Nephrol Dial Transplant (2008) 23: 1902–1909
doi: 10.1093/ndt/gfm825
Advance Access publication 25 March 2008
Original Article
Reproducibilities and responses to food intake of GFR measured with
chromium-51-EDTA and iohexol simultaneously and independently in
normal subjects
Nicholas J Bird1 , Christina Peters1 , A Robert Michell2 and A Michael Peters1
1
Department of Nuclear Medicine, Addenbrooke’s Hospital, Cambridge and 2 Department of Biochemical Pharmacology, William
Harvey Research Institute, St Bartholomew’s Medical School, London, UK
Abstract
Background. The aim was to evaluate the reproducibility of
glomerular filtration rate (GFR) measured with iohexol and
its response to food in a direct and independent comparison
with Cr-51-ethylenediaminetetraacetic acid (EDTA), and
examine the influence of two different whole body scaling
parameters, body surface area (BSA) and extracellular fluid
volume (ECV).
Methods. Fasting and non-fasting GFR were measured in
20 normal volunteers using Cr-51-EDTA and iohexol, simultaneously injected into opposite arms. In 10, the fasting
study was repeated. Venous samples obtained bilaterally 20,
40, 60, 120, 180 and 240 min after injection were assayed for
indicator injected contralaterally—Cr-51-EDTA by wellcounting and iohexol by X-ray fluorescence. GFR scaled
to BSA was measured from six samples (GFR/BSA6) and
from the last three (GFR/BSA3). GFR scaled to ECV was
calculated as the mean transit time of marker using six
samples (GFR/ECV6) or the last three (GFR/ECV3).
Results. GFR/BSA3 was reproducible (coefficient of variations of 7.4% for Cr-51-EDTA and 7.6% for iohexol). Using
Cr-51-EDTA, GFR/ECV3 (9.1%) and GFR/ECV6 (7.7%)
were as reproducible as GFR/BSA3 and GFR/BSA6 (both
8.1%). However, GFR/ECV3 measured with iohexol had
poorer reproducibility (16.8%). Food resulted in an increase
in scaled GFR of about 5 ml/min but this was statistically
significant only with respect to GFR/BSA (measured with
Cr-51-EDTA or iohexol) and not GFR/ECV.
Conclusions. Measured with Cr-51-EDTA, but not iohexol,
GFR/ECV was as reproducible as GFR/BSA. GFR/BSA,
measured with Cr-51-EDTA or iohexol, but not GFR/ECV,
significantly increased after food.
Keywords: Cr-51-EDTA; extracellular fluid volume;
glomerular filtration rate; iohexol; X-ray fluorescence
Correspondence and offprint requests to: A. M. Peters, Audrey Emerton Building, Royal Sussex County Hospital, Eastern Road, Brighton
BN2 5BE, UK. Tel: +44-1273-523360; Fax: +44-1273-523366; E-mail:
[email protected]
Introduction
Glomerular filtration rate (GFR) is the best measure of
overall renal function. Steady-state urinary clearance of
inulin is the gold standard method for measuring GFR but
it is a cumbersome and difficult technique to perform and
has consequently been replaced as the clinical gold standard
by the measurement of plasma clearance following bolus
injection of alternative filtration markers, most commonly,
at least in Europe, the radioactive compound, Cr-51ethylenediaminetetraacetic acid (EDTA). Plasma clearance
is most accurately measured from multiple blood samples
obtained without waiting for complete mixing of the
indicator throughout its distribution volume (which is
the extracellular fluid space), i.e. the two-compartment
approach. Sampling starting from after complete mixing
(which takes about 2 h) is almost as accurate (onecompartment approach) and is known as the slope–intercept
technique.
In order to compare individuals of different sizes, GFR
is conventionally scaled to body surface area (BSA). This,
however, has been challenged on several grounds [1] and alternative scaling variables suggested to replace it, including
extracellular fluid volume (ECV) [2]. From the viewpoint
of technical simplicity, ECV is an attractive option because
GFR per unit ECV requires the measurement of only the
mean waiting time of the marker in the extracellular fluid
space before filtration; i.e. it is a primary variable in its
own right and, in the one-compartment scenario, is simply
the rate constant of the terminal exponential of the plasma
clearance curve. Since it requires knowledge only of this
rate constant, this technique for measuring scaled GFR is
known as slope-only technique [3]. In contrast, BSA is
not always easy to measure or, in some patient groups,
appropriate, especially the obese [4] and children [5].
In some circumstances, a technically simple but accurate
method of measuring GFR using a non-radioactive indictor
would be advantageous. One such circumstance is the
remote measurement of GFR by courier or postal transportation of patient blood samples to a specialist centre for
analysis, as promoted recently by Niculescu-Duvaz et al.
C The Author [2008]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
For Permissions, please e-mail: [email protected]
Iohexol for measuring GFR
[6]. An obvious candidate for this approach is iohexol
because it is non-radioactive, stable, cheap, easily available and non-toxic in the required doses [7]. Moreover, it
has been validated for the measurement of GFR in several
comparisons with Cr-51-EDTA [7–14] or inulin clearance
[15,16]. The slope-only technique based on iohexol would
be a particularly advantageous approach for a postal service
because the amount of administered indicator would not
need to be known, significantly simplifying the procedure.
To be described as near-ideal, it is essential to show that
a filtration marker is reproducible in its measurement of
GFR and is easily able to show changes in GFR that follow physiological stimuli, such as a meal [17–19]. Whilst
the reproducibilities of GFR based on both iohexol [20,21]
and Cr-51-EDTA [22,23] have previously been separately
determined, head-to-head comparisons of these two indicators have generally been restricted to correlation analysis of
simultaneously measured GFR in patient cohorts with wide
ranges of renal function.
With the aim of further testing the reliability of not only
iohexol for measuring GFR but also the alternative scaling
variable, ECV, we simultaneously and independently compared the reproducibilities of GFR measured with iohexol
and scaled to BSA and ECV against the corresponding values measured with Cr-51-EDTA. Because of the potential
advantages of ECV as a whole body scaling parameter for
GFR, we also examined the reproducibility of GFR scaled
to ECV in both one-compartment and two-compartment
scenarios.
Methods
Subjects
Twenty healthy volunteers [7 male, 13 female; age range
30–59 (median 45); body mass index 18–34 (22) kg/m2
and BSA 1.48–2.09 (1.77) m2 ] with no history of allergy
to iodine-containing contrast agents were recruited. They
were all studied twice, once fasting and once non-fasting.
Ten volunteered for a second fasting study. For each subject,
the two or three studies were completed within 6 weeks.
All measurements were commenced at the same time in
the morning, the non-fasting measurement after breakfast.
Volunteers were instructed to fast from midnight until the
completion of the GFR measurement the following day.
For the non-fasting study, volunteers were asked to follow
their usual breakfast routines. All volunteers gave informed
consent. The study was approved by the local research ethics
committee and by the Administration of Radioactive Substances Advisory Committee of the United Kingdom.
Filtration markers
Cr-51-EDTA (Amersham Health, Bucks, UK) was diluted
with benzyl alcohol containing sodium chloride solution
(1% by volume, benzyl alcohol, 0.75% by weight, NaCl) to
make a stock solution with a concentration of 1 MBq/ml.
Iohexol (Omnipaque 300; Amersham Health, Bucks, UK)
was delivered in 20 ml vials containing 300 mg iodine/ml.
Syringes were weighed before and after drawing up the
1903
markers. Two millilitres of the Cr-51-EDTA stock with an
activity of ∼2 MBq and 19 ml iohexol, containing 5.7 g
iodine, were injected.
Administration of markers
The markers were injected i.v. in close sequence in separate
arms. The time half way through injection was noted as
the time of injection. The Cr-51-EDTA usually took <5 s
to administer but the larger volume of viscous iohexol took
up to 30 s to inject. The volume of administered Cr-51EDTA was measured from the total weight drawn up into the
syringe assuming no post-injection residue and a density of
the Cr-51-EDTA solution of 1 g/ml. The syringes containing
the marker were flushed several times with saline to ensure
complete administration of the weighed amount.
Blood sampling
In addition to a baseline sample, 10-ml blood samples were
taken from both lines nominally at 20, 40, 60, 120, 180
and 240 min after marker administration. Approximately
3 ml of liquid was withdrawn from the line and discarded
prior to the collection of each blood sample. The mid-time
point time of sampling was taken as the sampling time. Line
patency was maintained with heparin-saline. The samples
were transferred from the syringe into heparinized glass
tubes, centrifuged at 800–1000 g for 15 min and plasma
separated.
Sample analysis
Each marker was assayed in samples contralateral to the
side of injection.
51
Cr-EDTA. For each stock solution of Cr-51-EDTA,
1:1000 diluted standards were prepared. Aliquots of
plasma, standards and water (as a blank) were counted
in an automatic gamma counter (Wallac 1480 ‘Wizard 3’,
Turku, Finland) for 1000 s with an energy window suitable
for Cr-51. Appropriate corrections were made for counter
dead-time and background.
Iohexol. Iohexol was assayed by X-ray fluorescence. In
order to calibrate the analyser (Oxford Instruments; LabX3500; www.oxinst.com), a series of seven plasma standards were prepared containing iohexol concentrations of
0, 0.34, 0.69, 1.03, 1.36, 1.72 and 3.39 g/l (Figure 1). For
each sample, a holder was assembled in which the sample
is separated from the irradiation beam and the detector by
a taut piece of 6 µm thick film. Three millilitres of the
sample was transferred to the sample holder immediately
prior to analysis. Once the sample was inside the analyser,
air was displaced from the measuring chamber by helium
gas to prevent emission of characteristic X-rays from argon
in the air or absorption of the characteristic X-rays from
the iodine by argon. The sample was irradiated for 300 s,
the X-ray target being palladium, the voltage 9 kV and the
tube current 100 µA. The average counts per second (cps)
was reported using an energy window that was optimized
for the L-characteristic X-rays from iodine at ∼4–5 keV.
1904
N. J. Bird et al.
Fig. 1. Upper panel: Measurement of known iohexol concentrations by X-ray fluorescence to provide a standard calibration profile. Lower panel:
Histogram of measured sample concentrations (all 300 measurements) compared with standard concentrations used in the calibration profile.
For each sample, the irradiation was repeated three times to
confirm the reproducibility of the measurement, the average of the three being used. For each subject, the cps from
the baseline blood sample from that subject on that day
was used to correct the other sample measurements before
using the calibration to obtain an absolute concentration of
iohexol. The effective precision of the iohexol assay (i.e. in
the range of plasma concentrations encountered; Figure 1)
was estimated to be 4.9%.
Data analysis
Curve fitting. The plasma time–concentration curves for
both markers were bi-exponential between 20 and 240 min.
No samples were obtained before 20 min; otherwise the
clearance curve would have been a triple exponential [24].
A bi-exponential fit to the six-sample curve was therefore
performed using firstly a two-stage curve-stripping procedure and secondly iterative fitting. In the two-stage curve
stripping procedure, the fast and slow exponentials were
stripped in the conventional way. The fast exponential was
then subtracted from the last three sample concentrations
and a second slow exponential fitted to the resulting values. This was then subtracted from the initial three sample
concentrations to give a second fast exponential. The zerotime intercepts of fast and slow exponentials extracted by
these two curve-fitting procedures were denoted by A and
B, respectively, and the corresponding rate constants by α1
and α2 .
Measurement of GFR based on all six samples. GFR was
calculated from A, α1 , B and α2 using the following conventional formula that equates clearance to the quotient of
injected marker divided by the area under the clearance
curve between times zero and infinity:
GFR =
injected activity or dose
.
(A/α1 ) + (B/α2 )
(1)
Using the equation of Du Bois and Du Bois [25] to compute BSA from subject height and weight, GFR was then
scaled to 1.73 m2 to give GFR/BSA6.
Iohexol for measuring GFR
1905
Measurement of GFR based on the final three samples. A
single exponential with intercept B and rate constant α2 (to
be distinguished from B and α2 based on six-sample fitting)
was fitted to the last three sample points. The reciprocal of
B is the one-compartment distribution volume, Vd . Slope–
intercept GFR was then calculated as the product of α2 and Vd and scaled to 1.73 m2 . It was corrected for the onecompartment assumption using second-order polynomials
(of the form y = A + Bx + Cx2 ) based on the relations
between one-compartment and two-compartment GFR to
give GFR/BSA3. The formula for iohexol (in which A = 0,
B = 1.0022 and C = −0.00131) was derived from our own
database of 110 six-sample iohexol clearances. With respect
to Cr-51-EDTA data, three-sample GFR was corrected for
the one-compartment assumption using the second-order
polynomial formula of Brochner-Mortensen [26].
Measurement of mean marker residence time in the ECV
using six samples. Any individual molecule of indicator
within the extracellular fluid (assumed to be the same as the
indicator’s distribution volume) will wait a certain amount
of time before undergoing glomerular filtration. The mean
waiting time of all the molecules, i.e. transit time (T), can
be calculated from the following equation [27]:
T =
(A/[α1 ] ) + (B/[α2 ] )
.
(A/α1 ) + (B/α2 )
2
2
(2)
The reciprocal of T is the rate at which the ECV is ‘turned
over’ by GFR and is therefore a measure of GFR that is
already scaled to ECV. This scaled version of GFR is denoted by GFR/ECV or, as it was based on six samples,
GFR/ECV6. Multiplication of GFR (ml/min) by T (min)
gives ECV (ml). ECV was then scaled to a BSA of 1.73 m2 .
Measurement of mean marker residence time in the ECV
using the last three samples. The variable α2 (with units
of min−1 ) is also an estimate of GFR/ECV (ml/min/ml =
min−1 ), which, as it was based on the three samples between
2 and 4 h, was denoted by GFR/ECV3. As for GFR/BSA3,
GFR/ECV3 also needs correcting for the one-compartment
assumption. For iohexol data, GFR/ECV3 was therefore
corrected using a second-order polynomial derived from
our own database of 110 iohexol clearances in which A =
0, B = 1.0526 and C = 5.17. With respect to Cr-51-EDTA,
correction for the one-compartment assumption was based
on α2 using a formula in which A = 0, B = 1 and C = 15.4,
similar to that reported by Peters et al. [28].
Statistics
Reproducibility. Estimates of repeatability were made
from within-subject variance for the volunteers who had
repeat fasting measurements. The within-subject variance
(CV) was calculated from the residual mean square from
one-way analysis of variance [29]:
[(x1 − x2 )2 /2n]
CV =
(x1 + x2 )/2n
0.5
(3)
where x1 and x2 refer to first and second measurements and
n is number of paired measurements.
Comparison of repeatability between different methods
and markers was made from the F-test on the variance of
the differences between repeat measurements.
The standard deviation (SD) of the differences between
repeat measures of GFR approximates to [(SD of the real
differences)2 + 2(measurement error)2 ]0.5 .
Effect of food intake. The difference between fasting and
non-fasting GFR was evaluated for all measures of GFR
using the paired Student t-test. For volunteers with two
fasting measurements, the difference was based on the first
of the two.
Results
Curve stripping versus iterative fitting
Six-sample GFR based on iterative fitting was less reproducible than six-sample GFR based on two-stage curve
fitting, whether scaled to BSA or ECV (results not shown).
Six-sample GFR and ECV values are therefore reported
only for two-stage curve stripping.
Scaling ECV
ECV showed a marginally better correlation with BSA than
with its more traditional scaling parameter, body weight
(results not shown). Mean fasting ECV/BSA was 13.5
(SD 1.0) L for Cr-51-EDTA and 13.2 (1.5) L for iohexol.
Mean non-fasting ECV/BSA was 13.7 (1.5) for Cr-51EDTA and 13.6 (1.7) for iohexol, not significantly different
from fasting values. To obtain GFR/ECV numerically
comparable with GFR/BSA, GFR/ECV (measured with
either Cr-51-EDTA or iohexol) was therefore multiplied
by 13 500.
Reproducibility
Measured with Cr-51-EDTA, there were no significant
differences between the reproducibilities of GFR/BSA3,
GFR/BSA6, GFR/ECV3 and GFR/ECV6 (Table 1). For
iohexol, however, the within-subject variances of GFR/
ECV3 and GFR/ECV6 were both significantly higher
(P < 0.05) than GFR/BSA3, but not GFR/BSA6. The
within-subject variances of GFR/ECV3 and GFR/ECV6
based on iohexol were both significantly higher than the
corresponding variances recorded with Cr-51-EDTA, but
there were no significant differences between the two
markers with respect to GFR/BSA3 and GFR/BSA6.
The changes between the two separate, simultaneous and
independent measurements of fasting GFR values based
respectively on Cr-51-EDTA and iohexol correlated with
each other (Figure 2), although not significantly with
respect to GFR/ECV3, indicating that the variation between the two measurements partly reflects real changes in
filtration function.
Effect of a light meal
Food intake consistently resulted in small increases in GFR
values but these were significant only when scaled to BSA
(Table 2). The corresponding food-induced changes in GFR
1906
N. J. Bird et al.
Table 1. Reproducibilities of fasting GFR/BSA (ml/min/1.73
subjects with Cr-51-EDTA and iohexol
Cr-51-EDTA
GFR/BSA3
GFR/BSA6
GFR/ECV3
GFR/ECV6
Iohexol
GFR/BSA3
GFR/BSA6
GFR/ECV3
GFR/ECV6
m2 )
and GFR/ECV (ml/min/13.5 l) measured on two separate occasions in 10 normal
Mean
Mean difference
between replicates
SD of
differences
88.2
89.9
88.7
92.4
0.4
2.4
5.8
5.4
9.7
10.6
10.4
8.9
6.5
7.3
8.1
7.1
7.4
8.1
9.1
7.7
83.8
84.4
87.5
87.8
0.6
1.9
5.9
5.4
9.5
12.9
21.1a
17.0a
6.4
8.7
14.8
12.0
7.6
10.4
16.8b
13.2b
a Significantly different compared with corresponding variable
b Significantly different compared with GFR/BSA3 (iohexol).
Table 2. Fasting (first in those having two fasting studies) and non-fasting
values of GFR/BSA (ml/min/1.73 m2 ) and GFR/ECV (ml/min/13.5l) measured in 20 normal subjects with Cr-51-EDTA and iohexol
Mean increase after
food intake
CV%
measured with Cr-51-EDTA.
Fig. 2. Relations between corresponding individual fractional changes
in GFR values simultaneously measured with Cr-51-EDTA and iohexol
between the first and second fasting measurements in 10 normal subjects.
Mean fasted
Within subject
standard deviation
Fig. 3. Relations between corresponding individual fractional changes
in GFR values simultaneously measured with Cr-51-EDTA and iohexol
between the non-fasting and fasting measurements in 20 normal subjects
(the first in the 10 normal subjects who had two fasting measurements).
measured respectively with Cr-51-EDTA and iohexol correlated significantly with each other but only with respect
to GFR/BSA, not GFR/ECV (Figure 3).
P value
Discussion
Cr-51-EDTA
GFR/BSA3
GFR/BSA6
GFR/ECV3
GFR/ECV6
Iohexol
GFR/BSA3
GFR/BSA6
GFR/ECV3
GFR/ECV6
87.9
89.3
88.4
92.2
4.9
4.6
2.9
3.2
0.0045
0.0067
0.086
0.052
85.1
83.8
83.3
87.8
3.8
5.7
5.6
5.0
0.023
0.011
0.10
0.092
Many previous studies assessing the accuracy of GFR measurement against a gold standard have not performed the
measurements in a truly independent fashion. The novelty
of this study is that complete independence was assured
with respect to both the reproducibility and effect of food
studies. Although this was a cohort of normal volunteers,
average GFR was surprisingly low. We have no reason,
however, to suspect any measurement errors as the mean
Iohexol for measuring GFR
ratio of slope–intercept GFR, e.g. given by the two indicators in all 50 paired studies was 1.037 (SD 0.063).
1907
lation [37], suggesting that for a light meal, with small
increases in GFR, primary changes in ECV play a more
important role in the GFR increment. This would explain
why the increases in GFR/ECV were not significant.
Reproducibility
Reproducibility studies of Cr-51-EDTA and iohexol are
scarce, and we could find no direct comparisons between
them. Since chronic kidney disease (CKD) is slowly progressive, reproducibility is a key element for any clinical
measure of its rate of progression. There is an underlying diurnal variation in GFR [30–33] as well as changes
in response to external stimuli, e.g. exercise [34] and food
intake [9–11]. The diurnal rhythm in GFR, which reflects
changes in circulating atrial natriuretic peptide rather than
arterial pressure [32], is represented differently by inulin
and creatinine clearances [35,36].
Food intake
A protein-rich meal causes an increase in GFR of about
25% [18,19] but in the current study we observed a general
increase much less than this because the meal was generally
lighter. As our intention was to compare iohexol and Cr-51EDTA in a paired fashion, there was no need to standardize
the meal. Moreover, a light meal, resulting in a more modest
increase in GFR, is more discriminatory.
Three versus six samples
The coefficients of variation did not differ significantly
between three- and six-sample GFR measured with Cr-51EDTA, with respect either to GFR/BSA or GFR/ECV. For
iohexol, three-sample GFR was more reproducible when
scaled to BSA but the opposite was seen scaled to ECV.
There was little difference between three- and six-sample
GFR with respect to response to food intake. In terms of
both reproducibility and response to food intake, inclusion
of the initial three blood samples seemed to offer no improvement. So although not necessarily more accurate than
six-sample GFR, three-sample GFR provided good precision. This is presumably because the additional error generated by the inclusion of early samples is greater than the
normal variation that exists in the parameters on which the
first exponential is based. This may not of course apply to
an abnormal population or to the elderly with their reduced
muscle mass and possibly therefore also a reduced ECV.
Reliability of iohexol
There have been several previous comparisons between
iohexol and Cr-51-EDTA for measuring GFR [7–14], although they have not compared the two markers in a truly
independent fashion by basing their clearances on separate
sets of blood samples drawn from separate sites after simultaneous administration of markers via separate lines or
directly compared their reproducibilities, as we have done
here.
In the current study, the coefficients of variation did not
differ significantly between the two indicators, at least with
respect to the conventional slope–intercept approach (i.e.
GFR/BSA3). By showing a significant correlation between
individual changes in fasting GFR measured by the two
markers simultaneously and independently, it is evident
that a considerable proportion of the variability between
sequential measurements was the result of real changes in
GFR.
Other recent studies have assayed iohexol by HPLC
[6,7,16,20]. Notably though, Brandstrom et al. [7] recorded
correlation coefficients with Cr-51-EDTA that were
similar for both XRF (0.95) and HPLC (0.92). They
commented on the necessity of a minimum plasma iodine
concentration of 0.06 g/l (which corresponds to an iohexol
concentration of 0.27 g/l, less than our most dilute standard
(see Figure 1).
GFR values measured with Cr-51-EDTA showed statistically more significant responses compared with iohexol
in response to food. High protein meals act directly on the
kidney to cause an increase in GFR, which should therefore correlate inversely with any simultaneous change in
ECV. However, we previously observed a positive corre-
Reliability of scaling to ECV
The mean transit time, T, and its approximation, the reciprocal of the terminal rate constant, represent the mean
waiting time of individual molecules of marker in the extracellular fluid space compartment before filtration and
are therefore measures of GFR that are already scaled to
ECV. In other words, measuring GFR per unit ECV requires
measurement only of their ratio, and not of their individual values. BSA for scaling GFR is largely historical and
well entrenched but has nevertheless been criticized on several grounds, including sex differences in renal perfusion
and gross differences in body size [1,38] and compared
unfavourably with several alternatives, including ECV. In
addition to technical convenience, ECV as a whole body
scaling parameter offers the attraction of physiological validity [39], including an element of dependence of GFR
on ECV [39–41]. Further advantages of the exclusive use
of the terminal rate constant to measure GFR are the opportunity of the ‘real-time’ measurement of GFR (which
allows the detection of instantaneous changes) [42,43] and
convenience in the context of the remote measurement of
GFR.
The coefficients of variation did not differ significantly
between GFR values respectively scaled to BSA and ECV
with respect either to three- or six-sample Cr-51-EDTA
clearance. The slope–intercept technique, however, exposed
a clearer response to food intake than scaling to ECV which
could be explained by a simultaneous corresponding increase in ECV [37], limiting an increase in the ratio. Scaling to ECV also impaired reproducibility when iohexol was
the marker.
1908
Conclusion
GFR measured with iohexol has similar reproducibility to
GFR measured with Cr-51-EDTA using three samples and
the slope–intercept approach. Unless the volunteer’s weight
changes significantly between paired measurements, reproducibility and response to food intake of unscaled GFR
would, of course, be identical to BSA-scaled GFR, and
some of the disadvantages of BSA as a scaling parameter,
discussed above, would not be apparent in reproducibility
and acute response studies such as the current one. When
measured with Cr-51-EDTA, GFR/ECV was almost as reproducible as GRF/BSA, thereby in principle validating the
slope-only technique. GFR/ECV measured with iohexol,
however, was less reliable. A feature of the conventional
slope–intercept technique that is usually taken for granted,
but which relies on careful attention to technical detail,
is the administered amount of indicator, errors in which
are impossible to detect retrospectively. As far as a postal
service is concerned, therefore, scaling to ECV would be
highly favoured but critically dependent on the accuracy
of the iohexol assay. Technology advances rapidly, and although previous experiences with XRF have shown it to be
reliable, most authorities would agree that HPLC is the most
accurate method for assaying iohexol and its unavailability
to the current authors must be seen as a limitation in this
study. We believe, however, that the results reported here
provide a sound platform from which a postal service with
iohexol could be developed. Whether it should be based
on three or six samples awaits further work to confirm that
three-sample GFR/ECV measured with iohexol and HPLC
is as accurate as GFR/ECV measured with Cr-51-EDTA.
Acknowledgements. We are grateful to the normal subjects and patients
who kindly volunteered to participate in this study, which was supported
by the National Kidney Research Fund of the United Kingdom.
Conflict of interest statement. None declared.
References
1. Turner ST, Reilly SL. Fallacy of indexing renal and systemic hemodynamic measurements for body surface area. Am J Physiol 1995; 268:
R978–R988
2. Peters AM. Expressing glomerular filtration rate in terms of extracellular fluid volume. Nephrol Dial Transplant 1992; 7: 205–210
3. Bird NJ, Peters C, Michell AR et al. Comparison between slopeintercept and slope-only techniques for measuring glomerular filtration rate: use of two independent markers and an independent arbiter.
Nucl Med Commun 2007; 28: 711–718
4. Anastasio P, Spitali L, Frangiosa A et al. Glomerular filtration rate
in severely overweight normotensive humans. Am J Kidney Dis 2000;
35: 1144–1148
5. Bird NJ, Henderson BL, Lui D et al. Indexing glomerular filtration
rate to suit children. J Nucl Med 2003; 44: 1037–1043
6. Nicelescu-Duvaz I, D’Mello L, Maan Z et al. Development of an
outpatient finger-prick glomerular filtration rate procedure suitable
for epidemiological studies. Kidney Int 2006; 69: 1272–1275
7. Brandstrom E, Grzegorczyk A, Jacobsson L et al. GFR measurement
with iohexol and 51Cr-EDTA. A comparison of the two favoured GFR
markers in europe. Nephrol Dial Transplant 1998; 13: 1176–1182
8. Pucci L, Bandinelli S, Penno G et al. Iohexol plasma clearance in
determining glomerular filtration rate in diabetic patients. Ren Fail
1998; 20: 277–284
N. J. Bird et al.
9. Frennby B, Sterner G, Almen T et al. The use of iohexol clearance to
determine GFR in patients with severe chronic renal failure—a comparison between different clearance techniques. Clin Nephrol 1995;
43: 35–46
10. Pucci L, Bandinelli S, Pilo M et al. Iohexol as a marker of glomerular
filtration rate in patients with diabetes: comparison of multiple and
simplified sampling protocols. Diabet Med 2001; 18: 116–120
11. Lundqvist S, Hietala SO, Karp K. Experimental studies comparing
iohexol and 51 Cr-EDTA for glomerular filtration rate measurements.
Acta Radiol 1995; 36: 58–63
12. Rydstrom M, Tengstrom B, Cederquist I et al. Measurement of
glomerular filtration rate by single-injection, single-sample techniques, using 51 Cr-EDTA or iohexol. Scand J Urol Nephrol 1995;
29: 135–139
13. Frennby G, Sterner G, Almen T et al. Clearance of iohexol, 51 CrEDTA and endogenous creatinine for determination of glomerular
filtration rate in pigs with reduced renal function: a comparison between different clearance techniques. Scand J Clin Lab Invest 1997;
57: 241–252
14. Eriksson CG, Kallner A. Glomerular filtration rate: a comparison
between Cr-EDTA clearance and a single sample technique with a
non-ionic contrast agent. Clin Biochem 1991; 24: 261–264
15. Erley CM, Bader BD, Berger ED et al. Plasma clearance of iodine
contrast media as a measure of glomerular filtration rate in critically
ill patients. Crit Care Med 2001; 29: 1544–1550
16. Gaspari F, Perico N, Ruggenenti P et al. Plasma clearance of nonradioactive iohexol as a measure of glomerular filtration rate. J Am
Soc Nephrol 1995; 6: 257–263
17. Wilkinson J, Fleming JS, Waller DG et al. Effect of food and activity
on the reproducibility of isotopic GFR estimation. Nucl Med Commun
1990; 11: 697–700
18. De Santo NG, Anastasio P, Cirillo M et al. Sequential analysis of
variation of glomerular filtration rate to calculate the haemodynamic
response to a meat meal. Nephrol Dial Transplant 2002; 10: 1629–
1637
19. Laville M, Hadj-Aissa A, Pozet N et al. Restrictions on use of creatinine clearance for measurement of renal functional reserve. Nephron
1989; 51: 233–236
20. Gaspari F, Perico N, Matalone M et al. Precision of plasma clearance
of iohexol for estimation of GFR in patients with renal disease. J Am
Soc Nephrol 1998; 9: 310–313
21. James TJ, Lewis AZ, Tan GD et al. Validity of simplified protocols to
estimate glomerular filtration rate using iohexol clearance. Ann Clin
Biochem 2007; 44: 369–376
22. Bröchner-Mortensen J, Rödbro P. Selection of routine method for
determination of glomerular filtration rate in adult patients. Scand J
Clin Lab Invest 1976; 36: 35–43
23. Sambataro M, Thomaseth K, Pacini G et al. Plasma clearance rate
of 51 Cr-EDTA provides a precise and convenient technique for measurement of glomerular filtration rate in diabetic humans. J Am Soc
Nephrol 1996; 7: 118–127
24. Cousins C, Skehan SJ, Rolph S et al. Comparative microvascular
exchange kinetics of [77 Br]bromide and 99m Tc-DTPA in humans. Eur
J Nucl Med 2002; 29: 655–662
25. Du Bois D, Du Bois EF. A formula to estimate the approximate surface
area if height and weight are known. Arch Intern Med 1916; 17:
863
26. Brochner-Mortensen J. A simple method for the determination of
glomerular filtration rate. Scand J Clin Lab Invest 1972; 30: 271–274
27. Nosslin B. Determination of clearance and distribution volume with
the single injection technique. Acta Med Scand 1965; Suppl 442:
97–101
28. Peters AM, Henderson BL, Lui D et al. Appropriate corrections to
glomerular filtration rate and volume of distribution based on the bolus
injection and single compartment technique. Physiol Meas 1990; 20:
313–327
29. Altman DG. Practical Statistics for Medical Research. London: Chapman and Hall, 1991, 269–270
30. Sirota JH, Baldwin DS, Villarreal H. Diurnal variations of renal
function in man. J Clin Invest 1950; 29: 187–192
Iohexol for measuring GFR
31. Wesson LG Jr, Lauler DP. Diurnal cycle of glomerular filtration
rate and sodium and chloride excretion during responses to altered salt and water balance in man. J Clin Invest 1961; 40: 1967–
1977
32. Voogel AJ, Koopman MG, Hart AAM et al. Circadian rhythms
in systemic hemodynamics and renal function in healthy subjects
and patients with nephrotic syndrome. Kidney Int 2001; 59: 1873–
1880
33. Hansen HP, Hovind P, Jensen BR et al. Diurnal variations of glomerular filtration rate and albuminuria in diabetic nephropathy. Kidney Int
2002; 61: 163–168
34. Svarstad E, Myking O, Ofstad J et al. Effect of light exercise on renal hemodynamics in patients with hypertension and
chronic renal disease. Scand J Urol Nephrol 2002; 36: 464–
472
35. van Acker BA, Koomen GC, Koopman MG et al. Discrepancy between
circadian rhythms of inulin and creatinine clearance. J Lab Clin Med
1992; 120: 400–410
36. van Acker BA, Koomen GC, Arisz L. Drawbacks of the constantinfusion technique for measurement of renal function. Am J Physiol
1995; 268: F543–F552
1909
37. Bird NJ, Peters C, Michell R et al. Association between glomerular
filtration rate and extracellular fluid volume in normal subjects and
patients with renal impairment. Scand J Clin Lab Invest (in press)
38. McCance RA, Widdowson EM. The correct physiological basis on
which to compare infant and adult renal function. Lancet 1952; 1:
860–862
39. Cervenka L, Mitchell KD, Navar LG. Renal function in mice: effects
of volume expansion and angiotensin II. J Am Soc Nephrol 1999; 10:
2631–2636
40. Brown A, Zammit VC, Lowe SA. Capillary permeability and extracellular fluid volumes in pregnancy-induced hypertension. Clin Sci
1989; 77: 599–604
41. Neilsen OM, Engell HC. Increased glomerular filtration rate in patients after reconstructive surgery on the abdominal aorta. Br J Surg
1986; 73: 34–37
42. Bauman LA, Watson NE, Scuderi PE et al. Transcutaneous renal
function monitor: precision during unsteady hemodynamics. J Clin
Monit Comput 1998; 14: 275–282
43. Rabito CA, Moore RH, Bougas C et al. Noninvasive, real-time monitoring of renal function: the ambulatory renal monitor. J Nucl Med
1993; 34: 199–207
Received for publication: 27.6.07
Accepted in revised form: 17.10.07