⁶ pulmonary hypertension with a mean pulmonary artery pressure >25 mmHg, which could affect wall strain and hinder accurate VFM measurements; and poor CMR images, including those from phase-contrast analysis. Written informed consent was given by the patients or their parents. The study protocol was approved by the Institutional Review Board of Nagano Children’s Hospital (IRB s-02-16) and was conducted in accordance with the Declaration of Helsinki and the ethical standards of the responsible committee on human experimentation.

Assessment Using CMR

Using a Philips Multiva 1.5T scanner under ECG synchronization and standard methods,²⁷ we acquired 10–12 slices of 4-chamber, right ventricular longitudinal, and short-axis images (slice width: <10 kg 6 mm and ≥10 kg 8 mm) from the cardiac apex to the base. RVEDVI, RVESVI, and RVEF were calculated using the multiple disc summation method with Philips EWSR 2.6 software. The CMR-PRF was calculated from antegrade and retrograde blood flow crossings at the pulmonary valve position using the phase-contrast method, with CMR-PRF ≥40% defined as severe PR.²⁸

A CMR scan was performed with breath-holding for school-aged children and adolescents and without breath-holding for infants. The CMR scan parameters for the acquisition of ventricular volume using the multiple disc summation method were 360×360 mm field of view and 192×180 mm matrix size, with an RR interval of 20 ms for school-aged children and adolescents. For infants, the parameters were 200×200 mm field of view, 140×140 mm matrix size, and an RR interval of 20 ms. The CMR scan parameters for the phase-contrast method included 12° flip angle, 1 sample averaged, repetition time of 4.5, 360×360 mm field of view, 128×110 mm matrix size, slice thickness of 10 mm, and velocity encoding of 250 cm/s. Based on the CMR-PRF, the patients were divided into 2 groups: those with CMR-PRF ≥40% (S group) and those with CMR-PRF <40% (NS group).

Assessment Using Conventional 2D Echocardiography

We calculated the PR index (PRi), PR pressure half time (PR-PHT), ratio of PR jet width to the main pulmonary artery diameter (PR-JW/PAD), and PR color Doppler area (PR area), as previously reported.^18–21

The PRi was calculated from the continuous-wave Doppler signal as the ratio of the time from the onset of early diastole to the end of the PR wave (PR duration) and the time from the end of the antegrade flow to the onset of the next antegrade flow (total diastolic time). As Li et al. reported, PRi vales <0.77 were considered indicative of severe PR, corresponding to a CMR-PRF >24.5%.¹⁸ PR-PHT was measured as the deceleration slope from the continuous-wave Doppler signal of the PR jet. PR-PHT <100 ms was indicated by CMR-PRF ≥25%.¹⁹ PR-JW/PAD was obtained from the ratio of the PR jet width, measured via color Doppler, to the diameter of the pulmonary artery valve annulus in the parasternal short-axis view. PR-JW/PADs ≥0.5 and ≥0.7 were associated with CMR-PRF values ≥20% and ≥40%, respectively.²⁰ The PR area was qualitatively assessed from the parasternal short-axis image of the aorta and main pulmonary artery together with both pulmonary arterial branches. PR was classified as severe, with a corresponding CMR-PRF ≥40%, if the PR area extended into the peripheral pulmonary arteries.²¹

Assessment Using VFM

Basic Principles VFM data were calculated based on color flow mapping (CFM) and left ventricular wall velocity acquired via tissue tracking. By applying the mass conservation law of blood flow to the CFM velocity, the azimuthal velocities can be estimated under the 2D flow assumption. Details of the VFM derivation can be found in previous reports.^22,29–31 This section briefly describes the derivation using Figure 1A, which shows the imaged area scanned by an ultrasound sector probe, where subscripts r and θ denote the radial and azimuthal components, respectively.

Figure 1.

Schematic diagram of VFM algorithm using sector probe scan (A), time-change waveform of the flow rate (B) and schematic diagram of time-flow curve calculation (C). v represents the velocity, and subscripts, r, θ, and z denote the radial, azimuthal and through-plane components, respectively. Two boundary conditions of wall velocities at the cardiac wall a and b are v_θ,a and v_θ,b. Superscripts cw and ccw denote the calculation pathways in the clockwise and counterclockwise directions, respectively (A). The time-flow curve shows the time-change graph plotted from the sum of the flow rates on the profile line (B). The x-axis is distance, and the y-axis is flow velocity. The flow rate (Qi) when the flow velocity (Vi) flows at a constant time during the frame rate in the interval of Δd (Ex. systolic phase) (C). VFM, vector flow mapping.

The mass conservation law of fluid or the continuity equation is expressed in the cylindrical coordinate system as follows:

  

$$r{\partial }_r{v}_r+v_r+{\partial }_{\theta }{v}_{\theta }+r{\partial }_z{v}_z=0\left\{1\right\}$$

where v represents the velocity and the subscript z denotes the through-plane component. Given that the through-plane velocity is negligible, the last term in Eq. {1} can be neglected under the assumption of 2D flow. With 2 boundary conditions for wall velocities, v_θ,a and v_θ,b, at cardiac walls a and b (Figure 1A), Eq. {1} can be integrated with respect to vθ in 2 ways as follows:

  

$$v_{\theta }^{ccw}(r,\theta )=v_{\theta ,a}(r)+{\displaystyle {\int }_{\theta a}^{\theta }{\partial }_{\theta }{v}_{\theta }d\theta \left\{2\right\}}$$

and

  

$$v_{\theta }^{cw}(r,\theta )=v_{\theta ,b}(r)+{\displaystyle {\int }_{\theta b}^{\theta }{\partial }_{\theta }{v}_{\theta }d\theta \left\{3\right\}}$$

where the color Doppler velocity is used as the radial velocity v_r. Boundary conditions were measured using tissue tracking. Superscripts cw and ccw denote the calculation pathways in the clockwise and counterclockwise directions, respectively. Notably, blood flow velocity matches cardiac wall velocities due to the well-known “no-slip condition”. Using a linear weighted function, Eqs. {2} and {3} can be combined to improve the accuracy as follows:

  

$$v_{\theta }(r,\theta )=(1-W)v_{\theta }^{ccw}(r,\theta )+W{v}_{\theta }^{cw}(r,\theta )$$

  

$$W=\frac{\theta -{\theta }_a}{{\theta }_b-{\theta }_a}.\left\{4\right\}$$

Measuring Blood Flow Volume Using VFM

The diagnostic apparatus utilized was a FUJIFILM Healthcare Prosound F75 with a 5-MHz probe. Color Doppler flow images were obtained and stored in the parasternal longitudinal view of the right ventricular outflow tract to the pulmonary artery, with a frame rate of 35–60 frames/s. The maximal velocity scale was set to prevent aliasing in the area of interest.

The time-flow curve for the VFM captures the temporal changes in the blood flow rate through a line (profile line) positioned arbitrarily on the two-dimensional cross-section (Figure 1B). This profile line was set at the pulmonary valve level in the parasternal longitudinal view of the right ventricular outflow tract to the pulmonary artery. The blood flow rate at the profile line is calculated as the amount passing through a unit distance (∆d) per unit time (1 s). In this context, ∆d refers to the length of 1 segment when the profile line is divided into equal intervals.

The flow rate per section (Q_i) is calculated using Eq. {5} from the flow velocity (V_i) and the unit distance (∆d) in the segment (Figure 1C).

  

$${\text{Q}}_{\text{i}}[{\text{cm}}^2{\text{/s]=V}}_{\text{i}}[\text{cm/s}]\times \Delta \text{d[cm]}\text{{5}}$$

where i denotes the interval number. The total flow (Q_total) through the profile line per unit time was determined as the sum of the flow rates in each segment.

The flow rate (Q_total) was converted into the flow rate per frame using Eq. {6}.

  

$${\text{Q}}_{\text{i}}[{\text{cm}}^2{\text{]=Q}}_{\text{total}}[{\text{cm}}^2\text{/s}]\times \text{Frametime[s]}\text{{6}}$$

Here, frame time is the reciprocal of the frame rate [Hz].

A time-flow curve was created by arranging the blood flow rates obtained in each frame over time. Images were acquired at a frame rate of 35–60 frames/s. We calculated the VFM-PRF reference as the average ratio of the retrograde flow area to the antegrade flow area of the time-flow curve across three or four cardiac cycles at the pulmonary valve, using DAS-RS1 software (FUJIFILM Healthcare Corporation, Tokyo, Japan).

Statistical Analysis

The distribution for each dataset is reported as the median [interquartile range] or the mean±standard deviation. We used SPSS Statistics for Windows, Version 27.0 (IBM Corp., Armonk, NY, USA). Pearson correlation coefficients were employed to examine each correlation, and Bland-Altman analysis was used for assessing VFM errors and interobserver reliability. Fisher’s exact and Mann-Whitney U tests were applied to compare 2 groups. Receiver operating characteristic (ROC) curve analysis was conducted to evaluate the accuracy of echocardiographic measurements in estimating severe PR. Two-tailed P<0.05 was considered statistically significant.

Results

Demographic Data of the Study Patients

After applying the exclusion criteria, 34 patients were included: 33 had TOF, and 1 had a double-outlet right ventricle with pulmonary stenosis. Their ages ranged from 10.9 to 18.7 years (median: 14.4 years), and body weights ranged from 31.8 to 55.3 kg (median: 41.5 kg), with 24 male patients and 10 female patients (Table 1). There were no differences in age, height, body weight, body mass index, postoperative follow-up period, or operative procedures between groups. When comparing the CMR and 2D echocardiographic indices between groups, the S group had a higher CMR-RVEDVI (P=0.023) and CMR-RVESVI (P=0.028) than the NS group; however, there were no significant differences between the two groups for the other CMR or 2D echocardiographic parameters (Table 2).

Table 1.

Anthropometric Participant Characteristics

	NS group (CMR-PRF <40%)	S group (CMR-PRF ≥40%)	P value
No. participants	17	17
Diagnosis	TOF: 16, DORV + PS: 1	TOF: 17
M/F	12/5	12/5	1.000
Age at examination (years)	13.9 [7.6–17.9]	17.0 [12.7–18.7]	0.231
Height (cm)	148.9 [121.4–163.1]	156.8 [147.6–162.4]	0.438
Body weight (kg)	39.3 [21.7–51.0]	46.5 [33.6–60.9]	0.193
Body surface area (cm2)	1.29 [0.86–1.52]	1.43 [1.20–1.62]	0.278
Age at RVOTR (years)	1.3 [1.0–2.2]	1.5 [1.2–1.8]	0.890
Postoperative period (years)	10.8 [5.2–12.7]	15.1 [10.2–17.0]	0.073
Surgical procedure	Transannular patch: 14 Valve spared: 3	Transannular patch: 12 Valve spared: 4 Rastelli: 1

Values are median [interquartile range]. CMR, cardiac magnetic resonance imaging; DORV, double-outlet right ventricle; PRF, pulmonary regurgitant fraction; RVOTR, right ventricular outflow tract reconstruction; TOF, tetralogy of Fallot.

Table 2.

Comparison of Groups by CMR/Echocardiographic Characteristics

	NS group (CMR-PRF <40%)	S group (CMR-PRF ≥40%)	P value
CMR-LVEDVI (mL/m2)	75.5 [68.5–80.5]	78.4 [69.7–89.0]	0.513
CMR-LVESVI (mL/m2)	31.9 [26.6–37.3]	36.5 [32.1–44.6]	0.079
CMR-LVEF (%)	57.4 [56.0–61.2]	53.7 [51.3–56.9]	0.061
CMR-RVEDVI (mL/m2)*	120.3 [107.6–135.5]*	132.1 [115.7–159.6]*	0.023*
CMR-RVESVI (mL/m2)*	62.8 [52.7–74.2]*	71.5 [61.4–87.5]*	0.028*
CMR-RVEF (%)	47.4 [43.2–52.3]	45.1 [43.1–49.6]	0.293
CMR-PRF (%)*	33.2 [23.1–36.9]*	46.9 [45.0–49.9]*	<0.001*
LV diastolic diameter (mm)	38.8 [35.1–41.1]	38.5 [35.4–44.3]
(z-score)	(−0.4 [−1.2 to −0.2])	(−0.7 [−1.2 to 0.1])	1.000
LV fractional area change (%)	46 [42–49]	45 [41–51]	0.782
E wave velocity of transmitral flow (cm/s)	102.3 [92.1–126.4]	100.0 [90.4–128.0]	0.692
A wave velocity of transmitral flow (cm/s)	60.0 [38.0–82.7]	43.6 [39.7–48.8]	0.094
E/E′ in lateral of mitral valve	6.5 [5.8–8.5]	6.7 [5.5–8.1]	0.766
RV fractional area change (%)	41 [35–44]	40 [36–44]	0.889
RV S′ in tissue Doppler (cm/s)	8.2 [7.8–9.3]	9.8 [7.3–11.1]	0.180
RVOTS peak velocity (m/s)	2.0 [1.6–2.2]	2.0 [1.6–2.4]	0.492

Values are presented as median [interquartile range]. *The differences between groups were statistically significant (P<0.05). LV, left ventricle; LVEDVI, left ventricular end-diastolic volume index; LVESVI, left ventricular endsystolic volume index; LVEF, left ventricular ejection fraction; RV, right ventricle; RVEDVI, right ventricular end-diastolic volume index; RVESVI, right ventricular endsystolic volume index; RVEF, right ventricular ejection fraction. Other abbreviations as in Table 1.

Correlation Between CMR-PRF and CMR-RVEDVI/RVESVI/RVEF

The mean value of CMR-PRF was 38.4±12.9% in all patients. CMR-PRF showed a good correlation with CMR-RVEDVI (r=0.61, P<0.001, 95% confidence interval [CI] 0.34–0.79) and CMR-RVESVI (r=0.53, P<0.001, 95% CI: 0.23–0.73), consistent with previous reports³² (Figure 2A,B). However, CMR-PRF had no significant correlation with CMR-RVEF (r=−0.13, P=0.481, 95% CI: −0.44 to 0.22) (Figure 2C).

Figure 2.

Correlation between CMR-PRF and CMR-RVEDVI, CMR-RVESVI, and CMR-RVEF. Pearson correlation coefficients are calculated between CMR-PRF and CMR-RVEDVI (A), CMR-RVESVI (B), and CMR-RVEF (C). CMR, cardiac magnetic resonance imaging; PRF, pulmonary regurgitant fraction; RVEDVI, right ventricular end-diastolic volume index; RVESV, right ventricular endsystolic volume index.

Correlation Between VFM-PRF and CMR-PRF

The mean value of VFM-PRF was 42.3±13.6% in all patients, and VFM-PRF demonstrated a strong agreement and correlation with CMR-PRF across all patients (r=0.90, P<0.001, 95% CI: 0.82–0.95), with a bias of +1.5% and 95% limits of agreement (LOA) of −9.4% and +12.5% (Figure 3A,B). Notably, VFM-PRF and CMR-PRF were in very good agreement across a wide range of CMR-PRF values, from mild to severe, with particularly high agreement for patients with PRF ≥40%.

Figure 3.

Correlation between CMR-PRF and VFM-PRF (A), the Bland-Altmann plots (B) Pearson correlation coefficients are calculated between CMR-PRF and VFM-PRF (A). The Bland-Altmann analysis shows the difference in each value. The arithmetic mean (continuous line) and 95% limits of agreement (equal to +1.96 SD; dotted lines) are determined (B). VFM-PRF correlated with CMR-PRF, with a bias of +1.5% and 95% limits of agreement of −9.4% and +12.5%. CMR, cardiac magnetic resonance imaging; PRF, pulmonary regurgitant fraction; SD, standard deviation; VFM, vector flow mapping.

Comparison of VFM-PRF With Conventional Echocardiographic Parameters

Correlation Between Conventional Echocardiographic Parameters and CMR-PRF PRi and PHT, as semiquantification parameters, showed no correlation with CMR-PRF (r=0.097, P=0.591, 95% CI: −0.43 to 0.26 and r=0.10, P=0.578, 95% CI: −0.25 to 0.43, respectively), in contrast to previous reports. However, PR-JW/PAD (0.67±0.19) did correlate with CMR-PRF (r=0.54, P=0.001, 95% CI: 0.25–0.75) (Figure 4A–C).

Figure 4.

Correlation between previous conventional 2-dimensional echocardiographic parameters and CMR-PRF (A–C). Pearson correlation coefficients are calculated between CMR-PRF and PRi (A), PHT (B), and PR-JW/PAD (C). PRi and PHT, as semiquantification parameters, are not correlated with CMR-PRF. PR-JW/PAD is positively correlated with CMR-PRF. CMR, cardiac magnetic resonance imaging; PHT, pressure half time; PRi, pulmonary regurgitation index; PR, pulmonary regurgitation; PRF, pulmonary regurgitant fraction; PR-JW/PAD, PR jet width to the main pulmonary artery diameter; VFM, vector flow mapping.

PR Color Doppler Area Difference Between the S and NS Groups

According to the severity classification of PR by color Doppler area, 31 of the 34 cases (91.2%) were diagnosed as severe, reaching the pulmonary artery branches. Among these, 14 cases (45.2%) were in the NS group with CMR-PRF <40%. The PR area exhibited 54.8% sensitivity and 100% specificity for CMR-PRF ≥40% in this study. Because this technique is a qualitative measure based on color Doppler imaging, the results were influenced by imaging qualities, including color Doppler gain, velocity scale, and penetration. There was no significant difference in PR area between the S group and the NS group (P=0.227).

Predicting VFM-PRF Values to Estimate CMR-PRF ≥40% in Comparison With Conventional Echocardiographic Parameters

The predictive value for estimating CMR-PRF ≥40% for each echocardiographic index was analyzed using ROC. The parameters compared were PRi, PR area, PR-JW/PAD, and PHT, as well as VFM-PRF. The most accurate predictor was VFM-PRF, with a sensitivity of 76.5%, specificity of 81.2%, and an area under the curve (AUC) of 0.93 (P<0.001, 95% CI: 0.77–0.99). The other conventional qualitative and semiquantitative echocardiographic parameters demonstrated poor predictive value, with a lower AUC (Figure 5).

Figure 5.

ROC curves of VFM-PRF, PRi, PHT, and PR-JW/PAD (detecting CMR-PRF ≥40.0%). The highest prediction and accuracy to detect CMR-PRF ≥40% is achieved by VFM-PRF with a sensitivity of 76.5% and specificity of 81.2%, with an AUC of 0.93 (P<0.001, 95% CI: 0.77–0.99). AUC, area under the curve; CI, confidence interval; CMR, cardiac magnetic resonance imaging; PHT, pressure half time; PRi, pulmonary regurgitation index; PR, pulmonary regurgitation; PRF, pulmonary regurgitant fraction; PR-JW/PAD, PR jet width to the main pulmonary artery diameter; ROC, receiver operating characteristic curve.

Interobserver Reliability

The agreement of VFM-PRF measurement between 2 observers in 31 rTOF cases was quite good, with values of 44.8±14.5% vs. 47.1±13.6%. The bias was +0.1%, with LOA ranging from −5.6% to +5.3%. The interclass correlation coefficient was 0.97 (95% CI: 0.94–0.99) (Supplementary Figure).

Discussion

This report is the first to demonstrate the accurate quantification of PRF by echocardiography using VFM in pediatric patients with rTOF, showing good agreement with measurements acquired using CMR. Our results indicated that PRF evaluation by VFM could be used clinically as an alternative to CMR, which has been the gold standard for determining the indication and timing of pulmonary valve reintervention for pulmonary valve regurgitation in pediatric patients with rTOF.

Echocardiographic indices using pulse Doppler or color Doppler for assessing the severity of PR, such as PRi, PR-PHT, PR-JW/PAD, and PR area, are qualitative or semiquantitative evaluations that do not correlate well with the results of quantitative measures obtained through CMR. For instance, PRi, a pulse Doppler-derived index, was reported to have a significant correlation with CMR-PRF (r=−0.82, P<0.01). In addition, PRi <0.77 as a cutoff value could estimate the PR grade of CMR-PRF >24.5% with 100% sensitivity and 84.6% specificity.¹⁸ However, in our study, we did not observe such a strong correlation or predictive power for CMR-PRF ≥40%. Similarly, PR-PHT, a continuous-wave Doppler index, is also reported to have a negative correlation with CMR-PRF (r=−0.6, P<0.01) and could estimate CMR-PRF >20% with a sensitivity of 76% and specificity of 94% if the PR-PHT is <100 ms.¹⁹ In our study, however, PHT <100 ms yielded only 50% sensitivity and 50% specificity for detecting PRF ≥20%. The discrepancies between our findings and previous reports may be due to differences in the distribution of PR severity or ages within the study groups. PRi and PR-PHT are pulse Doppler-derived indices that assess blood flow at the right ventricular outlet, based on the instantaneous pressure gradient between the right ventricle and the main pulmonary artery during diastole. As these indices shorten, they reflect the increasing stiffness and elevated diastolic pressure of the right ventricle due to PR. However, our study population was younger than those in previous studies. As chronic volume overload of the right ventricle progresses, it leads to a gradual increase in stiffness and diastolic filling pressure;³³ therefore, some of our pediatric subjects may not exhibit the elevated diastolic pressure of the right ventricle, unlike adults. This difference may have made it difficult for these indices (PRi and PHT) to correlate with the severity of PR.

The PR-JW/PAD has been reported to have a good correlation with CMR-PRF (r=0.62, P<0.01).²⁰ Puchalski et al. reported that a PR-JW/PAD >0.5 could predict CMR-PRF >20% with a sensitivity of 94% and specificity of 100%, but it did not effectively differentiate further grading of PR severity.²⁰ Our study also showed a relatively good correlation with CMR-PRF (r=0.54, P=0.001), though it was less effective at predicting CMR-PRF ≥40%. Based on these observations, Zoghbi et al. recommended a multiparametric approach for grading the severity of PR, combining pulsed-wave and continuous-wave Doppler, as well as color Doppler measurements.¹³ Although all parameters that they advocated in their guidelines for grading PR severity are qualitative or semiquantitative, the simplest and most robust parameter was PR-JW/PAD and VC.¹³ Notably, no quantitative measures were included in their recommendations.

In contrast to other methods, VFM is a completely quantitative echocardiographic measurement of blood flow. VFM was originally developed from an analytical algorithm aimed at understanding the fluid hemodynamics of intracardiac blood flow while incorporating myocardial wall mechanics. VFM enables visualization of the velocity vector, streamline, and vortices of blood flow, and calculates their kinetic energy in the cardiac chambers and vessels. VFM has been used to analyze cardiac function and intracardiac dynamic blood flow kinetics in various diseases, including congenital heart disease, valvular disease, ischemic heart disease, and cardiomyopathy.^34–37 The VFM algorithm includes 2D continuity equations and cardiac wall velocity, obtained through tissue tracking, under several constraints such as flow angle, depth, and geometry.²⁹ By accurately measuring vascular flow using the mass conservation equation (assuming plane flow and precise Doppler and wall velocity measurements), VFM can quantify blood flow volume.³⁸ Here, we optimized VFM to obtain blood flow volume data by integrating the stream vectors of blood flow along an arbitrary set line throughout the cardiac cycle. As described in the Methods, the PR ratio was calculated by taking the ratio of the blood flow volume passing in the forward direction to that in the backward direction along the set line at the right ventricular outflow. This basic concept of measuring the PR ratio is fundamentally similar to the CMR measurement of blood flow volume across the setting plane using the 2D phase-contrast method. The key difference between the methods is that CMR uses the magnetic velocity vectors of blood flow passing through a setting plane, whereas VFM uses velocity vectors from color Doppler crossing the set line.

Our study confirmed that VFM-derived PRF measurements in patients with rTOF show a strong correlation and agreement with CMR-derived PRF, demonstrating good reproducibility. The ROC curve of echocardiographic parameters predicting CMR-PRF ≥40% revealed that VFM is the most accurate parameter, with an AUC of 0.93 compared with conventional echocardiographic indices. The sensitivity and specificity of VFM for CMR-PRF ≥40% were 76.5% and 81.2%, respectively. Interestingly, in cases of PRF <20%, there was approximately 10% variation between the VFM and CMR measurements. In contrast, for cases of PRF ≥40%, the 2 values were in good agreement. This finding indicated that VFM measurements are more accurate in severe cases, which is clinically important for determining the timing of reintervention without missing critical moments.

Despite these promising results, further studies are needed to improve the accuracy of VFM measurements by verifying the anatomic and hemodynamic factors that influence results in clinical settings. Notably, echocardiograms of rTOF sometimes involve acoustic shadowing due to postoperative changes in the right ventricular outflow tract.³⁹ Moreover, because VFM is an imaging modality based on color Doppler, special attention should be paid to imaging quality, as well as the physical limitations and artifacts inherent in color Doppler. Furthermore, color Doppler imaging must be optimized by adjusting the echo window and penetration, color Doppler flow velocity range scale, frame rate, and transmission frequency, while also eliminating image interference, such as aliasing, turbulence, calcification, and artifacts. Finally, although VFM is a useful new echo-imaging modality, equipment availability is limited, and this method is not currently implemented in routine practice. To ensure this valuable imaging technology becomes widely accessible, the development of VFM-enabled echo equipment for daily use is essential.

Study Limitations

First, it was a cross-sectional retrospective study with a small number of enrolled patients. Second, the number of patients with mild PR was too small to establish a more precise correlation between VFM-PRF and CMR-PRF, although the correlation observed was consistent. To overcome this limitation, a large-scale prospective comparative study is required. Third, the use of color Doppler signals in VFM introduces potential sources of error. Specifically, turbulent flow due to stenotic lesions or shunt flow at the right ventricular outflow tract to the pulmonary artery could affect the results of the VFM analysis. This is due to the Nyquist limit of color Doppler velocity, which may influence the accuracy of the measurements in the area of interest.

Conclusions

VFM is a new color Doppler-derived echocardiographic imaging technique that enables quantitative measurement of blood flow. In this study, we demonstrated that a quantitative measure of PVR is feasible in pediatric patients with rTOF. Furthermore, the quantitative assessment of PVR by VFM proved to be as accurate and reproducible as CMR measurements, with higher sensitivity and specificity. This makes VFM a valuable tool for determining the optimal timing for pulmonary valve reintervention. Moreover, unlike CMR, VFM has the advantages of shorter examination time and does not require deep sedation or anesthesia. This allows VFM to be easily repeated as needed in daily practice. As a result, VFM could substitute for CMR as a non-invasive imaging modality for the quantification of PRF.

Acknowledgments

None.

Disclosures

This research was funded by the FUJIFILM Healthcare Corporation. The funding company provided support in the form of salaries for authors T.N., T.T., and T.O., but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

IRB Information

Institutional Review Board of Nagano Children’s Hospital (IRB s-02-16).

Data Availability

The deidentified participant data will not be shared.

Supplementary Files

Please find supplementary file(s);

https://doi.org/10.1253/circj.CJ-24-0273

References

• 1.   Gatzoulis MA, Balaji S, Webber SA, Siu SC, Hokanson JS, Poile C, et al. Risk factors for arrhythmia and sudden cardiac death late after repair of tetralogy of Fallot: A multicentre study. Lancet 2000; 356: 975–981.
• 2.   Sugawara Y, Yoshihisa A, Takeishi R, Ohara H, Anzai F, Hotsuki Y, et al. Prognostic effects of changes in right ventricular fractional area change in patients with heart failure. Circ J 2022; 86: 1982–1989.
• 3.   Redington AN, Oldershaw PJ, Shinebourne EA, Rigby ML. A new technique for the assessment of pulmonary regurgitation and its application to the assessment of right ventricular function before and after repair of tetralogy of Fallot. Br Heart J 1988; 60: 57–65.
• 4.   Geva T. Indications for pulmonary valve replacement in repaired tetralogy of Fallot: The quest continues. Circulation 2013; 128: 1855–1857.
• 5.   Tatewaki H, Shiose A. Pulmonary valve replacement after repaired Tetralogy of Fallot. Gen Thorac Cardiovasc Surg 2018; 66: 509–515.
• 6.   Buechel ER, Dave HH, Kellenberger CJ, Dodge-Khatami A, Pretre R, Berger F, et al. Remodelling of the right ventricle after early pulmonary valve replacement in children with repaired tetralogy of Fallot: Assessment by cardiovascular magnetic resonance. Eur Heart J 2005; 26: 2721–2727.
• 7.   Bidviene J, Muraru D, Kovacs A, Lakatos B, Ereminiene E, Liptai C, et al. Global and regional right ventricular mechanics in repaired tetralogy of Fallot with chronic severe pulmonary regurgitation: A three-dimensional echocardiography study. Cardiovasc Ultrasound 2021; 19: 28.
• 8.   Therrien J, Provost Y, Merchant N, Williams W, Colman J, Webb G. Optimal timing for pulmonary valve replacement in adults after tetralogy of Fallot repair. Am J Cardiol 2005; 95: 779–782.
• 9.   Frigiola A, Tsang V, Bull C, Coats L, Khambadkone S, Derrick G, et al. Biventricular response after pulmonary valve replacement for right ventricular outflow tract dysfunction: Is age a predictor of outcome? Circulation 2008; 118(Suppl): S182–S190.
• 10.   Knauth AL, Gauvreau K, Powell AJ, Landzberg MJ, Walsh EP, Lock JE, et al. Ventricular size and function assessed by cardiac MRI predict major adverse clinical outcomes late after tetralogy of Fallot repair. Heart 2008; 94: 211–216.
• 11.   Geva T, Sandweiss BM, Gauvreau K, Lock JE, Powell AJ. Factors associated with impaired clinical status in long-term survivors of tetralogy of Fallot repair evaluated by magnetic resonance imaging. J Am Coll Cardiol 2004; 43: 1068–1074.
• 12.   Rebergen SA, Chin JG, Ottenkamp J, van der Wall EE, de Roos A. Pulmonary regurgitation in the late postoperative follow-up of tetralogy of Fallot: Volumetric quantitation by nuclear magnetic resonance velocity mapping. Circulation 1993; 88: 2257–2266.
• 13.   Zoghbi WA, Adams D, Bonow RO, Enriquez-Sarano M, Foster E, Grayburn PA, et al. Recommendations for noninvasive evaluation of native valvular regurgitation: A report from the American Society of Echocardiography developed in collaboration with the Society for Cardiovascular Magnetic Resonance. J Am Soc Echocardiogr 2017; 30: 303–371.
• 14.   Akazawa Y, Fujioka T, Ide H, Yazaki K, Honjo O, Sun M, et al. Impaired right and left ventricular function and relaxation induced by pulmonary regurgitation are not reversed by tardive antifibrosis treatment. Am J Physiol Heart Circ Physiol 2021; 321: H38–H51.
• 15.   Tretter JT, Friedberg MK, Wald RM, McElhinney DB. Defining and refining indications for transcatheter pulmonary valve replacement in patients with repaired tetralogy of Fallot: Contributions from anatomical and functional imaging. Int J Cardiol 2016; 221: 916–925.
• 16.   He F, Feng Z, Chen Q, Jiao Y, Hua Z, Zhang H, et al. Whether pulmonary valve replacement in asymptomatic patients with moderate or severe regurgitation after tetralogy of Fallot repair is appropriate: A case-control study. J Am Heart Assoc 2019; 8: e010689.
• 17.   Rutz T, Ghandour F, Meierhofer C, Naumann S, Martinoff S, Lange R, et al. Evolution of right ventricular size over time after tetralogy of Fallot repair: A longitudinal cardiac magnetic resonance study. Eur Heart J Cardiovasc Imaging 2017; 18: 364–370.
• 18.   Li W, Davlouros PA, Kilner PJ, Pennell DJ, Gibson D, Henein MY, et al. Doppler-echocardiographic assessment of pulmonary regurgitation in adults with repaired tetralogy of Fallot: Comparison with cardiovascular magnetic resonance imaging. Am Heart J 2004; 147: 165–172.
• 19.   Silversides CK, Veldtman GR, Crossin J, Merchant N, Webb GD, McCrindle BW, et al. Pressure half-time predicts hemodynamically significant pulmonary regurgitation in adult patients with repaired tetralogy of Fallot. J Am Soc Echocardiogr 2003; 16: 1057–1062.
• 20.   Puchalski MD, Askovich B, Sower CT, Williams RV, Minich LL, Tani LY. Pulmonary regurgitation: Determining severity by echocardiography and magnetic resonance imaging. Congenit Heart Dis 2008; 3: 168–175.
• 21.   Renella P, Aboulhosn J, Lohan DG, Jonnala P, Finn JP, Satou GM, et al. Two-dimensional and Doppler echocardiography reliably predict severe pulmonary regurgitation as quantified by cardiac magnetic resonance. J Am Soc Echocardiogr 2010; 23: 880–886.
• 22.   Itatani K, Okada T, Uejima T, Tanaka T, Ono M, Miyaji K, et al. Intraventricular flow velocity vector visualization based on the continuity equation and measurements of vorticity and wall shear stress. Jpn J Appl Phys 2013; 52: 07HF16.
• 23.   Li C, Zhang J, Li X, Zhou C, Li H, Tang H, et al. Quantification of chronic aortic regurgitation by vector flow mapping: A novel echocardiographic method. Eur J Echocardiogr 2010; 11: 119–124.
• 24.   Cai Y, Wei X, Li C, Zhang X, Tang H, Rao L. New echocardiographic method for chronic aortic regurgitation: Diastolic retrograde ratio in the descending aorta by vector flow mapping. Int J Cardiovasc Imaging 2019; 35: 461–468.
• 25.   To AH, Li VW, Ng MY, Cheung YF. Quantification of pulmonary regurgitation by vector flow mapping in congenital heart patients after repair of right ventricular outflow obstruction: A preliminary study. J Am Soc Echocardiogr 2017; 30: 984–991.
• 26.   Kronick MD, Chopra A, Swamy S, Brar V, Jung E, Abraham CZ, et al. Peak systolic velocity and color aliasing are important in the development of duplex ultrasound criteria for external carotid artery stenosis. J Vasc Surg 2020; 72: 951–957.
• 27.   Markl M, Frydrychowicz A, Kozerke S, Hope M, Wieben O. 4D flow MRI. J Magn Reson Imaging 2012; 36: 1015–1036.
• 28.   Mercer-Rosa L, Yang W, Kutty S, Rychik J, Fogel M, Goldmuntz E. Quantifying pulmonary regurgitation and right ventricular function in surgically repaired tetralogy of Fallot: A comparative analysis of echocardiography and magnetic resonance imaging. Circ Cardiovasc Imaging 2012; 5: 637–643.
• 29.   Garcia D, Del Alamo JC, Tanne D, Yotti R, Cortina C, Bertrand E, et al. Two-dimensional intraventricular flow mapping by digital processing conventional color-Doppler echocardiography images. IEEE Trans Med Imaging 2013; 29: 1701–1713.
• 30.   Asami R, Tanaka T, Kawabata KI, Hashiba K, Okada T, Nishiyama T. Accuracy and limitations of vector flow mapping: Left ventricular phantom validation using stereo particle image velocimetory. J Echocardiogr 2017; 15: 57–66.
• 31.   Tanaka T, Asami R, Kawabata KI, Hashiba K, Okada T, Nishiyama T. A posteriori accuracy estimation of ultrasonic vector-flow mapping (VFM). J Vis 2017; 20: 607–623.
• 32.   Gorter TM, van Melle JP, Freling HG, Ebels T, Bartelds B, Pieper PG, et al. Pulmonary regurgitant volume is superior to fraction using background-corrected phase contrast MRI in determining the severity of regurgitation in repaired tetralogy of Fallot. Int J Cardiovasc Imaging 2015; 31: 1169–1177.
• 33.   Ouyang R, Leng S, Chen L, Ma Y, Hu L, Sun A, et al. Assessment of right ventricular diastolic function in pediatric patients with repaired tetralogy of Fallot by cardiovascular magnetic resonance and echocardiography. Eur Radiol 2024; 34: 5487–5500.
• 34.   Avesani M, Degrelle B, Di Salvo G, Thambo JB, Iriart X. Vector flow mapping: A review from theory to practice. Echocardiography 2021; 38: 1405–1413.
• 35.   Lu J, Li W, Zhong Y, Luo A, Xie S, Yin L. Intuitive visualization and quantification of intraventricular convection in acute ischemic left ventricular failure during early diastole using color Doppler-based echocardiographic vector flow mapping. Int J Cardiovasc Imaging 2012; 28: 1035–1047.
• 36.   Uejima T, Koike A, Sawada H, Aizawa T, Ohtsuki S, Tanaka M, et al. A new echocardiographic method for identifying vortex flow in the left ventricle: numerical validation. Ultrasound Med Biol 2010; 36: 772–778.
• 37.   Zhang H, Zhang J, Zhu X, Chen L, Liu L, Duan Y, et al. The left ventricular intracavitary vortex during the isovolumic contraction period as detected by vector flow mapping. Echocardiography 2012; 29: 579–587.
• 38.   Zhang H, Liu L, Chen L, Ma N, Zhou L, Liu Y, et al. The evolution of intraventricular vortex during ejection studied by using vector flow mapping. Echocardiography 2013; 30: 27–36.
• 39.   Shimomiya Y, Nagao M, Kogure T, Asagai S, Inoue A, Yamamoto A, et al. Right ventricular strain with 4-dimensional computed tomography identifies pulmonary hypertension in adults with repaired tetralogy of Fallot. Circ J 2024; 89: 62–68.