Comparison of prenatal and postnatal MRI findings in the evaluation of intrauterine CNS anomalies requiring postnatal neurosurgical treatment.

AIM: To assess the diagnostic capability of fetal magnetic resonance imaging (MRI) in children suspected antenatally to harbor central nervous system (CNS) defects that require immediate postnatal neurosurgical treatment. MATERIALS AND METHODS: Between 2003 and 2005, 13 fetal MRI scans were performed in mothers suspected to have fetuses with congenital CNS defects that would require surgery soon after birth. Comparisons between antenatal and postnatal scans were made with emphasis on diagnostic accuracy of antenatal examinations. RESULTS: All mothers were scanned using heavily T2-weighted fat-saturated sequences, allowing rapid acquisitions to avoid movement artefacts. Imaging quality was satisfactory in all patients. Diagnoses made antenatally were: myelomeningocele in seven, meningocele in one, diastematomyelia in one, occipital meningocele in one, and isolated hydrocephalus in three children. Of the seven children with antenatal diagnosis of myelomeningocele, one proved to have spinal lipoma postnatally. The patient who antenatally was diagnosed with meningocele proved to have spinal lipoma postnatally. These two were early antenatal MR scans. Antenatal diagnosis of hydrocephalus was made in five of the six confirmed myelomeningocele patients, which was verified postnatally. Antenatal diagnosis of Chiari II malformation was made in all six confirmed myelomeningocele patients. The antenatal diagnoses of occipital meningo-encephalocele and isolated hydrocephalus were verified postnatally. Antenatal diagnosis of diastematomyelia was not verified postnatally. CONCLUSION: Fetal MRI scanning is an effective, noninvasive method of assessing in-utero CNS abnormalities. The diagnostic accuracy has improved to allow prediction of clinical outcome and counseling for possible treatment, but is not perfect yet to allow counseling for termination of pregnancy.

Incidence of vascular malformations in spontaneous intra-cerebral haemorrhage in children.

AIM: To assess the incidence of vascular malformations in children presenting with non-traumatic intra-cerebral haemorrhage and outline the need for radiological investigations. MATERIALS AND METHODS: Between 1993 and 2002, 26 children presented with spontaneous intra-cerebral haemorrhage at a mean age of 26 months (range 1-192 months). RESULTS: Twenty-two children had, as first treatment, surgical removal of the haematoma and any malformation found. One patient with no evidence of vascular malformation had conservative treatment, two had embolisation and one had stereotactic radiosurgery of arteriovenous malformations (AVMs). The diagnosis of vascular malformation was confirmed histologically and/or radiologically in 16 (61%) patients. Of these 16 patients, 7 were AVMs, 1 thrombosed middle cerebral artery (MCA) aneurysm, 1 cavernous angioma, 6 aggregates of abnormal vessels, 1 vein of Galen aneurysm. Pre-operatively, 12 patients had magnetic resonance imaging (MRI)/magnetic resonance angiography (MRA) and 7 were positive for vascular malformation (1 false positive) with no false negatives. Digital subtraction angiography (DSA) was performed pre-operatively in 7 patients, 4 were positive with no false negatives. There were no re-bleeds at the follow-up period. Five (19%) patients died from the haemorrhage, 7 (27%) had severe neurological deficit and 14 (54%) had no neurological deficit. Residual malformation after surgery requiring additional treatment was found in 3 (16%) patients. All patients with confirmed malformations were followed-up with DSA. CONCLUSION: MRI/MRA has high sensitivity and specificity in identifying vascular malformations in children presenting acutely with spontaneous intra-cerebral haemorrhage. This may prove useful when pre-operative DSA is not promptly available. After acute clot evacuation, there is high incidence of residual malformation and such patients should be followed-up with DSA.