Does partial surgical tumour removal influence the response to octreotide‐LAR in acromegalic patients previously resistant to the somatostatin analogue?

Objective To compare the intrapatient response to the same dose of slow‐release octreotide (OCT‐LAR) before and after noncurative surgery in acromegalic patients who did not attain disease control after primary treatment with OCT‐LAR. Design Prospective clinical study. Patients Eleven acromegalic patients (eight men, aged 42·45 ± 11·15 years, 10 macroadenomas) received OCT‐LAR (20 mg, n = 1; 30 mg, n = 10) every 28 days as the primary treatment (1stOCT‐LAR) for 11·3 ± 4·2 months, without IGF‐I normalization. They were subsequently submitted to surgery without cure and were then treated with the same dose of OCT‐LAR for 8·0 ± 6·5 months (2ndOCT‐LAR). Measurements GH and IGF‐I serum concentrations were obtained under basal conditions as well as during treatment. Pituitary tumour volume was assessed by magnetic resonance imaging (MRI) of the sella. IGF‐I was also expressed as a percentage of the upper limit of the normal age‐ and sex‐matched range (%ULNR IGF‐I). Results After 1stOCT‐LAR, there was a decrease in GH levels (P = 0·003) and %ULNR IGF‐I (P = 0·009) compared to baseline (B), but no IGF‐I normalization. Tumour shrinkage was observed in eight of 10 patients with macroadenomas (median 63·7%, range 24·5–75·5%). After surgery, mean levels of GH and %ULNR IGF‐I were lower than those at baseline (P = 0·0004 and P = 0·003, respectively), but not when compared to values during 1stOCT‐LAR (P = 1·000 and P = 0·957, respectively). MRI confirmed surgical tumour removal (median 64%, range 4·9–96·6%) in eight of the 10 patients. Comparing the 2ndOCT‐LAR results with postsurgical results, there were no significant decrease in %ULNR IGF‐I (P = 0·061) and GH levels (P = 0·414). Nine patients (82%) achieved IGF‐I normalization. The degree of surgical tumour reduction did not correlate with IGF‐I normalization (P = 0·794). When comparing the results between 1stOCT‐LAR and 2ndOCT‐LAR, there was a decrease, albeit not statistically significant, in serum GH levels (P = 0·059) and a significant decrease in %ULNR IGF‐I (P = 0·011). Conclusions Using strict criteria (same patient, same drug, same dose) our results strongly suggest that the surgical reduction of tumour mass can improve the outcome of OCT‐LAR treatment in acromegalic patients resistant to primary therapy with SA.     

Influence of ethnicity on IGF-I and procollagen III peptide (P-III-P) in elite athletes and its effect on the ability to detect GH abuse

Context A method based on the two GH dependent markers, IGF‐I and procollagen III peptide (P‐III‐P) has been proposed to detect exogenously administered GH. As previous studies involved predominantly white European elite athletes, it is necessary to validate the method in other ethnic groups. Objective To examine serum IGF‐I and P‐III‐P in elite athletes of different ethnicities within 2 h of competing at national or international events. Design Cross‐sectional observational study. Setting National and International sporting events. Subjects 1085 elite athletes of different ethnicities. Intervention Serum IGF‐I and P‐III‐P were measured and GH‐2000 discriminant function score was calculated. Effect of ethnicity was assessed. Results In men, IGF‐I was 21·7 ± 2·6% lower in Afro‐Caribbeans than white Europeans (P < 0·0001) but there were no differences between other ethnic groups. In women, IGF‐I was 14·2 ± 5·1% lower in Afro‐Caribbeans (P = 0·005) and 15·6 ± 7·0% higher in Orientals (P = 0·02) compared with white Europeans. P‐III‐P was 15·2 ± 3·5%, 26·6 ± 6·6% and 19·3 ± 5·8% lower in Afro‐Caribbean (P < 0·0001), Indo‐Asian (P < 0·0001) and Oriental men (P = 0·001), respectively, compared with white European men. In women, P‐III‐P was 15·7 ± 4·7% lower in Afro‐Caribbeans compared to white Europeans (P =0·0009) but there were no differences between other ethnicities. Despite these differences, most observations were below the upper 99% prediction limits derived from white European athletes. All GH‐2000 scores lay below the cut‐off limit proposed for doping. Conclusions The GH‐2000 detection method based on IGF‐I and P‐III‐P would be valid in all ethnic groups.     

Concentrations of the acute phase reactants high-sensitive C-reactive protein and YKL-40 and of interleukin-6 before and after treatment in patients with acromegaly and growth hormone deficiency

Background Acromegaly is accompanied by increased cardiovascular mortality and a cluster of proatherogenic risk factors. In the general population, ischaemic heart disease (IHD) is associated with elevated levels of inflammatory markers. The acute phase reactant (APR) C‐reactive protein (CRP) has been reported to be reduced in acromegaly and increase after treatment, suggesting that excess of GH/IGF‐I could have anti‐inflammatory effects. This is in accordance with results obtained in patients with growth hormone deficiency (GHD), where increased levels of CRP have been reported. Objective To investigate the hypothesis that the GH/IGF‐I system is a suppressive regulator of inflammatory processes. Subjects and methods Twenty‐one acromegalic patients and 19 GH‐deficient patients were studied. The two APRs CRP and YKL‐40 and the proinflammatory cytokine interleukin‐6 (IL‐6) were measured before and after treatment and in healthy matched controls. Results In acromegalic patients, serum concentrations of high‐sensitive CRP (hsCRP) and YKL‐40 were reduced compared to controls (P < 0·001) and increased (P < 0·001) after treatment, together with IL‐6 (P = 0·021), to levels comparable with controls. Pretreatment serum YKL‐40 and IL‐6 showed a significant inverse correlation with IGF‐I and GH. In GH‐deficient patients, hsCRP and YKL‐40 were elevated compared to controls (P = 0·001 and P = 0·048). During treatment, levels of both APRs showed a trend towards a decrease (P = 0·087 and P = 0·060), and after treatment, levels of YKL‐40 no longer differed from that of controls. Serum IL‐6 was not different from controls and did not change during GH treatment. Conclusion The results point to the possibility of a relationship between GH disturbances and inflammatory processes.     

GH sensitivity of GH-deficient adults is dependent on gender but not timing of onset

Background Females secrete 2–3‐fold greater amounts of GH compared with males despite maintaining similar IGF‐I levels. IGF‐I generation tests in healthy subjects suggest this discordancy results from relative resistance to GH in females. In GHD females the presumed relative insensitivity to GH is reflected by a lower basal IGF‐I and the need for higher GH maintenance doses during replacement. Adults with severe GHD of childhood‐onset (CO) have lower basal IGF‐I SDS and require higher GH maintenance doses compared with adult‐onset (AO) patients with GHD of equal severity. We hypothesised CO‐GHD adults to be less sensitive to GH than AO‐GHD patients. Methodology In a single site study we analysed the incremental change in IGF‐I (ΔIGF‐I) in 116 GHD adults following initiation of GH replacement. The data were corrected to provide ΔIGF‐I/mg GH because of slight variances in initial GH dose. Results Following GH replacement ΔIGF‐I was 230 ± 245 and 356 ± 278 ng/ml/mg GH in females and males, respectively (P = 0·01). In CO and AO patients ΔIGF‐I was 282 ± 206 and 294 ± 292 ng/ml/mg GH, respectively (P = 0·83). Further analysis after stratification by both gender and timing of onset of GHD showed ΔIGF‐I was 226 ± 164, 324 ± 228, 231 ± 268, and 373 ± 304 ng/ml/mg GH in the CO females, CO males, AO females, and AO males, respectively (AO males vs. AO females, P = 0·03; CO males vs. CO females, P = 0·17; AO males vs. CO males, P > 0·05; AO females vs. CO females, P > 0·05). Multiple linear regression with ΔIGF‐I as the dependent variable and age, gender, BMI, baseline IGF‐I level, and timing of onset as independent variables showed ΔIGF‐I to be dependent on gender alone (R = 0·28, P = 0·004). Age (P = 0·44), BMI (P = 0·54), baseline IGF‐I level (P = 0·63) and timing of onset (P = 0·61) had no effect on ΔIGF‐I. Conclusion We have shown gender to have a significant impact on GH sensitivity in GHD adults, which, at least in part, explains differences in maintenance dosages during replacement. None of the additional variables impacted significantly on GH sensitivity. The lower basal IGF‐I SDS and higher GH replacement requirement reported in CO compared with AO patients cannot be explained by differences in sensitivity to GH.     

Use of a GH receptor antagonist (GHRA) to explore the relationship between GH and IGF-I in adults with severe GH deficiency (GHD)

Objective At diagnosis, approximately 50% of adults with severe GH deficiency (GHD) have an IGF‐I within the reference range. It is unclear whether in such patients serum IGF‐I levels are regulated by factors other than GH. Design and patients We performed a double‐blind, randomized, placebo‐controlled, cross‐over study to investigate the effect of the GH receptor antagonist – pegvisomant (20 mg daily for 14 days) on GH and IGF‐I levels in three cohorts: patients with GHD and a normal IGF‐I (NORMS); patients with GHD and a low IGF‐I (LOWS) and healthy volunteers (CONS). Results Pegvisomant decreased IGF‐I in CONS and NORMS [158·5 (101–206) vs. 103 (77–125) µg/l, P < 0·01; 124 (81–136) vs. 95 (51–113) µg/l, P < 0·01 respectively], but not in LOWS [31 (< 31–32) vs. 34·5 (< 31–38) µg/l], and this was associated with an increase in mean 24 h GH in CONS [0·49 (0·12–0·89) to 1·38 (0·22–2·45) µg/l (P = 0·03)] and in NORMS [69 (0–320)% from 0·1 (< 0·1–0·13) to 0·17 (0·11–0·42) µg/l (P = 0·03)], but not in the LOWS. The peak GH response to arginine was increased by pegvisomant in CONS and NORMS [6·1 (0·8–9) vs. 20·4 (13·1–28·8) µg/l, P = 0·03; 0·4 (0·1–0·5) vs. 0·5 (0·3–0·6) µg/l, respectively], but not in LOWS. Conclusions These data indicate that patients with severe GHD with a normal IGF‐I are able to increase GH secretion in response to a pegvisomant‐induced fall in IGF‐I, whereas those with low IGF‐I levels are unable to increase GH secretion. Therefore circulating IGF‐I appears to be GH‐independent in GHD patients with a low IGF‐I, but remains partially GH‐dependant in GHD patients with a normal IGF‐I.     

Randomized controlled GH trial: effects on anthropometry, body composition and body proportions in a large group of children with Prader-Willi syndrome

Background Prader–Willi syndrome (PWS) children have impaired growth, and abnormal body composition. Previous 1‐year controlled studies showed improvement of height and body composition during GH‐treatment. Objective To evaluate growth, body composition and body proportions during GH‐treatment in a large group of PWS children. Design/patients We performed a randomized controlled GH trial in 91 prepubertal PWS children (42 infants, 49 children, aged 3–14 years). After stratification for age, infants were randomized to GH‐treatment (GH‐group; 1 mg/m²/day; n = 20), or no treatment (control group; n = 22) for 1 year. In the second year all infants were treated with GH. After stratification for BMI, children > 3 years of age were randomized to GH‐treatment (GH‐group; 1 mg/m²/day; n = 27) or no treatment (control group; n = 22) for 2 years. Anthropometric parameters were assessed once in every 3 months. Body composition was measured by Dual Energy X‐ray Absorptiometry. Results Median (interquartile range, iqr) height SDS increased during 2 years of GH in infants from –2·3 (–2·8 to –0·7) to –0·4 (–1·1–0·0) and in prepubertal children from –2·0 (–3·1 to –1·7) to –0·6 (–1·1 to –0·1). In non‐GH‐treated children height SDS did not increase. Head circumference completely normalized during 1 and 2 years of GH in infants and children, respectively. Body fat percentage and body proportions improved in GH‐treated children, but did not completely normalize. Lean body mass SDS improved compared to the control group. Serum IGF‐I increased to levels above the normal range in most GH‐treated children. Conclusions Our randomized study shows that GH‐treatment in PWS children significantly improves height, BMI, head circumference, body composition and body proportions. PWS children are highly sensitive to GH, suggesting that monitoring of serum IGF‐I is indicated.     

Changes and relationships of IGFS and IGFBPS and cytokines in coeliac disease at diagnosis and on gluten-free diet

Objective To evaluate changes and relationships of IGFs and IGFBPs, serum interleukin 6 (IL‐6) and tumour necrosis factor (TNF)‐α, and auxological parameters at diagnosis of coeliac disease (CD) and at 6 months and 12 months after starting a gluten‐free diet (GFD), compared with a control population. Patients Twenty patients were enrolled at diagnosis (9 male, 11 female; age 9·6 ± 0·8 years). A healthy population of 18 subjects (5 male, 13 female; age 11·3 ± 0·6 years) comparable for age, sex and pubertal status served as controls at baseline. Measurements Blood samples were taken at diagnosis, and at 6 months and 12 months after starting the GFD. Serum IGF‐I, IGF‐II, IGFBP‐1, IGFBP‐2, IGFBP‐3, IL‐6 and TNF‐α were measured using commercial kits. Height (Ht) standard deviation score (SDS), body mass index (BMI) SDS and Ht velocity SDS were evaluated at diagnosis and at 6 months and 12 months after starting GFD. Results In CD patients, both Ht SDS and BMI SDS increased during the first year of treatment, and Ht velocity SDS increased during the second 6 months of follow‐up (P < 0·05). At diagnosis, IGF‐I, IGF‐II and IGFBP‐3 were lower compared with controls, IGFBP‐1 was similar, IGFBP‐2, IL‐6 and TNF‐α were higher (P < 0·05). When on GFD, all peptides normalized and IGFBP‐1 decreased. The IGF‐I/IGFBP‐2 and IGF‐I/IGFBP‐3 molar ratios were significantly reduced at diagnosis compared with those of controls, but were increased for both groups when on GFD. Although there was no apparent abnormality at diagnosis, the IGF‐II/IGFBP‐2 molar ratio increased significantly on GFD. Ht velocity SDS was positively correlated with IGFBP‐3 (P < 0·05) and with the IGF‐I/IGFBP‐2 molar ratio (P < 0·05). Serum IL‐6 was negatively correlated with IGF‐I and positively with IGFBP‐1 (P < 0·05). Conclusions The data obtained from this study confirm changes in the IGF and cytokine systems at diagnosis of CD which tend to normalize on the gluten‐free diet. The two systems show relationships with each other and with linear growth.     

GH replacement reduces increased lipid peroxidation in GH-deficient adults

Background GH replacement improves numerous metabolic abnormalities in GH‐deficient patients; increased lipid peroxidation (LPO) has been observed in GH‐deficient patients; however, it is unknown if LPO is influenced by GH replacement. Aim and methods To evaluate the extent to which GH replacement might reverse the increased LPO in GH‐deficient adults and to analyse if this phenomenon might be involved in the improvement of metabolic disturbances due to GH treatment. Serum concentrations of malondialdehyde + 4‐hydroxyalkenals (MDA + 4‐HDA), as an index of LPO, were measured at baseline, and after 12 and 24 months of GH replacement in 40 adult patients with severe GH deficiency (both in adult‐ and childhood‐onset) and in 40 healthy volunteers, matched for sex, age and body mass index (BMI). Correlations were evaluated between LPO and lipids, IGF‐I, metalloproteinase‐2 and ‐9 (MMP‐2, ‐9), vascular endothelial growth factor (VEGF), BMI and GH dose. Results LPO values in GH‐deficient patients were several‐fold higher than in controls [55·36 ± 2·27 vs. 4·19 ± 0·42 nmol/mg protein (mean ± SEM), P < 0·0001] and decreased significantly over time with GH replacement to 38·61 ± 2·15 nmol/mg protein (i.e. by approximately 30%), though still remaining markedly elevated compared with controls (P < 0·0001). The proatherogenic lipid profile parameters correlated positively with LPO in the childhood‐onset subgroup before GH replacement. GH replacement restored the positive correlation between LPO and age in male patients (r = 0·57, P = 0·013; r = 0·8, P < 0·001, at 12 and 24 months of GH replacement, respectively). Conclusions GH replacement partially reverses the grossly abnormal LPO in GH‐deficient adults. It is highly probable, therefore, that oxidative mechanisms are involved in the overall improvement of metabolic changes due to GH replacement.     

Relationship between ghrelin and the metabolic syndrome in the elderly: a longitudinal population-based study

Context Ghrelin regulates energy homeostasis and may contribute to the development of the metabolic syndrome (MS) in the elderly. Objective To study the relationship between ghrelin and the MS, IGF‐I and life style factors over a 2‐year follow‐up. Design Longitudinal population‐based study, starting from 2002; 2 years follow‐up. Participants Three hundred and thirteen (153 men/160 women) individuals living independently older than 70 years. Results MS was found in 54·9% of men and 61% of women. In the 229 subjects available at follow‐up, ghrelin was higher in men than in women at basal (P = 0·002) and 2‐year follow‐up (P = 0·004). Ghrelin decreased over time in both genders (P < 0·01). Ghrelin was lower in individuals showing MS compared to non‐MS (P = 0·08), but this difference was more evident at 2‐year follow‐up (P = 0·016), mostly due to men with MS (P = 0·002) and even after adjustment for BMI, gender and age. Individuals with MS had an OR of 1·67 (95% CI: 1·0–2·78) for low ghrelin (< first tertile); when adjusting by BMI, gender and age, only high triglycerides with OR 1·8 (1·0–3·3), remained statistically significant among the MS components. IGF‐I showed a positive correlation with ghrelin only in individuals without MS (r_s 0·403, P < 0·001) with no gender differences; this relationship was not found in MS (r_s 0·120, P = 0·129). A positive association of ghrelin was found with academic level, alcohol consumption and smoking. Conclusions Ghrelin is higher in old men in comparison to women and decreases over time with a steeper decline in subjects with MS; moreover, in these subjects ghrelin/IGF‐I correlation is lost.     

Monitoring disease activity using GH and IGF-I in the follow-up of 501 patients with acromegaly

Context The aims of treatment in patients with acromegaly are to achieve serum GH/IGF‐I concentrations associated with cure or normalization of mortality and alleviation of symptoms. Objective and methods Using the West Midlands Acromegaly database (n = 501) we investigated the reliability of basal fasting GH in predicting nadir or mean GH during oral glucose tolerance test (OGTT) or GH day curve (GHDC), respectively, the degree of discordance between disease activity measured by GH and IGF‐I values and the effect of radiotherapy on the above relationships. In total 773 OGTT and 507 GHDC were performed. Results Basal fasting GH was strongly correlated with nadir/mean GH on OGTT/GHDC (r = +0·87, P < 0·0001, r = +0·93, P < 0·0001, respectively). A basal GH < 2·5 µg/l was associated with a nadir/mean GH during OGTT/GHDC < 2·5 µg/l in 98·6% and 88·2% of cases, respectively. Elevated IGF‐I was seen in 32·4% and 46·4% of patients with GH nadir values during OGTT < 1 and < 2·5 µg/l, respectively, and in 21·2% and 45·9% of GHDC with mean GH < 1 and < 2·5 µg/l, respectively. Radiotherapy increased the discordance in GH and IGF‐I as markers of disease activity at GH < 2·5 µg/l (elevated IGF‐I‐values when OGTT nadir GH < 2·5 µg/l: radiotherapy 55·5%vs. no radiotherapy 36·9%, P = 0·002). Conclusions There is a close relationship between a basal fasting GH < 2·5 µg/l and nadir/mean GH < 2·5 µg/l during OGTT/GHDC. There is a large discordance between disease activity when assessed by GH and IGF‐I which is further increased by radiotherapy. These observations illustrate the challenge of defining appropriate biochemical end‐points to achieve control of disease and normalization of mortality in acromegaly.     

Bone demineralization in adult thalassaemic patients: contribution of GH and IGF-I at different skeletal sites

Background and objective GH and IGF‐I exert an important role in the control of bone formation, as shown by decreased bone mineral density and increased fracture risk in adult hypopituitary patients untreated for GH deficiency (GHD). Different degrees of bone demineralization are frequently reported in patients affected by β‐thalassaemia. Considering the high prevalence of GHD recently observed by our group among adult thalassaemic patients, we elected to study the possible role of GH–IGF‐I abnormalities in the pathogenesis of the osteopenia/osteoporosis of this disease. Design Sixty‐four adult thalassaemic patients (49 with thalassaemia major and 15 with thalassaemia intermedia, 23 men and 41 women, aged 31·4 ± 6·8 years) were studied. Methods Bone mineral density was assessed by dual energy X‐ray absorptiometry at lumbar spine in 62 patients and at proximal femur in 58. All patients underwent GHRH (1 µg/kg as an i.v. bolus) plus arginine (0·5 g/kg as a 30‐min i.v. infusion) testing. Severe GHD was defined by GH peaks < 9 µg/l, whereas partial GHD was defined by GH peaks ranging from 9 to 16·5 µg/l. Blood samples for IGF‐I measurement were collected. Results Lumbar osteoporosis and osteopenia were demonstrated in 46/62 (74·1%) and 14/62 (22·5%) patients, respectively. Femoral osteoporosis and osteopenia were documented in 22/58 (37·9%) and 32/58 (55·1%) patients, respectively. Severe GHD was demonstrated in 16/64 patients (25%), while 11 additional patients (17·1%) displayed partial GHD. IGF‐I standard deviation score (SDS) was low, that is, below –1·88, in the majority (54·6%) of patients. Lumbar T‐score values were not correlated with either GH peaks or IGF‐I SDS values. Femoral T‐score values were positively correlated with GH peaks (r = 0·38, P < 0·005) and IGF‐I SDS values (r = 0·39, P < 0·005). Multiple regression analysis pointed to both GH peak and IGF‐I SDS as predictors of femoral T‐score. Furthermore, mean femoral T‐score was significantly lower in patients with severe GHD than in those with normal GH secretion (–2·94 ± 0·25 vs.–2·15 ± 0·12, P < 0·01). Conclusion This study, while confirming the high prevalence of both osteopenia/osteoporosis and somatotropin–somatomedin deficiency in adult thalassaemic patients, indicates that defective GH secretion and diminished serum IGF‐I levels may contribute to femoral demineralization in these patients. Further studies are worth carrying out to evaluate the efficacy of biosynthetic GH administration on bone abnormalities of GH‐deficient thalassaemic adults.     

Insulin resistance and body composition in preterm born children during prepubertal ages

Background Premature born children may show insulin resistance in childhood which may be due to intrauterine or postnatal adverse environmental factors. Objective Aim of this study was to evaluate insulin resistance and body composition in preterm born children born appropriate for gestational age (AGA) or small for gestational age (SGA) and relations with IGF‐I, IGFBP‐3 axis. Methods Ninety‐three preterm born children grouped as premature SGA (n = 30) and premature AGA (n = 63) were evaluated at age 4·6 ± 0·2 years and 4·7 ± 0·1 years with respect to their glucose, insulin, IGF‐I, IGFBP‐3, IGFBP‐1, leptin levels and body composition by dual‐energy X‐ray absorptiometry. Their data were compared to that of body mass index (BMI) matched term SGA (n = 42) and term AGA (n = 44) children of age 4·5 ± 0·2 and 3·8 ± 0·1 years. All children had height appropriate for their target height. Insulin resistance was evaluated by basal insulin and homeostasis model assessment for insulin resistance (HOMA‐IR). Results Basal insulin level was similar in preterm AGA (4·3 ± 1·4 pmol/l) and term AGA (7·9 ± 6·4 pmol/l) children at similar and normal BMI levels. Preterm SGA children had insulin levels (5·0 ± 3·6 pmol/l) similar to preterm AGA children but significantly lower than that in term SGA children (23·7 ± 20·8 pmol/l) (P = 0·001). Similar results were obtained for HOMA‐IR. Term SGA children had also significantly lower IGFBP‐1 levels. Body composition, leptin and IGFBP‐3 did not differ between the respective groups. IGF‐I was lower in preterm AGA (5·0 ± 0·6 nmol/l) than in term AGA (8·3 ± 1·2 nmol/l) (P < 0·001) children. Conclusions Premature born AGA and SGA children do not have insulin resistance when compared to term children if they have made a catch‐up growth appropriate for their target height and have normal BMI. The similar insulin levels in preterm SGA and preterm AGA children together with increased insulin levels in term SGA children points to the fact that it is the intrauterine restriction in the third trimester that has an adverse effect on future adverse metabolic outcome.     

Insulin-like growth factor-1 deficiency determines increased intima-media thickness at common carotid arteries in adult patients with growth hormone deficiency

objective To investigate the role of IGF‐1 on intima–media thickness (IMT) at common carotid arteries by Doppler ultrasonography. subjects Thirty‐nine patients (17 women, 22 men, aged 25–70 years) with severe GH deficiency (GHD), 19 with normal and 20 with low IGF‐1 levels, and 39 sex‐, age‐ and body mass index (BMI)‐matched healthy controls. results Patients with GHD showed abnormalities in lipid profile, and increased fibrinogen levels, mean IMT (0·88 ± 0·26 vs. 0·69 ± 0·14 mm, P < 0·001), and systolic and diastolic peak velocity (P < 0·001) compared to controls. Eight patients (18%) and one control (2·1%, P = 0·04) had well‐defined plaques. In controls, but not in patients with GHD, mean carotid IMT was correlated with age (r = 0·78, P < 0·001). In both controls (r = ?0·82; P < 0·0001) and patients with GHD (r = ?0·84, P < 0·0001), serum IGF‐1 levels were inversely correlated with mean IMT at common carotid arteries. At the stepwise multiple regression, the variables most significantly related to IMT in GH‐deficient patients were total cholesterol levels (t = 5·2, P < 0·001), followed by disease duration (t = 2·4, P = 0·02), while in controls the variables most significantly related to IMT were IGF‐1 levels (t = ?9·9, P < 0·001), followed by low density lipoprotein (LDL)‐cholesterol levels (t = ?2·3, P = 0·02). Compared to patients with normal IGF‐1 levels, those with low IGF‐1 levels had lower high density lipoprotein (HDL)‐cholesterol levels (1·0 ± 0·2 vs. 1·3 ± 0·2 mmol/l, P = 0·0002), and higher glucose (54·3 ± 6·1 vs. 48·9 ± 5·9 mmol/l, P = 0·008), insulin (25·2 ± 6·8 vs. 18·8 ± 6·0 mUl/l, P = 0·004), total cholesterol (7·1 ± 1·1 vs. 4·9 ± 0·6 mmol/l, P < 0·0001), total/HDL‐cholesterol ratio (7·2 ± 1·8 vs. 3·9 ± 0·7, P < 0·0001), fibrinogen levels (319·8 ± 56·9 vs. 241·8 ± 53·0 mg/dl, P < 0·0001) and mean IMT at common carotid arteries (1·05 ± 0·25 vs. 0·69 ± 0·07 mm, P < 0·0001). Atherosclerotic plaques were found only in GH‐deficient patients with low IGF‐1 levels. conclusions GH‐deficient patients have alterations in lipid profile with an increase in the total/HDL‐cholesterol ratio, which is an index of increased cardiovascular risk, but only patients with IGF‐1 deficiency have increased IMT.     

Glucocorticoid replacement is associated with hypertriglyceridaemia, elevated glucose and higher non-HDL cholesterol and may diminish the association of HDL cholesterol with the -629C>A CETP promoter polymorphism in GH-receiving hypopituitary patients

Objectives The effect of glucocorticoid substitution on the prevalence of metabolic syndrome components (NCEP ATP III criteria) and serum lipid levels was determined in GH‐replaced hypopituitary patients. As glucocorticoid replacement is associated with a pronounced decrease in plasma cholesteryl ester transfer protein (CETP) activity, we also tested associations of HDL cholesterol with the –629C>A CETP promoter polymorphism in subjects with and without ACTH deficiency. Design and patients In a university setting, we retrieved protocolized clinical and laboratory data from 165 adult hypopituitary patients, who had received GH for 1 year. Results After adjustment for age, sex and smoking, non‐HDL cholesterol (P = 0·05) and triglycerides (P = 0·004) were higher, but HDL cholesterol was not decreased in 117 glucocorticoid (mainly cortisone acetate in two divided doses) receiving subjects compared to 48 ACTH‐sufficient subjects. The prevalence of elevated plasma glucose and/or diabetes (P = 0·04) and hypertriglyceridaemia (P = 0·005), but not of other metabolic syndrome components, was higher in glucocorticoid‐replaced subjects. HDL cholesterol was higher in –629 A allele carriers compared to –629CC homozygotes in ACTH‐sufficient subjects (P = 0·04), but not in glucocorticoid‐treated subjects (P = 0·13). Multiple linear regression analysis demonstrated that only in ACTH‐sufficient subjects, HDL cholesterol was independently related to this CETP gene variation (P = 0·03). Conclusions In GH‐ and glucocorticoid‐replaced hypopituitary patients, serum non‐HDL cholesterol and triglycerides are higher and the prevalence of hyperglycaemia is increased, but HDL cholesterol is not decreased. Conventional glucocorticoid replacement appears to diminish the association of HDL cholesterol with a common CETP gene variation.     

The effects of growth hormone status on circulating levels of vascular growth factors

Background Vascular growth factors are important not only in angiogenesis but also for the maintenance of normal endothelial integrity and function. Elevated levels of vascular endothelial growth factor (VEGF), angiopoietin‐2, hepatocyte growth factor (HGF), endostatin and angiogenin have been associated with endothelial dysfunction and atherosclerosis. Both acromegaly and growth hormone deficiency (GHD) are associated with endothelial dysfunction and changes in blood vessel morphology. Aim To investigate the effect of GH status on the circulating levels of angiogenic factors. Design We measured the levels of six endothelial growth modulators, four angiogenic growth factors and two inhibitors of angiogenesis in 35 untreated acromegalics, 36 untreated GH‐deficient subjects and 101 normal control subjects. Fifteen GH‐deficient subjects were also studied before and 1 year after treatment with GH. Results Mean angiogenin concentrations were increased in acromegaly and decreased in GH‐deficient subjects compared to control subjects. Endostatin levels showed a similar pattern although the elevated levels in acromegalic subjects did not achieve statistical significance. Angiogenin and endostatin levels both correlated significantly with IGF‐I levels (R = 0·61, P < 0·001 and R = 0·22, P < 0·01, respectively). The relationship between angiogenin and IGF‐I levels remained significant even after correction for gender, age, body mass index (BMI) and insulin resistance. There were no significant differences in the levels of HGF, VEGF, VEGF‐C or angiopoietin‐2 between the three groups. VEGF‐D levels were elevated in both acromegalic and GH‐deficient male subjects. A similar pattern was apparent in female subjects. After GH treatment, a significant reduction in VEGF‐D levels and a significant rise in endostatin levels were observed in GH‐deficient subjects. A nonsignificant increase in angiogenin levels was also observed. Conclusion These data indicate that significant perturbations in the levels of vascular growth modulators are present in both acromegaly and GHD. While changes in endostatin and angiogenin levels appear to correlate with IGF‐I levels, VEGF‐D levels show similar perturbations in both acromegaly and GHD. Further studies are required to determine the relationship of the perturbations to endothelial dysfunction in these conditions.     

Measurement of basal growth hormone (GH) is a useful test of disease activity in treated acromegalic patients

Background Nadir GH during oral glucose tolerance test (OGTT) is the gold‐standard test of GH secretion in treated acromegaly. However, it was recently reported that variability in GH is reduced postradiotherapy, making basal GH a potential surrogate marker for nadir GH in such patients. Objective We aimed to investigate how predictive basal GH is of nadir GH and IGF‐I, and whether radiotherapy influenced these relationships. Design A total of 226 pairs of basal and nadir GH values from 76 treated acromegalic patients were analysed. Basal GH was defined as the fasting serum GH immediately prior to OGTT. Results A highly positive linear correlation (Pearson correlation = 0·955, P < 0·01) between basal and nadir GH was found. Negative predictive value for basal GH < 1 µg/l with respect to nadir GH > 1 µg/l was 100% (53/53 in radiotherapy group, 15/15 in nonradiotherapy group). Positive predictive values for basal GH > 2 µg/l with respect to nadir GH > 1 µg/l for patients treated and not treated with radiotherapy were 96·7% (88/91) and 95·2% (20/21), respectively. No significant difference between concordance of basal and nadir GH with IGF‐I in assessment of disease activity was found. Discordance between IGF‐I and nadir or basal GH < 1 µg/l was lower in the radiotherapy group than nonradiotherapy group, but this was nonsignificant. Conclusions Basal GH < 1 µg/l and > 2 µg/l are highly predictive of nadir GH < 1 µg/l and > 1 µg/l, respectively, regardless of previous radiotherapy. Basal GH is as good as nadir GH in concordance with IGF‐I. We therefore suggest basal GH is a useful test of disease activity in treated acromegaly, and can reliably replace OGTT unless basal GH is between 1 µg/l and 2 µg/l.     

The IGF system and cytokine interactions and relationships with longitudinal growth in prepubertal patients with cystic fibrosis

Objective Growth delay is a feature of patients with cystic fibrosis (CF). CF is a condition characterized by chronic inflammation that has been shown to modify the IGF system, which is essential for normal growth, and is related to pulmonary function in CF patients. We aimed to verify whether circulating levels of tumour necrosis factor (TNF)‐α, interleukin (IL)‐6, insulin and the IGF system were related and/or had relationships with linear growth in children with CF. Design and patients Seventeen prepubertal CF patients (nine males and eight females) in a stable clinical condition were enrolled. Auxological parameters, pulmonary function and the Shwachman–Kulczycki (S‐K) score were assessed, and serum samples were drawn at baseline and after 12 months. Measurements TNF‐α, IL‐6, IGF‐I, IGF‐II, IGFBP‐1, IGFBP‐2, IGFBP‐3 and insulin were assayed using specific commercial kits. Results At baseline, TNF‐α serum concentration was related to serum IGF‐I concentration (R = 0·53), IGF‐II bioactivity (IGF‐II/IGFBP‐3 molar ratio, R = +0·52) and insulin concentration (R = +0·63). Changes in serum IL‐6 and IGFBP‐2 concentrations during the 12‐month observation were positively correlated (R = +0·63). Changes in height standard deviation score (Ht SDS) were correlated with IGF‐I serum concentrations at baseline (R =+0·67) and after 12 months (R = +0·70), with IGF‐I bioavailability and with IGFBP‐1 serum concentrations (R = –0·88). Body mass index (BMI) SDS correlated with IGF bioavailability. Conclusions This study showed a relationship between inflammatory status and the IGF system, and an effect of these interactions on longitudinal growth. Moreover, a role for insulin in growth was identified. Better control of inflammation and preservation of insulin secretion could benefit these patients.     

Inflammatory markers and visceral fat are inversely associated with maximal oxygen consumption in women with polycystic ovary syndrome (PCOS)

Background We investigated whether several different inflammatory markers including C‐reactive protein (CRP) and fibrinogen and white blood cells (WBCs) count, are associated with maximal oxygen consumption (VO_2max) in women with polycystic ovary syndrome (PCOS). Methods In PCOS women (n = 124, 24·1 ± 4·5 year‐old) VO_2max was measured during symptom‐limited cardiopulmonary exercise test. Abdominal fat distribution was determined by ultrasound. Physical activity level was assessed by a standardized questionnaire. CRP was measured by immunoassays, fibrinogen by the Clauss method, and WBCs count with a Coulter counter. Results Pearson's analysis showed a significant correlation between VO_2max and logCRP (r = –0·437, P < 0·001), fibrinogen (r = –0·479, P < 0·001), and WBCs count (r = –0·438, P < 0·001). Multivariable logistic regression model showed that age (β = –0·127, P = 0·005), AUC_INS (β = –0·335, P < 0·001), HDL‐C (β = 0·390, P < 0·001), physical activity score (β = 0·238, P = 0·002), visceral fat (β =–0·184), P = 0·023), FAI (β = –0·291, P = 0·028); CRP (β = –0·216, P = 0·011), fibrinogen (β = –0·113, P = 0·008) and WBCs count (β = –0·177, P < 0·001) were significantly associated with VO_2max. Conclusions Acute‐phase reactants, such as CRP and fibrinogen, and WBCs count were independently and inversely associated with a direct measure of cardiorespiratory fitness (VO_2max) in women with PCOS, even after adjustment for physical activity level and other potential confounding factors. These findings add to the growing body of evidence linking inflammation to cardiorespiratory fitness in PCOS women.     

ORIGINAL ARTICLE: Associations of the growth hormone receptor (GHR) gene polymorphisms with adiposity and IGF‐I activity in adolescents

Objective: To explore the genetic effect of the GH receptor (GHR) on obesity and related metabolic parameters in Hong Kong Chinese adolescents. Context: Obesity is a growing global epidemic. Increasing evidence suggests that the GH‐IGF‐I axis plays an important role in regulating adiposity and insulin sensitivity. Design: We examined the associations of genetic variants of GHR with serum IGF‐I and IGFBP‐3 levels as well as obesity‐related metabolic traits in Hong Kong Chinese adolescents. Patients: Nine hundred and eighty‐one randomly selected Hong Kong Chinese adolescents from 14 schools. Measurements: We genotyped 17 single nucleotide polymorphisms (SNP) at GHR and measured serum IGF‐I and IGFBP‐3 levels as well as obesity‐related metabolic traits including fasting plasma glucose, insulin and lipid levels. Results: There were significant associations between rs4410646 and the body composition (P = 0·0044) and blood pressure factor scores (P = 0·00017). Carriers of the CC genotype had lower body mass index, percentage body fat, waist and hip circumferences than AC and AA genotype carriers (P = 0·00030–0·0094). There was also association between rs7703713 and the IGF‐I activity factor score (P = 0·0033). The GA and AA carriers of rs7703713 had higher serum IGF‐I, higher serum IGFBP‐3 and higher IGF‐I/IGFBP‐3 molar ratio (P = 0·00069–0·025). Haplotype analysis did not increase the significance of associations. Conclusion: Our results support the role of GHR gene polymorphisms in modulating adiposity and IGF‐I activity in adolescents. Examination of interactions of these SNPs with lifestyle, environmental and perinatal factors may provide further insights into their long‐term effects on obesity and metabolic risks.     

Primary hyperparathyroidism is associated with marked impairment of GH response to acylated ghrelin

Background Impaired GH secretion is a common finding in patients with primary hyperparathyroidism (PHP). Ghrelin displays strong GH‐releasing action, mainly at the hypothalamic level. Objective To evaluate secretory response of GH to ghrelin in PHP patients. Patients Fifteen patients [11 women/4 men, mean age 54 years, range 32–70 years, body mass index (BMI) 25·0 ± 0·7 kg/m²] affected with PHP due to single parathyroid adenoma and 35 normal age‐matched subjects (23 women/12 men, mean age 58 years, range 35–68 years, BMI 24·1 ± 1·1 kg/m²). Methods A measure of 1 µg/kg body weight i.v. acylated ghrelin or 1 µg/kg body weight i.v. GH releasing hormone (GHRH) followed by 0·5 g/kg body weight i.v. arginine (ARG) hydrochloride were administered to all subjects on alternate days in order to evaluate GH response. Results Mean serum GH peak after GHRH + ARG was 32·6 ± 7·8 and 17·4 ± 4·0 µg/l, in controls and PHP patients, respectively (P < 0·05). Mean serum GH peak after ghrelin was 70·4 ± 31·5 and 16·8 ± 1·9 µg/l, in controls and PHP patients, respectively, (P < 0·001). Using ROC curves, a serum GH peak > 22 µg/l after ghrelin stimulation might be considered as a cut‐off value for identifying normal subjects. Ten (67%) PHP patients have impaired GH response to GHRH + ARG and 13 (87%) to ghrelin. Serum GH peak after ghrelin or GHRH + ARG was unrelated to serum IGF‐1, PTH or ionized calcium concentrations. Conclusions The present data confirm that GH secretion is impaired in PHP patients using the potent GH secretagogue ghrelin and suggest that impaired GH secretion is likely due to a deleterious effect of hypercalcaemia at the hypothalamic level in PHP patients.