Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
587 Clinical Science (1991)81,587-592 Cardiovascular effects of growth hormone treatment in growth-hormone-deficient adults: stimulation of the renin-aldosterone system ROSS C. CUNEO', FRANC0 SALOMON', PETER WILMSHURST2, CHRIS BYRNE2, C. MARK WILES3, RICHARD HEW4 AND PETER H. SONKSEN' Divisions of 'Medicine, *Cardiologyand 3Neurology,United Medical and Dental Schools of Guy's and St Thomas' Hospitals, St Thomas' Hospital, London, and 4Divisionof Radio-Isotopes, Medical Research Council, Northwick Park Hospital, Harrow, Middlesex. U.K. (Received 22 January/l3 May 1991; accepted 3 June 1991) SUMMARY 1. In adult humans with growth hormone deficiency, treatment with growth hormone has recently been shown to have major anabolic effects and to improve exercise performance. The cardiovascular effects of growth hormone in adults with growth hormone deficiency were examined in 24 patients treated with recombinant human growth hormone (0.07 units/kg at night) in a doubleblind, placebo-controlled trial lasting 6 months. 2. Compared with placebo, resting M-mode echocardiography showed increases in left ventricular enddiastolic dimension and stroke volume in the group treated with recombinant human growth hormone. No differences were noted between the groups with respect to left ventricular end-systolic dimension, fractional shortening, wall thicknesses or mean arterial blood pressure. Left ventricular myocardial mass increased in the group given recombinant human growth hormone. 3. The supine plasma renin activity was increased and remained elevated over the 6 months, whereas the plasma aldosterone concentration was unchanged, after treatment with recombinant human growth hormone. Clinical signs of sodium retention were evident during the first 3 months of treatment with recombinant human growth hormone. 4. We conclude that treatment with recombinant human growth hormone in adults with growth hormone deficiency resulted in small increases in left ventricular pre-load, due to the sodium-retaining action of growth hormone. Activation of the re&-aldosterone system was involved in such changes. Myocardial hypertrophy was observed without changes in mean arterial pressure, reflectingthe anabolic action of growth hormone. Correspondence: Professor Peter H. Sonksen, Division of Medicine, St Thomas' Hospital, London SE1 7EH. Key words: aldosterone, echocardiography, insulin-like growth factor-1, lean body mass, plasma renin activity, somatotropin. Abbreviations: ANG I, angiotensin I; GH, growth hormone; IGF-1, insulin-like growth factor-1; IVS,interventricular septa1 thickness; LVED, left ventricular enddiastolic dimension; LVES, left ventricular end-systolic dimension; LVPW, left ventricular posterior wall thickness; PRA, plasma renin activity; rhGH, recombinant human growth hormone; V O ~ , , maximal , ~ ~ , , oxygen uptake. INTRODUCTION Growth hormone (GH) is a potent anabolic agent, with multiple effects on lipid and carbohydrate metabolism. Recently, the effects of recombinant human GH (rhGH) in adult man with GH deficiency have been examined in two controlled treatment trials. Major increases in lean body mass [l]and skeletal muscle mass [2, 31 have been demonstrated, reflecting the anabolic action of GH. Similar anabolic responses have been seen when GH has been given to adult patients with burns [4], after gastrointestinal surgery [5], with malnutrition [6] and with chronic obstructive airways disease [7], and when given to elderly male subjects [8]. GH treatment in adults with GH deficiency has resulted in major improvements in exercise performance [3,9]. Since cardiac output is generally considered to be a major determinant of exercise performance and GH has antinatriuretic effects, we report resting cardiac function, determined by M-mode echocardiography, and plasma levels of hormones regulating blood volume in a group of adults with GH deficiency before and after rhGH treatment, and we relate these to changes in exercise performance. 588 R. C. Cuneo et al. METHODS Patients The study reported here involves the same patients as previously reported [ l , 3, 91. Briefly, patients were selected if (a) G H deficiency (defined as a peak plasma G H concentration of < 3 m-units/l during an adequate insulin hypoglycaemia test) had been present for 12 months, (b)their age was between 18 and 52 years, and (c) if required, stable and optimal conventional pituitary hormone replacement was given for 12 months before entry into the trial. Doses of these hormone replacements remained unchanged during the trial, except in three patients in the rhGH group, where thyroxine doses were reduced at 1 month to maintain free tri-iodothyronine levels within the reference range [ 11. Patients with Cushing’s disease were all cured more than 12 months before entry to the trial. Patients with other illnesses or ingesting drugs or alcohol sufficient to affect myocardial function or sodium balance were excluded. Trial design The protocol was approved by the hospitals’ ethics committees and all patients gave written, informed consent. Patients were randomly assigned to receive either rhGH (Genotropin; 4 i.u./ml, 2.7 i.u./mg; KabiVitrum, Stockholm, Sweden) or identically presented placebo, for 6 months, given as a daily subcutaneous dose of 0.07 unit/kg body weight at 20.00 hours. Women were studied in the first half of their menstrual cycle where appropriate. Echocardiography Patients requiring pituitary hormone treatment took their usual medications on the morning of the study. Parasternal long-axis images were recorded just below the mitral valve using a Hewlett-Packard machine (model 77020 AC) with the patient in the left lateral position. A single observer, blinded to the treatment code and patient details, performed all measurements. Left ventricular end-diastolic wall thicknesses [interventricular septa1 (IVS) and posterior wall (LVPW)], and internal dimensions [end-diastolic (LVED) and end-systolic (LVES)] were recorded to the nearest 0.1 cm. Blood was collected via a plastic venous catheter after an overnight fast, at least 30 min after the subject had assumed the recumbent position. Plasma was separated and was stored at -20°C until assayed. Brachial blood pressure was then recorded, as the average of three sphygmomanometric recordings in the non-dominant arm at baseline and after 1, 3 and 6 months of treatment. Plasma renin activity (PRA) was measured by established methods [lo]. Plasma aldosterone concentration was measured by r.i.a. [ 111 after solvent extraction using the DPL Aldosterone Kit [DPL (U.K.) Ltd, Abingdon, Oxon, U.K.]. Intra- and inter-assay coefficients of variation for PRA were less than 5.2% and 7.0%, and for plasma aldosterone concentration were less than 6.9% and 7.8%, respectively. Maximal oxygen uptake ( V O ~ , , , ~ was ~,) measured on a cycle ergometer on the day preceding echocardiography, as previously reported [9]. Calculations Mean arterial blood pressure was calculated as: diastolic pressure + 1/3 X (systolic -diastolic pressure). Diastolic and systolic volumes were determined as: (7 x D3)/(2.4+ D), where D is the internal short diameter in cm [12]. Fractional shortening ( O h ) was calculated as: (LVED - LVES) X 100/LVED. Left ventricular wall mass was determined as: 1.04 X [(LVED+ 3 MMT)2- LVED3] - 13.6 g, where MMT represents mean myocardial thickness[MMT=(IVS+LVPW)/2] [13]. Statistics Results are expressed as means k SEM. Treatment responses were assessed by analysis of co-variance using baseline data as the co-variate, and where repeat measures were made a mean of treatment data was used for comparison. Significance was recognized at the 5% level. Reported percentage changes in variables are those in excess of changes in the placebo group. Relationships between changes in single variables in the two groups were explored with multiple linear regressions, using treatment group as a stratifying variable. RESULTS Randomization resulted in two evenly matched populations with respect to age, sex, body size and drug history (see Table 1).Two patients in each group had undergone a Table 1. Characteristics of the patients studied Characteristics of the rhGH and placebo groups at the beginning of the trial. Age, height and weight are shown as means k SEM. No statistically significant differences were observed between the two groups. rhGH group (n=12) Age (years) 39+3 8:4 Sex (M/F) Height (cm) 170.0 2.7 82.5 3.9 Weight (kg) Initial diagnosis (no.) Cushing’s disease 6 Prolactinoma 0 Chromophobe adenorna 3 Craniopharyngiorna 3 Idiopathic 0 Radiotherapy (medulloblastoma) 0 ** Pituitary replacement hormones (no.) Corticosteroids Thyroxine Sex steroids Desarnino-o-arginine vasopressin Fludrocortisone 10 Placebo group ( n = 12) 38+3 8:4 169.0 3.2 79.3 k 6.8 + 3 2 3 0 3 1 10 3 9 10 8 2 2 2 11 589 Cardiovascular effects of growth hormone bilateral adrenalectomy as treatment for Cushing’s disease. Before treatment, no patient had values of wall thicknesses or chamber dimension outside the adult reference range for our laboratory. After treatment, no changes were noted in left ventricular wall thicknesses (Table 2). LVED increased by 2% in the rhGH group (P=0.039). Fractional shortening was significantly lower in the rhGH group at baseline ( P = 0.004), but the increment after treatment was not different between the two groups (P= 0.77). Stroke volume increased by 6.1% in the rhGH group ( P =0.001). Left ventricular wall mass increased by 4.8% in the rhGH group (P=O.O45; see Fig. 1).Because there was no change in body weight after rhGH treatment [ 13, left ventricular wall mass expressed per body surface area showed equally significant changes (rhGH, 110 f 10 to 1 2 0 f 10 g/m2; placebo, 100 f 10 to 100 f 10 g/m2; P = 0.046). Mean arterial blood pressure increased slightly, but not significantly, at 1 month and was not different between the groups (rhGH, 87 f3, 96 f4, 91 f2 and 88 f 3 mmHg; placebo, 93 f3,95 f4,94 f 5 and 90 k 4 mmHg; see Fig. 2). One patient became hypertensive at 1 month (blood pressure 141/105 mmHg); subsequent halving of the rhGH dose resulted in a prompt normalization of blood pressure for the remainder of the trial. In two other Table 2. Echocardiographic results Echocardiographic measurements (means fSEM) in the rhGH and placebo groups before and after 6 months of treatment. Statistical significance (analysisof co-variance): *P<O.O5, **P<O.OOl compared with placebo; tP<O.Ol compared with placebo group before treatment. Normal values: LVED, 3.5-5.6 cm; LVES, 2.3-3.7 cm; IVS, < 1.2 cm; LVPW, < 1.1cm. Before treatment After 6 months of treatment LVED (cm) rhGH Placebo 4.9 k 0.1 4.9 f 0.2 5.1 k0.1* 5.0 k 0.2 LVES (cm) rhGH Placebo 3.3 k 0.1 2.9 k 0.1 3.3 k 0.1 3.0 k 0.0 IVS (cm) rhGH Placebo 1.1 k O . 1 1.OkO.1 1.1 kO.1 1.0 k 0.1 LVPW (cm) rhGH Placebo 0.9 k 0.1 0.9 k 0.1 1.0 kO.1 0.9 k 0.1 Fractional shortening (70) rhGH Placebo 34 k 21 40f1 35f2 39fl Stroke volume (ml) rhGH Placebo 71 k 4 79f6 78 k 4** 82f7 210f 10 200 f 20 230 k 20* 210f20 Left ventricular wall mass (gm) rhGH Placebo patients, the rhGH dose was halved due to carpal tunnel compression symptoms or peripheral oedema. One patient in the rhGH group withdrew after 3 days treatment due to agitation. The mean PRA in adults with GH deficiency before treatment was 2.43 f0.28 (range 1.38-6.1 1) pmol of angiotensin I (ANG I) h-’ ml-I in patients with intact zona glomerulosae (95% confidence intervals for normal samples collected after overnight recumbency, 1.14-2.65 pmol of ANG I h-I ml-I). In the patients with previous adrenalectomies, PRA was 2.84, 4.12, 5.55 and 17.2 pmol of ANG I h-I ml-*, and plasma aldosterone concentrations were undetectable. The plasma aldosterone concentration in the patients with intact adrenals was 199 f28 pmol/l (recumbent reference range, 100-400 pmol/l). In patients with intact adrenals after treatment, PRA rose in the rhGH group and remained elevated throughout the trial (rhGH, 2.64f0.32, 3.08f0.42, 3.18k0.42 and 3.02f0.38 pmol of ANG I h-I ml-’at entry, 1, 3 and 6 months, respectively; placebo, 2.22 f0.44, 1.86f0.22, 1.78f0.26 and 2.01f0.27pmolofANGI h-1 d - 1 . , p = 0.019; see Fig. 2). The plasma aldosterone concentration did not change appreciably in either group (rhGH, 185f34,144+24,163+34and 1 5 2 f 3 5 pmol/ 1; placebo, 2 1 1 f 4 5 , 2 0 0 f 4 1 , 1 6 8 f 3 2 and 1 9 8 f 2 5 pmol/l). Treatment data were analysed after excluding patients with prior adrenalectomies, although the conclusions were not altered by their inclusion. As previously reported [ 11, the plasma potassium concentration fell marginally after rhGH treatment (rhGH, 3.8 f0.1, 3.6f0.1, 3.7f0.1 and 3.6f0.1 mmol/l; placebo, 3.7k0.1, 3.7f0.1, 3.9f0.1 and 3.8f0.1 mmol/l at entry, 1, 3 and 6 months, respectively; P=O.O5). The plasma sodium concentration was normal and was unaffected by treatment with rhGH. As previously reported [9], vo2,max. during subjectlimited, maximal cycle ergometry increased in the rhGH group (rhGH, 1.88 f0.17 to 2.34 f0.20 l/min; placebo, 1.84 f0.17 to 1.98 f0.13 l/min at baseline and 6 months, respectively; P = 0.016). Increases in left ventricular 350 1 100 : 1 Pre I I Post Pre , i Post Fig. 1. Mean (bars) and individual values of left ventricular wall mass in the rhGH (m) and placebo (0)groups before and after 6 months treatment. R. C. Cuneo et al. 590 DISCUSSION 0 1 3 6 0 1 3 6 1 lo5 80 ' d i " "3' 6 Time (months) Fig. 2. Changes in PRA (a),plasma aldosterone concentration ( b ) and mean arterial blood pressure (c) during treatment with rhGH (m) or placebo ( 0 )for 6 months. Data are presented as means & SEM. Statistical significance (analysis of co-variance) between treatment groups: ( a ) P=O.O19, ( b )not significant and (c) not significant. Data presented for PRA and plasma aldosterone concentration include only patients with intact adrenal glands. diastolic volume, but not in stroke volume, were associated with increases in ( P = 0.03). Changes in left ventricular wall mass were associated with changes in stroke volume ( P = 0.01 3), but not with the changes in lean body mass, thigh muscle mass, plasma insulin-like growth factor-1 (IGF-1) concentration, volmax,, nor with the age or sex of the patients, nor with a history of cured Cushing's disease. Increases in left ventricular diastolic volume and wall mass were not associated with changes in mean arterial blood pressure from baseline to either 1 or 6 months. v02,max, GH treatment in adults with GH deficiency has been shown to increase and normalize lean body mass [l], to increase skeletal muscle mass [2, 31, to substantially reduce fat mass [ l , 21 and to increase maximal [2,9] and sub-maximal [9] exercise performance. We have shown an increase in left ventricular myocardial wall mass of approximately 5% after rhGH treatment in adults with GH deficiency, comparable with increases in thigh muscle (5-8'/0) and lean body (approximately looh)mass in the same patients [l, 31. Atrophy of visceral organs after hypophysectomy has been shown to include the heart [ 141. Treatment of hypophysectomized rats with either human GH or human IGF-1 significantly increased heart weight, suggesting that this anabolic action of GH is mediated by IGF-1 [15]. No correlation between the plasma IGF-1 concentration and the myocardial mass could be shown, due to either insufficient numbers of patients or to alterations in serum IGF-1binding proteins or in IGF-1 receptor status. While changes in lean body mass might be expected to be paralleled by changes in myocardial mass, the absence of such an association probably reflects the small number of patients studied. Cardiomegaly in human acromegaly is usually, but not always, associated with hypertension [ 16, 171. The cardiomegaly resulting from implantable GH-secreting tumours appears to be due to both direct cardiac effects (anabolism) and volume overload [ 181. Since no increase in training activities was recorded [9] and no statistically significant effect from changes in systemic arterial blood pressure were detected, the increase in left ventricular myocardial wall mass most likely reflected the anabolic action of rhGH. We attempted to minimize the variability involved in the measurement of left ventricular wall mass and volumes by echocardiography [12, 13, 371 by the use of a single echocardiographer, blinded as to the treatment code, and by using well-validated procedures. Due to the age of the patients and the successful completion of maximal exercise tests, ventricular asynergy was considered unlikely, allowing the assumptions in the calculations of ventricular volumes [12] and wall mass [13] to be met. The statistical design, testing for differences in mean responses between rhGH and placebo groups, accounts for this variability. The lack of change in myocardial wall mass in young GH-deficient adults treated with rhGH for 4 months [2] may thus reflect the lower dose and/or the shorter duration of treatment. This study demonstrated small increases in resting LVED. Admittedly, these changes were at the limit of detectable changes with echocardiography, but they are consistent with the clinical signs of sodium retention noted in this trial [l], particularly during the first 3 months. Sodium retention and/or reduced urinary sodium excretion have been noted after GH treatment in GHdeficient [2, 19, 201 and normal [21] human subjects. Extracellular fluid volume has been shown to increase with GH treatment in GH-deficient children [22]. Patients with Cardiovascular effects of growth hormone acromegaly have increased plasma and erythrocyte volumes [23], total body water and total body sodium contents [24], and extracellular/intracellular water ratio [25]. Thus the increase in LVED has been interpreted as an increase in pre-load secondary to increased circulating blood volume. That improved maximal exercise performance was associated with increased resting LVED suggests that the observed echocardiographic changes were beneficial. The increases in maximal oxygen pulse [9], an index of maximal stroke volume, supports the contention that some of the improvement in exercise performance was related to increased cardiac output. GH-induced sodium retention involves the renal tubules [26], a process stimulated directly by IGF-1 [27, 281. The antinatriuresis can occur independently of the adrenal glands [26, 291. The increase in PRA demonstrated in this study suggests an additional factor, activation of the renin-aldosterone system, after long-term GH replacement. The lack of suppression of plasma aldosterone concentrations in the face of an apparent increase in blood volume supports this contention. Similarly, the small reduction in plasma potassium concentration in the rhGH-treated group may have contributed to the reduced aldosterone response to the PRA stimulus. An increased plasma aldosterone concentration [21] or an increased urinary aldosterone excretion [30] in normal adults or hypopituitary humans, respectively, have been reported after GH treatment. Others have noted no change in aldosterone secretion rate after GH treatment in hypopituitary humans [311.In acromegaly, normal [32], suppressed [33] or stimulated [34] aldosterone systems, usually with suppressed PRA, have been reported. The variety of alterations in the renin-aldosterone system described in acromegaly may reflect the chronicity of volume expansion and the dual mechanisms of sodium retention by GH and IGF-1. The increase in stroke volume after rhGH treatment may reflect increased pre-load (the Starling effect) or possibly enhanced myocardial contractility. In neonatal rat cardiocytes, increased contractility has been noted after the addition of IGF-1 at physiological concentrations [35]. Increases in myocardial contractility (measured as increased fractional shortening) have been reported in some patients with acromegaly [17, 361, in normal subjects treated with human GH for 2 weeks [37], and in one patient with GH deficiency and cardiomyopathy [38]. No such change was noted in the present study. Whether the reported change in fractional shortening was due to increased contractility or to reduced after-load is uncertain. In summary, echocardiographic studies before and 6 months after rhGH treatment (0.07 unit/kg) in adults with GH deficiency showed an increased left ventricular wall mass, reflecting the anabolic action of GH. Secondly, small increases in LVED and stroke volume were noted, probably due to increased circulating blood volume. Finally, the antinatriuretic effect of rhGH treatment involves, in part, stimulation of the renin-aldosterone system. 591 ACKNOWLEDGMENTS We sincerely thank Ms Nadia Payne and the late Dr J. D. H. Slater, The Middlesex Hospital, London, for assaying PRA and plasma aldosterone concentrations. We are also grateful to Dr C. Lowy and Professor H. Jacobs for allowing us to study their patients, and to Ms Sue Chin and Mr Richard Morris for statistical advice. This study was supported in part by KabiVitrum, Stockholm, Sweden, who also provided the rhGH. F.S. was supported by a grant from the Swiss National Foundation for Scientific Research. REFERENCES 1. jalomon, F., Cuneo, R.C., Hesp, R. & Sonksen, P.H. The :ffects of treatment with recombinant human growth iormone on body composition and metabolism in adults vith growth hormone deficiency. N. Engl. J. Med. 1989; 321,1797-803. 2. brgensen, J.O.L., Pedersen, S.A., Thuesen, L. et al. Beneficial effects of growth hormone treatment in GHieficient adults. Lancet 1989; i, 1221-5. 3. Cuneo, R.C., Salomon, F., Wiles, C.M., Hesp, R. & Sonksen, P.H. Growth hormone treatment in growth hormone Ideficient adults. I. Effects on muscle mass and strength. J. Appl. Physiol. 1991; 70,688-94. 4. Soroff, H.S., Pearson, E., Green, N.L. & Artez, C.P. The 1effect of growth hormone on nitrogen balance at various levels of intake in burned patients. Surg. Gynecol. Obstet. 1960; 111,259-73. 5 . Ward, H.C., Halliday, D. & Sim, A.J.W. Protein and energy metabolism with biosynthetic human growth hormone after gastrointestinal surgery. Ann. Surg. 1987; 206,56-61. 6. Binnerts, A,, Wilson, J.H.P. & Lamberts, S.W.J. The effects of human growth hormone administration in elderly adults with recent weight loss. J. Clin. Endocrinol. Metab. 1988; 67,1312-16. 7. Suchner, U., Rothkopf, M.M., Stanislaus, G., Elwyn, D.H., Kvetan. V. & Askanazi, J. Growth hormone and pulmonary disease: metabolic effects in patients receiving -parenterd nutrition. Arch. Intern. Med. 1990; 150,1225-30. 8. Rudman, D., Feller, A.G., Nagraj, H.S. et al. Effects of human growth hormone in men over 60 years old. N. Engl. J. Med. 1990; 323, 1-6. 9. Cuneo, R.C., Salomon, F., Wiles, C.M., Hesp, R. & Sonksen, P.H. Growth hormone treatment in growth hormone deficient adults. 11. Effects on exercise performance. J. Appl. Physiol. 1991; 70,695-700. 10. Menard, J.C. & Catt, H.J. Measurement of renin activity, concentration and substrate in rat plasma by radioimmunoassay of angiotensin I. Endocrinology (Copenhagen) 1972; 90,422-30. 11. Jones, J.C., Cauter, G.D.C. & MacGregor, G.A. Interference by plasma metabolites in a direct radioimmunoassay for plasma aldosterone. Ann. Clin. Biochem. 1981; 18,54-9. 12. Teicholtz, L.E., Kreulen, T., Herman, M.V. & Gorlin, R. Problems in echocardiographic volume determinations: echocardiographic-angiographic correlations in the presence or absence of asynergy. Am. J. Cardiol. 1976; 37, 7-11. 13 Devereux, R.B. & Reichek, N. Echocardiographic determination of left ventricular mass in man: anatomic validation of the method. Circulation 1977; 55,613-8. 14. Hjalmarson, A., Isaksson, 0. & Ahren, K. Effects of growth hormone and insulin on amino acid transport in perfused rat heart. Am. J. Physiol. 1969; 217,1795-802. 15. Guler, H.-P., Zapf, J., Scheiwiller, E. & Froesch, E.R. Recombinant human insulin-like growth factor 1 stimulates - 592 R. C. Cum:oet al. growth and has distinct effects on organ size in hypophysectomized rats. Proc. Natl. Acad. Sci. U.S.A. 1988; 85, 4889-93. 16. McGuffin, W.L., Sherman, B.M., Roth, J. et al. Acromegaly and cardiovascular disorders. A prospective study. Ann. Intern. Med. 1974; 81, 11-8. 17. Smallridge, R.C., Rajfer, S., Davia, J. & Schaaf, M. Acromegaly and the heart. An echocardiographic study. Am. J. Med. 1979; 66,22-7. 18. Penny, D.G., Dunbar, J.C. & Baylerian, M.S. Cardiomegaly and haemodynamics in rats with a transplantable growth hormone-secreting tumour. Cardiovasc. Res. 1985; 19, 270-7. 19. Henneman, P.H., Forbes, A.P., Moldauer, M., Dempsey, E.F. & Carroll, E.L. Effects of human growth hormone in man. J. Clin. Invest. 1960; 39, 1223-38. 20. Rudman, D., Chyatte, S.B., Patterson, J.H. et al. Observations on the responsiveness of human subjects to human growth hormone: effects of endogenous growth hormone deficiency and myotonic dystrophy. J. Clin. Invest. 197 1; 50,1941-9. 21. Ho, K.Y. & Weissberger, A.J. The antinatriuretic action of biosynthetic human growth hormone in man involves activation of the renin-angiotensin system. Metab. Clin. Exp. 1990; 39,133-7. 22. Novak, L.P., Hayles, A.B. & Cloutier, M.D. Effect of hGH on body composition of hypopituitary dwarfs. Four-compartment analysis and composite body density. Mayo Clin. Proc. 1972; 41,241-6. 23. Strauch, G., Lego, A., Therain, F. & Bricaire, H. Reversible plasma and red cell volumes increases in acromegaly. Acta Endocrinol. 1977; 85,465-78. 24. Ikkos, D., Luft, R. & Sjogren, B. Body water and sodium in patients with acromegaly. J. Clin. Invest. 1954; 33,989-94. 25. Bengtsson, B.-A., Brummer, R.-J.M., Eden, S. & Bosaeus, I. Body composition in acromegaly. Clin. Endocrinol. 1989; 30,121-30. 26. Biglieri, E.R., Watlington, C.O. & Forsham, P.H. Sodium retention with human growth hormone and its subfractions. J. Clin. Endocrinol. Metab. 196 1; 2 1,36 1-70. 27. Blazer-Yost, B.L. & Cox, M. Insulin-like growth factor 1 stimulates renal epithelial Na+ transport. Am. J. Physiol. 1988; 255, C413-17. 28. Guler, H.-P., Eckardt, K.-U., Zapf, J., Bauer, C. & Froesch, E.R. Insulin-like growth factor 1 increases glomerular filtration rate and renal plasma flow in man. Acta Endocrinol. 1989; 121,101-6. 29. Stein, J.D., Bennett, L.L., Batts, A.A. & Li, C.H. Sodium, potassium and chloride retention produced by growth hormone in the absence of the adrenals. Am. J. Physiol. 1952; 171,587-91. 30. Beck, J.C., McGarry, E.E., Dyrenfurth, I. & Venning, E.H. The metabolic effect of human and monkey growth hormone in man. Ann. Intern. Med. 1958; 49, 1090-105. 31. Finkelstein, J.W., Kowarski, A., Spaulding, J.S. & Migeon, C.J. Effects of various preparations of human growth hormone on aldosterone secretion rate of hypopituitary dwarfs. Am. J. Med. 1965; 38,5 17-2 1. 32. Strauch, G., Vallotton, M.B., Touitou, Y. & Bricaire, H. The renin-angiotensin-aldosterone system in normotensive and hypertensive patients with acromenaly. - - N. E n-d J. Med. 1y72; 281,765-9. 33. Cain. J.P.. Williams. G.H. & Dluhv. ,, R.G. Plasma renin activity and aldosterone secretion in patients with acromegaly. J. Clin. Endocrinol. Metab. 1972; 34,73-81. 34. Karlberg, B.E. & Ottosson, A.-M. Acromegaly and hypertension: role of renin-angiotensin-aldosterone system. Acta Endocrinol. 1982; 100,581-7. 35. Vetter, U., Kepferschmidt, D., Lang, D. & Pentz, S. Insulinlike growth factors and insulin increase the contractility of neonatal rat cardiocytes in vitro. Basic Res. Cardiol. 1988; 83,647-54. 36. Thuesen, L., Christensen, S.E., Weeke, J., Orskov, H. & Henningsen, P. A hyperkinetic heart in uncomplicated active acromegaly. Acta Med. Scand. 1988; 223,337-43. 37. Thuesen, L., Christiansen, J.S., Sorensen, K.E., Jorgensen, J.O.L., Orskov, H. & Henningsen, P. Increased myocardial contractility following growth hormone administration in normal man. Dan. Med. Bull. 1988; 35, 193-6. 38. Cuneo, R.C., Wilmshurst, P., Lowy, C., McGauley, G. & Sonksen, P.H. Cardiac failure responding to growth hormone. Lancet 1989; i, 838-9. I ,