Vincenzo De Sanctis1, Ashraf T Soliman2, Giancarlo Candini3, Mohamed Yassin4, Giuseppe Raiola5, Maria Concetta Galati6, Rania Elalaily7, Heba Elsedfy8, Nicos Skordis9, Piernicola Garofalo10, Salvatore Anastasi11 , Saveria Campisi12 , Mehran Karimi13, Christos Kattamis14, Duran Canatan15, Yurdanur Kilinc16, Praveen Sobti17, Bernadette Fiscina18 and Mohamed El Kholy8
1 Pediatric and Adolescent Outpatient Clinic, Quisisana Hospital, Ferrara, Italy
2 Department of Pediatrics, Division of Endocrinology, Hamad General Hospital Doha, Qatar
3 Department of Medical Physics, St. Anna Hospital, Ferrara, Italy
4 Department of Hematology, Alamal Hospital, Hamad Medical Center, Doha, Qatar
5 Department of Paediatrics, Pugliese-Ciaccio Hospital, Catanzaro, Italy
6 Department of Haematology, Thalassaemia and Prenatal Diagnosis Regional Center, Pugliese-Ciaccio Hospital, Catanzaro, Italy
7 Department of Primary Health Care (PHC), Doha, Qatar
8 Department of Pediatrics, Ain Shams University, Cairo, Egypt
9 St. George’s University Medical School at the University of Nicosia, Cyprus
10 Endocrine Unit, Villa Sofia –Cervello Hospital, Palermo, Italy
11 Department of Thalassaemia, Garibaldi Hospital, Catania, Italy
12 Unit for the Diagnosis and Treatment of Thalassaemia, Umberto I Hospital, Siracusa, Italy
13 Hematology Research Center, Shiraz University of Medical Sciences, Shiraz, Iran
14 First Department of Paediatrics, University of Athens, Athens, Greece
15 Director of Hemoglobinopathy Diagnosis Center and President of Mediterranean Blood Diseases Foundation, Antalya, Turkey
16 Department of Pediatric Hematology, Faculty of Medicine, University of Çukurova, Adana, Turkey
17 Professor of Pediatrics, Christian Medical College and Hospital, Ludhiana, India
18 Department of Pediatrics, NYU School of Medicine, New York, USA
| This is an Open Access article distributed
under the terms of the Creative Commons Attribution License
(http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
deficiency in TM patients in children and adolescents has been attributed to
chronic anemia and hypoxia, chronic liver disease, iron overload and other
associated endocrinopathies, e.g. growth hormone deficiency (GHD). Few data are
available in the literature regarding adult TM patients and growth disorders.
The aim of this study was to measure IGF-1 values and other clinical data in a
large number of adult patients with TM to evaluate the possible relationships
IGF-1 is a small peptide (molecular weight 7649 Da), that circulates
in serum bound to high affinity binding proteins (IGFBPs), mostly in a
ternary complex with IGF-binding protein 3 (IGFBP-3) and acid-labile
subunit (ALS). This ternary complex is a storage form of IGF-1 in serum
and has a half-life of several hours. ALS is synthesized under direct
control of GH, primarily in hepatocytes. In contrast to pulsatile GH
secretion, circulating IGF-1 and IGFBP-3 are stable and reflect the
long-term status of GH secretion.[1,2]
Although the liver mainly produces IGF-1, every tissue can secrete IGF-1 for autocrine and paracrine purposes. IGF-1 possesses a large number of activities (anabolic, antioxidant, anti-inflammatory and cytoprotective actions), however, it is partly responsible for systemic GH actions. IGF-1 deficiency in TM patients in children and adolescents has been attributed to chronic anemia and hypoxia, chronic liver disease, iron overload and other associated endocrinopathies, e.g. growth hormone deficiency (GHD).[5-11]
Few data are available in the literature regarding adult TM patients. Therefore, the International Network of Clinicians for Endocrinopathies in Thalassemia and Adolescence Medicine (ICET-A), previously called the International Network on Endocrine Complications in Thalassemia (I-CET), for a better interpretation of IGF-1 values in thalassaemia made the following proposals:
(1) estimate the IGF-1 values in a large number of adult TM patients; (2) conduct the study in a single center, to minimize the considerable variation in the assay methods used; (3) compare the results with those for a reference population of healthy adults; (4) correlate the IGF-1 values of patients with TM with other clinical and laboratory parameters; (5) perform a provocative GH stimulation test with glucagon (GST) in a small group of TM subjects with normal or low IGF -1 levels (below the mean standard reference levels); (6) to review the current literature on AGHD in thalassaemia with the final goal of assisting clinicians in the management of TM patients with GH-IGF-1 axis related disorders.
Subjects and methods
From 2009 to 2013, we studied 120
consecutive adult patients with TM followed at our institution Patients
with thalassaemia intermedia, cirrhosis, cardiac and renal failure and
HIV positivity were excluded from the study.
An extensive medical history, including data on associated complications and current medications, was obtained, and a physical examination (pubertal status, weight, and height) for each patient was performed. Body mass index (BMI) was calculated as the body weight divided by the height squared (Kg/m2). A subject was considered overweight when the BMI was between 25 and 29.9 and obese when the BMI was 30 or higher.
All patients were on regular transfusions (mean haemoglobin level 11.5 g/dl) and iron chelation therapy with deferoxamine (64 patients: 30-45 mg/kg body weight, 4-6 days a week by slow subcutaneous infusion by pump, starting in 1977-1978), or oral deferiprone (22 patients: 75 mg/kg body weight daily), or deferiprone plus deferoxamine (25 patients; 75 mg/kg body weight daily and 40 mg/kg body weight, 3 days a week, by slow subcutaneous pump infusion) or oral deferasirox (9 patients: 20-30 mg/kg body weight daily).
The following clinical and laboratory data were also recorded: age at first transfusion, age at start of regular chelation therapy, duration of iron chelation therapy, compliance with treatment and the presence of associated growth and endocrine complications, as previously described.
The compliance was arbitrarily considered as good when the chelation therapy was given 5-6 days a week, moderate four days a week and poor less than three days a week.
Blood samples were drawn in the morning after an overnight fast and at least 2-3 weeks after the last blood transfusion to measure the serum concentrations of IGF-1, free thyroxine (FT4) and thyrotropin (TSH), urea, creatinine, electrolytes, glucose, calcium, phosphate and serum ferritin. As routine or to exclude severe liver pathology or decreased synthetic functions, alanine aminotransferase (ALT), gamma glutamyl transferase (γGT), alkaline phosphatase (ALP), total and direct bilirubin, albumin, prothrombin time (PT) and international normalization ratio (INR), and serologic screening assays for hepatitis C virus seropositivity (HCVab and HCV-RNA ) were also evaluated.
Iron overload was assessed at the beginning of the study, by serum ferritin level. It was arbitrarily categorized as mild, moderate or severe if the levels were <1000 ng/ml, from 1000 to 2000 ng/ml and >2000 ng/ml, respectively.
In eleven patients (3 females) the GH response, after glucagon stimulation test (GST), was evaluated. Blood samples were collected for 3 hours (at baseline, 30, 60, 90, 120, 150 and 180 minutes), because, according to the literature, the majority of GH peaks occur between 120 and 180 minutes (85%). Serum glucose, insulin and GH were measured; a GH peak below three μg/L was defined as severe GHD, according to Gomez et al.
For ethical reasons we used as a reference for normal the IGF-1 values determined in a large population of healthy subjects, reported in the literature using the same automated chemiluminescence immunoassay system (CLIA). A serum IGF-1 level below -2 SD was considered as deficiency.
Insulin resistance was also calculated in these patients using HOMA-IR (homeostasis model assessment method). Considering HOMA-IR values, the studied population was defined as a. (insulin sensitive) HOMA-IR < 2.24; b. (intermediate) 2.24 - ≤ 3.59; c. (insulin resistant) HOMA-IR > 3.59.
Biochemical, Hormonal and Cardiac Assessment: Fasting blood samples for circulating IGF-1 were collected and stored at −60°C until centrally assayed. Plasma total IGF-1 was determined on EDTA by chemiluminescent immunometric assay (CLIA) method (Nichols Institute Diagnostics, San Juan, CA). The assay was performed after separation of IGF-1 from binding proteins by Liaison® autoanalyser (DiaSorin SpA, Saluggia, Italy). The sensitivity of the test was six ng/ml, whereas the intra- and inter-assay coefficients of variation (CVs) of our in-house pooled serum control sample were 4.8% and 7.1%, respectively.
Serum GH concentrations were measured in duplicate at each time point with commercial solid-phase two-site chemiluminescent immunoassay. The inter- and intra-assay CVs were below 7%.
TSH and FT4 were assessed with an electro-chemoluminescence (ECLIA) assay with a normal range of 0.8-1.8 ng/dL for FT4 and 0.5-4.6 mIU/l for TSH. The minimal detectable levels of FT4 and TSH were 0.2 ng/dL, and 0.1 mIU/L, respectively. The inter-assay and intra-assay coefficients of variation of FT4 varied from 5.8% to 6.26%, and from 2.6% to 2.9%, respectively, and those of TSH were from 5.1% to 5.7%, and from 2.2% to 2.9%, respectively.
All biochemical and serologic tests were carried out in accordance with the routine procedures of the central laboratory. Serum ferritin was measured by electrochemiluminescence immunoassay. Reference range values were 30-350 μg/l in males and 15-150 μg/l in females.
Left ventricular ejection fraction (LVEF, %) was measured according to the recommendation of the American Society of Echocardiography, using freeze-frames from two-dimensional directed M-mode echocardiogram.
Associated endocrine complications were defined according to the I-CET position statement published in 2013.
Ethical Aspects: The study was made in accordance with the provisions of the Declaration of Helsinki. The prospective study was started at the beginning of 2009 by the Coordinator of I-CET (VDS) at the Thalassaemia Centre of Ferrara, and was completed at the end of 2013 at the Quisisana Pediatric and Adolescent Outpatient Clinic of Ferrara. All of the subjects gave their consent to participate in the study.
Statistical Analysis: Characteristics of the studied patients are reported as mean ± standard deviation (SD), median, number and range. Quantitative variables are reported as medians (range or percentiles) or mean (SD).
Statistical significance of the differences between variables was assessed using the unpaired two-tailed Student’s t test or Wilcoxon test using a software package program. The frequency distributions for age and sex were analyzed using the chi-square test while the multiple regression analysis was conducted using the multiple linear fitting with least squares method.
The distribution analysis of IGF1 values in males and females groups was performed including the following parameters: mean and SD, median, range, kurtosis, skewness, and percentiles.
Fisher's Exact test was used to calculate the probability value for the relationship between two dichotomous variables.
A p value < 0.05 was considered as significant.
A software program used for the statistical analysis was developed by Dr. Candini (Department of Medical Physics, St. Anna Hospital, Ferrara, Italy) and validated according to Alder and Roesser.
The demographic, clinical, and laboratory data of the TM study population are presented in Table 1. Rates of infection with hepatitis C are displayed in Table 2.
|Table 1. Demographic, clinical and laboratory features of a study population of 120 adults with TM and in 60 TM patients with IGF-1 level < -2SD based on reference values of healthy individuals matched for age and sex (Ref. 12)|
|Table 2. Prevalence of thalassaemia major patients infected with HCV enrolled in the study|
A statistical comparison between HCV positive and negative
TM patients was not done because only 3 TM patients were HCV negative (Table 2).
In addition, using a multivariate discriminant analysis the
classification error observed between HCV positive and negative
patients was 20% and therefore was considered not clinically acceptable.
An abnormal ALT value (>40 U/L) was observed in 51.8% of female TM patients with IGF-1 levels < -2SD and 23 % of patients with IGF-1 levels > -2SD. Similar results were observed in males (48.4% with IGF-1 levels < -2SD and 26.6% IGF-1 levels > -2SD).
Of the 120 patients, 58 (48%) were males and 62 (52%) females, with an age range of 26.0 - 53.2 (median 38) years for females and 20.8 - 51.2 years for males (median 37.05), 64.4 % of the patients were above 35 years of age. The mean body mass index (BMI) was 22.48 ± 3.34 kg/m2.
The mean value of BMI in TM female patients with IGF 1 <-2SD was 23.8 ± 4.3 (range: 18.6 - 40) and in patients with IGF 1 > -2SD was 23.2 ± 3.5 (range: 17.2 – 32.2. The mean value of BMI in TM male patients with IGF 1 <-2SD was 22.3 ± 2.7 (range: 17.4 – 28) and in patients with IGF 1 > -2SD was 22.1 ± 1.7 (range: 18.5 – 23.2).
Seven TM female patients were classified as overweight and three as obese; 5 TM male patients were classified as overweight and none as obese.
Of 63 patients with hypogonadism, 13 (37%) were on stable sex steroid therapy (Table 1). All patients with primary or central hypothyroidism or hypoparathyroidism were receiving levothyroxine or calcium and calcitriol. The type 1 diabetic patients were treated with insulin.
In males with TM, the serum concentrations of IGF-1 ranged between 18.3 and 147.7 ng/ml (mean 68.29 ± 33.5 ng/ml; median 62.5 ng/ml; kurtosis -0.5, skewness 0.57); whereas, in females with TM, IGF-1 ranged between 19.5 and 195.5 ng/ml (mean 76.46 ± 41.84 ng/ml; median 69.05 ng/ml; kurtosis 0.02; skewness 0.87).
The distribution of IGF-1 values in the two groups of TM patients compared to the percentiles of Brabant et. al are reported in Figures 1 and 2.
|Figure 1. IGF-1 as a function of age in female TM patients compared to the percentile of healthy subjects . Values are reported as mean, ±1 SD and ± 2 SD.|
|Figure 2. IGF-1 as a function of age in male TM
patients compared to the percentile of healthy subjects . Values are
reported as mean, ±1 SD and ± 2 SD.
No significant differences were observed between IGF-1
values in men and women with TM (t-test: 1.18; p: 0.249). In
multivariate regression analyses, height, weight, BMI, serum ferritin,
ALT, HCV serology and left ventricular ejection fraction (LVEF) had no
significant relationship with IGF-1 levels, but a significant multiple
correlation was found in females, with HCV-RNA positivity, between
IGF-1, ALT and serum ferritin (r= 0.504 , p= 0.043). The simple
correlation matrix was as follows: IGF1 vs ALT: r = 0.505 p < 0.05;
IGF1 vs serum ferritin: r = - 0.466 p < 0.05; ALT vs serum ferritin:
r = - 0.402 p < 0.05.
A statistical comparison of IGF 1 values between HCV positive and negative TM patients was not done because only three patients were HCV negative (Table 2). In addition, using a multivariate discriminant analysis the classification error between patients HCV positive and negative patients was 20% and therefore considered not statistically acceptable.
Analysis of individual IGF-1 levels in TM patients showed that IGF-1 levels were below -2SDs of normal values for healthy individuals in 60 (50 %) patients (33 males and 27 females).
The demographic, clinical and laboratory features of the TM patients with IGF-1 levels < -2SD below normal are reported in Table 1.
A comparison of clinical and laboratory features of male and female TM patients with IGF-1 < - 2SD and > - 2SD are reported in Tables 3 and 4.
|Table 3. Comparison of clinical and laboratory features of female TM patients with IGF-1 < - 2SD and > - 2SD of reference values of healthy individuals (Ref.12)|
|Table 4. Comparison of clinical and laboratory features of male TM patients with IGF-1 < - 2SD and > - 2SD of reference values of healthy individuals (Ref.12)|
There were significant differences in age, ALT levels, and
rates of primary hypothyroidism and insulin-dependent diabetes between
female patients with IGF-1 levels <2 SD and those with higher IGF-1
levels; significant differences only for ALT levels and rates of
primary hypothyroidism were found in males with IGF-1 <2 SD.
The left ventricular ejection fraction (LVEF) was reduced (< 50%) in 2 patients (1.6%).
AGHD was diagnosed in 6 (4 males) out of 11 patients who had glucagon stimulation tests (54.5 %). The mean age of patients with GHD was 39.3 years (range: 25-49 years) versus 35.8 years (range: 27-45 years) of non-GHD patients.
A positive correlation between GH peak after GST and IGF-1 levels was found (r =0.6409; p: < 0.05). In particular, a low GH peak after GST (range 0.28-1.3 ng/ ml) was found in 5 out of 8 patients (aged 25-49 years) with an IGF-1 level below -2SDs (range 16.7 – 52.5 ng/ml) and in 1 out of 2 patients with an IGF-1 level below -1SD (75 ng/ml) (Table 5). No correlation was observed between IGF-1 level, ALT and LVEF.
A HOMA-IR > 3.59 was found in one female TM patient with GHD (Table 5).
In general, in our patients the compliance to treatment was good or satisfactory. A serum ferritin level below 1000 ng/ml was observed in 36 (65.5%) female and 37 (74%) male TM patients. A ferritin level > 2000 ng/ml was observed in 11 (20%) female and 5 (10%) male patients. The remaining patients had a serum ferritin level between 1000 to 2000 ng/ml [11 females (20%) and 8 (16%) males]. However, we did not know with absolute certainty if the patients in the past had a severe iron overload.
|Table 5. Growth hormone (GH) peak (μg/L) after glucagon stimulation test (GST) given i.m., insulin-growth factor-1 (IGF-1), alanine aminotransferase (ALT) and left ventricular ejection fraction (LVEF) in 11 adult thalassaemia major patients.|