Characteristics and outcomes of critically ill patients with severe hyperammonemia.

PURPOSE: To determine the etiology and outcomes of critically ill patients with severe hyperammonemia. MATERIALS AND METHODS: Retrospective observational study of adults (18 years or older) admitted to a MICU from 2007 to 2016 who had a serum ammonia level >180 µmol/L (3 times the upper limit of normal). RESULTS: The 78 patients (45 male, 32 female) had a median age of 52 (interquartile range [IQR] 46-58) years. Hyperammonemia occurred most often with acute-on-chronic liver failure (ACLF) (49 %) or decompensated cirrhosis (27 %) and less often as a consequence of prior gastric bypass (9%), acute hepatic failure (6%), or valproic acid (3%). Median serum ammonia level was 201 µmol/L (IQR 126-265, range 18-736) on admission, with peak value of 245 µmol/L (IQR 205-336, range 185-842). Fifty (64%) patients died during the hospitalization. Cerebral edema was documented in 8 (10%) patients, only one of whom survived. Six of the 8 patients with cerebral edema had hyperammonemia related to ACLF, giving an incidence of 14% in this subset of patients. Neither mortality nor cerebral edema was associated with peak ammonia level. CONCLUSIONS: Critically ill patients with severe hyperammonemia have a high mortality rate and are at risk of developing cerebral edema.

Persistent Hyperinsulinemia Following High-Dose Insulin Therapy: A Case Report.

INTRODUCTION: Overdoses of beta-adrenergic antagonists and calcium channel antagonists represent an uncommonly encountered but highly morbid clinical presentation. Potential therapies include fluids, calcium salts, vasopressors, intravenous lipid emulsion, methylene blue, and high-dose insulin. Although high-dose insulin is commonly used, the kinetics of insulin under these conditions are unknown. CASE REPORT: We present a case of a 51-year-old male who sustained a life-threatening overdose after ingesting approximately 40 tablets of a mixture of amlodipine 5 mg and metoprolol tartrate 25 mg. Due to severe bradycardia and hypotension, he was started on high-dose insulin (HDI) therapy; this was augmented with epinephrine. Despite the degree of his initial shock state, he ultimately recovered, and HDI was discontinued. Insulin was infused for a total of approximately 37 hours, most of which was dosed at 10 U/kg/hour; following discontinuation, serial serum insulin levels were drawn and remained at supraphysiologic levels for at least 24 hours and well above reference range for multiple days thereafter. CONCLUSION: The kinetics of insulin following discontinuation of high-dose insulin therapy are largely unknown, but supraphysiologic insulin levels persist for some time following therapy; this may allow for simple discontinuation rather than titration of insulin at the end of therapy. Dextrose replacement is frequently needed; although the duration is often difficult to predict, prolonged infusions may not be necessary.