Proton MRS in Neonatal Asphyxia

Abstract & commentary

Synopsis: Conventional MR imaging significantly increases sensitivity to the early nonhemorrhagic parenchymal changes that may accompany birth asphyxia; single-voxel proton MR spectroscopy brings important additional information into the equation.

Source: Barkovich AJ, et al. Proton MR spectroscopy for the evaluation of brain injury in asphyxiated, term neonates. Am J Neuroradiol 1999;20:1399-1405.

Asphyxia is an unfortunate complication of birth with potential profound effects on infant survival and development. Clinical and laboratory criteria, however, are limited in their ability to predict neurologic and developmental outcome of asphyxiated neonates.

Standard MR imaging has been shown to be useful in the early evaluation of brain injury in asphyxiated neonates. In an attempt to improve prognostication for birth asphyxial injury, Barkovich and colleagues have begun to assess the role of proton MR spectroscopy in the evaluation of brain injury in asphyxiated term neonates. This study correlates proton MR spectroscopy (MRS) results with neurodevelopmental status at 12 months.

The 31 patients reported in this study are a subset of patients enrolled in an ongoing prospective study of the use of MR imaging of neonates who have had hypoxia or ischemia in the perinatal period. These patients are all term (> 36-week gestational age) neonates who had umbilical arterial pH less than 7.1, umbilical artery base deficit greater than 10, and/or 5-minute Apgar score less than or equal to 5. Patients with congenital malformations or infection were excluded. Age at the time of MR scanning ranged from 1-11 days, with a mean of 4.5 days. The MR protocol included sagittal T1-weighted and axial dual-echo spin-echo T2-weighted images, followed by two single-voxel proton spectra acquired using a commercially available pulse sequence, one centered in the deep gray matter nuclei and one centered in the frontal intervascular boundary zone ("watershed zone"). Metabolite ratios were calculated and tested for association with neurodevelopmental outcome as assessed by experienced pediatric neurologists.

The presence of lactate (a product of anaerobic respiration and, therefore, increased in the setting of asphyxia) was the most consistently noted spectral abnormality in these patients. Lactate was increased in both deep gray matter and watershed voxels, and its relative distribution correlated with the pattern of injury shown on conventional MR images (deep gray matter pattern of injury vs watershed injury). Reduced N-acetyl aspartate (NAA, a neuronal marker) was seen in some patients—usually those studied more than three days after birth. Overall, elevated lactate and reduced NAA were the most common findings in infants with neurologic and developmental abnormalities at age 12 months. An apparent false-negative finding (normal lactate/choline ratio but abnormal outcome) was likely due to prenatal injury with return to normal lactate levels by the time of scanning. False-positive findings in three children with apparent watershed injury but normal 12-month assessment may prove to be truly positive at further follow-up. Barkovich et al conclude that proton MRS is a valuable tool in the assessment of asphyxiated neonates and has a statistically significant association with both neurologic and cognitive status at 12 months.

Comment by Nancy J. Fischbein, MD

In addition to its profound effects on the affected neonate, birth asphyxia is a traumatic event for parents concerned about possible long-term effects on their child. The occurrence of birth asphyxia may also have medico-legal implications for physicians involved in the care of the baby and the mother. Both cranial sonography and CT scanning are relatively insensitive to the nonhemorrhagic parenchymal changes that may accompany an asphyxial event. Conventional MR imaging significantly increases sensitivity to the early nonhemorrhagic parenchymal changes that may accompany birth asphyxia, but these studies are often only subtly abnormal in the early days following the event and the changes may only be appreciated by an experienced observer. Single-voxel proton MR spectroscopy brings important additional information into the equation using a commercially available and easily performed technique. The simple determination of the presence or absence of lactate is an easily learned task and can contribute important additional information to the overall assessment of the asphyxiated term neonate. The presence of markedly elevated lactate levels may eventually factor into a decision of whether to continue life support. Mildly or moderately elevated levels of lactate may indicate that parents should be prepared for the possibility that their child may be neurodevelopmentally delayed, while normal lactate levels may be reassuring. An important caveat, however, is that lactate remains elevated for only a limited period and, therefore, as in the "false-negative" case mentioned by Barkovich et al, prenatal injury may result in normal lactate in the presence of an abnormal MRI and abnormal neurodevelopmental outcome. As the work presented in this small but important study is extended, the role of single-voxel proton MRS in birth asphyxia should become more clear.