The Importance of Endotoxin Testing

The Importance of Endotoxin Testing
Andrew Taylor, ARL Bio Pharma Microbiology Supervisor

In recent years, there have been instances of endotoxin-caused illness resulting from contaminated compounded sterile preparations. In 2015, an incident occurred where seven cases of endotoxin poisoning were related to contaminated glutathione infusions. In this case, patients were symptomatic within 2 hours of administration. Most cases reported fever, rigor, and headache. An investigation into the compounding pharmacy identified multiple issues with aseptic production of compounded products, including minimal compliance with standard quality assurance guidelines. Further testing of the glutathione powder itself revealed high levels of endotoxins.

Bacterial endotoxins are remnants of bacterial cells and are not detected by a sterility test. Potential sources of endotoxins include water, packaging components, equipment, and chemical / raw materials used during compounding of a drug product. Each drug has an allowable endotoxin limit, which specifies the amount of endotoxin that can safely be present. Criteria for acceptance is based on human tolerance and generally defined by the USP monograph, or by a calculation using the patient’s weight, route of administration, and maximum bolus dose. Some firms will also include a “safety factor” in their calculation to deliver a more conservative limit for a product.

It’s important that aseptic production of compounded products include in-process checks to monitor the presence of bacterial endotoxins. The FDA describes expectations in more detail by stating that “firms should take into account aspects of the manufacturing design, including consistency of a manufacturing process, impact of in-process hold times, endotoxin removal steps, and finished product endotoxin specifications” as they design their sampling plans. The plans should be dynamic; in that it may be necessary to adjust sampling when a deviation or an in-process change occurs. It’s also important to define an effective depyrogenation process as endotoxins are not removed by filter or steam sterilization. 

USP <797> requires a Bacterial Endotoxin Test (BET) for each preparation of:

  • Category 2 and Category 3 CSPs compounded from one or more nonsterile component(s)
  • Multiple-dose CSPs

USP <797> also requires a description of the depyrogenation process employed, including the temperature, pressure (if applicable), duration, permissible load conditions for each cycle; and the use of endotoxin challenge vials (ECVs) must be included in the facility’s SOPs. The chapter also states that if a CSP is dispensed or administered before Endotoxin testing results are known, a facility must have procedures in place to Immediately notify the prescriber of a test failure with the potential to cause patient harm.

The Food and Drug Administration (FDA) requires a BET for all 503B outsourcing facility drug products reported to be non-pyrogenic.

Even if it is not mentioned specifically in the regulatory documents, it is important to check for the presence of endotoxins in raw materials, at various points in the compounding process, and in finished products before administering a drug to the patient. Good quality practice requires control or monitoring of the endotoxin levels of contamination at all steps of the compounding process. This should also include a thorough investigation if bacterial endotoxin contamination is found. Appropriately monitoring the presence of bacterial endotoxin in a compounding pharmaceutical environment is essential for your patients’ safety.

For more information on endotoxin testing, contact ARL at 800-393-1595 or


United States Pharmacopeia <85>

United States Pharmacopeia <797>

Guidance for Industry: Pyrogen and Endotoxins Testing: Questions and Answers. U.S. Department of Health and Human Services Food and Drug Administration. June 2012.

Johnstone, T et al. “Seven cases of probable endotoxin poisoning related to contaminated glutathione infusions.” Epidemiology and infection vol. 146,7 (2018): 931-934. doi:10.1017/S0950268818000420