To the Editor.
—Hypertriglyceridemia, a common finding in diabetes mellitus, has been shown to falsely elevate glycosylated hemoglobin (GHbA1) levels when measured by cation-exchange column chromatography.1,2 Falko et al1 demonstrated this effect in vivo when triglyceride concentrations exceed 1,750 mg/dL (20 mmole/L) and confirmed this by the in vitro addition of lipemic plasma to a control sample. A triglyceride concentration of 9,316 mg/dL (105 mmole/L) produced an apparent eightfold increase in GHbA1.Estimation of GHbA1 levels by affinity chromatography as described by Hall et al3 is becoming increasingly used and it is, therefore, important to know whether the same effect is found.We have used this method to investigate the effects of lipemia on GHbA1 estimations. Glycosylated hemoglobin levels were estimated on grossly lipemic samples (with plasma triglyceride concentrations of 43, 48, and 89 mmole/L, respectively) from three patients, using both washed and