Background The accurate and precise measurement of triglycerides is important due to the adverse effects associated with hypertriglyceridaemia. Most laboratory methods are based on enzymatic hydrolysis of triglycerides with measurement of the total glycerol. An elevated free glycerol concentration may result in overestimation of triglyceride concentrations. The removal of free glycerol by blanking may therefore be of clinical importance. The aim of this study was to compare the glycerol blanking and non-glycerol blanking triglyceride methods. Methods This was a method comparison study of 1518 samples from both in-patients and out-patients at Charlotte Maxeke Johannesburg Academic Hospital. Triglycerides were measured in each sample using both the blanking and the non-blanking methods. Analytical performance was assessed based on the National Cholesterol Education Program (NCEP) goals. Clinical impact was assessed according to the NCEP Adult Treatment Program III (ATP III) risk classification. Results The method median was significantly higher in the non-blanking compared to the blanking method (1.33 vs. 1.12 mmol/L, p < 0.0001) in all patients. The average bias was above the total allowable error of 15% across all groups. There was a significant change in NCEP ATP III risk classification, with fewer patients classified as normal (67.6% vs. 74.6%, p < 0.0001) with the non-blanking method compared to the blanking method. Conclusions There was a significant error when glycerol blanking for triglyceride determination was not performed. The non-blanking triglyceride method overestimates triglyceride concentrations. This does not only exceed analytical performance goals, but also impacts on patient categorisation and clinical decision making in all patients.
Read full abstract