The Case

A 42-year-old woman status-post left pneumonectomy for aspergilloma was being treated with oral posaconazole for residual fungal disease. She presented to the outpatient infectious disease (ID) clinic with 2 weeks of pain, erythema, and purulent drainage around her sternotomy site. She was diagnosed with sternal osteomyelitis with multiple areas of fluid collections, felt to be due to contiguous spread from her aspergilloma. After surgical debridement, tissue cultures revealed Aspergillus fumigatus.

Per ID consultation, the patient was switched from posaconazole to intravenous (IV) liposomal amphotericin B (Ambisome) for presumed posaconazole treatment failure. Hours into the IV infusion, the patient developed nausea, vomiting, diaphoresis, and rigors. The next morning, it was discovered that she had been given conventional amphotericin B (Fungizone) at the intended 5 mg/kg liposomal amphotericin B dose. When treating aspergillosis, conventional amphotericin is dosed at a maximum of 1–1.5 mg/kg. Thus, the administered dose represented a 5-fold overdose. The patient developed acute kidney injury, which subsequently subsided, and she ultimately was discharged home with IV caspofungin to finish her course of antifungal therapy.

Multiple factors contributed to this error. First, the resident on the consulting ID team, unfamiliar with the different formulations of amphotericin B, did not distinguish between the two preparations in his progress note. He wrote for "amphotericin B" in his note, while the attending ID note specified "Ambisome" at 5 mg/kg. The primary service (thoracic surgery) inadvertently prescribed conventional amphotericin B (Fungizone) at the ID consult–recommended dose of 5 mg/kg. Second, the electronic prescribing system lacked an alert for the conventional amphotericin formulation that would have notified the prescribing physician that the dose was out of the recommended range. Third, the pharmacist filling the prescription was also unfamiliar with the different amphotericin formulations and did not recognize the toxic dose, either while compounding the medication or sending it to the floor. Finally, the nurse administering the infusion (given during shift change) did not recognize that the patient was having an infusion reaction. Consequently, the infusion was allowed to run to completion. The dosing error was not identified until the next morning.

While not irreversibly harmed, the patient experienced an infusion reaction and acute kidney injury. Her hospital stay was extended for 3 extra days. She was discharged home with a PICC line for prolonged IV caspofungin therapy (a suboptimal antifungal agent for treating aspergillosis), due to her fear of receiving more amphotericin. Had she received the proper medication at the proper dose during her admission, she may have not needed the PICC line at all.

The Commentary

by Jerod Nagel, PharmD, and Eric Nguyen

Adverse events are commonly observed in antifungal treatment for invasive Aspergillus infection. Medication errors in the use of these agents may increase the incidence and severity of these reactions.(1,2) Infusion-related adverse events, including fever, chills, and rigors, occur in 30%–72% of patients treated with amphotericin B deoxycholate (Fungizone).(1) Nephrotoxicity is also common, occurring in up to 53% in patients during amphotericin treatment of Aspergillus infections.(1) While these associations are well known, several measures can decrease the risk of both nephrotoxicity and infusion-related adverse events.

In review of the current case, it is important to note that adherence to evidence-based treatment recommendations might have reduced the risk of posaconazole failure. The Infectious Diseases Society of America guidelines for the treatment of Aspergillus infections recommend voriconazole as first-line therapy.(2) Although posaconazole has excellent in vitro activity against Aspergillus fumagatus, experience in the treatment of invasive Aspergillus infection is limited. In the recent past, other antibiotics have demonstrated excellent in vitro antimicrobial activity, but were associated with higher rates of clinical failure in randomized controlled trials.(3,4) Consequently, until a prospective randomized controlled trial demonstrates successful outcomes for invasive Aspergillus, posaconazole should be reserved as a second- or third-line therapy. It is unknown whether this patient had prior intolerance, unmanageable drug–drug interactions, or clinical failure with voriconazole; if not, that agent was the drug of choice. The formulation of posaconazole utilized is also not stated. Therapy with posaconazole suspension is associated with subtherapeutic serum concentrations compared to posaconazole delayed-release tablets, and treatment failure might have been associated with the use of the suspension and low serum concentrations.(5)

Specific to the use of amphotericin, hospitals should carefully control the ordering, dispensing, and administration process for all amphotericin B products to reduce the likelihood for medication errors. Given the frequency and severity of amphotericin B–related adverse events, safeguards should exist to decrease the risk of inappropriate product selection, dose selection, and infusion rates.

The typical daily dose of amphotericin B deoxycholate in the treatment of disseminated aspergillosis is 1–1.5 mg/kg every 24 hours. In contrast, the usual dose for liposomal amphotericin B is 5 mg/kg every 24 hours. High-dose alerts should be incorporated into the computerized physician ordering entry (CPOE) system to minimize order-entry errors. Similarly, high-dose alerts should be incorporated into the pharmacy computer system to alert pharmacists during the order verification process. Had a warning for amphotericin B deoxycholate for doses greater than 1.5 mg/kg existed in the CPOE system and pharmacy system, this medication error could have been prevented. Finally, nursing could have caught the error prior to administration if smart pumps with dose-related alerts had been available. Smart pumps may also decrease amphotericin B infusions-related adverse events, as these reactions are proportional to the speed and concentration of the infusion.

Although amphotericin B deoxycholate (Fungizone) and liposomal amphotericin B (Ambisome) contain the same active compound, their pharmacodynamics properties are significantly different, resulting in different dosing recommendations and rates of toxicity. Two additional formulations of amphotericin B exist; amphotericin B lipid complex (Abelcet), and amphotericin B colloidal dispersion (Amphotec, Amphocil). Consequently, a systematic approach that treats these four amphotericin B agents as sound-alike, look-alike medications would be prudent.

Look-alike and sound-alike medications are a common cause of medication errors, and creation of systems or algorithms can decrease the risk of medication errors and adverse effects.(6) Strategies to modify the spelling of specific types of medicines, using capitalized first and last letters, as well as integrating the use of upper case lettering in different parts of the name, i.e., "tall-man" letters, are strongly encouraged.(6) Additionally, computerized systems that use pop-ups and barcoding can reduce errors associated with look-alike and sound-alike medications.(6) While barcoding systems are considered the most promising strategy in decreasing look-alike and sound-alike medicine dispensing errors, they do not eliminate problems with confusion during the prescription phase.(6)

An antimicrobial stewardship program with a prior approval mechanism may have also prevented this medication error. Such programs often require a prior approval for antimicrobials that are broad-spectrum, high-cost, or associated with high likelihood for resistance. The prior approval process should involve a discussion of appropriate antimicrobial selection, correct dose and frequency, de-escalation plan, and optimal duration. Antimicrobial stewardship programs that have a step-wise approach to promoting appropriate drug selection, dose, and duration have demonstrated a significant reduction in inappropriate prescribing.(7)

Finally, the incidence of amphotericin B-associated nephrotoxicity and infusion-related adverse effects can be minimized with pre-medication. Administering 500–1000 mL bolus of normal saline before and after amphotericin B infusion can reduce the incidence and severity of nephrotoxicity.(8) Fevers, chills, and rigors are also minimized by providing acetaminophen, diphenhydramine, and/or hydrocortisone 30–60 minutes prior to amphotericin B administration.(8) Linking pre-medication recommendations to amphotericin B orders in CPOE is encouraged.

Ultimately, this is an unfortunate case of toxicity that could probably have been prevented by a systematic process to check the dose and provide high-dose alerts to the ordering prescriber and verifying pharmacist. Implementing steps to reduce errors related to look-alike and sound-alike medications, utilization of smart pumps, and pre-medication with normal saline, acetaminophen, and diphenhydramine would have also decreased the risk or extent of toxicity.

Take-Home Points

• The four available amphotericin B products should be treated as look-alike and sound-alike medications in the CPOE and pharmacy system.
• Amphotericin B adverse effects are frequent and sometimes severe. Implementation of weight-based maximum dose alerts in CPOE and the pharmacy system will reduce dose-related errors.
• Fevers, chills, and rigors are minimized by providing pre-medication with acetaminophen, diphenhydramine, and/or hydrocortisone 30–60 minutes prior to amphotericin B infusion. Linking pre-medication recommendations to amphotericin B orders in CPOE is encouraged.
• The incidence and severity of nephrotoxicity can be reduced by providing 500–1000 mL bolus of normal saline before and after amphotericin B infusion.
• The rate of nephrotoxicity, fevers, chills, and rigors are proportional to the dose and infusion rate. Slowing the infusion and utilizing smart pump may prevent or limit toxicity.

Jerod Nagel, PharmD Clinical Specialist, Infectious Diseases Clinical Assistant Instructor Director Infectious Diseases Residency University of Michigan Hospitals and Health System University of Michigan, College of Pharmacy

Eric Nguyen Medical Student American University of the Caribbean

References

1. Wingard JR, Kubilis P, Lee L, et al. Clinical significance of nephrotoxicity in patients treated with amphotericin B for suspected or proven aspergillosis. Clin Infect Dis. 1999;29:1402-1407. [go to PubMed]

2. Walsh TJ, Anaissie EJ, Denning DW, et al; Infectious Diseases Society of America. Treatment of aspergillosis: clinical practice guidelines of the Infectious Diseases Society of America. Clin Infect Dis. 2008;46:327-360. [go to PubMed]

3. Freire AT, Melnyk V, Kim MJ, et al; 311 Study Group. Comparison of tigecycline with imipenem/cilastatin for the treatment of hospital-acquired pneumonia. Diagn Microbiol Infect Dis. 2010;68:140-151. [go to PubMed]

4. Kollef MH, Chastre J, Clavel M, et al. A randomized trial of 7-day doripenem versus 10-day imipenem-cilastatin for ventilator-associated pneumonia. Crit Care. 2012;16:R218. [go to PubMed]

5. Krishna G, Ma L, Martinho M, O'Mara E. Single-dose phase I study to evaluate the pharmacokinetics of posaconazole in new tablet and capsule formulations relative to oral suspension. Antimicrob Agents Chemother. 2012;56:4196-4201. [go to PubMed]

6. Ostini R, Roughead EE, Kirkpatrick CMJ, Monteith GR, Tett SE. Quality Use of Medicines–medication safety issues in naming; look-alike, sound-alike medicine names. Int J Pharm Pract. 2012;20:349-357. [go to PubMed]

7. Toth NR, Chambers RM, Davis SL. Implementation of a care bundle for antimicrobial stewardship. Am J Health Syst Pharm. 2010;67:746-749. [go to PubMed]

8. Goodwin SD, Cleary JD, Walawander CA, Taylor JW, Grasela TH Jr. Pretreatment regimens for adverse events related to infusion of amphotericin B. Clin Infect Dis. 1995;20:755-761. [go to PubMed]