Department of Anaesthetics, Poole General Hospital, Poole BH15 2JB, UK*Corresponding author
Accepted for publication: October 2, 2000
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
Br J Anaesth 2000; 86: 43741
Keywords: acidbase equilibrium, acidosis: metabolism, lipid
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Case history |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
She deteriorated markedly in the hours after admission and became unrousable, tachypnoeic, hypotensive, and oliguric. Arterial blood analysis revealed a profound metabolic acidosis (pH 7.0, PO2 24 kPa, PCO2 4.2 kPa, base excess 18 mmol litre1) and her blood sugar had now decreased to an unrecordable concentration. Fifty mililitres of Dextrose 50% was administered with little effect on her conscious level and she was admitted to intensive care.
On admission to intensive care she developed severe circulatory failure (heart rate 146 beat min1, arterial pressure 70/30 mm Hg) and her breathing became progressively laboured. She was given volume resuscitation with colloid but respiratory failure required tracheal intubation and mechanical lung ventilation with a high minute volume. Oesophageal Doppler monitoring (after initial volume resuscitation) confirmed a high cardiac output and high-corrected flow times consistent with a low peripheral vascular resistance. Despite further volume resuscitation her arterial pressure continued to decrease and she did not respond to epinephrine and norepinephrine infusions. Review of her acidbase status revealed a persistent metabolic acidosis (base excess now 20 µmol litre1), an anion gap of 40 mmol litre1 and an arterial lactate of 0.9 mmol litre1. Severe cardiovascular instability with persistent acidosis prompted administration of bicarbonate 50 mM as a resuscitative measure, a bicarbonate infusion and continuous haemodiafiltration. After partial correction of her acidosis her cardiovascular system responded to dobutamine 5 µg kg min1 and norepinephrine 1 µg kg min1.
The diagnosis at this point was still unclear so we investigated the possibility of poisoning. Blood obtained before to resuscitation revealed a plasma ammonia of 52 µmol litre1 and measured plasma osmolality of 318 mOsm kg1 (calculated osmolarity of 298 mOsm litre1). Urine was sent for drug analysis on admission to intensive care and urgent microscopy did not reveal any oxalate crystals (metabolite of ethylene glycol). Ketones were absent in this urine sample. The presence of an increased osmolar gap raised the possibility of methanol toxicity and as a result an i.v. infusion of ethanol was commenced (10 g h1 by i.v. infusion, aiming for a plasma concentration of 12 g litre1). This was discontinued the following day when urinalysis, at our hospital, confirmed that ethanol, methanol, and ethylene glycol were not present. Specialist analysis of the urine was also requested from our regional centre the next morning and this revealed the presence of large quantities of dicarboxylic acids, small amounts of abnormal acylglycines, normal lactate, and mildly elevated ketones. This picture was consistent with the rare metabolic disorder multiple acyl-CoA dehydrogenase deficiency (MADD) and this diagnosis was reinforced later by results from ensuing outpatient investigation. Abnormal organic acid production was limited by a 10% Dextrose infusion (aiming to keep blood glucose between 5 and 10 mmol litre1), carnitine (100 mg kg day1) and riboflavin (300 mg day1) supplements. The rationale for this treatment is discussed below. At this stage the metabolic acidosis had started to resolve with supportive treatment and her inotropic support was decreased accordingly. The metabolic acidosis fully resolved after 3 days and she was weaned from cardiovascular, respiratory, and renal support over the next 2 weeks.
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
When determining the cause of a metabolic acidosis it is useful to calculate the anion gap. A high anion gap (as seen in this case) is most often accounted for by uraemia, ketoacidosis (diabetic, alcoholic), and lactic acidosis. Exogenous acids may also increase the anion gap and in the absence of an alternative, as in this case, poisoning with salicylate, methanol, or ethylene glycol should be considered.2 Salicylate poisoning was excluded but the diagnosis of alcohol poisoning is more difficult. Specialist laboratories are able to measure plasma/urinary alcohol but this is often not available immediately. The diagnosis of alcohol poisoning may be aided by estimation of the osmolar gap ([measured plasma osmolality][calculated plasma osmolarity]).3
To determine the osmolar gap the laboratory must measure plasma osmolality. This can be measured by two techniques: the vapour pressure technique and the depression of freezing point technique. When considering alcohol toxicity it is important to ensure that the measured osmolality is derived by depression of freezing point (the vapour pressure method will cause volatile alcohols to evaporate prematurely generating an inaccurate result).4
The osmolar gap is arbitrarily defined as raised when greater than 10 mOsm and in this case it was increased to 20 mOsm.3 The use of the osmolar gap in alcohol poisoning is controversial for several reasons. A normal gap does not exclude toxicity (especially important with ethylene glycol toxicity where toxic concentrations of the poison exert minimal osmotic influence)5 and a raised gap does not necessarily confirm the presence of alcohol. In the absence of an obvious cause for the acidosis and the presence of a raised osmolar gap we treated our patient for presumed methanol ingestion. Methanol is metabolized to formaldehyde (by alcohol dehydrogenase) and subsequently to formic acid (by aldehyde dehydrogenase). The symptoms and toxicity from methanol are related to the rapid production of the osmotically active aldehyde metabolite. Alcohol dehydrogenase metabolizes ethanol in preference to methanol and an infusion of ethanol will limit the production of formaldehyde and serve as a useful treatment for methanol poisoning. An ethanol infusion was commenced in this case until the laboratory was able to definitively exclude alcohol poisoning.
The correct diagnosis was conclusively reached when urine samples were analysed at a tertiary referral centre for the presence of abnormal organic acids. This was requested the day after admission to intensive care and the results were available the same day. The gas chromatography-mass spectrometry profile proved consistent with a defect in lipid metabolism, MADD. The pathophysiology of this condition is best explained by initial reference to how fatty acids are normally metabolized.
Normal fatty acid metabolism
Fatty acid metabolism provides a major source of energy for cardiac and skeletal muscle. Free fatty acids are bound to albumin in the plasma and diffuse across the cell membrane to enter the cytoplasm. The subsequent pathways involved in normal metabolism are shown in Fig. 1. Short (46 carbon atom) and medium (812 carbon atom) chain fatty acids are able to enter the mitochondria directly for metabolism but long chain fatty acids require a transport mechanism. The mechanism for translocating long chain fatty acids utilizes L-carnitine and an enzyme system. The fatty acid is transported as a fatty acyl-carnitine and when inside the mitochondria carnitine is released leaving fatty acyl-CoA. Metabolism is then possible using ß-oxidation, a four enzyme loop, cycles repeatedly liberating acetyl-CoA, FADH2, and NADH thus shortening the fatty acid by two carbon atoms each cycle.6 The ß-oxidation reactions are directly coupled to coenzyme Q of the electron transfer chain through electron transfer flavoprotein (ETF) and electron transfer flavoproteinubiquinone (co-enzyme Q) oxidoreductase (ETF-QO). These flavoproteins are also required by two similar mitochondrial enzymes isovaleryl-CoA and glutaryl-CoA dehydrogenase and defects in these flavoproteins will therefore affect oxidation of long, medium, and short chain fatty acids and the metabolism of the amino acids lysine and leucine. Alternative pathways for long chain fatty acid oxidation exist and these become heavily utilized when fatty acid metabolism is defective.
|
|
MADD has several documented forms that vary according to the degree of enzyme deficiency. The severe variety normally presents in the neonatal period (glutaric aciduria type II) and is associated with an absence of ETF-QO. Acidosis is profound and hypoglycaemia, coma, and multiple-organ dysfunction normally result in a fatal outcome. A milder variety has been documented (ethylmalonic-adipic aciduria) which tends to present in later life and has a better outcome.9
In MADD, ß-oxidation is defective so any condition that encourages increased fatty acid metabolism (starvation/gastroenteritis) will cause alternative metabolic pathways of lipid metabolism to be utilized and -oxidation of fatty acids will ensue.10 If the alternative enzyme system is overwhelmed then longer chain dicarboxylic acids will also appear. Acetyl-CoA production is impaired and gluconeogenesis is unable to maintain plasma glucose during the stress. This causes severe hypoglycaemic episodes that are typically associated with minimal ketone production. Unlike other acyl-CoA dehydrogenase deficiencies (medium chain being the most common with an incidence of 1:1020 000)9 MADD also produces a severe metabolic acidosis caused in part to the production of dicarboxylic acids but also caused by the defect in electron transport in the mitochondria and associated inability to produce water from hydrogen ions. The inability to utilize fatty acylcarnitine will cause an increase in plasma concentrations (detected by analysis of blood spots on Guthrie cards)11 and a secondary deficiency of carnitine.12 Carnitine is required to transport dicarboxylic acids from the microsome to the mitochondria and this secondary deficiency will encourage further short chain dicarboxylic acids to be produced. Carnitine treatment in this case was effective at eliminating isovaleryl carnitine but also encouraged production of longer chain carnitines that were excreted in urine inadequately.
The milder variant of MADD is also termed riboflavin responsive dicarboxylic aciduria.7 The partial loss of enzyme activity seen in this condition can be minimized by supplementing riboflavin (the cofactor for ETF and ETF-QO). Riboflavin was given in the case presented but it was discontinued later because of limited efficacy. Many features of severe MADD were present in the case presented even though this condition rarely presents in adulthood. The milder variant tends to produce minimal acidosis, a different urinary organic acid profile and a better response to riboflavin. It is likely that this case represents a variant of severe MADD but we are unsure why this case should present at this late stage. It is possible that the combination of an inadequate diet over the week before admission (she had drunk significant amounts of alcohol and eaten poorly while abroad), gastroenteritis, and a urinary tract infection created such a demand on lipid metabolism that the acidosis was precipitated. Before this event her limited ability to metabolize lipids by ß-oxidation and sufficient carbohydrate reserves prevented normal metabolic stresses causing an attack. After discharge she was advised to keep to a regular diet, avoid fasting, limit her alcohol intake, and was maintained on carnitine supplements.
When treating patients with a severe non-lactate acidosis on intensive care it is important to consider and exclude the common causes such as alcohol poisoning. However, if the diagnosis is still elusive then the importance of system support and basic intensive care principles cannot be overemphasized. The patient presented had started to respond with the supportive measures described to treat the acidosis before the diagnosis was confirmed. These measures allow time for more precise investigation and specific treatments to be implemented.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Cohen RD, Woods HF. Disturbances of acidbase homeostasis. In: Weatherall DJ, Ledingham JGG, Warrell DA, eds. Oxford Textbook of Medicine. Oxford: Oxford University Press, 1987; 9.174
3 Trummel J, Ford M, Austin P. Ingestion of an unknown alcohol. Ann Em Med 1996; 27: 36874[ISI]
4 Hoffman RS, Smilkstein MJ, Howland MA, Goldfrank LR. Osmol gaps revisited. Normal values and limitations. J Toxicol Clin Toxicol 1993; 31: 8193[ISI][Medline]
5 Glaser D. Utility of the serum osmol gap in the diagnosis of methanol or ethylene glycol ingestion. Ann Em Med 1996; 27: 3436[ISI]
6 Stryer L. Fatty acid metabolism. In: Stryer L. Biochemistry. New York: WH Freeman and Company, 1981; 383406
7 Walker V. Inherited organic acid disorders. In: Clayton BE, Round JM, eds. Clinical Biochemistry and the Sick Child. Oxford: Blackwell, 1994; 13143
8 Roe CR, Coates PM. Nuclear-encoded defects of mitochondrial respiratory chain, including glutaric acidemia type II. In: Scriver CR, ed. Metabolic and Molecular Basis of Inherited Disease. New York: McGraw-Hill, 1994; 161926
9 Buist NRM, Kennaway NG, Steiner RD. Disorders of mitochondrial energy generation. In: Campbell AGM, McIntosh N, eds. Forfar and Arneils Textbook of Pediatrics. Edinburgh: Churchill Livingstone, 1997; 112831
10 Divry P, David M, Gregersen N, et al. Dicarboxylic aciduria due to medium chain acyl CoA dehydrogenase defect. Aca Paedatri Scand 1983; 72: 9439
11 Stanley CA. Disorders of mitochondrial fatty acid oxidation. In: Nelson W, Behrman R, Kliegman R, Arvin A, eds. Nelsons Textbook of Pediatrics. Philadelphia: WB Saunders, 1996; 3603
12 Sluysmans T, Tuerlinckx D, Hubinont C, Verellen-Dumoulin C, Brivet M, Vianey-Saban C. Very long chain acyl-coenzyme A dehydrogenase deficiency in two siblings: evolution after prenatal diagnosis and prompt management. J Pediatr 1997; 131: 4446[ISI][Medline]