Clinical Science (1979) 51,89-92
A comparison of chromium sesquioxide and c5'Crlchromic
chloride as inert markers in calcium balance studies
R. H E S P , D. W I L L I A M S , M. R I N S L E R
A N D J. REEVE
Divisions of Radioisotopes and Clinical Chemistry, MRC Clinical Research Centre and NorfhwickPark
Hospital, Harrow, Middlesex. U.K.
(Received 31 October 1978; accepted 26 February 1979)
only be eliminated at some cost and inconvenience,
we decided to investigate the use of T r C l , as an
alternative to Cr,O,. Both markers were therefore
given in 14 studies, each of which included three or
more 4- or 6-day faecal collections. Thus we were
able to make a comparison between the two
markers to assess their relative merits.
Summary
1. Chromium sesquioxide (Cr,O,; 1.5 g/day)
and [slCr]chromic chloride [ WrCl, (0.3
pCi/day)l were compared as continuously administered non-absorbed markers for the correction
of faecal recoveries in 14 calcium balance studies
each lasting at least 18 days.
2. The mean recoveries of each, 98.4% for
Cr,O, and 101.9% for 51CrC13,were not significantly different from 100%.
3. The two markers reduced the uncertainty in a
typical 3 x 6 day calcium balance study to a
similar extent (SD = 1.4 mmol/day for Cr,O, and
SD = 1 5 mmol/day for s1CrC13).
4. WrCl, is a convenient and satisfactory
alternative to Cr,O, when the laboratory hazards
associated with estimating the latter cannot easily
be eliminated.
Patients and methods
In the course of conducting calcium balance
studies, continuous 18-day complete stool collections were made on each of 12 female patients with
osteoporosis. None had diarrhoea. Two patients
were studied twice. A 6-day equilibration period,
during which both markers were given, preceded
the collection periods in every case. In most studies
the collections were divided into three 6-day pools
but, in the two repeat studies, respectively six and
seven 4-day pools were obtained. The balance diets
were designed to approximate closely each patient's
daily intake of calcium and magnesium. The
procedure used has been previously described in
detail (Reeve, Tregear & Parsons, 1976).
Each patient received 0.5 g of Cr,O, orally,
together with a gelatin capsule containing 0-1 pCi
of 51CrC13,1.0 mg of CrCl, carrier and 5 0 0 mg of
lactose, three times per day with meals throughout
the equilibration and study periods. Thus the total
intake of W r was normally 7.2 pCi for each
patient (i.e. 0.3 pCi x 24 days) giving a total
radiation dose to the gastrointestinal tract of about
36 mrads, assuming a gut transit time of 2 days
(I.C.R.P., 1971).
-
Key words: calcium, chromic chloride, chromium
sesquioxide, metabolic balance studies.
Introduction
Chromium sesquioxide (Cr,O,) is widely used as a
non-absorbable marker in metabolic balance
studies to correct for incomplete recovery of faeces
and variations in intestinal transit times. Since the
estimation of chromium in faeces has been
associated with the risk of explosions, which can
Correspondence: Mr R. Hesp, Radioisotopes Division,
MRC Clinical Research Centre, Watford Road, Harrow,
Middlesex HA1 3UJ,U.K.
89
90
R. Hesp et al.
Some patients with a previously noted tendency
to constipation were prophylactically given 2-5 g
of Isogel (Allen and Hanburys Ltd) daily, in
divided doses with meals, from the start of their
study. Stool collection and intake of the two
markers ceased simultaneously at the end of each
study.
Each 6-day collection was transferred to a
sample container (external dimensions 15 cm
diameter x 16 cm high) and homogenized after
addition of an approximately equal volume of
distilled water. The "Cr (0.32 MeV gamma rays,
half-life 28 days) content of each bulk stool sample
was measured with a low-background shielded
counter, similar to that described by Cronquist,
MacKenzie & Smith (1973, which incorporated
two vertically opposed NaI (Tl) crystals (10 cm
diameter x 7.5 cm thick) with faces 40 cm apart.
The sample container was placed on a Perspex
platform which was equidistant from both crystals.
The radioactivity count rates from the two crystals
were added and the gamma-ray spectrum was
recorded by means of a multichannel analyser. A
portion of the "Cr dose (1 pCi, approximate
volume 10 ml) was used as reference standard, and
placed on the centre of the Perspex platform for
counting.
Calibration curves, which permit corrections to
be made for the weight of sample, are shown in Fig.
1 for the photopeak (0-27-0.37 MeV) and a wide
band ( 0 . 1 0 4 . 3 7 MeV). Background and counting
sensitivities were approximately 1 . 1 counts/s, and
90 counts s-l ,uCi-' for the photopeak, and 3.3
counts/s and 120 counts s-' PCi-' for the wide
energy band. For the latter, the SE for counting the
standard (counting times 1000 s each for standard
and background) was 0.34%, with a similar SE of
0.34% for the sample, and a SE of & 0.5% for
the weight correction factor (Fig. l), which was
constant for samples of 500 g to 2000 g. Thus
the composite SE for determining the "Cr content
of the sample was k 0.7%. Similar values for SE
were obtained by using photopeak radioactivity
counts.
Immediately before sampling for the chemical
determinations the homogenates were again
vigorously mixed by means of a Silverson homogenizer to ensure an even suspension of the Cr,O,.
About 5 g was then taken and boiled with 25 ml of
concentrated nitric acid to destroy all organic
matter. The digest was then boiled with 20 ml of
concentrated perchloric acid to convert the
chromium into the dichromate. The content of
chromium was measured titrimetrically by the
' '1
05
0
500
1000
1500
Wt. of sample (9)
1
2000
FIG.1. Variation of radioactivity counting sensitivity for
5 T r , with weight of sample, relative to that for a 10 ml
standard. 0 , Wide band ( 0 . 1 0 4 3 7 MeV). 0 ,
Photopeak (0.27-0.37 MeV).
method of Whitby & Lang (1960). Calcium was
determined in the digest by atomic absorption
flame spectrophotometry (Pybus, 1969).
In calculating our results, 'recovery' was defined
as marker outpudmarker intake for the period in
question.
Results
In spite of the precautions taken to ensure regular
stooling, the output of stool ranged from one-third
to twice the amount expected. In Fig. 2, a
comparison of the fractional recoveries of 51Cr and
Cr,O, in the 49 collections is shown. It can be seen
that good, but not complete, agreement was the
rule. The mean recoveries for the 14 complete
balance studies were 98.4% (SEM 2.1) for Cr,O,
and 10 1.9% (SEM 2.1) for WrCl,.
The two markers were further compared by
determining which one minimized the variability in
the daily faecal calcium output in the 12 studies
performed under constant therapeutic regimens.
Since no absolute calcium output rates were
available, individual values for each collection
period were compared with the mean of the three
results for the study to which they belonged. The
mean differences calculated in this way were, of
course, zero, with SD values of 1.89 and 2.15
mmol of Ca/day for Cr,O, and 5'CrCI, respectively. These figures represent reductions in the
variance of period to period calcium recoveries of
80-90%, since the equivalent figure for the
Chromium inert markers in calcium balance
Cr,O, fractional recovery
FIG.2. Relationship between W r recovery and Cr,O,
recovery. Line of identity is shown with 95% confidence
h k s (SD f 0.13). Regression equation of ”Cr fractional
recovery (y) upon Cr,O, fractional recovery (x) is: y =
0-84x+0.19+0.13;r=0.88.
uncorrected recoveries was 5.8 mmol of Ca/day.
The variance ratio calculated from the two
corrected figures was not statistically significant at
the 5% level, so the differences between them may
be due to chance. It was also observed that the
individual deviations from the mean values as
estimated by the two techniques were strongly
correlated ( r = 0.76, P < 0.001).
Only a small part of the differences between the
paired values of the corrected recoveries can be
attributed to known measurement errors. The SD of
the fractional W r recovery as estimated from the
Cr,O, recovery by the regression equation shown
in Fig. 2 represents an uncertainty of 13% of the
oral dose in any given period. The contribution to
this 13% arising from measurement uncertainties
and differences in the radioactivities contained
within individual lCrC1, capsules together account
for only a 2% uncertainty. Because Cr,O, minimizes the differences between serial stool calcium
recoveries at least as well as WrCI,, the methodological uncertainties in the estimation of Cr,O, are
unlikely to be any more significant as a source of
imprecision.
Discussion
For many years Cr,O, has been used as an inert
marker in balance studies (Whitby & Lang, 1960;
Rose, 1964; Davignon, Simmonds & Ahrens,
91
1968) as an alternative to other non-radioactive
markers (Soergel, 1968).
L”Cr20,, produced by neutron activation of
Cr,O, tablets, has been used as an inert marker for
absorption of radioactive tracers (Ditchburn, Smith
& Hayter, 1974) and for the measurements of
gastrointestinal transit times (Hansky & Connell,
1962; Davignon et al., 1968; Bernier, 1969).
Various combinations of inert markers have been
compared (Borgstrom, Dahlqvist, Lundh &
Sjovall, 1957; Rose, 1964; Donaldson & Barreras,
1966; Soergel, 1968; Ditchburn et al., 1974), but
little information is available on the use of 51CrCI,
as a continuous marker, although it has been found
to be almost entirely unabsorbed from the gastrointestinal tracts of a variety of patients (Donaldson
& Barreras, 1966).
The present studies have shown that 51CrC1,
behaves very similarly to Cr20, as a continuously
administered faecal marker for the correction of
faecal outputs in metabolic balance studies. In an
18-day calcium balance study with six duplicate
diets and three 6 d a y stool collections, we previously calculated that calcium intake is estimated
with an uncertainty (SD) of 2.08/\/6 = 0.85
mmol of Ca/day (Reeve et al., 1976). The overall
uncertainty (SD)in an 18-day balance study is thus
d[O.8S2 + (1.89/d3)21 = 1.4 mmol/day
if Cr,O, is used as a faecal marker, and
\/[0.8S2 + (2.15/d3)21 = 1.5 mmol/day
if 51CrCI, is used. That there are residual real
differences in calcium absorption between successive balance periods, as Fisher, Petrie, Tondowski & Joplin (1976) suggest, is supported by
the positive correlation between the paired corrected calcium recoveries.
We conclude that there is little to choose
between the two markers when the aim is to
achieve accuracy and precision in a balance study.
There appear to be small real differences between
the transit patterns of the two markers, but these
are insufficient to affect the choice between them,
which thus depends principally on the convenience
of estimation. The most commonly used technique
for the determination of stable chromium in stools
requires the prior oxidation of chromium to
chromate with perchlorate. Because of the hazard
of explosion of dry perchlorate salts, it is essential
to handle this material in a designated fume
cupboard with non-absorbent surfaces. Also great
care is necessary to flush away drained material to
avoid drying of significant amounts in the drainage
system.
92
R. Hesp et al.
Some investigators will therefore find it more
convenient to use WrCI3. Those who have only
the capacity to count small samples could, as an
alternative to bulk sample counting, ash each whole
sample and count the total ash (or a substantial
portion thereof) in a well-type scintillation counter.
A second option is to homogenize each stool
sample and count a portion before measuring its
calcium content.
Acknowledgment
We thank Mr J. D. Pearson for kindly preparing
the WrCI, capsules. This work formed part of a
study approved by the Hospital Ethical Committee,
to which each patient gave informed consent,
References
BERNIER,T.J. (1969) Study of rate of transit through the
intestine by "CrCI, dilution curve. (Comparison with
carmine test.) Biologie et Gastroenterologie. 2, 17 1-180.
BORGSTROM,
B., DAHLQVIST,
A., LUNDH,G. & SJOVALL,J.
(1957) Studies of intestinal digestion and absorption in the
human. Journal of Clinical Investigation, 36,1521-1536.
CRONQLIIST,
A.G., MACKENZIE,
J. & SMITH,T. (1975) A high
resolution bulk sample counter with variable geometry.
Internafional Journal of Applied Radiation and Isotops, 26,
89-9 1.
DAVIGNON,J., SIMMONDS,W.J. & AHRENS,E.H. (1968)
Usefulness of chromic oxide as an internal standard for
balance studies in formula-fed patients and for assessment of
colonic function. Journal of Clinical Investigation, 47, 121138.
DITCHBURN,R.K., SMITH,A.H. & HAYTER,CJ. (1974) The
assessment of fat absorption in man utilizing single stools or
incomplete faecal collections after oral administration of
radioactive triolein with an unabsorbed radioactive marker.
International Journal of Applied Radiation and Isotopes, 25,
167-176.
DONALDSON,R.M. & BARRERAS,R.F. (1966) Intestinal
absorption of trace quantities of chromium. Journal of
Laboratory and Clinical Medicine, 68,484-493.
FISHER, M.T., PETRIE, A., TONDOWSKI,A. & JOPLIN, G.F.
(1976) A method for performing a calcium balance study in
man and the interpretation of results. Clinica Chimica Acta,
70,297-3 11.
HANSKY,J. & CONNELL, A.M. (1962) Measurement of gastrointestinal transit using radioactive chromium. Gut, 3, 187188.
INTERNATIONAL COMMISSION ON RADIOLOGICAL
PROTECTION
(1971) Protection of the Patient in Radionuclide Investigations, p. 25. I.C.R.P. Publication no. 17. Pergamon Press,
Oxford.
PYBUS,J. (1969) The determination of calcium and magnesium
in serum and urine by atomic absorption spectrophotometry.
Clinica Chimica Acta, 23,309-3 11.
REEVE, J., TREGEAR, G.W. & PARSONS,J.A. (1976)
Preliminary trial of low doses of human parathyroid hormone
1-34 peptide in treatment of osteoporosis. Calcified Tissue
Research, 21 (Suppl.), 469477.
ROSE, G.A. (1964) Experiences with the use of interrupted
carmine red and continuous chromium sesquioxide marking
of human faeces with reference to calcium, phosphorus and
magnesium. Gut, 5,274-279.
SOERGEL, K.H. (Editorial) (1968) Inert markers. Gastroenterology, 54,449452.
WHITEY,L.G.& LANG,D. (1960) Experiences with the chromic
oxide method of faecal marking in metabolic balance investigations on humans. Journal of Clinical Investigation, 39,
854-863.