Abstract Background Cardiovascular diseases (CVDs) are the leading cause of death worldwide, causing over 17 million deaths each year. According to the World Health Organization (WHO), 85% of CVD deaths are due to heart disease and strokes, and over 75% occur in low- and middle-income countries (LMICs). Early detection and treatment are key to preventing consequences of CVD. Point-of-care testing (POCT) devices require minimal laboratory infrastructure and can be used to screen hard-to-reach populations for CVD risk. These devices are increasingly used in traditional care settings including hospitals and primary healthcare, and in non-traditional settings such as pharmacies. Although POCT devices offer fast and convenient testing, the analytical performance of these devices must produce accurate and reliable results, and have technical suitability for use. Several POCT devices for total cholesterol (TC) and other blood lipids are commercially available, however, the operational and analytical performance of these devices is not fully known. Methods Eight POCT devices for blood lipids were used in this study. For the technical assessment, a checklist was developed and applied, outlining the operational functionality and environmental conditions of the devices. Some relevant technical parameters in this checklist included requirements for temperature/humidity and power/voltage, storage conditions, and assay/test time. The analytical performance was assessed for parameters, such as accuracy and imprecision, using paired whole blood and serum donor samples for TC. In addition, these analytical parameters were assessed for serum High Density Lipoprotein-cholesterol (HDL-c) and Low Density Lipoprotein-cholesterol (LDL-c) samples on seven POCT devices. All samples were representative of a clinically relevant concentration range for TC, HDL-c, and LDL-c. Accuracy was compared to reference values obtained by the CDC reference measurement procedures. Results The mean bias for TC ranged among devices from -8.17% to 11.99% for whole blood and -7.89% to 7.02% for serum, respectively. The mean imprecision for TC in whole blood ranged from 1.94% to 11.63%, and in serum from 1.72% to 12.70%. The mean bias for serum HDL-c samples ranged from -7.33% to 17.36%, and for serum LDL-c samples from -20.92% to 5.04%. The mean imprecision for serum HDL-c samples ranged from 2.39% to 24.27% and for serum LDL-c samples ranged from 3.14% to 18.33%. For three of the eight devices, about 25% of the whole blood measurements and 50% of the serum measurements were reported outside of the measurement range and not included in the assessment. Comparing these non-reports with the reference values of these samples indicate that all of the reference values for whole blood and about 80% of the reference values for serum were actually within the measurement range of the device. Conclusions The technical assessment was helpful in building screening capacity in areas with limited laboratory infrastructure. The analytical assessment indicates notable differences in accuracy and precision for lipids measured in whole blood and serum, with serum measurements being more likely within performance requirements used by CDC's standardization programs. Additional studies to assess other devices and/or analytes, and future work to assist manufacturers improve the analytical performance of POCT devices are needed.
Read full abstract