National and Kapodistrian University of Athens
Η μαγνητική τομογραφία στην τελειόμηνη ωριμότητα ως προγνωστικός παράγοντας της νευροαναπτυξιακής εξέλιξης του πρόωρου νεογνού
Abstract
dc:descriptionIntroduction: Very low birth weight premature infants are at high risk for adverse neurodevelopmental outcomes. 5 to 15 percent present serious problems such as cerebral palsy, cognitive delay and severe neurosensory (visual or hearing) impairment. 25 to 50 percent present milder problems such as motor, cognitive behavioral and learning difficulties. Early identification of premature babies who are at risk of having serious long-term sequelae is crucial, as they can benefit from early intervention programs. Magnetic resonance imaging (MRI) of the neonatal brain demonstrates the lesions, mainly the diffuse ones, more precisely compared to Ultrasound. Cystic lesions and especially diffuse white matter damage are often in premature babies and are related with motor and cognitive difficulties. Aim: To evaluate the predictive value of MRI at term equivalent age performed at premature neonates admitted to a level III referral Neonatal Intensive Care Unit. Specifically: - To compare the incidence of brain lesions detected on cranial ultrasonography and MRI at term equivalent age. - To correlate the MRI findings at term equivalent age with motor and neurodevelopmental problems at two years corrected age. - To compare the predictive value of MRI and US at term equivalent age. Patients: This was a prospective study including an unselected cohort of premature neonates admitted to level III referral Neonatal Intensive Care Unit, from July 2007 to January 2010. Inclusion criteria to the study were premature babies born at 30 weeks gestation or less, or a birth weight less than 1250 g, or premature babies born at 30 to 34 weeks of gestation with brain lesions on early US scans. Methods: 73 neonates were initially recruited into the study. Demographic data and risk factors of adverse neurodevelopmental outcome were recorded. Sequential cranial US scans were performed according to a standardized protocol and brain lesions were recorded. Before discharge and around term equivalent age (36-40 weeks corrected), the following were performed: 155 1. Neurological evaluation (Dubowitz Neonatal Neurological Examination) 2. Cranial Ultrasound 3. MRI At two years corrected age, 55 infants underwent a complete neurodevelopmental assessment: 1. Neurological evaluation (Hammersmith Infant Neurological Examination) 2. Evaluation of development with use of the Griffiths Mental Development Scales- Extended Revised, 2006 3. Recording of neurosensory impairment 4. Measurement of the maximal occipitofrontal head circumference Results: The mean gestational age (±SD) of the initially studied 73 neonates, was 28.8±2.5 weeks and the mean birth weight (±SD) 1100±293 g. Lesions on US were compared with the corresponding ones on MRI. Abnormal MRI was seen in 62 infants (84.9%), while abnormal US was seen in 39 (53.4%). In 25 infants (34.2%) MRI had abnormal findings while US was normal. The reverse, i.e. abnormal findings on US and normal MRI was observed in only 2 infants (2.7%). Finally, in 37 infants (50.6%) both imaging techniques demonstrated abnormal findings and in 9 infants (12.3%) none of the methods showed any abnormal findings. US underestimated the diffuse white matter damage (6.9% versus 26.4% on MRI), the increase of the interhemispheric fissure and subarachnoid spaces (19.2% vs 43.1% on MRI) and the thinning of the corpus callosum (5.5% vs 27.4% on MRI). MRI was the only method to detect abnormalities in the posterior fossa (13.6%), basal ganglia (5.6%), cortex (6.9%) and the posterior limb of the internal capsule (57.5%). US detected the presence of cystic periventricular leukomalacia (9.7% vs 13.9% on MRI), ventricular enlargement (41.6% vs 40.3% on MRI), persistent germinal matrix/intraventricular hemorrhage (23.6% on both methods) and porencephalic cyst as a result of periventricular hemorrhagic infraction (9.7% vs 13.9% on MRI). When each perinatal variable was related separately with each of the abnormal findings on US and MRI, bronchopulmonary dysplasia, patent ductus arteriosus and postnatal administration of steroids were found to be potential risk factors, while antenatal administration of steroids acted in a protective manner. Regarding the neurological examination at term equivalent age, 17 (31%) out of 55 neonates had suboptimal scores, and 38 (69%) had optimal scores. The mean score was 29.2. 156 Most of the abnormal neurological items were found in the sections movements, posture and tone and tone patterns. The abnormal neurological items and sections of the Dubowitz Neonatal Neurological Examination were retrospectively studied in all infants who presented cerebral palsy at two years corrected age. 18 neurological items were found to have a significant predictive value for cerebral palsy. The main abnormal items were: Moro reflex, tendon reflexes, popliteal angle, arm traction, leg traction, leg recoil, ventral suspension, head control-anterior neck muscles (flexor tone) and head lag. The main abnormal section was tone patterns and specifically neck extensor tone (sitting), and assessment of extensor tone (horizontal). The predictive value of the findings on MRI and US in relation to the neurological examination at term equivalent age was also studied. Regarding the US, statistically significant findings for the prediction of an abnormal neurological examination were: intraventricular hemorrhage, dilatation-asymmetry of the ventricles and widening of interhemispheric fissure. Regarding the MRI, statistically significant findings for the prediction of an abnormal neurological examination were: cystic abnormalities, dilatation of the ventricles, delayed/impaired myelination (signal of the posterior limb of the internal capsule-PLIC), posthemorrhagic porencephalic cysts, posterior fossa lesions and abnormal signal intensity in cortical grey matter combined with delay of cortical gyration maturation. 55 infants were assessed at two years corrected age (75% of the total cohort). Regarding the neurological examination at two years corrected age, 23 (41.8%) out of 55 infants had suboptimal scores, and 32 (58.2%) had optimal scores. 12 (70.6%) out of the 17 neonates with suboptimal scores on the neurological examination at term, went on to have suboptimal scores at 24 months corrected age. 11(28.9%) out of the 38 neonates with optimal scores on the neurological examination at term, had suboptimal scores at 24 months corrected age. Only 12 (22%) out of the 55 infants that were assessed had suboptimal scores on both neurological examinations. It is worth noting that 3 of the infants who developed cerebral palsy, had optimal scores at term. The Dubowitz neonatal neurological examination was found to have sensitivity 52.2%, specificity 84.4%, positive predictive value 70.6% and negative predictive value 71.1% for identifying motor outcome. 11 infants (20%) developed cerebral palsy. According to the Gross Motor Function Classification System for Cerebral Palsy, 4 infants (36%) were classified in level I, 4 (36%) in level II, 1 (9%) in level III, 1 (9%) in level IV and 1 (9%) in level V. Topography was 157 distributed as 4 infants with diplegia, 5 with quadriplegia, 1 with triplegia and 1 with hemiplegia. The predictive value of the findings on MRI and US in relation to the development of cerebral palsy was also studied. Regarding the US, statistically significant findings for the prediction of cerebral palsy were: intraventricular hemorrhage, dilatation-asymmetry of the ventricles, cystic periventricular leukomalacia, and widening of interhemispheric fissure. Regarding the MRI, statistically significant findings for the prediction of cerebral palsy were: cystic abnormalities, dilatation of the ventricles, delayed/impaired myelination (signal of the PLIC), persistent/remnants of germinal layer/intraventricular hemorrhage, posthemorrhagic porencephalic cysts, posterior fossa lesions and abnormal signal intensity in cortical grey matter alone or combined with delay of cortical gyration maturation. Neurodevelopmental follow-up assessment with use of the Griffiths Mental Development Scales was performed, at a mean uncorrected age of 27.2 ± 0.9 months and at a mean corrected age of 24.5 ± 0.6 months. Both uncorrected and corrected quotients were calculated. The mean overall uncorrected quotient was 64.8 ± 17.3 and the median 70.4. The mean overall corrected quotient was 70.4 ± 19.2 and the median 76.9. Based on the chronological age, normal quotients were found in 3 infants (5.5%), mildly abnormal in 31 (56.4%), moderately abnormal in 11 (20%) and severely abnormal in 10 (18.2%). Based on the corrected age, normal quotients were found in 10 infants (18.2%), mildly abnormal in 27 (49.1%), moderately abnormal in 8 (14.5%) and severely abnormal in 10 (18.2%). The mean overall corrected quotient was significantly higher compared to the uncorrected one (p < 0.01). The same was seen in all the subscales mean quotients as well. Depending on how the evaluation was done, there were removals of children to different groups. Either ways, the same 10 children with severe delays were recognized. The predictive value of the findings on MRI and US in relation to severe neurodevelopmental delays was also studied. Regarding the US, statistically significant findings for the prediction of severe neurodevelopmental delays were: intraventricular hemorrhage, dilatation-asymmetry of the ventricles, cystic periventricular leukomalacia, and widening of interhemispheric fissure. Regarding the MRI, statistically significant findings for the prediction of severe neurodevelopmental delays were: cystic abnormalities, dilatation of the ventricles, thinning of the corpus callosum, posthemorrhagic porencephalic cysts, posterior fossa lesions and the abnormal signal intensity in cortical grey matter combined with delay of cortical gyration maturation. 158 A multivariate analysis of variance, with dependent variables the subscales and overall quotients and independent variables the findings on US and MRI was carried out. According to this analysis, lesions on US such as intraventricular hemorrhage and porencephalic cyst were found to have significant predictive value for locomotor delays. More lesions detected on MRI, such as foci of high signal intensity on T1 weighted imaging (punctuate white matter lesions), foci of high signal intensity on T2 weighted imaging, thinning of the corpus callosum, abnormal signal intensity in cortical grey matter, posthemorrhagic porencephalic cysts, basal ganglia and posterior fossa lesions, were found to have significant predictive value for delays in different areas of development. The overall quotient was mostly affected by the presence of foci of high signal intensity on T2 weighted imaging and thinning of the corpus callosum. Neurosensory impairment was also recorded. 22 (40%) out of 55 infants had visual problems of different severity: 2 blindness, 3 astigmatism, 5 hypermetropia, 13 strabismus, 2 nystagmus and 8 myopia. All 10 children with severe neurodevelopmental delays on Griffiths scales, had visual problems. 9 out of the 11 infants who developed cerebral palsy, had multiple and serious visual problems. Deafness was diagnosed in only 1 (1.8%) infant, who developed cerebral palsy and had severe neurodevelopmental delay. Concerning the measurement of the maximal occipitofrontal head circumference, 17 (30.9%) out of 55 infants, had a head circumference ≤ 3rd percentile. 10 of these 17 infants, had also a head circumference ≤ 3rd percentile at term equivalent age. 4 (36.3%) out of the 11 infants who developed cerebral palsy, presented microcephaly. 6 (60%) out of the 10 infants with severe neurodevelopmental delays on Griffiths scales, presented microcephaly. The remaining 11 out of the 17 with microcephaly had either mild or moderate difficulties on Griffiths Scales. None of the 17 infants had normal quotients. A statistically significant correlation was found between posterior fossa lesions and microcephaly (positive predictive value 85.7%, negative predictive value 76.6%). Conclusions: Magnetic Resonance Imaging at term equivalent age, compared to Ultrasound, demonstrates the exact site and extent of brain lesions of the premature baby, such as the diffuse white matter damage, the increase of the subarachnoid spaces and the interhemispheric fissure and the thinning of the corpus callosum. It is the only imaging modality that identifies brain lesions in deeper and more peripheral structures, such as posterior fossa, basal ganglia, cortex and the posterior limb of the internal capsule. 159 More brain lesions, detected by MRI, had a statistical significant predictive value regarding the development of cerebral palsy and the presence of severe neurodevelopmental delays. Neurological examination at term equivalent age helps in identifying preterm infants who are at risk of having serious long-term sequelae. Nevertheless, the relatively low sensitivity and positive predictive value of one examination must be noted. Additionally, many of premature babies’ problems become clinically detectable beyond the neonatal period or even months after discharge. Sequential neurological examinations within neurodevelopmental follow-up programs, combined with the additional information that MRI provides, can identify high risk premature babies. The evaluation of development with use of the Griffiths Scales must be based on the chronological (uncorrected) age of the premature baby. This way, fewer babies run the risk of missing early intervention programs. MRI at term cannot replace US, which remains the standard of care for routine neuroimaging in the Neonatal Intensive Care Unit. MRI is a complementary tool since it can improve brain assessment, detection of anatomic abnormalities and prognosis of the neurodevelopmental outcome of premature babies. Early identification of premature babies who are at risk of having serious long-term sequelae is crucial, because it results in early intervention and possible improvement of the outcome.
Degree
thesis:*- Grantor dc:publisher
- National and Kapodistrian University of Athens
- Year dc:date
- 2013
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Petropoulou, Chrissa
- Πετροπούλου, Χρυσούλα
Subjects
dc:subject × 16- Μαγνητική τομογραφία
- Υπερηχογράφημα
- Προωρότητα
- Τελειόμηνη ωριμότητα
- Προγνωστικός παράγων
- Νευροαναπτυξιακή εξέλιξη
- Magnetic resonance imaging
- Ultrasound
- Prematurity
- Term equivalent age
- Predictive factors
- Neurodevelopmental outcome
- Ιατρική και Επιστήμες Υγείας
- Κλινική Ιατρική
- Medical and Health Sciences
- Clinical Medicine
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/29836