0013-7227/02/$15.00/0
Printed in U.S.A.
The Journal of Clinical Endocrinology & Metabolism 87(4):1743–1749
Copyright © 2002 by The Endocrine Society
GH Increases Extracellular Volume by Stimulating
Sodium Reabsorption in the Distal Nephron and
Preventing Pressure Natriuresis
GUDMUNDUR JOHANNSSON, YRSA BERGMANN SVERRISDÓTTIR, LARS ELLEGÅRD,
PER-ARNE LUNDBERG, AND HANS HERLITZ
Research Center for Endocrinology and Metabolism (G.J.), Department of Clinical Neurophysiology (Y.B.S.), Department of
Clinical Nutrition (L.E.), Department of Clinical Chemistry (P.-A.L.) and Department of Nephrology (H.H.), Sahlgrenska
University Hospital, Göteborg SE-413 45, Sweden
Although sodium retention and volume expansion occur during GH administration, blood pressure is decreased or unchanged. The aim was to study the effect of short- and longterm GH replacement in adults on sodium balance, renal
hemodynamics, and blood pressure. Ten adults with severe
GH deficiency were included into a 7-d, randomized, placebocontrolled, cross-over trial followed by 12 months of open GH
replacement. All measurements were performed under metabolic ward conditions. Extracellular water (ECW) was determined using multifrequency bioelectrical impedance analysis. Renal plasma flow and glomerular filtration rate were
assessed using renal paraminohippurate and Cr51 EDTA
clearances, respectively. Renal tubular sodium reabsorption
was assessed using lithium clearance. Plasma renin activity
(PRA), plasma concentrations of angiotensin II, aldosterone,
atrial natriuretic peptides and brain natriuretic peptides
(BNP) and 24-h urinary norepinephrine excretion were mea-
T
HE SODIUM- AND water-retaining effect of GH has
been known for decades (1). Although the exact mechanism underlying the antinatriuretic action of GH is not fully
elucidated, several direct and indirect mechanisms have
been suggested. GH increases serum and tissue levels of
IGF-I, and both GH and IGF-I receptors are expressed in
renal tubules (2), making direct sodium- and water-retaining
effects of GH and IGF-I possible and plausible (3).
Indirect mechanisms for the sodium-sparing effect of GH
have also been proposed to be mediated by an interaction
with the renin-angiotensin-aldosterone system (RAAS) in
some (4 –7) but not all studies (8). Moreover, short-term GH
and IGF-I administration, respectively, has been reported to
suppress plasma atrial natriuretic peptide (ANP) and to impair the ANP response to a saline load, which might contribute to sodium retention (9, 10). This is not, however, a
consistent finding in all studies (8, 11).
GH may also increase glomerular filtration rate (GFR) and
renal plasma flow (RPF) (12, 13) by directly influencing the
local IGF-I and nitric oxide (NO) production in the kidney
Abbreviations: ANP, Atrial natriuretic peptide; BNP, brain natriuretic peptide; CLi, lithium clearance; Cr-EDTA, Cr51 EDTA; ECW,
extracellular water; GFR, glomerular filtration rate; Li, lithium; MF-BIA,
multifrequency bioelectrical impedance analysis; NE, norepinephrine;
NO, nitric oxide; PAH, paraminohippurate; PRA, plasma renin activity;
RAAS, renin-angiotensin-aldosterone system; RPF, renal plasma flow.
sured. Seven days of GH treatment decreased urinary sodium
excretion. Lithium clearance as a marker of proximal renal
tubular sodium reabsorption was unaffected by GH treatment. ECW was increased after both short- and long-term
treatment. This increase was inversely correlated to the decrease in diastolic blood pressure (r ⴝ ⴚ0.70, P ⴝ 0.02) between
baseline and 12 months. Short-term treatment increased PRA
and decreased BNP. The increase in PRA correlated with an
increase in 24-h urinary norepinephrine excretion (r ⴝ 0.77,
P < 0.01). Glomerular filtration rate and renal plasma flow did
not change during treatment. The sodium- and water-retaining effect of GH takes place in the distal nephron. The sustained increase in ECW in response to GH is associated with
an unchanged or decreased blood pressure. This together
with unchanged or decreased atrial natriuretic peptides and
BNP may prevent pressure-induced escape of sodium. (J Clin
Endocrinol Metab 87: 1743–1749, 2002)
(14, 15) or more indirect by increasing the extracellular water
(ECW) and plasma volume. GH may therefore affect both
renal hemodynamics and renal tubular function.
Despite the volume expansion induced by GH administration, blood pressure is reported to be either unchanged
(16 –21) or decreased (22, 23) in response to GH treatment. A
probable explanation is the decrease in peripheral resistance
(22), observed following GH replacement therapy, which in
turn may, at least in part, be explained by increased endothelial NO formation (21).
Thus, GH and IGF-I affect volume and pressure regulation
in a complex manner. The aim of this trial was to study the
effect of GH replacement on sodium and fluid retention by
exploring the short- and long-term effect of GH treatment on
ECW, sodium balance, renal hemodynamics, and blood pressure in adults with severe GH deficiency.
Materials and Methods
Patients
Ten adults (nine males and one female) with adult-onset hypopituitarism were recruited consecutively among patients being considered
for GH replacement therapy. The median age was 53 yr (range 48 – 69
yr), and their body mass index at entry was 27.3 kg/m2 (range 18.9 –30.2).
The hypopituitarism was a result of nonfunction pituitary adenoma or
its treatment in seven patients, and in three patients it was because of
treated Cushing’s disease, empty sella syndrome, and idiopathic hypopituitarism. Three of the 10 patients had panhypopituitarism and two
had isolated GH deficiency. A GH peak of less than 3 ␮g/liter during
1743
1744
J Clin Endocrinol Metab, April 2002, 87(4):1743–1749
insulin-induced hypoglycemia was used to confirm the GH deficiency.
Subjects with renal disease, hypertension, diabetes mellitus, previous
stroke, or polyneuropathy were not eligible for the study.
Study protocol
The study was designed as a two-phase trial. The initial phase of the
study was a 7-d randomized, double-blind, placebo-controlled, crossover trial with a 4-wk washout period in between followed by 12 months
of open GH treatment. Randomization was performed at the clinical trial
section at the Sahlgrenska University Hospital Pharmacy. The dose of
GH during the double-blind period was 9.5 ␮g/kg per day, and the dose
of GH during the open phase was individualized to normalize the serum
IGF-I level (24). Three months before the 12-month visit, the daily dose
of GH was kept stable. Other hormonal replacement therapy for hypopituitarism, such as glucocorticoids, l-thyroxine, and gonadal steroids, was kept stable for at least 3 months before entering the trial. Other
medication was not allowed.
Five patients were randomly allocated to receive GH in the first
period and placebo in the second period, and five patients were randomized to receive treatment in the reverse order. Before and at the end
of each treatment period in the placebo-controlled phase and after the
12-month open treatment period, the patients spent 3 d in a metabolic
ward unit. Collection of urine was initiated in the morning of d 1 and
ended in the morning of d 3, and the results presented are the mean of
two 24-h sampling periods. On the second day, multifrequency bioelectrical impedance analysis (MF-BIA) and an oral glucose tolerance test
using an oral glucose load of 75 g were performed. On the third day, after
an overnight fast and before leaving bed, blood samples were collected
and blood pressure measured in both the supine and after 30 min in the
upright position. On the third morning, measurements of renal paraminohippurate (PAH), renal Cr51 EDTA (Cr-EDTA), and lithium clearances
were performed.
Body weight was measured daily in the morning to the nearest 0.1 kg.
Body height was measured barefoot to the nearest 0.01 m. Body mass
index was calculated as the weight in kilograms divided by the height
in meters squared. Systolic and diastolic blood pressure was measured
to the nearest 5 mmHg using the sphygmomanometric cuff method.
Metabolic ward regimen
Three days before each metabolic ward period, the patients were
given sodium chloride capsules to keep the sodium intake constant. The
metabolic ward dietitian made a food history interview to customize the
metabolic ward menu for each patient. During the 3 d at the metabolic
ward, the patients were given a strictly controlled menu with the same
food items. Only the food on the menu was allowed, and the patients
were encouraged to eat all food that was served. All food was prepared
under metabolic ward kitchen conditions and weighed to nearest 0.1 g
on a scale (Sigma, St. Louis, MO). The same batches of food were used
for all metabolic ward periods for each patient, and food was kept deep
frozen until day of consumption. Intake of sodium was regulated to
median 150 mmol/d (range 149.75–150), intake of potassium was regulated to median 69 mmol/d (range 67–76), and median protein content
was 88 g (range 75–108 g) during the metabolic ward period.
ECW
ECW was determined using MF-BIA (25). In short, reactance and
resistance were determined by using a Xitron 4000B Bio-Impedance
spectrum analyzer (Xitron Technologies, San Diego, CA). The equipment was calibrated daily. Resistance and reactance were measured at
50 frequencies from 5kHz to 500kHz. Data were analyzed using a computer program supplied by the manufacturer (BIS 4000 system utility
version 1.00D) in which a semicircular function is fitted to the data in
a Cole-Cole plot. The resistances at frequency zero and infinity are
predicted and correspond to extracellular resistance and total body
water resistance, respectively (25). These predictions combined with
body weight, height, and resistivity of extracellular and intracellular
water are then used to calculate the ECW, intracellular and total body
water volume based on equations in the supplied computer program.
MF-BIA has been found to be a valid method for indirect measurement
of ECW in both healthy adults and adults with GH deficiency (26, 27).
Johannsson et al. • GH and Renal Sodium Handling
Renal hemodynamics
RPF and GFR were assessed using PAH and Cr-EDTA clearances,
respectively. The technique of continuous infusion and urine collection
was used. The patient initially received a priming dose of Cr-EDTA
(0.6⫻ body surface area ⫽ megabecquerel) and PAH (0.04⫻ body
weight/ml 20% solution). The bolus doses were followed by an iv
infusion of both at a rate of 0.83 ml/min to produce a plasma concentration of 500 counts/min per milliliter and 50 –100 ␮mol/liter of CrEDTA and PAH, respectively. The subjects were initially hydrated with
tap water (10 ml/kg body weight) to ensure diuresis. When urine flow
was established, the priming doses of Cr-EDTA and PAH was administered. The equilibration period started when the subject had voided
(approximately 45 min). The patients were supine throughout the study
but were allowed to stand up to void for each urine collection. This
procedure resulted in a complete bladder emptying according to ultrasound examination. Thereafter four 60-min periods followed in which
the subjects emptied the bladder at the end of each period. Between the
periods they drank the same volume of water as that of urine passed in
the preceding period. The mean of four measurements was used for the
renal hemodynamic assessment. Plasma and urine were assayed for
PAH and Cr-EDTA. Clearance values were expressed per 1.73 m2 body
surface area.
Renal tubular sodium reabsorption
At 2100 h the day before assessment, 600 mg (16.2 mmol) of lithium
(Li) carbonate was administered orally. Blood and urinary samples were
collected at 1000 h the next day and then every clearance period for
analysis of serum Na and Li. The mean serum value for each clearance
period was used in the calculation of renal clearance (milliliter per
minute). On the basis of the assumption that Li is absorbed solely in the
proximal tubules and to the same extent as Na and water, Li clearance
(CLi) equals the output of isotonic fluid from the proximal tubule (28).
Plasma and urinary concentrations of Li were measured by atomic
absorption and Na concentrations by flame photometry. Fractional Li
excretion was calculated as CLi/GFR.
Other assays
ANP and brain natriuretic peptide (BNP) were measured from
plasma that was instantly chilled and centrifuged at 4 C after collection
and thereafter stored at ⫺70 C until they were assayed. ANP and
BNP were measured using a solid-phase immunoradiometric assay
(SHIONOGI & Co. Ltd., Osaka, Japan) with a within-run coefficient of
variation of 6.3% (mean 18.9 ng/liter) and 2.5% (mean 22.1 ng/liter) and
detection limits of 2.5 ng/liter and 2.0 ng/liter, respectively.
RIA was used for determination of plasma renin activity (PRA; ReninRIA bead, Abbot Diagnostics Division, South Pasadena CA) and plasma
aldosterone (DiaSorin, Inc., Saluggia, Italy). Plasma angiotensin II concentration was assayed according to the methods of Kappelgaard et al.
(29) and Morton and Webb (30). PRA and angiotensin II had within-run
coefficient of variation of 8.8% and 5.1%, respectively.
The serum concentration of IGF-I was determined by a hydrochloric
acid-ethanol extraction RIA using authentic IGF-I for labeling (Nichols
Institute Diagnostics, San Juan Capistrano, CA). Serum insulin was
determined using a RIA (Phadebas, Pharmacia, Uppsala, Sweden), and
blood glucose was measured by the glucose-6-phosphate dehydrogenase method (Kebo Lab, Stockholm, Sweden). Urinary norepinephrine
(NE) and metoxycatecholamines were measured using HPLC.
Ethics
After oral and written information, informed consent was obtained
from all the patients. The Ethics Committee at the University of Göteborg
approved the study.
Statistical analysis
All analyses made in the blinded phase of the trial were performed
before the treatment code was broken. All descriptive statistical results
are presented as the median and the 25th and 75th percentiles. The
Wilcoxon’s matched pairs signed rank sum test was used to compare the
effects of GH treatment with the effects of placebo and baseline values
Johannsson et al. • GH and Renal Sodium Handling
J Clin Endocrinol Metab, April 2002, 87(4):1743–1749 1745
with values obtained after 12 months of open GH replacement therapy.
A carry-over effect was sought for by comparing baseline values of the
GH period with baseline values in the placebo period in the five subjects
who were first randomized to GH treatment. Correlations were sought
by calculating the Spearman rank coefficient. Significance was obtained
if the probability value was 0.05 or less.
Results
All subjects completed the blinded cross-over phase and
the 12-month open GH treatment without any noticeable side
effects. No carry-over effect was detected in the five subjects
who were randomized to GH during the first treatment period. The median daily dose of GH was 0.83 mg (range
0.50 –1.00) during the placebo-controlled period and 0.33
mg/day (range 0.27– 0.83) after 12 months of GH treatment.
Serum concentration of IGF-I increased in response to both
short-term treatment and 12 months of treatment of open GH
replacement (Table 1). The area under the curve for blood
glucose and plasma insulin during the 2-h oral glucose tolerance test were both increased after the short-term GH treatment (P ⬍ 0.01 for each, respectively) but returned to baseline
values following the 12-month GH replacement therapy
(data not shown). Median body weight increased (P ⬍ 0.05)
following the short-term GH period [0.7 (0.0 –1.0) kg], compared with placebo [0.1 (⫺0.8 – 0.7) kg], but it was unchanged
after the long-term treatment.
treatment in the controlled part of the trial (r ⫽ ⫺0.4, P ⫽ 0.3).
Blood pressure, 24-h urinary excretion of NE, aldosterone,
PRA, and angiotensin II
The supine diastolic blood pressure decreased from a median value of 88 (range 80 –90) mm Hg at baseline to 78 (range
70 – 80) mm Hg at 12 months of open treatment (P ⫽ 0.02).
During the short-term treatment, there was no significant
change in diastolic blood pressure [85 (range 80 –100) to 75
(range 70 –90) mm Hg], compared with placebo treatment [83
(range 70 –90) to 80 (range 70 – 80) mm Hg). The 24-h urinary
excretion of NE increased in response to short-term GH
treatment, compared with placebo (Table 1). This increase
was not sustained after 12 months of treatment. The 24-h
ECW, serum sodium, and 24-h urinary sodium excretion
During the short-term GH treatment, the ECW increased
and the 24-h urinary sodium excretion decreased, compared
with placebo (Table 1). Twelve months of GH replacement
therapy maintained the increased ECW, and the 24-h urinary
sodium excretion did not differ significantly from the baseline levels. The changes in ECW and diastolic blood pressure
were inversely related following 12 months of GH treatment
(Fig. 1) and a similar pattern was seen during the 7 d of
FIG. 1. The inverse correlation between changes in extracellular
fluid volume (ECW) and diastolic blood pressure between baseline
and 12 months of GH treatment in 10 hypopituitary adults. The
Spearman rank coefficient was calculated.
TABLE 1. Measurements of serum IGF-I, ECW and 24-h urinary sodium and norepinephrine excretion, and atrial and brain natriuretic
peptides during GH/placebo treatment in a randomized double blind crossover 7-d period, followed by 12 months of open replacement
therapy
Measure
Baseline
IGF-I (␮g/liter)
GH
Placebo
ECW (kg)
GH
Placebo
24-h U-sodium (mmol)
GH
Placebo
24-h NE (mmol/creat)
GH
Placebo
Atrial natriuretic peptide (ng/liter)
GH
Placebo
BNP (ng/liter)
GH
Placebo
119 (72–116)
125 (107–165)
128 (72–167)
21.2 (19.5–22.6)
20.9 (19.5–23.1)
21.9 (18.8 –23.5)
172 (123–190)
143 (123–166)
152 (106 –188)
25.0 (20.6 –36.4)
22.4 (16.1–26.0)
19.9 (18.2–25.5)
18.7 (10.1–26.8)
19.0 (12.1–26.6)
18.2 (9.1–24.6)
8.6 (4.8 –19.9)
9.7 (4.7–19.6)
6.9 (5.3–20.3)
Day 7
Median treatment
response
12 months
272 (212–318)a
372 (262– 433)
112 (94 –145)
22.5 (20.4 –24.7)
22.3 (19.2–23.8)
128 (106 –141)
169 (116 –188)
231 (206 –278)b
⫺4 (⫺22 to 3)
22.3 (20.7–23.9)a
1.1 (0.8 –1.6)b
0.3 (0.2– 0.5)
⫺13.8 (⫺43.6 –⫺6.7)b
6.0 (⫺12.7 to 30.9)
129 (120 –159)
21.6 (15.7–28.7)
26.4 (20.0 –29.1)
20.2 (18.2–22.4)
4.1 (1.0 –7.3)b
⫺1.1 (⫺1.9 – 0.0)
16.6 (11.7–19.0)
24.6 (9.8 –30.2)
⫺1.6 (⫺10.2 to 0.9)
1.8 (0.3–9.4)
7.5 (2.0 –14.5)
11.9 (4.3–19.8)
⫺1.9 (⫺2.7 to 0.0)b
0.5 (⫺0.7 to 4.5)
10.4 (7.0 –16.3)
5.6 (2.0 –9.0)
The first line reflects the baseline and 12-month values for all patients. Values are median and 25th and 75th percentiles.
P ⬍ 0.01 as compared with baseline.
b
P ⬍ 0.01 as compared with changes during the placebo treatment.
a
1746
J Clin Endocrinol Metab, April 2002, 87(4):1743–1749
Johannsson et al. • GH and Renal Sodium Handling
TABLE 2. Measurements of the RAAS in supine position and after 30 min of upright position during GH/placebo treatment in a
randomized double blind crossover 7-d period, followed by 12 months of open replacement
Measure
Supine
PRA (ngAI/ml䡠h)
GH
Placebo
Angiotensin II (pg/ml)
GH
Placebo
Aldosterone (pg/ml)
GH
Placebo
Upright posture
PRA (ngAI/ml䡠h)
GH
Placebo
Angiotensin II (pg/ml)
GH
Placebo
Aldosterone (pg/ml)
GH
Placebo
Baseline
0.54 (0.21– 0.70)
0.58 (0.26 – 0.71)
0.56 (0.21–1.04)
2.4 (2.0 –3.2)
2.9 (2.0 – 4.0)
3.8 (2.3–5.0)
92 (75–113)
81 (68 –113)
100 (75–118)
0.96 (0.35–3.02)
0.96 (0.26 –1.55)
1.67 (0.46 –3.02)
4.7 (2.0 –10.0)
4.7 (4.0 –5.8)
6.7 (3.5–10.0)
195 (121–240)
155 (121–247)
178 (101–240)
Day 7
Median treatment
response
12 months
0.77 (0.64 – 0.92)
0.94 (0.49 –1.29)
0.53 (0.20 –1.14)
0.29 (0.11– 0.78)a
⫺0.07 (⫺0.13 to 0.01)
4.9 (2.0 –7.2)
4.0 (2.0 – 4.6)
0.2 (0.0 – 4.6)
⫺0.1 (⫺1.3 to 1.9)
4.8 (2.0 – 6.4)
71 (41–96)
78 (52–97)
⫺9 (⫺38 to 4)
⫺3 (⫺12 to 0)
89 (78 –131)
1.42 (0.71–3.23)
3.26 (0.68 –3.64)
1.55 (0.64 –2.07)
1.98 (0.34 –2.37)a
⫺0.43 (⫺0.70 to 0.18)
5.8 (3.8 –9.8)
10.1 (5.5–13.5)
5.7 (3.0 –9.0)
6.2 (1.2–7.7)a
⫺0.2 (⫺5.0 to 1.0)
183 (168 –258)
216 (137–252)
174 (121–235)
40 (8 –132)
11 (⫺40 to 30)
The first line reflects the baseline and 12-month values for all patients. Values are median and 25th and 75th percentiles.
P ⬍ 0.01 as compared with changes during placebo treatment.
a
rise in stimulated PRA and angiotensin II (r ⫽ 0.77, P ⬍ 0.01)
following 7 d of GH treatment. No correlation, however, was
found between supine PRA and supine angiotensin II (r ⫽
0.14) or between 24-h urinary NE excretion and angiotensin
II (r ⫽ ⫺0.1) after the short-term GH treatment.
ANP and BNP (Table 1)
FIG. 2. The positive correlation found between changes in PRA and
24-h urinary NE excretion between baseline and 12 months of GH
replacement in 10 hypopituitary adults. The Spearman rank coefficient was calculated.
urinary excretion of aldosterone was not affected by the
short- or long-term GH treatment (data not shown).
The supine values of PRA increased in response to shortterm GH treatment, compared with placebo but not with
baseline values after 12 months of treatment (Table 2). The
supine plasma concentrations of angiotensin II and aldosterone were not affected by the treatment (Table 2). In standing
position, PRA and plasma concentration of angiotensin II
increased following 7 d of GH treatment, compared with
placebo, whereas plasma aldosterone concentration did not
change. The stimulated increment in PRA and angiotensin II
elicited by moving from supine to standing position was
more marked during GH treatment than during placebo
(P ⬍ 0.05).
A positive correlation was found between the increase in
PRA and 24-h urinary NE excretion (Fig. 2) and between the
Plasma ANP concentration tended to be reduced in response to GH treatment, compared with placebo (P ⫽ 0.06)
(Fig. 3A). A similar tendency was seen at 12 months, compared with baseline (P ⫽ 0.07). Plasma BNP concentration
decreased in response to GH, compared with placebo treatment (Fig. 3B), and tended to be reduced at 12 months,
compared with baseline (P ⫽ 0.09). No correlations were
found between changes in ANP or BNP and in ECW or any
other measurement in this trial.
Renal hemodynamics and tubular sodium reabsorption
GFR, RPF, CLi, and fractional Li excretion were all unaffected by both short- and long-term GH treatment. The filtration fraction increased in response to short-term GH treatment, compared with placebo, but this effect was not
sustained after 12 months of treatment (Table 3).
Discussion
The major finding in this study is that the sodium-retaining effect of GH seems to occur in the distal nephron. With
unchanged or even decreased blood pressure levels together
with decreased or unchanged plasma concentrations of natriuretic peptides, pressure or escape natriuresis does not
occur, allowing for increased ECW both during short- and
long-term GH replacement.
Decreased urinary sodium excretion in the presence of
unchanged CLi indicates that the increased renal tubular
sodium reabsorption in response to GH treatment took place
Johannsson et al. • GH and Renal Sodium Handling
FIG. 3. The median and 25th to 75th percentile change in ANP (A)
and BNP (B) in response to 7 d of GH and placebo in a randomized
cross-over trial with a 4-wk washout period in 10 hypopituitary
adults. *, P ⬍ 0.05, compared with placebo.
mainly in the distal nephron. In contrast to previous trials
(22, 31, 32), renal hemodynamics did not change in response
to GH treatment in this trial. Several factors may explain this
difference. Previous trials have used pharmacological doses
for a short time, whereas we have used more appropriate
replacement doses with less marked increment and lower
final serum levels of IGF-I. The standardized sodium and
protein intake during the period of measurements in this trial
is unique and may also help to explain the contrasting finding in this study, compared with previous ones. Of importance in this study, however, is the marked effect of GH/
IGF-I on renal tubular sodium reabsorption despite the lack
of effects on renal hemodynamics.
The enhanced tubular sodium reabsorption and the increased ECW did not offset pressure natriuresis, which
would be anticipated for elimination of the sodium load (33,
34). Diastolic blood pressure was reduced during the GH
treatment, an effect most likely explained by increased NO
formation and reduced peripheral vascular resistance (21,
22). The inverse relationship found between the decrease in
diastolic blood pressure and the increase in ECW between
baseline and 12 months of GH treatment favor blood pressure reduction as a potential mechanism in preventing the
J Clin Endocrinol Metab, April 2002, 87(4):1743–1749 1747
pressure natriuresis reported with other sodium-retaining
agents (34). Moreover, the escape from the sodium-retaining
action of mineralocorticoids is coincident with increased concentration of plasma ANP suggesting that natriuretic peptides play a role in this phenomenon (35). The trend for
reduced plasma ANP and BNP concentrations may therefore
be of importance in preventing sodium and water escape
during GH replacement. Increase ratio between total body
nitrogen and body cell mass has been observed (36), suggesting extracellular proteins to increase during GH administration. This may also contribute to the sustained increase
in ECW during more prolonged GH treatment and in patients with acromegaly (1).
Several factors may contribute to the increased renal tubular sodium reabsorption in response to GH. Our trial together with a previous one (8), also using a more appropriate
dose of GH for replacement in adults, does favor that stimulation of RAAS may be of some importance for the sodiumretaining effect of GH. The mechanism does not seem to be
a stimulation of the adrenal cortex (37) because plasma aldosterone levels did not change in response to GH. The
primary effect may be increased PRA, which in turn will
increase serum angiotensin II levels and thereby increase
directly the sodium retaining effect in the renal tubules. The
main shortcoming of this explanation is that the sodiumretaining effects of angiotensin II takes place mainly in the
proximal tubule (38), but our result form the lithium clearance suggests that this takes place in the distal nephron. More
recent lines of evidence suggest, however, that angiotensin
II directly increases sodium reabsorption in the more distal
part of the nephron (39).
Although contrasting effects in terms of angiotensin II and
aldosterone occur in previous studies, a consistent finding is
the increased PRA in response to GH administration (6 – 8, 11,
40). In genetically GH-deficient Lewis rats as well as in man,
GH administration increases plasma angiotensinogen concentration (5, 8), indicating that PRA is increased secondary
to increased renin substrate. GH treatment restores the attenuated renin secretion response to hypotension in hypophysectomized rats (4), and in this study it results in a more
marked rise in PRA and angiotensin II in response to an
orthostatic test. Renin release is amplified by renal sympathetic activity and circulating catecholamines (41). Our relationship between the increase in 24-h urinary NE excretion
and the increase in PRA may therefore indicate that shortterm GH treatment increases renal sympathetic activity,
which in turn may be responsible for the increase in PRA.
Although urinary NE excretion is not a specific measure of
renal sympathetic nervous function (42, 43), the increase in
urinary NE seen following the short-term GH treatment may
indicate an acute selective increase in renal sympathetic activity because sympathetic outflow to other vascular beds
remains unaffected by short-term GH treatment (Sverrisdóttir, Y. B., personal communication).
The decrease in insulin sensitivity and increase in insulin
levels during the short-term GH treatment may also have
contributed to the increased urinary excretion of NE (44).
However, all the above-mentioned effects are lost during
long-term GH replacement suggesting that the more prolonged effects of GH/IGF-I are mediated through direct ac-
1748
J Clin Endocrinol Metab, April 2002, 87(4):1743–1749
Johannsson et al. • GH and Renal Sodium Handling
TABLE 3. Measurements of renal hemodynamics and renal tubular function during GH/placebo treatment in a randomized double blind
crossover 7-d period, followed by 12 months of open replacement therapy
Measure
Baseline
GFR (ml/min)
GH
Placebo
RPF (ml/min)
GH
Placebo
Filtration fraction (%)
GH
Placebo
Li clearance (ml/min)
GH
Placebo
Fract. Li excretion (%)
GH
Placebo
91 (81–98)
94 (79 –103)
91 (83–98)
491 (427– 495)
520 (423–570)
468 (427– 495)
0.190 (0.184 – 0.200)
0.185 (0.168 – 0.190)
0.192 (0.190 – 0.200)
33.7 (25.6 –34.2)
33.7 (25.9 –34.8)
33.3 (28.3–35.0)
37 (29 –38)
33 (29 – 40)
34 (28 – 43)
Day 7
Median treatment response
12 months
89 (79 –98)
95 (87–103)
94 (77–109)
3 (⫺4 to 6)
⫺2 (⫺9 to 13)
460 (435–565)
493 (376 – 605)
⫺12 (⫺106 to 60)
25 (⫺26 to 112)
487 (398 –515)
0.194 (0.183– 0.196)
0.193 (0.189 – 0.205)
0.191 (0.187– 0.199)
0.023 (0.007– 0.031)a
⫺0.007 (⫺0.011 to 0.009)
33.6 (25.1–36.8)
33.7 (30.5–34.9)
32.0 (27.2–39.7)
30 (29 –37)
35 (29 – 41)
0.0 (⫺3.6 to 2.9)
3.4 (⫺0.3 to 6.3)
37 (29 – 42)
⫺1 (⫺4 to 0)
1 (⫺3 to 6)
The first line reflects the baseline and 12-month values for all patients. Values are median and 25th and 75th percentiles.
P ⬍ 0.01 as compared with changes during the placebo treatment.
a
tions on the distal nephron and not through the interaction
with renal sympathetic activity and RAAS. This is supported
by unchanged or even reduced activity of the RAAS in acromegaly patients (45, 46).
In contrast to most previous trials that have studied the
relationship between GH/IGF-I and natriuretic peptides, we
have used a solid-phase immunoradiometric assay to measure mature ANP and BNP (47). Both peptides affect blood
pressure, renal hemodynamics, renal tubular function, and
sodium and water homeostasis (48). Competitive blockage of
ANP and BNP results in increased blood pressure, increased
plasma concentrations of PRA, aldosterone, and catecholamines (49). It is therefore plausible that the reduction
seen in plasma ANP and BNP in response to short-term GH
treatment is a primary event that may explain both the increase in PRA and urinary NE excretion. The small reduction
observed in plasma ANP and BNP concentration is not likely
to solely explain the sodium-retaining effects in response to
GH treatment. This may be supported by the lack of any
association between changes in these peptides and changes
in ECW and urinary sodium excretion.
This study, performed under metabolic ward conditions,
suggests that the sodium- and water-retaining effect of GH
takes place mainly in the distal nephron by a direct action of
GH/IGF-I because other plausible indirect mechanisms are
only modestly or transiently affected. Of major importance
for the sustained increase in ECW in response to GH is the
unchanged or decreased blood pressure, which prevents
pressure-induced natriuresis. We may hypothesize that GH/
IGF-I reduces natriuresis and increases ECW and, by its
parallel action on blood pressure and plasma natriuretic peptides, prevents escape natriuresis.
Acknowledgments
We are indebted to dietitian Birgitta K. Lundgren of the Metabolic
Ward for excellent contribution to this study and the personnel at the
Endocrine Ward, the Research Center for Endocrinology and Metabolism, and the Research Laboratory of Nephrology at Sahlgrenska University Hospital for skillful technical support. Pharmacia kindly provided the GH and placebo preparations during the blinded phase of the
trial.
Received July 27, 2001. Accepted January 3, 2002.
Address all correspondence and requests for reprints to: Gudmundur
Johannsson, M.D., Ph.D., Research Center for Endocrinology and Metabolism, Sahlgrenska University Hospital, SE-413 45 Göteborg, Sweden.
E-mail: [email protected].
This work was supported by grants from the Swedish Medical Research Council (Project no. 11621, 12170, and 13192), Swedish Society for
Medical Research, and Swedish Heart and Lung Foundation.
References
1. Ikkos D, Luft R, Sjögren B 1954 Body water and sodium in patients with
acromegaly. J Clin Invest 33:989 –994
2. Chin E, Zhou J, Bondy CA 1992 Renal growth hormone receptor gene expression: relationship to renal insulin-like growth factor system. Endocrinology 131:3061–3066
3. Blazer-Yost BL, Cox M 1988 Insulin-like growth factor 1 stimulates renal
epithelial Na⫹ transport. Am J Physiol 255:C413–C417
4. Honeyman TW, Goodman HM, Fray JCS 1983 The effects of growth hormone
on blood pressure and renin secretion in hypophysectomized rats. Endocrinology 112:1613–1617
5. Wyse B, Waters M, Sernia C 1993 Stimulation of the renin-angiotensin system
by growth hormone in Lewis dwarf rats. Am J Physiol 265:E332–E339
6. Ho KY, Weissberger AJ 1990 The antinatriuretic action of biosynthetic human
growth hormone in man involves activation of the renin-angiotensin system.
Metabolism 39:133–137
7. Møller J, Møller N, Frandsen E, Wolthers T, Jørgensen JOL, Christiansen JS
1997 Blockade of the renin-angiotensin-aldosterone system prevents growth
hormone-induced fluid retention in humans. Am J Physiol 272(5 Pt 1):E803–
E808
8. Hoffman DM, Crampton L, Sernia C, Nguyen TV, Ho KK 1996 Short term
growth hormone (GH) treatment of GH-deficient adults increases body sodium and extracellular water, but not blood pressure. J Clin Endocrinol Metab
81:1123–1128
9. Møller J, Jørgensen JOL, Møller N, Hansen KW, Pedersen EB, Christiansen
JS 1991 Expansion of extracellular volume and suppression of atrial natriuretic
peptide after growth hormone administration in normal man. J Clin Endocrinol Metab 72:768 –772
10. Møller J, Jørgensen JOL, Marqversen J, Frandsen E, Christiansen JS 2000
Insulin-like growth factor I administration induces fluid and sodium retention
in healthy adults: possible involvement of renin and atrial natriuretic factor.
Clin Endocrinol (Oxf) 52:181–186
11. Møller J, Frandsen E, Fisker S, Jørgensen JOL, Christiansen JS 1996 Decreased plasma and extracellular volume in growth-hormone deficient adults
and the acute and prolonged effects of GH administration—a controlled experimental-study. Clin Endocrinol (Oxf) 44:533–539
12. Ikkos D, Ljunggren H, Luft R 1956 Glomerular filtration rate and renal plasma
flow in acromegaly. Acta Endocrinol (Copenh) 21:226 –236
13. Falkheden T, Sjögren B 1964 Extracellular fluid volume and renal function in
pituitary insufficiency and acromegaly. Acta Endocrinol (Copenh) 46:80 – 88
14. Guler HP, Schmid C, Zapf J, Froesch ER 1989 Effects of recombinant insulinlike growth factor I on insulin secretion and renal function in normal human
subjects. Proc Natl Acad Sci USA 86:2868 –2872
Johannsson et al. • GH and Renal Sodium Handling
15. Haylor J, Sing I, El Nahas AM 1991 Nitric oxide synthesis inhibitor prevents
vasodilation by insulin-like growth factor I. Kidney Int 39:333–335
16. Amato G, Carella C, Fazio S, et al. 1993 Body composition, bone metabolism,
and heart structure and function in growth hormone(GH)-deficient adults
before and after GH replacement therapy at low doses. J Clin Endocrinol Metab
77:1671–1676
17. Beshyah SA, Shahi M, Skinner E, Sharp P, Foale R, Johnston DG 1994
Cardiovascular effects of growth hormone replacement therapy in hypopituitary adults. Eur J Endocrinol 130:451– 458
18. Cittadini A, Cuocolo A, Merola B, et al. 1994 Impaired cardiac performance
in GH-deficient adults and its improvement after GH replacement. Am J
Physiol 267:E219 –E225
19. Fort S, Weaver JU, Monson JP, Mills PG 1995 The effects of low-dose recombinant human growth hormone on cardiovascular structure and function
in hypopituitary growth hormone-deficient adults. Endocrinol Metab 2:
119 –126
20. Valcavi R, Gaddi O, Zini M, Iavicoli M, Mellino U, Portioli I 1995 Cardiac
performance and mass in adults with hypopituitarism: effects of one year of
growth hormone treatment. J Clin Endocrinol Metab 80:659 – 666
21. Böger RH, Skamira C, Bode-Böger SM, Brabant G, von zur Mühlen A,
Frölich JC 1996 Nitric oxide may mediate the hemodynamic effects of recombinant growth hormone in patients with acquired growth hormone deficiency.
J Clin Invest 98:2706 –2713
22. Caidahl K, Edén S, Bengtsson B-Å 1994 Cardiovascular and renal effects of
growth hormone. Clin Endocrinol (Oxf) 40:393– 400
23. Johannsson G, Rosén T, Lindstedt G, Bosaeus I, Bengtsson B-Å 1996 Effects
of 2 years of growth hormone treatment on body composition and cardiovascular risk factors in adults with growth hormone deficiency. Endocrinol
Metab 3(Suppl A):3–12
24. Johannsson G, Rosén T, Bengtsson B-Å 1997 Individualized dose titration of
growth hormone (GH) during GH replacement in hypopituitary adults. Clin
Endocrinol (Oxf) 47:571–581
25. Gudivaka R, Schoeller DA, Kushner RF, Bolt MJG 1999 Single- and multifrequency models for bioelectrical impedance analysis of body water compartments. J Appl Physiol 87:1087–1096
26. Ellis KJ, Wong WW 1998 Human hydrometry: comparison of multifrequency
bioelectrical impedance with 2H2O and bromine dilution. J Appl Physiol 85:
1056 –1062
27. van Marken Lichtenbelt WD, Snel YEM, Brummer R-JM, Koppeschaar HPF
1997 Deuterium and bromide dilution, and bioimpedance spectrometry independently show that growth hormone-deficient adults have an enlarged
extracellular water compartment related to intracellular water. J Clin Endocrinol Metab 82:907–911
28. Thomsen K 1984 Lithium clearance: a new method for determining proximal
and distal tubular reabsorption of sodium and water. Nephron 37:217–223
29. Kappelgaard AM, Nielsen MD, Giese J 1976 Measurement of angiotensin II
in human plasma: technical modifications and practical experience. Clin Chim
Acta 67:299 –306
30. Morton JJ, Webb DJ 1985 Measurement of plasma angiotensin II. Clin Sci
(Colch) 68:483– 484
J Clin Endocrinol Metab, April 2002, 87(4):1743–1749 1749
31. Hirschberg R, Rabb H, Bergamo R, Kopple JD 1989 The delayed effect of
growth hormone on renal function in humans. Kidney Int 35:865– 870
32. Jørgensen JOL, Pedersen SA, Thuesen L, et al. 1989 Beneficial effect of growth
hormone treatment in GH-deficient adults. Lancet 1:1221–1225
33. Guyton AC, Coleman TG 1969 Quantitative analysis of the pathophysiology
of hypertension. Circ Res [Suppl I] 34:I1–I14
34. Knox FG, Burnett JCJ, Kohan DE, Spielman WS, Strand JC 1980 Escape from
the sodium-retaining effects of mineralcorticoids. Kidney Int 17:263–276
35. Yokota N, Bruneau BG, Kuroski T, de Bold ML, de Bold AJ 1994 Atrial
natriuretic factor significantly contributes to the mineralocorticoid escape phenomenon. Evidence for a guanylate cyclase-mediated pathway. J Clin Invest
94:1938 –1946
36. Bengtsson B-Å, Edén S, Lönn L, et al. 1993 Treatment of adults with growth
hormone (GH) deficiency with recombinant human GH. J Clin Endocrinol
Metab 76:309 –317
37. Biglieri EG, Watlington CO, Forsham PH 1961 Sodium retention with human
growth hormone and its subfractions. J Clin Endocrinol Metab 21:361–370
38. Harris PJ, Navar LG 1985 Tubular transport responses to angiotensin. Am J
Physiol 248:F621–F630
39. Hall JE, Brands MW, Henegar JR 1999 Angiotensin II and long-term arterial
pressure regulation: the overriding dominance of the kidney. J Am Soc Nephrol 10:S258 –S265
40. Cuneo RC, Salomon F, Wilmhurst P, et al. 1991 Cardiovascular effects of
growth hormone treatment in growth-hormone-deficient adults: stimulation
of the renin-aldosterone system. Clin Sci 81:587–592
41. DiBona GF, Kopp U 1997 Neural control of renal function. Physiol Rev
77:75–197
42. Baines AD, Drangova R 1986 Neural not tubular dopamine increases glomerular filtration rate in perfused rat kidneys. Am J Physiol 250:F674 –F679
43. Kopp U, Bradley T, Hjemdahl P 1983 Renal venous outflow and urinary
excretion of norepinephrine, epinephrine, and dopamine during graded renal
nerve stimulation. Am J Physiol 244:E52–E60
44. Reaven GM, Lithell H, Landsberg L 1996 Hypertension and associated metabolic abnormalities—the role of insulin resistance and the sympathoadrenal
system. N Engl J Med 334:374 –381
45. Karlberg BE, Ottosson AM 1982 Acromegaly and hypertension: role of the
renin-angiotensin-aldosterone system. Acta Endocrinol (Copenh) 100:581–587
46. Cain JP, Williams GH, Dluhy RG 1972 Plasma renin activity and aldosterone
secretion in patients with acromegaly. J Clin Endocrinol Metab 34:73– 81
47. Clerico A, del Ry S, Giannessi D 2000 Measurement of cardiac natriuretic
hormones (atrial natriuretic peptide, brain natriuretic peptides, and related
peptides) in clinical practice: the need for a new generation of immunoassay
methods. Clin Chem 46:1529 –1534
48. Levin ER, Gardner DG, Samson WK 1998 Natriuretic peptides. N Engl J Med
339:321–328
49. Hirata Y, Matsuoka H, Suzuki E, et al. 1993 Role of endogenous atrial natriuretic peptide in DOCA-salt hypertensive rats. Effects of a novel nonpeptide
antagonist for atrial natriuretic peptide receptor. Circulation 87:554 –561