PURPOSE: To investigate ammonia and glutamine levels in valproate (VPA)-related hyperammonemic encephalopathy (VHE). METHODS: We reviewed the medical records and EEG recordings of seven adults diagnosed with VHE. RESULTS: Venous ammonia levels were elevated in five (71%) of the seven patients. Elevated serum or cerebrospinal fluid (CSF) glutamine levels were found in four (80%) of five cases tested, including two who had normal ammonia levels. Initial behavioral signs included violent outbursts in three patients, paranoid ideation severe enough to require restraint in two cases, and milder abnormalities in two instances. The severity of encephalopathy was not related to any particular serum VPA level. In four patients serum VPA levels did not exceed 100 microg/ml, and in one case, VHE developed after taking only one 250-mg dose. Symptoms eventually cleared after reducing the dose of, or discontinuing, VPA. Liver-function tests were normal. Each of six patients tested had EEG findings that supported the diagnosis of VHE and excluded nonconvulsive status epilepticus. The rate of normalization of one patient's serum glutamine level and the EEGs of two cases correlated better with the timing of their delayed clinical recovery than did the more rapid rate of decline of the serum ammonia levels. CONCLUSIONS: Serum or CSF glutamine levels are initially elevated in a majority of patients with suspected VHE, sometimes in the absence of hyperammonemia. Glutamine levels may be useful adjunctive laboratory tests for the diagnosis of VHE.
PURPOSE: To investigate ammonia and glutamine levels in valproate (VPA)-related hyperammonemic encephalopathy (VHE). METHODS: We reviewed the medical records and EEG recordings of seven adults diagnosed with VHE. RESULTS: Venous ammonia levels were elevated in five (71%) of the seven patients. Elevated serum or cerebrospinal fluid (CSF) glutamine levels were found in four (80%) of five cases tested, including two who had normal ammonia levels. Initial behavioral signs included violent outbursts in three patients, paranoid ideation severe enough to require restraint in two cases, and milder abnormalities in two instances. The severity of encephalopathy was not related to any particular serum VPA level. In four patients serum VPA levels did not exceed 100 microg/ml, and in one case, VHE developed after taking only one 250-mg dose. Symptoms eventually cleared after reducing the dose of, or discontinuing, VPA. Liver-function tests were normal. Each of six patients tested had EEG findings that supported the diagnosis of VHE and excluded nonconvulsive status epilepticus. The rate of normalization of one patient's serum glutamine level and the EEGs of two cases correlated better with the timing of their delayed clinical recovery than did the more rapid rate of decline of the serum ammonia levels. CONCLUSIONS: Serum or CSF glutamine levels are initially elevated in a majority of patients with suspected VHE, sometimes in the absence of hyperammonemia. Glutamine levels may be useful adjunctive laboratory tests for the diagnosis of VHE.