Peak bone mass development of females: can young adult women improve their peak bone mass?

The purpose of this review is to summarize the effects of calcium intake on peak bone mass (PBM) accrual by premenopausal females during their 20s and possibly 30s. Prospective studies are highlighted because of the value of investigating the same subjects for one or more years. Results of cross-sectional studies are also summarized. Findings from both prospective and cross-sectional investigations suggest a positive benefit of adequate or supplemented calcium intakes on bone mineral content (BMC) in females during the third (20s) and fourth (30s) decades of life. PBM of subjects was found to be increased or maintained in comparison to PBM of controls in the five intervention studies that used calcium supplementation or the addition of calcium-rich foods. The results of cross-sectional studies, including a meta-analysis of approximately 20 such studies, also support the benefit to PBM of adequate calcium intakes. In summary, sufficient consumption of calcium during the various stages of the early life cycle, when combined with overall sound nutrition, regular physical activity, and possibly pregnancy, lactation, and child-rearing, may contribute to PBM accrual of women during the early adult decades.     

Estradiol is related to visual memory in healthy young men.

The influence of testosterone and estrogen on memory was investigated in 33 healthy young men. Tests of visual memory, visuospatial ability, verbal memory, and attention were administered, and circulating levels of estradiol and free testosterone were measured. Participants with high levels of estradiol performed better on 2 measures of visual memory than did those with normal but lower levels. There were no differences between individuals with high and low levels of testosterone on any cognitive measure. These results support the contention that estradiol influences memory in young men. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Plasma lipid response to hypolipidemic diets in young healthy non-obese men varies with body mass index

Lipid response to dietary fat is highly variable among individuals of a population. The aim of this study was to establish whether being overweight is one of the factors that determines this response. Forty-one non-obese healthy men were divided into two groups according to body mass index as follows: controls, <25 kg/m2; overweight, >25 kg/m2 but <30 kg/m2. After consuming a saturated fat-rich diet (SAT diet: 38% fat, 20% saturated) for 4 wk, subjects were switched to a low fat diet [National Cholesterol Education Program (NCEP)-I diet: 28% fat, 10% saturated] for 4 wk and then to a monounsaturated fat-rich diet (MUFA diet: 38% fat, 22% monounsaturated) for 4 wk. Data were analyzed by Student's t test and two-way ANOVA for repeated measures. After consuming the NCEP-I diet, the overweight subjects had a smaller decrease relative to the SAT diet period in plasma total cholesterol [-0.30 vs. -0.67 mmol/L (-7 vs. -16%), P < 0.02] and low density lipoprotein-cholesterol concentrations [-0.24 vs. -0.55 mmol/L (-9 vs. -21%), P < 0.04] than controls. However, in the overweight subjects, the MUFA diet produced a greater decrease in plasma triglycerides than in the controls relative to the SAT diet period [-0.36 vs. -0.03 mmol/L (-26 vs. -4%), P < 0.006] and to the NCEP-I diet period [-0.29 vs. 0. 01 mmol/L (-22 vs. 1%), P < 0.01). Plasma cholesterol concentrations changed to a lesser extent, and triglyceride concentration to a greater extent, in overweight but non-obese young men than in those of normal weight in response to changes in dietary fat composition. Our data suggest that in the diet treatment of obese hyperlipemic subjects, it is more important for them to lose weight than to change the fat composition of their diets.     

Feeding colostrum increases circulating insulin-like growth factor I in newborn pigs independent of endogenous growth hormone secretion

Our objective was to examine the influence of feeding and endogenous GH secretion on circulating IGF-I in colostrum-deprived newborn pigs fed colostrum (n = 4), formula (control, n = 4), or water (n = 4). In another four formula-fed pigs, GH was ablated (GRF-A) with two intravenous injections of a GH releasing-factor antagonist (N-Ac-Tyr1,D-Arg2)-GRF(1-29)-NH2. Blood was serially sampled in all pigs to measure plasma IGF-I and GH profiles. Feeding increased plasma IGF-I concentration two- to fourfold and decreased GH secretion. Despite a more than 80% decrease in the plasma GH in GRF-A pigs, the circulating IGF-I concentration was similar to that in control pigs. In colostrum-fed pigs, plasma IGF-I was higher than that in control pigs, despite equal nutrient intake and lower circulating GH. There were no differences in plasma IGF binding protein (IGFBP)-3 levels among the treatment groups. However, the relative abundance of plasma IGFBP-4 was lower, and that of IGFBP-1 higher, in unfed pigs than in any of the three fed groups. The plasma insulin concentration was not different among fed pigs, but it was lower in unfed pigs. Our results indicate that the circulating IGF-I concentration is more dependent on nutrient intake than on GH in newborn pigs, despite relatively high GH concentrations. However, because the nutrient content in the formula was designed to match that of colostrum, a factor other than nutrient intake and GH was responsible for the maximal increase in circulating IGF-I concentration observed in colostrum-fed pigs.     

Effect of growth hormone administration on circulating levels of luteinizing hormone, follicle stimulating hormone and testosterone in normal healthy men

A new area of growth hormone (GH) therapy in adults is the treatment of infertility. The aim of this study was to evaluate the effects of pharmacological GH administration on the secretion of pituitary and gonadal hormones in normal men. Eight healthy men, 23-32 years of age (mean 28.1 years), with a normal body mass index were studied in a double-blind, placebo-controlled crossover design. All participants had a normal semen analysis before entering the study. Each participant was treated with placebo and GH (12/IU/day, Norditropin; Novo Nordisk, Denmark) during two different 14-day periods, separated by a 6 week washout period. Administration of GH for 14 days resulted in a significant increase in serum insulin-like growth factor I (IGF-I; P < 0.01) but no changes occurred in IGF-I values during placebo treatment. The concentrations of follicle stimulating hormone and luteinizing hormone displayed no change during the two periods and did not differ between the GH treatment period and the placebo period. The concentration of testosterone was unchanged during the placebo/GH periods and there was no difference between the GH treatment period and the placebo period. We conclude that GH treatment for 14 days in normal healthy men does not affect gonadotrophin or testosterone patterns.     

Treatment with the oral growth hormone secretagogue MK-677 increases markers of bone formation and bone resorption in obese young males

The effect of 2 months of treatment with the oral growth hormone (GH) secretagogue MK-677 on markers of bone metabolism was determined in healthy obese male subjects. This was a randomized, double-blind, parallel, placebo-controlled study. Twenty-four healthy obese males, 19-49 years of age, with body mass index > 30 kg/m2 were treated with MK-677 (25 mg/day; n = 12) or placebo (n = 12) for 8 weeks. MK-677 increased markers of bone formation; a 23% increase in the carboxy-terminal propeptide of type I procollagen levels and a 28% increase in procollagen III peptide levels were seen with as little as 2 weeks of MK-677 treatment (p < 0.01 and p = 0.001 vs. placebo, respectively) while a 15% increase in serum levels of osteocalcin was not detected until 8 weeks of treatment (p < 0.01 vs. placebo). Markers of bone resorption were induced within 2 weeks of treatment with MK-677; serum levels of the carboxy-terminal cross-linked telopeptide of type I collagen were increased 26% at 8 weeks (p = 0.001 vs. placebo), and urine hydroxyproline/creatinine and calcium/creatinine ratios at 8 weeks were increased by 23% (p < 0.05 vs. placebo) and 46% (p < 0.05 vs placebo), respectively, MK-677 increased serum insulin-like growth factor binding protein-5 (IGFBP-5) by 43-44% after 2-8 weeks of treatment (p < 0.01 vs. placebo). Serum IGFBP-4 was increased by 25% after 2 weeks of treatment (p < 0.001 vs. placebo) but no significant change from baseline was observed after 8 weeks of treatment. Plasma interleukin-6 was not significantly changed by active treatment. In conclusion, short-term treatment of healthy obese male volunteers with the GH secretagogue MK-677 increases markers of both bone resorption and formation. Large increases in serum levels of IGF-1 and IGFBP-5 and a transient increase in serum IGFBP-4 were found. Future long-term studies are needed to investigate if prolonged treatment with MK-677 increases bone mass.     

Effects of sumatriptan on growth hormone releasing hormone-stimulated growth hormone secretion in acromegaly

Specific inhibition of mammalian genes is possible through the use of antisense oligonucleotides (AS ODNs) or ribozymes. These strategies have led to a better understanding of several cellular and molecular mechanisms, among which cancer development. Recently, these strategies have been applied also for therapeutical purposes in diseases such as AIDS and cancer. In some of these therapeutical trials the antisense strategy is combined with gene transfer technology: the AS ODN or the ribozyme are expressed within the cell by the use of adenoviral or retroviral vectors. However, many difficulties have still to be overcome before ODNs and ribozymes can be used routinely in the clinic.     

Normal growth hormone secretion in growth hormone insufficient children retested after completion of linear growth

Tuberculosis of the cervical spine is rare, comprising 3-5% of cases of tuberculosis of the spine. Eight patients with tuberculosis of the cervical spine seen during 1989-1992 were reviewed. They all presented with neck pain. The 4 children presented with a kyphotic deformity. In all the children the disease was extensive, with a large prevertebral abscess formation, while in the adults it was localised to one or two motion segments. Cord compression was present in 4 of the 8 patients. All the patients were treated with antituberculosis drugs and 6 underwent surgery. There was full neurological recovery in all patients. The kyphosis was improved though not fully corrected. There was a problem in stabilisation of severe involvement of the body and dens of C2. Surgery seems to play a major role in the treatment of tuberculosis of the cervical spine.     

Expression of expansin genes is correlated with growth in deepwater rice

Expansins are a family of proteins that catalyze long-term extension of isolated cell walls. Previously, two expansin proteins have been isolated from internodes of deepwater rice, and three rice expansin genes, Os-EXP1, Os-EXP2, and Os-EXP3, have been identified. We report here on the identification of a fourth rice expansin gene, Os-EXP4, and on the expression pattern of the rice expansin gene family in deepwater rice. Rice expansin genes show organ-specific differential expression in the coleoptile, root, leaf, and internode. In these organs, there is increased expression of Os-EXP1, Os-EXP3, and Os-EXP4 in developmental regions where elongation occurs. This pattern of gene expression is also correlated with acid-induced in vitro cell wall extensibility. Submergence and treatment with gibberellin, both of which promote rapid internodal elongation, induced accumulation of Os-EXP4 mRNA before the rate of growth started to increase. Our results indicate that the expression of expansin genes in deepwater rice is differentially regulated by developmental, hormonal, and environmental signals and is correlated with cell elongation.     

Twenty-four hour growth hormone secretion in a patient with Werner's syndrome

OBJECTIVE: To assess the 24-h endogenous secretory growth hormone (GH) profile and serum insulin-like growth factor-I (IGF-I) response to exogenous recombinant human growth hormone (rhGH) in a patient with Werner's syndrome. DESIGN: Blood sampling every 20 min for 24 h followed by three daily injections of growth hormone. SETTING: General Clinical Research Center. PATIENTS: Single patient with Werner's syndrome. MEASUREMENTS: Serum GH and IGF-I. RESULTS: Growth hormone pulses were absent during the 24-h monitoring period. Likewise, integrated GH concentrations were very low at 0.25 mu min/mL, and no peaks occurred after sleep onset. Following single daily administration of rhGH, serum GH and IGF-I rose. CONCLUSIONS: Our findings support previous but less extensive studies suggesting patients with Werner's syndrome have reduced growth hormone levels. Preliminary investigations using rhGH in patients with Werner's syndrome should be considered.     

Differential effect of insulin-like growth factor-1 and growth hormone on hypothalamic regulation of growth hormone secretion in the rat

Pharmacological administration of either growth hormone (GH) or insulin-like growth factor 1 (IGF-1) were reported to inhibit endogenous GH release in humans and in the laboratory animal. We have evaluated the short-term differential mechanisms whereby the two hormones affect hypothalamic regulation of GH secretion. Wistar male rats (90 days old) were injected i.p. with either GH (recombinant GH NIAMDD, Baltimore, MD, USA), rIGF-1 (Fujisawa Pharmaceutical Co. Ltd., Osaka, Japan) or saline. Animals were sacrificed at 15, 30, 60 and 120 minutes following injection. Hypothalami were dissected and extracted immediately and the levels of growth hormone-releasing hormone (GHRH) and somatostatin were determined using specific antisera. Trunk blood was collected for GH and IGF-1 determination by RIA. Administration of IGF-1 or GH markedly decreased hypothalamic somatostatin stores by 77% and 54% respectively, within 15 minutes. Concomitantly, the wide range of GH levels found in the control group was reduced in the IGF-1 treated group suggesting that the pulsatile pattern of GH secretion was suppressed. Growth hormone administration induced an increase in hypothalamic GHRH stores (60% at 120 minutes). During this period serum IGF-1 levels were not altered. It is suggested that short term modulation of hypothalamic neurohormones by GH and IGF-1 is mediated by rapid stimulation of somatostatin release by both hormones, and inhibition of GHRH release is induced only by GH.     

Contributions of endogenous inhibin and estradiol to the regulation of follicle-stimulating hormone and luteinizing hormone secretion in the pregnant rat

To examine the contributions of endogenous inhibin and estradiol to the regulation of FSH and LH secretion in the pregnant rat, some rats were passively immunized against inhibin and/or estradiol, and others were ovariectomized, on Days 5, 10, 15, and 20 of pregnancy. Ovarian and uterine venous blood was collected separately to confirm the sources of inhibin and steroid hormones during pregnancy. Immunoreactivity of inhibin in the placenta was also examined by RIA. Levels of inhibin in ovarian venous plasma were significantly higher than those in peripheral plasma during pregnancy. No difference was observed between the levels of inhibin in uterine venous plasma and peripheral plasma. No immunoreactivity of inhibin was detected in placental homogenate from rats at Days 10, 15, and 20. FSH secretion significantly increased after immunoneutralization of inhibin during pregnancy. A marked increase in FSH secretion was noted on Days 5 and 20, and the smallest increase was observed on Day 15. Administration of estradiol antiserum (AS) alone did not induce a significant increase in FSH secretion on any day of pregnancy. However, a synergistic effect of estradiol AS and inhibin AS was observed on Day 20. On Days 5, 10, and 20, administration of inhibin AS or estradiol AS induced a significant increase in LH secretion. A synergistic effect of inhibin AS and estradiol AS on LH secretion was observed on Day 5. On Days 5 and 10, significantly high LH secretion was noted in ovariectomized rats as compared with that in rats treated with both inhibin AS and estradiol AS, indicating that other ovarian hormones such as progesterone may be involved in the suppression of LH secretion in these stages of pregnancy. These data indicate that both inhibin and estradiol, predominantly secreted from the ovary, are involved in the regulation of gonadotropin secretion during pregnancy as during the estrous cycle in the rat.     

Fasting suppresses pulsatile luteinizing hormone (LH) secretion and enhances orderliness of LH release in young but not older men

BACKGROUND: Studies identifying genes that are differentially expressed following induction of acute ischemic injury have been useful in delineating the pathophysiology of acute renal failure. METHODS: A differential cDNA library screening technique was used to identify genes that are differentially expressed in rat kidney following induction of acute ischemic renal injury. RESULTS: Levels of mRNA with a high homology to that coding for Siva, a human proapoptotic protein, were increased approximately 4.5-fold in kidneys obtained from rats within 12 hours following ischemia, compared to kidneys from sham-operated rats. A partial cDNA sequence for the rat protein (rat Siva) was determined that overlaps 92% of the human open reading frame. The cDNA sequence predicts a protein 177 amino acids in length with 76% homology to human Siva. Levels of rat Siva in kidneys were elevated at one, five and seven days post-ischemia were not different from those in kidneys from sham-operated controls. In situ hybridization demonstrated that rat Siva mRNA was expressed in cells lining damaged sections in the S3 segment of the proximal tubule at 12 hours and one day post-ischemia. At five and seven days, Siva mRNA was located in epithelial cells of regenerating tubules including in papillary proliferations. TdT-mediated dUTP-biotin nick end-labeling (TUNEL)-positive cells colocalized with cells containing Siva mRNA. CD27, the receptor for Siva was localized by immunohistochemistry to sloughed cells in the lumens of damaged S3 segments at 12 hours post-ischemia and to cells within papillary proliferations at five days post-injury. CONCLUSIONS: Siva that is produced within the kidney could be a mediator of apoptosis post-ischemia via an interaction with CD27.     

Elevated growth hormone secretory rate in premature infants: deconvolution analysis of pulsatile growth hormone secretion in the neonate

Premature infants have higher circulating concentrations of growth hormone (GH) than term infants. Previous investigations of these differences have used sampling frequencies of every 30 min with subsequent application of pulse detection algorithms, such as the CLUSTER program, to assess serum GH pulse parameters. To determine differences in GH secretory rates or GH t1/2 values between premature and term infants, we have sampled 11 neonates at 15-min intervals. We performed deconvolution analysis of the resultant plasma GH values to estimate GH secretory and clearance parameters. Five premature infants (gestational age range 24-34 wk) and six term infants (gestational age range 38-42 wk) were sampled every 15 min for 6 h. All subjects had indwelling arterial catheters. GH was measured (in duplicate) by RIA using 10 microL of plasma. Premature infants had higher secretory burst amplitudes (2.2 +/- 0.13 micrograms/L/min versus 1.4 +/- 0.27 micrograms/L/min, p = 0.02), higher production rates (product of the total number of bursts and the mean mass of GH secreted per burst, 811 +/- 173 micrograms/L/6 h versus 283 +/- 77 micrograms/L/6 h, p = 0.03), and a higher mass of GH per secretory burst (106 +/- 25 micrograms/L versus 38 +/- 11 micrograms/L, p = 0.049) than term infants. The integrated plasma GH concentration exhibited a strong trend toward a higher value in the premature infants (18,100 +/- 800 micrograms/L versus 10,200 +/- 2,700 micrograms/L, p = 0.067).(ABSTRACT TRUNCATED AT 250 WORDS)     

The effect of recombinant follicle stimulating hormone (Gonal-F) on endogenous luteinizing hormone secretion in women

Parenteral administration of follicle stimulating hormone (FSH) has been shown to lower luteinizing hormone (LH) concentrations in women undergoing ovulation induction. This study was designed to explore the physiological mechanism of this effect. Seven healthy women were recruited into a double-blind placebo-controlled study. LH secretion, after the administration of variable i.v. boluses (37.5, 75 and 150 IU) of recombinant FSH (Gonal-F), was evaluated. LH was measured at 10 min intervals for 2 h before and 4 h after the FSH/placebo infusion. LH pulse frequency and amplitude were evaluated and there was no significant difference between control and trial cycles for each subject. A linear regression analysis revealed that in the group receiving 150 IU FSH, the mean plasma LH concentration decreased significantly due to a reduction tonic LH secretion. This could be a result of the suppression of secretion or an alteration of clearance. This decrease was not seen in the other dosage groups, revealing that above a dosage threshold, FSH reduced non-pulsatile LH secretion. Therefore the effect of FSH in this study exposed the likely presence of two components of LH concentration: an FSH-sensitive, non-pulsatile tonic secretion and a gonadotrophin-releasing hormone-stimulated, pulsatile release that is unaffected by FSH. Although an indirect effect involving ovarian regulation is not excluded, the rapidity of the effect suggests that FSH acts directly on the pituitary gland.     

Influence of cardiovascular risk status on coronary flow reserve in healthy young men

To detect changes in vascular physiology associated with early atherosclerosis, we studied whether alterations in coronary flow reserve, as assessed by positron emission tomography imaging with intravenous dipyridamole, would be related to risk factor variables in healthy young men. The number of conventional risk variables correlated significantly with coronary flow reserve (r = -0.58, p = 0.0007), suggesting that alterations in functional vascular reactivity are related to the cardiovascular risk status already in healthy young men.     

Insulin-like growth factor I in fetal serum obtained by cordocentesis is correlated with intrauterine growth retardation

We examined whether insulin-like growth factor-I (IGF-I) and one of its binding proteins (IGFBP-1) in fetal serum obtained by cordocentesis is correlated with intrauterine growth retardation (IUGR) and weight estimation by ultrasound. Cordocentesis sera from 27 fetuses suspected of having IUGR were analysed for IGF-I and IGFBP-1 by radioimmunoassay. The results showed that IGF-I concentrations were correlated significantly with birth weight (P < 0.001) and placenta weight (P < 0.05). Mean fetal concentrations of IGF-I were 38 +/- 18 microg/l. In patients (n = 11) with a weight deviation at delivery <-33%, IGF-I concentrations were 24.1 +/- 13.2 microg/l. IGFBP-1 was inversely correlated with birth weight (P < 0.006) and concentrations of IGF-I. Mean plasma concentrations of IGFBP-1 were 234.2 +/- 161.4 microg/l. Furthermore, IGF-I concentrations were correlated with the weight deviation estimated by ultrasonography at the time of cordocentesis (P < 0.007), as well as with the weight deviation at delivery (P < 0.0001). The actual weight deviation at delivery was correlated more strongly with fetal IGF-I concentrations than with the estimated weight deviation at cordocentesis. The lowest concentrations of IGF-I were found in patients with a weight deviation <-33%. Very low concentrations of IGF-I are thus associated with IUGR, indicating that IGF-I measured in fetal serum may increase the predictive value of ultrasonographic weight estimation.     

Effect of timing of growth hormone administration on plasma growth-hormone-binding activity, insulin-like growth factor-I and growth in children with a subnormal spontaneous secretion of growth hormone

Since normal pulsatile growth-hormone (GH) secretion displays a major and consistent surge during sleep, we studied the effect of timing of GH supplementation on plasma GH-binding protein activity (GH-BP), insulin-like growth factor-I (IGF-I) and growth. 34 prepubertal subjects (28 boys, 6 girls) aged 8-11 years, of short stature (< 2 SD for age), with a GH response to provocative test > 10 micrograms/l and a subnormal 24-hour GH secretion (< 3 micrograms/l), were randomly allocated to receive Bio-Tropin (recombinant GH, Bio-Technology, Israel) 0.81 IU/kg/week in 3 equally divided doses. GH was administered either at 8.00-10.00 h (M group), 14.00-16.00 h (AN group) or 19.00-21.00 h (NT group). Height velocity, IGF-I and GH-BP were determined prior to and after 6 and 12 months on GH therapy in the three groups. There was no significant difference between the three groups in the growth response, IGF-I and GH-BP increase, all of which increased significantly during GH therapy. Although GH levels after the injection decline to preinjection levels after 10 h, the changes induced by GH therapy, as reflected in IGF-I and GH-BP, last in the circulation long enough to prevent fluctuations in its action. The similarity of IGF-I and of GH-BP levels in the three treatment groups might explain the similar growth effects of the 3 protocols.     

The importance of growth hormone in the regulation of erythropoiesis, red cell mass, and plasma volume in adults with growth hormone deficiency

Total body water (TBW) is reduced in adult GH deficiency (GHD) largely due to a reduction of extracellular water. It is unknown whether total blood volume (TBV) contributes to the reduced extracellular water in GHD. GH and insulin-like growth factor I (IGF-I) have been demonstrated to stimulate erythropoiesis in vitro, in animal models, and in growing children. Whether GH has a regulatory effect on red cell mass (RCM) in adults is not known. We analyzed body composition by bioelectrical impedance and used standard radionuclide dilution methods to measure RCM and plasma volume (PV) along with measuring full blood count, ferritin, vitamin B12, red cell folate, IGF-I, IGF-binding protein-3, and erythropoietin in 13 adult patients with GHD as part of a 3-month, double blind, placebo-controlled trial of GH (0.036 U/kg.day). TBW and lean body mass significantly increased by 2.5 +/- 0.53 kg (mean +/- SEM; P < 0.004) and 3.4 +/- 0.73 kg (P < 0.004), respectively, and fat mass significantly decreased by 2.4 +/- 0.32 kg (P < 0.001) in the GH-treated group. The baseline RCM of all patients with GHD was lower than the predicted normal values (1635 +/- 108 vs. 1850 +/- 104 mL; P < 0.002). GH significantly increased RCM, PV, and TBV by 183 +/- 43 (P < 0.006), 350 +/- 117 (P < 0.03), and 515 +/- 109 (P < 0.004) mL, respectively. The red cell count increased by 0.36 +/- 0.116 x 10(12)/L (P < 0.03) with a decrease in ferritin levels by 39.1 +/- 4.84 micrograms/L (P < 0.001) after GH treatment. Serum IGF-I and IGF-binding protein-3 concentrations increased by 3.0 +/- 0.43 (P < 0.001) and 1.3 +/- 0.15 (P < 0.001) SD, respectively, but the erythropoietin concentration was unchanged after GH treatment. No significant changes in body composition or blood volume were recorded in the placebo group. Significant positive correlations could be established between changes in TBW and TBV, lean body mass and TBV (r = 0.78; P < 0.04 and r = 0.77; P < 0.04, respectively), and a significant negative correlation existed between changes in fat mass and changes in TBV in the GH-treated group (r = -0.95; P < 0.02). We conclude that 1) erythropoiesis is impaired in GHD; 2) GH stimulates erythropoiesis in adult GHD; and 3) GH increases PV and TBV, which may contribute to the increased exercise performance seen in these patients.