Article Text

Download PDFPDF

Pseudo-hypertriglyceridaemia: a measurement artefact due to glycerol kinase deficiency
  1. T A Walmsley,
  2. H C Potter,
  3. P M George,
  4. C M Florkowski
  1. Clinical Biochemistry Unit, Canterbury Health Laboratories, Christchurch, New Zealand
  1. Dr C M Florkowski, Clinical Biochemistry Unit, Canterbury Health Laboratories, Christchurch 8002, New Zealand; chris.florkowski{at}


A man presented with elevated plasma triglycerides and was commenced on fibrate treatment. The triglycerides did not fall and compliance was questioned. The triglyceride elevation was inconsistent with the observed lack of turbidity in the plasma sample. Triglyceride elevation was not confirmed by a different analytical method and lipoprotein electrophoresis showed a normal very low density lipoprotein (VLDL) band pattern. Glycerol kinase deficiency was suspected and was supported by elevated urine glycerol, and confirmed by reduced leucocyte enzyme activity and mutational analysis of the GK gene which showed a novel three base pair deletion. Demonstration of a point mutation also excludes a contiguous gene deletion syndrome.

  • glycerol kinase deficiency
  • leucocyte enzyme activity
  • lipaemia index
  • mutational analysis

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

A 46-year-old man was treated over many years after identification of an abnormal lipid profile with plasma cholesterol of about 6.3 mmol/l (Technicon AAII, Technicon, Tarrytown, New York, USA). Initially, triglycerides were 1.1 mmol/l, within reference limits using Hantzsch Colourimetric method (Technicon AAII). This subsequently rose to around 9.6 mmol/l (BM Hitachi, Hoffmann-La Roche, Basel, Switzerland) employing a non-glycerol blanked assay. This was consistent on repeat measurements.

With hypertension, obesity and impaired glucose tolerance (with random glucose of 10.3 mmol/l), the patient was considered to be at increased cardiovascular risk and was started on fibrate treatment, with gemfibrozil 600 mg twice daily. The triglycerides did not fall and compliance with medication was questioned. The patient had several admissions with atypical chest pain always associated with raised plasma creatine kinase in the order of 270 IU/l (reference range up to 175 IU/l) (Abbott Aeroset, Abbott Laboratories, Abbott Park, Illinois, USA). Plasma troponin T, however, was consistently undetectable (<0.01 μg/l) (Roche Elecsys, Roche, Basel, Switzerland). Subsequently, it was noted that the triglyceride elevation was inconsistent with the lack of turbidity in the plasma sample (fig 1).

Figure 1 Plasma sample from index patient (right) with triglyceride concentration of 9 mmol/l showing lack of expected turbidity, and (left) from control patient with similar triglyceride value and normal glycerol kinase activity.

Chylomicrons and very low density lipoprotein (VLDL) are large triglyceride-rich particles suspended in plasma that scatter light, producing cloudiness or turbidity, which is referred to as lipaemia. An estimate of the degree of lipaemia can be measured as the lipaemic index on the Abbott Aeroset analyser.1 This is an estimation of sample turbidity and would be expected to be raised in the presence of VLDL or chylomicrons along with significant triglyceride elevation. In the present case, it was consistently normal. Lipoprotein electrophoresis was done on buffered (pH 7.5) agarose gel on the semi-automated Hydrasys instrument (Sebia, Norcross, Georgia, USA). The electrophoresis system is capable of resolving chylomicrons, VLDL, low density lipoprotein (LDL), lipoprotein Lp(a) and high density lipoprotein (HDL) in plasma. The sample showed a normal pattern, without chylomicron or VLDL elevation and inconsistent with the reported triglyceride elevation. The triglyceride concentration was within reference limits at 1.0 mmol/l after correcting for a blank glycerol concentration of 9.9 mmol/l (serum glycerol reagents, Randox Laboratories, Crumin, County Antrim, UK). Raised plasma glycerol concentration was confirmed, with the differential diagnosis including glycerol kinase deficiency. Other potential causes include glycerol containing hyperalimentation or heparin therapy, both inapplicable in the present case. Further supporting evidence was provided by measurement of urine glycerol, which was found to be elevated at 71 mmol/l (normal <0.003 mmol/l). Glycerol kinase activity was found to be reduced in peripheral blood leucocytes at 0.8 pmol/min/mg protein (normal range 28–86 pmol/min/mg protein), consistent with glycerol kinase deficiency.

All exons encoding isoform b of the GK gene, including the intron-exon boundaries, were amplified using the polymerase chain reaction. All amplified products were sequenced in both directions on an Applied Biosystems 3130xl Genetic Analyzer (Applied Biosystems, Foster City, California, USA) and the sequence was compared to the Genebank reference NC_000023.9. This showed a novel three base pair deletion, c.1064_1066delCTT, predicted to cause deletion of the highly conserved serine at residue 355 (fig 2). Demonstration of a point mutation also helps to exclude a contiguous gene deletion syndrome.

Figure 2 DNA sequencing gel showing wild type sequence (top panel) from the glycerol kinase gene and patient sequence (bottom panel) showing novel three base pair deletion, c.1064_1066delCTT, predicted to cause deletion of the highly conserved serine at residue 355; p.Ser355del.


Glycerol kinase deficiency is a rarely diagnosed X-linked recessive disorder24 which may occur in isolation or as part of a contiguous gene syndrome with adrenal hypoplasia congenita and/or with Duchenne muscular dystrophy.5 In the isolated form, it can be asymptomatic or associated with intermittent hypoglycaemia and seizures.2 The prevalence of isolated glycerol kinase deficiency is difficult to estimate because of the existence of symptom-free cases,2 although in Sweden with about eight million inhabitants, it is thought that children with at least moderate/severe symptoms are diagnosed and at least 15 cases are documented.2

When triglycerides are measured by most assays, free glycerol is measured after action of specific lipase.6 This method also measures endogenous free glycerol, although the latter is usually insignificant (about 0.12 mmol/l)3 and does not significantly alter the measured triglycerides. Hence most current assays do not include “glycerol blanking”. High concentrations of free glycerol in plasma may result from glycerol contamination of blood collection tubes or glycerol kinase deficiency with “pseudohypertriglyceridaemia”, as in the present case. Although triglyceride assays that employ “glycerol blanking” effectively compensate for the presence of free glycerol, they are not routinely used, although historically such an assay was employed when our patient first presented without triglyceride elevation. Elevated free glycerol, however, can easily be measured in the urine as plasma triglyceride, where it is normally undetectable and values are typically >100 mmol/l.3 In this case, the absence of the expected visible turbidity in the plasma was a seminal observation and was confirmed by the normal lipaemic index on the Abbott Aeroset analyser. This index has been proposed as an automated screen for glycerol kinase deficiency,1 7 although some studies have shown poor correlation between lipaemic index and triglyceride elevation.1 The absence of a chylomicron band or an increased VLDL band on lipoprotein electrophoresis provided further supporting evidence. Endogenous glycerol is normally taken up by cells and either re-forms triglyceride or enters glucose metabolic pathways.8 In either case, it is converted to glycerophosphate by ATP and the action of glycerol kinase (ATP: glycerol phosphotransferase, EC Biological variation of triglycerides in some individuals may also be as high as nearly 50%, with resulting critical difference exceeding 100%9—thus another potential contributory factor to variability in results.

Learning points

  • Glycerol kinase deficiency is a recognised but rare cause of artefactual elevation of plasma triglycerides.

  • Lack of visible sample lipaemia (or positive indices of lipaemia on the laboratory analyser) in the presence of raised plasma triglycerides may suggest the presence of artefactual elevation.

  • An apparent lack of triglyceride reduction on fibrate treatment does not necessarily denote lack of compliance or efficacy.

  • Assay of urine glycerol (as urine triglyceride) may support the presence of glycerol kinase deficiency.

  • Mutational analysis of the GK gene provides a definitive diagnosis, and demonstration of a point mutation in the GK gene helps to exclude a contiguous gene deletion syndrome and also allows carrier testing.

Our case originally had insulin resistance and subsequently developed type 2 diabetes. This is a recognised complication of glycerol kinase deficiency and parallels changes seen in the glycerol kinase deficient mouse.10 In the more complex contiguous gene syndrome, glycerol kinase deficiency co-exists with Duchenne muscular dystrophy and/or adrenal hypoplasia congenita. These are unlikely to be overlooked and there was no clinical evidence of either condition in the present case. The persistently raised creatine kinase, however, suggested some kind of muscle pathology and also exacerbated concerns about possible acute coronary syndrome secondary to the apparently uncontrolled dyslipidaemia. The novel mutation identified in the present case, p.Ser355del, results in the deletion of a highly conserved residue in a control region where subunits interact. A missense mutation, C358Y at an adjacent residue has been shown to cause glycerol kinase deficiency2 and confirms that p.Ser355del is pathogenic. Demonstration of a discrete point mutation in the GK gene2 3 confirms the diagnosis of isolated glycerol kinase deficiency; it also enables identification of asymptomatic female carriers who may wish early diagnosis of their affected male children. The present case shows that apparent elevation of triglycerides may be an artefact of measurement, clinicians may be misled into inappropriate treatment decisions, and the clinical laboratory has the opportunity to make the definitive diagnosis of glycerol kinase deficiency.



  • Competing interests: None declared.

  • Patient consent: Obtained.