Cholesterol analyzers

We evaluated eight cholesterol analyzers from eight manufacturers, basing our ratings on the accuracy of the cholesterol results (from both fingerstick and venous samples) compared with the national reference method for cholesterol; repeatability (precision) of the cholesterol results; human factors design; suitability for off-site testing; sample analysis time and throughput; and the effects of sample interferences and electrical disturbances on cholesterol test results. Although many of the units perform additional tests, we did not evaluate performance for these tests; our ratings are based solely on the cholesterol tests performed in this evaluation. We rated seven units Acceptable. We rated one unit Unacceptable because it provided inaccurate and imprecise cholesterol results; also, its reagent system has inherent problems that contribute to human error and, thus, erroneous results. The manufacturer is no longer marketing this unit but is supporting the reagents. We found advantages and disadvantages for each Acceptable unit. To help users choose the appropriate unit for specific applications, we developed a Selection Factors table based on features and applications; this table is presented in the Discussion section. We recommend that all of these units be operated by qualified personnel to minimize the potential for human error. [See "Improved Regulations for Clinical Laboratory Tests."] For further discussion of related issues, see the Clinical Perspective, "Cardiovascular Disease: Reducing the Risk through Cholesterol Screening"; "Cholesterol Readers"; and "The Discontinued Johnson & Johnson CLA 200" in this issue.