Bacterial Endotoxin Testing (BET) for Medical Device Factory Release

Bacterial endotoxin testing (BET) is a mandatory quality-control step for medical devices that contact blood, the lymphatic system, or cerebrospinal fluid before they are released from the factory. This article explains the applicable standards, sampling strategies, LAL and rFC test methods, acceptance limits, and frequently asked questions related to endotoxin determination for medical devices at factory release.

Technical Overview

Bacterial endotoxins are lipopolysaccharide (LPS) complexes embedded in the outer membrane of Gram-negative bacteria. They are highly heat-stable, survive conventional steam sterilization, and cannot be removed by standard sterilization processes. When endotoxins enter the bloodstream or cerebrospinal fluid, they can trigger fever, inflammatory reactions, septic shock, and in severe cases death. For this reason, medical devices that directly or indirectly contact blood, the lymphatic system, or cerebrospinal fluid must be evaluated for endotoxin contamination before factory release.

The core analytical principle relies on Limulus Amebocyte Lysate (LAL), an extract of horseshoe crab amebocyte cells that contains an enzyme cascade activated specifically by endotoxin. Activation produces either a firm gel (gel-clot technique) or a measurable optical signal (turbidimetric or chromogenic techniques) proportional to the endotoxin concentration. Results are expressed in Endotoxin Units per milliliter (EU/mL) of extract or Endotoxin Units per device (EU/device).

Factory-release endotoxin determination is governed by several recognized standards, including GB/T 14233.2 for injection, infusion, and transfusion equipment in China, Chinese Pharmacopoeia General Chapter 1143, USP Chapter 85 (Bacterial Endotoxins Test), USP Chapter 161 (Medical Devices – Bacterial Endotoxin and Pyrogen Tests), European Pharmacopoeia Chapters 2.6.14 and 2.6.32, and AAMI ST72. Extraction design may reference ISO 10993-12 principles. Manufacturers may perform testing in-house or engage accredited third-party laboratories.

Testing Samples

Sterile Delivery and Administration Devices

Hypodermic syringes, infusion sets, transfusion sets, injection needles, needle-free connectors, and inline filters are high-priority samples because fluids passing through them enter the patient circulation directly. Both finished products and component rinses from production lines are routinely sampled.

Extracorporeal Circulation and Blood-Contacting Devices

Hemodialyzers, blood oxygenators, extracorporeal circulation tubing sets, blood collection and apheresis sets, and cardiopulmonary bypass kits carry large fluid volumes and therefore require strict per-device endotoxin control during factory release.

Catheters, Implants, and Neurosurgical Devices

Central venous catheters, umbilical catheters, intrathecal catheters, shunts, stents, guidewires, and devices contacting cerebrospinal fluid are subject to the strictest endotoxin limits. Implants and solid devices that cannot be filled are evaluated using the rinse or extraction method.

Process Water, Raw Materials, and Components

Purified water and water-for-injection used in washing, rinsing, and clean-in-place cycles, as well as elastomeric closures, polymer resins, and packaging components that contact product fluid paths, are monitored to control endotoxin at the source.

Testing Items

The primary testing item is the bacterial endotoxin content of the device extract or rinse solution, reported against the applicable acceptance limit. Key parameters verified during factory-release testing include:

• Endotoxin concentration of the extract (EU/mL) and total endotoxin burden per device (EU/device)
• Acceptance criteria: not more than 0.5 EU/mL or 20 EU/device for devices contacting the cardiovascular or lymphatic system, and not more than 0.06 EU/mL for devices contacting cerebrospinal fluid, consistent with USP Chapter 161 and GB/T 14233.2 practice
• Positive product control (PPC) recovery within the 50–200% range, confirming absence of interference
• Method suitability, also known as the inhibition or enhancement test, for each product family
• Standard curve linearity (correlation coefficient typically not less than 0.980) for photometric methods
• Lysate labeled sensitivity (lambda) confirmation using control standard endotoxin (CSE)

Testing Methods

Sample Preparation and Extraction

For hollow devices, the internal fluid path is rinsed with a defined volume of bacterial endotoxin test (BET) water, commonly 40 mL or a volume justified by validation. Solid or implantable devices are extracted in pyrogen-free containers, typically incubated at 37 ± 1 °C for a validated period with gentle agitation. All glassware and consumables must be depyrogenated, generally by dry heat at 250 °C for not less than 30 minutes, and the extract is tested without delay.

Gel-Clot Method

The gel-clot technique is a limit test based on the formation of a firm gel when endotoxin activates the LAL cascade. The sample is mixed with equal volumes of reconstituted lysate, incubated at 37 ± 1 °C for 60 ± 2 minutes, and inverted to observe gel integrity. The method is robust, inexpensive, and widely used for routine factory-release screening with lysate sensitivities typically between 0.03 and 0.5 EU/mL.

Kinetic Turbidimetric Method

The kinetic turbidimetric technique measures the time required for the reaction mixture to reach a defined optical density threshold. Reaction time is inversely proportional to endotoxin concentration, quantified against a multi-point standard curve. It offers good sensitivity and objective data recording for high-volume batch release.

Chromogenic Method

The chromogenic technique uses a synthetic peptide substrate that releases a yellow chromophore (p-nitroaniline) upon cleavage by the activated enzyme cascade, measured spectrophotometrically at approximately 405 nm. It is available in kinetic and endpoint formats and provides high sensitivity, wide linear range, and excellent reproducibility.

Recombinant Factor C (rFC) Method

The rFC assay uses a recombinantly produced Factor C protein instead of animal-derived lysate, offering a sustainable alternative that is inherently free of glucan-mediated false activation. It is accepted under European Pharmacopoeia Chapter 2.6.32 and current USP provisions, and is increasingly adopted for routine release testing.

Method Suitability and Quality Control

Each test run includes a negative control (BET water), a positive product control spiked at a mid-range concentration, and reference standard endotoxin. Valid acceptance requires negative control results below the detection limit, PPC recovery within 50–200%, and confirmed lysate sensitivity or standard curve validity. Where interference is observed, the extract is diluted, provided the dilution does not exceed the maximum valid dilution (MVD).

Applications

Endotoxin determination supports quality control and regulatory compliance across the medical device product lifecycle. Typical applications include:

• Routine lot-by-lot factory release testing of sterile single-use devices before shipment
• Validation and periodic requalification of depyrogenation cycles for components and glassware
• Incoming inspection of liquid-contacting raw materials, closures, and tubing
• Ongoing monitoring of purified water and water-for-injection distribution systems
• Supporting sterilization process validation, cleaning validation, and change control activities
• Root-cause investigation of pyrogenic complaints and adverse event follow-up
• Generation of technical data for regulatory submissions to NMPA, FDA, and CE-marking procedures, often in cooperation with accredited third-party laboratories

FAQ

Which medical devices require endotoxin testing at factory release?

Devices that directly or indirectly contact blood, the lymphatic system, or cerebrospinal fluid, such as syringes, infusion and transfusion sets, hemodialyzers, oxygenators, and catheters, require endotoxin testing. Devices with no fluid-path or blood contact generally do not require BET, subject to the risk assessment documented by the manufacturer.

What are the standard acceptance limits?

The widely applied limits are not more than 0.5 EU/mL of rinse solution, corresponding to a total of 20 EU/device, for devices contacting the cardiovascular or lymphatic system, and not more than 0.06 EU/mL for devices contacting cerebrospinal fluid. Contract-specific or pharmacopoeial limits may be stricter and always take precedence.

How are solid devices or implants tested?

Solid devices cannot be filled, so they are extracted by immersing the entire device, or a defined representative portion, in a validated volume of BET water at 37 ± 1 °C. The resulting extract is then assayed by the chosen BET method and the result is converted to EU per device.

What is the difference between the rabbit pyrogen test and the bacterial endotoxin test?

The rabbit pyrogen test is an in-vivo biological assay detecting all pyrogenic substances, whereas BET is an in-vitro assay specific to bacterial endotoxin. BET is faster, more sensitive, more reproducible, and does not use animals, making it the preferred method for routine factory release in most pharmacopoeial frameworks.

What should be done if a sample shows inhibition or enhancement?

Interference is managed by diluting the extract within the maximum valid dilution, adjusting pH to the 6.0–8.0 range, or applying a validated treatment such as filtration that does not remove endotoxin. Suitability must be demonstrated by PPC recovery within 50–200% before routine testing is accepted.

How often must the test method be revalidated?

Method suitability should be repeated whenever the product formulation, materials, manufacturing process, rinsing procedure, lysate lot characteristics, or test method changes significantly. In the absence of changes, periodic confirmation according to the laboratory quality system and applicable pharmacopoeial requirements is sufficient.