Relationship of continuous infusion lorazepam to serum propylene glycol concentration in critically ill adults* Alejandro C. Arroliga, MD; Nadine Shehab, PharmD; Kevin McCarthy, RCPT; Jeffrey P. Gonzales, PharmD, BCPS P ropylene glycol (PG) is used as a solvent for intravenous, oral, and topical pharmaceutical preparations, including intra- venous lorazepam. Chronic or large in- gestions of PG have been implicated in the development of hyperosmolar meta- bolic acidosis (1– 8) as well as serious toxicities, including renal dysfunction (4, 9 –12), intravascular hemolysis (13), car- diac arrhythmias (14), seizures (15), and central nervous system depression (16 – 18). The predominant manifestation of PG accumulation is a high anion gap metabolic acidosis with elevated osmol gap, most commonly reported with loraz- epam doses that exceed the upper limit of the recommended lorazepam dosage range (0.1 mg·kg -1 ·hr -1 ) (19). Therefore, patients requiring large doses of loraz- epam for sedation may be at risk for PG accumulation. Each vial of lorazepam (2 mg/mL) contains 0.8 mL (830 mg) of PG/mL, and critically ill patients could receive more than the recommended daily amount of PG (25 mg·kg -1 ·day -1 ) (20). Lorazepam is the medication of choice for maintenance sedation in the critically ill population (19). Currently, there is a paucity of data describing lorazepam- induced PG accumulation and/or toxicity in critically ill adults, with the majority of literature consisting of case reports or case series. Furthermore, interpretation of the existing literature is limited by a) the diversity of patients (age, renal fail- ure, and/or liver failure) in whom PG accumulation has been described; b) in- consistency in the criteria used to define PG accumulation; and c) variability in the lorazepam dose received and reported PG concentrations. Moreover, PG accumula- tion has been described over a wide range of cumulative lorazepam doses (1124 – 7226 mg), serum PG concentrations (12.0 –130.8 mg/dL), and infusion periods (2–24 days) (1– 8, 19). To help elucidate the occurrence of PG accumulation, we undertook a pro- spective, observational study in critically ill patients receiving high-dose loraz- epam (HD-LZ) by continuous infusion. We also wanted to determine whether PG accumulation occurred early in therapy (at 48 hrs) because most of the data de- scribing PG accumulation involve infu- *See also p. 1800. From The Cleveland Clinic Foundation, Department of Pulmonary and Critical Care Medicine (ACA, KM) and Department of Pharmacy (NS, JPG), Cleveland, OH. Address request for reprints to: Jeffrey P. Gonza- les, PharmD, BCPS, Cleveland Clinic Foundation, De- partment of Pharmacy/QQb5, 9500 Euclid Avenue, Cleveland, OH 44195. E-mail: gonzalj@ccf.org Copyright © 2004 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/01.CCM.0000134831.40466.39 Objectives: The primary objective was to evaluate the relation- ship between high-dose lorazepam and serum propylene glycol concentrations. Secondary objectives were a) to document the oc- currence of propylene glycol accumulation associated with continu- ous high-dose lorazepam infusion; b) to assess the relationship between lorazepam dose, serum propylene glycol concentrations, and propylene glycol accumulation; and c) to assess the relationship between the osmol gap and serum propylene glycol concentrations. Design: Prospective, observational study. Setting: Tertiary care, medical intensive care unit. Patients: Nine critically ill adults receiving high-dose loraz- epam (>10 mg/hr) infusion. Interventions: Cumulative lorazepam dose (mg/kg) and the rate of infusion (mg·kg 1 ·hr 1 ) were monitored from initiation of lorazepam infusion until 24 hrs after discontinuation of the high- dose lorazepam infusion. Serum osmolarity was collected at 48 hrs into the high-dose lorazepam infusion and daily thereafter. Serum propylene glycol concentrations were drawn at 48 hrs into the high-dose lorazepam infusion, and the presence of propylene glycol accumulation, as evidenced by a high anion gap (>15 mmol/L) metabolic acidosis with elevated osmol gap (>10 mOsm/ L), was assessed at that time. Measurements and Main Results: The mean cumulative high- dose lorazepam received and mean high-dose lorazepam in- fusion rate were 8.1 mg/kg (range, 5.1–11.7) and 0.16 mg·kg 1 ·hr 1 (range, 0.11– 0.22), respectively. A significant correlation between high-dose lorazepam infusion rate and serum propylene glycol concentrations was observed (r 2 .557, p .021). Osmol gap was the strongest predictor of serum propylene glycol concentrations (r 2 .804, p .001). Propylene glycol accumulation was observed in six of nine patients at 48 hrs. No significant correlation between duration of lorazepam infusion and serum propylene glycol concentra- tions was observed (p .637). Conclusions: Propylene glycol accumulation, as reflected by a hyperosmolar anion gap metabolic acidosis, was observed in criti- cally ill adults receiving continuous high-dose lorazepam infusion for >48 hrs. Study findings suggest that in critically ill adults with normal renal function, serum propylene glycol concentrations may be predicted by the high-dose lorazepam infusion rate and osmol gap. (Crit Care Med 2004; 32:1709 –1714) KEY WORDS: lorazepam; propylene glycol; toxicity; sedation; adults; osmol gap 1709 Crit Care Med 2004 Vol. 32, No. 8