Analytical Performance
Interference Effects
Biological media such as blood, plasma and urine are complex mixtures. In addition to the target analyte, they contain proteins, lipids and a variety of electrolytes. These chemicals can originate from a variety of sources. Examples include: (1) endogenous chemicals present in the body at normal levels or at elevated levels associated with certain medical conditions (e.g., albumin, lipids, bilirubin and lactate,) (2) metabolites of drugs prescribed or chemicals introduced during treatment (e.g., plasma expanders and anticoagulants,) and (3) drugs of abuse and other chemicals ingested by patients (e.g., alcohol and nutritional supplements.) These chemicals can interfere with the measurements of a quantitative medical device and can adversely affect its accuracy. In order to ensure the safety of patients and to provide reliable quantitative results to care providers, the effect of these interferences must be tested, quantified and evaluated.
The effects of interferences are typically evaluated at multiple levels of the analyte of interest (usually 2 or 3 clinically important levels) and at multiple levels of the potentially interfering substance. It is recommended that an interference screening test is conducted first using relatively high doses of the potential interference. In an interference screening test, the quantitative medical device is used to analyze pairs of samples. Each pair consists of a control sample and a test sample. The two samples are identical to each other except that the test sample contains the potentially interfering substance and the control sample does not. The difference between the average of the results obtained with the test samples and the average of the results obtained with the control samples (the interference effect) is then evaluated statistically. The number of sample pairs needed depends on the magnitude of the difference being detected and the desired level of statistical confidence and statistical power. Please note that a statistically significant difference (i.e., a significant interference effect) found during the screening test may not be clinically relevant. It must also be noted that the absence of an interference effect means only that there is no interference detected at the levels being tested. There may be an interference effect at other levels. If an interference effect is found to be statistically significant and clinically relevant, further studies are performed using multiple levels of the interfering substance in order to quantify the interference effect as a function of the amount of interfering substance.
Since the number of potential interferences can be quite large, interference testing requires intensive labor and can be time-consuming and costly. This is one of the reasons that specific or highly selective detection schemes are much more preferable than non-selective detection schemes.
Previous Page