Best Way to Dilute Specimens for HDL-Choiesterol Assay To the Editor: When high-density-lipoprotein

(HDL) cholesterol is determined by a precipitation technique, a turbid su-pernate is occasionally seen because of incomplete sedimentation of insoluble lipoproteins (1). This falsely increases values for HDL cholesterol, and, to avoid it, dilution of the specimen is usually recommended. There are two approaches to this dilution (1). One is to dilute the original specimen with an equal volume of isotonic saline before adding the precipitating reagents (pre-precipitation dilution). The second is to dilute the turbid supernate with an equal volume of isotonic saline that contains the same concentration of pre-cipitating reagent as does the superna-tant fluid (post-precipitation dilution). In our laboratory with a phospho-tungstate/MgC12 precipitating reagent and enzymatic quantitation, the pro-precipitation dilution produced HDL cholesterol results that averaged 10% lower than the post-precipitation dilu-tion. This discrepancy was greatest when HDL cholesterol values exceeded 400 mg/L. Because this same difference was not seen with aqueous calibrators, we concluded that the pre-precipitation dilution was probably affected by the change in protein concentration in the diluted specimen. Although turbid supernates most of-ten are associated with low HDL cho-lesterol concentrations (<400 mgfL), dilution is also sometimes used be-cause of insufficient sample. In these cases, the pro-precipitation dilution can result in a decrease in apparent HDL cholesterol that may be clinically significant. An HDL cholesterol value of 810 mg/L for an undiluted specimen was 780 mg/L with the post-precipita-tion dilution but only 620 mg/L with the pro-precipitation dilution, a differ-ence of 19.8%