| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Physiology, Cardiovascular Research Institute Maastricht, University of Maastricht, 6200 MD Maastricht, The Netherlands; 2 School of Physiology and Pharmacology of the University of New South Wales, Sydney, New South Wales 2260, Australia; and 3 Departments of Medicine and of Physiology and Biophysics, University of Washington, Seattle, Washington 98195
 |
ABSTRACT |
|---|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
The accuracy of the fluorescent (FM) and
radioactive microsphere (RM) techniques is similar in acute experiments
but has not been established in chronic experiments. In the present
study various combinations (at least pairs) of FM and/or RM
labels were injected simultaneously between 2 mo and 5 min before each
animal was killed. Blood flow was determined in many
organs. Intramethod mean difference and variation did not change over
time for FM but increased significantly for RM (from 1.8 ± 1.4 to
25.6 ± 21.8% and from 4.4 ± 3.2 to 32.4 ± 23.0% at 5 min
and 2 mo, respectively). Also the FM-RM intermethod mean difference and
variation increased (from 
0.5 ± 8.5 to 40.8 ± 23.8% and
from 23.6 ± 4.6 to 71.8 ± 34.3%, respectively). After 2 mo,
blood flow estimations were 20-50% lower with the various RM,
whereas brain and liver blood flow values varied even more between
isotopes. Underestimation started within 1 day for
51Cr and within 2 wk for
141Ce,
95Nb, and
85Sr. We conclude that FM are
superior to RM for blood flow determination in experiments lasting
longer than 1 day, presumably because of leaching of isotopes
from RM.
nonradioactive microspheres; fluorescence; organ flow; chronic animal experiments; cardiac output distribution
| |
INTRODUCTION |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
THE FLUORESCENT MICROSPHERES (FM) technique has been demonstrated to be an excellent alternative to the radioactive microspheres (RM) technique in acute experiments (2, 4, 14). The accuracy of the FM technique is comparable to that of its radioactive counterpart, and although the FM method is more time consuming, the lack of radiation for workers and the environment makes the fluorescent method safer and presents fewer legislative problems. The high costs of storage and disposal of waste are also avoided.
Nonradioactive microspheres may offer even greater advantages in chronic experiments. In the case of isotopes with short half-life, high specific activities are required at the time of injection. This causes a larger radiation load for animals and their environment during the early phase of the experiment. The release of isotopes with excreta from animals, although usually low, is also a concern.
Because the use of RM in chronic experiments has only been validated to a limited extent and because FM may offer the various benefits mentioned above, we compared the validity of the FM and RM techniques in chronic experiments. Two of the factors determining the accuracy of the microsphere method (5, 10, 13) deserve special emphasis in chronic experiments, i.e., the absence of both leaching of the label from the microspheres and disappearance of microspheres from the site of entrapment (10). In a study on leaching of radioactive microspheres from the heart, Consigny et al. (3) showed that microspheres appearing in the venous circulation are almost exclusively smaller than 12 µm in diameter. A similar conclusion was reached by a study of Medvedev et al. (12), who determined the content of 46Sc-labeled spheres with a diameter of 15 µm in donor hearts and in lungs of the recipients. The latter study also indicated some leaching of 46Sc from the beads. Losses of 125I (7) and 113Sn have also been reported (9).
The present experiments were performed in rabbits. Various combinations (at least pairs) of FM and RM labels were injected simultaneously at 2 mo, 1 mo, and 5 min before the animal was killed. The accuracy of the RM and FM methods was evaluated by quantification of the variation and mean difference of blood flow values between RM and FM and between different labels of RM as well as of FM in a large number of organs. Because RM lost label within 1 mo, a second set of animals was studied to compare the various isotopes in experiments lasting 1, 7, and 14 days.
| |
METHODS |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
In vitro test on leaching.
To test potential leaching of fluorescent labels from FM, we took 20 samples of 1 ml each from a stock solution containing a mixture of all
FM used in the present study; the spheres were suspended in Hemaccel.
Samples were stored in the dark for 3 days and for 2 mo, either at
37°C (the temperature at which microspheres stay while the animal
is still alive) or at 20°C (the temperature at which blood
samples are stored while the animal is still alive in the chronic
situation and at which all samples were stored between gamma counting
and processing for fluorimetry). At the end of the storage period the
samples were processed like the tissue and blood samples, and the
fluorescence was determined (see Sample processing and
blood flow determination). Because the
product information from the supplier states that decay of the
fluorescent labels is small (<1% after 6 mo of storage in the dark),
any decrease in the dye content during the 2 mo of storage was assumed
to be due to leaching of the dye from the beads during the storage.
20°C for at least
0.5 yr for decay of radioactivity. The microspheres were then isolated
by centrifugal sedimentation, and fluorescence was determined as
described previously (14).
Organ blood flow (Q,
ml · min
1 · g1)
was calculated as Qi = (Qref · Ii)/(Iref · Gi),
where Qref is the withdrawal speed
of the arterial reference sample (in ml/min), Ii and
Iref are the radioactivity or
fluorescence intensity in sample i and
the reference sample, respectively, and
Gi is the weight of the tissue
sample.
Statistics. Intermethod variability
(mean difference and variation) was determined by comparing the mean of
the blood flow values obtained with the two simultaneously injected RM
(
RM) to the mean values obtained with the two simultaneously injected FM
(FM)
for each individual sample. The values were
used rather than values from individual labels to obtain the most
precise blood flow value. Intramethod variability was determined by
comparing individual values of each label to the
value of the same kind (RM or FM).
Per-experiment intramethod mean difference and variation were
calculated as the mean value and SD of
QRM,i RM or
QFM,i
FM
of all samples, respectively, and expressed as percentage of
RM or
FM.
Intramethod mean difference was always taken as a positive value.
Intermethod mean difference and variation were calculated as the mean
value and SD of
FM QRM,i, so a
positive mean difference indicates lower values for RM than for FM.
Data were also compared by means of the analysis of Bland and Altman
(1), where the difference between the two methods (FM RM for
intermethod comparison and
QRM,i RM or
QFM,i FM for
intramethod comparison) was plotted against
RM or
FM,
respectively. Changes in mean difference and variation over time were
evaluated for significance using one-way ANOVA and subsequent post hoc
testing, with P < 0.05 considered significant.
| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
The in vitro studies, in which FM were stored for 2 mo at either
20 or 37°C, showed no significant leaching of fluorescent label, except for a 10% loss of blue-green and crimson at
37°C and a 13% loss of red at 20°C. Red also tended to
lose label at 37°C (Table 1).
|
In Fig. 1 the results obtained in one
typical experiment are presented and in Table
2 the pooled data for all
experiments are shown. In the acute measurements blood flow values
obtained from all microsphere labels correlated very well with each
other. Consequently, inter- and intramethod errors were small. In Fig. 1, A and
B, it can be seen that the intramethod
variability (mean difference and variation) for the FM increased
slightly over time, predominantly due to relatively low values for blue
and blue-green in samples from heart and lungs in this particular
experiment. Nevertheless, slope and intercepts of the regression
equations for FM intramethod comparison were not significantly
different from unity and zero, respectively
{y = 1.03x + 0.01 [r = 0.96, standard error of estimate
(SEE) = 0.06]} after 5 min,
y = 0.89x 0.07 (r = 0.90, SEE = 0.20) after 1 mo, and
y = 0.91x + 0.06 (r = 0.88, SEE = 0.34) after 2 mo]. For the whole group of experiments the FM-FM intramethod
variation did not change significantly over time (Table 2). Also, there
was no systematic mean difference for any of the FM labels at any time
point.
|
|
Figure 1, E and
F, shows an increase in the
intramethod variability for RM over time. Regression equations for RM
intramethod comparison were y = 1.05x + 0.01 (r = 0.97, SEE = 0.08) after 5 min,
y = 0.72x + 0.13 (r = 0.91, SEE = 0.19) after 1 mo, and y = 0.84x 0.02 (r = 0.93, SEE = 0.21) after
2 mo. After 1 mo the slope was significantly lower than unity. For the
whole group the increase in RM-RM intramethod variation was
statistically significant at 2 mo (Table 2).
In the experiment shown in Fig. 1, the intermethod mean difference
increased at 1 and 2 mo (Fig. 1, C and
D). This was due to a decrease of
the slope of the regression equation for the FM-RM comparisons from
0.86 and 0.92 at 5 min [not significantly different
from 1] to 0.81 and 0.57 after 1 mo and 0.49 and 0.57 after 2 mo
(all significantly lower than unity). Intercepts were not significantly
different from zero. For all experiments only the increase in mean
difference at 2 mo reached the level of significance. Because the
intermethod mean difference is defined as
(see
METHODS), the positive intermethod
mean difference indicates lower blood flow estimations by RM than by
FM. The increase in intramethod mean difference at 1 mo was not
significant because of higher RM blood flow estimates in one
experiment. In this particular experiment Cr proved to result in higher
flow values than the other RM and FM labels for several organs.
Intermethod variation significantly increased at 1 and 2 mo (Table 2).
After 2 mo blood flow values from RM were on average 40% lower than those obtained with FM (Table 2). This difference was smallest for Sn, Ce, and Ru (Fig. 2). In the case of Ce and Ru, however, this was due to disproportionally high values in brain (Ru and Ce) and liver (Ce). In the latter organ Ce blood flow was more than twice the FM values (Fig. 2). Blood flow values obtained with Cr and Nb were ~50% lower than those obtained with FM. This difference was similar in all organs for Nb, whereas brain flow values obtained with Cr were closer to the FM values (Fig. 2). These interisotope and interorgan differences in RM blood flow estimations explain the high FM-RM intermethod variability (Table 2). At the time of death significant amounts of Ce, Cr, Ru, and Nb were found in the urine and feces of some animals. Blood flow estimations from the various FM labels were similar in all organs (data not shown).
|
When RM were injected 1 day before the animal was killed, all labels except Cr gave similar blood flow values (Fig. 3). Injection of RM 7 days before the animal was killed resulted in 15-30% underestimation of blood flow by Cr, Sr, and Nb compared with Sc. During the experiments lasting 14 days a slightly more pronounced underestimation was observed for Sr, Cr, and Ce (Fig. 3).
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This study demonstrates that in chronic experiments of up to 2 mo blood flow estimation by FM is reproducible, whereas RM lead to variable results and usually lower values than those obtained with FM. Therefore, FM are to be preferred for blood flow measurements in experiments lasting longer than 1 day.
The mean difference and variation between any two fluorescent labels do not significantly change within 2 mo. This indicates that FM perform virtually as well in chronic as in acute studies and that FM are suitable to follow changes in absolute blood flow over time, at least for 2 mo. The reliability of FM is probably due to stability of labels. This idea is supported by the in vitro studies.
The present study cannot exclude dislodgment of spheres over time. Loss
of spheres from a tissue is likely to be reflected in a relatively high
lung blood flow value as the spheres are trapped by the lungs (6, 11).
In the present study, however, comparison of absolute lung flows
between various injection times is hampered by variable conditions
between injections, such as ambient temperature, which may influence
shunt flow. Although the absence of dislodgment cannot be proven in the
present study, three previous studies virtually exclude the likelihood
of dislodgment of microspheres in chronic studies. Hales and Cliff (8)
reported that losses from the rabbit ear or thoracic tissue had stopped within a few minutes after injection and were not detectable for up to
8 wk. Consigny et al. (3) demonstrated that over a period of 5 wk part
of the microspheres with diameters <12 µm disappear from the
myocardium but hardly any dislodging of microspheres with diameters
15 µm occurs. Medvedev et al. (12) counted activity of hearts from
donor rats that had previously been injected with Sc-labeled RM. The
hearts were retrogradely perfused by attaching the ascending aorta to
the abdominal aorta of the recipient animals. Within 4 wk these
investigators did not find accumulation of activity in the lungs of the
recipient animal (12).
The leaching of blue-green and crimson at 37°C in vitro may, in
theory, lead to underestimation of tissue blood flows determined with
these labels because FM in the tissues remain at this temperature while
the related reference blood samples were stored at 20°C. Figure 1, A and
B, shows some examples
of increased intramethod variability. However, in vitro leaching was
relatively small (10% in 2 mo) compared with the errors observed with
the RM method, and in vivo the increase in FM intramethod variability
over time was not statistically significant (Table 2). Red appeared to be leached from the spheres at a similar rate at 20 and
37°C. Therefore, this leaching would not result in a decline of
blood flow, calculated with the reference method.
The generally limited extent of dislodgment of 15-µm-diameter microspheres in general in combination with the minimal leaching of label from FM indicate that absolute blood flows can be determined accurately with FM in chronic experiments. The generally lower and more variable blood flow estimations by RM in chronic experiments, therefore, indicate poor performance of RM under these circumstances. While on average RM underestimate blood flow by up to 40% compared with FM, considerable differences are found between isotopes and between organs. This can be observed from the intermethod variation at 2 mo, which is considerably larger than both intramethod variations. Cr, Sr, and Nb underestimated blood flow most (Cr already within 1 day; Fig. 3). Although underestimation by Ce and Ru appeared to be less compared with other isotopes, this observation is, in part, due to selective overestimation of blood flow in organs like the brain and the liver.
This variability in blood flow estimation by RM is most likely not caused by dislodgment of the spheres, because in that case blood flow to the lungs should be higher for RM than for FM. Instead, lung flows were underestimated by RM as much as other organs like the heart. Therefore, the most likely explanation is the loss of radioactive label from the beads. While loss of activity from the beads in most organs can explain the underestimation of tissue blood flow, selective binding of isotopes in some organs can explain overestimation, as, for example, by Ce in the liver and Ce and Ru in the brain.
The interorgan variation in leaching of the isotopes suggests that leaching is dependent on the environment. This environment is different in the various organs, because microspheres move to the interstitium within 1-3 wk (3, 8), thus becoming subject to the chemical environment of each particular organ. In the present study leaching of label is also suggested by the presence of Ce, Cr, Ru, and Nb in urine and feces at the time some animals were killed. The data of the present study also indicate that the leaching process is different between isotopes. Hales et al. (9) attributed falsely high blood flow values for baboon liver obtained with Sn (manufactured by 3M) to minute losses of that label from tissues in general and uptake in the liver. Similarly, 125I appeared to be lost from spheres entrapped in the kidney, gastrointestinal tract, and bone and taken up by thyroid and fat in sheep (7). These observations suggest that coating and/or chemical binding to the resin of the bead is of variable efficiency.
The present results are not in contradiction with those of Consigny et al. (3), who studied the loss of microspheres from the myocardium. These investigators determined the chronic loss of microspheres from the myocardium from the radioactivity counted within and outside the heart rather than counting microspheres. In this approach loss of label from spheres will not be observed if loss is equal within and outside the heart.
The large variability in blood flow estimations between the various radioactive labels, as well as the increasing underestimation of flow with time, starting within a week after injection of the spheres, makes blood flow data from RM in chronic experiments unreliable. Although the interisotope variation was limited to 30% in organs like heart, lungs, and kidneys, variation in the brain amounted to as much as a factor three.
This poor performance of RM stands in contrast to the good correlation
between RM and FM blood flow estimation in acute experiments (<1
day), as has also been demonstrated in previous studies (4, 14).
Actually, the accuracy of the RM was slightly better than that of the
FM, although in previous studies the accuracies of both methods were
found to be similar. This may be due to the fact that in the present
study determination of radioactivity was performed soon after the
animal was killed, whereas the fluorescence measurements were performed
after storage of the samples for up to 1 yr. The in vitro studies show
that long-term storage of samples at 20°C may cause some
leaching or destruction of spheres.
We conclude that in chronic animal experiments the FM method is superior to the RM method, presumably because after periods of 1 day and longer several isotopes are leached from spheres in the tissue and may accumulate in other tissues. The accuracy of organ blood flow determination with FM does not deteriorate within 2 mo. This advantage in the accuracy of blood flow measurements comes in addition to other advantages of FM, like lack of radiation and minimal decay of the label over time.
| |
ACKNOWLEDGEMENTS |
|---|
The authors are indebted to Gari Gazibarich for assistance with the animal experiments, Annita Rousseau for performing the fluorescent microsphere assays, and Robert S. Reneman for critically reviewing the manuscript.
| |
FOOTNOTES |
|---|
Present address of J. R. S. Hales: Dept. of Veterinary Clinical Sciences, Univ. of Sydney, Camden, New South Wales, Australia.
Address for reprint requests: F. W. Prinzen, Dept. of Physiology, Univ. of Maastricht, PO Box 616, 6200 MD Maastricht, The Netherlands.
Received 20 October 1997; accepted in final form 16 March 1998.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
1.
Bland, J. M.,
and
D. G. Altman.
Statistical methods for assessing agreement between two methods of clinical measurement.
Lancet
1:
307-310,
1986[Medline].
2.
Chien, G. L.,
C. G. Anselone,
R. F. Davis,
and
D. M. Van Winkle.
Fluorescent vs. radioactive microsphere measurement of regional myocardial blood flow.
Cardiovasc. Res.
30:
405-412,
1995[Medline].
3.
Consigny, P. M.,
E. D. Verrier,
B. D. Payne,
G. Edelist,
J. Jester,
R. W. Baer,
G. J. Vlahakes,
and
J. I. E. Hoffman.
Acute and chronic microsphere loss from canine ventricular myocardium.
Am. J. Physiol.
242 (Heart Circ Physiol. 11):
H392-H404,
1982.
4.
Glenny, R. W.,
S. Bernard,
and
M. Brinkley.
Validation of fluorescent-labeled microspheres for measurement of regional organ perfusion.
J. Appl. Physiol.
74:
2585-2597,
1993
5.
Hales, J. R. S.
Practical considerations in the use of microspheres for evaluation of cardiovascular and thermoregulatory interactions.
Biomed. Thermol.
13:
127-145,
1994.
6.
Hales, J. R. S.
Radioactive microsphere measurement of cardiac output and regional tissue blood flow in the sheep.
Pflügers Arch.
344:
119-132,
1973[Medline].
7.
Hales, J. R. S.,
J. W. Bennett,
and
A. A. Fawcett.
Effects of acute cold exposure on the distribution of cardiac output in the sheep.
Pflügers Arch.
366:
153-157,
1976[Medline].
8.
Hales, J. R. S.,
and
W. J. Cliff.
Direct observations of the behaviour of microspheres in microvasculature.
Bibl. Anat.
15:
87-91,
1977.
9.
Hales, J. R. S.,
R. B. King,
and
A. A. Fawcett.
Observations on the validity of using NEN-Trac microspheres for measuring organ blood flow.
Pflügers Arch.
379:
295-296,
1979[Medline].
10.
Heymann, M. A.,
B. D. Payne,
J. I. E. Hoffman,
and
A. M. Rudolph.
Blood flow measurements with radionuclide-labeled microspheres.
Prog. Cardiovasc. Dis.
20:
55-79,
1977[Medline].
11.
Kaihara, S.,
P. D. Van Heerden,
T. Migita,
and
H. N. Wagner.
Measurement of distribution of cardiac output.
J. Appl. Physiol.
25:
696-700,
1968
12.
Medvedev, O. S.,
E. R. Martynova,
R. S. Akchurin,
and
V. Y. Khalatov.
Experimental estimation of chronic microsphere loss from the rat myocardium.
Biull. Eksp. Biol. Med.
103:
8-10,
1987[Medline].
13.
Prinzen, F. W.,
and
R. W. Glenny.
Developments in non-radioactive microsphere techniques for blood flow measurement.
Cardiovasc. Res.
28:
1467-1475,
1994[Medline].
14.
Van Oosterhout, M. F. M.,
H. M. M. Willigers,
R. S. Reneman,
and
F. W. Prinzen.
Validation of fluorescent microsphere technique for measurement of organ perfusion with improved sample processing method.
Am. J. Physiol.
269 (Heart Circ. Physiol. 38):
H725-H733,
1995
This article has been cited by other articles:
|
|
 |
|
  R. W. Glenny, S. L. Bernard, D. L. Luchtel, B. Neradilek, and N. L. Polissar The spatial-temporal redistribution of pulmonary blood flow with postnatal growth J Appl Physiol, March 1, 2007; 102(3): 1281 - 1288. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. N. Athanasopoulos, M. Economopoulou, V. V. Orlova, A. Sobke, D. Schneider, H. Weber, H. G. Augustin, S. A. Eming, U. Schubert, T. Linn, et al. The extracellular adherence protein (Eap) of Staphylococcus aureus inhibits wound healing by interfering with host defense and repair mechanisms Blood, April 1, 2006; 107(7): 2720 - 2727. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. K. M. Decking, V. M. Pai, E. Bennett, J. L. Taylor, C. D. Fingas, K. Zanger, H. Wen, and R. S. Balaban High-resolution imaging reveals a limit in spatial resolution of blood flow measurements by microspheres Am J Physiol Heart Circ Physiol, September 1, 2004; 287(3): H1132 - H1140. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. C. Alders, A. B. J. Groeneveld, F. J. J. de Kanter, and J. H. G. M. van Beek Myocardial O2 consumption in porcine left ventricle is heterogeneously distributed in parallel to heterogeneous O2 delivery Am J Physiol Heart Circ Physiol, September 1, 2004; 287(3): H1353 - H1361. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Anetzberger, E. Thein, G. Loffler, and K. Messmer Fluorescent microsphere method is suitable for chronic bone blood flow measurement: a long-term study after meniscectomy in rabbits J Appl Physiol, May 1, 2004; 96(5): 1928 - 1936. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Matsumoto, H. Tachibana, T. Asano, M. Takemoto, Y. Ogasawara, K. Umetani, and F. Kajiya Pattern differences between distributions of microregional myocardial flows in crystalloid- and blood-perfused rat hearts Am J Physiol Heart Circ Physiol, April 1, 2004; 286(4): H1331 - H1338. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Anetzberger, E. Thein, M. Becker, A. K. Walli, and K. Messmer Validity of fluorescent microspheres method for bone blood flow measurement during intentional arterial hypotension J Appl Physiol, September 1, 2003; 95(3): 1153 - 1158. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Conhaim, K. E. Watson, D. M. Heisey, G. E. Leverson, and B. A. Harms Perfusion heterogeneity in rat lungs assessed from the distribution of 4-{micro}m-diameter latex particles J Appl Physiol, February 1, 2003; 94(2): 420 - 428. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. H. Traverse, P. Melchert, G. L. Pierpont, B. Jones, M. Crampton, and R. J. Bache Regulation of Myocardial Blood Flow by Oxygen Consumption Is Maintained in the Failing Heart During Exercise Circ. Res., March 5, 1999; 84(4): 401 - 408. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
) in blood flow of FM pair and of RM pair. In D,
x-axis corresponds to mean flow of all FM labels used.
Bottom: absolute errors as a function of mean
blood flow. Data and symbols of B,
D, and
F correspond to regression plots of
A, C,
and E, respectively. Measurements are
at 5 min (filled symbols), 1 mo (open symbols), and 2 mo before animals
were killed (+ and ×). In this particular experiment some samples
obtained from heart and lungs showed underestimation by B (2 mo) and BG
(1 mo) compared with R and C, respectively. However, overall the slope
and intercept of regression equations were not statistically
different from 1 and 0, respectively. RM-RM intramethod variation
increased with time, which becomes especially clear in
F. The 2 outliers for Ce (see
C-F)
are from the 2 kidney samples in this experiment.
