Proc.฀Intl.฀Soc.฀Mag.฀Reson.฀Med฀9฀(2001) 585 ) MR spectroscopic changes in pediatric acute hydrocephalus, sub-acute hydrocephalus and cortical atrophy Miriam SCADENG 1 , Rex MOATS 2 , S YAMADA 3 , Marvin D NELSON Jr 2 , J Gordon McCOMB 3 1 Department of Radiology & Rudi Schulte Research Institute, Childrens Hospital Los Angeles, 4650 Sunset Blvd, Los Angeles, CA USA; 2 Department of Radiology, Childrens Hospital Los Angeles, 4650 Sunset Blvd, Los Angeles, CA USA; 3 Department of Neurosurgery & Rudi Schulte Research Institute, Childrens Hospital Los Angeles, 4650 Sunset Blvd, Los Angeles, CA USA; Introduction Hydrocephalus (HC) is a potentially fatal condition, frequently characterised by raised intracranial pressure (ICP), which results in compression and thinning of the cortical mantle, and abnormal accumulation of water within the brain. Presenting symptoms range from increasing head size to headache, vomiting and drowsiness. The diagnosis is usually made from the clinical symptoms and signs and confirmed on imaging studies. In patients with less acute symptoms it can be difficult to differentiate between dilated ventricles due to raised pressure who would benefit from surgery, and those with other diseases causing cortical thinning. This is a particular problem in the pediatric population. Aim To determine the utility of MRS in identifying those patients with large ventricles who would benefit from surgical cerebrospinal fluid (CSF) diversion. Methods Patients with clinical suspicion of HC and enlarged ventricles on MRI were divided into three groups: 1. Acute group (n=8). Rapid deterioration with classical symptoms of acute raised ICP (nausea, vomiting, headache, drowsiness). Ventricles reduced in size post CSF diversion. 2. Sub acute group (n=6). Chronic symptoms/signs of HC (increasing head circumference, chronic headache). Ventricles reduced in size post CSF diversion. 3. Non-operated `Atrophy group´ (n=10). Patients with enlarged ventricles on imaging studies at presentation who were not operated upon as it was decided that the changes in the brain were not due to impairement of normal CSF draining pathways, and that these patients would not benefit form CSF diversion. Children in groups 1 and 2 were also divided by age into above/below 18months of age. Age matched control data for metabolite ratios was obtained from composite data charts from normal subjects. Imaging parameters: 1.5T Siemens Magneton SP4000 scanner. The imaging protocol T1w (TR/TE) 608/14 and T2w (3500/90. Localised proton MR spectra, short echo time STEAM, selective water suppression (2000/20, 30 ms mixing time). 256 averages. Voxel placement periventricular white matter or occipital grey matter voxel size less than 12.5cm 3 . Statistical significance was ascertained using student t-test for each group. The results presented as percentage of the expected normal value (corrected for age). Results The mI/Cr and Cho/Cr ratios were normal for all groups. Acute group: NAA/Cr reduced p= 0.0034. Significant lactate peak in 2/8. Sub-acute group: NAA/Cr not reduced p= 0.6248. Atrophy group: NAA/Cr reduced p= 0.0005 Significant lactate peak in 1/10. 7/8 of the acute group were age 18 months or older. (NAA/Cr reduced - p=0.0017). 6/6 of the sub-acute group were under 18 months of age. (NAA/Cr normal - p=0.7650). Discussion The effect of hydrocephalus on the metabolite ratios has been studied in both animal models and in human patients. Of particular interest is the NAA/Cr as NAA is a marker of functioning neurons. Previous human pediatric studies have shown that the NAA/Cr ratio is preserved in HC, whereas it is reduced in patients with a thinned cortical mantle for other reasons. Much of this work has been done in children under the age of two years and in mixed acute and sub acute groups. Studies on animal models of acute HC demonstrate a reduced NAA/Cr ratio and the presence of lactate. This may be due to ischaemic injury secondary to the tissue hypoperfusion caused by compression of the brain matter with neuronal dysfunction occuring once the cerebral perfusion pressure drops below a critical level. Anaerobic brain metabolism leads to the formation of lactate. The NAA/Cr ratio was reduced in our group of patients who presented with acute symptoms of raised ICP; in fact 6/8 of these patients had reduced NAA/Cr ratio. 2 of these patients presented as emergencies (both had brain tumors and were previously well). In addition to having markedly reduced NAA/CR ratios, they both also had lactate in their spectra. Seven of the eight children in the acute group were older than 18 months. This suggests that at least under the age of 18 months patient's brains may be protected to a degree from compression and hence ischaemia by being able to increase their head size. The only child under 18 months of age with acute symptoms had shunt failure and therefore there may not have been enough time to accommodate for the sudden change in intracranial pressure. The patients who presented with more chronic symptoms did not necessarily have less dilated ventricles but the were all under the age of 18 months. Individual NAA/Cr ratios varied which is likely to reflect underlying disease processes, however as a group NAA/Cr was not reduced. The children in the atrophy group were not operated upon as it was felt that the dilated ventricles and thin cortices were satisfactorily explained by clinical assessment and imaging. The diverse range of cerebral pathologies, which caused the cortical destruction, explains the reduced NAA/Cr. The patient in this group in whom lactate was found was a premature infant who had suffered from an intraventricular haemorrhage. Lactate is a normal finding at this age. Since changes of water content in the brain in Hc complicate the quantification of metabolite concentrations we have not included absolute quantitation in this study. Conclusions 1. The decision on whether to operate on children with sub acute symptoms and signs of HC may be made easier if the NAA/Cr is normal as the imaging changes are not due to neuronal loss/destruction. 2. Patients with clinical symptoms, signs and imaging consistent with acute HC are obvious and do not require MRS. The reduced NAA/Cr ratio may reflect neuronal ischaemia. 3. Acute symptoms occurred mostly in children older than 18 months. Children under the age of 18 months may be protected to some degree from the effects of raised ICP by having a potentially expandable skull. Child with acute-onset HC