Clinical Science (I997) 92,5 1-56 (Printed in Great Britain)
51
Renal handling of urate and sodium during acute physiological
hyperinsulinaemia in healthy subjects
J. C. TER MAATEN, A. VOORBURG, R. J. HEINE, P. M. TER WEE, A. J. M. DONKER and R. 0. B. GANS
Department of Medicine, ICaR-VU, Cardiovascular Research School, University Hospital Vrije Universiteit,
Amsterdam, The Netherlands
(Received 17 October 1995/21 August 1996; accepted 22 August 1996)
1. The renal effects of insulin may play a central
role in the association between insulin resistance,
hypertension and hyperuricaemia. After a 2-h
baseline period, we investigated the effects of exogenous insulin for 4 h (50 m-units h-' kg-') on
fractional renal sodium and urate excretion in 13
healthy subjects, using the euglycaemic clamp and
lithium clearance technique, and performed a
control experiment in eight of the subjects.
2. Insulin caused a decline in both fractional renal
sodium excretion, from 1.13 f0.41% to 0.88 k0.58%
(control study: 0.81 f0.35 to 1.35 k0.49%; P c 0.001,
insulin versus control), and fractional renal urate
excretion, from 6.72 k 1.87% to 5.71 +2.02% (control
study: 7.03 k2.06 to 7.05 f1.94%; P = 0.085, insulin
versus control). The changes in fractional renal
sodium and urate excretion were positively correlated (r = 0.71, P<O.Ol). Estimated fractional distal
sodium reabsorption increased during insulin
infusion from 93.7 f2.8% to 96.7 f1.9% (control
study: 95.7 f1.5% to 93.6 f1.1%; P <0.001, insulin
versus control). Estimated fractional proximal
tubular sodium reabsorption fell from 81.0 f0.5% to
73.7 f4.7% during insulin infusion, but less in the
control study (81.5 L- 4.3% to 79.3 f4.8%; P = 0.056,
insulin versus control). The changes in fractional
proximal tubular sodium reabsorption and fractional distal sodium reabsorption during insulin
infusion were inversely correlated (r = -0.59,
P = 0.03).
3. During the course of the insulin infusion experiment a n inverse correlation between the changes in
fractional sodium and urate excretion, and the
insulin-mediated glucose disposal, became gradually
evident (r = -0.73,
P<O.Ol, and r = -0.71,
P <0.01, respectively; fourth hour of the insulin
infusion period).
4. We conclude that exogenous insulin acutely
decreases renal sodium and urate excretion, and
that this effect is probably exerted at a site beyond
the proximal tubule.
INTRODUCTION
Insulin resistance is associated with hypertension
and hyperuricaemia [l-51. It has been proposed that
the attendant compensatory hyperinsulinaemia may
contribute to the pathogenesis of elevated blood
pressure and hyperuricaemia through its renal
effects [6, 71. This hypothesis is supported by several
observations. Exogenous administration of insulin
reduces sodium and urate excretion in acute experiments [8-131. In addition, the sodium-retaining
action of insulin has been shown to be preserved in
insulin-resistant states such as essential hypertension, obesity, and non-insulin-dependent diabetes
mellitus (NIDDM) [14-171. Notably, both the
severity of hypertension and the elevation in serum
uric acid levels are related to the degree of insulin
resistance [2, 71. Furthermore, hypertension associated with NIDDM and obesity often displays salt
sensitivity, as does essential hypertension [18-211.
It is not clear where in the nephron insulin exerts
its action on sodium and urate excretion. Insulin's
antinatriuretic effect has been localized in the distal
tubule in most studies [9-12, 17, 22, 231. However,
an increased proximal tubular sodium reabsorption
during insulin administration has also been reported
[24]. On the other hand, the renal handling of urate
occurs almost exclusively in the proximal tubule [25].
Indeed, the results of a recent population-based
study showed that elevated serum urate levels are
independently associated with increased proximal
tubular sodium reabsorption as denoted by
decreased fractional lithium excretion [26]. The
notion that insulin affects sodium in the distal
tubule while urate is handled in the proximal tubule
is difficult to reconcile with the recent observation
that acute, physiological hyperinsulinaemia decreases sodium and urate excretion in a coupled
fashion [13]. Moreover, an increase in fractional
lithium excretion, indicating a decreased proximal
sodium reabsorption, has been demonstrated to
accompany insulin-induced sodium retention in
several studies of similar design [ l l , 12, 17, 22,231.
Key words: insulin, insulin resistance, lithium, renal sodium excretion, urate.
Abbreviations: ANOVA. analysis of variance: FE, fractional excretion; NIDDM, non-insulin-dependentdiabetes mellitus.
Correspondence: Dr J. C. ter Maaten, Department of Medicine, University Hospital Vrije Universiteit, De Boelelaan I I 17, 1081 HV Amsterdam, The Netherlands.
52
J.C.ter Maaten et al.
As the relationship between the effects of insulin
on renal urate handling and segmental tubular
sodium reabsorption has not been studied, we
assessed the effect of acute, physiological hyperinsulinaemia on renal sodium, lithium and urate
handling in healthy subjects.
METHODS
Subjects
Thirteen healthy male Caucasian subjects, mean
age 21 (range 18-23) years, were studied. All were
normotensive (blood pressure less than 140/90
mmHg) with a mean weight of 74.2 (range
64.3-88.0) kg, a body mass index (meanfSD) of
21.8f1.7 kg/m2 and a waist-hip ratio (meanfSD)
of 0.84f0.04, and none was taking medication.
Three subjects had a first-degree relative and four
others a second-degree relative with hypertension,
and two subjects had a second-degree relative with
NIDDM. Normal glucose tolerance according to
World Health Organziation criteria was confirmed
by an oral glucose tolerance test. Informed consent
was obtained from all subjects. The protocol had
been approved by the local ethics committee, and
the study was carried out in accordance with the
Declaration of Helsinki.
In the week before the insulin infusion study and
the control study all subjects adhered to a diet
containing 200 mmol of sodium; the extra amount of
sodium, necessary in all but one subject, was
supplied by capsules containing 8.5 mmol of sodium
chloride. Compliance with the diet was confirmed by
measurement of 24-h urinary sodium excretion
during the last 2 days before each study.
Insulin infusion experiment
The subjects received 300 mg of lithium carbonate
orally at 22.00 hours the evening before the insulin
infusion and control study. For practical reasons the
experiments were performed in the afternoon.
Therefore, subjects were allowed to eat one slice of
bread and to drink one cup of tea without sugar at
08.00 hours. No food or drinks except water were
allowed thereafter. This intake will not perturb the
fasting state to a significant extent. All subjects
refrained from smoking. The subjects came to the
outpatient clinic at noon. Studies were conducted in
a room with a constant temperature of 22°C. Polytetrafluoroethylene cannulae (Venflon; Viggo,
Helsinborg, Sweden) were inserted for intermittent
blood sampling and infusions as described previously
PI.
At the beginning of the study the subjects were
asked to empty their bladder. They drank 300 ml of
tap water each hour to ensure adequate diuresis.
After a basal clearance period of 2 h, four timed
urine collections of 1 h were performed. During the
entire study the subjects were in a supine position,
except when voiding.
After the basal period of 120min, a 4-h euglycaemic hyperinsulinaemic clamp was performed.
Insulin (Velosulin; NovoNordisk, Bagsvaerd,
Denmark), diluted to 50ml with 45ml of 0.9%
sodium chloride and 5 m l of 20% human albumin,
was infused at a rate of 50 m-units h-' kg-l after a
priming dose (0.1 x kg x the desired rise in plasma
insulin level in m-units) had been given. Fasting
blood glucose level was maintained by adjustment of
the infusion rate of a 20% D-glucose solution (1.11
mol/l), using a manual method as described previously [27]. The blood glucose level was measured
every 5 min with an automated glucose oxidase
method (Yellow Springs Instruments, Yellow
Springs, OH, U.S.A.).
Systolic and diastolic blood pressure and heart
rate were measured every 10 min with a semicontinuous blood pressure-measuring device (Nippon
Colin BP 103 N Sphygmomanometer, Hayashi,
Komaki-City, Japan).
Control experiment
Only eight of the subjects were willing and available for the control experiment. This experiment
was carried out in an identical fashion as the insulin
infusion study with infusion of the same amount of
insulin solvent and with blood sampling at the same
time intervals, including blood sampling for glucose
measurements. Control experiments had to be
performed after the insulin clamp experiments
because we could not determine beforehand the
amount of 20% glucose to be infused each hour to
maintain euglycaemia. To correct for any (nonspecific) change in sodium excretion or change in
blood pressure, heart rate or both, due to volume
expansion as the result of 20% glucose infusion to
maintain euglycaemia during the insulin infusion
experiment, a corresponding amount of water was
given orally each hour. Absorption of water from
the gastrointestinal tract is rapid and complete, so
that no relevant lag in volume expansion was to be
expected when compared with intravenous administration of 20% glucose. Control experiments
enabled us to correct for any circadian variation in
the variables under evaluation.
Blood samples for measurements of the various
substances were drawn halfway through each clearance period. Urine and serum concentrations of
sodium, potassium, urate and creatinine were determined by standard laboratory techniques. Serum
and urinary concentrations of lithium were
measured by atomic absorption (Atomic Absorption
Spectrophotometer, Perkin Elmer, Nonvalk, CT,
U.S.A.). Blood samples for measurement of plasma
insulin were drawn four times during the second and
fourth hour of the insulin infusion period. Plasma
insulin concentrations were measured by radio-
Insulin and urate excretion
immunoassay (Immunoradiometric Assay, Medgenix
Diagnostics, Fleurus, Belgium). A quantitative
estimate of insulin sensitivity was provided by the
mean glucose infusion rate in the second, third, and
fourth hour of the euglycaemic clamp (M value (mg
min-' kg-I)) [28] and expressed per unit of plasma
insulin concentration (MII value), thereby correcting
for differences in steady-state plasma insulin levels
[29]. For convenience, the MII ratio was multiplied
by 100. To calculate the MII value during the second
and fourth hour of the hyperinsulinaemic clamp, we
used the average value of the four plasma insulin
concentrations obtained during each of these
periods. The average of the plasma insulin concentrations during the second and fourth hour was used
to calculate the MII value during the third hour of
the clamp.
53
ance period had been shown not to differ. Correlation between changes in the fractional excretion
rates was performed using the average of the values
obtained during the second, third and fourth clearance periods versus the baseline values. A P-value
of ~ 0 . 0 5 was considered significant. Data are
expressed as means f:SD, unless stated otherwise.
All analyses were performed on a personal
computer using the statistical software package
SPSS version 6.0 (SPSS, Chicago, IL, U.S.A.).
RESULTS
The subjects showed a good compliance with the
diet.The average urinary sodium excretion per day
amounted to 244 f:52 mmol before the insulin
infusion study (n = 13) and 210+42 mmol before
the control experiment (n = 8). Clamp characteristics and insulin sensitivity variables are shown in
Table 1. The M-value increased from the second to
the fourth hour of the clamp. The MII value did not
change because plasma insulin levels demonstrated
a small, albeit insignificant (P= 0.10; two-sample
t-test), increase over the infusion period. Mean
blood glucose levels during the control experiment
(n =8) were not different from those during the
insulin clamp experiment (n = 13).
Measurements of plasma concentrations of
solutes during the experiments are listed in Table 2.
Plasma potassium declined significantly during
insulin infusion. Plasma urate showed a similar
decline during both studies. The clearance data are
shown in Tables 3 and 4. No changes in creatinine
clearance were observed during both experiments
(P = 0.29). Fractional excretion rate of sodium
(FENa) decreased by 22% during the clamp
(P<O.OOl), whereas an increase of 66% (P = 0.002)
was noted during the control experiment as
compared with baseline. Fractional excretion of
lithium (FELi) increased during the clamp,
indicating that proximal tubular sodium reabsorption decreased. FELi also tended to increase over
time during the control study. Estimated fractional
distal tubular sodium reabsorption increased during
the clamp but decreased during the control experiment. The changes in estimated distal and proximal
Calculations
Sodium, potassium urate, lithium and creatinine
clearances were calculated according to standard
formulas. Fractional clearances were preferred to
absolute clearances, because they correct for
changes in glomerular filtration rate as well as for
dead space or incomplete voiding. Hence, fractional
proximal tubular sodium reabsorption was calculated as 1-((ClithiumjCcreatinine) x loo%, and the
fractional distal tubular sodium reabsorption as
1- (Csodium/Clithium)X 100%.
Statistical analysis
Data are expressed as means of the measurements
obtained during each clearance period. All variables
were analysed by the method of analysis of variance
(ANOVA) for repeated measurements to detect
differences over time, and differences between the
control and study days. Correlation and linear
regression analysis and Fisher's test for the
comparison of correlations were applied when
appropriate. Correlation between the fractional
excretion rates of solutes was performed using the
average of the values obtained in all clearance
periods after data point distributions in each clear-
Table 1. Clamp characteristics and insulin sensitivity variables. Values are means_+SDand an+ed
time (ANOVA).
Characteristic
Insulin infusion study
Mean glucose concentration (mmoVI)
coefficient of variation (%)
Insulin concentration (pmoVI)
M-value (mg min-' kg-I)
M/I d u e (mg min-' kg-' per prnoVl x IOO)
Control experiment
Mean glucose concentration (mmoVI)
Baseline
4. I & 0.4
32.0 f 7.9
4.2 & 0.3
2h
4. I & 0.5
9.8k2.1
270.2 & 59.2
10.4 If:4.4
3.94_+ 1.79
4. I _+ 0.2
3h
4.0k0.3
9. I f 2.3
to detect differences over
4h
P-value
(4.09k 1.53)
4.1 k 0 . 4
9.7+ 1.8
293.9k48.5
12.4 k3. I
4.29_+ 1.18
<0.001
0.005
0.45
4.2k0.4
4.2 f0.5
0.73
-
I 1.3 k 3 . 7
0.23
J.C. ter Maaten eta].
54
Table 2. Plasma concentrations of solutes at baseline and during hyperinsulinaemic euglycaemic clamp (n = 13) and timecontrol studies (n = 8). Values are meansf SD. Column A denotes significant changes over time during the separate studies, and
column B denotes significant differencesover time between the insulin and control studies.
P-due
Baseline
Sodium (mmolll)
Insulin
Control
Ih
2h
3h
4h
B
A
0.08
0.87
140k1.3
140+ 1.3
140+ 1.7
140f 1.5
140+ 1.1
140k 1.6
141 f2.0
140f 1.0
141 f2.2
14Of1.1
Potassium (mmolil)
Insulin
Control
4.02k0.29
3.98& 0.24
3.77f0.25
4.00+0.30
3.42+0.18
3.79f0.28
3.31 k0.25
3.76k0.30
3.38f0.30
3.68k0.33
<0.001
Urate (pmolil)
Insulin
Control
345 f46
323 k29
336f41
317k33
328 k42
299 f 28
321 f47
291 f35
312+48
286 k35
<0.00I
0.46
0.12
0.003
<0.001
0.40
M/I value became apparent. The correlations with
the M/I value reached significance during the third
hour of the clamp for the changes in both FEN,
(second hour: r = -0.31. P = 0.30; third hour:
r = -0.56, P = 0.048; fourth hour: r = -0.73,
P<O.Ol) and FEur (second hour: r = -0.44,
P = 0.13; third hour: r = -0.56, P = 0.049; fourth
hour: r = -0.71, P<O.Ol). Thus, the higher the
insulin-mediated glucose disposal, the more sodium
and urate was retained (Figs 4 and 5).
Both systolic and diastolic blood pressure
increased more during the insulin infusion study
(from 120.5k6.8 to 129.1k8.6 mmHg and from
62.7 f 6.0 to 68.9 k7.4 mmHg respectively) than
during the control experiment (from 127.0& 7.5 to
131.2k10.4 mmHg and from 67.9k6.0 to 69.8f8.8
mmHg respectively; P = 0.04 and P = 0.03, insulin
sodium reabsorption during the insulin infusion
study were inversely correlated (r = -0.59, P = 0.03;
Fig. 1).
As expected, changes in the fractional excretion of
urate (FEU,) paralleled changes in fractional excretion of sodium (FEN,) throughout the clamp
(r=O.71, P<O.Ol; Fig. 2). In addition, FEur and
FELi were positively correlated during the insulin
infusion experiment (r = 0.70, P<O.Ol; Fig. 3).
However, in contrast to the increase in FELi, FEur
decreased during the hyperinsulinaemic clamp.
Compared with the control experiment, the decline
in FEU, approached statistical significance
(P = 0.085). Notably, the changes in FEur were not
related to changes in FELi.
During the course of the study, an inverse correlation between the changes in FEN, and FEur, and the
Table 3. Clearance data of solutes at baseline and during hyperinsulinaemic euglycaemic clamp (n = 13) and time-control
studies (n = 8). Values are means+SD. Abbreviations: C, clearance: FE, fractional excretion; I -FEU, fractional proximal tubular
sodium reabsorption; I -(CNJCU), fractional distal tubular sodium reabsorption. Column A denotes significant changes over time
during the separate studies, and column B denotes significant differencesover time between the insulin and control studies.
P-due
B
Baseline
Ih
2h
3h
4h
A
Urine flow (ml/min)
Insulin
Control
4.52f2.5 I
7.21 k3.53
4.99 f I.87
10.6k3.40
8.92& 2.73
7.24 If:I .64
8.73 f2.97
8.39 fI.85
7.80f4. I7
8.26f2.60
<0.00I
C N (ml/min)
~
InsuI in
Control
I .59& 0.46
I.36k0.46
1.65 f0.66
1.55 50.56
I .I0 k0.62
I.80f 0.62
I.I4 k0.70
I.89f0.75
I .28f0.82
2.23 0.82
<0.00I
0.002
<0.001
145k23
176532
153f23
167+20
145+26
171 13
*
144f19
l57f I 8
147k22
165f20
0.64
0.23
0.29
0.075
<0.001
CO (mllmin0
Insulin
Control
FEN, (%)
InsuIin
Control
FEK(96)
Insulin
Control
I.I3 f0.41
0.81 k0.35
1.11 kO.44
0.93 5 0.32
0.77k0.39
1.06k0.38
0.81 k0.51
1.21 f0.48
0.88kO.58
1.35 k0.49
<O.OOl
18.7 f 7.0
18.6 f7.2
16.5k6.5
17.3k4.9
8.0 k2.9
15.0f4.6
6.6k2.3
12.9 k4.8
6.05 2.9
10.5 k4.5
<O.OOl
0.002
0.005
<O.OOl
0.017
55
Insulin and urate excretion
Table 4. Clearance data of markers of segmental tubular solute reabsorption at baseline and during hyperinsulinaemic
euglycaemic clamp (n = 13) and time-control studies (n = 8). Values are meanskSD. Abbreviations: C, clearance; FE, fractional
excretion; I -FEU, fractional proximal tubular sodium reabsorption; I -(CNJCu), fractional distal tubular sodium reabsorption. Column
A denotes significant changes over time during the separate studies, and column B denotes significant differences over time between
the insulinand control studies.
P-value
FEU(%)
Insulin
Control
Baseline
Ih
2h
3h
4h
A
19.0k5.9
18.5 24.3
22.1 k 4 . 9
19.6 k3.8
24.3k6.5
17.5 k4.0
24.6k5.3
21.4 k5.6
26.3k4.7
20.7 k 4 . 8
<O.OOl
8i.ok5.9
81.5k4.3
n.9k4.9
80.423.8
75.7k6.5
82.5k4.0
75.4k5.3
78.6k5.6
73.7k4.7
79.3k4.8
<0.00 I
93.7k2.8
95.751.5
95.1k1.4
95.3k1.4
96.7k1.9
94.0k1.4
96.8k1.5
94.2k1.8
96.7k1.9
93.6kl.l
<O.OOl
0.001
6.72k1.87
7.03k2.06
6.37k1.35
6.66+ 1.52
5.54k1.56
6.54k 1.59
5.43k1.78
7.06+ 1.98
5.71k2.02
7.05+ 1.94
8
0.09
0.056
I -FEU (%)
Insulin
Control
0.09
0.056
I -(CNJCU) (%)
Insulin
Control
FEU, (%)
Insulin
Control
versus control). We found a weak correlation
between mean arterial pressure at baseline and the
M/Z value (r = 0.48, P = 0.10). Notably, no relations
were observed between the changes in blood
pressure and the changes in F E N a and changes in
segmental tubular sodium reabsorption during the
clamp studies. After correction for changes in mean
arterial blood pressure, the inverse correlation
between the M/Z value and changes in F E N , and
Feur persisted.
<O.OOl
0.00 I
0.73
0.085
DISCUSSION
Recently, it has been shown that acute, physiological hyperinsulinaemia induces a joint reduction
in sodium and urate excretion [13]. In that study,
however, the conclusions regarding both sodium and
urate handling were hampered by the absence of a
time-control experiment and an analysis of
segmental tubular sodium reabsorption [13]. The
results of the present study demonstrate that insulin
increases both sodium and urate reabsorption at a
30
5
20
0
-g-
10
-5
-
0
TI
-20
8
s
-10
1
W
lii
2 -10
d
-W
0
-15
-30
-40
a
-20
-50
-25
I
I
I
I
1
I
- 4 - 2
0
2
4
6
8
I
I
I
1 0 1 2 1 4
delta I-(CNXLI)(%)
Fig. I.Scatterplot showing correlation between changes in I-FEUand
I-(CNJCU). Each value represents the averages of the values obtained
during the second, third and fourth hour of hyperinsulinaemic euglycaemic
clamp studies comparedwith baseline values. r = -0.59, P = 0.03.
-80
I
I
I
I
I
I
60
-40
-20
0
20
40
delta FENa (%)
Fig. 2. Scatterplot showing correlation between changes in the
fractional excretion rates of sodium ( h a ) and urate (FEW). Each
value represents the averages of the values obtained during the second,
third and fourth hour of the hyperinsulinaemic(50 m-units h-' kg-I) euglycaemic clamp studies comparedwith the baseline values. r = 0.71, P<O.Ol.
J. C. ter Maaten et al.
56
40
32
30
20
28
26
h
5
8 24
3
u.
0
5
L
-9
22
g -20
20
18
-40
e @
16
-60
14
3
4
5
6
7
8
9
1
0
Fig. 3. Scatterplot showing correlation between the fractional excretion rates of urate (FEU,) and lithium (FEU). Each value represents the
averages of the values obtained in all timed periods before and during
hyperinsulinaemic euglycaemicclamp studies. r = 0.70, P<O.Ol.
site beyond the proximal tubule, with a concomitant
decrease in estimated fractional proximal tubular
sodium reabsorption.
It is well known that the kidney regulates sodium
and urate reabsorption in a parallel fashion under
several (patho)physiological conditions [30] and that
1
401
20 h
55
0-
i
L
@
\
:
-20 -
U
-80
I
I
1
I
I
I
2
3
4
5
6
7
M/I value (mglkglmin per pmoI/l*lOO)
FEur (%)
6o
1
I
I
I
I
I
I
I
1
2
3
4
5
6
7
M/I value (mglkglmin per pmolll'l00)
Fig. 4. Scatterplot showing correlation between rates of insulinmediated glucose uptake and changes in the fractional excretion
rates of sodium (FENS during the fourth hour of hyperinsulinaemic
euglycaemic clamp studies. Insulin-mediatedglucose uptake is expressed
as the MI1 value (mg min-' kg-' per pmoVl x 100). Changes in FEN, represent the values during the fourth hour of the clamp studies compared with
baseline values. r = 0.73,P<0.021.
Fig. 5. Scatterplot showing correlation between rates of insulinmediated glucose uptake and changes in the fractional excretion
rates of urate (FEU,) during the fourth hour of hyperinsulinaemic
euglycaemic clamp studies. Insulin-mediatedglucose uptake is expressed
as the M/l value (mg min-' kg-' per pmoVl x 100). Changes in FEU, represent the values during the fourth hour of the clamp studies compared with
baselinevalues. r = -0.71, P<O.Ol.
filtered urate is almost totally absorbed in the
proximal tubule [25]. This has been confirmed by
the observation of a clear (positive) correlation
between the fractional excretion rates of lithium and
urate in a population study [26]. We also observed a
similar correlation between FELi and FEur (Fig. 3).
However, during acute insulin administration the
fractional excretion of lithium and urate changed in
opposite directions (Table 4). The cause of this
discrepancy most likely resides in the acute nature
of our experiments as compared with the observational character of the previous studies. It might be
that under chronic conditions renal urate handling is
in some way correlated with the degree of insulin
resistance and not with hyperinsulinaemia per se.
Preliminary evidence suggests that, although the
sodium-retaining effect of insulin is preserved, the
decreased proximal tubular sodium reabsorption, as
denoted by the increased lithium clearance during
acute insulin administration, is absent in insulinresistant conditions [17, 231. The absence of an
increase in atrial natriuretic peptide and/or renal
plasma flow has been implicated as possible
mechanisms. This notion gains some support from
an earlier study among essential hypertensive
patients, which showed that renal blood flow was
lower and renal vascular resistance increased in
patients with high uric acid levels [31].
The renal handling of urate is a complex process,
consisting of glomerular filtration, tubular reabsorption, tubular secretion and post-secretory reabsorption [32, 331. It has been proposed that the final rate
of urate excretion is determined by the last two
Insulin and utate excretion
processes, which may take place both in the late part
of the proximal tubule and in the distal tubule [32,
331. As no appreciable changes in glomerular filtration rate occurred in our study, the finding of a
correlation between the changes in sodium and
urate excretion, together with the observation that
insulin increased distal tubular sodium reabsorption
with a concomitant decrease in proximal sodium
reabsorption, suggests that insulin modifies urate
handling by enhancing the post-secretory reabsorption of urate beyond the proximal tubule. It is less
likely that insulin decreased urate excretion by a
decrease in filtration or tubular secretion as the
glomerular filtration rate did not change and a
similar decrease in plasma urate levels was observed
during the insulin and control experiment. It is also
unlikely that some, possibly hypokalaemia-related,
distal lithium secretion contributed to the increments in the fractional excretion of lithium in the
course of the insulin infusion study [34]. In insulinresistant states, e.g. NIDDM and the nephrotic
syndrome, the absence of a decreased proximal
sodium reabsorption (or increased lithium excretion) has been shown despite the preservation of the
insulin-lowering effect on potassium [17, 231.
Evidence for direct and coupled renal tubular
handling of sodium and urate is lacking. It has been
proposed that the tubular transport of sodium and
urate transport is indirectly coupled by anion
exchange mechanisms [35]. Possibly, insulin
promotes a parallel increase in anion reabsorption,
including urate, by activation of the sodiumhydrogen exchanger [36, 371. One may argue that
lithium is less than an optimum marker for proximal
tubular sodium reabsorption [38]. It is important to
remember that lithium clearance gives an estimate
of end-proximal sodium delivery and not a precise
quantitative measurement [39]. However, we applied
appropriate precautions to improve the usefulness
of the method [39], such as the provision of a
relatively small amount of lithium and performance
of the experiments in the sodium-replete state.
A surprising result of our study is the inverse
correlation between insulin-mediated glucose
disposal and the changes in sodium and urate excretion. In previous studies, comparing groups, similar
antinatriuretic effects have been found in insulinresistant and insulin-sensitive subjects [14, 15, 17, 22,
23, 36, 401. The discrepant observation in the
present study may be related to the longer duration
of the clamp experiment, as the correlation between
insulin’s renal effects and the insulin-mediated
glucose disposal did not reach significance until the
third hour of the clamp. Time-dependent changes in
the renal handling of sodium during insulin infusion
have been documented before [12]. Also, it has been
shown that changes in muscle blood flow do not
reach a steady state until the third hour of an insulin
infusion experimenty [41]. It is possible that the
attenuated antinatriuretic response in the subjects
with the lowest insulin-mediated glucose disposal in
57
our study is related to pre-existing (insulin induced
or insulin resistance related) sodium retention and
volume expansion. Alternatively, our results might
imply that resistance to the glucose-lowering effects
of insulin is coupled with resistance to its antinatriuretic and antiuricosuric effects. This view is
supported by the results of a recent experiment
which showed that the response of the tubular
cation transport to insulin in rats is abolished by
fructose-induced hypertension [42].
We observed a rise in blood pressure during the
clamp, contrary to previous studies during acute,
physiological hyperinsulinaemia. It is unlikely that
the change in blood pressure confounded our results
as changes in sodium excretion were not related to
changes in blood pressure, and because the inverse
relation between the insulin-mediated glucose
disposal and changes in sodium and urate excretion
was still present after correction for the changes in
blood pressure.
In conclusion, insulin acutely decreases both urate
and sodium excretion by an effect beyond the
proximal tubule. The exact localization in the
nephron and the mechanisms involved are not clear.
However, the significance of insulin’s antinatriuretic
and antiuricosuric effects for the pathogenesis of
hypertension and hyperuricaemia in insulin-resistant
states remains to be demonstrated.
ACKNOWLEDGMENT
This study was in part made possible by grant C
94.1378 from the Dutch Kidney Foundation (Nier
Stichting Nederland).
REFERENCES
I. Modan M, Halkin H, Almong S, Lusky A, Eshkol A, Shell M. Hyperinsulinemia.
Alink between hypertension, obesity and glucose intolerance. J Clin Invest 1985;
75: 809- 17.
2. Fenannini E, Bmigoli G, Bonadonna R, et al. Insulin resistance in essential
hypertension. N Engl J Med 1987; 317: 350-7.
3. Modan M, Halkin H, Karasik A, Lusky A. Elevated serum uric acid - a facet of
hyperinsulinemia. Diabetologia 1987; 3 0 713-18.
4. Vumorinen-Markkola H, Yki-JMnen H. Hyperuricemiaand insulin resistance.
J Clin Endocrinol Metab I994;78: 25-9.
5. Cigolini M, Targher G, Tonoli M, Manara F, Muggeo M, Desandre G.
Hyperuricemia: relationships to body fat distribution and other components of
the insulin resistance syndrome in 38-year-old men and women. Int J Obes 1995;
I 9 92-6.
6. De Fronzo RA, Ferrannini E. Insulin resistance: a multifaceted syndrome
responsible for IDDM, obesity, hypertension, dyslipidemia, and atherosclerotic
cardiovascular disease. Diabetes Care 1991; 14: 173-94.
7. Facchini F, Chen YDI, Hollenbeck CD, Reven GM. Relationship between
resistance to insulin-mediated glucose uptake, urinary uric acid clearance, and
plasma uric acid concentration. JAMA I99 I;266 3008-1 I.
8. Gans ROB, van der Toorn L, Bilo HJG, NautaJJP, Heine RJ, Donker AIM. Renal
and cardiovascular effects of exogenous insulin in healthy volunteers. Clin Sci
1991; 8 0 219-25.
9. DeFronzo R4, Cooke CR, Faloona GR, Davis PJ. The effect of insulin on the
renal handling of sodium, potassium, calcium, and phosphate in man. J Clin Invest
1975; 55: 845-55.
10. Skott P, Hother-Nielsen 0, Bruun NE, et al. Effects of insulin on kidney function
and sodium excretion in healthy subjects. Diabetologia 1989; 32: 694-9.
58
I. C.ter Maaten et al.
I I. FriedbergCE, Van Buren M. Bijlsma]A, Koomans HA Insulin increases sodium
reabsorption in diluting segment in humans: evidence for indirect mediation
through hypokalemia. Kidney Int I991;4 0 25 1-6.
12. Stenvinkel P, BolinderJ, Alvestrand A Effects of insulin on renal haemodynamics
and the proximal and distal tubular sodium handling in healthy subjects.
Diabetologia 1992; 3 5 1042-8.
13. Galvan AQ, Natali A, Baldi S, et al. Effect of insulin on uric acid excretion in
humans. Am J Physiol 1995; 2 6 8 E 1-5.
14. Natali A, Galvan AQ, Santoro D, et al. Relationshipbetween insulin release,
antinatriuresis and hypokalaemia after glucose ingestion in normal and
hypertensive man. Clin Sci 1993; 8 5 327-35.
15. Shimamoto K, HirataA, Fukuoka, et al. Insulin sensitivity and the effects of insulin
on renal sodium handling and pressor systems in essential hypertension.
Hypertension 1994; 23 (Suppl. I):1-29-33.
16. RocchiniAP,Katch V, Kveselis D, et al. Insulin and renal sodium retention in
obese adolescents. Hypertension 1989; 1 4 367-74.
17. Gans ROB, Bilo HJG, Donker AIM. The renal response to exogenous insulin in
non-insulindependent diabetes mellitus in relation to blood pressure and
cardiovascular hormonal status. Nephrol Dial Transplant 1996; I I: 794-802.
18. Tuck ML, Cony DB, Trujillo A Salt sensitive blood pressure and exaggerated
vascular reactivity in the hypertensionof diabetes mellitus. Am J Med 1990; 8 8
210- 16.
19. Rocchini AP,Key J, Bondie D, et al. The effect of weight loss on the Sensitivity of
blood pressure to sodium in obese adolescents. N Engl J Med 1989; 321: 580-5.
20. Sullivan JM. Salt-sensitivity. Definition, conception, methodology, and long-term
issues. Hypertension 1991; 17 (Suppl. I):1-61-8.
21. Sharma AM, Schorr U, Distler A Insulin resistance in young salt-sensitive
normotensivesubjects. Hypertension 1993; 21: 273-9.
22. Abouchacra S, Baines AD, Zinman B, Skorecki KL, Logan AG. Insulin blunts the
natriuretic action of atrial natriuretic peptide in hypertension. Hypertension 1994;
23: 1054-8.
23. Stenvinkel P. Ottosson-Seeberger A, Alvenrand A Renal hemodynamics and
sodium handling in moderate renal insufficiency: The role of insulin resistance and
dyslipidemia. J Am SOC Nephrol 1995; 5 I75 1-60.
24. Trevisan R, Fioretto P, Semplicini, et al. Role of insulin and atrial natriuretic
peptide in sodium retention in insulin-treated IDDM patients during isotonic
volume expansion. Diabetes 1990; 39: 289-98.
25. Steele TH. Control of uric acid excretion. N Engl J Med 197 I; 2&1: I 193-6.
26. Cappuccio FP, Stranullo P, Farinaro E, Trevisan M. Uric acid metabolism and
tubular sodium handling. JAMA 1993; 270 354-9.
27. Ponchner M, Heine RJ. Pernet A, et al. A comparison of the artificial pancreas
(glucose-controlled insulin infusion system) and a manual technique for assessing
insulin sensitivity during euglycaemic clamping. Diabetologia 19&1; 2 6 420-5.
28. DeFronzo RA, Tobin ID, Andres R Glucose clamp technique: a method for
quantifying insulin secretion and resistance. Am J Physiol 1979; 237: E214-23.
29. Sherwin RS. h e r KJ, Tobin JD, et al. A model of the kinetics of insulin in man.
J Clin Invest 1974; 5 3 I48 1-92.
30. Holmes WE, Kelley NW, WyngaardenJB. The kidney and uric acid excretion in
man. Kidney Int 1972; 2 115-18.
3 I. Messerli FH, Frohlich ED, Dreslinski GR, Suarez DH, Aristimuno GG. Serum uric
acid in essential hypertension: an indicator of renal vascular involvement. Ann
Intern Med 1980 93: 817-21.
32. Diamond HS, PaolinoIS. Evidence for a postsecretory reabsorptivesite for uric
acid in man. J Clin Invest 1973; 5 2 1491-9.
33. Rieselbach RE, Steele TH. Influence of the kidney upon urate homeostasis in
health and disease. Am J Med 1974; 5 6 665-75.
34. Boer WH, Hen6 RJ, Koomans HA, Dorhout Mees EJ. Discrepancy between
lithium and free water clearance in patients with Bamer’s syndrome. Nephron
1994; 67: 82-7.
35. Kahn AM. Indirect coupling between sodium and urate transport in the proximal
tubule. Kidney Int 1989; 3 6 378-84.
36. BlomstedtJW,Aronson PS. pH gradient-stimulatedtransport of urate and paraaminohippurate in dog renal microvillus membrane vesicles. J Clin Invest 1980;
6 5 93 1-4.
37. Mahnensmith RL, Aronson PS. The plasma membrane sodium-hydrogen
exchanger and its role in physiological and pathophysiological processes. Circ Res
1985; 57: 773-88.
38. Koomans HA, Boer WH, Dorhout Mees EJ. Evaluation of lithium clearance as a
marker of proximal tubule sodium handling. Kidney Int 1989; 3 6 2-12.
39. Shirley DG, Walter SJ. Is lithium a useful tool in studies on renal sodium handling?
J Nephrol 1996; 9 126-32.
40. Endre T, Mattiason I, Berglund G, Hulthh UL. Insulin and renal sodium retention
in hypertension-pronemen. Hypertension 1994; 23: 3 13-1 9.
41. Laakso M, Edelman SV, Baron AD. Decreased effect of insulin to stimulate
skeletal muscle blood flow in obese man. J Clin Invest 1990; 8 5 1844-52.
42. FBraille E, Many S, Barlet-Bas C. et al. Insulin responsiveness of tubular cation
transport during fructose-inducedhypertensionin rats. Clin Sci 1995; 8 8 293-9.