Limulus Amebocyte Lysate (LAL) assays are widely used for detection and quantification of bacterial endotoxins in pharmaceuticals and medical devices. However, there are only a few studies on the measurement uncertainty of LAL assays. The aim of this work was to identify and quantify the main sources of measurement uncertainty for end point and kinetic-chromogenic LAL assays. Response surface methodology was used to study how the release of p-nitroaniline (pNA) is affected by reaction time and proportion of endotoxin and LAL reagent in end point and kinetic-chromogenic LAL assays, respectively. Increased release of pNA was observed when reaction time was increased. In addition, if different volumes of sample (or endotoxin standard) and LAL reagent are used, the pNA release rate will be affected. These results may be due to the increased interaction between the bacterial endotoxin and LAL-activated enzyme. Final measurement uncertainties (95% confidence interval) were 90-120% and 90-127% of bacterial endotoxin content for end point and kinetic-chromogenic assays, respectively. These values are reasonable for the scope of the method and allow the application of these measurement uncertainties in routine analysis of pharmaceuticals and medical devices.