Lipoprotein-cholesterol analysis during screening: accuracy and reliability.

OBJECTIVE

To evaluate the accuracy and reliability of lipoprotein-cholesterol measurements obtained during screening.

DESIGN

Cross-sectional study.

PARTICIPANTS

From November 1989 to January 1990, 154 adults were screened.

MEASUREMENTS

Split venous samples from fasting participants were analyzed for total cholesterol, triglyceride, and high-density-lipoprotein (HDL) cholesterol with screening and standardized laboratory methods. Low-density-lipoprotein (LDL)-cholesterol levels were calculated using the Friedewald equation. Split venous samples from nonfasting participants were analyzed for total cholesterol. Capillary blood samples were analyzed for total cholesterol with the screening method.

MAIN RESULTS

Total cholesterol measurements in screening venous blood samples were 5.4% and 3.8% lower than the laboratory values in samples from fasting and nonfasting participants, respectively. Triglyceride and HDL-cholesterol values in venous samples obtained from fasting participants were, on average, 9.8% and 11.2% lower than the respective laboratory measurements. Screening HDL-cholesterol values varied, differing from the laboratory values by as much as 40% in 95% of participants. In fasting participants, total cholesterol in capillary samples averaged 5.5% higher than in venous samples; in nonfasting participants the capillary samples were 3.1% higher. Screening for either total cholesterol or LDL cholesterol identified 93% of the persons with LDL-cholesterol values of 3.36 mmol/L (130 mg/dL) or higher.

CONCLUSIONS

Total cholesterol can be reliably measured in samples from fasting or nonfasting persons. The values in capillary blood samples were slightly higher than those in venous samples. Screening HDL-cholesterol values were too variable to establish the HDL-cholesterol level reliably. Participants with high LDL-cholesterol levels were identified as accurately by measuring total cholesterol only when compared with calculating the LDL-cholesterol level from total cholesterol, triglyceride, and HDL-cholesterol concentrations.