Cerebral Lactate and Neurological Outcome in Brain-Injured Children

ABSTRACT & COMMENTARY
Source: Ashwal S, et al. 1H-Magnetic Resonance Spectroscopy-determined cerebral lactate and poor neurological outcomes in children with central nervous system disease. Ann. Neurol 1997;41:470-481.

The early and accurate prediction of neurological outcome in brain-injured children is an important clinical issue, as it may predict which children are most likely to benefit from intensive medical and rehabilitative efforts. Ashwal et al have used proton magnetic resonance spectroscopy (1H-MRS) to determine the presence of detectable lactate in occipital and parietal cortex in 97 significantly brain-injured children, and correlated these findings to neurological outcome at six and 12 months after injury. They find cerebral lactate to be a very good predictor of long-term neurological outcome.

Cerebral lactate forms from pyruvate by the NADH-dependent enzyme lactate dehydrogenase (LDH); thus, cerebral lactic acidosis reflects impaired cerebral pyruvate and/or NADH metabolism. NADH metabolism occurs primarily in mitochondria, via the electron transport chain, which uses oxygen as a final electron acceptor. Thus, a range of conditions, including a number of inborn errors of metabolism, meningitis, and hypoxic-ischemic encephalopathy (HIE), could result in impaired cerebral pyruvate and/or NADH metabolism. Technically, lactate can be detected in situ with 1H-MRS; lactate produces a characteristic doublet with a 1.3 ppm chemical shift that is quite characteristic with little interference. Thus, 1H-MRS allows one to probe specific aspects of cerebral metabolism in vivo.

All of the 97 children studied by Ashwal et al had suffered acute brain injury of significant degree to be considered at risk for neurodevelopmental sequelae. Ashwal et al chose to analyze their data by subdividing patients by age (newborns, infants, and older children) and according to whether there was (Lac+) or was not (Lac-) detectable lactate by 1H-MRS. Within each subgroup, there were a range of etiologies. Among infants and children, approximately 55% of the patients with detectable cerebral lactate had sustained cardiac arrest, while only 20% of patients without detectable cerebral lactate had sustained cardiac arrest. Cardiac arrest rate was only 17% in newborns with detectable cerebral lactate, suggesting that intrinsic disorders of pyruvate and NADH metabolism might be a more significant cause of cerebral lactic acidosis in this group.

Of the 36 patients with detectable cerebral lactate, considering all age groups, 75% had poor neurological outcome, defined as severe disability, persistent vegetative state, or death at six- or 12-month follow-up. In contrast, only 19% of children without detectable cerebral lactate had such a poor neurological outcome. The finding of detectable cerebral lactate was especially ominous in patients who had sustained evident HIE, as all 23 of such patients had a poor neurological outcome. However, of the HIE patients without elevated lactate, 72% still had a poor neurological outcome, suggesting that the finding of detectable cerebral lactate is sufficient but not necessary to guarantee a poor outcome.

COMMENTARY

The Loma Linda group headed by Dr. Ashwal has made major contributions over the past few years regarding the epidemiology of coma, persistent vegetative state, and severe brain injury in children. These new findings represent the start of efforts to determine useful brain biochemical markers that correlate strongly with outcome. Patients are divided between "detectable" and "non-detectable" levels of lactate, so it is possible that additional information could be gleaned by quantitatively correlating lactate signals to outcome. One point that was not directly addressed in the present study was the relationship between more readily accessible measures of pyruvate metabolism, such as blood and CSF lactate, with 1H-MRS findings. It is not clear what advantages 1H-MRS would have over blood measures in the acutely brain injured population. —rt