Advertisement
The Journal of Heart and Lung Transplantation
International Society for Heart and Lung Transplantation.

Model for screening adult congenital heart disease surgery eligibility with echocardiography parameters

  • Author Footnotes
    # These authors contributed equally to this work.
    Yang Zi-yang
    Footnotes
    # These authors contributed equally to this work.
    Affiliations
    Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou, Guangdong, China

    Southern Medical University, The Second School of Clinical Medicine, Guangzhou, Guangdong, China

    Guangdong Provincial Key Laboratory of South China Structural Heart Disease, Guangzhou, Guangdong, China
    Search for articles by this author
  • Author Footnotes
    # These authors contributed equally to this work.
    Li Hezhi
    Footnotes
    # These authors contributed equally to this work.
    Affiliations
    Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou, Guangdong, China
    Search for articles by this author
  • Xie Nanshan
    Affiliations
    Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou, Guangdong, China

    Southern Medical University, The Second School of Clinical Medicine, Guangzhou, Guangdong, China

    Guangdong Provincial Key Laboratory of South China Structural Heart Disease, Guangzhou, Guangdong, China
    Search for articles by this author
  • Zhou Yin
    Affiliations
    Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou, Guangdong, China

    Guangdong Provincial Key Laboratory of South China Structural Heart Disease, Guangzhou, Guangdong, China
    Search for articles by this author
  • Luo Dongling
    Affiliations
    Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou, Guangdong, China
    Search for articles by this author
  • Fei Hongwen
    Correspondence
    Reprint requests: Zhang Caojin, MD. Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, 106 Zhongshan Er Road, Guangzhou 510100, China
    Affiliations
    Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou, Guangdong, China

    Southern Medical University, The Second School of Clinical Medicine, Guangzhou, Guangdong, China
    Search for articles by this author
  • Zhang Caojin
    Correspondence
    Reprint requests: Zhang Caojin, MD. Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, 106 Zhongshan Er Road, Guangzhou 510100, China
    Affiliations
    Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences, Guangzhou, Guangdong, China

    Southern Medical University, The Second School of Clinical Medicine, Guangzhou, Guangdong, China

    Guangdong Provincial Key Laboratory of South China Structural Heart Disease, Guangzhou, Guangdong, China
    Search for articles by this author
  • Author Footnotes
    # These authors contributed equally to this work.
Open AccessPublished:August 27, 2022DOI:https://doi.org/10.1016/j.healun.2022.08.016

      Objectives

      This study aimed to screen for the eligibility of correction in cases of adult congenital heart disease (CHD). Pulmonary to systemic flow ratios (Qp/Qs) > 1.5 and pulmonary to systemic vascular resistance ratios (Rp/Rs) < 1/3, acquired by right heart catheterization (RHC), are two essential parameters. Nonetheless, performing RHC at every follow-up is impractical and even harmful. Thus, it is important to establish a model to predict Qp/Qs and Rp/Rs status before a RHC confirmation, using echocardiography parameters.

      Methods

      A total of 1,785 patients with adult CHD were enrolled and randomly assigned to the derivation or validation groups. Echocardiogram parameters of the 974 patients in the derivation group were considered candidate predictors for surgery eligibility (Qp/Qs > 1.5 and Rp/Rs < 1/3). Binary logistic regression analyses were performed to identify the independent predictors and establish a scoring system. The scoring system was further examined in the validation group using a receiver operating characteristic (ROC) analysis.

      Results

      Estimated pulmonary artery systolic pressure, velocity through the pulmonary valve, and diameters of the left and right atria were identified as independent predictors. The area under the ROC curve of the predictive value in the validation group and its pre- and post-tricuspid valve malformation subgroups were 0.87 (95% confidence interval [CI]: 0.84−0.90, p < 0.01), 0.86 (95% CI: 0.82−0.91, p < 0.01), and 0.85 (95% CI: 0.79−0.90, p < 0.01), respectively.

      Conclusions

      This scoring system could augment flexibility and convenience for pre-screening CHD patients’ eligibility for surgery, before RHC.

      Graphical Abstract

      KEYWORDS

      Abbreviations:

      CHD (congenital heart disease), CI (confidence interval), ePASP (estimated pulmonary artery systolic pressure), LA (left atria anterior-posterior diameters), OR (odds ratio), PAH (pulmonary artery hypertension), PV (highest blood flow velocity through pulmonary valve), Qp/Qs (pulmonary to systemic flow ratio), RA (Right atrial diameter), RHC (right heart catheterization), Rp/Rs (pulmonary to systemic vascular resistance ratio), ROC (receiver operating characteristic)
      Congenital heart disease (CHD) is a common cause of congenital malformations in newborns, with an incidence of approximately 0.8%.
      • Dolk H
      • Loane M
      • Garne E
      European Surveillance of Congenital Anomalies Working G
      Congenital heart defects in Europe: prevalence and perinatal mortality, 2000 to 2005.
      Due to advancements in cardiac surgery, most patients with CHD can survive to adulthood.
      • Warnes CA
      The adult with congenital heart disease: born to be bad?.
      In America and Europe, the number of adult patients with CHD exceeded that of children in this century.
      • Hoffman JI
      • Kaplan S
      • Liberthson RR
      Prevalence of congenital heart disease.
      ,
      • Moons P
      • Engelfriet P
      • Kaemmerer H
      • et al.
      Delivery of care for adult patients with congenital heart disease in Europe: results from the Euro Heart Survey.
      Nonetheless, approximately 20% of CHD patients will require heart surgery in their adult years, and approximately 70% of these surgeries are first-time operations.
      • Zomer AC
      • Verheugt CL
      • Vaartjes I
      • et al.
      Surgery in adults with congenital heart disease.
      In less developed areas that lack adequate cardiac screening for newborns, the proportion of first-time surgeries in adults might be even larger.
      To screen for eligibility for surgery or percutaneous intervention, pulmonary artery pressure, pulmonary vascular resistance (PVR), pulmonary to systemic flow ratio (Qp/Qs), and pulmonary to systemic vascular resistance ratio (Rp/Rs) are essential parameters. Although the detailed standard for PVR is still debatable, Qp/Qs > 1.5 with Rp/Rs < 1/3, are two critical terms for eligibility, according to the existing guidelines.
      • Stout KK
      • Daniels CJ
      • Aboulhosn JA
      • et al.
      2018 AHA/ACC Guideline for the management of adults with congenital heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.
      ,
      • Baumgartner H
      • De Backer J
      • Babu-Narayan SV
      • et al.
      2020 ESC Guidelines for the management of adult congenital heart disease: the Task Force for the management of adult congenital heart disease of the European Society of Cardiology (ESC). Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC), International Society for Adult Congenital Heart Disease (ISACHD).
      Right heart catheterization (RHC) is the gold standard for acquiring Qp/Qs, Rp/Rs, and other hemodynamic parameters in patients with CHD. Nevertheless, performing RHC at every follow-up can be impractical and even harmful due to X-ray exposure. To avoid unnecessary invasive measurements, a pre-screening for patients’ eligibility for surgery before RHC is needed. Currently, the prediction of surgery eligibility and the decision of RHC confirmation are mainly based on the experiences of CHD specialists. If a scoring system based on echocardiography parameters were available, it would provide greater usability to patients and physicians. There are several methods now available to estimate pulmonary circulation hemodynamics by echocardiogram.
      • Kitabatake A
      • Inoue M
      • Asao M
      • et al.
      Noninvasive evaluation of pulmonary hypertension by a pulsed Doppler technique.
      • Dabestani A
      • Mahan G
      • Gardin JM
      • et al.
      Evaluation of pulmonary artery pressure and resistance by pulsed Doppler echocardiography.
      • Chemla D
      • Castelain V
      • Humbert M
      • et al.
      New formula for predicting mean pulmonary artery pressure using systolic pulmonary artery pressure.
      • Rudski LG
      • Lai WW
      • Afilalo J
      • et al.
      Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography.
      • Aduen JF
      • Castello R
      • Daniels JT
      • et al.
      Accuracy and precision of three echocardiographic methods for estimating mean pulmonary artery pressure.
      • Wang B
      • Feng Y
      • Jia LQ
      • et al.
      Accuracy of Doppler echocardiography in the assessment of pulmonary arterial hypertension in patients with congenital heart disease.
      However, there is no existing model or formula for Qp/Qs and Rp/Rs in patients with CHD. Thus, we intended to establish a model to predict whether CHD patients had Qp/Qs > 1.5 and Rp/Rs < 1/3 using echocardiogram parameters.

      Methods

      Patient selection criteria

      This retrospective single-center study was conducted in the Department of Adult Congenital Heart Disease of the Guangdong Provincial People's Hospital, Guangzhou, Guangdong, China. Patients diagnosed with CHD with shunt(s) who received RHC from January 1, 2010 to December 31, 2020 were initially included. Exclusion criteria were as follows: 1) Single ventricle, single atrium, and/or complete atrioventricular septal defects; 2) anomalous great artery connection, such as transposition of the great arteries, double outlet right ventricle; 3) residual shunt due to previous surgery or intervention; 4) echocardiogram received more than three months before RHC; 5) missing echocardiogram data; and 6) age < 18 years. Two-thirds of the patients were randomly assigned to the derivation group by systematic sampling and the rest were placed in the validation group.

      Echocardiogram

      Complete two-dimensional Doppler echocardiography was performed by unspecified operators at our center. All operators were trained and supervised by the consultant doctors. The echocardiographic equipment was all commercially available, including GE VIVID E9, GE VIVID 7(GE Vingmed, Horten, Norway), Philips IE 33, Philips IE ELITE (Philips Medical Systems, Boston, MA, USA), Siemens SC2000, and Siemens Acuson Sequoia 512 (Siemens Medical Solutions USA, Inc., Mountain View, CA, USA). All Doppler-echocardiographic images were recorded in native DICOM format and centralized, after anonymization, at the Echo Laboratory at Guangdong Cardiovascular Institute, Guangdong Provincial People's Hospital, Guangdong Academy of Medical Sciences. A minimum of three cardiac cycles were recorded for analysis. Using the parasternal short axis view, the pulmonary artery diameters were measured at end-diastole, between the pulmonary valve annulus, and the highest blood flow velocity through the pulmonary valve (PV) was recorded (Figure S1). Using the parasternal long axis view, the left atria anterior-posterior diameters (LA), ascending aortic diameters and left ventricular end-diastolic diameters were measured at end-diastole (Figure S2). The right atria diameters (RA) and right ventricular longitudinal diameters were assessed at end-systole, by measuring the superior-inferior axes from the apical four-chamber view (Figure S3). From the apical five-chamber view, the highest blood flow velocity through the aortic valve was recorded (Figure S4). The estimated pulmonary artery systolic pressure (ePASP) was calculated by adding the right atrial pressure to the transtricuspid pressure gradient. The right atrial pressure was estimated to be 5, 10, or 15 mm Hg, based on the inferior vena cava diameter variation.
      • Brennan JM
      • Blair JE
      • Goonewardena S
      • et al.
      Reappraisal of the use of inferior vena cava for estimating right atrial pressure.
      The transtricuspid pressure gradient was calculated from the velocity of tricuspid regurgitation using the simplified Bernoulli equation.
      • Rudski LG
      • Lai WW
      • Afilalo J
      • et al.
      Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography.
      The left ventricular ejection fraction was also calculated using the Teichholz method.
      • Teichholz LE
      • Cohen MV
      • Sonnenblick EH
      • Gorlin R
      Study of left ventricular geometry and function by B-scan ultrasonography in patients with and without asynergy.
      Transtricuspid pressure gradient = 4*(Tricuspid regurgitation Velocity)2

      Right heart catheterization

      The RHC procedure has been described in detail in our previous study.
      • Zhang C
      • Huang T
      • Huang X
      • et al.
      Inaccuracy of doppler echocardiographic estimates of pulmonary artery pressures in adult atrial septal defect patients with pulmonary arterial hypertension.
      Before RHC, weight and height were routinely measured and recorded. RHC was performed in the catheter laboratory under electrocardiographic monitoring. After local anesthesia, a 6-French MPA 2 catheter (Cordis Inc., Miami, FL, USA) was inserted into the right heart system through the right femoral vein by placement of a 6-French vascular sheath. The catheter was manipulated to the correct position under fluoroscopy. We measured the pressure in the right atrium, pulmonary artery, and wedged pulmonary artery. Before measuring the pressure, the transducers were calibrated to zero at atmospheric pressure. At the same time, blood from the superior vena cava, inferior vena cava, pulmonary artery, and systemic circulation were aspirated for blood gas assays. After oxygen saturation was measured, cardiac output and pulmonary output were calculated using the indirect Fick method. Vascular resistance was derived using the following formula: vascular resistance = pressure gradient/blood flow. Subsequently, Qp/Qs and Rp/Rs were calculated, where patients with a Qp/Qs > 1.5 and Rp/Rs < 1/3, were considered eligible for CHD closure or repair.

      Statistical analyses

      Data was analyzed using SPSS 22 software (IBM Corp., Armonk, NY, USA) and presented as number (percentage) or mean ± standard deviation. Descriptive data analysis was conducted to examine the data distribution. Proportions were compared using chi-square (χ2) tests. Continuous data was compared using the Kolmogorov–Smirnov test. The relationships between the parameters and eligibility were initially evaluated using univariate binary logistic regression. Further, multivariate binary logistic regression was performed on parameters with statistical significance. Independent predictors were further identified through the multivariate analysis and only parameters with statistical significance were selected for the final model. After categorizing the independent predictors based on their quartiles, these categorical parameters were entered into the final multivariable binary logistic regression model. Scores applied to different independent predictors were determined by the relative size of β coefficient from the final model, and the predictive values for each patient were the sum of their scores. Receiver operating characteristic (ROC) curves of the predictive values were used for validation. A stratified scoring system of surgery eligibility were further constructed based on the positive predictive value. All tests were 2-tailed, and statistical significance was set at p < 0.05.

      Missing data

      Since < 5% of the data were missing in the analyzed variables in this study, we used complete case analysis in the main analysis. To confirm the robustness of the data set, we repeated our analysis, where missing data was imputed and replaced with the according average value.

      Ethics

      This study was approved by the Research Ethics Committee of Guangdong Provincial People's Hospital, Guangdong, China, on July 24, 2015 (No. GDREC2015254H[R1]). Data from all participants were de-identified, and informed consent was waived for this study. Patients or the public were not involved in the design, conduct, reporting, or dissemination plans of our research.

      Results

      Patient population and clinical characteristics

      A flowchart of the enrollment process is presented in Figure 1. Based on the study criteria, 1,461 patients were included in the study. Among them, 974 and 487 patients were included in the derivation and validation groups, respectively. The diagnoses and hemodynamic and echocardiographic parameters are shown in Tables 1 and 2, respectively. The diagnosis distribution and clinical characteristics indicated no statistical difference between the derivation and validation groups (p > 0.05).
      Figure 1
      Figure 1Flow chart of the enrollment process. his figure illustrates the patient enrollment process. Each box includes the criteria and the related patient number. CHD: congenital heart disease; RHC: right heart catheterization.
      Table 1Types of CHD in Participants
      TypeDerivationValidation
      Total, n974487
      Pre-tricuspid valve
      ASD, n (%)580 (59.5%)286 (58.7%)
      APVC, n (%)2 (0.2%)4 (0.8%)
      Post-tricuspid valve
      VSD, n (%)197 (20.2%)99 (20.3%)
      PDA, n (%)142 (14.6%)67 (13.8%)
      AVSD, n (%)48 (4.9%)27 (5.5%)
      RASV, n (%)5 (0.5%)4 (0.8%)
      APVC: anomalous pulmonary venous connection; ASD: atrial septal defect; AVSD: atrioventricular septal defect; CHD: congenital heart disease; PDA: patent ductus arteriosus; RASV: ruptured aneurysm of the sinus of Valsalva; VSD: ventricular septal defect.
      Table 2Patient Clinical Characteristics
      DerivationValidation
      Sample Size, n974487
      Eligible, n (%)544 (55.9%)260 (53.4%)
      Sex: Men, n (%)293 (30.1%)158 (32.4%)
      Age (years)39.49±14.0139.21±14.07
      BSA (m2)1.49±0.161.51±0.18
      sPAP (mmHg)78.42±34.3380.75±33.52
      dPAP (mmHg)28.59±17.9230.00±18.17
      mPAP (mmHg)47.13±23.3849.08±23.46
      PAWP (mmHg)10.22±3.8610.21±4.13
      SAP (mmHg)122.28±20.53122.44±21.19
      DAP (mmHg)71.79±12.4272.22±12.51
      MAP (mmHg)89.88±13.2790.39±14.49
      SaO2 (%)93.75±5.1093.75±5.42
      Qp (L/min)9.90±5.559.86±5.28
      Qs (L/min)5.11±1.84.98±1.51
      PVR (WU)5.94±6.66.23±6.45
      SVR (WU)17.95±6.2718.23±6.04
      Qp/Qs2.02±1.092.05±1.06
      Rp/Rs0.34±0.390.36±0.38
      ePASP (mmHg)79.26±28.7782.55±27.99
      LA (mm)37.23±9.1236.76±8.46
      LVEDD (mm)43.30±10.7742.84±10.67
      RA (mm)57.94±13.7257.54±13.15
      RV (mm)62.10±9.9261.87±9.79
      LVEF (%)64.98±8.2164.97±8.68
      PV (m/s)1.31±0.591.27±0.52
      AV (m/s)1.15±0.361.15±0.35
      PA (mm)35.38±9.3735.71±8.72
      AO (mm)25.27±3.8025.27±3.93
      Values of parameters are presented as mean ± standard deviation.
      AO: aortic diameter; AV: highest blood flow velocity through the aortic valve; BSA: body surface area; DAP, diastolic artery pressure; dPAP: diastolic pulmonary artery pressure; ePASP: estimated pulmonary artery systolic pressure; LA: left atrial diameter; LVEDD: left ventricular end diastolic diameter; LVEF: left ventricular ejection fraction; MAP, mean artery pressure; mPAP, mean pulmonary artery pressure; Qp/Qs: pulmonary blood flow to systemic blood flow ratio; PA: pulmonary artery diameter; PAWP: pulmonary arterial wedge pressure; PV: highest blood flow velocity through the pulmonary valve; PVR: pulmonary vascular resistance; Qp: quantity of pulmonary blood flow; Qs: quantity of systemic blood flow; RA: right atrial diameter; Rp/Rs: pulmonary to systemic vascular resistance ratio; RV: right ventricular longitudinal diameter; SaO2: arterial oxygen saturation; SAP: systolic artery pressure; sPAP: systolic pulmonary artery pressure; SVR: systemic vascular resistance.

      Model derivation

      The parameters univariately associated with surgery eligibility are presented in Table 3. In the multivariate regression analysis, the ePASP, LA, RA, and PV remained independent significant predictors associated with surgery eligibility (Table S1). The variance inflation factors were 1.03, 1.37, 1.39, and 1.07, respectively. As they were all within the permissible range, the multicollinearity effect was negated. After categorizing these predictors based on quartiles, a scoring system was developed through the final multivariate analysis (Table 4). Using an arithmetic sum of the scores of ePASP, LA, RA, and PV presented in Table 4, a predictive value was calculated.
      Table 3Univariate Binary Logistic Regression Analysis of Factors for Surgery Eligibility
      UnitOR95% CIp
      Sex (Men)1.090.83–1.440.54
      Age1 years1.051.04–1.06<0.01
      BSA0.1 m210.92–1.080.97
      ePASP
      Patients without tricuspid valve regurgitation or those with minimal regurgitation were not included. AO, ascending aortic diameter; AV: highest blood flow velocity through the aortic valve; BSA: body surface area; ePASP: estimated pulmonary artery systolic pressure; LA: left atrial diameter; LVEDD: left ventricular end diastolic diameter; LVEF: left ventricular ejection fraction; OR: odds ratio; PA: pulmonary artery diameter; PV: highest blood flow velocity through the pulmonary valve; RA: right atrial diameter; RV: right ventricular longitudinal diameter.
      1 mmHg0.970.96–0.98<0.01
      LA1 mm1.091.07–1.11<0.01
      LVEDD1 mm1.000.99–1.010.96
      RA1 mm1.071.06–1.09<0.01
      RV1 mm1.051.03–1.06<0.01
      LVEF1%0.990.98–1.010.27
      PV0.1 m/s1.221.17–1.26<0.01
      AV0.1 m/s1.010.98–1.050.46
      PA1 mm10.99–1.020.68
      AO1 mm1.010.98–1.040.58
      low asterisk Patients without tricuspid valve regurgitation or those with minimal regurgitation were not included.AO, ascending aortic diameter; AV: highest blood flow velocity through the aortic valve; BSA: body surface area; ePASP: estimated pulmonary artery systolic pressure; LA: left atrial diameter; LVEDD: left ventricular end diastolic diameter; LVEF: left ventricular ejection fraction; OR: odds ratio; PA: pulmonary artery diameter; PV: highest blood flow velocity through the pulmonary valve; RA: right atrial diameter; RV: right ventricular longitudinal diameter.
      Table 4Multivariable Binary Logistic Regression Analysis of Independent Predictors for Surgery Eligibility
      ScoreOR95% CIpβ coefficient
      ePASP≤6056.894.73–10.03<0.011.93
      (mmHg)61-8045.293.63–7.72<0.011.67
      81-10011.8301.28–2.63<0.010.61
      >10001 (reference)
      RA≤4801 (reference)
      (mm)49-5621.891.31–2.72<0.010.64
      57-6433.282.2–4.89<0.011.19
      >6433.532.37–5.26<0.011.26
      LA≤3001 (reference)
      (mm)31-3611.71.18–2.46<0.010.53
      37-4222.861.89–4.32<0.011.05
      >4233.352.19–5.14<0.011.21
      PV≤0.901 (reference)
      (m/s)0.91-1.2033.762.57–5.51<0.011.33
      1.21-1.5058.415.53–12.8<0.012.13
      >1.5610.737.1–16.24<0.012.37
      *Score=β coefficient*2.35, rounded to the nearest integer.
      ePASP: estimated pulmonary artery systolic pressure; LA: left atrial diameter; OR: odds ratio; PV: highest blood flow velocity through the pulmonary valve; RA: right atrial diameter.

      Model validation

      The ROC curves of the predictive value for screening eligibility in the validation group are presented in Figure 2 and Table 5. The area under the curve (AUC) was 0.87 (95% confidence interval [CI]: 0.84-0.90, p < 0.01), demonstrating that the prediction model had high accuracy. We also confirmed that, the final model had a similar performance to the models without categorizing parameters or applying integral scores (Table S2). In subgroup analyses, the AUC of the pre-tricuspid valve and post-tricuspid valve malformation both reached 0.85, indicating that the model was compatible with different kinds of CHD, without the complex malformations mentioned in the exclusion criteria.
      Figure 2
      Figure 2Predictive values for screening surgery eligibility in the validation group. (A) ROC curve of the predictive value for screening surgery eligibility in the validation group. (B) ROC curve of the predictive value for screening surgery eligibility in the pre-TV subgroup. (C) ROC curve of the predictive value for screening surgery eligibility in the post-TV subgroup. These figures present the ROC curves of the predictive values in the validation group and its subgroups. Each point on the curve represents different threshold settings and shows the sensitivity and specificity for screening surgery eligibility. ROC: receiver operating characteristic; TV: tricuspid valve.
      Table 5Area Under the ROC Curve of Predictive Values for Screening Surgery Eligibility in Validation Group
      Malformation typeAUC95% CIp
      All0.870.84–0.90<0.01
      Pre-tricuspid valve0.860.82–0.91<0.01
      Post-tricuspid valve0.850.79–0.90<0.01
      AUC: Area under the curve.

      Positive predictive value stratifying

      After the validation, we stratified patients’ possibility of eligibility based on the positive predictive value for better use (Table 6). An example using the scoring system is presented in Figure 3. We further confirmed that, when PVR < 5 wood units was also taken into the eligibility standard, the positive predictive value of the predictive score fitted in the according range.
      Table 6Eligible Possibility Between Different Predictive Value Cut-Points
      Predictive valuePossibility
      Derived from positive predictive value.
      Class
      ≤ 6< 20%Low
      7–820%–50%Medium low
      9–1150%–80%Medium high
      ≥ 12> 80%High
      low asterisk Derived from positive predictive value.
      Figure 3
      Figure 3Example of using this scoring system. This figure provides an example of how to use this scoring system. Each box with black border provides the value of ePASP, RA, LA and PV. The boxes with red border indicate the scores, arithmetic sum and the according eligible possibility. ePASP: estimated pulmonary artery systolic pressure, RA: Right atrial diameter, LA: left atria anterior-posterior diameters, PV: highest blood flow velocity through pulmonary valve.

      Robustness confirmation

      The results from the imputed dataset were similar to the results from the primary analysis, and the scoring system had a similar performance when predictive scores were ≤ 6 or ≥ 12 (Table S3 - 6).

      Discussion

      In this study, we established a model based on echocardiogram parameters to predict Qp/Qs and Rp/Rs status in patients with CHD. Among the independent significant predictors in the final model, ePASP and PV have greater weight in the scoring system. In pulmonary artery hypertension (PAH), significantly elevated pulmonary artery pressure represents a later stage of the disease course, and ePASP is already widely used in PAH assessment.
      • Lau EM
      • Tamura Y
      • McGoon MD
      • Sitbon O
      The 2015 ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension: a practical chronicle of progress.
      As for PV, it is highly related to the real-time pulmonary circulation volume,
      • Stewart WJ
      • Jiang L
      • Mich R
      • Pandian N
      • Guerrero JL
      • Weyman AE
      Variable effects of changes in flow rate through the aortic, pulmonary and mitral valves on valve area and flow velocity: Impact on quantitative doppler flow calculations.
      while the pulmonary blood flow volume is positively related to Qp/Qs and negatively related to Rp/Rs. The other two parameters included in this scoring system are LA and RA. RA would be enlarged due to a left-to-right shunt,
      • Du ZD
      • Cao QL
      • Koenig P
      • Heitschmidt M
      • Hijazi ZM
      Speed of normalization of right ventricular volume overload after transcatheter closure of atrial septal defect in children and adults.
      and LA could represent the return blood volume from the pulmonary veins, which are related to the Qp/Qs and Rp/Rs status. In the validation group, and its two subgroups, the AUC all reached 0.85, indicating high reliability.
      To our knowledge, this is the first model to predict Qp/Qs and Rp/Rs in patients with CHD based on echocardiograms. In patients with PAH but without shunts, several pulmonary hemodynamic parameters can be estimated using ultrasound, including ePASP,
      • Rudski LG
      • Lai WW
      • Afilalo J
      • et al.
      Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography.
      mean pulmonary artery pressure,
      • Kitabatake A
      • Inoue M
      • Asao M
      • et al.
      Noninvasive evaluation of pulmonary hypertension by a pulsed Doppler technique.
      • Dabestani A
      • Mahan G
      • Gardin JM
      • et al.
      Evaluation of pulmonary artery pressure and resistance by pulsed Doppler echocardiography.
      • Chemla D
      • Castelain V
      • Humbert M
      • et al.
      New formula for predicting mean pulmonary artery pressure using systolic pulmonary artery pressure.
      ,
      • Aduen JF
      • Castello R
      • Daniels JT
      • et al.
      Accuracy and precision of three echocardiographic methods for estimating mean pulmonary artery pressure.
      cardiac output,
      • Rudski LG
      • Lai WW
      • Afilalo J
      • et al.
      Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography.
      and PVR.
      • Abbas AE
      • Fortuin FD
      • Schiller NB
      • Appleton CP
      • Moreno CA
      • Lester SJ
      A simple method for noninvasive estimation of pulmonary vascular resistance.
      • Kouzu H
      • Nakatani S
      • Kyotani S
      • Kanzaki H
      • Nakanishi N
      • Kitakaze M
      Noninvasive estimation of pulmonary vascular resistance by Doppler echocardiography in patients with pulmonary arterial hypertension.
      • Opotowsky AR
      • Clair M
      • Afilalo J
      • et al.
      A simple echocardiographic method to estimate pulmonary vascular resistance.
      • Li M
      • Wang Y
      • Li H
      • et al.
      A prediction model of simple echocardiographic variables to screen for potentially correctable shunts in adult patients with pulmonary arterial hypertension associated with atrial septal defects: a cross-sectional study.
      Regardless, estimating pulmonary circulation volume and resistance in patients with CHD-PAH is challenging due to the irregular shape of the right heart. In addition, to achieve high reproducibility and usability, the parameters in this model are widely used in clinical practice, and the predictive value could be easily calculated. Although other echocardiogram data also had a high accuracy in assessing PAH hemodynamics, such as time-velocity integral and pulmonary artery acceleration time, their strength in assessing CHD hemodynamics, especially Qp/Qs, might need further evaluation due to the shunt interference.
      • Abbas AE
      • Fortuin FD
      • Schiller NB
      • Appleton CP
      • Moreno CA
      • Lester SJ
      A simple method for noninvasive estimation of pulmonary vascular resistance.
      ,
      • Levy PT
      • Patel MD
      • Groh G
      • et al.
      Pulmonary artery acceleration time provides a reliable estimate of invasive pulmonary hemodynamics in children.
      Moreover, when patients are transferred to a CHD center, these advanced parameters might not be available, as they were not mandatory measurements. Thus, we established this model using the fundamental elements of the echocardiogram for better generalization.
      In real-world clinical practice, physicians could use this scoring system to have an initial prediction of patients’ eligibility for surgery at their first visit, which could allow the medical team to better prepare treatment plans after RHC confirmation. In addition to the feasibility at the baseline evaluation, this scoring system could also be helpful during the follow-up. Currently, patients who were ineligible for surgery at first RHC were prescribed targeted PAH vasodilators during the follow-up, and considered for CHD correction surgery, after another RHC confirmed that patient hemodynamics reached standards. Nonetheless, the timing for reassessment could be challenged: a premature RHC could lead to a failure of eligible hemodynamics and result in unnecessary incisions and X-ray exposure, while a delayed RHC might increase the possibility of adverse events before CHD correction. With this scoring system, doctors might be able to pre-screen for patients with a high possibility of eligibility for surgery and better decide the RHC timing. Lastly, it should be once again noted that, this scoring system should not be used as a substitute for RHC, as it is the gold standard to confirm CHD surgery eligibility. Moreover, even in patients who were ineligible for surgery due to severe CHD-PAH, RHC is indispensable when accurate hemodynamics data is needed, such as for monitoring or changing prescriptions.
      This study has some limitations. First, more than half of the diagnoses were atrial septal defects, which might have limited the effectiveness of post-tricuspid valve malformation. Other complex CHDs with single atria, single ventricle, complete atrioventricular septal defects, and/or abnormal great artery connection were not included, as the valve velocity was difficult to define. Second, this was a retrospective study and did not undergo strict inspections during echocardiogram examinations, RHC, data collection, and randomization. However, this might be more similar to real-world practices. Under this study design, accuracy was still satisfactory, which indicates that this model might be practical in real clinical settings. Third, the model was not externally verified. As mentioned above, this was a retrospective study, did not have a unified protocol, and the systemic error might have already been limited. Lastly, this model did not include critically ill patients who were unable to receive an echocardiogram or RHC and this might have weakened the predictive accuracy for these patients.

      Author contributions

      This study was mainly undertaken by Dr. Yang Zi-yang and Dr. Li Hezhi and supervised by Dr. Zhang Caojin and Dr. Fei Hongwen. Dr. Zhou Yin and Dr. Xie Nanshan helped with hemodynamic data collection and analysis. Dr. Luo Dongling provided assistance with the statistical analysis.

      Declaration of Competing Interest

      The authors declare no other potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

      Acknowledgments

      We sincerely thank our colleagues in the echocardiogram department for their assistance in echocardiogram data collection. The figures in this study were created with BioRender.com.

      Financial conflict of interest statement

      The authors disclose receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Nature Science Foundation of Guangdong Province (No. 2018A030313764 ), National Key Research and Development Program (No. 2018YFC1002600 ), the Science and Technology Department of Guangdong Province (No. 2014A050503048 , 2017A070701013 , 2017B090904034 , and 2017030314109 ), Guangdong Provincial People's Hospital (No. Y012017008 , No. KJ012019443 ), and the National Nature Science Foundation of China (No. U1401255 ).

      References

        • Dolk H
        • Loane M
        • Garne E
        • European Surveillance of Congenital Anomalies Working G
        Congenital heart defects in Europe: prevalence and perinatal mortality, 2000 to 2005.
        Circulation. 2011; 123: 841-849
        • Warnes CA
        The adult with congenital heart disease: born to be bad?.
        J Am Coll Cardiol. 2005; 46: 1-8
        • Hoffman JI
        • Kaplan S
        • Liberthson RR
        Prevalence of congenital heart disease.
        Am Heart J. 2004; 147: 425-439
        • Moons P
        • Engelfriet P
        • Kaemmerer H
        • et al.
        Delivery of care for adult patients with congenital heart disease in Europe: results from the Euro Heart Survey.
        Eur Heart J. 2006; 27: 1324-1330
        • Zomer AC
        • Verheugt CL
        • Vaartjes I
        • et al.
        Surgery in adults with congenital heart disease.
        Circulation. 2011; 124: 2195-2201
        • Stout KK
        • Daniels CJ
        • Aboulhosn JA
        • et al.
        2018 AHA/ACC Guideline for the management of adults with congenital heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.
        J Am Coll Cardiol. 2019; 73: e81-e192
        • Baumgartner H
        • De Backer J
        • Babu-Narayan SV
        • et al.
        2020 ESC Guidelines for the management of adult congenital heart disease: the Task Force for the management of adult congenital heart disease of the European Society of Cardiology (ESC). Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC), International Society for Adult Congenital Heart Disease (ISACHD).
        Eur Heart J. 2020; 42: 563-645
        • Kitabatake A
        • Inoue M
        • Asao M
        • et al.
        Noninvasive evaluation of pulmonary hypertension by a pulsed Doppler technique.
        Circulation. 1983; 68: 302-309
        • Dabestani A
        • Mahan G
        • Gardin JM
        • et al.
        Evaluation of pulmonary artery pressure and resistance by pulsed Doppler echocardiography.
        Am J Cardiol. 1987; 59: 662-668
        • Chemla D
        • Castelain V
        • Humbert M
        • et al.
        New formula for predicting mean pulmonary artery pressure using systolic pulmonary artery pressure.
        Chest. 2004; 126: 1313-1317
        • Rudski LG
        • Lai WW
        • Afilalo J
        • et al.
        Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography.
        J Am Soc Echocardiogr. 2010; 23 (quiz): 86-88
        • Aduen JF
        • Castello R
        • Daniels JT
        • et al.
        Accuracy and precision of three echocardiographic methods for estimating mean pulmonary artery pressure.
        Chest. 2011; 139: 347-352
        • Wang B
        • Feng Y
        • Jia LQ
        • et al.
        Accuracy of Doppler echocardiography in the assessment of pulmonary arterial hypertension in patients with congenital heart disease.
        Eur Rev Med Pharmacol Sci. 2013; 17: 923-928
        • Brennan JM
        • Blair JE
        • Goonewardena S
        • et al.
        Reappraisal of the use of inferior vena cava for estimating right atrial pressure.
        J Am Soc Echocardiogr. 2007; 20: 857-861
        • Teichholz LE
        • Cohen MV
        • Sonnenblick EH
        • Gorlin R
        Study of left ventricular geometry and function by B-scan ultrasonography in patients with and without asynergy.
        N Engl J Med. 1974; 291: 1220-1226
        • Zhang C
        • Huang T
        • Huang X
        • et al.
        Inaccuracy of doppler echocardiographic estimates of pulmonary artery pressures in adult atrial septal defect patients with pulmonary arterial hypertension.
        Chinese Med J. 2014; 127: 3389-3395
        • Lau EM
        • Tamura Y
        • McGoon MD
        • Sitbon O
        The 2015 ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension: a practical chronicle of progress.
        Eur Respir J. 2015; 46: 879-882
        • Stewart WJ
        • Jiang L
        • Mich R
        • Pandian N
        • Guerrero JL
        • Weyman AE
        Variable effects of changes in flow rate through the aortic, pulmonary and mitral valves on valve area and flow velocity: Impact on quantitative doppler flow calculations.
        J Am CollCardiol. 1985; 6: 653-662
        • Du ZD
        • Cao QL
        • Koenig P
        • Heitschmidt M
        • Hijazi ZM
        Speed of normalization of right ventricular volume overload after transcatheter closure of atrial septal defect in children and adults.
        Am J Cardiol. 2001; 88 (A9): 1450-1453
        • Abbas AE
        • Fortuin FD
        • Schiller NB
        • Appleton CP
        • Moreno CA
        • Lester SJ
        A simple method for noninvasive estimation of pulmonary vascular resistance.
        J Am Coll Cardiol. 2003; 41: 1021-1027
        • Kouzu H
        • Nakatani S
        • Kyotani S
        • Kanzaki H
        • Nakanishi N
        • Kitakaze M
        Noninvasive estimation of pulmonary vascular resistance by Doppler echocardiography in patients with pulmonary arterial hypertension.
        Am J Cardiol. 2009; 103: 872-876
        • Opotowsky AR
        • Clair M
        • Afilalo J
        • et al.
        A simple echocardiographic method to estimate pulmonary vascular resistance.
        Am J Cardiol. 2013; 112: 873-882
        • Li M
        • Wang Y
        • Li H
        • et al.
        A prediction model of simple echocardiographic variables to screen for potentially correctable shunts in adult patients with pulmonary arterial hypertension associated with atrial septal defects: a cross-sectional study.
        Int J Cardiovasc Imaging. 2021; 37: 1551-1562
        • Levy PT
        • Patel MD
        • Groh G
        • et al.
        Pulmonary artery acceleration time provides a reliable estimate of invasive pulmonary hemodynamics in children.
        J Am Soc Echocardiogr. 2016; 29: 1056-1065