CMRI-detected brain injuries and clinical key risk factors associated with adverse neurodevelopmental outcomes in very preterm infants

Abstract

Neurological impairment is high after preterm birth. This study evaluates the impact and interplay of cMRI-detected brain injuries (BI) and clinical risk factors on neurodevelopmental outcomes and extracts the most important key factors. A retrospective analysis was conducted on risk factors (perinatal/neonatal, cMRI-detected BI) for adverse motor (MO) and cognitive (CO) outcomes (Bayley Scales of Infant Development, 24 months corrected age) in a tertiary center cohort (2009-2018) of very preterm infants (< 32 weeks of gestation) using uni-/multivariable regression models. We included 342 infants (mean gestational age:28.0 ± 2.3 weeks; male:49%). Significant clinical predictors for MO/CO included GA, birthweight, APGAR score, catecholamine treatment, ventilation, retinopathy of prematurity, transfusion of red blood cells (RBCs), bronchopulmonary dysplasia, surgery, and patent ductus arteriosus interventions (all p < 0.01/p < 0.01), surfactant (MO: p = 0.037), and sepsis (p < 0.001/p = 0.016). (Severe) cMRI-detected BIs (> 1, all p < 0.05) and not only severe intraventricular hemorrhage (IVH) III°/III°+PVHI and ventricular dilatation (VD) (all p < 0.05), but also mild/moderate injuries like IVH II° (p < 0.001/p < 0.024), cerebellar hemorrhage (CO: p = 0.028), and moderate VD (MO: p = 0.005) significantly impacted outcomes. Independent key factors were > 1 severe cMRI-detected BI (MO/CO:-11.27/-10.3 score points (sp), p = 0.021/0.043), APGAR score (10 min, MO/CO:+5.3/+4.45 sp/point, p < 0.001/p < 0.001), surfactant administration (MO:+4.88 sp, p = 0.031), and transfusion of RBCs (MO/CO:-1.69/-1.96 sp/transfusion, p = 0.006/p < 0.001). In conclusion, combining imaging and clinical (key) risk factors is important for risk stratification of preterm infants. Even mild BI, like IVH II°, significantly contributes to adverse outcomes, underlining the importance of cMRI.

Keywords: Brain injuries; Magnetic resonance imaging; Neurodevelopmental outcomes; Pediatric research; Preterm infants.

Fig. 1

Overview: severe brain injuries. (A) IVH III° (SWI-sequence, white arrows), (B) PVHI (SWI-sequence), (C) Ventricular dilatation (T2-sequence), (D) CBH III° (SWI-sequence) (E) CBH IV° (SWI-sequence, F) bilateral cPVL (T2-sequence). CBH: cerebellar hemorrhage, cPVL: cystic periventricular leukomalacia, IVH: intraventricular hemorrhage, PVHI: periventricular hemorrhagic infarction.

Fig. 2

Flowchart of the study population. CO: cognitive outcome, MO: motor outcome.

Fig. 3

Association between cMRI-detected brain injuries and neurodevelopmental outcomes. For motor outcome: IVH grades, IVH III°+PVHI, ventricular dilatation, number of BI, > 1 BI, severe BI, number of severe BI, > 1 severe BI. For cognitive outcome: IVH grades, IVH III°+PVHI, ventricular dilatation, CBH, number of BI, <1BI, severe BI, number of severe BI, > 1 severe BI. BI: brain injury, CBH: cerebellar hemorrhage, cPVL: cystic periventricular leucomalacia, IVH: intraventricular hemorrhage, PVHI: periventricular hemorrhagic infarction, PWML: punctate white matter lesions, sBI: severe brain injury, VD: ventricular dilatation, BI: IVH I°; II°; III°, PVHI, moderate/severe VD, CBH I°-IV°, PWML, cPVL) and severe BIs: IVH III°, PVHI, CBH III°+IV°, severe VD, cPVL, significant: p < 0.05.

Fig. 4

Key factors for motor and cognitive outcome of preterm infants born < 32 weeks of gestation. For motor outcome: (1) Apgar score at 10 min (p = 0.615), (2) Surfactant application in the delivery room (p = 0.43), (3) transfusion of RBCs (p < 0.001), distribution and quantity of transfusion of RBCs are shown, (4) more than one severe brain injury (p < 0.001). For cognitive outcome: (1) Apgar score at 10 min (p = 0.153), (2) transfusions of RBCs (p < 0.001), the numeric distribution and quantity of transfusion of RBCs are shown in the figure, and (3) presence of more than one severe brain injury. Descriptive values of the key factors were used for the figure. Please note: Surfactant application was associated with a negative coefficient in the descriptive analysis (shown in the figure) and in the multivariable regression analysis the coefficient became positive.