Nephrol Dial Transplant (1996) 11: 275-281
Nephrology
Dialysis
Transplantation
Original Article
Micropuncture and clearance measurements of segmental reabsorption
by the rat nephron
E. Bartoli, G. Romano and G. Favret
Medicina Interna, Universita di Udine, Udine, Italy
Abstract
Background. We wanted to verify whether the calculations of segmental tubular reabsorption obtained
during water diuresis were supported by direct micropuncture measurements.
Methods. Experiments were performed on 18 rats
during baseline water diuresis (B) and after the administration of frusemide (F), lOmg/kg, by whole-kidney
clearance measurements and micropuncture collections
from early distal (ED) and last proximal (LP) tubular
segments.
Results. GFR was 957 + 79 in B, 1053 + 77 ul/min in
F, 7>>0.13. SNGFR was 38 + 1 in 166 and
38 + 1 nl/min in 165 tubules respectively, P>0.77. In
LP collections the percentage reabsorption was 71+2
in B and 76 + 2% during F (P>0.07) in 99 and 95
samples respectively. The absolute proximal reabsorption was not changed by F (27.6 ±1.5 versus
27.7 + 1.3 nl/min, P>0.96). The data were superimposable when the analysis was restricted to paired data.
The difference between ED and LP resorption was
17 + 3 during B and fell significantly (P< 0.008) to
5 ±3% during F. The percentage of GFR excreted
during F, measured by clearance techniques, and the
percentage delivery of filtrate beyond the proximal
tubule, measured independently by micropuncture,
were not different (27 ±2 versus 24 + 2%, P>0.10),
while they were significantly correlated (Z'<0.04). The
calculations of segmental Na reabsorption along the
different nephron segments by clearance techniques
were not statistically different from and were significantly correlated with the reabsorptions measured directly by micropuncture.
Conclusions. The present experiments validate the calculations of reabsorption by techniques applicable to
human studies of clinical physiology.
Key words: diluting ability; free water clearance; frusemide; micropuncture; proximal reabsorption; SNGFR
Correspondence and offprint requests to: Ettore Bartoli, Medicina
Interna, Policlinico Universitario, Piazzale S. M. Misericordia 1,
33100 Udine, Italy.
Introduction
Clinical physiology studies in humans are useful to
disclose the pathophysiology of diseases. Several
attempts were made to measure separately proximal
from distal events. Traditionally these studies are performed during maximal water diuresis (WD) and are
based on the assumption that the peak urine flow rate
approximates proximal delivery. Other assumptions
are that free water clearance (CH2O) is a quantitative
estimate of solute reabsorption along the diluting
segment.
In previous studies we analysed the quantitative
inadequacy of this approach and proposed a new
technique that allows to measure the Na + reabsorption
by each segment of the human nephron in vivo [1,2].
This new method requires the measurement of renal
diluting power during baseline conditions (B) and
during the superimposed action of frusemide (F) and
it is critically dependent on the key assumption that F
does not act on the proximal tubule. A micropuncture
study on rats demonstrated that F does not exert any
proximal effect [3], validating the basic assumption of
our human studies.
To gain a wider acceptance of this method we
designed the present studies to confirm the results
obtained by the indirect calculations based on clearance
experiments with direct measurements performed at
proper sampling sites by micropuncture techniques.
Subjects and methods
The experiments were performed on 18 Wistar rats weighing
220-260 g (Morini, San Polo D'Enza, Italy). Detailed
descriptions of the surgical technique and the preparation of
the left kidney for micropuncture were previously reported
together with the thorough discussion of the clearance and
micropuncture methods [3,4]. The relevant aspects, specific
for the present study, are outlined below. The ureters from
both kidneys were cannulated and connected to a reservoir
from which, via a Silastic tubing and a peristaltic pump, the
urine was returned to the animal through the femoral vein.
The pump rate was continuously adjusted manually to keep
the level in the reservoir constant. Immediately after the
insertion of the jugular catheter, the animals were primed
© 1996 European Dialysis and Transplant Association-European Renal Association
276
with a hypotonic solution that contained NaCl 75 mmol/1,
glucose 5 mmol/1. After a priming infusion of 10 ml/kg, WD
was induced by a continuous maintenance infusion of the
same solution at a rate of 50 ml/kg per h. This rate of
infusion lasted on average 21 min, and was reduced to
20 ml/kg per h after the urine reinfusion had begun, and was
kept constant throughout the experiment as it was found in
pilot studies to assure constancy of haematocrit, blood
pressure, and urine flow rate. It was meant to maintain WD
while replacing the losses of fluids due to sampling and to
leakage from surgical openings and to insensible losses. In
fact the haematocrit averaged 43+ 1 during B, and 42+ 1%
during F (/)>0.40, w=18). The systolic BP averaged
123 + 3 mmHg (n=18) throughout the experiment, without any difference between B and F. Immediately after
the beginning of urine reinfusion, the glomerular marker
(14-C- Carboxy-inulin, Amersham International pic,
Buckinghamshire, UK) was added to the urine reservoir at
the dose of 135uCi/kg. From exhaustive pilot studies we
found that these anaesthetized animals never achieved a
minimal urine osmolality and that they never achieved a
maximal WD such as that obtained in conscious animals
and humans. Therefore we proceeded with the experimental
measurements only in those animals which were haemodynamically stable and could dilute the urine in the presence
of a brisk urine flow. At the end of the experiment the
animals were sacrificed by exsanguination, and their bodies
disposed of according to the regulations issued by the Italian
Department of Health.
The experiments were started when the urine flow rate (V)
had reached a stable value above the predetermined lower
limit of 50 ul/min for one kidney. We started by carrying out
micropuncture sampling. After 2-4 collections we performed
a clearance measurement by a timed urine collection into a
calibrated glass capillary and by collecting a blood sample
of 30-50 ul from the femoral artery. At least two clearance
periods were obtained and three blood samples were taken
from each rat during baseline, and at least six tubular
collections were performed between the clearance periods.
This experimental period, called baseline water diuresis (B),
lasted on average 80 (48-131) min. Then F was injected i.v.
at the dose of lOmg/kg, while a maintenance infusion was
not necessary since this was automatically provided by the
continuous urine reinfusion. When V had reached a new
steady value, usually within 25 min from the time of injection,
the micropuncture collections were performed again during
the action of F, interspersed between at least three clearance
periods and three to four blood samplings. This second
experimental period obtained during WD and the superimposed action of F lasted 89 (59-122) min.
The data presented in this study were collected from 18
animals in which the experimental procedures were completed. The experiment was interrupted in three rats because
of haemodynamic instability, haemorrhage and respiratory
failure. No measurements were performed. One animal had
to be discarded because of erratic plasma inulin values due
to malfunction of the urine reinfusion system which was not
recognized at the time of sampling.
The micropuncture measurements were performed with
the technique of total collection of tubular fluid during oil
blockade, as already described in detail elsewhere [4]. The
pipettes were mounted on Leitz micromanipulators. Two
trained operators worked simultaneously together under the
double-headed microscope. The micropuncture recollection
technique from the same sampling site was checked separately
on 83 tubules: SNGFR averaged 41+2 in the first
collection 42 ± 2 nl/min in the immediate recollection,
E. Bartoli el al.
)
/ >0.28. The correlation coefficient was 0.78, /)<0.0001.
These data show a normal distribution and a 96.5% reproducibility, in agreement with those of a previous study [4].
The different experimental procedures carried out in the
course of each experiment were the following:
(i) ED collections during B, and re-collections from the
same sites during F;
(ii) LP collections during B, and re-collections from the same
sites during F;
(iii) ED collections, with immediate LP re-collections from
the same nephrons during B;
(iv) ED collections, with immediate LP re-collections from
the same nephrons during F. These tubules were different
from those punctured during B;
(v) unpaired collections taken either from ED or LP sites,
during B or F. These unpaired samples were due to
tubules which could not be re-collected for technical
reasons, or tubules where either the first collection or the
re-collection had been lost in sample processing.
In the nephrons to be re-collected the oil block was
aspirated when possible with a very thin pipette after the
total collection had been completed, to avoid persistent
blocking and the consequent hydronephrosis. When the
tubule could not be decompressed we always checked
carefully for dilated upstream loops before re-collection,
and did not recollect when obstruction was suspected.
The observation of reverse flow along the distal tubule
was an important clue to persistence of the proximal oil
block. Otherwise it was assumed that the oil block had
run through the whole nephron, and the re-collection
was performed. At the end of aspiration the tip of the
collecting pipette was sealed with a thin droplet of either
tubular blocking oil or surface mineral oil, removed from
the micromanipulator holder and inserted into the transferring apparatus. Under microscopic observation, the
sample was quantitatively transferred into a constantbore precalibrated capillary tubing. The length of the
sample was measured through a calibrated micrometre
advancing apparatus that used the sample meniscus as a
target for length measurement. Then the sample was
quantitatively delivered into the scintillation counting
solution and counted [3]. The reproducibility of 83
duplicate countings was 99.6 + 0.7%. The measurements
of nanoliter samples by constant bore capillaries differed
from true values by 0.6 + 2%. From the TF and P count
per nanolitre per unit time we computed the TF/Pin ratio,
from the sample volume in nanolitres and the collection
time we computed the collection rate (CR) in nl/min.
The SNGFR was calculated as
SNGFR = CR-TF/Pin;
(1)
(2)
(3)
(4)
(5)
The calculation of segmental Na + reabsorption by wholekidney clearance data was the object of two previous
studies on humans and it represents the background of
the present work [1,2]. The theory and the method of
calculation were extensively discussed in those studies.
The nomenclature used in the present paper is the
following:
CH2O = free water clearance measured during baseline
water diuresis (solute free water excreted);
CH2O-T = total free water generated by the diluting
system;
CH2O-HL = free water generated by the loop of Henle;
CH2O-DT = free water generated by the distal convoluted
tubule (given by CH2Of, the free water excreted during
F);
CH2O-BD = free water back-diffusion.
Segmental Na + transport in the rat nephron
277
Table 1. Clearance data
mEqr
UNa +
Hi min '
GFR
Baseline
n
pBvsF
(paired /)
Frusemide
n
957 + 79
18
>0.13
1053 + 77
18
V
CH,O
92 ±8
18
HS
45 + 5
18
>0.27
283 + 26
18
37 + 8
18
1
mosm kg
i
U/P in
Uosm
Posm
19 + 5
15
HS
145 + 10
18
HS
291+6
18
>0.1
11.6+1.0
18
HS
80 + 2
17
266 + 7
18
301+4
18
4.3+0.4
17
Overall clearance data of segmental reabsorplion
ml min~l GFR x 100
PR
Vf
72.4 + 2.1
27.6 + 2.1
CH2O
5.2 + 0.7
CH2O-BD
CH2O-T
17.2+1.9
' 22.4 + 2.0
CH,O-HL
18.8+1.4
CH2O-DT
3.6 + 0.7
not cause any change in SNGFR, percentage, and
absolute rate of reabsorption. The SNGFRs measured
during B were significantly correlated with the paired
values obtained in the same 50 tubules during F (R =
0.82, P<0.00\). The mean difference between these
paired values was —0.3 + 2.7 nl/min, which indicates
an error close to that of clearance techniques. The
Table also reports on the left hand side the data from
paired collections during B and re-collections sampled
during F at the same ED tubular sites. We obtained
26 paired samples in the 18 animals. In these 26 tubules
frusemide did not change SNGFR, while percentage
reabsorption fell slightly. SNGFR measured during B
Results
was significantly correlated with the paired values
obtained in the same 26 tubules during F (R = 0.72,
Table 1 reports the whole kidney clearance data. P<0.00\).
During B, GFR was not significantly different from
Table 3 shows the paired data obtained by paired
the paired measurements obtained during F. The urine
+
+
collections
from ED and then from LP sampling sites
flow rate (V), the urinary Na concentration (UNa )
+
the
same
nephrons: 36 paired samples were collected
of
and the urine Na excretion rose strikingly during F.
during
B,
and
other 36 paired samples were obtained
Table 1 also shows the additional clearance data that
from
an
equal
number
of different nephrons during F.
can be calculated with our method [2]. These data, as
outlined in the Subjects and Methods section, reflect Again, the main figures derived from this table confirm
reabsorption by the different segments of the nephron that SNGFR is not different when measured at ED
calculated indirectly, without direct micropuncture with respect to LP sites both during B and F. The
assessment. The average values measured, factored by fractional reabsorptions also confirm the results
showed by the previous Table: there is a significant
GFR, are very close to those obtained in humans.
Table 2 shows on the right hand side the paired data difference during B between ED and LP samples (87 + 2
which vanishes during F
measured in 50 tubules at the LP sampling sites. F did versus 70 + 3%, P< 0.0001)
(79±3 versus 73 ±2%, />>0.08). Finally, the absolute
rates of reabsorption (AR) measured at the end of the
Table 2. Paired collections-re-collections
proximal tubules are not different by unpaired / test.
Table 4 reports the mean values generated by a single
ED
LP
average number computed on each rat, during B and
F. This was necessary to calculate the correlations
nl/min
PR
nl/min
PR
between micropuncture and clearance data. Even this
SNGFR
SNGFR
different clustering of the results confirms those of
Baseline
87 + 2
36.3 + 2.2
46.0 + 6.7
72 + 3 previous tables.
50
26
26
50
The delivery out of the proximal tubule can be
>0.91
>0.67
p B vs F
>0.05
<0.001
computed both by clearance techniques (as
(paired /)
Frusemide
75±3
48.1 ±5.9
36.7±2.2
77±2 Vf/GFR x 100) and by micropuncture (as 100 minus
percentage proximal resorption during F). In Tables 4
All measurements were processed statistically. The
means ± the standard error of the mean were computed, and
their differences tested by paired or unpaired t test.
Regressions and correlations between clearance data, micropuncture data, and between data obtained with both micropuncture and clearance techniques were performed by
standard procedures, and their significance tested. Tests for
normal distributions were also performed. All statistical
analyses were carried out by using the Stat Work statistical
program applied to a Macintosh LC personal computer, and
by the BMDP Statistical Software package applied to a AST
Bravo 386SX/20 personal computer.
278
E. Bartoli ei al.
Table 3. Paired ED-LP collections
ED
n
P ED vs LP
(paired t)
LP
n
Baseline
Frusemide
unpaired P values: B vs F
nl/min
nl/min
SNGFR
AR
PR
SNGFR
AR
PR
SNGFR
AR
PR
34.0 + 2.5
36
>0.07
29.8 + 2.3
36
>0.59
87 + 2
36
<0.06
36.5 + 3.1
36
>0.83
28.3 + 2.4
36
>0.66
79 + 3
36
>0.08
>0.52
>0.65
< 0.005
40.2 + 3.0
36
28.4 + 2.6
36
70 + 3
36
37.2 + 3.0
36
27.3+2.3
36
73 + 2
36
>0.47
>0.75
>0.40
Table 4. Overall micropuncture data calculated by average data in individual animals
nl/min
SNGFR
P ED vs LP
(paired i)
Baseline
n
PBvsF
Frusemide
n
AR
Mean
ED
40.0 + 2.2
17
>0.85
39.5 + 2.7
17
> 0.520
38.8 + 2.8
40.6 + 2.8
17
18
>0.65
>0.49
38.5 + 2.3
40.4 + 3.9
17
18
LP
ED
PR
ED
LP
ED
27.7 + 2.8
18
<0.02
23.8 + 2.1
18
88 + 2
17
< 0.008
80 + 2
17
30
CL
V/GFR F
Fig. 1. Correlation between proximal deliveries measured by micropuncture and clearance techniques. In the abscissa are the percentage
deliveries of filtrate beyond the proximal tubule measured by clearance techniques as the urine flow rates during F (V/GFR F, the
percent urine flow excreted during F), plotted against the paired
average values, obtained in each rat, by micropuncture techniques
(LP-delivery F, calculated as 100 minus percentage proximal reabsorption). The regression line traced is calculated with zero intercept.
The correlation coefficient is significant (see Table 5).
HS
HS
12.9+1.6
17
<0.01
19.3 + 2.3
17
72 + 3
18
<0.02
76 + 2
18
ies (R = 0.67, P<0.002), since these are not significantly different (P>0.10) while they are significantly
correlated (R = 0.52, P<0.04). Finally Table 5 reports
the correlations found in the present experiments
between different measurements. The table indicates
both dependent and independent variables, the correlation coefficients and their significance.
Table 5. Correlations
measurements
U_
LP
LP
>0.195
29.2 + 2.5
33.9 + 2 .6
17
18
>0.37
>0.89
31.6 + 2.8
28.9+1.7
17
18
and 1 these two figures average 27 + 2 and 24 + 2%
respectively, and they are not significantly different by
paired t test (/ ) >0.10). Moreover, they are significantly
correlated (R = 0.49, / > <0.04), as shown in Figure 1.
Similar results are obtained by plotting the delivery
out of the distal tubule during F (R = 0.69, P<0.002),
or the mean values between proximal and distal deliver-
% delivery beyond
between
micropuncture
and
clearance
Independent
variable(X)
Dependent
variable(Y)
R
P
n
SNGFR
SNGFR-LP
LP Delivery
Average delivery
Average delivery
Average delivery
V/GFR F
V/GFR F
V/GFR F
V/GFR F
V/GFR F
V/GFR F
DPR B
D-PR B
DPR B
Vb
GFR
SNGFR-ED
ED delivery F
CH2O-BD
CH2O-HL
CH2O-DT
LP delivery F
ED delivery F
Average delivery
CH2O-BD
CH2O-HL
CH2O-DT
CH2O-BD
CH2O-HL
CH2O-DT
CH2O/>
CH2O/
C 0sm6
COsra/
0.351
0.370
0.520
0.470
0.610
0.470
0.489
0.690
0.672
0.882
0.958
0.786
0.577
0.663
0.516
0.657
0.800
0.794
0.591
< 0.040
< 0.001
< 0.040
< 0.050
< 0.007
< 0.050
< 0.040
< 0.002
< 0.002
< 0.001
< 0.001
< 0.001
< 0.020
< 0.005
< 0.040
< 0.003
<0.001
< 0.001
<0.0I0
18
17
17
18
18
18
18
17
18
18
18
18
17
17
17
18
18
18
18
v/
Vb
v/
Segmental Na + transport in the rat nephron
Discussion
The aim of the present study was to obtain, simultaneously from the same animals, clearance and micropuncture data during WD. This was necessary in order
to ascertain whether the calculations of solute reabsorption by proximal tubule, loop of Henle and distal
tubule with a method applicable during WD in humans
[1,2], could be verified in vivo by direct sampling from
appropriate nephron sites. This technique requires
performing clearance periods before and during F, and
it would be unreliable if the renal haemodynamics were
unstable from baseline WD to the superimposed effect
of the diuretic. Our data demonstrate that both whole
kidney GFR and SNGFR were stable during the study,
possibly because of the effect of urine reinfusion, which
ensured a constancy in bodyfluidcompartments during
the experiment, as evidenced by the stability of the
haematocrit. An additional important finding was that
the SNGFR was not different when measured at the
ED with respect to the LP sampling site, in agreement
with a previous study [4]. This allows a meaningful
comparison of reabsorptions without the confounding
effect of a difference in filtration along the length of
the nephron.
The analysis of data obtained by micropuncture
with respect to those calculated by clearance measurements is useful, as it may indicate that a clearance
technique, applicable to the bedside, can give informations on segmental nephron function as detailed as
those requiring a strictly experimental technique, like
micropuncture. With respect to this goal, two results
seem important.
Firstly, F did not affect absolute and fractional
proximal reabsorption. The data were unequivocal
whether they were obtained in the overall nephron
population sampled (27.7 + 1.3, n = 95 versus
27.6+1.5 nl/min, n = 96, P>0.96) or limited to the
more meaningful paired comparison of the LP samples
collected before and recollected after F from the same
site (Table 2). Since this was the basic assumption on
which our method was based, the demonstration of a
lack of proximal effect of F, at least at the dose used
and under the conditions of the present experiments,
seems particularly meaningful. It is also in agreement
with the findings from us [3] and others [5-8].
The second relevant result is given by the fact that,
during the action of F, the fractional delivery of filtrate
out of the proximal tubule and the fractional renal
excretion of water are not significantly different and
are significantly correlated (Figure 1). The proximal
delivery, according to the different numbers chosen
(overall, paired or unpaired data) differs by 1-3% from
the fractional urine flow rate, a close match well within
the experimental error of the techniques. Since the
proximal reabsorption was the same during water
diuresis alone (B) and water diuresis during F, it is
entirely warranted from these data to compute proximal reabsorption by the difference between GFR and
the urine flow rate measured during F. The scatter of
279
the data of Figure 1 is compatible with the experimental error of the techniques used and with interanimal variability. To generate this correlation we needed
to perform calculations in individual animals (Table 4),
where the coefficient of variability of micropuncture
data is 20% with 5 or more measurements [9,10], This
is the reason that we needed repeated measurements
in each rat, and a fairly large number of experiments.
It is important to recognize that all correlations predicted by the theory were significant (Table 5). To
reach the statistical significance required a large
number of animals both in the previous [3] and the
present study, while the scatter of data in each regression is within the limits of error of the methods. These
correlations, however, are not intended to extrapolate
exact values from rats to humans. They merely demonstrate that the coincidence of the means values calculated independently with clearance and micropuncture
techniques stems from the same physiological events.
Such coincidence is then accounted for by the same
general theory and it is also present in individual
animals, though with some scatter.
Thus the present study demonstrates that our
method described in humans [1,2] and reproduced in
these animals allows the calculation of reabsorption
by different nephron segments which more closely
approximates true estimates as compared to more
widely used methods of clinical physiology based on
WD alone without the use of F. This conclusion is
unequivocal and stems from the clearance values
(Table 1), from the paired data obtained at the ED or
LP tubules (Table 2), and from the paired distal-proximal collections-re-collections performed during B and
during F respectively (Table 3). It would be important
in the future to compare our method with the lithium
technique [11], which cannot be applied during the
action of F [8].
This conclusion can be reached even without having
attained a maximal water diuresis in our animals.
Maximal water diuresis is necessary in humans to
calculate exactly the numerical values of free water
clearances. The aim of the present experiments in
animals was only to compare, during water diuresis,
the free water clearances calculated by clearance with
those measured by micropuncture. This comparison is
independent of the attainment of maximal water diuresis, though this is necessary in man to compute
reproducible values and to compare them in different
clinical conditions and disease states. To the same
extent the extrapolation to human studies is not intended for the numerical values, even though it is rewarding to find them very similar. What can be extrapolated
to human studies is the internal consistency of the
technique that can then be applied to humans. The
true normal values for man can be obtained only in
•experiments of clinical physiology at the bedside
[12-14]. Owing to the similarity of transport processes
and of the general functioning of the countercurrent
system, it is likely that F will not affect proximal
reabsorption even in humans and that its blockade of
E. Bartoli et al.
280
+
Na transport on Henle's loop will abolish the interstitial hypertonicity and the abstraction of CH2O-BD.
An additional point to be discussed concerns proximal reabsorption, which cannot be measured at the
very end of the proximal tubule, which is not accessible
to sampling. Therefore, the true percentage proximal
absorption must lie somewhere between LP and ED
values. During F, which does not affect LP values, the
estimates are very close and they match the fractional
water excretion. However, ED reabsorption is higher
than LP before F, indicating frusemide-inhibitable
volume absorption either along the thin descending
limb or the early distal convoluted tubule. Since F
abolishes the medullary and papillary hypertonicity
[15], it will abolish volume absorption along the thin
descending limb: this volume, in our theory, will be
computed as CH2O-BD. However, volume absorption
along this segment was demonstrated only in juxtamedullary nephrons [16], while our experiments could be
performed only in superficial nephrons. Some volume
absorption along the early distal segment, upstream to
the sampling site, is possible even during WD, since
the fluid leaving the thick ascending limb of Henle's
loop is hypotonic [17] and a large transtubular osmotic
gradient is present across an epithelium which, in the
distal tubule, is to some extent water permeable even
in the absence of ADH [18,19]. By abolishing Henle'
loop transport [6,18] F will abolish even this gradient
and the attendant volume flow. Therefore the disappearance of the ED versus LP difference in reabsorption during F can be accounted for by these events,
which are predicted by our theory.
These conclusions are valid only to the extent that
Henle's loop transport is completely abolished by the
drug concentrations used. The pertinent studies indicate that this occurs at tubular fluid concentrations
[20,21] which are reached in man with the dose of
1 mg/kg plus maintenance [22]. Since the putative
proximal effect of F was reported at much higher
concentrations [21], a 10-fold higher dose was chosen
in this study. Since we could not demonstrate a proximal effect of F even with the present high amounts,
we are confident that the loop transport was completely
abolished and that our technique is valid, due to a
lack of proximal effect of F, over the wide drug
concentration range that can be encountered in clinical studies.
Thus the present work demonstrates that, due to the
existence of a large volume of free water dissipated
even during water diuresis (CH2O-BD), the free water
generated (CH2O-T) is much larger than that excreted,
and the latter then underestimates by more than 50%
the true distal reabsorption, which our improved
method more correctly approximates. Tables 1 and 5
show that the reabsorption by Henle's loop, given by
CH2O-HL as the equivalent volume of free water
generated, corresponds to that estimated by micropuncture techniques. This value is much higher than
the entire distal reabsorption estimated by CH2O alone,
as suggested by the traditional approach.
Thus with these experiments we obtained all the
information necessary to compare independently, on
the same animals and under the same experimental
conditions, the clearance measurements on which our
calculations of Na + reabsorption are based, and the
direct measurements with micropuncture. We demonstrated that F shows no proximal effect, in agreement
with others [5-8], that reabsorption from ED and LP
segments are almost superimposable during F and
WD, and that the estimates of proximal fractional
reabsorption and of free water back diffusion with the
two methods are practically identical and significantly
correlated. Finally, micropuncture estimates of
CH2O-DT can be inferred by.our data and they are in
reasonable agreement with clearance data.
Thus, Na+ reabsorption and solvent flow can be
estimated in vivo, in man and animals, along different
nephron segments even without micropuncture.
Acknowledgements. This work was financially supported by grants
from Consiglio Nazionale delle Ricerche, Rome, Italy, and from the
Ministero dell'Universita e Ricerca Scientifica, 40% (Rome, Italy)
and 60% (Universita di Udine, Italy).
References
1. Bartoli E, Branca GF, Satta A, Faedda R. Sodium reabsorption
by the Henle loop in humans. Nephron 1987; 46: 288-300
2. Bartoli E, Satta A, Faedda R, Olmeo NA, Soggia G, Branca G.
A furosemide test in the functional evaluation of the human
nephron in vivo. J Clin Pharmacol 1983; 23: 56-64
3. Romano G, Favret G, Bartoli E. Micropuncture study of the
effect of furosemide on proximal and distal tubules of the rat
nephron. Renal Phisiol Biochem 1995; 18: 209-218
4. Bartoli E. Earley LE. Measurements of nephron filtration rate
in the rat with and without occlusion of the proximal tubule.
Kidney Int 1973; 3: 372-380
5. Kirchner KA. Prostaglandin inhibitors alter loop segment chloride uptake during furosemide diuresis. Am J Physiol 1985; 248:
F698-F704
6. Taniguchi J, Shirley DG, Walter SJ, Imai M. Simulation of
lithium transport along the thin segment of Henle's loop. Kidney
Int 1993; 44: 337-343
7. Tucker BJ, Blantz RC. Effect of furosemide administration on
glomerular and tubular dynamics in the rat. Kidney Int 1984;
26: 112-121
8. Walter SJ, Shirley DG. Effect of furosemide on lithium clearance
and proximal tubular reabsorption in anaesthetized rats.
J Phvsiol 1991; 437: 85-93
9. Oken DE. Wolfert AI, Laveri LA, Choi SC. Effects of intraanimal nephron heterogeneity on studies of glomerular
dynamics. Kidney Int 1985; 27: 871-878
10. Jackson B, Oken DE. Internephron heterogeneity of filtration
fraction and disparity between protein- and hematocrit-derived
values. Kidney Int 1982; 21: 309-315
11. Thomsen K, Holstein Rathlou NH, Leyssac PP. Comparison of
three measures of proximal tubular reabsorption:lithium clearance, occlusion time, and micropuncture. Am J Physiol 1991;
241: F348-F355
12. Satta A, Faedda R, Olmeo NA, Soggia G, Branca GF, Bartoli
E. Studies on the nephron, segment with reduced sodium
reabsorption during starvation natriuresis. Renal Physiol 1984;
7: 283-292
13. Zoccali C, Bartoli E, Curatola G, Maggiore Q. The renal tubular
defect of Bartter's syndrome. Nephron 1982; 32: 140-148
14. Satta A, Faedda R, Chiandussi L, Bartoli E. Fluid and electrolytes in liver disease. Postgrad Med J 1983; 59(4): 64-72
15. Beck FX, Sone M, Dorge A, Thurau K. Effect of loop diuretics
on organic osmolytes and cell electrolytes in the renal outer
medulla. Kidney Int 1992; 42: 843-850
Segmental Na + transport in the rat nephron
16. De Roufignac C, Morel F. Micropuncture study of water,
electrolytes and urea movements along the loops of Henle in
psammomys. J Clin Invest 1969; 48: 474-486
17. Gottschalk CW, Mylle M. Micropuncture study of the mammalian urinary concentrating mechanisnrevidence for the countercurrent hypothesis. Am J Physiol 1959; 196: 927-936
18. Greger R, Velasquez H. The cortical ascending limb and the
early distal convoluted tubule in the urinary concentrating
mechanism. Kidney lnt 1987; 31: 590-596
19. Schaefer JA, Andreoli TE. Cellular constraints to diffusion: the
effect of antidiuretic hormone on water flow in isolated mammalian collecting tubules. J Clin Invest 1972; 51: 1264-1278
281
20. Holzgreve H. Action of diuresis in chronic renal insufficiency.
In: Siegenthaler W, Beckerhoff R, Vetter W, ed. Diureticts in
Research and Clinics. Georg Thieme, Stuttgart, 1977; 178-183
21. Meng K, O'Dea K. Peritubular and intraluminal concentration
of diuretics effecting isotonic fluid absorption in the kidney
tubule. Pharmacology 1973; 9: 193-200
22. Bartoli E, Arras S, Faedda R, Soggia G, Satta A, Olmeo NA.
Blunting of furosemide diuresis by aspirin. J Clin Pharmacol
1980; 20: 452-458
Received for publication: 17.7.95
Accepted in revised form: 23.10.95