Pediatric Nephrology
Pediatr Nephrol (1991) 5:509-512 9 IPNA 1991
Endocrine changes Review article
Influence of growth hormone and insulin-like growth faetor-I on kidney function and kidney growth Eberhard Ritz1, Burkhard T'6nshoff2, Stefan Worgall 2, Gabor Kovacs 2, and Otto Mehls 2 Departments of Internal Medicine1 and Paediatrics 2, University of Heidelberg, Heidelberg, Federal Republic of Germany Received November 23, 1990; accepted December 28, 1990
Abstract. Decreased glomerular filtration rate (GFR) in hypopituitarism and increased GFR in acromegaly suggest that growth hormone (GH) has a substantial effect on renal haemodynamics. Extractive and recombinant human (rh) GH in healthy volunteers increased effective renal plasma flow (ERPF) and GFR by 10% and 15% respectively. Renal response to GH was delayed and occurred at the same time as an increase in plasma insulin-like growth factor (IGF)-I values, whereas infusion of rhIGF-I promptly increased GFR and ERPF, indicating that the haemodynarnic response of the kidney to GH is mediated by IGF-I. In chronic renal failure (CRF), the acute effect of GH on GFR is obliterated. This might protect the diseased kidney against the undesired consequences of hyperfiltration. Indeed, rhGH treatment for 1 year in children with CRF did not lead to an accelerated decline in GFR compared with the year before treatment. GH and IGF-I also affect renal growth. Exposure to excessive GH in transgenic mice causes renomegaly and progressive glomerular sclerosis. In acromegalic humans, increased renal size and weight and increased glomerular diameter are well known, whereas renal failure is not a long-term hazard. At least in normal and hypophysectomized rats treated with doses comparable with the therapeutic regimens used in stunted children, rhGH increased renal weight but in proportion to the increase in body weight indicating an isometric effect of GH on renal growth. From these data, major renal longterm side effects of rhGH treatment in children with CRF appear unlikely.
Renal function and renal morphology in hypersomatotropism and hyposomatotropism The renal effects of growth hormone (GH), i.e. a decreased glomerular filtration rate (GFR) in hypopituitarism and increased GFR in acromegaly were recognized several decades ago [1, 2]. These could not be explained by changes in renal perfusion pressure or extracellular volume [3]. Falkheden and Sjrgren [3] found increased GFR and effective renal plasma flow (ERPF) in patients with acromegaly, and this was associated with an increased extracellular fluid space. In contrast, in hypophysectomized patients on hormonal replacement therapy other than GH, a marked decrease in GFR and ERPF was found despite no significant reduction in extracellular fluid space (Table 1). This led to the early conclusion that the changes in renal haemodynamics are due to a direct effect of GH on the kidney. It is well known that renal size and renal weight are increased in acromegaly. Gershberg et al. [4] observed that the in vivo increase in GFR was not proportional to the increase in renal weight observed on post-mortem examination. This suggested functional rather than structural causes of the alterations in glomerular haemodynamics. Visceromegaly was accompanied by increased glomerular size. The lack of correlation between GFR and renal size Table 1. Extracellular volume (ECV) and renal haemodynamics in hy-
persomatotropism and hyposomatotropism in man (after [3])
Key words: Growth hormone - Insulin-like growth factor - Renal growth - Hyperfiltration - Acromegaly
Offprint requests to: E. Ritz, Medizinische Universit~ts-Klinik, Bergheimer Strasse 58, W-6900 Heidelberg, Federal Republic of Germany
ECV (1)
GFR (ml/min per 1.73 m2)
RPF (ml/min per 1.73 m2)
11.4
Controls (n = 13)
(9.4-17.4)
104 (89-120)
588 (387-695)
Acromegaly (n-= 13)
16.8 (12.2-25.6)
134 (76-174)
751 (453-892)
Hypophysectomy (n = 12)
10.2 (7.5-12.7)
53 (45-88)
340 (207-477)
GFR, Glomerular filtration rate; RPF, renal plasma flow
510 was also found by Falkheden and Wickbom [5] who showed that the decrease in GFR and ERPF following hypophysectomy did not strictly correlate with the decline in renal weight assessed by urogram. Furthermore, Gershberg et al. [4] noted that acromegalic patients exhibit increased proximal tubular function, i.e. raised maximum tubular rate of transport for sulphate, para-aminohippuric acid and glucose with no change in distal tubular functions (free water clearance). They suggested preferential proximal tubular hypertrophy. This observation is of particular interest in view of the recent demonstration of GH receptors in proximal, but not distal tubules.
Renal haemodynamics under the influence of extractive or recombinant human GH and insulin-like growth factor-I According to the hyperfiltration theory [6], a rise in GFR (or one of its more distant consequences) in a patient with impaired renal function may accelerate the progression of renal failure. Consequently, the renal haemodynamic consequences have become a source of concern when recombinant human GH (rhGH) is used in the treatment of growth failure, particularly in children with renal disease. This has prompted several recent studies on the actions of GH on the kidney. Since the early studies of Parving et al. [7] it has been known that there is no acute change in GFR and ERPF in healthy individuals after short-term GH infusion, despite a tenfold elevation in plasma GH levels. In contrast, when extractive GH was administered for 1 week to male volunteers, GFR increased significantly (from 114 + 5 to 125 _+4 ml/min per 1.73 m 2) and this was not associated with changes in renal size or albuminuria [8]. This suggested that the renal response to GH is delayed which agreed with results of previous studies in dogs [9] and humans [10]. Recently Hirschberg et al. [11] gave a single in= tramuscular injection of GH (0.15 mg/kg) to normal adults and observed that, despite an acute rise in plasma GH levels, GFR and ERPF were unchanged during the first 24 h but rose subsequently. This delayed rise in GFR and ERPF was paralleled by an increase in plasma insulin-like growth factor (IGF)-I levels. GH causes an increase in circulating IGF-1 by stimulation of IGF-1 synthesis predominantly in the liver. This suggested that IGF-I could be a mediator of the haemodynamic response of the kidney to GH, and this has been supported by recent experiments with recombinant human IGF-I (rhlGF-I). Infusion of rhIGF-I caused an acute reversible increase in creatinine excretion paralleled by a fall in serum creatinine when administered to normal human subjects [ 12]. Later studies, using iodothalamate and iodohippurate clearance techniques, confirmed a prompt 20%-30% increase in GFR and ERPF in response to 20 gg/kg per hour rhIGF-I [13]. The relative roles of circulating versus locally formed IGF-I have not been defined. Potential mediators of the haemodynamic effects of IGF-I have been the subject of investigation. Intrarenal angiotensin II (although considered as one possible mediator [14] is unlikely to play an essential role, since in our
180
E'160
8.1-7 y
E
8
~,140 ._=
E ~120
,00 80
Con'tro[
3 GH Enatapril
GH+enalapril
Fig. 1. Changes in insulin clearance (C/R) with recombinant human grown hormone (rhGH) administrationwithout (left panel) and with (right panel) enalaprilpretreatment.Figuresrepresentprobands
studies [15] GH caused a similar median increase in GFR in the absence and presence of enalapril, an angiotensin converting enzyme inhibitor (Fig. 1). As shown in Fig. 1, when 4.5 units rhGH were given twice daily (bid) for 3 days to eight healthy normotensive probands by subcutaneous injection, median inulin clearance (Cin) rose 72 h after the start of rhGH administration from 114 (91-158) to 135 (108-167) ml/min per 1.73 m 2. With enalapril pretreatment Cin rose from 111 (88-153) to 13l (100-173) ingrain per 1.73 m2. However. in experimental animals very good evidence has been provided that the effect of IGF-I on GFR and ERPF is dependent on vasodilatory eicosanoids [16]. The increase in GFR on administration of 25 gg/kg IGF-I was blocked by indomethacin, but not by somatostatin. Whether this also occurs in humans is currently under investigation. Since the infusion of somatostatin did not suppress the IGF-I-induced rise in GFR and ERPF, these renal haemodynamic changes seem to be independent of peptide hormones like glucagon or insulin.
The effect of rhGH on GFR in renal failure Since an increase in GFR may not be desirable in the patient with renal failure [6], the action of GH on the kidney of individuals with impaired renal function is of interest. As shown in Table 2. acute injection of rhIGF-I (or rhGH) fails to affect creatinine clearance (Ccr) or Cin in sub-totally nephrectomized rats. in contrast to the consistent increase of approximately 20% observed in pair-fed or ad libitum fed sham-operated control animals. This prompted us to examine the effect of rhGH (4.5 units bid) o n Gin in healthy adult probands and patients with renal failure [17]. While GFR increased from 120 (range 91 158) to 133 (range 108-167) ml/min per 1.73 m2 in healthy volunteers, GFR was unaffected in patients with chronic renal failure: Cin was 21 (range 15-32) before rhGH and 22 (range 15-32) ml/min per 1.732 after rhGH. Interaction of rhGH with target organs was shown in these patients by the pronounced effects on serum cholesterol
511 Table 2. Effect of recombinant human growth hormone (rhGH) mid recombinant human insulin-like growth factor-I (rhlGF-I) on GFRa
1,6-
1./,,-
3
Ccr (ml/min per 100 g) Uraemia
Control
Solvent rhGH (1.5 IU/150 g per day)
353_+61 333 • 68
833 • 156 1098 _+295*
Solvent rhIGF-I (300 gg/150 g per day)
349_+48 351 _+49
956_+ 67 1246+_ 84*
--
J
1.2-
& 1.o0.8 ,,..,
O~ 0.6-
* P
Pediatr Nephrol (1991) 5:509-512 9 IPNA 1991
Endocrine changes Review article
Influence of growth hormone and insulin-like growth faetor-I on kidney function and kidney growth Eberhard Ritz1, Burkhard T'6nshoff2, Stefan Worgall 2, Gabor Kovacs 2, and Otto Mehls 2 Departments of Internal Medicine1 and Paediatrics 2, University of Heidelberg, Heidelberg, Federal Republic of Germany Received November 23, 1990; accepted December 28, 1990
Abstract. Decreased glomerular filtration rate (GFR) in hypopituitarism and increased GFR in acromegaly suggest that growth hormone (GH) has a substantial effect on renal haemodynamics. Extractive and recombinant human (rh) GH in healthy volunteers increased effective renal plasma flow (ERPF) and GFR by 10% and 15% respectively. Renal response to GH was delayed and occurred at the same time as an increase in plasma insulin-like growth factor (IGF)-I values, whereas infusion of rhIGF-I promptly increased GFR and ERPF, indicating that the haemodynarnic response of the kidney to GH is mediated by IGF-I. In chronic renal failure (CRF), the acute effect of GH on GFR is obliterated. This might protect the diseased kidney against the undesired consequences of hyperfiltration. Indeed, rhGH treatment for 1 year in children with CRF did not lead to an accelerated decline in GFR compared with the year before treatment. GH and IGF-I also affect renal growth. Exposure to excessive GH in transgenic mice causes renomegaly and progressive glomerular sclerosis. In acromegalic humans, increased renal size and weight and increased glomerular diameter are well known, whereas renal failure is not a long-term hazard. At least in normal and hypophysectomized rats treated with doses comparable with the therapeutic regimens used in stunted children, rhGH increased renal weight but in proportion to the increase in body weight indicating an isometric effect of GH on renal growth. From these data, major renal longterm side effects of rhGH treatment in children with CRF appear unlikely.
Renal function and renal morphology in hypersomatotropism and hyposomatotropism The renal effects of growth hormone (GH), i.e. a decreased glomerular filtration rate (GFR) in hypopituitarism and increased GFR in acromegaly were recognized several decades ago [1, 2]. These could not be explained by changes in renal perfusion pressure or extracellular volume [3]. Falkheden and Sjrgren [3] found increased GFR and effective renal plasma flow (ERPF) in patients with acromegaly, and this was associated with an increased extracellular fluid space. In contrast, in hypophysectomized patients on hormonal replacement therapy other than GH, a marked decrease in GFR and ERPF was found despite no significant reduction in extracellular fluid space (Table 1). This led to the early conclusion that the changes in renal haemodynamics are due to a direct effect of GH on the kidney. It is well known that renal size and renal weight are increased in acromegaly. Gershberg et al. [4] observed that the in vivo increase in GFR was not proportional to the increase in renal weight observed on post-mortem examination. This suggested functional rather than structural causes of the alterations in glomerular haemodynamics. Visceromegaly was accompanied by increased glomerular size. The lack of correlation between GFR and renal size Table 1. Extracellular volume (ECV) and renal haemodynamics in hy-
persomatotropism and hyposomatotropism in man (after [3])
Key words: Growth hormone - Insulin-like growth factor - Renal growth - Hyperfiltration - Acromegaly
Offprint requests to: E. Ritz, Medizinische Universit~ts-Klinik, Bergheimer Strasse 58, W-6900 Heidelberg, Federal Republic of Germany
ECV (1)
GFR (ml/min per 1.73 m2)
RPF (ml/min per 1.73 m2)
11.4
Controls (n = 13)
(9.4-17.4)
104 (89-120)
588 (387-695)
Acromegaly (n-= 13)
16.8 (12.2-25.6)
134 (76-174)
751 (453-892)
Hypophysectomy (n = 12)
10.2 (7.5-12.7)
53 (45-88)
340 (207-477)
GFR, Glomerular filtration rate; RPF, renal plasma flow
510 was also found by Falkheden and Wickbom [5] who showed that the decrease in GFR and ERPF following hypophysectomy did not strictly correlate with the decline in renal weight assessed by urogram. Furthermore, Gershberg et al. [4] noted that acromegalic patients exhibit increased proximal tubular function, i.e. raised maximum tubular rate of transport for sulphate, para-aminohippuric acid and glucose with no change in distal tubular functions (free water clearance). They suggested preferential proximal tubular hypertrophy. This observation is of particular interest in view of the recent demonstration of GH receptors in proximal, but not distal tubules.
Renal haemodynamics under the influence of extractive or recombinant human GH and insulin-like growth factor-I According to the hyperfiltration theory [6], a rise in GFR (or one of its more distant consequences) in a patient with impaired renal function may accelerate the progression of renal failure. Consequently, the renal haemodynamic consequences have become a source of concern when recombinant human GH (rhGH) is used in the treatment of growth failure, particularly in children with renal disease. This has prompted several recent studies on the actions of GH on the kidney. Since the early studies of Parving et al. [7] it has been known that there is no acute change in GFR and ERPF in healthy individuals after short-term GH infusion, despite a tenfold elevation in plasma GH levels. In contrast, when extractive GH was administered for 1 week to male volunteers, GFR increased significantly (from 114 + 5 to 125 _+4 ml/min per 1.73 m 2) and this was not associated with changes in renal size or albuminuria [8]. This suggested that the renal response to GH is delayed which agreed with results of previous studies in dogs [9] and humans [10]. Recently Hirschberg et al. [11] gave a single in= tramuscular injection of GH (0.15 mg/kg) to normal adults and observed that, despite an acute rise in plasma GH levels, GFR and ERPF were unchanged during the first 24 h but rose subsequently. This delayed rise in GFR and ERPF was paralleled by an increase in plasma insulin-like growth factor (IGF)-I levels. GH causes an increase in circulating IGF-1 by stimulation of IGF-1 synthesis predominantly in the liver. This suggested that IGF-I could be a mediator of the haemodynamic response of the kidney to GH, and this has been supported by recent experiments with recombinant human IGF-I (rhlGF-I). Infusion of rhIGF-I caused an acute reversible increase in creatinine excretion paralleled by a fall in serum creatinine when administered to normal human subjects [ 12]. Later studies, using iodothalamate and iodohippurate clearance techniques, confirmed a prompt 20%-30% increase in GFR and ERPF in response to 20 gg/kg per hour rhIGF-I [13]. The relative roles of circulating versus locally formed IGF-I have not been defined. Potential mediators of the haemodynamic effects of IGF-I have been the subject of investigation. Intrarenal angiotensin II (although considered as one possible mediator [14] is unlikely to play an essential role, since in our
180
E'160
8.1-7 y
E
8
~,140 ._=
E ~120
,00 80
Con'tro[
3 GH Enatapril
GH+enalapril
Fig. 1. Changes in insulin clearance (C/R) with recombinant human grown hormone (rhGH) administrationwithout (left panel) and with (right panel) enalaprilpretreatment.Figuresrepresentprobands
studies [15] GH caused a similar median increase in GFR in the absence and presence of enalapril, an angiotensin converting enzyme inhibitor (Fig. 1). As shown in Fig. 1, when 4.5 units rhGH were given twice daily (bid) for 3 days to eight healthy normotensive probands by subcutaneous injection, median inulin clearance (Cin) rose 72 h after the start of rhGH administration from 114 (91-158) to 135 (108-167) ml/min per 1.73 m 2. With enalapril pretreatment Cin rose from 111 (88-153) to 13l (100-173) ingrain per 1.73 m2. However. in experimental animals very good evidence has been provided that the effect of IGF-I on GFR and ERPF is dependent on vasodilatory eicosanoids [16]. The increase in GFR on administration of 25 gg/kg IGF-I was blocked by indomethacin, but not by somatostatin. Whether this also occurs in humans is currently under investigation. Since the infusion of somatostatin did not suppress the IGF-I-induced rise in GFR and ERPF, these renal haemodynamic changes seem to be independent of peptide hormones like glucagon or insulin.
The effect of rhGH on GFR in renal failure Since an increase in GFR may not be desirable in the patient with renal failure [6], the action of GH on the kidney of individuals with impaired renal function is of interest. As shown in Table 2. acute injection of rhIGF-I (or rhGH) fails to affect creatinine clearance (Ccr) or Cin in sub-totally nephrectomized rats. in contrast to the consistent increase of approximately 20% observed in pair-fed or ad libitum fed sham-operated control animals. This prompted us to examine the effect of rhGH (4.5 units bid) o n Gin in healthy adult probands and patients with renal failure [17]. While GFR increased from 120 (range 91 158) to 133 (range 108-167) ml/min per 1.73 m2 in healthy volunteers, GFR was unaffected in patients with chronic renal failure: Cin was 21 (range 15-32) before rhGH and 22 (range 15-32) ml/min per 1.732 after rhGH. Interaction of rhGH with target organs was shown in these patients by the pronounced effects on serum cholesterol
511 Table 2. Effect of recombinant human growth hormone (rhGH) mid recombinant human insulin-like growth factor-I (rhlGF-I) on GFRa
1,6-
1./,,-
3
Ccr (ml/min per 100 g) Uraemia
Control
Solvent rhGH (1.5 IU/150 g per day)
353_+61 333 • 68
833 • 156 1098 _+295*
Solvent rhIGF-I (300 gg/150 g per day)
349_+48 351 _+49
956_+ 67 1246+_ 84*
--
J
1.2-
& 1.o0.8 ,,..,
O~ 0.6-
* P
