Pituitary

, Volume 12, Issue 2, pp 143–152

GH/GHRH axis in HIV lipodystrophy

Authors

  • Takara L. Stanley
    • Program in Nutritional Metabolism and Neuroendocrine UnitMassachusetts General Hospital and Harvard Medical School
    • Program in Nutritional Metabolism and Neuroendocrine UnitMassachusetts General Hospital and Harvard Medical School
Article

DOI: 10.1007/s11102-008-0092-8

Cite this article as:
Stanley, T.L. & Grinspoon, S.K. Pituitary (2009) 12: 143. doi:10.1007/s11102-008-0092-8
  • 99 Views

Abstract

Approximately half of patients with HIV-infection develop abnormal body fat distribution, characterized by increased abdominal, breast, and dorsocervical adiposity and decreased fat in the limbs and face in association with antiretroviral therapy. Changes in fat distribution are associated with dyslipidemia, insulin resistance, and increased cardiovascular risk in patients with HIV lipodystrophy. Growth hormone secretion is reduced and responses to standardized stimulation testing altered, suggesting relative growth hormone deficiency in this population. Growth hormone secretion is characterized by normal pulse frequency, but decreased pulse amplitude, pulse width, and trough GH levels compared to weight matched, non-HIV-infected patients. Abnormalities in GH secretion are strongly associated with body composition and metabolic abnormalities in patients with HIV lipodystrophy, particularly with increased visceral fat and elevated free fatty acids. Increased somatostatin tone and decreased ghrelin concentrations may also contribute to reduced GH levels. Administration of exogenous GH or growth hormone releasing hormone (GHRH) to normalize growth hormone concentrations is effective to reduce visceral fat and improve lipid parameters in HIV-infected patients. Treatment with supraphysiologic GH is limited by side effects and exacerbation of insulin resistance, whereas administration of physiologic doses of GH demonstrates more modest treatment effects but fewer adverse effects. Initial studies of GHRH also show significant reductions in visceral adipose tissue (VAT) with potentially fewer adverse effects. GHRH may be particularly useful to normalize GH dynamics in patients with HIV lipodystrophy by increasing endogenous GH pulse height, GH pulse width, and trough GH levels, while preserving the negative feedback of IGF-I on pituitary GH secretion.

Keywords

HIVLipodystrophyGrowth hormoneGrowth hormone releasing hormone

Introduction

Patients with HIV infection commonly experience changes in fat distribution, including increased abdominal adiposity, breast enlargement, development of a dorsocervical fat pad (“buffalo hump”), and loss of fat in the limbs and face [1, 2]. This HIV lipodystrophy syndrome, which occurs in approximately one-half of patients with HIV, is particularly prevalent among patients receiving highly active antiretroviral therapy (HAART) [1, 2]. HIV lipodystrophy is accompanied by metabolic abnormalities, including insulin resistance, hyperlipidemia, and increased free fatty acids [37]. In addition, patients with HIV lipodystrophy have increased systemic inflammation, indicated by elevated c-reactive protein (CRP), IL-6, and monocyte chemoattractant protein-1 (MCP-1), as well as decreased adiponectin [8, 9]. HIV-infection and, independently, treatment with HAART also confer increased risk of cardiovascular disease, demonstrated by increased carotid intimal medial thickness (IMT) and increased incidence of myocardial infarction [1013]. Importantly, patients with HIV lipodystrophy have alterations in growth hormone dynamics that may both result from and exacerbate the increased visceral fat in these patients. In addition, reduced growth hormone secretion in these patients may contribute to the metabolic abnormalities and increased cardiovascular risk that accompany HIV lipodystrophy. Consequently, potential strategies for normalizing growth hormone secretion may be useful to reduce visceral adiposity and improve metabolic abnormalities in this population.

GH secretion in patients with HIV

Growth hormone dynamics in patients with HIV-infection vary according to body composition. In one of the first investigations of pulsatile GH secretion in patients with HIV, Heijligenberg et al. demonstrated that GH secretion in men with HIV or AIDS did not differ from healthy controls with respect to pulsatile pattern and pulse area [14]. Patients with HIV who experience weight loss or AIDS wasting, however, demonstrate a pattern of GH resistance, with elevated GH levels and reduced IGF-I [15, 16]. Overnight frequent sampling demonstrates increased mean overnight GH levels, with increased GH pulse frequency and increased basal GH levels in patients with HIV-associated weight loss [16]. Conversely, HIV-infected patients with fat redistribution and abdominal adiposity exhibit relative GH deficiency as described in detail below.

GH dynamics in HIV lipodystrophy

Men with HIV lipodystrophy demonstrate relative GH deficiency with apparently normal or increased GH sensitivity. Overnight frequent sampling demonstrates that mean overnight GH levels are reduced approximately 50% in HIV-infected men with fat redistribution compared to similar weight non HIV-infected subjects and men with HIV-infection but without fat redistribution (Fig. 1) [17]. Pulsatility analysis demonstrates that lower mean GH concentrations result from reductions in GH pulse amplitude, pulse width, and trough GH levels, without significant reduction in pulse frequency [7, 17]. As in the non-HIV infected population [18], body composition is a significant determinant of mean overnight GH secretion in these men, with visceral fat being the single most significant predictor of mean overnight GH concentrations in multivariate regression modeling (Fig. 2) [17].
https://static-content.springer.com/image/art%3A10.1007%2Fs11102-008-0092-8/MediaObjects/11102_2008_92_Fig1_HTML.gif
Fig. 1

Mean overnight growth hormone concentrations in patients with HIV lipodystrophy (LIPO, n = 21), patients with HIV-infection but no fat redistribution (NONLIPO, n = 20), and HIV-negative controls (CONTROL, n = 20). * P < 0.05 vs. NONLIPO;  P < 0.05 vs. CONTROL [17]. With permission from The Endocrine Society

https://static-content.springer.com/image/art%3A10.1007%2Fs11102-008-0092-8/MediaObjects/11102_2008_92_Fig2_HTML.gif
Fig. 2

Association between visceral fat area and mean overnight GH and GH area under the curve in response to GHRH stimulation [17]. With permission from The Endocrine Society

Men with HIV lipodystrophy also show decreased GH response to standard GHRH-arginine stimulation test. Almost 40% of men with lipodystrophy demonstrate a peak GH response less than 7.5 ng/ml [19, 20], and 18% demonstrate a deficient response using a stringent GH response criterion of <3.3 ng/ml (Fig. 3) [20]. Peak GH response in patients with HIV is negatively associated with body composition indices, including waist-to-hip ratio (WHR) [19] and visceral adipose tissue (VAT) (Fig. 2) [17, 20]. In men with HIV lipodystrophy, peak GH response is also inversely related to insulin levels and free fatty acids in a multivariate model controlling for age and BMI [20].
https://static-content.springer.com/image/art%3A10.1007%2Fs11102-008-0092-8/MediaObjects/11102_2008_92_Fig3_HTML.gif
Fig. 3

Percentage of failing responses to GHRH-arginine stimulation testing at various cutoffs in patients with HIV lipodystrophy (n = 39), HIV-infected patients without lipodystrophy (n = 17), and HIV-negative controls (n = 16). * P ≤ 0.05 by compared to HIV-infected non-lipodystrophic group;  P ≤ 0.05 by comparison with control group [20]. With permission from The Endocrine Society

In contrast to GH, significant variability has been shown in IGF-I in men with HIV lipodystrophy, which may relate to the unique characteristics of individual study populations, nutritional status, gonadal function and also to assay differences among studies. Andersen et al. report similar baseline GH levels but increased IGF-I and IGFBP-3 levels in men with lipodystrophy compared to HIV-infected non-lipodystrophic controls, suggesting that lipodystrophy is accompanied by increased GH sensitivity [21]. Two prior studies demonstrated decreased GH levels but similar IGF-I levels in men with HIV lipodystrophy compared to BMI-matched HIV-infected non-lipodystrophic controls. In this situation, the relative preservation of normal IGF-I concentrations may result from increased GH sensitivity and/or increased insulin levels in the lipodystrophic population [7, 17]. Alternatively, lower IGF-I and IGFBP-3 levels have been reported in men with HIV and fat redistribution compared to HIV-infected men without lipodystrophy and healthy controls [20]. Taken together, the evidence to date demonstrates reduced GH secretion without significant reductions in IGF-I in most studies of men with HIV lipodystrophy.

Few studies have characterized GH dynamics in women with HIV lipodystrophy. In contrast to men, differences in GH secretion are less obvious between HIV-infected women and BMI-matched control subjects. Nonetheless, clear gender differences exist in GH secretion between HIV-infected men and women, and body composition is an important determinant of GH secretion among female HIV-infected patients. In response to GHRH-arginine stimulation testing, HIV-infected women have significantly higher peak GH and GH area-under-the-curve (AUC) compared to HIV-infected men (36.4 ± 7.3 vs. 18.9 ± 2.0 ng/ml, female vs. male, P = 0.003) [19]. IGF-I levels, however, are significantly lower than those in HIV-infected males (265 ± 27 vs. 356 ± 13 ng/ml, female vs. male, P = 0.004) [19]. Differences in steroid milieu between men and women are likely to contribute to these differences, with estrogen suppressing production of IGF-1, thereby reducing pituitary inhibitory feedback to GH release. Importantly, compared to HIV-negative women, women with HIV-infection show similar peak GH response to GHRH-arginine stimulation [19]. As is the case with males, peak GH in HIV-infected women is negatively associated with BMI, waist circumference, and WHR [19]. Overnight pulsatile GH secretion has not yet been investigated in HIV-infected women.

Etiology of altered GH dynamics in HIV lipodystrophy

The etiology of decreased GH secretion in HIV lipodystrophy appears to be threefold: increased somatostatin tone, decreased ghrelin, and direct suppression of GH by elevated free fatty acids (Fig. 4) [7]. These factors were investigated in a study comparing weight matched HIV-infected men with lipodystrophy, HIV-infected men without lipodystrophy, and healthy controls [7]. Subjects in this study underwent a GH stimulation test with GHRH alone and, on a separate occasion, stimulation with GHRH combined with arginine to block somatostatin tone. Peak GH response to GHRH alone was significantly reduced in men with HIV lipodystrophy compared to both HIV-infected and healthy controls. Peak GH response to combined GHRH-arginine stimulation was also significantly lower in men with HIV lipodystrophy, but the percent increase in GH response with addition of arginine was significantly greater in men with lipodystrophy. Using a subtraction algorithm, it was demonstrated that men with lipodystrophy had a 247% increase in peak GH with addition of arginine, compared to increases of 102% and 137%, respectively, in the HIV-infected non-lipodystrophic group and in healthy controls. Similarly, GH area under the curve increased 261% with addition of arginine in the lipodystrophy group, compared to 113% and 132% in HIV-infected non-lipodystrophy and healthy controls groups [7]. The two-fold greater augmentation of GH response with addition of arginine in HIV lipodystrophy compared to controls suggests that increased somatostatin tone may contribute to lower GH secretion in this group.
https://static-content.springer.com/image/art%3A10.1007%2Fs11102-008-0092-8/MediaObjects/11102_2008_92_Fig4_HTML.gif
Fig. 4

Schema for the mechanisms of reduced GH secretion in HIV lipodystrophy [7]. With permission from The American Physiological Society

Decreased ghrelin also leads to lower GH concentrations in men with HIV lipodystrophy. In the cohort described above, fasting ghrelin concentrations were significantly lower in men with HIV lipodystrophy compared to HIV-infected men without lipodystrophy and to healthy controls. Moreover, in men with HIV lipodystrophy, fasting ghrelin was positively associated with mean overnight GH secretion (r = 0.73, P = 0.01) and GH pulse area (r = 0.59, P = 0.05), whereas these associations were not evident in the control group or the HIV-infected non-lipodystrophy group. As demonstrated in HIV-negative patients, elevated fasting insulin likely contributes to decreased ghrelin in these patients. In the HIV lipodystrophy group, there was a significant negative association between fasting ghrelin and both fasting insulin (r = −0.57, P < 0.01) and insulin area under the curve following 75 g oral glucose tolerance test (r = −0.51, P = 0.02) [7].

Elevated free fatty acid concentrations also decrease GH secretion in patients with HIV lipodystrophy. In men with HIV lipodystrophy, significantly higher fasting free fatty acid levels have been shown compared to both HIV-infected patients without lipodystrophy and controls [7]. Overnight administration of acipimox decreased free fatty acids and augmented GH response to GHRH. Moreover, increase in GH response was highest in those patients who had the greatest reduction in free fatty acids [7]. The change in peak GH following GHRH stimulation pre- and post-acipimox was significantly associated with the decrease in free fatty acids in the HIV lipodystrophy group (r = −0.61, P = 0.04), whereas this association was not evident in the healthy control group or in patients with HIV but no lipodystrophy [7].

The physiologic regulation of GH in HIV is similar in many ways to that in the non-HIV-population. Multiple studies have reported increased somatostatin tone related to obesity and/or visceral fat [18, 2225], and administration of arginine to block somatostatin tone in obese patients partially restores GH response to GHRH [22, 23, 26]. Moreover, visceral fat is a significant negative predictor of GH response to submaximal GHRH-arginine stimulus [22]. Pharmacologic arginine administration does not completely restore normal GH response to GHRH, however, suggesting that visceral fat also decreases GH secretion via non-somatostatinergic pathways [22, 23, 26, 27]. Elevated insulin levels during euglycemic hyperinsulinemic clamp reduce ghrelin in healthy subjects [28], and hyperinsulinemia is a probable mechanism for lower ghrelin concentrations in patients with HIV lipodystrophy. Finally, the finding that free fatty acids diminish GH secretion in patients with HIV lipodystrophy is consistent with in vitro and in vivo studies in rat pituitary demonstrating that free fatty acids directly inhibit pituitary somatotroph response to GHRH [29, 30]. Maccario et al. report corroborating findings in a human study, in which acipimox significantly increases GH response to stimuli in obese patients but has no effect on normal-weight controls [31].

Treatment strategies to normalize GH secretion in HIV lipodystrophy

While metabolic and body composition alterations appear to contribute to reduced GH secretion in HIV lipodystrophy, relative GH deficiency may also exacerbate abdominal obesity, dyslipidemia, and increased cardiovascular risk in these patients. Patients with pituitary GH deficiency demonstrate increased abdominal adiposity along with dyslipidemia [3234], insulin resistance [35], and increased carotid IMT [33, 34], all of which are common in HIV lipodystrophy. In addition, adults with hypopituitarism who are given conventional hormone replacement but not treated with GH replacement demonstrate increased risk of cardiovascular disease and consequent mortality [36, 37]. Addressing relative GH deficiency in patients with HIV lipodystrophy may therefore be useful to prevent a vicious cycle whereby fat redistribution and metabolic abnormalities result in decreased growth hormone, which further potentiates abdominal adiposity, hyperlipidemia, inflammation, and cardiovascular disease.

Use of GH or GHRH to normalize GH concentrations in patients with HIV lipodystrophy has multiple potential benefits. Growth hormone treatment in non-HIV-infected patients with abdominal obesity decreases visceral fat and improves lipid parameters [38, 39]. Likewise, GH replacement in GH deficient adults significantly decreases visceral fat [40] and lowers total cholesterol [41, 42]. In addition, physiologic GH replacement in adults with GH deficiency decreases IL-6 and CRP [43, 44] and decreases carotid IMT [33, 45]. Paradoxically, GH treatment also has the potential to increase insulin sensitivity by reducing visceral fat. A recent long-term study of five years of physiologic GH replacement in GH deficient adults showed that insulin sensitivity initially worsened with treatment. However, mean 2 h glucose levels following 75 g glucose challenge actually decreased over the last two years of the study, suggesting an increase in insulin sensitivity that likely resulted from significant decreases in VAT with long-term GH administration [46]. Similarly, Franco et al. demonstrate an increase in glucose disposal during euglycemic hyperinsulinemic clamp after one year of relatively low dose GH (0.67 mg/day) in post-menopausal women with abdominal obesity [38], and Johannsson et al. show higher glucose disposal rates after nine months of low dose (9.5 mcg/kg) GH treatment in abdominally obese men [39]. These results suggest that normalization of growth hormone secretion in HIV lipodystrophy may also potentially improve a number of the metabolic abnormalities associated with visceral fat accumulation in the HIV population.

Recombinant human GH in treatment of HIV lipodystrophy

Growth hormone was first administered to patients with HIV-infection for the treatment of AIDS wasting, and it is FDA approved for this indication. In patients with HIV-associated weight loss, subcutaneous GH administration at a very high dose of 6 mg daily has been used to overcome GH resistance and to increase body weight [4753], build lean body mass [48, 5053], and decrease protein oxidation [47, 54]. The effects of GH to increase lean mass and decrease abdominal adiposity have also made it an attractive potential treatment for HIV lipodystrophy, and numerous clinical trials confirm that GH improves body composition and lipid parameters in patients with HIV lipodystrophy (Table 1) [5569]. In the two largest of these trials, GH 4 mg daily for 12 weeks decreased visceral fat by approximately 20%, lowered total cholesterol levels, and increased HDL cholesterol [58, 59, 63]. However, IGF-I levels achieved in these studies were supraphysiological. In smaller trials, GH has also shown benefit in decreasing dorsocervical fat accumulation [56, 60, 63] and improving quality of life indices, albeit inconsistently [57]. Almost universally, however, GH treatment in HIV lipodystrophy also decreases insulin sensitivity, which is already impaired in this population [5658, 64]. Studies investigating GH administration at 3 mg/day demonstrate decreased insulin-stimulated glucose uptake [56], impaired insulin-induced suppression of gluconeogenesis and lipolysis [64], and increased endogenous hepatic glucose production in patients with HIV lipodystrophy [64]. Furthermore, the increased lipolysis and decreased hepatic de novo lipogenesis shown with growth hormone administration [64] may decrease not only visceral adipose tissue, but also subcutaneous adipose tissue and limb fat [57], which are already abnormally reduced in this population.
Table 1

Trials of GH treatment in HIV lipodystrophy

Author

N

RCT

Rx

LM

FM

VAT

SAT

HDL

LDL

IS

Wanke et al. [55]

10

N

6 mg/day × 12 weeks

(↓ WHR)

    

Lo et al. [56]

8

N

3 mg/day × 6 months

 

 

Engelson et al. [57]

30

N

6 mg/day × 24 weeks

  

  

Kotler et al. [58, 59]

245

Y

4 mg/day vs. 4 mg QOD × 12 weeks

↓ 18.8% vs. 16.8%

Lo et al. [60]

5

N

1 mg/day × 6 months

Andersen et al. [21]

6

N

0.7 mg/day × 16 weeks

(↓ WHR)

 

Luzi et al. [61]

30

Y

0.2 IU/kg/week vs. placebo × 6 months

 

    

Bickel et al. [62]

26

Y

4 mg/day vs. 4 mg TIW × 12 weeks

 

↓ 27.4% vs. 28.5%

 

Grunfeld et al. [63]

326

Y

4 mg/day vs. placebo × 12 weeks

 

↓ 20.3%

Abbreviations: Rx, treatment and duration in randomized phase of each trial; LM, lean mass; FM, fat mass; VAT, visceral adipose tissue; SAT, subcutaneous adipose tissue; HDL, high density lipoprotein; LDL, low density lipoprotein; IS, insulin sensitivity, assessed in different trials by varying methods, including fasting insulin and glucose, insulin and glucose AUC following OGTT, or glucose utilization during euglycemic hyperinsulinemic clamp

The optimal dose of GH for patients with HIV lipodystrophy has not yet been determined. The dose of 6 mg daily used in HIV-associated weight loss was very high and designed to overcome GH resistance seen among patients with AIDS wasting. In contrast, patients with HIV lipodystrophy appear to have normal or even increased sensitivity to GH. Initial studies using a GH dose of 6 mg daily in lipodystrophy showed dramatic reductions in VAT along with decreased insulin sensitivity and relatively frequent treatment-limiting side effects, including arthralgias, myalgias, hypoesthesia, and fluid retention, likely related to the supraphysiological IGF-I levels achieved [55, 57]. A relatively lower GH dose of 4 mg daily decreases VAT by approximately 20% or more, but this dose also worsens glucose tolerance and causes side effects in a significant percentage of patients [58, 59, 62, 63]. Alternate day dosing (4 mg every other day) appears to yield similar benefits in terms of VAT reduction while causing fewer side effects [58, 59, 62]. Significantly lower, physiologic GH doses designed to maintain IGF-I levels in the normal range have been studied in three small trials, all of which show modest effect on body composition but, importantly, do not demonstrate worsened insulin sensitivity [60, 61, 66].

The treatment effect of growth hormone is not sustained after treatment discontinuation, with body composition changes returning to baseline after 12 weeks in several studies [5759]. While GH certainly has beneficial effects on body composition and lipids at high doses, its efficacy is limited by adverse effects on insulin sensitivity, treatment limiting side effects, and lack of durability. GH treatment is not currently FDA approved for treatment of HIV lipodystrophy.

GHRH in treatment of HIV lipodystrophy

An alternative treatment paradigm in HIV lipodystrophy is to use GHRH, which augments endogenous GH pulsatility and preserves the negative feedback of IGF-I on the pituitary. In contrast to the non-pulsatile administration of GH via subcutaneous injection, treatment with GHRH improves a number of the abnormalities in pulsatile GH secretion found in HIV lipodystrophy, namely reduced GH pulse area and decreased GH valley mean levels. In the first study of GHRH in patients with HIV, 31 men and women with HIV lipodystrophy were randomized to GHRH1–29 (EMD Serono Inc., Rockland MA) 1 mg SC BID vs. placebo for 12 weeks [70]. GHRH treatment increased valley mean and nadir GH concentrations, but did not affect GH pulse frequency. In the GHRH treatment group, IGF-I increased significantly by an average of 104 ng/ml. Lean body mass increased by an average of 0.9 kg, and trunk fat measured by DEXA decreased by an average of 0.4 kg in the treatment group, with both changes reaching significance compared to placebo. Visceral fat measured by cross-sectional abdominal CT scan also decreased significantly, with a reduction of 9% in the GHRH group compared to a 1% increase in the placebo group. In contrast to studies of high dose GH, there was no change in parameters of insulin sensitivity, including fasting glucose, fasting insulin, HbA1c, and glucose and insulin areas under the curve following OGTT [70]. Subsequent analysis demonstrated that bone markers also changed significantly in response to GHRH, with increase in c-terminal peptide and n-terminal propeptide of type 1 procollagen [71]. GHRH1–29 is now only available for diagnostic use.

Two recent studies have investigated the use of GHRH1–44 (TH9507, Theratechnologies, Inc., Montreal, Canada) in HIV lipodystrophy. In the first trial, 61 men with HIV and increased waist circumference and waist-to-hip ratio were randomized to 1 mg vs. 2 mg of TH9507 SC daily for 12 weeks [72]. The 2 mg dose proved more efficacious, resulting in a decrease of 9.2% in trunk fat, which was significant compared to placebo. Lean body mass and the ratio of visceral to subcutaneous adipose tissue were also significantly decreased in both treatment groups. As in the previous study of GHRH1–29, there were no changes in glucose in the treatment groups [72]. A larger, multicenter randomized placebo-controlled trial recently examined GHRH1–44 2 mg daily vs. placebo for 26 weeks in 412 patients with HIV and abdominal adiposity [73]. After 26 weeks of treatment, visceral fat decreased by 15.2% in the treatment group compared to a 5% increase in the placebo group (Fig. 5). As shown in Fig. 5, the magnitude of reduction in VAT was directly associated with the degree of visceral adiposity at baseline. In addition, triglycerides decreased significantly by 50 mg/dl in the treatment group, along with a reduction in total cholesterol and an increase in HDL [73]. In an extension phase of this study, patients initially randomized to TH9507 were re-randomized to TH9507 vs. placebo for an additional 26 weeks, and patients initially randomized to placebo were switched to TH9507 [74]. Patients who received TH9507 for the entire study period of 52 weeks demonstrated an overall 18.1% decrease in VAT and did not demonstrate any changes in glucose tolerance [74]. Patients who had initially received active treatment and were switched to placebo demonstrated reaccumulation of VAT to baseline levels over 26 weeks, suggesting that the beneficial effect of TH9507 to reduce VAT is not maintained after treatment discontinuation [74]. In addition, significant improvements in quality of life were reported, particularly related to distress of increased abdominal girth. In contrast to the studies with high dose GH, and consistent with the physiological nature of the GHRH dosing, GH related side effects, such as arthralgias and edema, were not seen.
https://static-content.springer.com/image/art%3A10.1007%2Fs11102-008-0092-8/MediaObjects/11102_2008_92_Fig5_HTML.gif
Fig. 5

Changes in visceral adipose tissue over 26 weeks with use of Tesamorelin (TH9507). Panel a shows mean % change in visceral adipose tissue in the Tesamorelin (TH9507) group, treated with 2 mg TH9507 daily for 26 weeks, vs. the placebo group. Panel b shows the relationship between changes in visceral adipose tissue and baseline VAT values in the treatment (lower line) and placebo (upper line) groups [73]. With permission from the Massachusetts Medical Society

Together these trials suggest that GHRH may be an effective treatment to normalize endogenous GH pulse dynamics and decrease visceral fat while preserving insulin sensitivity. Like GH, however, the beneficial effects of GHRH do not appear to be sustained after discontinuation, suggesting that long-term treatment may be needed. GHRH is not currently FDA approved for use in HIV lipodystrophy.

Conclusions

Men with HIV lipodystrophy demonstrate reduced GH secretion secondary to decreased GH pulse height, pulse width, and valley mean levels with preserved GH pulse generation. The etiology of altered GH dynamics in these patients relates to the body composition and metabolic abnormalities of the lipodystrophy syndrome. Visceral fat contributes to increased somatostatin tone, hyperinsulinemia decreases ghrelin, and elevated free fatty acids directly inhibit GH release from somatotrophs. In women with HIV lipodystrophy, GH dynamics are not well characterized, but women with HIV-infection appear to have better preserved peak GH responses to GHRH-arginine stimulation.

Currently, there are no FDA approved treatments for HIV lipodystrophy. Two strategies to normalize GH secretion, namely GH and GHRH, have proven effective in reducing visceral fat, although neither appears to have sustained benefit beyond treatment discontinuation. Growth hormone at supraphysiologic doses exacerbates insulin resistance and results in treatment limiting side effects. GHRH does not appear to affect insulin sensitivity and may potentially normalize endogenous GH secretion by increasing GH pulse area. Further studies of both treatments are needed, and studies of long-term low dose, physiologic GH will be important. In addition, weight loss has the potential to normalize GH secretion in non-HIV infected patients with obesity [75, 76], and lifestyle modification remains an important treatment strategy for patients with HIV lipodystrophy.

Acknowledgements

The authors Takara Stanley and Steven Grinspoon were funded by NIH T32 HD052961-03 and NIH R01 DK63639, respectively.

Conflict of interest statement

Dr. Grinspoon has received consulting fees and research support from EMD Serono and Theratechnologies, Inc.

Copyright information

© Springer Science+Business Media, LLC 2008