Glycerol blanking in triglyceride assays: is it necessary?

As evidenced by the article by Jessen et al. in this issue, the question of whether to blank triglyceride analyses for glycerol is still alive and well and, as ever, controversial. The value of glycerol blanking has been debated in several such articles over the years, and, because of complexities involved, will probably be a source of disagreement in the foreseeable future. A laboratory director wishing to make a rational decision about incorporating a glycerol blank in a triglyceride analysis must consider well several issues. Essentially all commonly used clinical methods for determining triglyceride concentration measure the glycerol that has been chemically or enzymatically hydrolyzed from triglyceride, rather than the triglyceride itself. Overestimation of triglyceride occurs when glycerol from sources other than from hydrolyzed triglyceride is also measured. Increased concentrations of interfering glycerol in plasma may arise from a wide variety of sources: metabolic disorders, stress, parenteral nutrition, and use of glycerolcontaining intravenous medications or of blood-collection tubes with glycerol-coated stoppers. In some situations, activated lipases in the blood continue to hydrolyze triglyceride to glycerol and free fatty acids after the blood sample has been taken from the patient, so that the glycerol concentration varies with time in vitro (1). Given these possible sources of error, the routine inclusion of a glycerol blank for all triglyceride analyses seems prudent. However, is such action always necessary? Suitable reagent systems for glycerol-blanked triglyceride measurements are available for most clinical analyzers. Why then are not all triglyceride measurements glycerolblanked as a matter of course? The answer is simple: time and money. The available glycerol-blanking systems function in one of two ways. The “internal blanking” method uses a two-part reagent formulation in such a way that the free glycerol in the sample in a single cuvette is consumed before the addition of the second reagent, which contains lipase to hydrolyze the triglyceride. The liberated glycerol is then measured as the “true triglyceride concentration.” The major drawback of this method is that it requires an analyzer capable of multiple reagent additions. In addition, the magnitude of the glycerol blank is not determined, although this value may be of some clinical interest. The “external blanking” method requires analyses on duplicate samples, one determined with the lipase component and one lacking it; the difference between the two measurements is the “true triglyceride concentration.” This method can be run on any analyzer, but requires two separate reagents and the analysis of two samples, essentially doubling the cost and time required for analysis of a single plasma sample. However, this method does provide a quantitative measure of the glycerol blank. In most samples, particularly from free-living, healthy subjects, the amount of free glycerol is small, generally less than that from a 0.23 mmol/L (200 mg/L) concentration of triglyceride. Some researchers have suggested that, because it represents only a small and generally inconsequential error in the measurement of triglyceride, the glycerol from nontriglyceride sources can be compensated for effectively by subtracting a calculated portion of the measured total triglyceride concentration. Although this method is simple and suffices in most situations, it seems more appropriate to report the measured value and to indicate that it contains an unmeasured amount of glycerol. Otherwise, the recipient of the information may put unwarranted trust in the validity of the value reported. Purists may argue that if triglyceride concentration is to be measured at all, it should be done with the greatest accuracy possible; pragmatists invoke the point of diminishing returns. In the real world, the pragmatista seem to win more often, as evidenced by the incredibly small number of laboratories that perform glycerol-blanked triglyceride analysis. But what may appear to be a denial of good laboratory practice is complicated by the fact that, for the vast majority of triglyceride analyses, the blanked value has no more clinical importance than the nonblanked value, owing to the small amount of glycerol present in most samples. Therefore, from an economic standpoint, blanked measurements may validly be reserved for only those situations in which the glycerol blank is truly required. In such cases, a request for a glycerol-blanked analysis must be made by the ordering physician, or a “reflex” system should be instituted for retesting samples with abnormally high triglyceride concentrations. Although ideally the laboratory would always include a glycerol blank for each triglyceride measurement, such may not be economically feasible, as noted above. When all samples are not routinely blanked for glycerol, the laboratory must determine why the sample was submitted for analysis and must know the population from which it was drawn, if it is to make the correct decision whether to include a glycerol blank. Because such information is generally not readily available to the laboratory, inconsistent application of glycerol blanking will undoubtedly ensue. However, if the information is available, the laboratory director should consider the following points in the decision-making process: Screening for triglyceride status. The current guidelines for the assessment of triglyceride status define normotriglyceridemia as a trigly#{235}eride concentration of 2.82 mmnol/L in fasting plasma, whereas in definite hypertriglyceridemnia the triglyceride concentration exceeds 5.65 mmolJL (2, 3). Triglyceride concentrations between 2.82 and 5.65 mmol/L are considered borderline and require further monitoring of the subject. if the purpose of an ordered test is to assess triglyceride status, a result of 2.82 mmoIJL would correctly classify the subject as normotriglyceridemic, even if the endogenous glycerol content of the sample was very high. Of course, values >2.82 mmol/L would require glycerol blanking for proper diagnosis of abnormalities. In such situations, “reflex” glycerol blanking would be a suitable alternative to routine glycerol blanking for all samples.