Incidental diagnosis of glycerol kinase deficiency during investigation of hyponatraemia and acute kidney injury

Introduction

Glycerol kinase deficiency (GKD) is a rare X-linked recessive disorder caused by pathogenic variants in the GK gene located on chromosome Xp21. ¹ The GK gene encodes the enzyme glycerol kinase, which catalyses the phosphorylation of glycerol to glycerol-3-phosphate – a critical step in glycerol metabolism. ² The enzyme is predominantly expressed in the kidney, skeletal muscle, liver and small intestine. GKD is characterised by the accumulation of free glycerol in plasma and urine due to reduced or absent activity of the mitochondrial enzyme glycerol kinase. Three phenotypic variants are recognised: infantile (complex), juvenile (intermediate) and adult (benign or asymptomatic). While the more severe forms present in infancy with developmental delay, adrenal hypoplasia, or myopathy, the benign adult form may go undetected for decades, with diagnosis often made incidentally during routine biochemical testing. ³

This case illustrates how assay interference can lead to misdiagnosis and inappropriate treatment of pseudohypertriglyceridaemia. Awareness of GKD and recognition of the biochemical hallmarks of hyperglycerolaemia are crucial to avoid unnecessary pharmacological interventions.

Case presentation

A 52-year-old male with a background of alcohol dependence attended the emergency department following a minor road traffic accident. An incidental finding of chronic urinary retention with bilateral moderate hydronephrosis was identified. Renal function was normal with an eGFR greater than 90. The patient was catheterised and a referral to urology made for further review.

Two weeks later, the patient was admitted to hospital with severe abdominal pain, vomiting and diarrhoea. He was diagnosed with right pyelonephritis, urinary retention and AKI secondary to catheterisation.

Previous past medical history was unremarkable. Hypertension was diagnosed several years prior to this admission alongside an incidental finding of hypertriglyceridaemia (triglycerides 11.4 mmol/L, reference interval <2.0 mmol/L), which was attributed to alcohol excess and Atorvastatin was initiated. A chronic mild hyponatraemia was also observed, again attributed to alcohol excess.

The biochemistry results from the day of admission revealed significant hyponatraemia (sodium 124 mmol/L, reference interval 133–146 mmol/L) and AKI stage 2 (creatinine 198 µmol/L, reference interval 58–110, Abbott Alinity Jaffe method). Blood gas results were unremarkable (Table 1).

OPEN IN VIEWER

Table 1.

Biochemistry results from the day of admission.

Test	Units	Reference range	Day of admission	+2 days	+4 days
Sodium	mmol/L	133–146	124 (L)	113 (L)	120 (L)
Potassium (serum)	mmol/L	3.5–5.3	4.0	4.4	4.9
Urea	mmol/L	2.5–7.8	11.7 (H)	19.3 (H)	27.9 (H)
Creatinine (Jaffe)	µmol/L	58–110	198 (H)	281 (H)	459 (H)
eGFR (MDRD)	ml/min/1.73 m2		31	21	12
Osmolality (serum)	mmol/kg	275–295			326 (H)
Osmolality (urine)	mmol/kg				299
Sodium (urine)	mmol/L				<20
Triglyceride	mmol/L	<2.0			41.1 (H)
Cholesterol	mmol/L				1.64
Total protein	g/L	60–80	68		48 (L)
Albumin	g/L	35–50	24 (L)		15 (L)
Lipaemic-index			0.01	0.00	−0.04
pH (venous)		7.32–7.42	7.37		7.07 (L)
pCO2 (venous)	kPa		4.87		3.6
pO2 (venous)	kPa		2.04		5.2
Actual bicarbonate	mmol/L		20.5		8.0
Standard bicarbonate	mmol/L		20.0		9.0
Base excess	mmol/L		−3.9		−20.8
Oxygen saturation	%		23.1		70.9
Haemoglobin	g/L	120–175	125		107 (L)
Lactate (venous)	mmol/L	0.5–2.2	1.9		0.5
Ketones (POCT)	mmol/L				2.8 (H)
Glucose (POCT)	mmol/L		7.7		5.9

The cause of hyponatraemia was further investigated by requesting serum and urine osmolality 4 days after admission. Upon reviewing the results, the duty biochemist noted an osmolar gap of approximately 50 mmol/kg with no identified hyperglycaemia. Serum triglyceride was appended, and the patient was found to have hypertriglyceridaemia (41 mmol/L) with an inappropriately low Lipaemic index (L-index) of −0.04. (Table 1). Visual inspection of the sample showed a clear plasma, and further analysis found adequate total cholesterol (1.6 mmol/L). On the same day, venous blood gas results confirmed severe anion gap metabolic acidosis (Table 1).

Serum triglyceride was retrospectively appended to other samples stored in the laboratory – a pattern of increasing triglyceride concentration as the eGFR decreased was observed (Table 2). Suspecting pseudohypertriglyceridaemia due to glycerol interference with the triglyceride assay, the duty biochemist deployed several methods to confirm this suspicion as detailed in Table 2. Triglycerides are not excreted by the kidney; therefore, high concentrations in the urine can be attributed to glycerol. ⁴ A urine sample was therefore requested for triglyceride measurement using the serum assay and found to have a high triglyceride concentration (128 mmol/L). Urine organic acids were performed by using gas-chromatography mass spectrometry (GC-MS), with a marked increase in glycerol detected (result not shown). Finally, a subsequent serum sample with a triglyceride concentration of 30.1 mmol/L was referred for free glycerol measurement using GC-MS, the presence of which was confirmed at a concentration of 29.87 mmol/L.

OPEN IN VIEWER

Table 2.

Plasma and urine measurement of triglyceride and glycerol.

Test	Units	Reference range	3 years pre- admission		Admission					Post-admission
			May	June	Day of admission	+2 days	+4 days	+6 days	+8 days	+3 months
Triglyceride	mmol/L	<2.0	11.4 (H)	12.3 (H)	24.4 (H)	28.7 (H)	41.1 (H)	42.6 (H)	30.1 (H)	11.8 (H)
Cholesterol	mmol/L		4.6	3.7			1.6			4.1
HDL-C	mmol/L	>1.0	1.3	2.0
eGFR (MDRD)	ml/min/1.73 m2		78	>90	31	21	12	11	21	57
Lipaemic index			0.04	−0.01			−0.04			0.05
Urine triglyceride	mmol/L								128
Urine organic acids									Glycerol +++
Free plasma glycerol	mmol/L	0.08-0.28							29.87 (H)

These results were suggestive of pseudohypertriglyceridaemia secondary to hyperglycerolaemia. Genetic testing confirmed that the patient was hemizygous for a likely pathogenic variant in the GK gene on the X chromosome (c.552+5G>A, p.[(Lys140_Ser185del; Ser139Leufs*70)]), consistent with a genetic diagnosis of glycerol kinase deficiency. Variant classification was obtained by previous RNA studies performed by Wessex Genomic Laboratory Services. They demonstrated on a previous patient with the same variant which results in two aberrant splice transcripts, the first of which leads to the loss of an exon likely to affect critical gene function.

Discussion

The commercial triglyceride assay is an enzymatic method which hydrolyses the triglyceride in the sample to Free Fatty Acids (FFAs) and glycerol; the free glycerol is then phosphorylated and oxidised, producing hydrogen peroxide which is detected in a colour change reaction in which the absorbance is proportional to the concentration of triglyceride present in the sample. In normal subjects, there is minimal interference from free plasma glycerol (reference interval 0.08–0.28 mmol/L); however in cases with high free plasma glycerol, interference will lead to overestimation of serum triglycerides. ⁵

There are many potential causes of hyperglycerolaemia to consider in cases of pseudohypertriglyceridaemia, including (but not limited to) critical illness (stress), heparin infusions, glycerol-containing medications, intravenous lipids, peritoneal dialysis solutions, excess alcohol consumption and genetic glycerol kinase deficiency. ⁵ Chronic consumption of glycerol-containing alcoholic beverages ⁶ may lead to mild increases in the median serum glycerol in comparison with normal healthy controls, especially if associated with reduced glycerol clearance and reduced glycerol kinase activity due to liver impairment. ⁷ Glycerol intoxication in adults, while less common than in children, may still occur after ingestion of large quantities of glycerol. ⁸ Key biochemical characteristics of glycerol intoxication include hypoglycaemia, anion gap metabolic acidosis, large osmolar gap and pseudohypertriglyceridaemia.

The glycerol kinase enzyme facilitates glycerol phosphorylation for subsequent entry into glycolytic and gluconeogenic pathways. It is broadly expressed in the kidney and small intestine, with the kidney contributing to glycerol reabsorption and metabolism. ² Deficiency in this enzyme leads to accumulation of free glycerol in blood and urine, as evidenced by elevated serum and urinary glycerol in our patient.

It is unclear whether the significantly elevated glycerol levels observed in our patient are cause or effect of the severe metabolic acidosis, AKI and hyponatraemia. It is possible that the hyperglycerolaemia may have contributed to osmotic nephropathy. Attributed to its high molecular weight and osmolality, glycerol can lead to intracellular dehydration, tubular dysfunction and oxidative stress. ⁹ Experimental models demonstrate that glycerol infusion can precipitate AKI through free radical-mediated damage and renal vasoconstriction. ¹⁰ In our case, existing renal compromise from urinary retention and sepsis may have been exacerbated by hyperglycerolaemia. In the absence of direct pathological evidence, such as a kidney biopsy, it is difficult to confirm whether the elevated glycerol levels contributed to the renal injury observed. Conversely, the concurrent illness with renal impairment and metabolic stress may exacerbate the hyperglycerolaemia above the patient’s usual baseline.

Although no specific pharmacological treatment exists for GKD, during periods of acute illness, metabolic decompensation may occur. Supportive measures include avoidance of prolonged fasting and adoption of a low-fat, high-carbohydrate diet to reduce glycerol generation from lipolysis. ¹¹ These strategies aim to prevent further elevations in glycerol and mitigate potential metabolic and renal complications.

Given that the diagnosis has been made incidentally, it is now important to inform our patient, and the GP of his true serum triglyceride levels without interference from endogenous glycerol. Unfortunately, commercial glycerol-blanking assays are not available in the UK. However, subtracting free glycerol from total triglycerides would give an approximate estimation of true triglyceride level of 0.23 mmol/L. Statin therapy, initially prescribed due to presumed hyperlipidaemia, should be re-evaluated. Genetic counselling should be offered, and due to the X-linked recessive inheritance of this genetic condition, any daughters of the patient can be offered genetic screening for carrier status.

This case highlights the critical need for biochemical vigilance in patients with discordant lipid profiles and unexplained metabolic derangements. GKD should be considered in the differential diagnosis of pseudohypertriglyceridaemia, particularly when glycerolaemia coexists with renal dysfunction and osmotic abnormalities. Enhanced awareness and recognition can prevent misdiagnosis, reduce unnecessary pharmacological interventions, and guide appropriate genetic and dietary management strategies.