RESUMO
INTRODUCTION: Hyperammonemia (HA) is a potential side-effect of valproate (VPA) treatment, which has been described during long-term administration. The aim of this study was to evaluate the incidence, the impact and the risk factors of HA in critically ill patients. METHODS: We reviewed the data of all adult patients treated in our mixed 35-bed Department of Intensive Care over a 12-year period (2004-2015) who: a) were treated with VPA for more than 72 h and b) had at least one measurement of ammonium and VPA levels during the ICU stay; patients with Child-Pugh C liver cirrhosis were excluded. HA was defined as ammonium levels above 60 µg/dl. RESULTS: Of a total of 2640 patients treated with VPA, 319 patients met the inclusion criteria (median age 64 years; male gender 55%); 78% of them were admitted for neurological reasons and ICU mortality was 30%. Median ammonium levels were 88 [63-118] µg/dl. HA was found in 245 (77%) patients. For those patients with HA, median time from start of VPA therapy to HA was 3 [2-5] days. In a multivariable analysis, high VPA serum levels, mechanical ventilation and sepsis were independently associated with HA during VPA therapy. In 98/243 (40%) of HA patients, VPA was interrupted; VPA interruption was more frequent in patients with ammonium levels > 100 µg/dl than others (p = 0.001). HA was not an independent predictor of ICU mortality or poor neurological outcome. CONCLUSIONS: In this study, HA was a common finding during treatment with VPA in acutely ill patients. VPA levels, sepsis and mechanical ventilation were risk factors for HA. Hyperammonemia did not influence patients' outcome.