Ribera et al. BMC Pediatrics (2019) 19:326 https://doi.org/10.1186/s12887-019-1689-y STUDY PROTOCOL Open Access Multicenter prospective clinical study to evaluate children short-term neurodevelopmental outcome in congenital heart disease (children NEURO- HEART): study protocol I. Ribera1, A. Ruiz1, O. Sánchez1,2, E. Eixarch3, E. Antolín2,4, E. Gómez-Montes2,5, M. Pérez-Cruz2,3, M. Cruz-Lemini1, M. Sanz-Cortés3, S. Arévalo1, Q. Ferrer6, E. Vázquez7, L. Vega6, P. Dolader6, A. Montoliu9, H. Boix8, R. V. Simões3, N. Masoller3, J. Sánchez-de-Toledo2,10, M. Comas11, J. M. Bartha2,4, A. Galindo2,5, J.M. Martínez3, L. Gómez-Roig2,3, F. Crispi3, O. Gómez3, E. Carreras1, L. Cabero1, E. Gratacós3 and E. Llurba2,12* Abstract Background: Congenital heart disease (CHD) is the most prevalent congenital malformation affecting 1 in 100 newborns. While advances in early diagnosis and postnatal management have increased survival in CHD children, worrying long-term outcomes, particularly neurodevelopmental disability, have emerged as a key prognostic factor in the counseling of these pregnancies. Methods: Eligible participants are women presenting at 20 to < 37 weeks of gestation carrying a fetus with CHD. Maternal/neonatal recordings are performed at regular intervals, from the fetal period to 24 months of age, and include: placental and fetal hemodynamics, fetal brain magnetic resonance imaging (MRI), functional echocardiography, cerebral oxymetry, electroencephalography and serum neurological and cardiac biomarkers. Neurodevelopmental assessment is planned at 12 months of age using the ages and stages questionnaire (ASQ) and at 24 months of age with the Bayley-III test. Target recruitment is at least 150 cases classified in three groups according to three main severe CHD groups: transposition of great arteries (TGA), Tetralogy of Fallot (TOF) and Left Ventricular Outflow Tract Obstruction (LVOTO). Discussion: The results of NEURO-HEART study will provide the most comprehensive knowledge until date of children’s neurologic prognosis in CHD and will have the potential for developing future clinical decisive tools and improving preventive strategies in CHD. Trial registration: NCT02996630, on 4th December 2016 (retrospectively registered). Keywords: Congenital heart disease, Neurodevelopment, Predictive markers, Cardiac function and fetal brain MR * Correspondence: ellurba@santpau.cat 2 Spain Maternal and Child Health Development Network, RETICS funded by the PN I+D+I 2013-2016 (Spain), ISCIII- Sub-Directorate General for Research Assessment and Promotion and the European Regional Development Fund (ERDF), ref. RD16/0022, Madrid, Spain 12 Director of Obstetrics and Gynaecology Department, St Creu and St Pau Hospital, Sant Antoni Mª Claret, 167, 08025 Barcelona, Spain Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Ribera et al. BMC Pediatrics (2019) 19:326 Page 2 of 10 Background until date of children’s neurologic prognosis in CHD Among congenital anomalies, congenital heart disease and will have the potential for developing future clinical (CHD) is the leading cause of childhood mortality and decisive tools and improving preventive strategies in morbidity, affecting up to 1% of all live births [1]. More CHD to be used by cardiologists and obstetricians for than 36,000 infants with CHD are born in Europe every parental assessment. year and approximately 3000 more die in utero either spontaneously through legal abortion or during neonatal Objectives life [2]. Prognosis in these patients has significantly Aims improved in recent years, with survival up to 95%, The NEURO-HEART study aims to: 1) Describe the thanks to advances in diagnostic techniques and surgical neuro-developmental outcome of patients with complex management [3, 4]. CHD at 24 months of age and identify a subgroup with Up to 50% of children with CHD present deficits in at poorer outcome. 2) Evaluate the utility of fetal and postna- least one area of neurodevelopment (learning ability, motor tal (preoperative and postoperative) diagnostic techniques skills, language, etc.) [5, 6]. These findings seem to be more for early recognition of patients at risk for suboptimal pronounced in CHD types associated with reduced oxygen- neurologic outcome. ated blood brain delivery, such as transposition of great arteries and left outflow tract obstruction; however, poor Methods neurodevelopment has also been reported in children with Participants/eligibility criteria heart defects with normal cerebral oxygen delivery [6]. The participation of different referral centers will permit: Most studies evaluating neurological abnormalities in 1. sufficient recruitment of fetuses to be distributed into children with CHD have focused on factors associated 3 groups depending on their CHD and prospective fol- with surgical repair [7], as abnormal neurodevelopment low-up and integration of data from the prenatal period was long believed to occur as a result of the procedures to early childhood. associated with open surgery, mainly extra-corporeal The study protocol was approved by each of the Ethic circulation. However, signs of brain injury on magnetic Committee’s Centers, and written consent was obtained resonance imaging and ultrasound have also been shown from all women to participate in the study. in prenatal studies [8, 9]. Additional studies report that CHD fetuses have smaller head biometries and signs of brain sparing from the second trimester of pregnancy, Settings/locations regardless of the type of CHD, supporting the recent hypothesis of early onset appearance of noxa mecha-  Hospital Universitari Vall d’Hebrón de Barcelona, nisms that could lead to a poorer neurodevelopment Spain later in life in these children [10, 11].  Hospital Universitario La Paz de Madrid, Spain Moreover, pregnancies affected with a CHD fetus have  Hospital 12 de Octubre de Madrid, Spain a higher rate of placental-related complications such as  Hospital Maternitat-Clínic de Barcelona, Spain preeclampsia and preterm birth [12], which increases the  Hospital Sant Joan de Déu de Barcelona, Spain chance of a poorer prognosis after birth and also after surgery [13]. Some authors have described an existing Recruitment and collection of data difference in plasma biomarkers in CHD, with higher Target recruitment is 150 cases classified in three groups ratio of anti-angiogenic factors compared to pro- according to the specific type of CHD. After reviewing angiogenesis factors and also increased blood levels of existent literature and a variety of cohort classifications brain and heart hypoxemic biomarkers [14], leading [15–17] and being aware of the capacity of recruitment to a possible related pathway between placental com- due to our multicentric project, we decided to analyze plications and CHD, without being clearly defined. patients with very similar diagnosis as it appeared Despite the importance and frequency of these compli- inaccurate to create subgroups including different CHD cations, the prenatal and postnatal risk factors impacting types and will also make future parental assessment on neurodevelopment remain poorly understood. Stud- easier. Groups were made with our three most common ies from a clinical viewpoint have been based on small diagnosis: TGV, TOF and LVOTO. These groups are series. Consequently, research into whether differences known for having different patterns of blood supply to exist in the spectrum of neurological damage according the brain, being TGA and hypoplasic left heart to type of cardiac defect and what kind of neurodevelop- syndrome (HLHS) the ones with a higher oxygen impair- mental deficits these children present is needed. ment [15], so we expect to obtain differences depending Our hypothesis is that the results of NEURO-HEART on the CHD type. Cases are summarized in Table 1 in study will provide the most comprehensive knowledge Appendix. Ribera et al. BMC Pediatrics (2019) 19:326 Page 3 of 10  Group 1: Transposition of great arteries. At delivery, a cord blood and a maternal blood sample  Group 2: Tetrallogy of Fallot. will be taken.  Group 3: Left outflow tract obstruction. Mode of delivery, perinatal complications, biometric data such birth weight and head circumference, need for A control group of 50 low-risk pregnancies will also intubation or need for drugs after birth will also be re- be recruited. These patients will be followed up once a corded. A FCUS will be performed during the ICU stay. month with fetal ultrasound at each visit; functional Blood samples will be taken and continuous EEG for 2 h echocardiography and MRI, performed at the same times will be performed before and after surgery. Brain oxim- as in the case group. etry will be measured from 12 h before to 48 h after See Table 1 in Appendix. Classification according to surgery. CHD type. Children’s follow-up will take place at 12 and 24 months of corrected age. On both appointments, elec- Exclusion criteria troencephalogram (EEG) and functional echocardiog- Pregnancies with gestational age < 20 weeks or > 38 weeks, raphy will be performed and a blood sample will be associated non-cardiac malformations, presence of obtained. At 2 years of age, a MRI will be performed. chromosomal abnormalities, associated arrhythmia and Neurodevelopment follow-up will be performed at 12 maternal conditions that might affect fetal hemodynamics months of age using the Ages & Stages Questionnaires such as diabetes, thyroid disease or preeclampsia, multiple (ASQ) and at 24 months of age using Bayley-III Scale, pregnancy and fetal anemia. Minimum age of inclusion applied by a previously trained neuropsychologist. These for the study will be 18 years old. scores are widely used to determine the child’s perform- ance compared with norms taken from typically devel- Database oping children of the same age (in months). The local research coordinator and/or the staff from Study procedures will be performed and organized as participating hospitals will identify eligible women and shown in Fig. 1. after counseling and reading the information form, pa- tients will be asked for written consent. Ultrasound evaluation At study recruitment, demographic, obstetric and Gestational age at scan will be calculated based on the medical history data will be recorded into a web-based crown-rump length obtained at first trimester screening Case Report Form (CRF) that will be accessible through [18]. A complete ultrasound examination will be per- a restricted website. Details on delivery, maternal and formed in each fetus using high quality ultrasound neonatal assessments during pregnancy or post-partum systems. data will be recorded in the CRF. The collected data will Fetal biometric parameters assessed will include: bipar- be coded and processed with adequate precaution to en- ietal diameter (BPD), head circumference (HC), abdom- sure patient confidentiality with the following measures: inal circumference (AC) and femoral length (FL). Initials of participants as well as a local patient number Estimated fetal weight will be calculated according to will be recorded in the electronic database. Linking the method of Hadlock et al. [19]; both estimated fetal names with patients’ numbers will only be available in weight and birth weight centile will be obtained using the local clinics. Each participating clinic will receive a local reference curves [20]. Doppler recordings will be login name and password to gain access to the web-se- made in the absence of fetal movements with voluntarily cured database. Database access will be restricted to cli- suspended maternal breathing. Measurements of pulsed nicians with electronic password. Full access to the Doppler parameters will be taken automatically from entire database will also be restricted to some members three or more consecutive waveforms, with an angle of of the research staff. isonation as close to 0° as possible. Umbilical artery (UA) will be evaluated in a free loop of the umbilical Procedures and interventions cord; middle cerebral artery (MCA) will be measured in Data collection a transverse view of the fetal skull at the level of its ori- Patients will be followed up at the obstetrics and fetal gin from the circle of Willis [21]. Cerebroplacental ratio cardiology unit every 4 weeks with a multidisciplinary (CPR) will be calculated by dividing MCA by UA pulsa- approach. At inclusion, a blood sample from both par- tility index (PI) [22]. Uterine artery (UtA) will be evalu- ents will be taken. During pregnancy, visits will be ated with the probe placed on the lower quadrant of the scheduled every 4 weeks, including fetal biometries and abdomen, angled medially, with identification by color Doppler. At 28–32 weeks of pregnancy, a functional Doppler imaging of the apparent crossover with the ex- echocardiography will be performed and finally, at 35– ternal iliac artery. Mean uterine artery PI will be calcu- 37 weeks, a MRI for evaluating fetal brain will be done. lated as the average PI of right and left arteries [23]. Ribera et al. BMC Pediatrics (2019) 19:326 Page 4 of 10 Fig. 1 Study procedures flowchart Fetal echocardiographic examination will be performed and cardiac index (CI)) will also be measured. With a 3- according to the International Society of Ultrasound in vessel and trachea view in 2D, the aortic isthmus and Obstetrics and Gynecology guideline [24]. Briefly, cardiac ductus arteriosus diameters will be measured. In the axis and situs, pericardial effusions, ventricular morph- same plane with pulsed Doppler, aortic isthmus PI, sys- ology, veno-atrial, atrioventricular and ventricule-arterial tolic and diastolic VTI (calculation of isthmus flow index connections, size and relationships of left and right (IFI)) and PI index will be recorded. ventricular outflow tracts, ductal and aortic arches, atrial Myocardial peak systolic (S′) and diastolic (E’, A’) and interventricular septum and flow through atrioven- velocities and myocardial performance index (MPI’) in tricular and semilunar valves will be evaluated using two-di- mitral, septal and tricuspid annuli will be also obtained mensional, color and pulsed Doppler ultrasound. Prenatal by tissue Doppler imaging with described previous diagnosis of CHD will be confirmed using postnatal echo- methodology [26]. cardiography or autopsy in cases of neonatal death. Functional echocardiographic scan will include several MRI evaluation measures [25]: from a transverse 4-chamber view in M- Pregnant women will undergo fetal MRI on 1.5 Tesla mode: end-diastolic diameter (EDD), end-systolic diam- scanners at 35 to 37 weeks of gestational age following eter (ESD), septum, myocardial wall thickness in systole the American College of Radiology guidelines for the use and diastole (calculation of shortening (SF) and ejection of medical imaging during pregnancy and lactation [27]. fractions (EF)). From an apical/basal 4-chamber view in T2-weighted images (HASTE) will be acquired in 2D: cardiothoracic ratio, atrial areas, AV valve diameters, sagittal, axial and coronal sections including all the brain ventricular longitudinal and transverse diameters (calcu- at 3.5 mm of thickness, as reported previously [28]. MR lation of sphericity indices). With an apical/basal 4- spectroscopy (MRS) images will also be acquired in the chamber view, using pulsed Doppler, we will measure frontal lobe. Anatomical 2D images in sagittal, axial and early (E) and late (A) peaks of transvalvular filling veloci- coronal sections will be acquired for volumetric recon- ties of left and right ventricles (E/A ratio calculation) struction. An anatomical control sequence (HASTE) will and the presence of mitral/tricuspid insufficiency and its be taken to verify the position of the fetal head has not velocity. From an apical/basal 4-chamber view in M- significantly changed, in which case the MRS images will mode: mitral and tricuspid annular-plane systolic excur- have to be repeated. Diffusion and MRS data will sion (MAPSE, TAPSE) will be measured. With an additionally be acquired if possible. If the quality of the apical/basal 5-chamber view in 2D, aortic valve diameter images is suboptimal, sequences will be repeated. in systole will be measured. Finally, with an apical/basal Structural MRI images will be reviewed for the 5-chamber view adding pulsed Doppler, we will measure presence of anatomic abnormalities by an experienced aortic peak systolic velocity, measurement of the velocity neuroradiologist. Parameters evaluated will be: brain time integral (VTI), left fetal heart rate (calculation of calcifications if present, hypoxic or ischemic damage. left cardiac output (CO)), left isovolumetric contraction Brain biometric measurements will be performed using (ICT) and relaxation (IRT) times and ejection time (ET) the semiautomatic Analyze 9.0 software (Biomedical (calculation of myocardial performance index (MPI)). Imaging Resource; Mayo Clinic, Kansas City, KS) by an Pulmonary valve diameter in systole, pulmonary artery experienced examiner blinded to diagnosis. Cortical peak, systolic velocity, VTI and right fetal heart rate (cal- fissure delineations in the fetal MRI will be performed culation of right CO, combined cardiac output (CCO) adapting previously described methodology to assess Ribera et al. BMC Pediatrics (2019) 19:326 Page 5 of 10 cortical development in fetal ultrasound [29, 30] and will different time points: 1. after birth (cord blood) 2. during include the following measures, performed as they have the pre-surgery period and 3. immediately after surgery. previously been described for our group [31]: BPD, All blood samples will be processed within 1 h. Plasma fronto-occipital diameter (FOD), occipital fossa, cerebel- will be separated by centrifuge at 1400 g for 10 min and lum diameter and lateral ventricular size, Sylvian fissure stored at 4 °C. depth, parieto-occipital suture depth and insula measure- Angiogenic Placental Growth Factor (PlGF) and ments. In a middsagital plane, corpus callosum longitu- antiangiogenic soluble fms-like tyrosine kinase-1 and dinal measurement and vermix diameter biometries will soluble endothelin (sFlt1, sEng) factors will be mea- be performed. Calcarine and Cingular fissures depth will sured in pregnant patients from both case and control be measured in coronal plane. All biometric parameters groups. In cord blood, the following biochemical will be corrected by BPD and therefore expressed as ratios. markers will be measured: a. markers of cardiac func- Cortical fissure depths will be measured bilaterally. tion and injury: Troponin I, N-Terminal pro-brain MRS data will be acquired from the semi oval centre, natriuretic peptide (NT-proBNP), endothelin-1 and with single voxel PRESS localization and will be proc- heart fatty acid-binding protein (h-FABP); b. neuro- essed in 2 ways using MATLAB R2010a (MathWorks logical damage markers S-100 proteins and neuron- Inc., Natick, MA): method A, standard averaging of 128 especific enolase (ENE) and c. angiogenic PlGF and transients (first free-induction-decay block discarded); antiangiogenic sFlt1 factors. and method B, spectral sorting (discarding data with All plasma biomarkers will be determined by ELISA gross artifacts and lipid contaminations) and alignment commercial kits following the manufacturer’s instruc- (based on the residual water peak), before averaging. tions. Angiogenic factors (PlGF, sFlt-1 and sEng) are Spectra will then be analyzed for quality assessment and obtained from R& D Systems (R&D Systems selection: both with jMRUI v4.0 (MRUI Consortium, Europe Ltd., Abington, UK). Markers of cardiac func- EU), spectral pattern well resolved for major peaks, such tion and injury are obtained from different manufac- as choline compounds, total creatine, and N-acetylaspar- turers: Troponin I from AbFRONTIER (Young In tate, and amplitude ratio of lipid (1.3–0.9 ppm) to cho- Frontier Co. Ltd. Seoul, Korea), NT-proBNP from BG line peak (3.21 ppm) < 10%; and with LCModel v6.3 (BlueGene biotech CO. Ltd., Shanghai, China), (Stephen Provencher Inc., Oakville, Canada), peak full- Endothelin-1 from R&D Systems (R&D Sys- width at half-maximum < 0.1 ppm and signal-to-noise tems Europe Ltd, Abington, UK) and h-FABP from ratio [32]. Hycult Biotech (Hycult Biotech, Uden, The Postnatal MRI will be performed with a Siemens Netherlands). Markers neurological damage (S100B Magnetom Trio 3.0 Tesla (Erlangen, Germany). T1 and NSE) are obtained from BG (BlueGene biotech sequences will be taken for anatomical data. Qualita- CO. Ltd., Shanghai, China). tive analysis on T2 and FLAIR frequencies will be made with FSL software (functional magnetic reson- ance imaging of the brain’s Software Library) [33]. Neonatal evaluation and neurodevelopment MPRAGE imaging will be used for grey matter quan- During hospital stage, data on intensive care unit and titative measurements and DTI images for white hospital length stay, need for mechanic ventilation, matter measurements. need for vasoactive drugs and mortality will be The presence of myelinisation in conventional sequences, recorded. brain cortical development, ischemic or hemorrhagic Surgical duration, time on extracorporeal support, lesions and total grey /white matter volume will be cardiac ischemia time and the need and duration of cir- recorded. In addition, metabolic analysis will be made culatory arrest will be also recorded. Need for renal and compared to data published in previous studies replacement therapies, extracorporeal life support and [34]. All volumetric estimations will be obtained using cardiac arrest before or after surgery will be also Cavalieri’s principle 27 by a multiplanar analysis con- recorded. sidering a slice thickness of 3.5 mm with no gap Monitoring for oximetry and EEG will be recorded be- interval between them. For brain maturation evalu- fore and after surgery. Neurosonography will be used to ation, specific MRI visual scores will be used [35]. evaluate brain anatomy and findings such as: ventriculo- megaly (VMG) and / or increased subarachnoid space; Plasma biomarkers analysis signs of periventricular leukomalacia (PVL) signs of in- At the time of inclusion, a maternal venous blood sam- traventricular hemorrhage (IVH); presence of calcifica- ple will be drawn. Additionally, a second blood sample tions and alterations in cortical development. will be taken from the mother at the time of delivery. A baseline EEG of 12 channels will be performed Blood samples from children with CHD will be drawn at preoperatively and within the first 48 h postoperatively. Ribera et al. BMC Pediatrics (2019) 19:326 Page 6 of 10 Presence and number of power crisis and overall dur- taps understanding and expression of language and 3. ation of delta waves will be recorded. the Motor Scale, which assesses gross and fine motor Following previous literature by Mebius et al. [36], skills. The scaled score ranges from 1 to 19 points, with regional cerebral oximetry using technology Near-in- a mean of 10 points in the validated population (USA), frared Spectroscopy (NIRSS) will be recorded continu- and a SD of 3 points. Some studies question the capacity ously. NIRSS has been described as a useful predictor of the Bayley Scale, especially the 3rd edition, for detect- for cardiac arrest in patients with complex CHD pre- ing impaired neurodevelopment due to a tendency to vious to surgery and a reliable way to measure brain undervalue alterations [37]. For this reason, our study oxygen saturation. Bilateral forehead sensors will be defined impaired cognitive neurodevelopment as a score placed preoperatively (2 h prior to surgery) and will of 8 or less on the scale (25th percentile, equivalent to be kept for at least 24 h post-operatively. Global per- the mean minus 0.66 SD). centage of time below a threshold of 45 and percent- FCUS and MRI will also be performed at 2 years of age of time below 20% of the baseline measurement age. will be recorded. At the same period of time, low-risk pregnant women Children will be followed up for 2 years. Cognitive attending a routinely 20-weeks scan at Vall d’Hebron neurodevelopment will be assessed at 1 year of age Obstetrics Department would be asked for participation using the ASQ. This test is validated and extensively as part of the control group of the study. In this group used as a screening method for detecting impaired of women, a blood sample at the time of the inclusion neurodevelopment in high-risk children. Research and during delivery, as well as cord blood sample, will nurses will be notified when a patient will approach be obtained. During pregnancy, visits will be scheduled the corrected age of 1 year by an email generated every 4 weeks and will include fetal biometries and from our database. After this alert, children’s parents Doppler measurements. At 28–32 weeks of pregnancy, a will be called to complete again the questionnaire. In functional echocardiography will be performed and case the parents do not return the questionnaire, a finally, at 35–37 weeks, a fetal brain MRI will be done. reminder will be sent. We will be using the validated No further maternal or neonatal tests will be performed. Spanish translation of the ASQ, covering the age- range 4–60 months. Outcome measures In the ASQ 3rd edition surveys, parents are asked to Primary outcome mesures answer “YES” (10 points), “SOMETIMES” (5 points) or “NOT YET” (0 points) to a series of items of the follow- ing categories of global neurodevelopment: communica- A. Neurodevelopmental outcome according to CHD tion, large movements, fine movements, problem solving types. and socio-individual competence. A total score is ob- tained from the sum of the items of each category. There is a different baseline for every age and category under 1. Abnormal ASQ: ASQ score results will be which an alteration in this part of the neurodevelopment compared to the normal Bell curve results for the is considered to exist. In our study, results were analyzed same age. Those close to the cut-off of 2 SD will be as: a) impairment in each category; or b) impairment of considered at risk for neurodevelopment at least one category. Children scoring under 2 standard impairment. deviations (SD) will be referred to diagnostic assessment. 2. Bayley-III test will be considered at risk for At 2 years of age, Bayley-III Scale will be performed by neurodevelopment delay when scored 8 or less on a previously trained neuropsychologist. This measure the scale (25th percentile, equivalent to the mean consists on different developmental play tasks adminis- minus 0.66 SD) tered between 45 to 60 min and derives in a develop- mental quotient (DQ) rather than an intelligence Secondary outcome mesures quotient. Raw scores of successfully completed items are converted to scale scores and to composite scores. These 1. Hemodynamic changes and Doppler fetal cardiac scores are used to determine the child’s performance function in fetuses with CHD by type. compared with norms taken from typically developing 2. Cortical brain development assesed by MRI such as children of their age (in months). The most recent changes in fissure depths, head biometries or brain edition, the Bayley-III, has three main subtests: 1. the metabolism spectroscopic parameters. Cognitive Scale, which includes items such as attention 3. Fetal head biometries and perfusional parameters. to familiar and unfamiliar objects, looking for a fallen 4. Biochemical markers in cord blood associated with object, and pretend play, 2. the Language Scale, which neurological damage in children with CHD. Ribera et al. BMC Pediatrics (2019) 19:326 Page 7 of 10 5. Parameters of perioperative brain. analysis will be based on single variable descriptive 6. Severe neonatal or child morbidity will be taken models. In case of finding associations with into consideration as primary outcome in our confusion factors, multivariable models will be trial. The presence of one or several of the considered. following complications will define severe All statistical comparative analysis will be done using morbidity: severe respiratory distress, SPSS 2.13 (SPSS for Windows, SPSS Inc., Chicago, Ill., intraventricular hemorrhage grades III-IV, treated USA) and MatLab statistical programs (2007b, The ductus arteriosus persistence, renal dysfunction, MathWorks Inc., Natick, USA). In hypothesis testing for necrotizing enterocolitis, intestinal perforation, population inference, statistical significance of 0.05 will early and late sepsis, retinopathy of prematurity be postulated. treated with laser, bronchopulmonary dysplasia, For the analysis of blood sample biomarkers all param- periventricular leucomalacia, postnatal eters will be transformed to MoMs as well as ultrasound administration of corticosteroids or inotropic biometric and Doppler parameters will be converted to drugs and/or death. z-scores using previously published data [38–40] in 7. Statistical association of all the parameters order to compare data obtained in different gestational described in 1–5 with neurological development. age. MRI measurements will be corrected by BPD and 8. Regression statistical measures to select the most birth weight centile. valuable variables related to neurological For multiple comparsions between groups one- impairment in order to identify risk factors for way ANOVA tests will be performed. In case of brain damage in a preoperative stage. non-parametric data, Mann-Whitney tests will be used. All these measures will be analyzed in an exploratory Also multivariate analysis using logistic regression manner. Analysis will be performed to investigate the will be made with the support of the Unidad de utility of pre- and post-natal markers in the short-term Soporte en Metodología para la Investigación Biomé- prediction of adverse neurological outcomes in order to dica (USMIB) using the application STATA v.11.Dif- create prognostic tools for early detection of patients at ferent analyses will be carried out in order to risk. establish the association and prediction power of the characteristics analyzed in all tests independently and Statistical analysis also combined by implementing classical statistic Sample size methods (linear and logistic regression) and inference All participating centers together have over 25,000 deliv- methods with and without a priori assumptions eries / year and more than 150 births of congenital heart (Bayesian networks, decision trees, principal compo- defects babies We have planned a 3 year-period recruit- nent analysis among others). After that, the results ment that, considering our previous birth statistics, cases will be implemented in a learning machine for of interruption of pregnancy and an estimated 10% loss algorithms in MATLAB (2007b, The MathWorks Inc., of follow up, will allow us to recruit 60 patients diag- Natick, USA). The type of analysis will depend on nosed with TGA, 30 patients with TOF and 30 patients each specific objective. for the LVOTO group, the largest series published so far. Limitations of the study Data analysis Establishing specific associations to a single CHD type Storage of variables will be done in a database spe- may require a larger sample size than we will be able cially designed for our study by the Department of to recruit, especially in the Fallot and LVOTO Bioinfomatics and Clinical Pharmacology (HUVH). groups. Although our time to include patients may Once the sample collection and conducting of labora- not be enough to obtain the necessary number of tory determinations will be finished, we will proceed cases, we believe they will still be remarkable series to the evaluation and analysis of the results obtained and a unique opportunity to create a CHD population with help and advice by Bioinformatics Research In- cohort. stitute Unit and the Support for Investigation Unit We are aware that the LVOTO group is intrinsically (USMI) from Vall d’Hebron Hospital. heterogenical with different prognosis and probably also different brain development impairment between a coar- Statistical analysis tation of aorta and a hypoplasic heart syndrome case, Data will be collected, digitalized and stored in a but this pathophysiological classification allows a reason- data base designed for the study. The statistical able approach. Ribera et al. BMC Pediatrics (2019) 19:326 Page 8 of 10 Inclusion of healthy controls with the same follow up Appendix has its logical limitations, and due to ethical considerations, Table 1 Classification according to CHD type no control children would be used for biochemical Group 1: Transposition of great arteries markers. • With or without VSD. Group 2: Tetralogy of Fallot Discussion Group 3: LVOTO Congenital heart disease is one of the leading causes of • Coartation of aorta congenital malformation. While advances in early diag- • Aortic arch hypoplasia nosis and postnatal management have increased survival in CHD children [3], worrying long-term outcomes, par- • Double-inlet left ventricle with coartation/aortic arch hypoplasia ticularly neurodevelopmental disability, have emerged as • Double outlet right ventricle with coartation/aortic arch hypoplasia a key prognostic factor in the counseling of these preg- nancies [41, 42]. IQ: Intelligence Quotient; IRT: Isovolumetric relaxation time; A recent study showed that a high proportion of IVH: Intraventricular hemorrhage; LVOTO: Left Ventricular Outflow Tract fetuses with CHD already have a smaller head and Obstruction; m: Months; MCA: Middle cerebral artery; MPI: Myocardial increased brain perfusion in the second trimester of performance index; MRI: Magnetic resonance imaging; NT-proBNP: N- Terminal pro-brain natriuretic peptide; PI: Pulsatility index; PlGF: Angiogenic pregnancy [15]. Additionally, a substantial percentage placental growth factor; PVL: Periventricular leukomalacia; SD: Standard of newborns with CHD have signs of brain injury on deviations; sEng: Soluble endothelin; SF: Shortening fraction; magnetic resonance imaging and reduced cranial size sFlt: Antiangiogenic soluble fms-like tyrosine kinase-1; TGA: Transposition of the great arteries; TOF: Tetralogy of Fallot; UA: Umbilical artery; UtA: Uterine [43], suggesting an early onset of the mechanisms artery; VMG: Ventriculomegaly; VSD: Interventricular communication; leading to poorer neurodevelopment later in life VTI: Velocity time integral; W: Weeks [44]. It has been hypothesized that altered cerebral Acknowledgements perfusion is one of the main contributors to abnor- Special thanks to Ms. Christine O’Hara, who helped in the correction of mal neurodevelopment in fetuses with CHD. Al- English version of this protocol. though abnormal head size was more pronounced in Authors’ contributions fetuses with compromised blood delivery to the EL, AG, LC, JB, LG, EG and JST planned the study. EL, AR, IR, QF and JST will brain, it was also present in milder forms of CHD, be responsible of creating database, analysis and interpretation of data and suggesting that there could be additional mecha- spreading results. EL, QF, SA, AR, IR, EC, EM, MC, EA, MP and EE will be in charge for patient recruitment. EL, AR, IR, SA, QF, EM, EA, EE, JM, NM, FC and nisms that contribute to abnormal neurodevelopment MP will do the maternal clinical follow-up, ultrasound scan and postpartum in CHD cases [12, 45–47]. sample collection. QF, LV, PD, OG, AM, HB and JST will do the neonatal fol- No prospective studies have been performed to allow low up, oximetry monitoring, pediatric postnatal tests and echocardiography and blood sample compilation. JST, VL, JC, QF, HB, PD and LV will be in clinicians to do a good assessment on neurological prog- charge for surgical monitoring. EL and OS will be analyzing blood samples. nosis with compelling evidence until now. Differences in RS, EE, MSC, MCL and EV are in charge for taking magnetic resonance im- brain development markers have not been reported in aging. IR and MP will analyze MR images. All authors read and approved the final manuscript. previous series, but we believe those results might be re- lated to sample size [48]. Funding The Neuro-Heart trial aims to compare and de- RETICS funded by the PN 2018-2021 (Spain), ISCIII- Sub-Directorate General for Research Assessment and Promotion and the European Regional Devel- scribe preoperative markers on CHD affected fetuses opment Fund (FEDER), reference RD16/0022. Funding body had no role in both prenatal as well as postnatal brain functional the design, collection or analysis of the study, neither on writing the monitoring during the perioperative period and manuscript. through cardiac surgery. At the same time, this multi- Availability of data and materials centric trial will allow CHD patients included to be The datasets generated during and/or analyzed during the current study are classified into specific CHD groups in the largest not publicly available due to patient confidentiality, but are available from the corresponding author on reasonable request. series published so far. Ethics approval and consent to participate Abbreviations Ethics approval to conduct the study was obtained from every participant AC: Abdominal circumference; ASQ: Ages and stages questionnaire; hospital ethics committee. Vall d’Hebrón Hospital was named coordinating AVSD: Atrioventricular septal defect; BPD: Biparietal diameter; centre for the study. Ethic approval in this centre was obtained from the CCO: Combined cardiac output; CHD: Congenital Heart Disease; CI: Cardiac Investigation Ethics Committee on February 13th, 2014 (PR (AMI) 317/2012) index; CO: Cardiac output; CPR: Cerebroplacental ratio; DQ: Developmental by. All patients included gave written informed consent. quotient; EDD: End-diastolic diameter; EEG: Electroencephalogram; EF: Ejection fraction; ENE: Neuron-especific enolase; ESD: End-systolic Consent for publication diameter; ET: Ejection time; FCUS: Functional cardiac ultrasonography; Not applicable. FL: Femoral length; FOD: Fronto-occipital diameter; HC: Head circumference; h-FABP: Heart fatty acid-binding protein; HLHS: Hypoplasic Left Heart Competing interests Syndrome; ICT: Isovolumetric contraction time; IFI: Isthmus flow index; The authors declare that they have no competing interests. Ribera et al. BMC Pediatrics (2019) 19:326 Page 9 of 10 Author details 14. Llurba E, Sanchez O, Ferrer Q, Nicolaides KH. Ruiz a et al maternal and 1 Department of Obstetrics, Vall d’Hebron University Hospital, Universitat foetal angiogenic imbalance in congenital heart defects. European Autònoma De Barcelona, Barcelona, Spain. 2Spain Maternal and Child Health Heart J. 2014;35(11):701–7. Development Network, RETICS funded by the PN I+D+I 2013-2016 (Spain), 15. Masoller N, Martínez JM, Gómez O, Bennasar M, Crispi F, Sanz-Cortés M, ISCIII- Sub-Directorate General for Research Assessment and Promotion and Egaña-Ugrinovic G, Bartrons J, Puerto B, Gratacós E. Evidence of second the European Regional Development Fund (ERDF), ref. RD16/0022, Madrid, -trimester changes in head biometry and brain perfusion in fetuses with Spain. 3BCNatal, Hospital Clínic of Barcelona and Hospital Sant Joan de Déu, congenital heart disease. Ultrasound Obstet Gynecol. 2014;44:182–7. Barcelona, Spain. 4Division of Maternal and Fetal Medicine, Department of 16. Ruiz A, Cruz-Lemini M, Masoller N, Sanz-Cortés M, Ferrer Q, Ribera I, Obstetrics and Gynaecology, Hospital Universitario La Paz, Madrid, Spain. Martínez JM, Crispi F, Arévalo S, Gómez O, Pérez-Hoyos S, Carreras E, 5 Hospital Universitario 12 de Octubre, Universidad Computense de Madrid, Gratacós E, Llurba E. Longitudinal changes in foetal biometries and Madrid, Spain. 6Department of Paediatric Cardiology, Vall d’Hebron University cerebroplacental haemodynamics in foetuses with congenital heart disease. Hospital, Universitat Autònoma De Barcelona, Barcelona, Spain. 7Department Ultrasound Obstet Gynecol. 2017;49(3):379–86. of Pediatric Radiology, |Vall d’Hebron University Hospital, Universitat 17. Rychik J, Goff D, McKay E, Mott A, Tian Z, Licht D, Gaynor JW. Characterization Autònoma De Barcelona, Barcelona, Spain. 8Department of Pediatrics, Vall of the placenta in the newborn with congenital heart disease: distinctions d’Hebron University Hospital, Universitat Autònoma De Barcelona, Barcelona, based on type of cardiac malformation. Pediatr Cardiol. 2018;39(6):1165–71. Spain. 9Department of Neuropsicology, Vall d’Hebron University Hospital 18. Robinson HP, Sweet EM, Adam AH. The accuracy of radiological Barcelona, Barcelona, Spain. 10Department of Cardiology, Hospital Sant Joan estimates of gestational age using early fetal crown-rump length de Déu, Barcelona, Barcelona, Spain. 11Universitary Hospital Germans Trias i measurements by ultrasound as a basis for comparison. Br J Obstet Pujol, Barcelona, Spain. 12Director of Obstetrics and Gynaecology Gynaecol. 1979;86(7):525–8. Department, St Creu and St Pau Hospital, Sant Antoni Mª Claret, 167, 08025 19. Hadlock FP, Harrist RB, Shah YP, King DE, Park SK, Sharman RS. Estimating Barcelona, Spain. fetal age using multiple parameters: a prospective evaluation in a racially mixed population. Am J Obstet Gynecol. 1987;156(4):955–7. Received: 8 May 2018 Accepted: 26 August 2019 20. Carrascosa A, Fernandez JM, Fernandez C, Ferrandez A, Lopez-Siguero JP, Sanchez E, et al. Spanish growth studies 2008. New anthropometric standards. Endocrinol Nutr. 2008;55(10):484–506. 21. Kurmanavicius J, Wright EM, Royston P, Wisser J, Huch R, Huch A, et al. References Fetal ultrasound biometry: 1. Head reference values. Br J Obstet 1. Hoffman JI, Kaplan S. The incidence of congenital heart disease. J Am Coll Gynaecol. 1999;106(2):126–35. Cardiol. 2002;39:1890–900. 22. Kurmanavicius J, Wright EM, Royston P, Zimmermann R, Huch R, Huch A, 2. Moons P, Sluysmans T, De Wolf D, Massin M, Suys B, Benatar A, Gewillig M. et al. Fetal ultrasound biometry: 2. Abdomen and femur length reference Congenital heart disease in 111 225 births in Belgium: birth prevalence, values. Br J Obstet Gynaecol. 1999;106(2):136–43. treatment and survival in the 21s century. Acta Paediatr. 2009;98:472–7. 23. Figueras F, Meler E, Iraola A, Eixarch E, Coll O, Figueras J, et al. Customized 3. Bakiler AR, Ozer EA, Kanik A, Kanit H, Aktas FN. Accuracy of prenatal birthweight standards for a Spanish population. Eur J Obstet Gynecol diagnosis of congenital heart disease with fetal echocardiography. Fetal Reprod Biol. 2008;136(1):20–4. Diagn Ther. 2007;22:241–4. 24. International Society of Ultrasound in O, Gynecology, Carvalho JS, Allan 4. Majnemer A, Limperopoulos C, Shevell M, Rosenblatt B, Rohlicek C, LD, Chaoui R, Copel JA, et al. ISUOG Practice Guidelines (updated): Tchervenkov C. Long-term neuromotor outcome at school entry of sonographic screening examination of the fetal heart. Ultrasound infants with congenital heart defects requiring open-heart surgery. J Obstet Gynecol. 2013;41(3):348–59. Pediatr. 2006;148:72–7. 25. Crispi F, Valenzuela-Alcaraz B, Cruz-Lemini M, Gratacós E. Ultrasound assessment 5. Bellinger DC, Jonas RA, Rappaport LA, Wypij D, Wernovsky G, Kuban KCK, of fetal cardiac function. Australas J Ultrasound Med. 2013;16(4):158–67. Barnes PD, Holmes GL, Hickey PR, Strand RD, Walsh AZ, Helmers SL, 26. Comas M, Crispi F. Assessment of fetal cardiac function using tissue Doppler Constantinou JE, Carrazana EJ, Mayer JE, Hanley FL, Castaneda AR, Ware JH, techniques. Fetal Diagn Ther. 2012;32(1-2):30–8. Newburger JW. Developmental and neurologic status of children after heart 27. Tremblay E, Therasse E, Thomassin Naggara I, Trop I. Quality initiatives: surgery with hypothermic circulatory arrest or low-flow cardiopulmonary guidelines for use of medical imaging during pregnancy and lactation. bypass. N Engl J Med. 1995;332:549–55. Radiographics. 2012;32:897–911. 6. Massaro AN, Glass P, Brown J, Chang T, Krishnan A, Jonas RA, Donofrio MT. 28. Damodaram MS, Story L, Eixarch E. Foetal volumetry using magnetic Neurobehavioral abnormalities in newborns with congenital heart disease resonance imaging in intrauterine growth restriction. Early Hum Dev. 2012; requiring open-heart surgery. J Pediatr. 2011;158:678–81. 88(Suppl 1):S35–40. 7. Newburger JW, Jonas RA, Wernovsky G, et al. A comparison of the 29. Pistorius LR, Stoutenbeek P, Groenendaal F, de Vries L, Manten G, Mulder E, perioperative neurologic effects of hypothermic circulatory arrest versus Visser G. Grade and symmetry of normal fetal cortical development al low-flow cardiopulmonary bypass in infant heart surgery. N Engl J Med. longitudinal ultrasound study. Ultrasound Obstet Gynecol. 2010;36(6):700–8. 1993;329:54–9. 30. Alonso I. Depth of brain fissures in normal fetuses by prenatal 8. Rosenblatt B. Monitoring the central nervous system in children with ultrasound between 19 and 30 weeks of gestation. Ultrasound Obstet congenital heart defects: clinical neurophysiological techniques. Semin Gynecol. 2010;36:693–9. Pediatr Neurol. 1999;6(1):27–31. 31. Egaña G, Sanz-Cortés M, Figueras F, Bargalló N, Gratacós E. Differences in 9. McQuillen PS, et al. Temporal and anatomic risk profile of brain injury with cortical development assessed by fetal MRI in late-onset intrauterine growth neonatal repair of congenital heart defects. Stroke. 2007;38:736–41. restriction. AJOG. 2013;209(2):126.e1–8. 10. Limperopoulos C, Majnemer A, Shevell M, Rosenblatt B, Rohlicek C, 32. Simoes RV, et al. Feasibility and technical features of fetal brain Tchervenkov C. Neurologic status of newborns with congenital heart magnetic resonance spectroscopy in 1.5 T scanners. Am J Obstet defects before open heart surgery. Pediatrics. 1999;103:402–8. Gynecol. 2015;213(5):741–2. 11. Limperopoulos C, Majnemer A, Shevell MI, Rosenblatt B, Rohlicek C, 33. Smith SM, Jenkinson M, Woolrich MW, Beckmann CF, Behrens TE, Johansen- Tchervenkov C. Neurodevelopmental status of newborns and infants Berg H, Bannister PR, De Luca M, Drobnjak I, Flitney DE, Niazy RK, Saunders with congenital heart defects before and after open heart surgery. J J, Vickers J, Zhang Y, De Stefano N, Brady JM, Matthews PM. Advances in Pediatr. 2000;137:638–45. functional and structural MR image analysis and implementation as FSL. 12. Ruiz A, Ferrer Q, Sanchez O, Ribera I, Arevalo S, Alomar O, Mendoza M, Neuroimage. 2004;23(Suppl 1):S208-19. Review. Cabero L, Carreras E, Llurba E. Placental related complications in women 34. Bartha A, et al. The normal neonatal brain: MR imaging, diffusion tensor carrying a foetus with congenital heart disease. J Fetal Maternal Neonatal imaging, and 3D MR spectroscopy in healthy term neonates. AJNR Am J Medicine. 2016;29(20):3271–5. Neuroradiol. 2007;28(6):1015–21. 13. Gaynor J, Parry S, Moldenhauer J, Simmons R, et al. The impact of the 35. Vossough A, Limperopoulos C. Developmentand Validation of a maternal-foetal environment on outcomes of surgery for congenital heart semiquantitative brain maturation score on fetal MR images. Radiology. disease in neonates. Eur J Cardiothorac Surg. 2018;54:1–6. 2013;268(1):200–7. Ribera et al. BMC Pediatrics (2019) 19:326 Page 10 of 10 36. Mebius MJ, et al. Near-infrared spectroscopy as a predictor of clinical deterioration: a case report of two infants with duct-dependent congenital heart disease. BMC Pediatr. 2017;17(1):79. 37. Milne S, McDonald J, Comino EJ. The use of the Bayley Scales of Infant and Toddler Development III with clinical populations: a preliminary exploration. Phys Occup Ther Pediatr. 2012;32(1):24–33 Epub 2011 Aug 4. 38. Cruz-Martinez R, Figueras F, Hernandez-Andrade E, Benavides-Serralde A, Gratacos E. Normal reference ranges of fetal regional cerebral blood perfusion as measured by fractional moving blood volume. Ultrasound Obstet Gynecol. 2011;37:196–201. 39. Arduini D, Rizzo G. Normal values of Pulsatility index from fetal vessels: a cross-sectional study on 1556 healthy fetuses. J PerinatMed. 1990;18:165–72. 40. Baschat AA, Gembruch U. The cerebroplacental Doppler ratio revisited. Ultrasound Obstet Gynecol. 2003;21:124–7. 41. Donofrio MT, Massaro AN. Impact of congenital heart disease on brain development and neurodevelopmental outcome. Int J Pediatr 2010;2010: 359390. 42. 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. 43. Bellinger DC, Rappaport LA, Wypij D, Wernovsky G, Newburger JW. Patterns of developmental dysfunction after surgery during infancy to correct transposition of the great arteries. J Dev Behav Pediatr. 1997;18:75–83. 44. Bellinger DC, Wypij D, Kuban KC, et al. Developmental and neurological status of children at 4 years of age after heart surgery with hypothermic circulatory arrest or low-flow cardiopulmonary bypass. Circulation. 1999;100:526–32. 45. Hovels-Gurich HH, Seghaye MC, Dabritz S, Messmer BJ, von Bernuth G. Cognitive and motor development in preschool and school-aged children after neonatal arterial switch operation. J Thorac Cardiovasc Surg. 1991;114:578–85. 46. Blackwood M, Haka-Ikse K, Steward D. Developmental outcome in children undergoing surgery with profound hypothermia. Anesthesiology. 1986;65:437–40. 47. Ferry PC. Neurologic sequelae of cardiac surgery in children. Am J Dis Child. 1990;144:369–73. 48. Sanchez-de-Toledo J, et al. Cerebral regional oxygen saturation and serum neuromarkers for the prediction of adverse neurologic outcome in pediatric cardiac surgery. Neurocrit Care. 2014;21(1):133–9. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.