Postnatal Cerebral Findings in Infants With CHD

Study (First Author, Journal, Year of Publication)Study Design, No. InfantsCHDAntenatal DiagnosisMethodsFindings (Compared With Healthy Controls and/or Reference Values, Unless Otherwise Stated)
Brossard-Racine et al, ANJR Am J Neuroradiol, 2016Cohort study, N = 103Mixed100%MRI (structural)32% brain injury (26% acquired)
WMI most common injury (5 mild and 10 moderate or severe)
WMI located in the periventricular white matter, centrum semiovale, and frontal white matter
Second most common injury: nonhemorrhagic parenchymal injury
McCarthy et al, Pediatr Res, 2015Retrospective study, N = 72MixedUMRI (structural)18% PVL
The majority of PVL classified as moderate
Bertholdt et al, Eur J Cardiothorac Surg, 2014Case-control study, N = 30/20Mixed17%MRI (structural)23% WMI or stroke, 47% intracranial hemorrhage (subdural hematoma or choroid plexus)
Low Spo2 risk factor for brain injury, BAS not associated with brain injury
Brain injury associated with poorer neurologic functioning (82% abnormal assessment)
Owen et ala, J Pediatr, 2014Cohort study, N = 35Mixed51%MRI (structural)46% evidence of injury or immaturity on MRI (most common: hemorrhage)
71% suspect or abnormal neurobehavioral assessment (16 suspect, 9 abnormal)
Goff et al, J Thorac Cardiovasc Surg, 2014Cohort study, N = 57HLHS86%MRI (structural)19% PVL preoperatively
Brain immaturity and male sex independent strong predictors of PVL
Andropoulos et ala, Paediatr Anaesth, 2014Retrospective study, N = 59MixedUMRI (structural)46% preoperative brain injury
WMI most common injury (31%; 8 mild, 3 moderate, 1 severe)
Beca et ala, Circulation, 2013Cohort study, N = 153Mixed59%MRI (structural)26% brain injury (20% WMI, 5% stroke, 4% hemorrhage)
WMI associated with brain immaturity but not with BAS, diagnostic group, or GA at birth
WMI and stroke not associated with postoperative brain injury
Mulkey et al, Pediatr Cardiol, 2013Retrospective study, N = 73Mixed32%MRI (structural)47% ≥1 type of brain injury, 26% 2–4 injury types
25% brain injury if hemorrhage was excluded
Lower Apgar score at 5 min associated with brain injury
Ortinau et al, J Pediatr, 2013Case-control study, N = 15/12MixedUMRI (structural)Reduced cortical surface area and gyrification index for left and right hemispheres
46% focal signal abnormalities in the white matter
Glass et al, Cardiol Young, 2011Cohort study, N = 127MixedUMRI (structural)24% white matter injury
Infants with TGA and blood stream infection might have a higher risk of developing WMI (not significant in the whole group but significant when stroke was excluded)
Block et al, J Thorac Cardiovasc Surg, 2010Cohort study, N = 92TGAUMRI (structural)43% brain injury (23 stroke, 21 WMI, and 7 IVH)
SVABAS doubled the risk for brain injury
Higher Spo2 protective factor for brain injury (OR = 0.96)
Andropoulos et al, J Thorac Cardiovasc Surg, 2010Cohort study, N = 67Mixed44%MRI (structural)28% brain injury (single ventricle and 2 ventricles)
Brain immaturity associated with preoperative WMI and late death
58% of lesions partially or completely resolved at late MRI scan (3–6 mo)
Beca et al, J Am Coll Cardiol, 2009Cohort study, N = 64Mixed32%MRI (structural)30% brain injury (27% WMI and 5% stroke)
No differences between cardiac diagnoses
No association between BAS and brain injury
Petit et al, Circulation, 2009Retrospective study, N = 26 (14 BAS)TGAUMRI (structural)38% PVL, 0 strokes
Arterial oxygen saturation and time to surgery risk factors for brain injury
No association between BAS and brain injury
Licht et al, J Thorac Cardiovasc Surg, 2009Cohort study, N = 42TGA HLHS83% HLHS 39% TGAMRI (structural)21% PVL, 9.5% stroke, 86% incomplete closure of the opercular space (brain immaturity)
Lower total maturation scores (10.15), ∼1 mo younger than their actual GA
McQuillen et al, Circulation, 2006Cohort study, N = 29TGAUMRI (structural)41% brain injury (5 stroke, 2 WMI, 1 IVH, 4 combination of lesions)
5 min Apgar score, lowest Spo2, and BAS (12 of 19 infants with BAS had brain injury, 0 of 10 without BAS had brain injury) are risk factors for brain injury
Durandy et al, Artif Organs, 2011Cohort study, N = 21TGA57%MRI (structural)42% brain injury (4 infarct, 4 WMI, and 5 hemorrhages in 9 infants)
55% brain injury in antenatal diagnosis compared with 33% in postnatal diagnosis
Tavani et al, Neuroradiology, 200310Cohort study, N = 24MixedUMRI (structural)62% of infants delivered vaginally had hemorrhage on MRI
11 subdural hematomas
6 blood in the subdural space along the tentorium and falx or more laterally
7 blood in the choroid plexus
No relation between intracranial hemorrhage and abnormal neurologic examination
von Rhein et al, J Pediatr, 2015Case-control study, N = 19/19MixedUMRI, volume21% total brain volume reduction, all regions affected
Smallest difference: mesencephalon 7.7% smaller
Biggest difference: cortical gray matter 29.5% smaller and occipital lobes 28.5% smaller
Ortinau et al, Pediatr Cardiol, 2012Cohort study, N = 57/36MixedUMRI, volumeSmaller frontal, parietal, cerebellar, and brain stem measures
Brain growth rate not different
Differences in volume persisted at 3 mo, except for cerebellar measures
Somatic growth the greatest predictor of brain growth
Ortinau et al, J Thorac Cardiovasc Surg, 2012Cohort study, N = 67/36MixedUMRI (structural, volume)42% focal WMI
Smaller frontal, parietal, cerebellar, and brain stem
Frontal and brain stem most affected
Delayed maturation at the microstructural level
Makki et al, AJNR Am J Neuroradiol, 201311Case-control study, N = 15/11TGAUMRI (DTI)Higher apparent diffusion coefficient, lower FA genu corpus callosum
Lower FA splenium corpus callosum
Hagmann et al, J Child Neurol, 2016Case-control study, N = 22/22MixedUMRI (volume, DTI)Corpus callosum 25% (splenium) to 35% (genu) smaller
Total corpus callosum and splenium significantly smaller
Splenium lower FA, higher radial diffusion, diffusion coefficient not significant
No differences in other substructures of the corpus callosum
Mulkey et al, Pediatr Neurol, 2014Pilot study, N = 19MixedUMRI (structural, DTI)52% brain injury (WMI or stroke)
Lower FA in multiple major white matter tracts in infants with brain injury compared with infants without brain injury
Partridge et al, Ann Neurol, 2006Cohort study, N = 25MixedUMRI (structural, DTI)28% brain injury (focal or multifocal)
Brain injury associated with less change in FA over time in the pyramidal tract compared with newborns with 2 normal MRI scans
Infants with brain injury had the least dramatic changes with age detected by DTT
Trend in FA maturation rates across the 3 injury groups: newborns with normal scans had the most rapid changes, those with postoperative injury had intermediate maturation rates, and those with preoperative injury had the least rapid changes
No differences in directionally averaged diffusion coefficients
Sethi et al, Pediatr Res, 2013Cohort study, N = 36 CHDSVA61%MRI (structural, MRS)36% brain injury (4 mild WMI, 4 moderate WMI, 2 severe WMI, 6 focal strokes, 5 IVH)
Higher mean average diffusivity for gray matter and lower FA in the white matter regions
Lower mean NAA/Cho ratios and higher mean Lac/Cho ratios
Delayed microstructural brain development
Park et ala, Pediatr Cardiol, 2006Case-control study, N = 16/15TGAUMRI (structural, MRS)No abnormal findings on preoperative MRI
Altered metabolism in parietal white matter (increased Cho/Cr) and occipital gray matter (increased Cho/Cr and Ino/Cr)
Altered metabolism persisted 1 y after ASO in parietal white matter and normalized for occipital gray matter
Miller et al, Ann Thorac Surg, 2004Cohort study, N = 10TGAUMRI (structural, MRS)40% brain injury (stroke or hemorrhage)
Higher Lac/Cho
Similar NAA/Cho between TGA and healthy controls, but those with brain injury on MRI had lower NAA/Cho
0% focal deficits on neurologic examination
Abnormalities in tone or reflexes common in newborns with and without brain injury
Mahle et al, Circulation, 2002Cohort study, N = 24Mixed63%MRI (structural, MRS)25% ischemic lesions (small cortical watershed infarct, small infarct of the caudate, PVL)
4% hemorrhagic injury
16% elevated lactate with diffuse distribution, 25% lactate localized to the basal ganglia, 4% lactate in the peri-insular region
Elevation of brain lactate associated with brain injury
Dimitropoulos et al, Neurology, 2013Cohort study, N = 120Mixed33%MRI (structural, DTI, MRS)41% brain injury
Lower WM FA and lower NAA/Cho associated with higher injury severity preoperatively
Higher SNAP-PE, lower Spo2, hypotension, and BAS predictive for higher injury severity
Shedeed and Elfaytouri, Pediatr Cardiol, 2011Case-control study, N = 38/20MixedUMRI (structural, DTI, MRS)24% white matter injury (PVL and stroke)
Lower NAA/Cho ratio (0.55 vs 0.67)
Higher Lac/Cho ratio (0.14 vs 0.09)
Higher average diffusivity (1.41 vs 1.27)
Lower white matter FA (0.19 vs 0.25)
Miller et al, N Engl J Med, 2007Case-control study, N = 41/16SVA17%MRI (structural, DTI, MRS)32% brain injury
Decreased NAA/Cho (10%), increased average diffusivity (4%), decreased FA (12%), increased Lac/Cho (28%)
Nagaraj et al, J Pediatr, 2015Case-control study, N = 43/58Mixed100%MRI (structural, ASL)32.6% brain injury (64.3% WMI) compared with 0.6% in controls
All cerebral blood flow parameters lower but not significantly different
Lower global cerebral blood flow and regional cerebral blood flow in SVA
Lower regional thalamic cerebral blood flow in cyanotic CHD and lower cerebral blood flow in thalami, occipital white matter, and basal ganglia compared with acyanotic CHD
Licht et al, J Thorac Cardiovasc Surg, 2004Cohort study, N = 25MixedUMRI (volume, ASL)Mean cerebral blood flow 19.7 ± 9.1 mL/100 g per min compared with 50 ± 3.4 mL/100 g per min in controls
5 neonates cerebral blood flow <10 mL/100 g per min (moderate ischemic changes in piglets)
24% microcephaly
Low Hb associated with higher baseline cerebral blood flow
28% PVL, associated with lower cerebral blood flow and less reactivity to hypercarbia
Van der Laan et al, Pediatr Res, 2013Retrospective study, N = 21 (12 BAS)TGAUNIRSPreductal Spo2 increased immediately after BAS (72%–85%) and stabilized afterward (86%)
rcSo2 increased immediately after BAS and continued increasing during 24 h after BAS (42%–48% 2 h after BAS to 64% 24 h after BAS)
Lower baseline rcSo2 in the BAS group, whereas post-BAS rcSo2 was higher compared with infants who did not undergo BAS (64% vs 58%)
Uebing et al, J Thorac Cardiovasc Surg, 2011Cohort study, N = 53HLHS, TGAUNIRS10 h before surgery, HLHS infants had higher rcSo2 than TGA infants (61% vs 56%)
In HLHS infants, rcSo2 decreased after CPB and recovered to preoperative values within 48 h after CPB
In TGA infants, rcSo2 decreased after CPB and increased ∼20% above preoperative values within 48 h after CPB
Kurth et al, Ann Thorac Surg, 2001Case-control study, N = 91/19MixedUNIRSLower rcSo2 (immediately before surgery in the operating room, 1 min recordings)
Infants with PA had lowest rcSo2 values (38% ± 8%)
Latal et ala, Dev Med Child Neurol, 2015Cohort study, N = 77Mixed27%CUS29% brain injury (16% brain edema, 6% PVL, 4% ventricular dilatation, 3% IVH grade I)
Clinical variables not associated with brain injury
BAS associated with brain edema (32% vs 6%)
Gunn et ala, Ann Thorac Surg, 2012Cohort study, N = 39SVA95%aEEG33% EA, commonly left-sided, predominantly occurring during CPB
0% preoperative EA
Gunn et ala, Intensive Care Med, 2012Cohort study, N = 150MixedUaEEG3% preoperative EA
Te Pas et al, Acta Paediatr, 2005Retrospective study, N = 50MixedUCUS42% abnormal CUS (26% widening ventricles or subarachnoidal space, 8% ischemic changes, 6% lenticulostriate vasculopathy)
Abnormalities on CUS tended to occur more frequently in HLHS or CoA (63%) than TGA (14%)
Sigler et al, Ann Thorac Surg, 2001Cohort study, N = 35TGAUCUS3% preoperative brain injury (enhanced subependymal echogenicity)
65% resolved within 2 wk after operation
Neuron specific enolase not associated with brain injury
Combination of techniques
Mulkey et al, Pediatr Neurol, 2015Cohort study, N = 24Mixed100%aEEG63% abnormal aEEG (42% mildly abnormal, 21% severely abnormal)
MRI (structural)Abnormal aEEG associated with lower Apgar score at 5 min, CHD surgery at an older age, and male sex
50% brain injury (infarct and/or white matter injury)
Infants with brain injury higher odds of having abnormal aEEG (OR = 3.0)
33% brain atrophy
Severely abnormal aEEG background pattern associated with brain atrophy (OR = 15.0)
Dehaes et al, Biomed Opt Express, 2015Case-control study, N = 11/13SVAUNIRSLower cerebral oxygen metabolism index, cerebral blood flow index, cerebral oxygen saturation index, and hemoglobin
DCSHigher cerebral oxygen extraction
Jain et al, J Cereb Blood Flow Metab, 201412Cohort study, N = 32MixedUMRILower resting state oxygen extraction fraction, cerebral blood flow, and cerebral metabolic rate for oxygen
Lynch et al, J Thorac Cardiovasc Surg, 2014Cohort study, N = 37HLHSUMRI (structural)22% PVL
DOS (NIRS) DCSLonger time to surgery associated with postoperative PVL
Lower rcSo2 and higher blood flow index associated with postoperative PVL
Longer time to surgery associated with lower rcSo2 and higher FTOE
Rios et al, Pediatrics, 2013Cohort study, N = 167MixedUMRI (structural)3% brain injury ultrasound (4 hemorrhage, 1 PVL)
CUS26% brain injury MRI (WMI most common)
4 infants with hemorrhage on CUS had normal MRI suggesting 80% false positives and a positive predictive value for brain injury of only 20% for HUS before surgery
Andropoulos et ala, Ann Thorac Surg, 2012Cohort study, N = 30Mixed43%MRI (structural)33% brain injury
Mean preoperative rcSo2 56.5% (53.0%–61.9%)
NIRSrcSo2 < 45% area under the curve 9 (0–191) min
Williams et ala, Ultrasound Obstet Gynecol, 2012Pilot study, N = 13Mixed100%Ultrasound (Doppler)MCA-PI −0.75 TGA, −2.01 TOF, −2.4 HLHS
EEGCPR < 1 40% TGA, 67% TOF, 60% HLHS
MCA-PI positive correlation with neonatal EEG left frontal polar and left frontal β power
CPR < 1 associated with lower left frontal polar en left frontal β power
Ter Horst et al, Early Hum Dev, 2010Cohort study, N = 62Mixed15%aEEG40% normal aEEG, 45% mildly abnormal (DNV), 15% severely abnormal (BS, CLV, FT)
CUSSimilar rate of severely abnormal aEEG in cyanotic and acyanotic CHD (13% vs 16%)
19% EA, more frequently observed in acyanotic CHD (OR 9.37)
58% SWC within 72 h
In acyanotic CHD, SWC more frequent in CoA than in HLHS (92% vs 48%)
9% ischemia on HUS
Trend for more severely abnormal background patterns in abnormal HUS (OR 5.4)
Severely abnormal background pattern and EA associated with more profound acidosis (low pH, more negative base excess, higher lactate)
McQuillen et al, Stroke, 2007Cohort study, N = 62MixedUMRI (structural)39% preoperative brain injury (18% WMI, 21% stroke, 8% IVH)
Risk factors for preoperative brain injury: BAS and 5 min Apgar score
NIRSPreoperative brain injury more common in 2 ventricle anomalies
Toet et ala, Exp Brain Res, 2005Cohort study, N = 20TGAUNIRSLower rcSo2 (27%–52%) 12 h before CPB
aEEG100% normal aEEG, 0% EA
No difference in duration to normalization in aEEG after surgery between preoperative low or high rcSo2
Robertson et ala, Cardiol Young, 2004Cohort study, N = 47MixedUEEG11% preoperative abnormal EEG (2.8% clinical seizure)
CUSNadir CBF velocity 2 h post CPB
No association between CBF velocity and EEG
  • ASL, arterial spin labeling; ASO, arterial switch operation; BS, burst suppression; CLV, continuous low voltage; CoA, coarctation of the aorta; CPB, cardiopulmonary bypass; CUS, cranial ultrasound; DCS, diffuse correlation spectroscopy; DNV, discontinuous normal voltage; DTI, diffusion tensor imaging; DTT, diffuse tensor tractography; FA, fractional anisotropy; FT, flat trace; FTOE, fractional tissue oxygen extraction; GA, gestational age; IVH, intraventricular hemorrhage; Ino/Cr, myo-inosinotol/creatinine; MRS, magnetic resonance spectroscopy; OR, odds ratio; PVL, periventricular leukomalacia; SNAP-PE, Score for Neonatal Acute Physiology–Perinatal Extension; SpO2, pulse oxygen saturation; SVA, single ventricle anomaly; SWC, sleep-wake cycling; TOF, tetralogy of Fallot; U, unknown; WMI, white matter injury.

  • a Articles that also address neurodevelopmental outcome.