Acute Tubular Necrosis Associated With Propylene Glycol From Concomitant Administration of Intravenous Lorazepam and Trimethoprim-Sulfamethoxazole

Marybeth Hayman, Pharm.D., Edward C. Seidl, Pharm.D., Median Ali, M.D., Khalid Malik, M.D., FCCP


Pharmacotherapy. 2003;23(9) 

In This Article

Case Report

A 46-year-old obese (208 kg) Caucasian man with increasing shortness of breath was transferred directly from his primary care physician's office to the emergency department. Two days earlier, he began to experience a productive cough with yellow sputum, chest tightness, and wheezing. His medical history was notable for asthma, morbid obesity, and an unclear history of narcolepsy, perhaps associated with sleep apnea. He took no drugs and reported no known drug allergies. He did report smoking 1.5 packs of cigarettes/day.

Initial assessment, including chest radiograph, revealed acute respiratory distress and bilateral pulmonary infiltrates. In the emergency department, after blood, urine, and sputum cultures were obtained, the patient began therapy with intravenous antibiotics and steroids. Despite administration of 100% oxygen by nonrebreather mask, the patient was intubated because of respiratory fatigue and was transferred to the medical intensive care unit for further management.

On admission, the patient had a temperature of 38.5°C, heart rate 95 beats/minute, respiratory rate 20 breaths/minute, and blood pressure 146/59 mm Hg. Laboratory tests revealed a white blood cell count of 11 x 103/mm3 (normal range 4.4-11.3 x 103/mm3) with 77.5% neutrophils, 12.9% lymphocytes, 8.7% monocytes, 0.6% eosinophils, 0.3% basophils, and zero bands. Serum electrolytes were within normal limits, blood glucose was 182 mg/dl (70-110 mg/dl), and baseline blood urea nitrogen (BUN) and serum creatinine were 15 mg/dl (7-26 mg/dl) and 0.8 mg/dl (0.7-1.5 mg/dl), respectively. Breathing treatments with albuterol and ipratropium were administered, along with intravenous methylprednisolone 40 mg every 8 hours, and propofol was given for sedation. Clindamycin, azithromycin, and cefotaxime were administered to provide coverage for both community-acquired pneumonia and possible aspiration during intubation.

Over the next 5 days, the patient continued to deteriorate with increasing oxygen requirements. Blood, urine, and sputum cultures remained negative, and an angiographic study ruled out pulmonary embolism. The methylprednisolone dosage was increased to 125 mg every 8 hours. Despite aggressive medical management, there was no change in the patient's pulmonary examination. On day 6, the patient still required a high fraction of inspired oxygen, and he developed diffuse pulmonary infiltrates consistent with acute respiratory distress syndrome. Despite fulminant respiratory failure, he remained hemodynamically stable with normal electrolytes and normal hepatic and renal function.

On day 7, tracheostomy and bronchoscopy were performed. On the evening of day 8, the patient was given an intravenous bolus of lorazepam 6 mg, along with continuous-infusion lorazepam at 4 mg/hour. The lorazepam dosage was titrated to a sedative effect, and propofol was titrated down. On day 9, an infectious diseases specialist was consulted. The patient's therapy was then changed from empiric antimicrobial therapy to scheduled imipenem and a single dose of tobramycin in order to cover ventilator-associated pneumonia pathogens. Over the next 3 days, the patient became increasingly febrile (temperature 40°C) and developed a white blood cell count of 20.5 x 103/mm3 (80% neutrophils, 1% bands). A bronchial wash specimen grew Stenotrophomonas maltophilia and Acinetobacter calcoaceticus. Laboratory data collected on day 13 revealed mild elevations in serum osmolality (312 mOsm/kg [normal range 270-290 mOsm/kg]), lactic acid 2.8 mmol/L (0.5-2.2 mmol/L), and serum glucose 334 mg/dl. Renal and hepatic functions were within normal limits, with serum creatinine 0.8 mg/dl and total bilirubin 0.8 mg/dl (0.2-1.2 mg/dl). On day 14, steroid therapy and an infectious process were thought to be contributing to the patient's hyperglycemia. To ensure tighter blood glucose control, continuous-infusion insulin was started with regular insulin at 5 U/hour, and antibiotic therapy was tailored to intravenous levofloxacin 500 mg/day plus intravenous trimethoprim 320 mg-sulfamethoxazole 1600 mg every 6 hours.

On day 17, 9 days after the start of continuous-infusion lorazepam and 3 days after the start of intravenous trimethoprim-sulfamethoxazole, laboratory tests revealed sudden, severe changes: a rise in serum creatinine to 3.1 mg/dl (Figure 1), nonoliguric renal failure, and marked metabolic acidosis. A nephrologist was consulted. The calculated fractional excretion of sodium was 0.47%. Urinalysis was negative for glucose, ketones, protein, nitrites, and leukocyte esterase. Renal sonography revealed a normal right kidney without hydronephrosis, echogenic stones, or perinephric collections. The left kidney was not assessed because of the patient's habitus. Urine sediment and Hansel stain revealed numerous granular (muddy brown) casts and no eosinophils. The results did not indicate interstitial nephritis and were consistent with ATN. In addition to intravenous lorazepam 9 mg/hour and intravenous trimethoprim 320 mg-sulfamethoxazole 1600 every 6 hours, during the previous 4 days the patient received intravenous furosemide 20 mg twice/day, continuous-infusion insulin, lansoprazole 30 mg/day, and levofloxacin. The patient had no episodes of hypotension. Documentation in the medical record and from a cardiac consultation provided no indications that hypovolemia or hypoperfusion were contributing factors to the patient's renal failure. Subsequently, the cumulative dose of propylene glycol was calculated, and propylene glycol-associated renal toxicity was suspected.

The amount of propylene glycol infused and the patient's serum creatinine levels during his hospital course after intravenous lorazepam was started on day 8. TMP-SMX = trimethoprim-sulfamethoxazole.

On day 18, trimethoprim-sulfamethoxazole was discontinued; continuous venovenous hemofiltration was started that evening. Morning laboratory data collected on day 19 revealed hyperosmolality, serum osmolality 346 mOsm/kg, sodium 140 mEq/L (normal range 135-145 mEq/L), BUN 128 mg/dl, glucose 284 mg/dl, mild hyperlactatemia (serum lactate 2.4 mmol/L), serum propylene glycol level 30 mg/dl, and an osmol gap of 15. These laboratory data were collected simultaneously, and the osmol gap (the difference between measured and estimated osmolalities) was calculated using the following equation: (1.86 x serum sodium [mEq/L]) + (glucose [mg/dl]/18) + (BUN [mg/dl]/2.8) + 9.[16]

The lorazepam infusion was changed to continuous-infusion midazolam, and continuous venovenous hemofiltration was continued. Five days later, on day 24, the patient died from hypoxic respiratory failure. Renal biopsy showed disturbed brush borders of the proximal renal tubules, consistent with resolving ATN.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.