Analysis of humidified water and other material samples by two different methods for endotoxin measurement

Project No. BGIA 2053


completed 12/2005


Endotoxins are constituents of the cell wall of bacteria. Depending upon their concentration in the atmosphere, they may trigger acute symptoms of an inhalation fever in humans, e.g. coughing, fever, and muscle and joint pain. Persons exposed to endotoxins for longer periods of time may contract chronic bronchitis. The aerosols which are produced during the use of colonized materials may contain endotoxins. A wide range of natural fibres potentially contaminated in this way are employed within the textile industry. Endotoxins may also be released in aerosol form from bacterially contaminated water. Many plants in the textile industry are equipped with ventilation systems for production reasons. The humidifier water in such installations also constitutes a potential source of endotoxins.

The limulus amoebocyte lysate (LAL) assay is employed as standard for measurement of the endotoxin content of solutions (in this case, humidifier water from ventilation installations in textile plants). In recent years, the in-vitro pyrogen assay has increasingly been available as an alternative. The LAL assay is employed in the microbiological laboratory of the BG Institute for Occupational Safety and Health (BGIA), the IPT assay in the Immunology department of the BG Research Institute for Occupational Medicine (BGFA). The objective of the project was to compare the suitability of the two test methods for the analysis of relevant samples, and if possible to correlate the results.


Partial samples from the same humidifier water were studied by means of both methods in order to permit conclusions regarding the comparability of the two methods for measurement of the endotoxin content of such samples. Based upon these data, the correlation was also examined between the parameters returned by the two methods (LAL assay: endotoxin units/ml solution; in-vitro pyrogen assay: IL 1 or IL 6 [pg/ml]). Partial samples of other material samples (e.g. lining material for safety gloves, allergy test solutions) were likewise studied by means of both methods in order to determine their suitability for these applications.

Humidifier water samples were taken in the plants concerned by authorized staff of the institution for statutory accident insurance and prevention in the textile and clothing industry (TBBG), and dispatched cooled (4 °C) within 24 hours to the two laboratories for the respective tests.

In the BGIA's microbiological laboratory, the LAL assay was performed in the form of a chromogenic-kinetic assay. The reaction time in the assay is dependent upon the endotoxin content of the sample, the pH value, and the temperature of the reactant. The principle of the test is that an enzyme reaction catalyses production of a yellow pigment from a colourless substrate in the test solution, and that the pigment content can be determined photometrically. The rate of pigment formation in this case is directly proportional to the quantity of endotoxin present in the solution. The results are stated in endotoxin units (EUs).

Conversely, the in-vitro pyrogen assay is performed on human blood. The principle of measurement is based in this case upon the detection of cytokines, which are released following incubation for several hours of whole human blood with the sample containing pyrogen. Interleukin(IL)-6 and interleukin(IL)-1 beta are measured in the assay. The whole- blood assay thus permits conclusions regarding the property of the sample under analysis to bring about the release of fever-inducing substances.


The studies were prompted by moves to use the in-vitro pyrogen assay exclusively in future for measurement of the endotoxin content in relevant samples, in place of the LAL assay, which to date has been the standard method. In order for results from previous studies to be included in data analyses where appropriate, the relationship had to be found between results obtained by the two methods on one and the same sample. Results from the two methods cannot be compared directly. Both methods indicated the same trend, however, i.e. low or high concentrations of fever-inducing substances were measured by both, provided the solutions studied contained no interfering substances which had a negative influence upon the photometric analysis method employed with the LAL assay.

The in-vitro pyrogen assay is better suited to the study of material samples, particularly where pigmented samples (e.g. allergy test solutions) or samples which produce turbid solutions when suspended in pyrogen-free water are concerned, as the factors of discolouration or clouding do not influence evaluation of the assay (release of cytokines), whereas they have a substantial influence upon photometric evaluation of the development of a yellow pigment.

Where the test blood originates from different donors, the results are only of limited comparability. This constitutes a constraint upon interpretation of the results obtained by means of the whole-blood assay. In the future, this limitation could however be overcome by the use of cryo-conserved pooled blood which constitutes a mixed sample from several discrete samples from suitable donors. The results of endotoxin activity obtained by means of a LAL-Test can be influenced by several factors. A series of literature references confirm our own experience in this regard. Results are most readily comparable which have been obtained within a single laboratory by means of the same laboratory method and with the use of the same reagents from one manufacturer.

Correlation of the results from the LAL assay and the whole-blood assay from all results obtained to date from comparative studies of atmospheric and material samples (n = 44) indicates that endotoxins represent a significant element within the fever-inducing components of the samples studied. The LAL assay and the whole-blood assay are thus complementary in their conclusions. The results from the whole-blood assay relate to a test system derived directly from the human body. They yield information beyond that on the endotoxin activity alone in the form of sum parameters for quantification of the fever-inducing characteristics. Conversely, no advantage exists in studying samples by means of the whole-blood assay in place of the LAL assay where the endotoxin activity is of explicit interest, since the specific component of this activity cannot be determined by this method in the sum of the fever-inducing reactions measured by the whole-blood assay.

Further informations:

Last Update:

8 Mar 2006


Financed by:
  • Hauptverband der gewerblichen Berufsgenossenschaften (HVBG)
Research institution(s):
  • Berufsgenossenschaftliches Institut für Arbeitsschutz - BGIA
  • Berufsgenossenschaftliches Institut für Arbeitsmedizin - BGFA


Type of hazard:

biological agents


Exposition, Biologische Arbeitsstoffe, Messverfahren

Description, key words:

water samples from humidifiers, air conditioning (AC) systems in textile plants, material samples, lining material for gloves, allergy test solutions, endotoxin content, Limulus amoebocyte lysate assay (LAL), in-vitro pyrogen assay (IPT)