вход по аккаунту



код для вставки
Biomed. Chromatogr. 20: 1386–1389 (2006)
2 November
2006 in Wiley InterScience
( DOI: 10.1002/bmc.712
P. J. Hilton et al.
Unexplained acidosis of malnutrition: a study by
ion-exchange chromatography/mass spectrometry
P. J. Hilton,1* W. McKinnon,1 G. A. Lord,2 J.-M. R. Peron3 and L. G. Forni4
Renal Research Laboratory, Department of Medicine, 4th Floor, North Wing, St Thomas’s Hospital, Lambeth Palace Road, London SE1 7EH, UK
MRC Bioanalytical Science Group, School of Biological and Chemical Sciences, Birkbeck, University of London, Malet Street, London WC1 7HX,
School of Pharmacy and Chemistry, Kingston University, Kingston-upon-Thames, Surrey KT1 2EE, UK
Department of Critical Care Medicine, Worthing Hospital, West Sussex BN11 2DH, UK
Received 24 May 2006; revised 2 June 2006; accepted 5 June 2006
ABSTRACT: Keto-acidosis is usually associated with uncontrolled diabetes and typically poses few diagnostic problems when
presenting as hyperglycaemia, metabolic acidosis and a high anion gap. An emaciated patient suffering from Duchenne Muscular
Dystrophy and volume depletion presented with acidosis of unknown origin. Preliminary investigations appeared to rule out lactic
acidosis, diabetic keto-acidosis and acidosis due to base loss. We have previously reported a technique utilizing liquid chromatography coupled to mass spectrometry (LC-MS) which can be used to characterize the underlying aetiology of acidosis and applied
it to ultrafiltrate derived from a blood sample taken from this patient. The anion profile obtained on the chromatogram showed
elevated levels of acetoacetate and hydroxybutyrate but no evidence of lactic acidosis, nor was the profile typical of that seen in
‘unexplained’ acidosis. We concluded that the patient was suffering from keto-acidosis associated with starvation and dehydration,
the biochemical features being obscured by both the patient’s chronic malnutrition and minimal muscle mass. A combination of
enteral feeding and rehydration led to prompt resolution of the patient’s metabolic acidosis. Copyright © 2006 John Wiley &
Sons, Ltd.
KEYWORDS: acidosis; keto-acidosis; malnutrition; ion exclusion; LC-MS
Diabetic keto-acidosis is one of the most commonly
observed types of acidosis in clinical medicine. Diagnosis of diabetic acidosis usually relies on signs and
symptoms, plus laboratory findings of hyperglycaemia
(blood glucose greater than 250 mg/dL), a low serum
bicarbonate concentration (less than 18 mmol/L), a high
anion gap (greater than 15 mmol/L) and a metabolic
acidosis (Naunheim et al., 2006), the levels of keto- and
hydroxy-acids per se rarely being measured. We were
presented with a 16-year-old male with Duchenne
muscular dystrophy and a long-standing history of
dysphagia who was admitted to hospital with a respiratory tract infection. Upon routine blood examination
the patient was found to be suffering from a metabolic
*Correspondence to: P. J. Hilton, Renal Research Laboratory,
Department of Medicine, 4th Floor, North Wing, St Thomas’s
Hospital, Lambeth Palace Road, London SE1 7EH, UK.
E-mail: [email protected]
Contract/grant sponsor: The Special Trustees for St Thomas’
Hospital, London.
Copyright © 2006 John Wiley & Sons, Ltd.
acidosis. Arterial blood gas analysis on admission
revealed a pH of 7.22, a pCO2 of 3.6 kPa and a standard bicarbonate of 11.2 mmol/L, giving a base excess of
−15 mmol/L. Plasma concentration of sodium was
142 mmol/L,
104 mmol/L,
12 mmol/L and potassium 4.0 mmol/L. Venous blood
analysis revealed a blood urea of 1.3 mmol/L, serum
creatinine of 31 µmol/L and serum albumin of 47 g/L.
Arterial blood lactate was 0.9 mmol/L and the blood
glucose concentration was 7.8 mmol/L. The calculated
anion gap on admission was 26 mmol/L.
The underlying cause of the patient’s high anion gap
metabolic acidosis was unclear given that the initial
observations appeared to rule out both lactic acidosis
and diabetic keto-acidosis as the underlying cause. We
have previously shown that the concentrations of various anions of intermediary metabolism can increase in
patients with acidosis in a manner that may be specific
to the underlying aetiology of the acidosis (Forni et al.,
2005) and have described a method using liquid chromatography coupled with negative ion electrospray
ionization mass spectrometry (LC-ESI/MS) to rapidly
determine the profiles of these anions (McKinnon et al.,
2006). We were requested to use this method to determine the aetiology of the acidosis in this patient.
Biomed. Chromatogr. 20: 1386–1389 (2006)
DOI: 10.1002/bmc
Unexplained acidosis of malnutrition
To determine the aetiology of the underlying acidosis we
used an Agilent HPLC system (Agilent 1100) with an on-line
degasser coupled to a Series 1100 mass spectrometer fitted
with electrospray ionization and operating in ‘negative ion’
mode (Agilent Technologies UK Ltd, Wokingham, Berkshire,
UK). Sample analysis was performed on an Aminex HPX87H Ion Exclusion Column (300 × 7.8 mm, Bio-Rad, Hemel
Hempstead, Herts, UK), which had been washed with water
titrated to pH 3.2 with concentrated hydrochloric acid (HCl)
for 170 h at a flow rate of 0.8 mL/min. Whilst in use, the
column was maintained at 31°C. Post-column, the eluent was
split using a T piece so that approximately one-twelfth of the
flow entered the electrospray nebulizer, the remaining eleventwelfths of the flow being diverted to waste. Prior to entering
the MS source, the HCl in the stream was partially neutralized by the addition of 10 mM ammonium acetate in a 50:50
(v/v) methanol–water mixture at a flow rate of 0.09 mL/h
through a second T piece.
A 5 mL aliquot of the patient’s venous blood was withdrawn into a Kodak (SST II) Vaccutainer. This was chilled
and transported rapidly to the laboratory where it was centrifuged. The resulting plasma was withdrawn and passed
through a 30,000 Da cut-off Amicon centrifugal filter to form
ultrafiltrate. A 20 µL aliquot of this plasma ultrafiltrate was
diluted 1:9 (v/v) with a water–HCl mix at pH 3.2, before
200 µL were fractionated using a mobile phase that ramped
linearly from water–HCl at pH 3.2 to water–HCl at pH 2.6
over 25 min at a flow rate of 0.8 mL/min.
The ultrafiltrate was immediately analysed as previous
work undertaken by other authors (Guth et al., 1999) and
by staff in this laboratory had highlighted the need for
swift sample assay due to an observed rapid decrease in
concentrations of the measured anions in frozen plasma and
an appreciable, though slower, decline in frozen plasma
Retention times of the various acids were determined by
spiking human plasma ultrafiltrate with individual standards
of citric, isocitric, fumaric, pyruvic, succinic, acetoacetic,
lactic, 3OH-butryric, α-ketoglutaric and malic acids. Once the
individual retention times were established, the acids were
chromatographed as a mixture. Standards were prepared
freshly before use. When viewed as selected ion chromatograms at the appropriate (M − H)− m/z value, this allowed
unequivocal identification of the peaks resulting from each
anion. Oxaloacetate could not be measured as a result of
its short half-life (approximately 69 s) in aqueous systems
(Tsai, 1967).
The anion profile obtained from LC-ESI/MS of the
plasma ultrafiltrate obtained was typical of that seen in
keto-acidosis, showing increased levels of acetoacetate
and hydroxybutyrate. Figure 1 shows three-dimensional
offset selected ion chromatograms for acetoacetic acid
and hydroxybutyrate from the patient and a normal
individual. Figure 2 displays three-dimensional selected
ion chromatograms for acetoacetic acid, malic acid
hydroxybutyrate and α-ketoglutaric acid from a normal
individual. Figure 3 shows similar chromatograms to
Fig. 2, but from a diabetic ketoacedotic patient.
Figure 4 shows selected ion chromatograms for
hydroxybutyrate from a typical moderately severe
keto-acidotic diabetic, the patient described in this
paper and a normal individual.
Examination of the ultraviolet absorbance of the
OH-butyrate peak at 210 nm suggested this patient had
an approximate plasma OH-butyrate concentration of
4.5 mmol/L. The results obtained from the LC-MS
technique provided no evidence of either D- or L-lactic
acidosis, nor were the changes in the anion profile obtained typical of that seen in ‘unexplained’ or ‘missing
anion’ acidosis. We concluded that the main cause of
this patient’s metabolic acidosis was a combination of
keto-acidosis secondary to starvation as well as prerenal impairment secondary to volume depletion.
Rehydration caused the serum creatinine to attain the
patient’s normal baseline of 6 µmol/L, this very low
creatinine value resulting from the patient’s minimal
muscle mass and confirming that renal function was
Figure 1. Three-dimensional offset selected ion chromatograms (M − H)− for hydroxybutyrate at m/z 103 and acetoacetic acid at m/z 101 from the patient and a normal
Copyright © 2006 John Wiley & Sons, Ltd.
Biomed. Chromatogr. 20: 1386–1389 (2006)
DOI: 10.1002/bmc
P. J. Hilton et al.
Figure 2. Three-dimensional offset selected ion chromatograms (M − H)− for acetoacetic
acid at m/z 101, malic acid at m/z 133, hydroxybutyrate at m/z 103 and α-ketoglutaric
acid at m/z 145 from a normal individual.
Figure 3. Three-dimensional offset selected ion chromatograms (M − H)− for acetoacetic
acid at m/z 101, malic acid at m/z 133, hydroxybutyrate at m/z 103 and α-ketoglutaric
acid at m/z 145 from a patient with diabetic ketoacidosis.
Figure 4. Three-dimensional offset selected ion chromatograms for hydroxybutyrate
(M − H)− at m/z 103, from (1) a typical moderately severe keto-acidotic diabetic, (2) the
patient described in this paper and (3) a normal individual.
significantly reduced on admission, which also contributed to the acidosis. A combination of enteral feeding
and rehydration led to a prompt resolution of the
acidosis. There was no evidence of impaired glucose
tolerance to suggest diabetes on follow-up of this
Copyright © 2006 John Wiley & Sons, Ltd.
Keto-acidosis is often suspected when an acidotic
patient presents with hyperglycaemia. As this patient
was not overtly hyperglycaemic, keto-acidosis was not
initially considered as a possible cause of the acidosis
Biomed. Chromatogr. 20: 1386–1389 (2006)
DOI: 10.1002/bmc
Unexplained acidosis of malnutrition
observed in this patient. However, he did present with
a metabolic acidosis with an unexplained increased
anion gap. The anion gap represents the difference
between the commonly measured cations (Na+ and K+)
and anions (HCO3− and Cl−) and an elevated value
indicates the presence of an anion not routinely
measured in clinical medicine. Our subsequent analyses
suggested keto-acidosis resulting from malnutrition but
complicated by other features resulting from dehydration. There are increasing reports of keto-acidosis in
malnourished patients (Chen et al., 2006) and these
cases may be difficult to diagnose as a result of the
patient’s atypical presentation. The results obtained
from this patient support the use of the LC-MS technique in the rapid diagnosis of the aetiology of obscure
acidosis. Since the concentrations of 3-OH butyric acid
and acetoacetic acid are seldom measured in clinical
practice, it is questionable whether this component of
this patient’s acidosis would have been determined by
any other method. An advantage of the application of
the LC-MS technique is that it allows the plasma profile of virtually all anions involved in metabolism to be
determined simultaneously
Since the blood sample taken for the determination
of 3-OH butyrate was simultaneous with that used for
other analyses, both the anion gap and base deficit are
greater than that which would be predicted if 3-OH
butyric acid at an approximate concentration of
4.5 mmol/L was the sole cause of the acidosis. Acetoacetate occurs at lower concentrations than 3-OH
butyrate in keto-acidosis and cannot account for more
than a small portion of the missing anions. In the
absence of evidence of other possible candidates, it has
to be assumed that renal failure was the other relevant
factor in the acidosis.
This brief report illustrates that the keto-acid acetoacetic acid and more importantly the hydroxy acid 3OH butyrate may also contribute significantly to the
generation of acidosis in cases of non-diabetic ketoacidosis which can occur in starvation. Application of
the LC-MS technique described allows rapid and easy
assessment of malnourished patients to be undertaken
and enables early detection of perturbations in the
anion profile, allowing appropriate treatment to be
implemented promptly. However, such advanced and
expensive techniques need not necessarily be employed,
as this case illustrates the need for routine urinalysis.
Copyright © 2006 John Wiley & Sons, Ltd.
When presented with a patient with an unexplained
acidosis, performing urinalysis to check for the presence of ketone bodies may help diagnostically. Absent
or low-level ketonuria effectively excludes the situation
we have described. What is more difficult is to predict
how relevant a diagnosis of keto-acidosis is when
a strong positive result is obtained. Such is the sensitivity of this colorimetric test that a strong positive
reaction is seen before a clinically significant ketoacidosis is present. Some further information on this
can be obtained by diluting the urine until the strength
of the colour development begins to decline. It is
important to note that the urine ketone test is affected
only by acetone (which does not contribute to the
acidosis) and aceto-acetic acid (which contributes only
a little). The dominant anion, 3-OH butyric acid, does
not possess a ketone group and does not itself participate in generating a positive colour reaction but
its concentration tends to parallel that of the ketone
bodies that the test detects. Therefore in cases of
unexplained metabolic acidosis, simple tests such as
urinalysis may provide useful clues as to the underlying
The work in this laboratory is supported by The Special
Trustees for St Thomas’s Hospital, London.
Chen T-S, Smith W, Rosenstock JL and Lessnau K-D. A
life-threatening complication of Atkins diet. Lancet 2006; 367:
Forni LG, McKinnon W, Lord GA, Treacher DF, Peron J-MR
and Hilton PJ. Circulating anions usually associated with the Krebs
cycle in patients with metabolic acidosis. Critical Care 2005; 9(5):
Guth HJ, Zschiesche M, Panzig E, Rudolph PE, Jager B and Kraatz
G. Which organic acids does hemofiltrate contain in the presence
of acute renal failure? International Journal of Artificial Organs.
1999; 22: 805.
McKinnon W, Lord GA, Forni LG, Peron J-MR and Hilton PH.
A rapid LC-MS method for determination of plasma anion profiles
of acidotic patients. Journal of Chromatography (B) 2006; 833:
Naunheim R, Jang TJ, Banet G, Richmond A and McGill J. Pointof-care test identifies diabetic ketoacidosis at triage. Academic
Emergency Medicine 2006; 13: 683.
Tsai CS. Spontaneous decarboxylation of oxaloacetic acid. Canadian
Journal of Chemistry 1967; 45: 873.
Biomed. Chromatogr. 20: 1386–1389 (2006)
DOI: 10.1002/bmc
Без категории
Размер файла
168 Кб
Пожаловаться на содержимое документа