The Automated Determination of Acetoacetate

in Serum and Urine

Downloaded from https://academic.oup.com/clinchem/article/12/9/606/5674425 by guest on 04 November 2020

Bernard Klein and Morris Okiander

A procedure based on the Rothera reaction is presented for the automated determi-
nation of acetoacetate in serum and urine. The mean serum acetoacetate concen-
tration determined by this procedure for normal human subjects is 0.20 mM; S.D. is
0.059 mM. Recovery experiments demonstrate excellent reliability. The interfer-
ences are identified and their effects evaluated.

FT UMAX KETONEMIA and ketonum-ia are manifestations of marked

metabolic dysfunction arising fi-ommi mci-eased lipid nietabohism, fre-
quently caused by several disease entities, and indicating the iimabihity
of time periplmeral tissues’ to imandle time mci-ease(1 output of these meta-
bolic products by time liver. At time presehit time, it is customam-y to
follow a patient’s progm-ess by qualitative or semiquaimtitative testing
of the seruni or urimie for the presence of acetone and acetoacetate,
usually by time Rotimera reaction (Ketostix or Acetest*).
A procedure which would rapidly amid accurately quantify these
substances in blood and urine would be of considerable assistance in
the recognition and management of this serious clinical situation. Re-
cently, Schilke and Johimson (1) demonstm-ated that the Rothera i-eaction
could be applied to the quaiititative determination of acetoacetate in
urine and plasma. In connection with another imivestigation in progress
in this laboratory, it was felt that an automated version of this reaction
would aid the rapid collection of desired information. Timis paper re-
ports the development of this automated procedure for use with the
AutoAnalyzer.

From the Autolnation Research Laboratory, Veterans Administration Hospital, Bronx,

N. Y. 10468.
Tile authors gratefully acknowledge the capable clerical contributions of Mrs. Jean Meyer.
Reeeived for publication Mar. 8, 1966; accepted for publication June 20, 1966.
*Ames Company, Elkhart, md.

606
Vol. 12, No. 9, 1966 DETERMINATION OF ACETOACETATE 607

Experimental
Reagents

Giycine-pbosphale buffer, pH 8.6 A solution of 60.0 gm. glycine

and 14.7 gm. dibasic sodium phosphate in 800 ml. water is adjusted to
pH 8.6 with 0.1 N NaOH (pH nmeter) and diluted to 1 L.

Downloaded from https://academic.oup.com/clinchem/article/12/9/606/5674425 by guest on 04 November 2020

Sodium nit rof erricyan ide 4% (w/v) in water.
Acetoacetafe stock standard Redistilled ethyl acetoacetate, 3.0 ml.,
is diluted to 100 ml. with 0.2 N NaOH ammd refrigerated (0-8#{176}) for
48 hr. A stock solutioum containing 9.44 mM is prepared by dilutiimg 4 ml.
of the hydrolysate to 100 ml. witim water. The stock standard is stable
for about 1 month.
Acetoaeet ate standards, diluted Working standards are pmepared
by diluting 5.3, 10.6, 15.9, and 21.2 ml. stock standard to 50 ml. This
provides standards containing 1.0, 2.0, 3.0, and 4.0 mM. Working stand-
ards are kept i-efi-igerated and renewed weekly.

Instrumental Considerations

A 1.6-mm I.D., 40-foot coil provides time time delay needed for the
color development. Chart paper ruled in absorbance ummits is preferred.
The chai-t speed is 18 in./hr. An interference filter absombing maximally
at 550 mt is used.

SAMPLER K

H-3
A
BUFFER 2.50

AIR 0.42

SINGLE SODIUM
NITROFERRI- 0.60
40’ COIL
CYAN IDE
AT ROOM
TEMPERATURE

F/C 250

Fig. 1. Flow diagram for AutoAnalyzer in automated determination of acetoacetate.A and

B indicate single mixing coils.
608 KLEIN & OKLANDER Clinical Chemistry

Operating Procedure

The flow diagram for use with time AutoAmialyzer is shown in Fig. 1.
With the sample tube aspirating water, all reagents are pumped, and
when the combined reaction stream passes timrough the flow cell, the
recorder base line is adjusted to 0.01 absorbance. Standards arid speci-

Downloaded from https://academic.oup.com/clinchem/article/12/9/606/5674425 by guest on 04 November 2020

memis are sampled and the absorbance of the purple reactiomm product
imi each ihmstance is measured at 550 m and recorded on the strip chart.
A calibration curve is prepam-ed by plotting the absorbance of the
standard acetoacetate solutioims against concemmtration. A typical cali-
bratiomm curve is shown in Fig. 2. The acetoacetate content of specimens
is obtained by reference to this plot.

Results
In Fig. 3 ai-e showmm recordings obtained by time analysis of 33 un-
selected serum specimens from normal subjects. The mean concentra-
tion is 0.20 mM; S.D. 0.059 mM (2.04 ± 0.6 mg./l00 ml.), expressed as
acetoacetate. This is consistemmt with serum values ranging from 0.08

0.

w
C-)
z
4
0
U)

0.

1.0 2.0 3.0

mM/L ETOACETATE
Fig. 2. Calibration curve and strip chart recording from which it was prepared, showing
absorbance of standard solutions of acetoacetate.
Vol. 12, No. 9, 1966 DETERMINATION OF ACETOACETATE 609

to 0.48 mM (recalculated from (lata based omm acetomme) reported by

Madollia (2), usimig the Botliera reaction, arid by Nadeau (3) aimd
Henry (4), based omi the reactiomi of total acetone (pi-efornmed as well
as that resultimig from thermal (lecompositiomi of acetoacetate) with
alkaline vanillimi. lodometric titration (s) gave 1Iolmal serum concen-
trtion in tli tnnp nlagnitude.

Downloaded from https://academic.oup.com/clinchem/article/12/9/606/5674425 by guest on 04 November 2020

In Fig. 4 and Table 1 are shown time results of recovery experiments.
A nmeami 1-ecovery of 100.4% was obtaimied, amid demomistrated consider-
able analytical reliability.

Discussion

The quantitative determination of acetoacetate (or $-hydroxy-

butyrate) and acetone, which constitute the combined “ketone bodies”
has been the subject of much interest aimd aimalytical effort. Most pro-
cedures actually determined acetone by a variety of means, either alone
or accompanied by the acetone formed by the decomposition of aceto-
acetate. The quantitative determination ot these substances utilized
nonspecific reactions which were complex and technically involved,
especially wheim fractiomiation into acetoacetate (or $-hydroxybutyrate)
and acetone was attempted. Tlmese methods have beemi reviewed, with
critical evaluation, by Henry (4).

The Rothera reactioim with acetoacetate as used in the presemit pro-

cedure yielded colomed solutions wimose absorbances were linear with
concentration (Fig. 2). lt was recognized that other pimysiological con-

- r: f

- -- 4__._____ -‘s--.-- --- - a

-- - --------- ---------------------------.-- I

r -- I

- ---- --.------ ---- __

-- H--- - H- - --- F

- _

Fig. 3. Strip chart recording of analyses of serum specimens from normal human sui)jects.
 Downloaded from https://academic.oup.com/clinchem/article/12/9/606/5674425 by guest on 04 November 2020

C
.
C-C
CC

CC

CC
Cr
C)

C)

C)

C)

C.-
C)

C)

C)

CC

00
C)

C)
C)

C)
0

‘Jo
0.
0

C)

C)
0

0
C”

0
,)

Ut
VoL 12, No. 9. I966 DETERMINATION OF ACETOACETATE 611

Table 1. REcovERY or AcETocr.Tr._&nnI.:n TO P0OLKI) SIKuM

Acetoacetate added Found Cubulated Recovery

Erper. No. (,,i mat e/L. ) ( n ,nofr/L. ) (in ama/p/fl.) ( /,

1 0 0.13 - -

2 0.59 e.72 0.72 100.0

3 1.18 1.31 1.31 100.0

Downloaded from https://academic.oup.com/clinchem/article/12/9/606/5674425 by guest on 04 November 2020

4 1.77 1.92 1.90 101.0
5 2.36 2.50 2.49 100.5
MEAN 100.4

stitueimts of serunm also reacted witim sodium nitroferricvanide and could

contribute eri-oneous inci-enients to the measured absorbance. The
more conmmon intem-fem-ences are pym-uvate, sulfimy(lryl compounds,
pyrrols, creatimline amid acetone.
At pH 8.6, pyruvate and sulfhvdrvl compoumids did imot react. A
highly icteric specimen (Fig. 4, Specinien ii) yielded a slightly turbid
solutiomm aimd gave a value of 0.66 mM, although time patieimt was mmot in
acidosis. This suggested bilirubimm (pyrrol?) interfei-ence which could
be corrected by deterniination of the serunm blank. Creatimmine, at a level
of 100 mg./100 ml. (assumed to be a meaim urinary excretory concen-
tratiomi) gave aim absom-bance of 0.013 (see Fig. 4), offering minimal
interferemice in urine analysis and none in serum analysis.
Acetomme reacted readily with time Rotimera reagent although its molar
extinction was estinmated as 1/37 that of acetoacetate (1). Aim extreme
example is shown in Fig. 4 (Specimneli A). Specimens 7, 17, almd 24 are
repeated analyses of a pooled serum comltailling about 0.5% (v/v) added
acetoime. Time precisiomm attained was witiiimm 1%. Iii severe acidosis,
acetone contributed to time color pioduced amid would he measured also.
Urine Specimens 13-16 aiid 18-22 were collected fi-omn patients in mild
acidosis with glucosum-ia. Specimemi 16 also gave a positive acetone test
with Ketostix. These specimens contained 0.96-1.68 mM and 0.18-1.12
mM acetoacetate respectively. Casual urine specinmens collected from
several normal subjects showed acetoacetate concentrations of 0.08 to
0.64 mM.
Hemolysis produced serious imitem-ference. Teim blood specimens were
divided, and one portion was hemolyzed by mechanical disruption of
the red cells prior to centrifugation and removal of time seium. Time
specimens containing hemoglobin upon analysis yielded absorbances
2-4 times timat given by the clear serunm. Time iimcrements appeared pro-
portional (the proportioim was not quantified) to the degree of visual
hemolysis. This may be time combined result of the release of aceto-
612 KLEIN & OKLANDR Clinical Chems+ry

acetate om- other nitroferricyamiide-reactiiig substamices from the red

cells and the absorhance of oxyhenioglobimi (pH 8.6) at 350 mp.. The
present authors consider the latter coimtribution more significant. It
is clear that hemolyzed plasma or serum specimens cannot be used.

References

Downloaded from https://academic.oup.com/clinchem/article/12/9/606/5674425 by guest on 04 November 2020

1. Schilke, B. E., and Johnson, B. E., A colorimetric method for estimating acetoneetate.
Am. J. Chin. Pathol. 43, 539 (1965)
2. Madonia, J. P., Acetone and acetone bodies in serum. Am. J. Chin. Pathol. 39, 206 (1963).
3. Nadeau, G., The interference of acetone in blood alcohol determinations. Canad. M.A.J. 67,
158 (1952).
4. Henry, B. J., Clinical Chemistry. Principles and Technics. Hoeber, New York, 1964, p. 693.
5. Weichselbaum, T. E., and Somogyi, M., A method for the determination of small amounts
of ketone bodies. J. Biol. Chem. 140, 5 (1941).