Ultrasound Obstet Gynecol 2015; 45: 649–656 Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/uog.14798
Assessment by three-dimensional power Doppler ultrasound of cerebral blood flow perfusion in fetuses with congenital heart disease S. ZENG*, J. ZHOU*, Q. PENG*, L. TIAN*, G. XU*, Y. ZHAO*, T. WANG† and Q. ZHOU* *Department of Ultrasonography, The Second Xiangya Hospital, Central South University, Hunan, PR China; †Department of Pediatrics, The Second Xiangya Hospital, Central South University, Changsha, Hunan, PR China
K E Y W O R D S: cerebral blood perfusion; congenital heart disease; fetal; power Doppler; three-dimensional ultrasound
ABSTRACT Objectives To use three-dimensional (3D) power Doppler ultrasound to investigate cerebral blood flow perfusion in fetuses with congenital heart disease (CHD). Methods The vascularization index (VI), flow index (FI) and vascularization flow index (VFI) in the total intracranial volume and the main arterial territories (middle cerebral artery (MCA), anterior cerebral artery (ACA) and posterior cerebral artery (PCA)) were evaluated prospectively and compared in 112 fetuses with CHD and 112 normal fetuses using 3D power Doppler. Correlations between the 3D power Doppler indices and neurodevelopment scores at 12 months of age were assessed in a subset of the CHD group, and values were compared with those of controls. Results Compared with the controls, the VI, FI and VFI of the total intracranial volume and the three main arteries were significantly higher in fetuses with hypoplastic left heart syndrome and left-sided obstructive lesions (P < 0.001), and the 3D power Doppler values in the ACA territory were significantly higher in fetuses with transposition of the great arteries (P < 0.01). The largest proportional increase in the blood flow perfusion indices in the fetuses with CHD relative to controls was observed in the ACA territory (P < 0.05). Among 41 cases with CHD that underwent testing, the mean Psychomotor Development Index (PDI) and Mental Development Index (MDI) scores were significantly lower than in 94 of the controls that were tested (P < 0.001). Among these CHD cases, total intracranial FI was positively correlated with PDI (r = 0.342, P = 0.029) and MDI (r = 0.339, P = 0.030), and ACA-VI and ACA-VFI were positively correlated with PDI (r = 0.377 and 0.389, P = 0.015 and 0.012, respectively) but were not correlated with MDI (r = 0.243 and 0.203, P = 0.126 and 0.204, respectively).
Conclusions Cerebral blood flow perfusion was increased relative to controls in most fetuses with CHD and was associated with neurodevelopment scores at 12 months. Prenatal 3D power Doppler ultrasound might help to identify cases of brain vasodilatation earlier and inform parental counseling. Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
INTRODUCTION Previous studies by our group and others have demonstrated the occurrence of impaired brain development and metabolism in fetuses with complex congenital heart disease (CHD)1 – 4 , particularly in fetuses with hypoplastic left heart syndrome (HLHS). Although the underlying causes of compromised prenatal neurological conditions in fetuses with CHD are unknown, one possible explanation is that such delays in brain maturation may be caused by decreased oxygen and nutrient delivery due to abnormal cardiac circuits and/or function. Several studies have documented abnormal cerebral blood flow in fetuses with complex heart anomalies5 – 7 . Nevertheless, most of these studies of the fetal cerebral circulation have focused only on the assessment of middle cerebral artery (MCA) Doppler parameters, such as pulsatility or resistance indices. These indices are indicators of vascular impedance and cannot estimate subtle changes in blood movement within the small vessels that are responsible for variations in cerebral blood flow perfusion8 . In addition, cerebral perfusion in other arterial territories, such as the anterior cerebral artery (ACA) and posterior cerebral artery (PCA), has not been studied in fetuses with complex CHD, although changes to perfusion in these areas appear to present during the cerebral blood flow redistribution process in fetuses with intrauterine growth restriction (IUGR)9 – 11 .
Correspondence to: Dr J. Zhou, Department of Ultrasound, Second Xiangya Hospital of Central South University, 139 Renmin Road (M), Changsha 410011, China (e-mail: [email protected]) Accepted: 15 January 2015
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
ORIGINAL PAPER
650
Three-dimensional (3D) power Doppler ultrasonography is a technique that enables the quantification of vascularization and flow of a region of interest (ROI). With this technology, three perfusion indices based on the power Doppler signal for each voxel can be calculated within the rendered volume, specifically the vascularization index (VI, percentage of color voxels in the volume), the flow index (FI, mean voxel intensity in the volume) and the vascularization flow index (VFI)12 . 3D power Doppler can offer a more complete perspective on regional hemodynamic changes. However, until now this technique has only been used for the evaluation of cerebral circulation in normal and IUGR fetuses13,14 , but not in fetuses with CHD. The objective of the present study was to evaluate fetal cerebral blood flow perfusion using 3D power Doppler ultrasound in fetuses with CHD and to investigate the possible association between fetal cerebral 3D power Doppler measurements and neurodevelopment in the population with CHD.
PATIENTS AND METHODS A prospective cross-sectional study was conducted at the Second Xiangya Hospital of Central South University in China between February 2011 and March 2014. Mothers of fetuses with confirmed CHD were recruited into the study. The inclusion criteria required one of the following fetal cardiac diagnoses: (1) HLHS; (2) left-sided obstructive lesions (LSOL); (3) right-sided obstructive lesions (RSOL); or (4) transposition of the great arteries (TGA). Gestational age-matched (±1 week) uncomplicated normal pregnancies were analyzed as controls. Exclusion criteria were: multiple-gestation pregnancies, chromosomal abnormalities or syndromes, the presence of a pre-existing neurological deficit that was not related to the cardiovascular defect, the presence of anomalies in other organ systems, persistent fetal arrhythmia and maternal complications including gestational diabetes, pre-eclampsia and thyroid disease, or the presence of an IUGR fetus. Written informed consent was obtained from all families, and the study was approved by the ethics committees of Second Xiangya Hospital. For all fetuses, routine obstetric ultrasound and fetal echocardiographic examinations were performed by one expert (Q.Z.) using a Voluson E8 (GE Healthcare Ultrasound, Milwaukee, WI, USA) ultrasound machine system with a RAB 4–8-D curvilinear probe. Gestational age was estimated from the day of the last menstrual period and was confirmed by ultrasound measurement during the first trimester. Fetal biometry was performed, including measurement of biparietal diameter (BPD), head circumference, abdominal circumference, femoral length and transverse cerebellar diameter. Fetal weight was estimated using Hadlock’s formula15 . Doppler waveforms of the MCA were obtained in the absence of fetal movements. The pulsatility index of the middle cerebral artery (MCA-PI) was calculated and converted into
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
Zeng et al. Z-scores using previously published normative data16 . Multiple views of the heart were obtained to evaluate fetal cardiac anatomy. Brain volume images were acquired using 3D ultrasound by a trained operator (Q.P.), who was blinded to the group status of each subject. The volume sample box was adjusted to include the complete fetal head on the high-resolution real-time two-dimensional ultrasound image. The volume sweep angle was set at 80◦ , and the highest acquisition quality was used. The volume originating from the level of the BPD plane was obtained. The power Doppler characteristics applied were normal quality, low wall motion filter of 1, pulse repetition frequency of 0.9 kHz and balance 150. All subjects were examined under the same conditions. Brain volumes were acquired from each subject in the absence of fetal movement and the images were stored for offline analysis. Fetal brain vascularization and blood flow calculations were performed by one observer (S.Z.), who was blinded to the clinical characteristics, using the Virtual Organ Computer-aided AnaLysis (VOCAL™) and shell histogram analysis software (4D View Version 10.0, GE Healthcare Ultrasound). To evaluate regional cerebral flow perfusion in the main arterial territories, using Box A (axial view) in the multiplanar mode, the reference plane was selected as the one with maximum vascular density in the circle of Willis. The ROIs were the MCA, ACA and PCA territories. As previously described13 , contour points were positioned on the base and apex of the MCA nearest the transducer, and the sphere mode of VOCAL was selected to calculate the volume of this ROI. Then, the VOCAL shell histogram switch was activated to calculate automatically the vascularization and blood flow indices (VI, FI and VFI) (Figure 1). Using the same method, the ACA and PCA territories closest to the transducer were evaluated (Figures 2 and 3). The distances between the contour points were the same; thus, the sample volumes were equal in each ROI. To evaluate total intracranial blood flow, the standard BPD plane (Box A, axial view) was selected as the reference plane. As previously described14 , the two contour points were placed on the anterior and posterior fetal parietal bones. The total intracranial volume was measured using manual mode and a 30◦ rotation angle. Total intracranial vascularization and blood flow indices were calculated automatically with the VOCAL shell histogram analysis software (Figure 4). All measurements were made three times, and the mean was recorded. A single expert (T.W.), who was blinded to the cardiac diagnosis and the fetal cerebral blood flow results, performed standardized 12-month neurodevelopmental testing using the Bayley Scales of Infant Development, Second Edition (BSID-II)17 . The BSID-II is approved for assessing children aged 1–42 months; it provides two summary scores, the Psychomotor Development Index (PDI) and the Mental Development Index (MDI). The normative mean ± SD for both the MDI and the PDI is 100 ± 15.
Ultrasound Obstet Gynecol 2015; 45: 649–656.
Cerebral perfusion in fetuses with CHD
651
Figure 1 Three-dimensional power Doppler evaluation of the territory of the middle cerebral artery (MCA). (a) Two contour points were positioned, one on the base and one on the apex of the MCA nearest the transducer, and the VOCAL sphere mode was selected to calculate the volume of this region of interest. (b) The VOCAL shell histogram switch was then activated to calculate automatically the vascularization and blood flow indices. FI, flow index; VFI, vascularization flow index; VI, vascularization index.
Figure 2 Three-dimensional power Doppler evaluation of the anterior cerebral artery (ACA). (a) The VOCAL sphere mode was used to select the territory of the ACA. (b) The VOCAL shell histogram switch was selected to calculate all vascularization and blood flow indices. FI, flow index; VFI, vascularization flow index; VI, vascularization index.
Statistical analysis Data are reported as mean ± SD or n (%). To account for the varying numbers of fetal echocardiograms per subject and to minimize the likelihood of a type-I error while maximizing our sample size, we limited these primary analyses to data recorded at the time of the first fetal echocardiogram. The clinical characteristics and neurodevelopmental testing scores were compared between fetuses with CHD and controls using Student’s t-test or the χ2 test. The cerebral blood flow perfusion indices (VI, FI and VFI) and MCA-PI Z-scores were compared between groups using one-way analysis of variance (ANOVA), with post-hoc Games–Howell testing to determine between-group differences. The proportional increase in each variable in CHD cases relative to their gestational age-matched control was calculated using the following formula: proportional increase = value in fetus with CHD – value in matched control / value in matched control .
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
The proportional increase in blood flow perfusion indices in the ACA, MCA and PCA territories in fetuses with CHD were also compared using one-way ANOVA, with post-hoc Games–Howell testing. Correlations between the prenatal cerebral blood flow perfusion indices and BSID neurodevelopment scores were tested using Pearson’s correlation coefficient; P < 0.05 was considered to be statistically significant. All statistical analysis was performed using PASW statistics software (PASW (SPSS) statistics, version 18.0; IBM, Armonk, NY, USA).
RESULTS During the study period, 236 pregnancies were enrolled, but we excluded 12 (seven with CHD and five normal fetuses) because of excessive fetal motion, which resulted in poor-quality images. A total of 224 fetuses were included in this study: 28 with HLHS, 32 with LSOL, 29 with RSOL, 23 with TGA and 112 gestational age-matched controls. All prenatal diagnoses of CHD
Ultrasound Obstet Gynecol 2015; 45: 649–656.
652
Zeng et al.
Figure 3 Three-dimensional power Doppler evaluation of the posterior cerebral artery (PCA). (a) The VOCAL sphere was used to select the territory of the PCA. (b) The VOCAL shell histogram switch was selected to calculate all vascularization and blood flow indices. FI, flow index; VFI, vascularization flow index; VI, vascularization index.
Figure 4 Three-dimensional power Doppler evaluation of the total intracranial volume. (a) Two contour points were placed, one on the anterior and one on the posterior fetal parietal bone. The total intracranial volume was then measured using the manual mode and a 30◦ rotation angle. (b) The VOCAL shell histogram switch was selected to calculate all vascularization and blood flow indices. FI, flow index; VFI, vascularization flow index; VI, vascularization index.
were in agreement with the postnatal echocardiograms or autopsy findings. Demographic and diagnostic details of the fetuses are listed in Table 1. The gestational age and estimated fetal weight at the time of echocardiography were not significantly different between fetuses with CHD and controls. The MCA-PI Z-scores and cerebral 3D power Doppler values are shown in Table 2 for both the total intracranial volume and specific ROIs. One-way ANOVA showed a significant difference between normal fetuses and CHD subgroups for MCA-PI Z-scores and all 3D power Doppler indices (all P < 0.001). Post-hoc analysis demonstrated that the fetuses with HLHS had significantly lower MCA-PI Z-scores than did the controls (P < 0.001). The MCA-PI Z-score was smaller in fetuses with LSOL than in controls, but this difference did not reach the level of statistical significance (P = 0.32). The MCA-PI Z-score did not differ between fetuses with RSOL and TGA and controls (P = 0.90 and P = 0.99, respectively). Total intracranial blood flow perfusion and perfusion of the territories of the three main arteries were significantly
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
increased in the fetuses with HLHS compared with normal controls (all P < 0.001). Fetuses with LSOL also had significantly higher 3D power Doppler values than did controls (all P < 0.001). The blood flow perfusion indices in the ACA territory were significantly higher in the fetuses with TGA than in the controls (all P < 0.01). There were no significant differences in 3D power Doppler indices between fetuses with RSOL and controls. In the fetuses with CHD, the increases in VI, FI and VFI relative to controls were highest in the ACA territory (P < 0.05) (Figure 5). There were no significant differences in the relative increase in blood flow perfusion between the total intracranial volume and the specific vascular territories. Among the 112 CHD cases enrolled in the study, 41 (36.6%) (six fetuses with HLHS, 10 with LSOL, 11 with RSOL and 14 with TGA) underwent BSID-II testing at a mean age of 12.43 ± 0.31 months. Among the controls, 94 (83.9%) underwent neurological examination at a mean age of 12.17 ± 0.26 months. The mean PDI and MDI scores for the CHD group (n = 41) were
Ultrasound Obstet Gynecol 2015; 45: 649–656.
Cerebral perfusion in fetuses with CHD
653
Table 1 Demographic parameters of cohort of fetuses with congenital heart disease (CHD) and normal control fetuses Variable Maternal age (years) Nulliparous ART pregnancy GA at diagnosis (weeks) EFW at diagnosis (g) Type of CHD HLHS LSOL Aortic hypoplasia/ coarctation Aortic stenosis Interrupted aortic arch RSOL TOF Valvar pulmonary stenosis PA-IVS TGA
CHD (n = 112)
Controls (n = 112)
P
27.60 ± 3.74 (18–42) 72 (64.3) 21 (18.8) 25.69 ± 2.74 (19.6–30.3) 875 ± 343 (255–1633)
27.46 ± 4.54 (20–38) 83 (74.1) 18 (16.1) 25.67 ± 2.69 (20.0–30.7) 869 ± 332 (245–1780)
0.80 0.15 0.73 0.95 0.90
28 (25.0) 32 (28.6) 18 (16.1) 4 (3.6) 10 (8.9) 29 (25.9) 15 (13.4) 10 (8.9) 4 (3.6) 23 (20.5)
Data given as mean ± SD (range) or n (%). ART, assisted reproduction technology; EFW, estimated fetal weight; GA, gestational age; HLHS, hypoplastic left heart syndrome; LSOL, left-sided obstructive lesions; PA-IVS, pulmonary atresia with intact ventricular septum; RSOL, right-sided obstructive lesions; TGA, transposition of the great arteries; TOF, tetralogy of Fallot. Table 2 Summary of middle cerebral artery (MCA) pulsatility index (PI) Z-scores and cerebral three-dimensional power Doppler indices in fetuses with congenital heart disease (CHD), according to subtype, and controls Parameter MCA-PI Z-score Total intracranial VI FI VFI ACA VI FI VFI MCA VI FI VFI PCA VI FI VFI
Normal (n = 112)
HLHS (n = 28)
LSOL (n = 32)
RSOL (n = 29)
TGA (n = 23)
P∗
−0.63 ± 0.77
−2.01 ± 1.31†
−1.04 ± 1.14
−0.46 ± 0.95
−0.55 ± 1.02
< 0.001
1.67 ± 0.42 32.92 ± 3.91 0.57 ± 0.21
3.83 ± 1.33† 78.72 ± 15.81† 1.43 ± 0.60†
3.02 ± 1.14† 52.39 ± 10.45† 1.06 ± 0.50†
1.61 ± 0.46 32.36 ± 4.57 0.56 ± 0.26
2.05 ± 0.83 35.93 ± 6.08 0.72 ± 0.31
< 0.001 < 0.001 < 0.001
1.76 ± 0.53 28.61 ± 3.42 0.52 ± 0.20
5.68 ± 1.69† 92.35 ± 16.27† 1.76 ± 0.76†
4.07 ± 1.31† 60.10 ± 9.73† 1.61 ± 0.79†
1.72 ± 0.54 29.29 ± 4.18 0.53 ± 0.21
2.70 ± 0.74‡ 33.27 ± 5.68‡ 0.79 ± 0.28‡
< 0.001 < 0.001 < 0.001
8.63 ± 2.58 36.33 ± 4.86 3.24 ± 1.30
17.25 ± 6.96† 77.18 ± 14.55† 7.20 ± 3.17†
14.65 ± 4.22† 54.50 ± 9.17† 5.57 ± 2.81†
7.97 ± 3.36 34.84 ± 7.07 2.93 ± 1.32
10.67 ± 4.21 40.43 ± 8.03 3.99 ± 1.37
< 0.001 < 0.001 < 0.001
6.43 ± 3.00 33.04 ± 5.29 1.90 ± 0.59
16.07 ± 9.63† 85.91 ± 13.22† 5.33 ± 3.09†
12.86 ± 8.15† 59.59 ± 12.62† 4.02 ± 1.77†
5.94 ± 2.90 32.21 ± 6.80 1.79 ± 0.76
9.08 ± 5.30 36.60 ± 6.66 2.52 ± 0.99
< 0.001 < 0.001 < 0.001
Data given as mean ± SD. ∗ ANOVA. †P < 0.001 vs normal. ‡P < 0.01 vs normal. ACA, anterior cerebral artery; FI, flow index; HLHS, hypoplastic left heart syndrome; LSOL, left-sided obstructive lesions; PCA, posterior cerebral artery; RSOL, right-sided obstructive lesions; TGA, transposition of the great arteries; VFI, vascularization flow index; VI, vascularization index.
72.81 ± 13.35 and 85.24 ± 11.96, respectively. These values were significantly lower than the normative values (99.37 ± 11.66 and 99.11 ± 12.49; P < 0.001). There was no significant correlation between MCA-PI and BSID score (P = 0.20 and P = 0.09 for PDI and MDI, respectively). Total intracranial FI was positively correlated with the PDI score (r = 0.342, P = 0.029) and the MDI score (r = 0.339, P = 0.030) (Figure 6), while ACA-VI and ACA-VFI were positively correlated with PDI (r = 0.377, P = 0.015; r = 0.389, P = 0.012, respectively) (Figure 7), but did not correlate with the MDI score (r = 0.243, P = 0.126; r = 0.203, P = 0.204, respectively). These correlations are reported according to CHD subtype in Table S1.
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
DISCUSSION To the best of our knowledge, this study is the first to evaluate cerebral blood flow perfusion redistribution in fetuses with CHD using 3D power Doppler ultrasound. Overall, fetuses with CHD had significantly increased cerebral blood flow perfusion compared with the controls. The VI, FI and VFI of the total intracranial volume and the three main arterial territories were higher in the fetuses with HLHS or LSOL, and the 3D power Doppler indices of the ACA territory were higher in fetuses with TGA. Fetal brain development is a function of oxygen and substrate delivery, which depends on the volume and content of blood delivered to the brain. In HLHS or aortic hypoplasia,
Ultrasound Obstet Gynecol 2015; 45: 649–656.
Zeng et al.
654 1.8
*
Mean proportional increase
1.6 1.4 *
1.2 1.0
*
0.8 0.6 0.4 0.2 0
Total
ACA
MCA
PCA
Figure 5 Mean proportional increase in cerebral three-dimensional power Doppler indices (vascularization index ( ), flow index ( ) and vascularization flow index ( )) in fetuses with congenital heart disease relative to gestational age-matched normal control fetuses, shown for the total intracranial volume and the territories of the anterior cerebral artery (ACA), middle cerebral artery (MCA) and posterior cerebral artery (PCA). There were no significant differences in the relative increase in blood flow perfusion between the total intracranial volume and specific vascular territories. *P < 0.05 vs total intracranial and other territories.
impaired left ventricle filling and/or diminished ventricular outflow results in reduced or absent flow into the ascending aorta. In TGA, the aorta arises from the right ventricle and receives deoxygenated blood returning from systemic circulation streams4 . It is possible that decreased cerebral volume and oxygen supply in fetuses with specific congenital heart defects results in autoregulation of the blood flow, which enhances cerebral perfusion18 . The MCA is considered to be the clinical standard for hemodynamic evaluation of the fetal brain. However, the clinical utility of MCA evaluation alone is limited by regional variations in brain perfusion and sensitivity in revealing neurological complications. In our study, ACA-VI, ACA-FI and ACA-VFI were increased in fetuses with TGA, whereas MCA-PI showed no change. These findings indicate that other vascular territories, such as the ACA and PCA, might provide additional information concerning the onset of the brain-sparing effect. Figueroa-Diesel et al.10 studied major fetal cerebral arteries in IUGR fetuses and suggested the existence of regional brain redistribution processes. Benavides-Serralde et al.9 found that ACA segments presented earlier signs of vasodilatation than did the MCA in IUGR fetuses. Cruz-Martinez et al.11 also demonstrated that the ACA-PI becomes abnormal 1 week before the MCA-PI does in small-for-gestational age fetuses. Additionally, in our study all 3D power Doppler indices were increased, while MCA-PI did not show a reduction, in the fetuses with LSOL, indicating that 3D power Doppler indices may be more sensitive than conventional MCA Doppler for the detection of fetuses with abnormal cerebral circulation. This concept was also explored by Bartha et al.14 , who found that the proportion of fetuses with hemodynamic redistribution was higher when using 3D power Doppler
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
indices than was the proportion of such fetuses detected using MCA-PI. In our study, cerebral blood flow perfusion increased most in the ACA territory in fetuses with CHD, indicating that the redistribution favored the frontal lobes instead of the lateral and posterior parts of the brain. The frontal region of the brain, for which the ACA is the main supply route, is believed to play a significant role in intellectual and emotional behavior, whereas the MCA supplies areas that are primarily responsible for motor control and sensory function. The PCA primarily supplies the occipital part of the brain19 . During hypoxia, cerebral circulation might change in a hierarchical manner, depending on the relative importance of the various vessels for long-term survival20 . Dubiel et al.20 demonstrated that signs of brain sparing in the ACA occur more frequently than those in the MCA in pregnancies complicated by pregnancy-induced hypertension. This physiological observation is important and may be clinically significant. Brain perfusion increments in the ACA territory in fetuses with CHD might be a mechanism for protecting important regions in the fetal brain. Such a protective response is also demonstrated by the neurodevelopmental testing in our study, in which higher ACA-VI and ACA-VFI were associated with higher PDI scores. In our study, total intracranial FI was positively correlated with PDI and MDI scores at 12 months of age, indicating that increased brain perfusion might predict better neurodevelopmental outcomes in the CHD population. Williams et al.21 found a correlation between lower MCA-PI Z-scores and higher PDI scores in patients with single-ventricle lesions. However, there was no significant association between MCA-PI Z-scores and PDI or MDI scores in our cohort. This discrepancy could be related to the different subject populations and differences in gestational age between the two studies. This relationship is the opposite to that observed with IUGR. In fetuses with IUGR, increased FI has been associated with higher perinatal mortality rates16 . The differences in pathophysiology between fetal disease states may explain this contradiction. In IUGR, the primary problem is placental insufficiency, which leads to inadequate gas and substrate exchange. However, the primary pathological mechanism of CHD is the hemodynamic consequence of an altered pathway or diminished ventricular outflow, which leads to the delivery of relatively deoxygenated blood to the brain. It is possible that the autoregulatory increase in brain perfusion associated with diminished oxygen delivery is sufficient to meet the cerebral metabolic demands in the early stages of CHD. It is interesting to note that the relationship between 3D power Doppler measurements and development scores appears to vary between CHD subgroups (Table S1). However, the numbers in the subgroups were limited for an in-depth analysis. There are a number of limitations to our study. First, measurements of 3D power Doppler indices based on static volumes are dependent on gain settings and attenuation at acquisition22,23 and there have been contradictory
Ultrasound Obstet Gynecol 2015; 45: 649–656.
Cerebral perfusion in fetuses with CHD
(b)
110
100
110
100 Mental Development Index score
Psychomotor Development Index score
(a)
655
90
80
70
90
80
70
60
60
50
50 20
40
60
80
100
120
20
40
Total-FI
60
80
100
120
Total-FI
Figure 6 Scatterplots with overall linear regression lines ( ) showing relationship between total intracranial flow index (Total-FI) on three-dimensional power Doppler in fetuses with congenital heart disease (CHD) and neurodevelopmental scores at 12 months: (a) Psychomotor Development Index (r = 0.342, P = 0.029) and (b) Mental Development Index (r = 0.339, P = 0.030). Individual data points are plotted according to CHD subtype: hypoplastic left heart syndrome ( ); left-sided obstructive lesions ( ); right-sided obstructive lesions ( ); and transposition of the great arteries ( ).
(b)
110
100
Psychomotor Development Index score
Psychomotor Development Index score
(a)
90
80
70
60
50
110
100
90
80
70
60
50 0.0
2.0
4.0
6.0
8.0
ACA-VI
0.0
1.0
2.0
3.0
4.0
ACA-VFI
Figure 7 Scatterplots with overall linear regression lines ( ) showing relationship between: (a) anterior cerebral artery (ACA) flow indices on three-dimensional power Doppler in fetuses with congenital heart disease (CHD) and Psychomotor Development Index at 12 months: (a) vascularization index (VI) (r = 0.377, P = 0.015) and (b) vascularization flow index (VFI) (r = 0.389, P = 0.012). Individual data points are plotted according to CHD subtype: hypoplastic left heart syndrome ( ); left-sided obstructive lesions ( ); right-sided obstructive lesions ( ); and transposition of the great arteries ( ).
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
Ultrasound Obstet Gynecol 2015; 45: 649–656.
Zeng et al.
656
reports regarding their reproducibility24,25 because of differences in machine settings, ROI, image acquisition and analysis. In this study, therefore, we attempted to keep the acquisition and measurement settings as consistent as possible. Second, additional studies including a greater number of cases are needed to ascertain the differences between CHD subgroups. Neurodevelopmental outcomes are influenced by other variables, such as socioeconomic status, illness severity and treatment strategies. Therefore, our data only show an association between fetal cerebral blood perfusion and neurodevelopment, not causation. In conclusion, cerebral blood flow perfusion was significantly higher in fetuses with CHD than in controls, and this redistribution occurred most prominently in the ACA territory. Higher fetal cerebral 3D power Doppler measurements, namely total intracranial FI, ACA-VI and ACA-VFI, were associated with higher neurodevelopmental test scores in the CHD population. Although further studies are needed to elucidate the predictive value of ultrasound parameters for each type of CHD, the data presented in this report indicate that 3D power Doppler ultrasound may help to identify cases of brain vasodilatation earlier and thus inform parental counseling.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
ACKNOWLEDGMENTS 16.
This study was supported by the National Clinical Key Subject (Medical Imageology) Construction Project of China, the State Natural Sciences Foundation of China (no. 81271593) and the Hunan Province Science & Technology program (no. 2012FJ4142, 2013SK3035).
17. 18. 19.
20.
REFERENCES 1. Limperopoulos C, Tworetzky W, McElhinney DB, Newburger JW, Brown DW, Robertson RL Jr, Guizard N, McGrath E, Geva J, Annese D, Dunbar-Masterson C, Trainor B, Laussen PC, du Plessis AJ. Brain volume and metabolism in fetuses with congenital heart disease: evaluation with quantitative magnetic resonance imaging and spectroscopy. Circulation 2010; 121: 26–33. 2. Clouchoux C, du Plessis AJ, Bouyssi-Kobar M, Tworetzky W, McElhinney DB, Brown DW, Gholipour A, Kudelski D, Warfield SK, McCarter RJ, Robertson RL Jr, Evans AC, Newburger JW, Limperopoulos C. Delayed cortical development in fetuses with complex congenital heart disease. Cereb Cortex 2013; 23: 2932–2943. 3. Khalil A, Suff N, Thilaganathan B, Hurrell A, Cooper D, Carvalho JS. Brain abnormalities and neurodevelopmental delay in congenital heart disease: systematic review and meta-analysis. Ultrasound Obstet Gynecol 2014; 43: 14–24. 4. Zeng S, Zhou Q, Zhou J, Li M, Long C, Peng Q. Impaired brain growth of intracranial structures in fetuses with complex congenital heart disease as measured
21.
22.
23.
24.
25.
by 3D ultrasonography. Ultrasound Obstet Gynecol 2014. DOI: 10.1002/uog.14677. [Epub ahead of print] Donofrio MT, Bremer YA, Schieken RM, Gennings C, Morton LD, Eidem BW, Cetta F, Falkensammer CB, Huhta JC, Kleinman CS. Autoregulation of cerebral blood flow in fetuses with congenital heart disease: the brain sparing effect. Pediatr Cardiol 2003; 24: 436–443. Jouannic JM, Benachi A, Bonnet D, Fermont L, Le Bidois J, Dumez Y, Dommergues M. Middle cerebral artery Doppler in fetuses with transposition of the great arteries. Ultrasound Obstet Gynecol 2002; 20: 122–124. Kaltman JR, Di H, Tian Z, Rychik J. Impact of congenital heart disease on cerebrovascular blood flow dynamics in the fetus. Ultrasound Obstet Gynecol 2005; 25: 32–36. Hernandez-Andrade E, Jansson T, Figueroa-Diesel H, Rangel-Nava H, Acosta-Rojas ´ E. Evaluation of fetal regional cerebral blood perfusion using power R, Gratacos Doppler ultrasound and the estimation of fractional moving blood volume. Ultrasound Obstet Gynecol 2007; 29: 556–561. Benavides-Serralde A, ScheierM, Cruz-Martinez R, Crispi F, Figueras F, Gratacos E, Hernandez-Andrade E. Changes in central and peripheral circulation in intrauterine growth-restricted fetuses at different stages of umbilical artery flow deterioration: new fetal cardiac and brain parameters. Gynecol Obstet Invest 2011; 71: 274–280. Figueroa-Diesel H, Hernandez-Andrade E, Acosta-Rojas R, Cabero L, Gratacos E. Doppler changes in the main fetal brain arteries at different stages of hemodynamic adaptation in severe intrauterine growth restriction. Ultrasound Obstet Gynecol 2007; 30: 297–302. ´ E. Cruz-Martinez R, Figueras F, Hernandez-Andrade E, Puerto B, Gratacos Longitudinal brain perfusion changes in near-term small-for gestational-age fetuses as measured by spectral Doppler indices or by fractional moving blood volume. Am J Obstet Gynecol 2010; 203: 42.e1–e6. Dubiel M, Hammid A, Breborowicz A, Pietryga M, Sladkevicius P, Olofsson PA, Breborowicz GH, Gudmundsson S. Flow index evaluation of 3-D volume flow images: an in vivo and in vitro study. Ultrasound Med Biol 2006; 32: 665–671. ´ Milani HJ, de Sa´ Barreto EQ, Araujo Junior E, Haratz KK, Rolo LC, Nardozza LM, Moron AF. Assessment of cerebral circulation in normal fetuses by three-dimensional power Doppler ultrasonography. Eur J Radiol 2012; 81: 514–521. Bartha JL, Moya EM, Herv´ıas-Vivancos B. Three-dimensional power Doppler analysis of cerebral circulation in normal and growth-restricted fetuses. J Cereb Blood Flow Metab 2009; 29: 1609–1618. Hadlock FP, Harrist RB, Sharman RS, Deter RL, Park SK. Estimation of fetal weight with the use of head, body, and femur measurements-a prospective study. Am J Obstet Gynecol 1985; 151: 333–337. Arduini D, Rizzo G. Normal values of pulsatility index from fetal vessels: a cross-sectional study on 1556 healthy fetuses. J Perinat Med 1990; 18: 165–172. Bayley N. Bayley Scales of Infant Development (2nd edn). The Psychological Corporation: San Antonio, TX, 1993. Donofrio MT. The heart–brain interaction in the fetus: Cerebrovascular blood flow in the developing human. Progr Pediatr Cardiol 2006; 22: 41–51. Lockhart RD, Hamilton GF, Fyfe FW. Arteries of the central nervous system. In Anatomy of the Human Body. Faber and Faber Limited: London, UK, 1965: 607–609. Dubiel M, Gunnarsson GO, Gudmundsson S. Blood redistribution in the fetal brain during chronic hypoxia. Ultrasound Obstet Gynecol 2002; 20: 117–121. Williams IA, Fifer C, Jaeggi E, Levine JC, Michelfelder EC, Szwast AL. The association of fetal cerebrovascular resistance with early neurodevelopment in single ventricle congenital heart disease. Am Heart J 2013; 165: 544–550.e1. Miyague AH, Raine-Fenning NJ, Pavan TZ, Polanski LT, Baumgarten MN, Nastri CO, Martins WP. Influence of gain adjustment on 3-dimensional power Doppler indices and on spatiotemporal image correlation volumetric pulsatility indices using a flow phantom. J Ultrasound Med 2013; 32: 1831–1836. Miyague AH, Pavan TZ, Grillo FW, Teixeira DM, Nastri CO, Martins WP. Influence of attenuation on three-dimensional power Doppler indices and STIC volumetric pulsatility index: a flow phantom experiment. Ultrasound Obstet Gynecol 2014; 43: 103–105. Lai PK, Wang YA, Welsh AW. Reproducibility of regional placental vascularity/perfusion measurement using 3D power Doppler. Ultrasound Obstet Gynecol 2010; 36: 202–209. ˜ Filho HA, Mattar R, Araujo Junior ´ Guimaraes E, da Costa LL, de Mello Junior CF, Nardozza LM, Nowak PM, Moron AF. Reproducibility of three-dimensional power Doppler placental vascular indices in pregnancies between 26 and 35 weeks. Arch Gynecol Obstet 2011; 283: 213–217.
SUPPORTING INFORMATION ON THE INTERNET The following supporting information may be found in the online version of this article: Table S1 Linear correlation coefficients for the relationship between three-dimensional power Doppler measurements and neurodevelopment scores at 12 months according to subtype of congenital heart disease
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
Ultrasound Obstet Gynecol 2015; 45: 649–656.
Assessment by three-dimensional power Doppler ultrasound of cerebral blood flow perfusion in fetuses with congenital heart disease S. ZENG*, J. ZHOU*, Q. PENG*, L. TIAN*, G. XU*, Y. ZHAO*, T. WANG† and Q. ZHOU* *Department of Ultrasonography, The Second Xiangya Hospital, Central South University, Hunan, PR China; †Department of Pediatrics, The Second Xiangya Hospital, Central South University, Changsha, Hunan, PR China
K E Y W O R D S: cerebral blood perfusion; congenital heart disease; fetal; power Doppler; three-dimensional ultrasound
ABSTRACT Objectives To use three-dimensional (3D) power Doppler ultrasound to investigate cerebral blood flow perfusion in fetuses with congenital heart disease (CHD). Methods The vascularization index (VI), flow index (FI) and vascularization flow index (VFI) in the total intracranial volume and the main arterial territories (middle cerebral artery (MCA), anterior cerebral artery (ACA) and posterior cerebral artery (PCA)) were evaluated prospectively and compared in 112 fetuses with CHD and 112 normal fetuses using 3D power Doppler. Correlations between the 3D power Doppler indices and neurodevelopment scores at 12 months of age were assessed in a subset of the CHD group, and values were compared with those of controls. Results Compared with the controls, the VI, FI and VFI of the total intracranial volume and the three main arteries were significantly higher in fetuses with hypoplastic left heart syndrome and left-sided obstructive lesions (P < 0.001), and the 3D power Doppler values in the ACA territory were significantly higher in fetuses with transposition of the great arteries (P < 0.01). The largest proportional increase in the blood flow perfusion indices in the fetuses with CHD relative to controls was observed in the ACA territory (P < 0.05). Among 41 cases with CHD that underwent testing, the mean Psychomotor Development Index (PDI) and Mental Development Index (MDI) scores were significantly lower than in 94 of the controls that were tested (P < 0.001). Among these CHD cases, total intracranial FI was positively correlated with PDI (r = 0.342, P = 0.029) and MDI (r = 0.339, P = 0.030), and ACA-VI and ACA-VFI were positively correlated with PDI (r = 0.377 and 0.389, P = 0.015 and 0.012, respectively) but were not correlated with MDI (r = 0.243 and 0.203, P = 0.126 and 0.204, respectively).
Conclusions Cerebral blood flow perfusion was increased relative to controls in most fetuses with CHD and was associated with neurodevelopment scores at 12 months. Prenatal 3D power Doppler ultrasound might help to identify cases of brain vasodilatation earlier and inform parental counseling. Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
INTRODUCTION Previous studies by our group and others have demonstrated the occurrence of impaired brain development and metabolism in fetuses with complex congenital heart disease (CHD)1 – 4 , particularly in fetuses with hypoplastic left heart syndrome (HLHS). Although the underlying causes of compromised prenatal neurological conditions in fetuses with CHD are unknown, one possible explanation is that such delays in brain maturation may be caused by decreased oxygen and nutrient delivery due to abnormal cardiac circuits and/or function. Several studies have documented abnormal cerebral blood flow in fetuses with complex heart anomalies5 – 7 . Nevertheless, most of these studies of the fetal cerebral circulation have focused only on the assessment of middle cerebral artery (MCA) Doppler parameters, such as pulsatility or resistance indices. These indices are indicators of vascular impedance and cannot estimate subtle changes in blood movement within the small vessels that are responsible for variations in cerebral blood flow perfusion8 . In addition, cerebral perfusion in other arterial territories, such as the anterior cerebral artery (ACA) and posterior cerebral artery (PCA), has not been studied in fetuses with complex CHD, although changes to perfusion in these areas appear to present during the cerebral blood flow redistribution process in fetuses with intrauterine growth restriction (IUGR)9 – 11 .
Correspondence to: Dr J. Zhou, Department of Ultrasound, Second Xiangya Hospital of Central South University, 139 Renmin Road (M), Changsha 410011, China (e-mail: [email protected]) Accepted: 15 January 2015
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
ORIGINAL PAPER
650
Three-dimensional (3D) power Doppler ultrasonography is a technique that enables the quantification of vascularization and flow of a region of interest (ROI). With this technology, three perfusion indices based on the power Doppler signal for each voxel can be calculated within the rendered volume, specifically the vascularization index (VI, percentage of color voxels in the volume), the flow index (FI, mean voxel intensity in the volume) and the vascularization flow index (VFI)12 . 3D power Doppler can offer a more complete perspective on regional hemodynamic changes. However, until now this technique has only been used for the evaluation of cerebral circulation in normal and IUGR fetuses13,14 , but not in fetuses with CHD. The objective of the present study was to evaluate fetal cerebral blood flow perfusion using 3D power Doppler ultrasound in fetuses with CHD and to investigate the possible association between fetal cerebral 3D power Doppler measurements and neurodevelopment in the population with CHD.
PATIENTS AND METHODS A prospective cross-sectional study was conducted at the Second Xiangya Hospital of Central South University in China between February 2011 and March 2014. Mothers of fetuses with confirmed CHD were recruited into the study. The inclusion criteria required one of the following fetal cardiac diagnoses: (1) HLHS; (2) left-sided obstructive lesions (LSOL); (3) right-sided obstructive lesions (RSOL); or (4) transposition of the great arteries (TGA). Gestational age-matched (±1 week) uncomplicated normal pregnancies were analyzed as controls. Exclusion criteria were: multiple-gestation pregnancies, chromosomal abnormalities or syndromes, the presence of a pre-existing neurological deficit that was not related to the cardiovascular defect, the presence of anomalies in other organ systems, persistent fetal arrhythmia and maternal complications including gestational diabetes, pre-eclampsia and thyroid disease, or the presence of an IUGR fetus. Written informed consent was obtained from all families, and the study was approved by the ethics committees of Second Xiangya Hospital. For all fetuses, routine obstetric ultrasound and fetal echocardiographic examinations were performed by one expert (Q.Z.) using a Voluson E8 (GE Healthcare Ultrasound, Milwaukee, WI, USA) ultrasound machine system with a RAB 4–8-D curvilinear probe. Gestational age was estimated from the day of the last menstrual period and was confirmed by ultrasound measurement during the first trimester. Fetal biometry was performed, including measurement of biparietal diameter (BPD), head circumference, abdominal circumference, femoral length and transverse cerebellar diameter. Fetal weight was estimated using Hadlock’s formula15 . Doppler waveforms of the MCA were obtained in the absence of fetal movements. The pulsatility index of the middle cerebral artery (MCA-PI) was calculated and converted into
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
Zeng et al. Z-scores using previously published normative data16 . Multiple views of the heart were obtained to evaluate fetal cardiac anatomy. Brain volume images were acquired using 3D ultrasound by a trained operator (Q.P.), who was blinded to the group status of each subject. The volume sample box was adjusted to include the complete fetal head on the high-resolution real-time two-dimensional ultrasound image. The volume sweep angle was set at 80◦ , and the highest acquisition quality was used. The volume originating from the level of the BPD plane was obtained. The power Doppler characteristics applied were normal quality, low wall motion filter of 1, pulse repetition frequency of 0.9 kHz and balance 150. All subjects were examined under the same conditions. Brain volumes were acquired from each subject in the absence of fetal movement and the images were stored for offline analysis. Fetal brain vascularization and blood flow calculations were performed by one observer (S.Z.), who was blinded to the clinical characteristics, using the Virtual Organ Computer-aided AnaLysis (VOCAL™) and shell histogram analysis software (4D View Version 10.0, GE Healthcare Ultrasound). To evaluate regional cerebral flow perfusion in the main arterial territories, using Box A (axial view) in the multiplanar mode, the reference plane was selected as the one with maximum vascular density in the circle of Willis. The ROIs were the MCA, ACA and PCA territories. As previously described13 , contour points were positioned on the base and apex of the MCA nearest the transducer, and the sphere mode of VOCAL was selected to calculate the volume of this ROI. Then, the VOCAL shell histogram switch was activated to calculate automatically the vascularization and blood flow indices (VI, FI and VFI) (Figure 1). Using the same method, the ACA and PCA territories closest to the transducer were evaluated (Figures 2 and 3). The distances between the contour points were the same; thus, the sample volumes were equal in each ROI. To evaluate total intracranial blood flow, the standard BPD plane (Box A, axial view) was selected as the reference plane. As previously described14 , the two contour points were placed on the anterior and posterior fetal parietal bones. The total intracranial volume was measured using manual mode and a 30◦ rotation angle. Total intracranial vascularization and blood flow indices were calculated automatically with the VOCAL shell histogram analysis software (Figure 4). All measurements were made three times, and the mean was recorded. A single expert (T.W.), who was blinded to the cardiac diagnosis and the fetal cerebral blood flow results, performed standardized 12-month neurodevelopmental testing using the Bayley Scales of Infant Development, Second Edition (BSID-II)17 . The BSID-II is approved for assessing children aged 1–42 months; it provides two summary scores, the Psychomotor Development Index (PDI) and the Mental Development Index (MDI). The normative mean ± SD for both the MDI and the PDI is 100 ± 15.
Ultrasound Obstet Gynecol 2015; 45: 649–656.
Cerebral perfusion in fetuses with CHD
651
Figure 1 Three-dimensional power Doppler evaluation of the territory of the middle cerebral artery (MCA). (a) Two contour points were positioned, one on the base and one on the apex of the MCA nearest the transducer, and the VOCAL sphere mode was selected to calculate the volume of this region of interest. (b) The VOCAL shell histogram switch was then activated to calculate automatically the vascularization and blood flow indices. FI, flow index; VFI, vascularization flow index; VI, vascularization index.
Figure 2 Three-dimensional power Doppler evaluation of the anterior cerebral artery (ACA). (a) The VOCAL sphere mode was used to select the territory of the ACA. (b) The VOCAL shell histogram switch was selected to calculate all vascularization and blood flow indices. FI, flow index; VFI, vascularization flow index; VI, vascularization index.
Statistical analysis Data are reported as mean ± SD or n (%). To account for the varying numbers of fetal echocardiograms per subject and to minimize the likelihood of a type-I error while maximizing our sample size, we limited these primary analyses to data recorded at the time of the first fetal echocardiogram. The clinical characteristics and neurodevelopmental testing scores were compared between fetuses with CHD and controls using Student’s t-test or the χ2 test. The cerebral blood flow perfusion indices (VI, FI and VFI) and MCA-PI Z-scores were compared between groups using one-way analysis of variance (ANOVA), with post-hoc Games–Howell testing to determine between-group differences. The proportional increase in each variable in CHD cases relative to their gestational age-matched control was calculated using the following formula: proportional increase = value in fetus with CHD – value in matched control / value in matched control .
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
The proportional increase in blood flow perfusion indices in the ACA, MCA and PCA territories in fetuses with CHD were also compared using one-way ANOVA, with post-hoc Games–Howell testing. Correlations between the prenatal cerebral blood flow perfusion indices and BSID neurodevelopment scores were tested using Pearson’s correlation coefficient; P < 0.05 was considered to be statistically significant. All statistical analysis was performed using PASW statistics software (PASW (SPSS) statistics, version 18.0; IBM, Armonk, NY, USA).
RESULTS During the study period, 236 pregnancies were enrolled, but we excluded 12 (seven with CHD and five normal fetuses) because of excessive fetal motion, which resulted in poor-quality images. A total of 224 fetuses were included in this study: 28 with HLHS, 32 with LSOL, 29 with RSOL, 23 with TGA and 112 gestational age-matched controls. All prenatal diagnoses of CHD
Ultrasound Obstet Gynecol 2015; 45: 649–656.
652
Zeng et al.
Figure 3 Three-dimensional power Doppler evaluation of the posterior cerebral artery (PCA). (a) The VOCAL sphere was used to select the territory of the PCA. (b) The VOCAL shell histogram switch was selected to calculate all vascularization and blood flow indices. FI, flow index; VFI, vascularization flow index; VI, vascularization index.
Figure 4 Three-dimensional power Doppler evaluation of the total intracranial volume. (a) Two contour points were placed, one on the anterior and one on the posterior fetal parietal bone. The total intracranial volume was then measured using the manual mode and a 30◦ rotation angle. (b) The VOCAL shell histogram switch was selected to calculate all vascularization and blood flow indices. FI, flow index; VFI, vascularization flow index; VI, vascularization index.
were in agreement with the postnatal echocardiograms or autopsy findings. Demographic and diagnostic details of the fetuses are listed in Table 1. The gestational age and estimated fetal weight at the time of echocardiography were not significantly different between fetuses with CHD and controls. The MCA-PI Z-scores and cerebral 3D power Doppler values are shown in Table 2 for both the total intracranial volume and specific ROIs. One-way ANOVA showed a significant difference between normal fetuses and CHD subgroups for MCA-PI Z-scores and all 3D power Doppler indices (all P < 0.001). Post-hoc analysis demonstrated that the fetuses with HLHS had significantly lower MCA-PI Z-scores than did the controls (P < 0.001). The MCA-PI Z-score was smaller in fetuses with LSOL than in controls, but this difference did not reach the level of statistical significance (P = 0.32). The MCA-PI Z-score did not differ between fetuses with RSOL and TGA and controls (P = 0.90 and P = 0.99, respectively). Total intracranial blood flow perfusion and perfusion of the territories of the three main arteries were significantly
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
increased in the fetuses with HLHS compared with normal controls (all P < 0.001). Fetuses with LSOL also had significantly higher 3D power Doppler values than did controls (all P < 0.001). The blood flow perfusion indices in the ACA territory were significantly higher in the fetuses with TGA than in the controls (all P < 0.01). There were no significant differences in 3D power Doppler indices between fetuses with RSOL and controls. In the fetuses with CHD, the increases in VI, FI and VFI relative to controls were highest in the ACA territory (P < 0.05) (Figure 5). There were no significant differences in the relative increase in blood flow perfusion between the total intracranial volume and the specific vascular territories. Among the 112 CHD cases enrolled in the study, 41 (36.6%) (six fetuses with HLHS, 10 with LSOL, 11 with RSOL and 14 with TGA) underwent BSID-II testing at a mean age of 12.43 ± 0.31 months. Among the controls, 94 (83.9%) underwent neurological examination at a mean age of 12.17 ± 0.26 months. The mean PDI and MDI scores for the CHD group (n = 41) were
Ultrasound Obstet Gynecol 2015; 45: 649–656.
Cerebral perfusion in fetuses with CHD
653
Table 1 Demographic parameters of cohort of fetuses with congenital heart disease (CHD) and normal control fetuses Variable Maternal age (years) Nulliparous ART pregnancy GA at diagnosis (weeks) EFW at diagnosis (g) Type of CHD HLHS LSOL Aortic hypoplasia/ coarctation Aortic stenosis Interrupted aortic arch RSOL TOF Valvar pulmonary stenosis PA-IVS TGA
CHD (n = 112)
Controls (n = 112)
P
27.60 ± 3.74 (18–42) 72 (64.3) 21 (18.8) 25.69 ± 2.74 (19.6–30.3) 875 ± 343 (255–1633)
27.46 ± 4.54 (20–38) 83 (74.1) 18 (16.1) 25.67 ± 2.69 (20.0–30.7) 869 ± 332 (245–1780)
0.80 0.15 0.73 0.95 0.90
28 (25.0) 32 (28.6) 18 (16.1) 4 (3.6) 10 (8.9) 29 (25.9) 15 (13.4) 10 (8.9) 4 (3.6) 23 (20.5)
Data given as mean ± SD (range) or n (%). ART, assisted reproduction technology; EFW, estimated fetal weight; GA, gestational age; HLHS, hypoplastic left heart syndrome; LSOL, left-sided obstructive lesions; PA-IVS, pulmonary atresia with intact ventricular septum; RSOL, right-sided obstructive lesions; TGA, transposition of the great arteries; TOF, tetralogy of Fallot. Table 2 Summary of middle cerebral artery (MCA) pulsatility index (PI) Z-scores and cerebral three-dimensional power Doppler indices in fetuses with congenital heart disease (CHD), according to subtype, and controls Parameter MCA-PI Z-score Total intracranial VI FI VFI ACA VI FI VFI MCA VI FI VFI PCA VI FI VFI
Normal (n = 112)
HLHS (n = 28)
LSOL (n = 32)
RSOL (n = 29)
TGA (n = 23)
P∗
−0.63 ± 0.77
−2.01 ± 1.31†
−1.04 ± 1.14
−0.46 ± 0.95
−0.55 ± 1.02
< 0.001
1.67 ± 0.42 32.92 ± 3.91 0.57 ± 0.21
3.83 ± 1.33† 78.72 ± 15.81† 1.43 ± 0.60†
3.02 ± 1.14† 52.39 ± 10.45† 1.06 ± 0.50†
1.61 ± 0.46 32.36 ± 4.57 0.56 ± 0.26
2.05 ± 0.83 35.93 ± 6.08 0.72 ± 0.31
< 0.001 < 0.001 < 0.001
1.76 ± 0.53 28.61 ± 3.42 0.52 ± 0.20
5.68 ± 1.69† 92.35 ± 16.27† 1.76 ± 0.76†
4.07 ± 1.31† 60.10 ± 9.73† 1.61 ± 0.79†
1.72 ± 0.54 29.29 ± 4.18 0.53 ± 0.21
2.70 ± 0.74‡ 33.27 ± 5.68‡ 0.79 ± 0.28‡
< 0.001 < 0.001 < 0.001
8.63 ± 2.58 36.33 ± 4.86 3.24 ± 1.30
17.25 ± 6.96† 77.18 ± 14.55† 7.20 ± 3.17†
14.65 ± 4.22† 54.50 ± 9.17† 5.57 ± 2.81†
7.97 ± 3.36 34.84 ± 7.07 2.93 ± 1.32
10.67 ± 4.21 40.43 ± 8.03 3.99 ± 1.37
< 0.001 < 0.001 < 0.001
6.43 ± 3.00 33.04 ± 5.29 1.90 ± 0.59
16.07 ± 9.63† 85.91 ± 13.22† 5.33 ± 3.09†
12.86 ± 8.15† 59.59 ± 12.62† 4.02 ± 1.77†
5.94 ± 2.90 32.21 ± 6.80 1.79 ± 0.76
9.08 ± 5.30 36.60 ± 6.66 2.52 ± 0.99
< 0.001 < 0.001 < 0.001
Data given as mean ± SD. ∗ ANOVA. †P < 0.001 vs normal. ‡P < 0.01 vs normal. ACA, anterior cerebral artery; FI, flow index; HLHS, hypoplastic left heart syndrome; LSOL, left-sided obstructive lesions; PCA, posterior cerebral artery; RSOL, right-sided obstructive lesions; TGA, transposition of the great arteries; VFI, vascularization flow index; VI, vascularization index.
72.81 ± 13.35 and 85.24 ± 11.96, respectively. These values were significantly lower than the normative values (99.37 ± 11.66 and 99.11 ± 12.49; P < 0.001). There was no significant correlation between MCA-PI and BSID score (P = 0.20 and P = 0.09 for PDI and MDI, respectively). Total intracranial FI was positively correlated with the PDI score (r = 0.342, P = 0.029) and the MDI score (r = 0.339, P = 0.030) (Figure 6), while ACA-VI and ACA-VFI were positively correlated with PDI (r = 0.377, P = 0.015; r = 0.389, P = 0.012, respectively) (Figure 7), but did not correlate with the MDI score (r = 0.243, P = 0.126; r = 0.203, P = 0.204, respectively). These correlations are reported according to CHD subtype in Table S1.
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
DISCUSSION To the best of our knowledge, this study is the first to evaluate cerebral blood flow perfusion redistribution in fetuses with CHD using 3D power Doppler ultrasound. Overall, fetuses with CHD had significantly increased cerebral blood flow perfusion compared with the controls. The VI, FI and VFI of the total intracranial volume and the three main arterial territories were higher in the fetuses with HLHS or LSOL, and the 3D power Doppler indices of the ACA territory were higher in fetuses with TGA. Fetal brain development is a function of oxygen and substrate delivery, which depends on the volume and content of blood delivered to the brain. In HLHS or aortic hypoplasia,
Ultrasound Obstet Gynecol 2015; 45: 649–656.
Zeng et al.
654 1.8
*
Mean proportional increase
1.6 1.4 *
1.2 1.0
*
0.8 0.6 0.4 0.2 0
Total
ACA
MCA
PCA
Figure 5 Mean proportional increase in cerebral three-dimensional power Doppler indices (vascularization index ( ), flow index ( ) and vascularization flow index ( )) in fetuses with congenital heart disease relative to gestational age-matched normal control fetuses, shown for the total intracranial volume and the territories of the anterior cerebral artery (ACA), middle cerebral artery (MCA) and posterior cerebral artery (PCA). There were no significant differences in the relative increase in blood flow perfusion between the total intracranial volume and specific vascular territories. *P < 0.05 vs total intracranial and other territories.
impaired left ventricle filling and/or diminished ventricular outflow results in reduced or absent flow into the ascending aorta. In TGA, the aorta arises from the right ventricle and receives deoxygenated blood returning from systemic circulation streams4 . It is possible that decreased cerebral volume and oxygen supply in fetuses with specific congenital heart defects results in autoregulation of the blood flow, which enhances cerebral perfusion18 . The MCA is considered to be the clinical standard for hemodynamic evaluation of the fetal brain. However, the clinical utility of MCA evaluation alone is limited by regional variations in brain perfusion and sensitivity in revealing neurological complications. In our study, ACA-VI, ACA-FI and ACA-VFI were increased in fetuses with TGA, whereas MCA-PI showed no change. These findings indicate that other vascular territories, such as the ACA and PCA, might provide additional information concerning the onset of the brain-sparing effect. Figueroa-Diesel et al.10 studied major fetal cerebral arteries in IUGR fetuses and suggested the existence of regional brain redistribution processes. Benavides-Serralde et al.9 found that ACA segments presented earlier signs of vasodilatation than did the MCA in IUGR fetuses. Cruz-Martinez et al.11 also demonstrated that the ACA-PI becomes abnormal 1 week before the MCA-PI does in small-for-gestational age fetuses. Additionally, in our study all 3D power Doppler indices were increased, while MCA-PI did not show a reduction, in the fetuses with LSOL, indicating that 3D power Doppler indices may be more sensitive than conventional MCA Doppler for the detection of fetuses with abnormal cerebral circulation. This concept was also explored by Bartha et al.14 , who found that the proportion of fetuses with hemodynamic redistribution was higher when using 3D power Doppler
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
indices than was the proportion of such fetuses detected using MCA-PI. In our study, cerebral blood flow perfusion increased most in the ACA territory in fetuses with CHD, indicating that the redistribution favored the frontal lobes instead of the lateral and posterior parts of the brain. The frontal region of the brain, for which the ACA is the main supply route, is believed to play a significant role in intellectual and emotional behavior, whereas the MCA supplies areas that are primarily responsible for motor control and sensory function. The PCA primarily supplies the occipital part of the brain19 . During hypoxia, cerebral circulation might change in a hierarchical manner, depending on the relative importance of the various vessels for long-term survival20 . Dubiel et al.20 demonstrated that signs of brain sparing in the ACA occur more frequently than those in the MCA in pregnancies complicated by pregnancy-induced hypertension. This physiological observation is important and may be clinically significant. Brain perfusion increments in the ACA territory in fetuses with CHD might be a mechanism for protecting important regions in the fetal brain. Such a protective response is also demonstrated by the neurodevelopmental testing in our study, in which higher ACA-VI and ACA-VFI were associated with higher PDI scores. In our study, total intracranial FI was positively correlated with PDI and MDI scores at 12 months of age, indicating that increased brain perfusion might predict better neurodevelopmental outcomes in the CHD population. Williams et al.21 found a correlation between lower MCA-PI Z-scores and higher PDI scores in patients with single-ventricle lesions. However, there was no significant association between MCA-PI Z-scores and PDI or MDI scores in our cohort. This discrepancy could be related to the different subject populations and differences in gestational age between the two studies. This relationship is the opposite to that observed with IUGR. In fetuses with IUGR, increased FI has been associated with higher perinatal mortality rates16 . The differences in pathophysiology between fetal disease states may explain this contradiction. In IUGR, the primary problem is placental insufficiency, which leads to inadequate gas and substrate exchange. However, the primary pathological mechanism of CHD is the hemodynamic consequence of an altered pathway or diminished ventricular outflow, which leads to the delivery of relatively deoxygenated blood to the brain. It is possible that the autoregulatory increase in brain perfusion associated with diminished oxygen delivery is sufficient to meet the cerebral metabolic demands in the early stages of CHD. It is interesting to note that the relationship between 3D power Doppler measurements and development scores appears to vary between CHD subgroups (Table S1). However, the numbers in the subgroups were limited for an in-depth analysis. There are a number of limitations to our study. First, measurements of 3D power Doppler indices based on static volumes are dependent on gain settings and attenuation at acquisition22,23 and there have been contradictory
Ultrasound Obstet Gynecol 2015; 45: 649–656.
Cerebral perfusion in fetuses with CHD
(b)
110
100
110
100 Mental Development Index score
Psychomotor Development Index score
(a)
655
90
80
70
90
80
70
60
60
50
50 20
40
60
80
100
120
20
40
Total-FI
60
80
100
120
Total-FI
Figure 6 Scatterplots with overall linear regression lines ( ) showing relationship between total intracranial flow index (Total-FI) on three-dimensional power Doppler in fetuses with congenital heart disease (CHD) and neurodevelopmental scores at 12 months: (a) Psychomotor Development Index (r = 0.342, P = 0.029) and (b) Mental Development Index (r = 0.339, P = 0.030). Individual data points are plotted according to CHD subtype: hypoplastic left heart syndrome ( ); left-sided obstructive lesions ( ); right-sided obstructive lesions ( ); and transposition of the great arteries ( ).
(b)
110
100
Psychomotor Development Index score
Psychomotor Development Index score
(a)
90
80
70
60
50
110
100
90
80
70
60
50 0.0
2.0
4.0
6.0
8.0
ACA-VI
0.0
1.0
2.0
3.0
4.0
ACA-VFI
Figure 7 Scatterplots with overall linear regression lines ( ) showing relationship between: (a) anterior cerebral artery (ACA) flow indices on three-dimensional power Doppler in fetuses with congenital heart disease (CHD) and Psychomotor Development Index at 12 months: (a) vascularization index (VI) (r = 0.377, P = 0.015) and (b) vascularization flow index (VFI) (r = 0.389, P = 0.012). Individual data points are plotted according to CHD subtype: hypoplastic left heart syndrome ( ); left-sided obstructive lesions ( ); right-sided obstructive lesions ( ); and transposition of the great arteries ( ).
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
Ultrasound Obstet Gynecol 2015; 45: 649–656.
Zeng et al.
656
reports regarding their reproducibility24,25 because of differences in machine settings, ROI, image acquisition and analysis. In this study, therefore, we attempted to keep the acquisition and measurement settings as consistent as possible. Second, additional studies including a greater number of cases are needed to ascertain the differences between CHD subgroups. Neurodevelopmental outcomes are influenced by other variables, such as socioeconomic status, illness severity and treatment strategies. Therefore, our data only show an association between fetal cerebral blood perfusion and neurodevelopment, not causation. In conclusion, cerebral blood flow perfusion was significantly higher in fetuses with CHD than in controls, and this redistribution occurred most prominently in the ACA territory. Higher fetal cerebral 3D power Doppler measurements, namely total intracranial FI, ACA-VI and ACA-VFI, were associated with higher neurodevelopmental test scores in the CHD population. Although further studies are needed to elucidate the predictive value of ultrasound parameters for each type of CHD, the data presented in this report indicate that 3D power Doppler ultrasound may help to identify cases of brain vasodilatation earlier and thus inform parental counseling.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
ACKNOWLEDGMENTS 16.
This study was supported by the National Clinical Key Subject (Medical Imageology) Construction Project of China, the State Natural Sciences Foundation of China (no. 81271593) and the Hunan Province Science & Technology program (no. 2012FJ4142, 2013SK3035).
17. 18. 19.
20.
REFERENCES 1. Limperopoulos C, Tworetzky W, McElhinney DB, Newburger JW, Brown DW, Robertson RL Jr, Guizard N, McGrath E, Geva J, Annese D, Dunbar-Masterson C, Trainor B, Laussen PC, du Plessis AJ. Brain volume and metabolism in fetuses with congenital heart disease: evaluation with quantitative magnetic resonance imaging and spectroscopy. Circulation 2010; 121: 26–33. 2. Clouchoux C, du Plessis AJ, Bouyssi-Kobar M, Tworetzky W, McElhinney DB, Brown DW, Gholipour A, Kudelski D, Warfield SK, McCarter RJ, Robertson RL Jr, Evans AC, Newburger JW, Limperopoulos C. Delayed cortical development in fetuses with complex congenital heart disease. Cereb Cortex 2013; 23: 2932–2943. 3. Khalil A, Suff N, Thilaganathan B, Hurrell A, Cooper D, Carvalho JS. Brain abnormalities and neurodevelopmental delay in congenital heart disease: systematic review and meta-analysis. Ultrasound Obstet Gynecol 2014; 43: 14–24. 4. Zeng S, Zhou Q, Zhou J, Li M, Long C, Peng Q. Impaired brain growth of intracranial structures in fetuses with complex congenital heart disease as measured
21.
22.
23.
24.
25.
by 3D ultrasonography. Ultrasound Obstet Gynecol 2014. DOI: 10.1002/uog.14677. [Epub ahead of print] Donofrio MT, Bremer YA, Schieken RM, Gennings C, Morton LD, Eidem BW, Cetta F, Falkensammer CB, Huhta JC, Kleinman CS. Autoregulation of cerebral blood flow in fetuses with congenital heart disease: the brain sparing effect. Pediatr Cardiol 2003; 24: 436–443. Jouannic JM, Benachi A, Bonnet D, Fermont L, Le Bidois J, Dumez Y, Dommergues M. Middle cerebral artery Doppler in fetuses with transposition of the great arteries. Ultrasound Obstet Gynecol 2002; 20: 122–124. Kaltman JR, Di H, Tian Z, Rychik J. Impact of congenital heart disease on cerebrovascular blood flow dynamics in the fetus. Ultrasound Obstet Gynecol 2005; 25: 32–36. Hernandez-Andrade E, Jansson T, Figueroa-Diesel H, Rangel-Nava H, Acosta-Rojas ´ E. Evaluation of fetal regional cerebral blood perfusion using power R, Gratacos Doppler ultrasound and the estimation of fractional moving blood volume. Ultrasound Obstet Gynecol 2007; 29: 556–561. Benavides-Serralde A, ScheierM, Cruz-Martinez R, Crispi F, Figueras F, Gratacos E, Hernandez-Andrade E. Changes in central and peripheral circulation in intrauterine growth-restricted fetuses at different stages of umbilical artery flow deterioration: new fetal cardiac and brain parameters. Gynecol Obstet Invest 2011; 71: 274–280. Figueroa-Diesel H, Hernandez-Andrade E, Acosta-Rojas R, Cabero L, Gratacos E. Doppler changes in the main fetal brain arteries at different stages of hemodynamic adaptation in severe intrauterine growth restriction. Ultrasound Obstet Gynecol 2007; 30: 297–302. ´ E. Cruz-Martinez R, Figueras F, Hernandez-Andrade E, Puerto B, Gratacos Longitudinal brain perfusion changes in near-term small-for gestational-age fetuses as measured by spectral Doppler indices or by fractional moving blood volume. Am J Obstet Gynecol 2010; 203: 42.e1–e6. Dubiel M, Hammid A, Breborowicz A, Pietryga M, Sladkevicius P, Olofsson PA, Breborowicz GH, Gudmundsson S. Flow index evaluation of 3-D volume flow images: an in vivo and in vitro study. Ultrasound Med Biol 2006; 32: 665–671. ´ Milani HJ, de Sa´ Barreto EQ, Araujo Junior E, Haratz KK, Rolo LC, Nardozza LM, Moron AF. Assessment of cerebral circulation in normal fetuses by three-dimensional power Doppler ultrasonography. Eur J Radiol 2012; 81: 514–521. Bartha JL, Moya EM, Herv´ıas-Vivancos B. Three-dimensional power Doppler analysis of cerebral circulation in normal and growth-restricted fetuses. J Cereb Blood Flow Metab 2009; 29: 1609–1618. Hadlock FP, Harrist RB, Sharman RS, Deter RL, Park SK. Estimation of fetal weight with the use of head, body, and femur measurements-a prospective study. Am J Obstet Gynecol 1985; 151: 333–337. Arduini D, Rizzo G. Normal values of pulsatility index from fetal vessels: a cross-sectional study on 1556 healthy fetuses. J Perinat Med 1990; 18: 165–172. Bayley N. Bayley Scales of Infant Development (2nd edn). The Psychological Corporation: San Antonio, TX, 1993. Donofrio MT. The heart–brain interaction in the fetus: Cerebrovascular blood flow in the developing human. Progr Pediatr Cardiol 2006; 22: 41–51. Lockhart RD, Hamilton GF, Fyfe FW. Arteries of the central nervous system. In Anatomy of the Human Body. Faber and Faber Limited: London, UK, 1965: 607–609. Dubiel M, Gunnarsson GO, Gudmundsson S. Blood redistribution in the fetal brain during chronic hypoxia. Ultrasound Obstet Gynecol 2002; 20: 117–121. Williams IA, Fifer C, Jaeggi E, Levine JC, Michelfelder EC, Szwast AL. The association of fetal cerebrovascular resistance with early neurodevelopment in single ventricle congenital heart disease. Am Heart J 2013; 165: 544–550.e1. Miyague AH, Raine-Fenning NJ, Pavan TZ, Polanski LT, Baumgarten MN, Nastri CO, Martins WP. Influence of gain adjustment on 3-dimensional power Doppler indices and on spatiotemporal image correlation volumetric pulsatility indices using a flow phantom. J Ultrasound Med 2013; 32: 1831–1836. Miyague AH, Pavan TZ, Grillo FW, Teixeira DM, Nastri CO, Martins WP. Influence of attenuation on three-dimensional power Doppler indices and STIC volumetric pulsatility index: a flow phantom experiment. Ultrasound Obstet Gynecol 2014; 43: 103–105. Lai PK, Wang YA, Welsh AW. Reproducibility of regional placental vascularity/perfusion measurement using 3D power Doppler. Ultrasound Obstet Gynecol 2010; 36: 202–209. ˜ Filho HA, Mattar R, Araujo Junior ´ Guimaraes E, da Costa LL, de Mello Junior CF, Nardozza LM, Nowak PM, Moron AF. Reproducibility of three-dimensional power Doppler placental vascular indices in pregnancies between 26 and 35 weeks. Arch Gynecol Obstet 2011; 283: 213–217.
SUPPORTING INFORMATION ON THE INTERNET The following supporting information may be found in the online version of this article: Table S1 Linear correlation coefficients for the relationship between three-dimensional power Doppler measurements and neurodevelopment scores at 12 months according to subtype of congenital heart disease
Copyright © 2015 ISUOG. Published by John Wiley & Sons Ltd.
Ultrasound Obstet Gynecol 2015; 45: 649–656.
