1Section of Endocrinology, 2Section of Internal Medicine, Department of Internal Medicine and Medical Specialties, 3Center of Excellence for Biomedical Research, University of Genoa, Italy
OBJECTIVE: This prospective study
aimed to evaluate the impact of growth hormone deficiency (GHD) on cardiac
autonomic tone and on cardiovascular risk and the changes after 12 months of GH
replacement therapy (GHRT). GHD is associated with increased cardiovascular
morbidity and mortality. This has been attributed to increased markers of
cardiovascular risk and to abnormalities of both the cardiac and peripheral
autonomic nervous systems. The autonomic cardiac nervous system (ACNS) can be
indirectly evaluated by analysis of heart rate variability (HRV) in
clinostatism and orthostatism.
DESIGN: We compared 14 GHD patients at baseline
and after 12 months of GHRT and 17 healthy controls. We analyzed a number of
cardiovascular risk factors and we used analysis of HRV during the Tilt Test
that identified high frequency (HF) and low frequency (LF), representing
parasympathetic and sympathetic activity, respectively.
RESULTS: Compared with the
control group, in either clinostatism or in orthostatism our patients showed a
significantly lower value of LF (P=0.047; P=0.004, respectively), whereas HF
was significantly reduced in orthostatism (P=0.037), and indicatively in
clinostatism (P=0.065). These values remained unchanged after 12 months of
GHRT. No statistically significant differences were found between the LF/HF
ratio in untreated and treated patients. In GHD patients, there was a
significant reduction of cardiovascular risk in 12 months of replacement
therapy (P=0.002).
CONCLUSIONS: Our study highlights the absence of
sympathovagal imbalance in GHD patients; GHRT does not change ACNS during the
first year of GH treatment but it reduces the absolute cardiovascular risk in
these patients.
Autonomic nervous system, GHD, IGF-I
INTRODUCTION
It is known that patients with hypopituitarism have an increased risk of morbidity and mortality from cardiovascular and cerebrovascular disease.1-3 This elevated risk has been attributed to increased cardiovascular risk factors,4 such as abnormal body composition,5 dyslipidemia,6-8 hypertension, increased premature atherosclerosis and cardiac dysfunction.1,9-12 However, two studies have reported that abnormalities both in the cardiac autonomic nervous system13 and in peripheral sympathetic activity may contribute to the increased risk of cardiovascular morbidity in growth hormone deficiency (GHD) patients.14 Indeed, in hypopituitarism, sympathovagal imbalance is generally due to an inadequate replacement therapy with glucocorticoid, thyroid hormones and gonadal steroids, whereas the involvement of GH is still controversial.15
The cardiovascular function in GHD can be negatively influenced directly by the decrease of diastolic filling, cardiac performance and contractility, and indirectly by the induced hypercoagulability, insulin resistance, truncal obesity, decreased muscle performance, dyslipidemia, reduced pulmonary capacity and early features of atherosclerosis.16 Indeed, increase of intima-media thickness of the carotid arteries and decreased physical capacity have been shown in these patients.17 Conversely, the positive effects of GH replacement therapy are well known and several studies have documented that GH administration may have a trophic effect on health.18-20
A significant relationship has been reported in many pathophysiological conditions between autonomic nervous system imbalance and cardiovascular mortality, including sudden cardiac death.21-23 Some studies have shown that cardiac autonomic tone could be altered in patients with myocardial infarction24,25 and heart failure26 undergoing percutaneous transluminal coronary angioplasty.27 Sympathovagal imbalance is also a common finding in diabetes and the usefulness of standard cardiovascular tests in assessment of cardiac autonomic neuropathy in diabetes is well documented.28
There is evidence implicating autonomic system dysfunction in cardiovascular disease. Indeed, cardiovascular dysfunction has been linked to diabetic autonomic neuropathy in diabetic patients.29 It has been suggested that early recognition of autonomic dysfunction is important in cardiovascular risk assessment and management.30
Experimental data are available on the impact of GH excess on heart rate variability in transgenic mice overexpressing bovine GH. These animals, representing an experimental model of acromegaly, GH suppresses sympathetic tone, most probably by reducing the availability of norepinephrine.31 A recent study by our group has demonstrated that acromegalic patients present with sympathovagal imbalance (due to vagal hypertone) similar to that found in diabetics.32 There are indications that the somatotropic axis is involved in the regulation of sympathetic activity.33-35
One common and easy technique for the study of the autonomic nervous system is the evaluation of heart rate variability (HRV) during postural changes. Clinical and experimental data suggest that this technique is a powerful predictor of future arrhythmic events.36 There is one important study showing an impairment of sympathetic tone in adult patients with GHD that normalized after 6-12 months of GHRT; however, in this work the time of GHD onset in the enrolled patients was mostly longer than 2 years (median about 13 years).37
The aim of the present study was to prospectively evaluate the heart autonomic tone in patients with recent onset of GHD and the effect of a 12 month GH replacement therapy on it.
SUBJECTS AND METHODS
Patients and protocol
The study group comprised 14 patients with a diagnosis of GHD (7 women, 7 men; age range 28-81 years) selected from a cohort of 150 hypopituitary patients. Highly restrictive inclusion/exclusion criteria were used in order to enroll only cases without additional cardiovascular risk factors.
Inclusion criteria were: diagnosis of adult GHD established by GH response to the insulin tolerance test (ITT) <3 ng/ml or by an inadequate peak after GHRH/arginine stimulation (Guidelines of the Growth Hormone Research Society, 2007); subject older than 18 years old. Any other deficient pituitary axes were adequately replaced in all patients before study entry.
Exclusion criteria were: GHD onset more than 2 years previously, prior rGH therapy, obesity, diabetes mellitus, myocardial infarction, heart failure, hypertension, smoking or any medication that might interfere with autonomic regulation, family history of cardiovascular disease. The GHD patients were compared to 17 healthy controls (8 women, 9 men; age range 26-75 years) selected using the same criteria.
Apart of the assessment of heart autonomic tone, body mass index (BMI), systolic and diastolic blood pressure, IGF-I, lipid profile, blood glucose and glycated haemoglobin (Hb1Ac) were also measured at baseline and after 12 months of rGH treatment (average initial dose of 1.2 ±0.19 mg/week) titrated on the basis of IGF-I value at 3-6-12 months. Our ethics committee had approved the protocol and written informed consent was obtained from the patients.
Autonomic tests
Autonomic tests were performed by power spectral analysis of heart rate variability in clinostatism and orthostatism, using a frequency domain method. HRV, which is the variability in R-R interval from one heartbeat to the next, is a measure of autonomic nervous system balance.38
An electrocardiographic recording of R-R intervals was made at 10:00 h in a quiet room reserved for this purpose, at least 1 h after venipuncture for routine hormonal evaluation. Subjects were placed supine on a mechanically driven tilt-table and instructed to relax, stay awake, breathe regularly and not to speak. After supine resting for about 10 min to stabilize blood pressure and heart rate, clinostatic R-R intervals were obtained (clinostatism values). Immediately thereafter, each patient was slowly and passively brought into the upright position by rising the table over a 30-sec period and the recording was repeated (orthostatism values). In both clinostatic and orthostatic postures, the EKG was recorded for 330 sec by means of an electrocardiograph connected to a PC equipped with software that sampled the analogical signal at about 200 Hz using an analogic/digital converter. Each R-R interval was measured in milliseconds and memorized as a tachogram (R-R interval duration vs. number of heartbeats). Two series of data, corresponding to clinostatic (in tachogram A) and orthostatic (in tachogram B), R-R intervals, were analyzed in all subjects using a parametric method based on the autoregressive model for the quantification of HRV signals. The main power densities in the high frequency (HF; 0.15– 0.4 Hz) and low frequency (LF; 0.04–0.15 Hz) bands were identified for each density spectrum. In addition, LF/HF ratios were calculated in both clinostatism and orthostatism. An LF/HF greater than 1 is considered normal.
Power spectral analysis identifies 3 peaks of power: a peak of HF, which expresses vagal activity; a peak of LF, which expresses sympathetic activity; and a peak of very LF, which is of uncertain significance. LF and HF are quantitative indicators of neural control of the sinoatrial node; in particular, the LF component is a marker of sympathetic modulation, whereas the HF component is a marker of vagal modulation. LF/HF is regarded as an index of sympathovagal balance in the frequency domain.24,25,30,39 All HRV parameters were measured in square milliseconds.
Coronary risk
Absolute cardiovascular risk assessment provides an estimate of the probability, expressed as a percentage, of developing a first major cardiovascular event over the next 10 years based on a number of risk factors. Cardiovascular risk charts have been elaborated from these mathematical functions, considering major cardiovascular risk factors: sex, age, blood pressure, total and HDL cholesterol levels and cigarette smoking.40 This chart (Risk Card 2009) offers the option of evaluating both protective factors (as for instance HDL cholesterol levels) and risk factors in the cardiovascular risk assessment.41
Statistical analysis
Statistical analysis was performed using the SPSS 11.0 program. All data are presented as mean SDS. The observed values for each variable in patients and controls group were analyzed using the Mann-Whitney test for independent data. The observed values for each variable in patients before and after treatment were analyzed using the Wilcoxon test for paired data. The correlations between variables were assessed using the Pearson correlation coefficient. A p value <0.05 was considered statistically significant.
RESULTS
As expected, IGF-I levels before treatment were significantly lower than normal and increased significantly after 1 year of GHRT (IGF1SDS -1.28±0.40; absolute values 82.61 ± 12.13 ng/ml vs. IGF1SDS 0.39±1.07; absolute values 156.76 ± 16.04 ng/ml, respectively. P IGF1SDS= 0.002; P IGF1 absolute values= 0.002).
Autonomic tests
HRV values in patients, before and after 1 year of therapy, and controls are shown in Table 1. HF (vagal activity) decreased in all subjects during the test. Compared with the control group, in clinostatism the patients showed a significantly lower value of LF (P=0.047) and a reduced, though non-significantly, HF value (P=0.065); in orthostatism the values of LF and HF were significantly reduced in GHD patients compared to the normal subjects (P=0.004 and P=0.037, respectively) (Figure 1). No statistically significant differences were found between the LF/HF ratio in untreated patients and controls, both in clinostatism and orthostatism (Figure 2). LF, unlike what was found in controls, was reduced in the transition from clinostatism to orthostatism, although not significantly (Figure 1). No significant changes were observed after 12 months of GH replacement therapy (Figures 1, 2).
Figure 1. Box and whisker plots of LF and HF in both clinostatism (a) and orthostatism (b) in the study groups. The lines in the boxes represent the median; the boxes are the 25-75th percentile range and the whiskers are the 10-90th percentile range. LF: low frequency; HF: high frequency.
Figure 2. Box and whisker plots of LF/HF in both clinostatism and orthostatism in the study groups. The lines in the boxes represent the median; the boxes are the 25-75th percentile range and the whiskers are the 10-90th percentile range. LF: low frequency; HF: high frequency.
Cardiovascular risk factors
BMI was between 25 and 30 kg/m² at baseline and after 1 year of treatment. Total cholesterol, and triglycerides levels were slightly above normal ranges (200 mg/dl and 150 mg/dl, respectively); HDL cholesterol was normal (>40-50 mg/dl in males and females, respectively) and did not significantly change after therapy (P=0.662; P=0.696; P=0.181, respectively). However, a downward trend in total cholesterol and upward trend in HDL was evident. Conversely, LDL cholesterol was significantly reduced (P=0.022) (Table 2).
Systolic pressure was normal (<140 mmHg) but higher before therapy start, while it was significantly decreased after one year of treatment (136.43±3.41 mmHg vs. 126.8±3.87 mmHg, respectively, P=0.005). In addition, the diastolic pressure showed a significant decrease after treatment (82.50±2.76 mmHg vs. 78.21±1.45 mmHg, respectively, P=0.003) and reached values comparable to those found in the control group (76.71±1.46 mmHg).
Blood glucose levels did not significantly increase (P=0.421), whereas Hb1Ac values were significantly increased (P=0.023), but remained within the normal range (<6%).
All patients were treated with levothyroxine (dose range: 50-150 mcg daily) and cortisone acetate (dose range: 25-75 mg daily) and displayed adequate levels of fT4, sodium - potassium, and urinary 24 hours free cortisol (Table 3). Three of seven female patients were taking replacement estrogen-progestinic therapy given in the premenopausal period (one transdermal, two oral route). All male patients were taking substitutive testosterone therapy (four transcutaneous and three by intramuscular route). During the 12 months of the study, no changes in the doses of these therapies were necessary since the hormonal assessment constantly displayed an adequate replacement of all the axes. Although the 10-year absolute coronary risk was not elevated in any of the patients (8.8±7.4%) at diagnosis, the value of risk after one year of therapy was reduced in 12 of the 14 patients with GHD (6.6±4.6%), and particularly in those displaying the higher values before the start of treatment with rGH. Therefore, these data show a significant reduction of cardiovascular risk in 12 months of replacement therapy (P=0.002).
DISCUSSION
GHD can lead to cardiovascular events and thus to increased mortality.1-3 Early diagnosis, treatment and follow-up enable the reduction of this risk.18-20 As in this study we have considered only non-diabetic, non-smoking patients without any known heart disease or history of cardiovascular events, the role of GH per se was highlighted without confounding factors. In our study, we prospectively evaluated the autonomic system, lipid and metabolic profile and blood pressure to explore the “pure” role of GHD status on cardiovascular risk in an early phase of disease. In fact, the effect on cardiac function in patients treated with GH is unknown so far. The data available in the literature primarily refer to long-term treated patients who already display impaired cardiovascular function. Surprisingly, the rather precocious lipid profile and blood pressure modifications in these patients suggest that even a short-term period of GHD in patients without clinically evident vascular alterations results in an increased risk of coronary disease. Moreover, the lack of correlation between the GHRT and autonomic function suggests that increased cardiovascular mortality, reported in patients with GHD, does not seem dependent on dysregulation of the autonomic system, this being at variance with acromegalic and diabetic patients.29,31,32 Nevertheless, it must be underlined that our data describe no clinically significant alterations in sympatovagal balance in an early phase of GHD onset, while other studies have demonstrated it in longer disease duration.37 This could be explained by the fact that GHD disease duration is the most important risk factor; indeed, the differences (HF and LF values significantly reduced) observed in our patients could predict, after more years of untreated GHD, significant alterations also from the clinical point of view (like a sympatovagal imbalance if reduction in LF became more substantial than that in HF). An important study analyzing the role of GHD in the autonomic nervous system demonstrated a reduced beta adrenoreceptor responsiveness in GHD rats despite a normal myocardial beta-adrenoreceptor density.42 The latter, together with impairment of left ventricular performance, which is known to be present in GHD patients, may identify alterations in the autonomic nervous system if GHD is not adequately treated.
Moreover, 12 months’ treatment with rGH significantly reduced the risk of a coronary event in 10 years due to improvement of modifiable risk factors (blood pressure and lipid profile), even at an early stage of GHD disease. Consistently, an improvement in blood pressure and lipidic profile after GHRT was demonstrated in other studies.16,43
The risk reduction induced by therapy, even in patients slightly compromised, stresses the importance of starting GH replacement at early stages also in patients apparently free from risk factors. One limitation of the study is the small number of patients, justified by the rarity of the disease and the strict criteria of recruitment. However, our series is perfectly adequate for the purpose of the study and devoid of confounding factors as well.
In conclusion, our data show that, in a meticulously selected group of patients, one year of rGH therapy improves cardiovascular risk by preventing the onset of atherosclerotic processes in patients with recent-onset GHD. The study also demonstrates that patients with recent GHD do not exhibit sympathovagal imbalance. However, we cannot exclude that GHD status untreated for more than two years could affect the autonomic nervous system, as previous studies have reported. Larger and longer-term case studies may help to explain the pathophysiological mechanisms that regulate the influence of GH and IGF-I on the autonomic system, particularly in patients with long-standing GHD. Indeed, it is likely that the discrepancies described in the literature by other authors are due to longer periods of GHD in their series.
ACKNOWLEDGEMENT
This paper is dedicated to Prof. Francesco Minuto, our beloved “maestro” who passed away a few months ago.
REFERENCES
1. Rosen T, Bengtsson BA, 1990 Premature mortality due to cardiovascular disease in hypopituitarism. Lancet 336: 285-288.
2. Bates AS, Van’t HW, Jones PJ, Clayton RN, 1996 The effect of hypopituitarism on life expectancy. J Clin Endocrinol Metab 81: 1169-1172.
3. Tomlinson JW, Holden N, Hills RK, et al, 2001 Association between premature mortality and hypopituitarism. West Midlands Prospective Hypopituitary Study Group. Lancet 357: 425-431.
4. Cuneo RC, Salomon F, McGauley GA, Sonksen PH, 1992 The growth hormone deficiency syndrome in adults. Clin Endocrinol Oxf 37: 387-397.
5. Jorgensen JO, Pedersen SA, Thuesen L, et al, 1989 Beneficial effects of growth hormone treatment in GH-deficient adults. Lancet 1: 1221-1225.
6. Rosen T, Eden S, Larson G, Wilhelmsen L, Bengtsson BA 1993 Cardiovascular risk factors in adult patients with growth hormone deficiency. Acta Endocrinol Copenh 129: 195-200.
7. De Boer H, Blok GJ, Voerman HJ, Phillips M, Schouten JA 1994 Serum lipid levels in growth hormone-deficient men. Metabolism 43: 199-203.
8. Cuneo RC, Salomon F, Watts GF, Hesp R, Sonksen PH, 1993 Growth hormone treatment improves serum lipids and lipoproteins in adults with growth hormone deficiency. Metabolism 42: 1519-1523.
9. Boschetti M, Goglia U, Teti C, et al, 2008 Replacement therapy and cardiovascular diseases. J Endocrinol Invest 31: 85-90.
10. Salomon F, Cuneo RC, Umpleby AM, Sonksen PH, 1994 Interactions of body fat and muscle mass with substrate concentrations and fasting insulin levels in adults with growth hormone deficiency. Clin Sci Colch 87: 201-206.
11. Gomberg-Maitland M, Frishman WH, 1996 Recombinant growth hormone: a new cardiovascular drug therapy. Am Heart J 132: 1244-1262.
12. Lombardi G, Colao A, Ferone D, Marzullo P, Orio F, Longobardi S, Merola B, 1997 Effect of growth hormone on cardiac function. Horm Res 48: Suppl 4: 38-42. Review
13. Leong KS, Mann P, Wallymahmed M, MacFarlane IA, Wilding JP, 2001 The influence of growth hormone replacement on heart rate variability in adults with growth hormone deficiency. Clin Endocrinol (Oxf) 54: 819-826.
14. Sverrisdottir YB, Elam M, Herlitz H, Bengtsson BA, Johannsson G, 1998 Intense sympathetic nerve activity in adults with hypopituitarism and untreated growth hormone deficiency. J Clin Endocrinol Metab 83: 1881-1885.
15. Beshyah SA, Johnston DG, 1999 Cardiovascular disease and risk factors in adults with hypopituitarism. Clin Endocrinol (Oxf) 50: 1-15.
16. Colao A, Di Somma C, Rota F, et al, 2005 Short-term effects of growth hormone (GH) treatment or deprivation on cardiovascular risk parameters and intima-media thickness at carotid arteries in patients with severe GH deficiency. J Clin Endocrinol Metab 90: 2056-206.
17. Pfeifer M, Verhovec R, Zizek B, Prezel J, Poredos P, Clayton RN, 1999 Growth hormone (GH) treatment reverses early atherosclerotic changes in GH-deficient adults. J Clin Endocrinol Metab 84: 453-457.
18. Colao A, Di Somma C, Cuocolo A, et al, 2001 Improved cardiovascular risk factors and cardiac performance after 12 months of growth hormone (GH) replacement in young adult patients with GH deficiency. J Clin Endocrinol Metab 86: 1874-1881.
19. Amato G, Carella C, Fazio S, et al, 1993 Body composition, bone metabolism, and heart structure and function in growth hormone (GH)-deficient adults before and after GH replacement therapy at low doses. J Clin Endocrinol Metab 77: 1671-1676.
20. Ezzat S, Fear S, Gaillard RC, et al, 2002 Gender-specific responses of lean body composition and non-gender-specific cardiac function improvement after GH replacement in GH-deficient adults. J Clin Endocrinol Metab 87: 2725-2733.
21. Malliani A, Lombardi F, Pagani M, Cerutti S, 1994 Power spectral analysis of cardiovascular variability in patients at risk for sudden cardiac death. J Cardiovasc Electrophysiol 5: 274-286.
22. Vanoli E, Schwartz PJ, 1990 Sympathetic-parasympathetic interaction and sudden death. Basic Res Cardiol 85: Suppl 1: 305-321.
23. Zipes DP, Heger JJ, Prystowsky EN, 1983 Pathophysiology of arrhythmias: clinical electrophysiology. Am Heart J 106: 812-828.
24. Kjellgren O, Gomes JA, 1993 Heart rate variability and baroreflex sensitivity in myocardial infarction. Am Heart J 125: 204-215.
25. Singh N, Mironov D, Armstrong PW, Ross AM, Langer A, 1996 Heart rate variability assessment early after acute myocardial infarction. Pathophysiological and prognostic correlates. GUSTO ECG Substudy Investigators. Global Utilization of Streptokinase and TPA for Occluded Arteries. Circulation 93: 1388-1395.
26. Lombardi F, Mortara A, 1998 Heart rate variability and cardiac failure. Heart 80: 213-214.
27. Osterhues HH, Kochs M, Hombach V, 1998 Time-dependent changes of heart rate variability after percutaneous transluminal angioplasty. Am Heart J 135: 755-761.
28. Spallone V, Uccioli L, Menzinger G, 1995 Diabetic autonomic neuropathy. Diabetes Metab Rev 11: 227-257.
29. Spallone V, Menzinger G, 1997 Autonomic neuropathy: clinical and instrumental findings. Clin Neurosci 4: 346-358.
30. Kleiger RE, Stein PK, Bigger JTJr, 2005 Heart rate variability: measurement and clinical utility. Ann Noninvasive Electrocardiol 10: 88-101.
31. Andersson IJ, Barlind A, Nystrom HC, et al, 2004 Reduced sympathetic responsiveness as well as plasma and tissue noradrenaline concentration in growth hormone transgenic mice. Acta Physiol Scand 182: 369-378.
32. Resmini E, Casu M, Patrone V, et al, 2006 Sympathovagal imbalance in acromegalic patients. J Clin Endocrinol Metab 91: 115-120.
33. Leong KS, Mann P, Wallymahmed M, MacFarlane IA, Wilding JP, 2000 Abnormal heart rate variability in adults with growth hormone deficiency. J Clin Endocrinol Metab 85: 628-633.
34. Scott EM, Greenwood JP, Stoker JB, Mary DA, Gilbey SG 2002 Sympathetic nerve hyperactivity is associated with increased peripheral vascular resistance in hypopituitary patients with growth hormone deficiency. Clin Endocrinol 56: 759-763.
35. Sverrisdóttir YB, Elam M, Caidahl K, Söderling AS, Herlitz H, Johannsson G, 2003 The effect of growth hormone (GH) replacement therapy on sympathetic nerve hyperactivity in hypopituitary adults: a double-blind, placebo-controlled, crossover, short-term trial followed by long-term open GH replacement in hypopituitary adults. J Hypertens 21: 1905-1914.
36. Barron HV, Lesh MD, 1996 Autonomic nervous system and sudden cardiac death. J Am Coll Cardiol 27: 1053-1060.
37. Tanriverdi F, Eryol NK, Atmaca H, et al, 2005 The effects of 12 months of growth hormone replacement therapy on cardiac autonomic tone in adults with growth hormone deficiency. Clin Endocrinol 62: 706-712.
38. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996 Heart rate variability: standards of measurement, physiological interpretation and clinical use. Circulation 93: 1043-1065.
39. Task Force of the European society of Cardiology and The North American Society of Pacing and Electrophysiology, 1996 Heart rate variability. Guidelines. Standards of measurements, physiological interpretation and clinical use. European Heart Journal 17: 354-381.
40. Menotti A, Lanti M, Agabiti-Rosei E, et al, 2005 Riskard 2005: new tools for prediction of cardiovascular disease risk derived from Italian population studies. Nutr Metab Cardiovasc Dis 15: 426-440.
41. Proceedings of the IV National Conference of cardiovascular prevention. G Ital Cardiol (Rome) 15: 253-263.
42. Cittadini A, Strömer H, Vatner DE, et al, 1997 Consequences of growth hormone deficiency on cardiac structure, function, and beta-adrenergic pathway: studies in mutant dwarf rats. Endocrinology 138: 5161-5169.
43. Abdu TA, Elhadd TA, Buch H, Barton D, Neary R, Clayton RN, 2004 Recombinant GH replacement in hypopituitary adults improves endothelial cell function and reduces calculated absolute and relative coronary risk. Clin Endocrinol 61: 387-393.
Address for correspondence:
Mara Boschetti, MD, PhD, Endocrinology, Department of Internal Medicine and Medical Specialties, University of Genoa, Viale Benedetto XV 6, 16132 Genoa, Italy, Tel.: +39 010 353 7954, Fax: +39 010 353 8977, E-mail: mara.boschetti@unige.
Received 05-03-2014, Accepted 05-06-2014