Estimation of uncertainty in the detection of bacterial endotoxin by gel-clot method

Abstract

Since the publication of ISO 17025:1999, the interest in methods for estimation of the uncertainty in qualitative analysis, such as 'pass/fail', have became more important. The usual form of estimating and informing the uncertainty in this kind of analysis is the use of false-response rates, particularly false-positive and false-negative, determinated from Bayes theorem. The aim of this paper is establish a method for estimation of the uncertainty in the detection of bacterial endotoxins by in vitro Limulus Amebocyte Lysate (LAL) test. Considering the confirmation of LAL sensitivity and the validation of the test, the probability of a false-response corresponds to the sum of false-negative and false-positive result probabilities. From results obtained was verified that the confirmation of LAL sensitivity contributed to the uncertainty in a more significant way (67,6%) than the validation of the test (32,4%). Through this simple procedure and data obtained from the confirmation of LAL sensibility and the validation of the test is possible to obtain a reasonable estimation of the uncertainty of the detection of bacterial endotoxins by gel-clot test.