Abstract
Objective
To determine the relationship between echocardiography indices of heart function and adverse short-term outcomes in infants with congenital diaphragmatic hernia (CDH).
Study design
This was a retrospective cohort study of infants with CDH from four neonatal surgical centres between 2018 and 2021. Serial targeted neonatal echocardiography was performed to assess myocardial function. The composite primary outcome was death or invasive mechanical ventilation beyond postnatal day 14.
Results
We identified 44 infants, and 18 (40.9%) met the composite primary outcome. They had lower pre-operative left ventricular output (110 vs. 157 mL/kg/min, p = 0.002), right ventricular output (114 vs. 187 mL/kg/min, p = 0.008), and peak systolic myocardial velocities measured at the mitral annulus (4.6 vs. 5.4 cm/s, p = 0.006) and interventricular septum (4.2 vs. 4.8 cm/s, p = 0.001). They exhibited lower post-operative right ventricular output (155 vs. 237 mL/kg/min, p = 0.021) and lower myocardial velocities on tissue Doppler imaging.
Conclusion
Echocardiographic assessments may provide critical prognostic information regarding post-operative respiratory trajectories.
Subject terms: Outcomes research, Paediatrics
Introduction
Congenital diaphragmatic hernia (CDH) is characterized by the herniation of abdominal organs into the thoracic cavity, occurring in approximately 1 in 4000 to 5000 live births [1]. A recent report from the Congenital Diaphragmatic Hernia Study Group demonstrated an improvement in both risk-adjusted mortality rates and observed-to-expected mortality rates for CDH, which are presently estimated to be between 26 and 31% [2]. The pathophysiological complexities of CDH are primarily driven by pulmonary hypoplasia. As a consequence, many infants with CDH require prolonged respiratory support. It has been shown that in Canada, CDH survivors required an average of 12.1 days of mechanical ventilation [3]. The underdeveloped pulmonary vasculature often leads to the development of acute pulmonary hypertension (PH), which is often accompanied by ventricular dysfunction and serve as a pivotal determinant of clinical outcomes [4]. The prompt identification of abnormal hemodynamic parameters is paramount in guiding therapeutic interventions and providing both the care team and parents a realistic understanding of prognosis [5, 6].
The adoption of bedside targeted neonatal echocardiography (TNE) for routine hemodynamic surveillance and in response to clinical deterioration, both prior to and following CDH surgery, represents a significant advancement in CDH care [6, 7]. Lusk et al. reported that severity of PH as defined by traditional echocardiographic parameters at the second week of life strongly predicts short-term morbidity and death [8]. Of studies utilising tissue Doppler imaging (TDI) techniques, averaged right ventricular (RV) early diastolic myocardial velocity on days 1 and 2 of < 4.6 cm/s predicted duration of respiratory support > 21 days, with 100% sensitivity and 88% specificity [9]. Among the studies exploring ventricular function in CDH, the sample size was often small, the outcomes were not analysed according to the site and severity of anatomical defect in CDH (i.e. intra-abdominal herniation), and ventricular dysfunction was not well defined [10, 11]. The objective of this study was to determine the echocardiography predictors of adverse short-term outcomes in infants with CDH.
Methods
Study population
In our cohort, we included infants with CDH, born at gestational age (GA) of 34 weeks or greater in one of the four neonatal surgical centres, namely British Columbia Women’s Hospital from British Columbia, Canada, Alberta Children’s Hospital from Alberta, Canada, Monash Children’s Hospital, from Victoria, Australia, and University of Iowa Hospital from Iowa, United States during 2018–2021. Infants with major congenital anomalies apart from CDH or chromosomal abnormalities were excluded. Ethics approval was obtained from all participating sites: University of British Columbia (H17-01620; August 1, 2017), University of Calgary (REB17-1694; September 26, 2017), University of Iowa (M18-00338; March 26, 2019), and Monash University (RES-17-0000014L; January 27, 2017), with data transfer agreements obtained at all participating centres, as needed. Post-acquisition analysis was performed separately at each centre and anonymous data without patient identifiers were transferred to British Columbia Women’s Hospital for data analyses. Written informed consent for the study was obtained from the legal guardian of the neonates, as appropriate.
Imaging protocol
As per standard of care, serial TNE was performed by TNE-trained physicians pre-operatively, within 24–48 h post-operation, and again at approximately two weeks postnatal age, per standard protocols. Utilising either a 6- or 12-MHz high-frequency phased-array transducer, standard two-dimensional, M-mode, colour Doppler, pulse-wave Doppler, and continuous-wave Doppler images were obtained as part of a comprehensive evaluation. The following parameters were measured:
-
(i)
Left ventricular (LV) systolic function: LV fractional shortening; ejection fraction; left ventricular output (LVO);
-
(ii)
LV diastolic function: isovolumic relaxation time, the interval between aortic valve closure and mitral valve opening, and the E/A ratio [11];
-
(iii)
RV systolic function: tricuspid annular plane systolic excursion (TAPSE) to characterize longitudinal apex-to-base shortening; right ventricular output (RVO);
-
(iv)
PH assessment: RV ejection time (RVET) and pulmonary artery acceleration time (PAAT) to calculate the RVET/PAAT ratio; interventricular septal curvature at the end-systole at the papillary muscle level; presence of pulmonary insufficiency or tricuspid regurgitation, to estimate mean pulmonary artery pressure and RV systolic pressure; presence or absence of patent ductus arteriosus and its flow directionality; patent foramen ovale/ atrial septal defect, and their flow directionality;
-
(v)
Pulsed TDI: at tricuspid annulus, interventricular septum (IVS) and mitral annulus, measuring peak systolic (S′), early diastolic (E′), and late diastolic (A′) velocities.
Outcomes
The primary composite outcome was death or invasive ventilation beyond postnatal day 14 [3]. Secondary outcomes included death before discharge, need for extracorporeal membrane oxygenation (ECMO), pre-operative inotropic support, and duration of invasive mechanical ventilation.
Statistical analyses
Descriptive statistics were used to summarise the clinical and echocardiographic parameters of the infants. Results were reported as mean ± standard deviation for parametric data and median (interquartile range) for non-parametric data. Normally distributed data were analysed using the Student’s t-test, whereas non-normally distributed data were analysed using the Mann-Whitney U test. Categorical variables were evaluated using the Chi-square test or Fisher’s exact test, as appropriate. Serial changes in echocardiography parameters over time were assessed using either Repeated Measures ANOVA (post hoc Bonferroni correction) or the Friedman test, as appropriate. Comparisons of clinical and echocardiography parameters between infants with and without the primary outcome were conducted using the Mann-Whitney U test, Chi-square test, or Fisher’s exact test, as appropriate. Receiver operating characteristic (ROC) curves were constructed to evaluate the predictive value of echocardiography parameters distinguishing infants with and without the primary outcome. Youden’s Index was used to find the optimal threshold that maximizes both sensitivity and specificity. All statistical analyses were performed using R software (version 4.2.1). A p-value < 0.05 was considered statistically significant.
Results
Forty-four infants were included in this cohort, of whom 43 (97.7%), 38 (86.4%) and 39 (88.6%) patients had TNE performed at the median postnatal age of 2 [2–2], 6 [4–7], and 14 [13–16] days, respectively. Infants were born at median gestational age and birth weight of 38 weeks [38–39] and 3042 grams [2888–3360], respectively. We found 63.6% (28/44) patients had bowel and liver herniated into the thoracic cavity (“liver-up” category), of whom 23 (82.1%) had left-sided defect.
In total, 18 (40.9%) infants developed the composite primary outcome of mortality or invasive mechanical ventilation beyond 14 postnatal days. These infants were more likely to have herniation of both gut and liver into the thoracic cavity (“liver-up” group; 15 (83.3%) vs. 13 (50.0%), p = 0.030) and pre-operative vasopressor/ inotrope use (15 (83.3%) vs. 7 (26.9%), p < 0.001) (Table 1). The median age of death among non-survivors (n = 5) was 22 days [21.5–140].
Table 1.
Baseline demographics of all infants (n = 44) with and without composite primary outcome.
| Infants with composite primary outcome | Infants without composite primary outcome | p-value | |
|---|---|---|---|
| Number (%) | 18 (40.9) | 26 (59.1) | |
| GA (weeks), median [IQR] | 38 [37–39] | 38.5 [38–39] | 0.152 |
| Birthweight (g), median [IQR] | 3017 [2817–3365] | 3074 [2901–3348] | 0.559 |
| Liver-up, n (%) | 15 (83.3) | 13 (50.0) | 0.030 |
| Left sided defect, n (%) | 14 (77.8) | 25 (96.) | 0.142 |
| Pre-operative vasopressor/inotrope use, n (%) | 15 (83.3) | 7 (26.9) | <0.001 |
| Age of repair, days, median [IQR] | 5 [4–6] | 4 [3–5] | 0. 173 |
| ECMO, n (%) | 5 (27.8) | – | – |
| Deceased, n (%) | 5 (27.8) | – | – |
| Age of death, days, median [IQR] | 22 [21.5–140] | – | – |
Composite primary outcome was death or invasive ventilation beyond postnatal day 14.
Numeric comparison: Mann-Whitney U test (non-parametric); Categorical comparison: Chi-square (parametric) or Fisher’s (non-parametric).
Serial TNE evaluation showed an increase in LA:Ao ratio, RVO, TAPSE, and myocardial velocities on TDI (MV S′, MV E’, IVS S′, IVS E′, and IVS A′), and decrease in the proportion of bidirectional and right-to-left atrial shunting after surgical intervention (Supplementary Table 1).
Infants with the composite primary outcomes had lower pre-operative LVO (110 vs.157 mL/kg/min, p = 0.002), RVO (114 vs.187 mL/kg/min, p = 0.008), as well as lower myocardial velocities on TDI (MV S′, MV E′, MV A′, and IVS S′) (Table 2). On the first post-operative TNE, they exhibited lower RVO (155 vs. 237 mL/kg/min, p = 0.021), lower TAPSE (7.3 vs. 9.4 mm, p = 0.005), lower pulmonary vein D-wave (55.7 vs. 77.2 cm/s, p = 0.014), higher RVET:PAAT ratio (4.0 vs. 3.3, p = 0.007), and lower myocardial velocities (MV E′, MV A′, IVS S′, IVS E′, and TV E’) (Table 3).
Table 2.
Echocardiographic parameters among infants with and without composite primary outcome–pre-operative echo (n = 43).
| Infants with composite adverse outcome (N = 17)a | Numbers entered: nb | Infants without composite primary outcome (N = 26)a | Numbers entered: nb | p-value | |
|---|---|---|---|---|---|
| DoL, days | 2 [1–2] | – | 2 [2–2] | – | 0.285 |
| Left heart | |||||
| PV S-wave, cm/s | 47.7 [35.6–61.9] | 8 (47.1) | 54.2 [47.1–58.6] | 7 (26.9) | 0.693 |
| PV D-wave, cm/s | 59.4 [42.3–63.8] | 7 (42.1) | 48.1 [46.1–63.2] | 7 (26.9) | 1.000 |
| LV FS, % | 34.5 [25.2–48.3] | 14 (82.4) | 34.5 [32.0–42.2] | 26 (100.0) | 0.691 |
| LV EF, % | 69.0 [65.5–73.5] | 11 (64.7) | 74.6 [66.4–76.6] | 9 (34.6) | 0.424 |
| LA:Ao ratio | 1.05 [0.99–1.21] | 8 (47.1) | 1.25 [1.06–1.30] | 6 (23.1) | 0.194 |
| LVO, mL/kg/min | 110 [74.2–118] | 12 (70.6) | 157 [134–188] | 9 (34.6) | 0.002 |
| LV-E/A ratio | 1.00 [0.90–1.14] | 11 (64.7) | 0.81 [0.80–0.92] | 21 (80.8) | 0.065 |
| LV-IVRT, ms | 54.0 [44.3–59.5] | 11 (64.7) | 48.0 [42.0–53.0] | 21 (80.8) | 0.662 |
| Shunts | |||||
| PDA size, mm | 3.3 [2.9–4.4] | 14 (82.4) | 3.0 [2.8–3.2] | 21 (80.8) | 0.262 |
| Bidirectional or right-to-left shunt, n/N (%) | 13/14 (92.9) | – | 17/21 (81.0) | – | 0.332 |
| ASD/PFO, n/N (%) | 11/13 (84.6) | 13 (76.5) | 21/21 (100) | 21 (80.8) | 0.139 |
| Left to right, n (%) | 2/11 (18.2) | 6/21 (28.6) | |||
| Bidirectional, n (%) | 8/11 (72.7) | – | 15/21 (71.4) | – | 0.477 |
| Right to left, n (%) | 1/11 (9.1) | 0/21 (0) | |||
| Right heart | |||||
| TAPSE, mm | 6.97 [6.16–8.29] | 16 (94.1) | 8.35 [7.48–8.95] | 24 (92.3) | 0.075 |
| RVET:PAAT ratio | 4.3 [3.5–4.5] | 12 (70.6) | 3.6 [3.1–4.0] | 23 (88.5) | 0.085 |
| RVO, mL/kg/min | 114 [91.5–118] | 11 (64.7) | 187 [164–222] | 7 (26.9) | 0.008 |
| Tissue Doppler Imaging | |||||
| MV S’, cm/s | 4.64 [4.27–4.88] | 10 (58.8) | 5.35 [5.07–5.60] | 20 (76.9) | 0.006 |
| MV E’, cm/s | 5.21 [3.43–6.00] | 13 (76.5) | 6.80 [5.76–7.00] | 22 (84.6) | 0.009 |
| MV A’,cm/s | 5.64 [4.23–6.60] | 14 (82.4) | 7.70 [6.58–7.90] | 23 (88.5) | 0.004 |
| IVS S’, cm/s | 4.15 [3.62–4.51] | 14 (82.4) | 4.80 [4.60–4.90] | 23 (88.5) | 0.001 |
| IVS E’, cm/s | 4.62 [4.20–5.10] | 9 (52.9) | 5.60 [5.00–5.80] | 21 (80.8) | 0.080 |
| IVS A’,cm/s | 5.30 [5.00–6.00] | 9 (52.9) | 6.40 [5.90–6.80] | 21 (80.8) | 0.093 |
| TV S’, cm/s | 5.74 [5.30–6.05] | 10 (58.8) | 6.50 [6.30–7.00] | 21 (80.8) | 0.079 |
| TV E’, cm/s | 6.75 [6.02–8.48] | 10 (58.8) | 7.90 [7.00–8.20] | 21 (80.8) | 0.554 |
| TV A’, cm/s | 7.50 [6.78–8.84] | 10 (58.8) | 8.80 [8.50–9.10] | 21 (80.8) | 0.212 |
Composite primary outcome was death or invasive ventilation beyond postnatal day 14.
aN: total number of patients per group.
bn: represent the number of measurements obtained.
Table 3.
Echocardiographic parameters among infants with and without composite primary outcome–first post-operative echo (n = 38).
| Infants with composite adverse outcome (N = 12)a | Numbers entered: nb | Infants without composite primary outcome (N = 26)a | Numbers entered: nb | p-value | |
|---|---|---|---|---|---|
| DoL, days | 7 [5–7] | 10 (83.3) | 6 [4–7] | 11 (42.3) | 0.499 |
| Left heart | |||||
| PV S-wave, cm/s | 56.6 [44.7–61.1] | 6 (50.0) | 58.8 [57.1–72.8] | 7 (26.9) | 0.234 |
| PV D-wave, cm/s | 55.7 [42.4–65.3] | 6 (50.0) | 77.2 [67.4–81.8] | 7 (26.9) | 0.014 |
| LV FS, % | 41.0 [39.2–45.0] | 10 (83.3) | 36.0 [34.0–40.0] | 25 (96.2) | 0.149 |
| LV EF, % | 74.5 [72.4–77.5] | 8 (66.7) | 75.0 [72.0–80.2] | 6 (23.1) | 0.795 |
| LA:Ao ratio | 1.29 [1.22–1.37] | 6 (50.0) | 1.60 [1.40–1.75] | 3 (11.5) | 0.300 |
| LVO, mL/kg/min | 136 [122–156] | 8 (66.7) | 138 [133–168] | 8 (30.8) | 0.279 |
| LV-E/A ratio | 0.94 [0.85–1.07] | 6 (50.0) | 0.87 [0.81–1.01] | 22 (84.6) | 0.433 |
| LV-IVRT, ms | 46.9 [40.7–53.2] | 6 (50.0) | 45.0 [42.0–50.0] | 21 (80.8) | 0.704 |
| Shunts | |||||
| PDA size, mm | 4.4 [3.0–5.2] | 9 (75.0) | 3.0 [2.8–3.1] | 21 (80.8) | 0.074 |
| Bidirectional or right-to-left shunt, n/N (%) | 9/9 (100) | – | 14/22 (63.6) | – | 0.068 |
| ASD/PFO, n/N (%) | 7/8 (87.5) | 8 (66.7) | 26/26 (100) | 26 (100) | 0.235 |
| Left to right, n (%) | 5/7 (71.4) | 17/26 (65.4) | |||
| Bidirectional, n (%) | 0/7 (0) | – | 1/26 (3.8) | – | 1.000 |
| Right to left, n (%) | 2/7 (28.6) | 8/26 (30.8) | |||
| Right heart | |||||
| TAPSE, mm | 7.3 [7.0–8.5] | 10 (83.3) | 9.4 [8.3–10.0] | 25 (96.2) | 0.005 |
| RVET:PAAT ratio | 4.0 [3.6–4.0] | 9 (75.0) | 3.3 [2.9–3.7] | 24 (92.3) | 0.007 |
| RVO, mL/kg/min | 155 [125–187] | 8 (66.7) | 237 [206–251] | 9 (34.6) | 0.021 |
| Tissue Doppler Imaging | |||||
| MV S’, cm/s | 4.80 [4.80–6.01] | 5 (41.7) | 5.70 [5.50–5.90] | 19 (73.1) | 0.431 |
| MV E’, cm/s | 5.04 [4.24–5.95] | 10 (83.3) | 7.00 [6.40–7.20] | 22 (84.6) | <0.001 |
| MV A’,cm/s | 6.00 [4.58–6.30] | 10 (83.3) | 8.00 [7.15–8.20] | 23 (88.5) | 0.001 |
| IVS S’, cm/s | 4.51 [4.20–5.00] | 11 (91.7) | 5.30 [5.10–5.60] | 21 (80.8) | 0.038 |
| IVS E’, cm/s | 4.96 [4.23–5.07] | 6 (50.0) | 6.45 [6.02–6.60] | 18 (69.2) | 0.030 |
| IVS A’,cm/s | 6.95 [5.62–7.98] | 6 (50.0) | 7.30 [7.00–7.48] | 18 (69.2) | 0.764 |
| TV S’, cm/s | 6.10 [5.70–6.46] | 6 (50.0) | 6.90 [6.60–7.25] | 20 (76.9) | 0.058 |
| TV E’, cm/s | 6.70 [6.65–6.70] | 3 (25.0) | 8.20 [7.68–8.55] | 20 (76.9) | 0.032 |
| TV A’,cm/s | 10.6 [7.10–12.0] | 5 (41.7) | 9.10 [8.60–9.55] | 19 (73.1) | 0.477 |
Composite primary outcome was death or invasive ventilation beyond postnatal day 14.
aN: total number of patients per group.
bn: represent the number of measurements obtained.
On the day-14 TNE, infants with composite primary outcome had lower LVO (124 vs. 188 mL/kg/min, p = 0.019), higher LV-E/A ratio (1.0 vs. 0.85, p = 0.014), larger PDA size (3.35 vs. 1.5 mm, p = 0.012), more infants with bidirectional shunting of the ASD/PFO (4 (42.9%) vs. 0 (0%), p = 0.007), lower TAPSE (7.3 vs. 9.25 mm, p < 0.001), and had lower myocardial velocities (MV A′, IVS E′, IVS A′, TV S′, TV E′, and TV A′) (Table 4).
Table 4.
Echocardiographic parameters among infants with and without composite primary outcome–echo at 2 weeks of age (n = 39).
| Infants with composite adverse outcome (N = 13)a | Numbers entered: nb | Infants without composite primary outcome (N = 26)a | Numbers entered: nb | p-value | |
|---|---|---|---|---|---|
| DoL, days | 15 [14–16] | 11 (84.6) | 14 [13–15] | 11 (42.3) | 0.288 |
| Left heart | |||||
| PV S-wave, cm/s | 55.0 [39.9–67.8] | 8 (61.5) | 62.5 [60.0–66.9] | 6 (23.1) | 0.438 |
| PV D-wave, cm/s | 48.0 [42.3–63.0] | 8 (61.5) | 56.6 [51.3–62.7] | 6 (23.1) | 0.249 |
| LV FS, % | 40.0 [37.4–40.6] | 11 (84.6) | 35.7 [33.5–37.9] | 26 (100) | 0.100 |
| LV EF, % | 74.0 [70.0–76.0] | 9 (69.2) | 65.6 [63.6–71.0] | 9 (34.6) | 0.353 |
| LA:Ao ratio | 1.35 [1.30–1.40] | 6 (46.2) | 1.43 [1.40–1.49] | 6 (23.1) | 0.368 |
| LVO, mL/kg/min | 124 [95.2–179] | 10 (76.9) | 188 [180–230] | 10 (38.5) | 0.019 |
| LV-E/A ratio | 1.00 [0.94–1.07] | 7 (53.8) | 0.85 [0.80–0.92] | 19 (73.1) | 0.014 |
| LV-IVRT, ms | 46.2 [42.1–58.2] | 8 (61.5) | 44.0 [42.0–51.0] | 21 (80.8) | 0.694 |
| Shunts | |||||
| PDA size, mm | 3.35 [2.28–5.12] | 8 (61.5) | 1.50 [1.20–1.65] | 3 (11.5) | 0.012 |
| Bidirectional or right-to-left shunt, n/N (%) | 8/8 (100) | – | 2/3 (66.7) | – | 0.273 |
| ASD/PFO, n/N (%) | 9/10 (90.0) | 10 (76.9) | 18/23 (78.3) | 23 (88.5) | 0.640 |
| Left-to-right | 5/9 (55.6) | 18/18 (100) | 0.007 | ||
| Bidirectional | 4/9 (44.4) | – | 0 | – | |
| Right-to-left | 0 | 0 | |||
| Right heart | |||||
| TAPSE, mm | 7.30 [6.60–7.85] | 11 (84.6) | 9.25 [8.85–10.7] | 26 (100) | <0.001 |
| RVET:TPV ratio | 3.50 [3.10–3.90] | 9 (69.2) | 3.10 [2.92–3.38] | 22 (84.6) | 0.137 |
| RVO, mL/kg/min | 100 [79–272] | 7 (53.8) | 286 [239–289] | 9 (34.6) | 0.100 |
| Tissue Doppler Imaging | |||||
| MV S’, cm/s | 5.25 [5.00–5.89] | 6 (46.2) | 5.70 [5.60–5.90] | 17 (65.4) | 0.276 |
| MV E’, cm/s | 6.73 [5.74–8.20] | 8 (61.5) | 7.05 [6.95–7.42] | 20 (76.9) | 0.779 |
| MV A’,cm/s | 5.90 [5.24–6.65] | 7 (53.8) | 7.80 [7.48–8.20] | 20 (76.9) | 0.001 |
| IVS S’, cm/s | 4.62 [4.38–5.12] | 8 (61.5) | 5.45 [5.25–5.88] | 22 (84.6) | 0.090 |
| IVS E’, cm/s | 5.45 [5.10–5.57] | 6 (46.2) | 6.60 [6.25–6.85] | 18 (69.2) | 0.002 |
| IVS A’,cm/s | 5.45 [5.08–5.90] | 6 (46.2) | 7.40 [7.10–7.50] | 17 (65.4) | 0.003 |
| TV S’, cm/s | 6.05 [5.10–6.56] | 6 (46.2) | 7.00 [6.70–7.40] | 21 (80.8) | 0.014 |
| TV E’, cm/s | 6.60 [5.00–6.60] | 5 (38.5) | 8.20 [7.60–8.80] | 21 (80.8) | 0.005 |
| TV A’,cm/s | 7.00 [6.90–7.00] | 5 (38.5) | 8.75 [8.20–9.28] | 18 (69.2) | 0.017 |
Composite primary outcome was death or invasive ventilation beyond postnatal day 14.
aN: total number of patients per group.
bn: represent the number of measurements obtained.
Among infants of liver-up groups (liver herniation into the chest), ROC was performed using LVO, RVO, MV S′, MV E′, MV A′ and IVS S′ from the pre-operative echocardiogram (Supplementary Table 2). We found that pre-operative LVO less than 127 mL/kg/min (sensitivity=100%, specificity=100%), RVO less than 116 mL/kg/min (sensitivity=75.0%, specificity=100%) and MV S′ less than 4.95 cm/s (sensitivity=85.7%, specificity=84.6%) were predictive of the development of adverse primary outcome.
Discussion
In our cohort of 44 infants diagnosed with CDH, patients with liver herniation into the thoracic cavity demonstrated a higher likelihood of adverse composite outcomes, including mortality or the requirement for invasive mechanical ventilation beyond 14 postnatal days. These infants exhibited reduced pre-operative and early post-operative right ventricular output, as well as reduced myocardial velocities, particularly over the mitral annulus, as assessed using TDI.
Low ventricular output and PH often coexist in CDH, leading to hypoxemia, metabolic acidosis, systemic hypotension, and adverse neonatal outcomes [12, 13]. These pathophysiologic consequences are closely linked to pulmonary hypoplasia, a hallmark of CDH which is characterised by thickened alveolar walls, increased interstitial tissue, and reduced alveolar spaces available for gas exchange [14]. The associated PH arises from altered pulmonary vasoreactivity and pathological vascular remodelling and is an important component of the pathophysiology of CDH-associated PH [15]. Pulmonary hypoplasia, PH, and cardiac dysfunction intersect in those liver-up neonates, with a significant proportion of mortality and morbidity attributed to PH [16]. Pulmonary vasoconstriction contributes to an increased RV afterload, which leads to RV dysfunction under sustained wall stress [17, 18]. Consequently, it is unsurprising to observe lower RVO and lower TAPSE among infants who manifest adverse primary outcomes.
LV hypoplasia has been increasingly reported with left-sided CDH, attributed to the preferential streaming of the ductus venosus towards the right heart. This results from the rightward displacement of the heart, which alters the orientation of the vena cava and ductus venosus flow towards the right atrium, rather than the patent foramen ovale, leading to chronic underfilling of the LV [11]. As its growth is volume-dependent, reduced blood flow through the LV can result in hypoplasia in severe cases of fetal CDH [6, 19, 20].
Altered fetal haemodynamics, characterised by decreased LVO, occur in both right- and left-sided CDH cases, with an additional compressive effect on the left heart when the hernia is left-sided [21]. Our study identified lower pre-operative LVO and reduced myocardial velocities of the mitral annulus on TDI before and after the operation among those with adverse composite outcomes. We postulate that these findings reflect the extent of LV dysfunction, resulting from either hypoplasia or significant external compression.
The signs of both systolic and diastolic altered function were observed in fetuses with CDH with spectral tissue Doppler imaging, and a significant positive correlation was observed between fetal cardiac systolic function and the severity of pulmonary hypoplasia [22]. Postnatally, significant reductions in systolic and early diastolic TDI velocities in the RV have been reported in a prospective case-control study including 11 CDH infants [23]. In a cohort of 44 infants from Lucile Packard Children’s Hospital during 2009–2016, decreased LV and RV performance as measured by LV ejection fraction, right ventricular fractional area change, TAPSE, left ventricular outflow tract, right ventricular outflow tract velocity time integral, and left ventricular and right ventricular strain were significantly associated with need for ECMO [24]. Fraga et al. reported a dose-response effect between increasing severity of left heart disease, ECMO use, and mortality [25]. They found that LV dysfunction with and without LV hypoplasia had the highest odds of ECMO use and death, after adjustment for CDH severity [25]. Our study addressed gaps in the existing literature by demonstrating significant associations between lower myocardial velocities at the mitral annulus and/or interventricular septum, measured preoperatively and in the early postoperative period, and adverse outcomes of mortality or mechanical ventilation beyond 14 days.
While ductal shunting is typically bidirectional or right-to-left on echocardiogram, significant LV dysfunction can increase pressure in the left atrium, leading to a left-to-right inter-atrial shunt which is a pathognomonic and distinguishing feature of this phenotype [11]. However, the distorted cardiac axis in CDH often complicates the acquisition of optimal transthoracic views for studying atrial shunt flow. Our study is among the largest to quantify LV performance through myocardial velocities on TDI, providing valuable insights into the prediction of adverse outcomes beyond the flow direction at the atrial level.
That said, certain limitations warrant consideration. First, our study encompasses data from four quaternary centres around the world (Canada, USA, and Australia) providing intensive care for CDH patients. Although echocardiography was performed according to standard protocols (as outlined in the Methods), inter-observer variability cannot be excluded. Second, unified management protocols for CDH, including the use of ECMO, were not in place across the four participating centres. Consequently, some variations in baseline demographics collections and interventions may reflect differences in institutional preferences rather than true clinical variations, such as the use of inhaled nitric oxide or intravenous sildenafil. Third, the institutional restrictions on in-person interactions and the temporary unavailability of research personnel during the COVID pandemic led to unavoidable interruptions in recruitment procedures. The small sample size limited our ability to perform further analyses using multivariate methods. As echocardiograms were collected as part of routine clinical care, only a subset of infants underwent follow-up imaging, particularly at the two-week mark, limiting our ability to correlate these findings with adverse outcomes. Fourth, although TNE plays a critical role in assessing cardiovascular reserve and guiding decisions on the initiation and management of ECMO, the small size of the ECMO cohort in our study precluded subgroup analyses to evaluate its impact on cardiac function, even though LV dysfunction itself represents a significant risk factor for the need for ECMO [26]. Fifth, while infants with major congenital anomalies were excluded, the possibility remains that some had underlying genetic abnormalities without phenotypic expression at birth other than CDH, potentially impacting prognosis.
In conclusion, our study demonstrated an association between low ventricular outputs, lower RV systolic function, reduced myocardial performance indices at the mitral annulus and interventricular septum, and increased mortality or prolonged invasive mechanical ventilation at 14 postnatal days. Future prospective studies would benefit from incorporating standardised CDH severity scoring and advanced imaging modalities, such as speckle-tracking echocardiography and TDI, to further characterise the prognostic significance of pre-operative ventricular parameters on post-surgical and long-term outcomes.
Supplementary information
Acknowledgements
We are thankful to Ms Jennifer Claydon for her support and guidance in developing this multi-centre study.
Author contributions
JYT, ASe, MTYK, ASo, EV, MC, and PJM conceptualised and designed the study. JYT, ASe, MTYK, ASo, EV, and PJM collected the data. MTYK and JYT analysed the data. All authors interpreted the data. JYT drafted the initial manuscript. All authors critically reviewed and revised the manuscript for important intellectual content.
Funding
Dr. Joseph Y. Ting was supported by Investigator Grant Award Program of British Columbia Children’s Hospital Research Institute, Vancouver BC, Canada and Start-up funding from Women and Children’s Health Research Institute of University of Alberta, Edmonton AB, Canada.
Competing interests
The authors declare no competing interests.
Ethics approval and consent to participate
The study was conducted in accordance with the Declaration of Helsinki and approved by the University of British Columbia Research Ethics Boards (H17-01620), University of Calgary (REB17-1694), University of Iowa (M18-00338), and Monash University (RES-17-0000014L). Written informed consent was obtained from the legal guardian of the neonates.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
The online version contains supplementary material available at 10.1038/s41372-026-02670-5.
References
- 1.Paoletti M, Raffler G, Gaffi MS, Antounians L, Lauriti G, Zani A. Prevalence and risk factors for congenital diaphragmatic hernia: A global view. J Pediatr Surg. 2020;55:2297–307. [DOI] [PubMed] [Google Scholar]
- 2.Politis MD, Bermejo-Sanchez E, Canfield MA, Contiero P, Cragan JD, Dastgiri S, et al. Prevalence and mortality in children with congenital diaphragmatic hernia: a multicountry study. Ann Epidemiol. 2021;56:61–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Network. TCPS. CAPSNET Annual Report; 2009.
- 4.Schmidt AF, Goncalves FL, Nassr AC, Pereira LA, Farmer D, Sbragia L. Antenatal steroid and tracheal occlusion restore vascular endothelial growth factor receptors in congenital diaphragmatic hernia rat model. Am J Obstet Gynecol. 2010;203:184.e13–20. [DOI] [PubMed] [Google Scholar]
- 5.Pammi M, Kelagere Y, Koh S, Sisson A, Hagan J, Kailin J, et al. Prognostic value of echocardiographic parameters in congenital diaphragmatic hernia: a systematic review and meta-analysis. Arch Dis Child Fetal Neonatal Ed. 2023;108:631–7. [DOI] [PubMed] [Google Scholar]
- 6.Sehgal A, Tan K, Ferguson P. Cardiac Function Assessments in Left Bochdalek’s Hernia: Clinical Relevance. Pediatr Cardiol. 2018;39:829–36. [DOI] [PubMed] [Google Scholar]
- 7.Kuan MTY, Yadav K, Castaldo M, Tan J, Chan NH, Traynor M, et al. The impact of a care bundle with an emphasis on hemodynamic assessment on the short-term outcomes in neonates with congenital diaphragmatic hernia. J Perinatol. 2023;44:348–53. [DOI] [PubMed] [Google Scholar]
- 8.Lusk LA, Wai KC, Moon-Grady AJ, Steurer MA, Keller RL. Persistence of pulmonary hypertension by echocardiography predicts short-term outcomes in congenital diaphragmatic hernia. J Pediatr. 2015;166:251–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Moenkemeyer F, Patel N. Right ventricular diastolic function measured by tissue Doppler imaging predicts early outcome in congenital diaphragmatic hernia. Pediatr Crit Care Med. 2014;15:49–55. [DOI] [PubMed] [Google Scholar]
- 10.Prasad R, Saha B, Kumar A. Ventricular function in congenital diaphragmatic hernia: a systematic review and meta-analysis. Eur J Pediatr. 2022;181:1071–83. [DOI] [PubMed] [Google Scholar]
- 11.Surak A, Mahgoub L, Ting JY. Hemodynamic management of congenital diaphragmatic hernia: the role of targeted neonatal echocardiography. World J Pediatr Surg. 2024;7:e000790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Siebert JR, Haas JE, Beckwith JB. Left ventricular hypoplasia in congenital diaphragmatic hernia. J Pediatr Surg. 1984;19:567–71. [DOI] [PubMed] [Google Scholar]
- 13.Jain A, McNamara PJ. Persistent pulmonary hypertension of the newborn: Advances in diagnosis and treatment. Semin Fetal Neonatal Med. 2015;20:262–71. [DOI] [PubMed] [Google Scholar]
- 14.Donahoe PK, Longoni M, High FA. Polygenic Causes of Congenital Diaphragmatic Hernia Produce Common Lung Pathologies. Am J Pathol. 2016;186:2532–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Pugnaloni F, Capolupo I, Patel N, Giliberti P, Dotta A, Bagolan P, et al. Role of microRNAs in Congenital Diaphragmatic Hernia-Associated Pulmonary Hypertension. Int J Mol Sci. 2023;24:6656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.De Bie FR, Avitabile CM, Joyeux L, Hedrick HL, Russo FM, Basurto D, et al. Neonatal and fetal therapy of congenital diaphragmatic hernia-related pulmonary hypertension. Arch Dis Child Fetal Neonatal Ed. 2022;107:458–66. [DOI] [PubMed] [Google Scholar]
- 17.Pinson CW, Morton MJ, Thornburg KL. An anatomic basis for fetal right ventricular dominance and arterial pressure sensitivity. J Dev Physiol. 1987;9:253–69. [PubMed] [Google Scholar]
- 18.Patel N, Massolo AC, Kraemer US, Kipfmueller F. The heart in congenital diaphragmatic hernia: Knowns, unknowns, and future priorities. Front Pediatr. 2022;10:890422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tan CMJ, Lewandowski AJ. The Transitional Heart: From Early Embryonic and Fetal Development to Neonatal Life. Fetal Diagn Ther. 2020;47:373–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Baumgart S, Paul JJ, Huhta JC, Katz AL, Paul KE, Spettell C, et al. Cardiac malposition, redistribution of fetal cardiac output, and left heart hypoplasia reduce survival in neonates with congenital diaphragmatic hernia requiring extracorporeal membrane oxygenation. J Pediatr. 1998;133:57–62. [DOI] [PubMed] [Google Scholar]
- 21.Byrne FA, Keller RL, Meadows J, Miniati D, Brook MM, Silverman NH, et al. Severe left diaphragmatic hernia limits size of fetal left heart more than does right diaphragmatic hernia. Ultrasound Obstet Gynecol. 2015;46:688–94. [DOI] [PubMed] [Google Scholar]
- 22.Kaya B, Tayyar A, Sezer S, Kaya S. The assessment of cardiac function with tissue Doppler imaging in fetuses with congenital diaphragmatic hernia. J Matern Fetal Neonatal Med. 2020;33:1233–8. [DOI] [PubMed] [Google Scholar]
- 23.Patel N, Mills JF, Cheung MM. Assessment of right ventricular function using tissue Doppler imaging in infants with pulmonary hypertension. Neonatology. 2009;96:193–9. [DOI] [PubMed] [Google Scholar]
- 24.Altit G, Bhombal S, Van Meurs K, Tacy TA. Ventricular Performance is Associated with Need for Extracorporeal Membrane Oxygenation in Newborns with Congenital Diaphragmatic Hernia. J Pediatr. 2017;191:28–34. [DOI] [PubMed] [Google Scholar]
- 25.Fraga MV, Hedrick HL, Rintoul NE, Wang Y, Ash D, Flohr SJ, et al. Congenital diaphragmatic hernia patients with left heart hypoplasia and left ventricular dysfunction have highest odds of mortality. J Pediatr. 2024;271:114061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Guner YS, Hammond JD, Keene S, Gray B. The role of ECLS in the management of congenital diaphragmatic hernia. Semin Pediatr Surg. 2024;33:151440. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.