thalassemia, biology

genetics-project.pdf

Home >Biology homework help >thalassemia, biology

Differences in the prevalence of growth, endocrine and vitamin D abnormalities among the various thalassaemia syndromes in North America

Maria G. Vogiatzi,1 Eric A. Macklin,2

Felicia L. Trachtenberg,3 Ellen B. Fung,4

Angela M. Cheung,5 Elliott Vichinsky,4

Nancy Olivieri,5 Melody Kirby,6 Janet L.

Kwiatkowski,7 Melody Cunningham,8

Ingrid A. Holm,8 Martin Fleisher,9

Robert W. Grady 1 , Charles M. Peterson

and Patricia J. Giardina 1

for the

Thalassemia Clinical Research Network* 1Department of Pediatrics, Weill Medical College

of Cornell, New York, NY, 2Massachusetts General

Hospital Biostatistics Center, Massachusetts

General Hospital and Department of Medicine,

Harvard Medical School, Boston, MA, 3 New

England Research Institutes, Inc., Watertown,

MA, 4 Children’s Hospital Oakland, Oakland, CA,

USA, 5Department of Medicine, University Health

Network, Toronto General Hospital, Toronto, ON, 6Toronto Hospital for Sick Children, Toronto, ON,

Canada, 7Division of Hematology, Children’s

Hospital of Philadelphia and Department of

Pediatrics, University of Pennsylvania School of

Medicine, Philadelphia, PA, 8 Division of

Hematology & Oncology, Children’s Hospital

Boston, Boston, MA, 9Memorial Sloan-Kettering

Cancer Cancer, New York, NY, and 10National

Heart, Lung, and Blood Institute, NIH, Bethesda,

MD, USA

Received 13 March 2009; accepted for

publication 7 May 2009

Correspondence: Maria G. Vogiatzi, Pediatric

Endocrinology, New York Presbyterian

Hospital/Weill Medical College of Cornell

University, 525 E 68th Street, Box 103, New

York, NY 10065, USA.

E-mail: [email protected]

*This is publication number seven of the

Thalassemia Clinical Research Network

(TCRN). A list of TCRN member institutions

and staff appears in Appendix 1.

Summary

This study aimed to determine differences in the rates of growth, endocrine-

and calcium-related abnormalities in the various thalassemia syndromes in

North America treated with current therapies. Medical history, physical

examinations and blood and urine collections were obtained from patients

with all thalassemia syndromes age 6 years and older in the Thalassemia

Clinical Research Network. 361 subjects, 49% male, mean age 23Æ2 years (range 6Æ1–75 years) were studied. Approximately 25% of children and adults, regardless of the thalassemia syndrome, had short stature. Overall

growth in children was mildly affected. Final height was close to midparental

height (z = )0Æ73 ± 1Æ24). Patients with beta thalassemia major (TM) had higher rates of hypogonadism, multiple endocrinopathies, worse

hyperglycaemia, subclinical hypoparathyroidism and hypercalciuria.

Hypogonadism remained the most frequent endocrinopathy and was

frequently under-treated. 12Æ8% of the subjects had 25 vitamin D concentrations less than 27 nmol/l and 82% less than 75 nmol/l, regardless

of the thalassemia syndrome. Adolescents had lower 25 vitamin D levels than

children and adults. Compared to patients with other thalassemia syndromes,

those with beta TM suffered from higher rates of multiple endocrinopathies,

abnormal calcium metabolism and hypercalciuria. Vitamin D abnormalities

were high among adolescents.

Keywords: thalassemia, short stature, hypogonadism, endocrinopathies,

vitamin D, hypercalciuria.

research paper

First published online 13 July 2009 doi:10.1111/j.1365-2141.2009.07793.x ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 146, 546–556

Poor growth and multiple endocrinopathies, including

hypogonadotropic hypogonadism, growth hormone deficiency

and diabetes, are known complications in beta thalassemia

major (TM), and are considered the result of iron overload

(De Sanctis et al, 1994; Jensen et al, 1997; Roth et al, 1997).

Iron chelation therapy with deferoxamine (DFO), introduced

in the mid 1970’s, combined with a transfusion regimen that

maintains near normal pre-transfusion haemoglobin concen-

trations, initiated in the 1980’s, has dramatically changed the

course of beta TM and prolonged survival (Brittenham et al,

1994; Calleja et al, 1998; Cunningham et al, 2004). Even with

iron chelation, the rate of endocrinopathies remains high

among patients with beta TM, with hypogonadism reported in

approximately 40–60% of patients, short stature in 30% and

diabetes in 5–14% (Caruso-Nicoletti et al, 2004; De Sanctis

et al, 2004; Gamberini et al, 2004; Al-Rimawi et al, 2006).

Little is known, however, about the current prevalence of

various endocrinopathies and growth disturbances among

patients with other thalassemia syndromes treated with

contemporary therapies. One would expect better outcomes

in patients with mild to moderate disease who do not require

regular transfusions, such as beta thalassemia intermedia (TI)

and Hemoglobin H disease (HbH), compared to those

transfused regularly, including beta TM, E-beta thalassemia

(E-beta), homozygous alpha thalassemia (Hz a) and the co-inheritance of HbH and hemoglobin Constant Spring

(HbH/CS).

The Thalassemia Clinical Research Network (TCRN) con-

sists of five thalassemia centres in North America and their

associated satellite sites with access to patients with both alpha

and beta thalassemia syndromes. The TCRN performed a

cross-sectional study to address the above questions. Particular

emphasis was placed on the study of calcium and vitamin D

metabolism, because of the recently described high prevalence

of low bone mass across all thalassemia syndromes (Pollak

et al, 2000; Vogiatzi et al, 2006; Vogiatzi et al, 2009). Finally,

we sought to determine associations between endocrinopa-

thies, anaemia, transfusion history and chelation-related

parameters, and thalassemia co-morbidities to better under-

stand their evolution and aetiology.

Methods

Study protocol

TCRN patients of all thalassemia syndromes, age 6 years and

older, were eligible for this study. Exclusion criteria included

pregnancy and any known pre-existing medical condition

known to require chronic systemic administration of steroids.

The protocol was approved by the TCRN Data and Safety

Monitoring Board and by the ethical review boards of all

TCRN institutions. Informed consent, or assent in the case of a

minor, was obtained.

Medical history was obtained by interview and review of

medical records, and a complete physical examination was

performed. A fasting morning blood sample was obtained for

measurement of free thyroxine (T4) and TSH, insulin–like

growth factor 1 (IGF1), insulin–like growth factor binding

protein 3 (IGFBP3), serum gonadotropins (i.e. luteinizing

hormone [LH] and follicle-stimulating hormone [FSH]),

testosterone (only in males), 25-hydroxy vitamin D (25

vitamin D), 1,25-dihydroxy vitamin D (1,25 vitamin D),

intact parathyroid hormone (PTH), serum calcium, phospho-

rus, ferritin and transferrin receptor concentrations. A 24-h

urine collection was performed for measurement of calcium

excretion. Dietary calcium intake was estimated from a

46-item self-completed food frequency questionnaire. Bone

age was determined from x-rays of the left hand and wrist of

participants younger than 20 years and read locally according

to the method of Greulich and Pyle (1959).

Calculated variables. Patients with beta thalassemia were

classified as having TM if they had received eight or more

transfusions during the 12 months prior to entering the study

or as TI if they had been transfused less than eight times the

year before.

Body mass index (BMI) was calculated as kg/m2. Anthro-

pometric Z-scores were calculated relative to age- and gender-

specific norms for Caucasians produced by the Center for

Disease Control (CDC) from National Health and Nutrition

Examination Survey (NHANES) III data. Mid-parental heights

were calculated as the average parental height ±6Æ5 cm for boys and girls, respectively. Presence of an endocrinopathy (i.e.

hypogonadism, hypothyroidism, diabetes mellitus, growth

hormone deficiency and hypoparathyroidism) was defined as

having an identified clinical history or prescribed treatment.

Among females, hypogonadism was defined as lack of menses

after the age of 16 years, lack of spontaneous menarche, having

a history of using hormone replacement therapy (HRT) for

failure to proceed through puberty, loss of menses before aged

40 years, or current use of HRT. Hypogonadism in males was

determined by the prescription of HRT (testosterone or

human chorionic gonadotrophin [HCG]) or by having a

serum testosterone concentration lower than established

norms for age. Inadequate gonadal steroid replacement in

hypogonadal males was confirmed by having a serum testo-

sterone concentration lower than established norms for age

despite prescribed HRT. In addition to clinical history,

hypoparathyroidism was defined as an intact PTH level below

the normal range (10 ng/l) in the presence of albumin-

adjusted hypocalcemia (serum calcium below 2Æ1 mmol/l). Vitamin D deficiency was defined as a 25 vitamin D

concentration of less than 27 nmol/l, and insufficiency as

between 27 and 75 nmol/l. 24-h urine calcium excretion above

4 mg/kg/d was consistent with hypercalciuria. Fasting hyper-

glyacemia (blood glucose level above 70 mmol/l), having a

history of fasting hyperglycaemia or diabetes, having a history

of prescribed therapy with oral hypoglycaemics or insulin, or

current therapy with oral hypoglycaemics or insulin were used

to determine the presence of diabetes. The diagnosis of growth

Endocrinopathies and Vitamin D Abnormalities in Thalassemia

ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 146, 546–556 547

hormone deficiency in children with growth failure was made

after endocrine referral to a participating site and appropriate

testing (i.e. growth hormone provocative test). Daily calcium

intake (dietary + supplements) was considered adequate if it

was above 800 mg for those aged of 6–8 years, 1300 mg for

9–18 years, 1000 mg for 19–50 years and 1200 mg for 51 years

and older.

Laboratory assays

Urine and serum samples from each participant were stored at

)80�C and analysed as a batch at a central facility. Serum IGF1 and IGFBP3 were measured by solid-phase, enzyme-linked

immunoassays (ELISA), 25 vitamin D by competitive radio-

immunoassay following extraction, 1,25 vitamin D by column

chromatography and radioimmunoassay, intact PTH by

immunochemiluminometric assay, and free T4, LH, FSH,

and TSH by high sensitivity heterogeneous sandwich separa-

tion assays. Testosterone was measured by a solid-phase,

competitive radioimmunoassay. Serum ferritin levels were

determined by a radioimmunoassay (RIA, T-14; Ramco

Laboratories, Houston, TX, USA). Serum transferrin receptor

concentrations were measured by an enzyme immunoassay

(EIA, T-94; Ramco Laboratories).

Statistical analysis

Very rare subgroups of participants (homozygous a thalasse- mia, n = 3, and recipients of bone marrow transplants, n = 6)

were excluded from most analyses that included a thalassemia

diagnosis as a predictor to avoid bias in variance estimates and

model instability due to empty cells.

General linear models were used to model the effects of age,

gender, race, thalassemia syndrome, hypogonadism, growth

hormone deficiency, serum IGF1 and IGFBP3 concentrations,

ferritin concentration, transferrin receptors, and hepatitis C on

the following outcomes: height, weight, BMI, difference in

height from midparental height, and difference between

chronological age and bone age. Logistic regression was used

to model the effect of the same variables on short stature

(failure to thrive). Additional general linear models examined

the effect of gender, thalassemia syndrome, hypogonadism,

ferritin concentration, and growth hormone therapy on serum

IGF1 and IGFBP3 concentrations, controlling for age as a

covariate using linear splines with two knots at 11 and 20 years

of age.

Logistic regression was used to model the effects of age,

gender, thalassemia syndrome, and serum ferritin concentra-

tion on the following outcomes in patients over 10 years of

age: hypogonadism, hypothyroidism, diabetes, and the pres-

ence of multiple endocrinopathies. The effect of various

gonadal steroid replacement therapies on serum testosterone

concentration was modelled with a general linear model,

controlling for age, in medicated hypogonadal males. Addi-

tionally, a general linear model was also used to assess the

effects of gender, thalassemia syndrome, hypogonadism, and

ferritin concentration on fasting blood glucose concentration,

controlling for age and hours fasting as covariates using linear

splines with two knots each.

General linear models were used to determine the effects of

age, race, and thalassemia syndrome on the following

outcomes: dietary calcium intake, calcium supplementation,

total calcium intake, vitamin D sufficiency, and serum

concentration of intact PTH. Additional predictors of vitamin

D sufficiency were season, and urinary calcium excretion.

Analysis of variance was used to model the effect of vitamin D

sufficiency and diagnosis (TM vs. TI) on intact PTH. A Fisher

exact test and logistic regression were used to model the effect

of diagnosis (TM vs. TI) and vitamin D sufficiency on

hypercalciuria, respectively.

Analyses were generally exploratory with the aim of

describing observed patterns in the data. Corrections for

multiple comparisons were not made, and alpha £0Æ05 was considered statistically significant. All analyses were conducted

using SAS (version 9.1.3; SAS Institute, Cary, NC, USA).

Results

Patient characteristics

A total of 361 subjects, 176 males (48Æ8%) and 185 females (51Æ2%), mean age 23Æ2 years (range 6Æ1–75Æ4 years) were enrolled in the study (Table I). Of these, 236 were diagnosed

with beta TM, 43 with beta TI, 43 with E-b, 19 with HbH, and 20 with HbH/CS or other non-deletional mutations.

There were three patients with Hz a thalassemia, who were not included in this analysis because of the small sample

size. Details on the demographics of the participants, and

their transfusion and chelation regimens are presented in

Table I.

Anthropometric parameters

Growth abnormalities were common among children and

adolescents (age < 20 years): height z-score: )1Æ20 ± 1Æ16; mean ± SD; range )5Æ08 to 1Æ56 (Fig 1). Twenty-five percent were of short stature (height z-score < )2), compared to <3% of the general population. Their mean weight and BMI

were in the normal range (weight z-score: )0Æ89 ± 1Æ25, range )4Æ47 to 2Æ78; BMI z-score: )0Æ26 ± 1Æ1; range )4Æ5 ± 2Æ71). The final height z-score (calculated in partic- ipants >20 years) was )1Æ34 ± 1Æ2; range )4Æ8 to 1Æ81. Short stature occurred in 28Æ5%. As in the case of children, mean adult weight and BMI were in the normal range (weight

z-score = )0Æ72 ± 1Æ2, range )4Æ57 to 2Æ48; BMI z-score = )0Æ07 ± 1Æ05, range )4Æ51 to 2Æ44). The difference of final height from midparental height was )0Æ72 ± 1Æ23 SDs or )4Æ3 ± 7Æ4 cm.

The type of thalassemia syndrome had no effect on growth

before the age of 20 years, final height or difference of final

M. G. Vogiatzi et al

548 ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 146, 546–556

T a

b le

I. C

h ar

ac te

ri st

ic s

o f

st u

d y

p ar

ti ci

p an

ts *.

C h

ar ac

te ri

st ic

A ll

p ar

ti ci

p an

ts (n

= 3

6 1

) B

et a

T M

(n =

2 3

6 )

B et

a T

I (n

= 4

3 )

E /b

et a

(n =

4 3

) H

b H

D is

ea se

(n =

1 9

) H

b H

/C S

(n =

2 0

)

D em

o gr

a p

h ic

A g e

(y ea

rs )

2 3 Æ2

± 1

2 Æ1

(6 Æ1

to 7

5 Æ4

) 2

4 Æ4

± 1

1 Æ6

(6 Æ1

to 5

3 Æ2

) 2

6 Æ3

± 1

6 Æ2

(6 Æ1

to 7

5 Æ4

) 1

9 Æ5

± 9 Æ3

(7 Æ0

to 4

8 Æ2

) 1

5 Æ1

± 8 Æ7

(6 Æ9

to 3

9 Æ6

) 1

7 Æ1

± 1

0 Æ0

(6 Æ2

to 5

1 Æ2

)

0 6

to 1

0 y ea

rs 6

9 (1

9 Æ1

% )

3 4

(1 4 Æ4

% )

1 0

(2 3 Æ3

% )

1 1

(2 5 Æ6

% )

8 (4

2 Æ1

% )

6 (3

0 Æ0

% )

1 1

to 1

9 y ea

rs 9

4 (2

6 Æ0

% )

5 5

(2 3 Æ3

% )

9 (2

0 Æ9

% )

1 2

(2 7 Æ9

% )

8 (4

2 Æ1

% )

1 0

(5 0 Æ0

% )

2 0

y ea

rs o

r o

ld er

1 9

8 (5

4 Æ8

% )

1 4

7 (6

2 Æ3

% )

2 4

(5 5 Æ8

% )

2 0

(4 6 Æ5

% )

3 (1

5 Æ 8

% )

4 (2

0 Æ0

% )

G en

d er

(m al

e) 1

7 6

/3 6

1 (4

8 Æ8

% )

1 1

6 /2

3 6

(4 9 Æ2

% )

2 0

/4 3

(4 6 Æ5

% )

2 0

/4 3

(4 6 Æ5

% )

7 /1

9 (3

6 Æ8

% )

1 3

/2 0

(6 5 Æ0

% )

R a

A si

an 1

5 2

(4 2 Æ1

% )

7 8

(3 3 Æ1

% )

6 (1

4 Æ0

% )

4 0

(9 3 Æ0

% )

1 0

(5 2 Æ6

% )

1 8

(9 0 Æ0

% )

C au

ca si

an 1

8 6

(5 1 Æ5

% )

1 5

4 (6

5 Æ3

% )

3 0

(6 9 Æ8

% )

1 (2

Æ3 %

) 1

(5 Æ3

% )

0 (0

Æ0 %

)

O th

er 2

3 (6

Æ4 %

) 4

(1 Æ7

% )

7 (1

6 Æ3

% )

2 (4

Æ7 %

) 8

(4 2 Æ1

% )

2 (1

0 Æ0

% )

A n

th ro

p o m

et ri

H ei

g h

t z-

sc o

re )

1 Æ2

6 ±

1 Æ1

7 ()

5 Æ0

8 to

1 Æ8

1 )

) 1 Æ3

8 ±

1 Æ1

7 ()

5 Æ0

8 to

1 Æ5

6 )

) 0 Æ5

2 ±

0 Æ9

5 ()

2 Æ1

7 to

1 Æ8

1 )

) 1 Æ5

4 ±

1 Æ1

8 ()

4 Æ1

2 to

0 Æ5

7 )

) 0 Æ9

1 ±

1 Æ1

1 ()

3 Æ1

0 to

1 Æ2

4 )

) 1 Æ2

9 ±

0 Æ9

9 ()

3 Æ2

1 to

0 Æ3

0 )

W ei

g h

t z-

sc o

re )

0 Æ8

0 ±

1 Æ2

3 ()

4 Æ5

7 to

2 Æ7

8 )

) 0 Æ7

3 ±

1 Æ2

6 ()

4 Æ5

7 to

2 Æ7

8 )

) 0 Æ5

1 ±

0 Æ9

2 ()

2 Æ8

0 to

1 Æ2

0 )

) 1 Æ3

8 ±

1 Æ2

2 ()

4 Æ0

3 to

0 Æ9

6 )

) 0 Æ7

6 ±

1 Æ5

9 ()

3 Æ7

2 to

2 Æ0

6 )

) 1 Æ1

5 ±

0 Æ8

5 ()

2 Æ9

4 to

0 Æ3

2 )

B M

I z-

sc o

re )

0 Æ1

7 ±

1 Æ0

9 ()

4 Æ5

2 to

2 Æ7

1 )

) 0 Æ0

4 ±

1 Æ0

9 ()

4 Æ5

1 to

2 Æ7

1 )

) 0 Æ2

8 ±

0 Æ8

9 ()

3 Æ0

8 to

1 Æ2

2 )

) 0 Æ5

2 ±

0 Æ9

1 ()

2 Æ8

9 to

1 Æ2

3 )

) 0 Æ4

8 ±

1 Æ7

7 ()

4 Æ5

2 to

1 Æ9

3 )

) 0 Æ5

1 ±

0 Æ9

4 ()

3 Æ2

1 to

0 Æ8

8 )

B o

n e-

ag e

(y ea

rs ,

o n

ly if

< 2

0 y ea

rs )

1 1 Æ5

± 4 Æ7

(4 Æ5

to 2

0 Æ0

) 1

1 Æ7

± 4 Æ6

(5 Æ0

to 2

0 Æ0

) 1

0 Æ2

± 4 Æ7

(5 Æ0

to 2

0 Æ0

) 1

0 Æ 8

± 4 Æ1

(5 Æ8

to 1

8 Æ0

) 1

2 Æ3

± 4 Æ3

(7 Æ8

to 2

0 Æ0

) 1

3 Æ1

± 7 Æ9

(4 Æ5

to 2

0 Æ0

)

C h

ro n

o lo

g ic

al ag

e –

b o

n e

ag e

(y ea

rs ,

o n

ly if

< 2

0 y ea

rs )

0 Æ6

0 ±

1 Æ5

1 ()

4 Æ4

4 to

4 Æ5

4 )

0 Æ5

9 ±

1 Æ4

6 ()

2 Æ7

3 to

4 Æ5

4 )

0 Æ8

8 ±

1 Æ1

2 ()

0 Æ8

7 to

3 Æ4

9 )

1 Æ1

7 ±

1 Æ3

9 ()

1 Æ2

7 to

4 Æ0

5 )

) 0 Æ4

0 ±

1 Æ5

8 ()

4 Æ0

0 to

1 Æ6

9 )

) 0 Æ0

0 ±

2 Æ6

6 ()

4 Æ4

4 to

1 Æ8

2 )

H ei

g h

t z-

sc o

re –

m id

-p ar

en ta

h ei

g h

t z-

sc o

re ,

o n

ly if

> 2

0 y ea

) 0 Æ7

3 ±

1 Æ2

4 ()

4 Æ8

2 to

2 Æ6

3 )

) 0 Æ9

8 ±

1 Æ2

2 ()

4 Æ8

2 to

2 Æ6

3 )

0 Æ0

4 ±

0 Æ8

5 ()

1 Æ2

6 to

2 Æ0

1 )

) 0 Æ0

1 ±

1 Æ1

9 ()

3 Æ5

8 to

1 Æ2

3 )

0 Æ5

7 ±

0 Æ7

0 (0

Æ0 8

to 1 Æ0

7 )

0 Æ3

1 ±

0 Æ6

7 ()

0 Æ4

6 to

0 Æ7

0 )

T ra

n sf

u si

o n

s &

ch el

a ti

o n

H ae

m o

g lo

b in

(g /l

) 9

7 ±

1 4

(6 0

to 1

4 7

) 1

0 1

± 1

1 (7

3 to

1 4

7 )

9 1

± 1

8 (6

0 to

1 3

6 )

8 7

± 1

7 (6

2 to

1 2

1 )

9 5

± 9

(7 9

to 1

1 3

) 9

2 ±

1 7

(6 8

to 1

2 4

)

T ra

n sf

er ri

re ce

p to

r (m

g /l

)

2 6 Æ1

± 1

7 Æ6

(2 Æ5

to 1

1 5 Æ9

) 2

0 Æ0

± 1

1 Æ4

(2 Æ5

to 6

6 Æ3

) 4

2 Æ6

± 1

8 Æ1

(1 4 Æ6

to 8

2 Æ2

) 3

9 Æ1

± 2

4 Æ1

(6 Æ7

to 1

1 5 Æ9

) 1

6 Æ3

± 5 Æ8

(8 Æ0

to 2

8 Æ8

) 4

4 Æ7

± 1

7 Æ8

(6 Æ3

to 6

6 Æ8

)

T ra

n sf

u si

o n

ra te

(m l/

k g /m

o n

th )

9 Æ8

± 6 Æ9

(0 Æ0

to 3

1 Æ1

) 1

3 Æ2

± 4 Æ4

(0 Æ9

to 3

1 Æ1

) 2 Æ5

± 5 Æ6

(0 Æ0

to 2

5 Æ0

) 6 Æ5

± 7 Æ1

(0 Æ0

to 2

5 Æ3

) N

/A 2 Æ5

± 6 Æ4

(0 Æ0

to 2

1 Æ3

)

A g

e st

ar te

d 8

tr an

sf u

si o

n s/

y ea

(y ea

rs )

3 Æ1

± 6 Æ5

(0 Æ0

to 4

5 Æ0

) 2 Æ6

± 6 Æ4

(0 Æ0

to 4

5 Æ0

) 1

0 Æ5

± 1

0 Æ2

(0 Æ0

to 2

7 Æ0

) 4 Æ7

± 4 Æ4

(0 Æ0

to 1

7 Æ0

) N

/A 2 Æ6

± 3 Æ7

(0 Æ0

to 8 Æ0

)

Y ea

rs o

n 8

tr an

sf u

si o

n s/

y ea

1 7 Æ8

± 1

0 Æ9

(0 Æ 0

to 4

4 Æ0

) 1

9 Æ5

± 1

0 Æ7

(1 Æ0

to 4

4 Æ0

) 3 Æ8

± 2 Æ7

(0 Æ0

to 7 Æ0

) 1

0 Æ5

± 6 Æ4

(0 Æ0

to 2

5 Æ0

) N

/A 3 Æ0

± 0 Æ0

(3 Æ0

to 3 Æ0

)

F er

ri ti

n (l

g /l

) 1

6 6

2 ±

1 7

4 1

(1 0

to 1

2 2

8 0

) 1

9 9

2 ±

1 6

6 8

(9 4

to 1

1 9

9 5

) 7

3 6

± 8

6 5

(1 7

to 3

8 6

5 )

1 8

9 8

± 2

4 5

4 (1

4 to

1 2

2 8

0 )

9 2

± 7

8 (1

0 to

3 1

1 )

6 8

0 ±

1 0

3 8

(6 7

to 3

9 4

0 )

Ir o n

ch el

a ti

o n

U n

ch el

at ed

9 5

(2 6 Æ4

% )

8 (3

Æ4 %

) 2

8 (6

5 Æ1

% )

2 3

(5 4 Æ8

% )

1 9

(1 0

0 Æ0

% )

1 7

(8 5 Æ0

% )

D F

O 2

4 1

(6 6 Æ9

% )

2 0

7 (8

7 Æ7

% )

1 4

(3 2 Æ6

% )

1 9

(4 5 Æ2

% )

0 (0

Æ0 %

) 1

(5 Æ0

% )

IC L

2 1

(5 Æ8

% )

1 8

(7 Æ6

% )

1 (2

Æ3 %

) 0

(0 Æ0

% )

0 (0

Æ0 %

) 2

(1 0 Æ0

% )

D F

P 3

(0 Æ8

% )

3 (1

Æ3 %

) 0

(0 Æ0

% )

0 (0

Æ0 %

) 0

(0 Æ0

% )

0 (0

Æ0 %

)

Endocrinopathies and Vitamin D Abnormalities in Thalassemia

ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 146, 546–556 549

T a

b le

I. (C

o n

ti n

u ed

C h

ar ac

te ri

st ic

A ll

p ar

ti ci

p an

ts (n

= 3

6 1

) B

et a

T M

(n =

2 3

6 )

B et

a T

I (n

= 4

3 )

E /b

et a

(n =

4 3

) H

b H

D is

ea se

(n =

1 9

) H

b H

/C S

(n =

2 0

)

A g

e st

ar te

d D

F O

th er

ap y

(y ea

rs )

9 Æ0

± 9 Æ2

(0 Æ0

to 5

1 Æ0

) 7 Æ6

± 7 Æ6

(0 Æ0

to 4

6 Æ0

) 2

1 Æ7

± 1

2 Æ3

(6 Æ0

to 5

1 Æ0

) 1

3 Æ2

± 1

1 Æ1

(0 Æ0

to 4

3 Æ0

) N

/A 1

4 Æ3

± 2

0 Æ7

(2 Æ0

to 4

5 Æ0

)

Y ea

rs o

n D

F O

1 3 Æ4

± 8 Æ3

() 1 Æ0

to 3

0 Æ0

) 1

4 Æ5

± 8 Æ1

(0 Æ0

to 3

0 Æ0

) 7 Æ7

± 8 Æ2

() 1 Æ0

to 2

2 Æ0

) 7 Æ1

± 6 Æ2

(0 Æ0

to 2

4 Æ0

) N

/A 2 Æ0

± 1 Æ7

(1 Æ0

to 4 Æ0

)

P re

sc ri

b ed

D F

d o

se (m

g /k

g /d

)

3 2 Æ2

± 1

1 Æ5

(3 Æ7

to 6

2 Æ5

) 3

2 Æ2

± 1

1 Æ5

(3 Æ7

to 6

2 Æ5

) 2

7 Æ7

± 1

1 Æ2

(5 Æ4

to 4

4 Æ4

) 3

7 Æ5

± 8 Æ8

(2 2 Æ5

to 5

6 Æ9

) N

/A 1

3 Æ2

± .

(1 3 Æ2

to 1

3 Æ2

)

T h

a la

ss em

ia co

m p

li ca

ti o n

H y p

o g o

n ad

is m

N o

rm al

2 0

7 (5

8 Æ5

% )

1 0

1 (4

3 Æ9

% )

3 5

(8 1 Æ4

% )

3 3

(7 8 Æ6

% )

1 8

(9 4 Æ7

% )

2 0

(1 0

0 Æ0

% )

D el

ay ed

p u

b er

ty 1

9 (8

Æ2 %

) 1

1 (9

Æ3 %

) 4

(1 0 Æ3

% )

3 (8

Æ1 %

) 1

(5 Æ3

% )

0 (0

Æ0 %

)

H y p

o g

o n

ad al

1 2

8 (3

6 Æ2

% )

1 1

8 (5

1 Æ3

% )

4 (9

Æ3 %

) 6

(1 4 Æ3

% )

0 (0

Æ0 %

) 0

(0 Æ0

% )

G ro

w th

h o

rm o

n e

d efi

ci en

3 3

/3 4

4 (9

Æ6 %

) 2

9 /2

2 3

(1 3 Æ0

% )

2 /4

2 (4

Æ8 %

) 2

/4 2

(4 Æ8

% )

0 /1

8 (0

Æ0 %

) 0

/1 9

(0 Æ0

% )

H y p

o th

y ro

id is

m 3

1 /3

5 8

(8 Æ 7

% )

2 8

/2 3

4 (1

2 Æ0

% )

2 /4

3 (4

Æ7 %

) 0

/4 2

(0 Æ0

% )

0 /1

9 (0

Æ0 %

) 1

/2 0

(5 Æ0

% )

H y p

o p

ar at

h y ro

id is

m 5

/3 5

8 (1

Æ4 %

) 5

/2 3

4 (2

Æ1 %

) 0

/4 2

(0 Æ0

% )

0 /4

3 (0

Æ0 %

) 0

/1 9

(0 Æ0

% )

0 /2

0 (0

Æ0 %

)

D ia

b et

es M

el li

tu s

3 3

/3 5

8 (9

Æ2 %

) 3

3 /2

3 4

(1 4 Æ1

% )

0 /4

3 (0

Æ0 %

) 0

/4 2

(0 Æ0

% )

0 /1

9 (0

Æ0 %

) 0

/2 0

(0 Æ0

% )

2 5

(O H

)V it

< 2

7 n

m o

l/ l

4 2

(1 2 Æ0

% )

2 7

(1 1 Æ7

% )

5 (1

2 Æ2

% )

6 (1

4 Æ3

% )

0 (0

Æ0 %

) 4

(2 0 Æ0

% )

2 7

– 7

5 n

m o

l/ l

2 4

5 (6

9 Æ8

% )

1 4

9 (6

4 Æ8

% )

3 1

(7 5 Æ6

% )

3 5

(8 3 Æ3

% )

1 5

(8 3 Æ3

% )

1 5

(7 5 Æ0

% )

‡7 5

n m

o l/

l 6

4 (1

8 Æ2

% )

5 4

(2 3 Æ5

% )

5 (1

2 Æ2

% )

1 (2

Æ4 %

) 3

(1 6 Æ7

% )

1 (5

Æ0 %

)

M ed

ic at

ed h

ea rt

d is

ea se

2 0

/3 5

9 (5

Æ6 %

) 2

0 /2

3 5

(8 Æ5

% )

0 /4

3 (0

Æ0 %

) 0

/4 2

(0 Æ0

% )

0 /1

9 (0

Æ0 %

) 0

/2 0

(0 Æ0

% )

C ir

rh o

si s

1 3

/3 5

2 (3

Æ7 %

) 1

2 /2

3 1

(5 Æ2

% )

1 /4

2 (2

Æ4 %

) 0

/4 0

(0 Æ 0

% )

0 /1

9 (0

Æ0 %

) 0

/2 0

(0 Æ0

% )

H ep

at it

is C

8 4

/3 5

9 (2

3 Æ4

% )

7 5

/2 3

5 (3

1 Æ9

% )

3 /4

3 (7

Æ0 %

) 6

/4 2

(1 4 Æ3

% )

0 /1

9 (0

Æ0 %

) 0

/2 0

(0 Æ0

% )

D F

O ,

d ef

er o

x am

in e;

IC L

, d

ef er

as ir

o x

(o ra

l ir

o n

ch el

at o

r) ;

D F

P ,

d ef

er ip

ro n

e (o

ra l

ir o

n ch

el at

o r)

; B

M I,

B o

d y

m as

s In

d ex

*V al

u es

fo r

co n

ti n

u o

u s

v ar

ia b

le s

ar e

m ea

n ±

S D

(r an

g e)

. V

al u

es fo

r ca

te g o

ri ca

l v ar

ia b

le s

ar e

N /T

o ta

l (%

). T

o ta

ls fo

r so

m e

m ea

su re

s ar

e le

ss th

an th

e to

ta l

n u

m b

er o

f p

ar ti

ci p

an ts

d u

e to

m is

si n

g d

at a.

M. G. Vogiatzi et al

550 ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 146, 546–556

height from midparental height. Among children and adoles-

cents (age < 20 years), hypogonadism was associated with

decreased height (P = 0Æ04), weight (P < 0Æ01) and BMI (P = 0Æ02) in a model that included age, gender, race and type of thalassemia as co-variants. Growth hormone deficiency

was also a significant negative predictor of height (P = 0Æ05) and weight (P = 0Æ03). Further multivariate analyses that, in addition, included IGF1 concentrations and serum ferritin and

transferrin receptor concentrations as co-variants showed a

1Æ3-fold increase in the odds of short stature with advancing age (P = 0Æ022) and a 4Æ5-fold increase in Asians compared to Caucasians (P = 0Æ045). Similar analyses in adults found that the only significant predictor of short stature was growth

hormone deficiency (OR = 3Æ8 [1Æ2Æ12Æ4]); P = 0Æ025). Among participants younger than 20 years, bone age (BA)

was close to chronological age (CA; Table I). The difference

between CA and BA was greater in patients with hypogona-

dism (P = 0Æ01) and among patients with beta thalassemias compared to HbH disease (P = 0Æ02). Gender, race, and serum IGF1 concentrations were not significantly associated with

delayed BA.

Serum IGF1 and IGFBP3 concentrations were plotted

against normative values for sex and age (Fig 2). Seventy-one

percent of all patients had IGF1 concentrations below normal

(57Æ7%, 56Æ8% and 83Æ3% for the age groups 6–11, 12–19 and >20 years, respectively). IGFBP3 concentrations were below

normal in 34Æ4% of the subjects. Increased ferritin concentra- tions (P £ 0Æ001), male gender (P £ 0Æ02), the presence of hypogonadism (P £ 0Æ01), and TI (P £ 0Æ02) were all associ- ated with lower serum IGF1 and IGFBP3 concentrations. Of

interest, IGF1 and IGFBP3 concentrations were not found to

be predictors of short stature.

Hypogonadism and other endocrinopathies

Hypogonadism was the most frequent endocrinopathy and

affected both genders. It was present in 14Æ3% of females and 25Æ5% of males younger than 20 years. Its frequency was significantly higher after 20 years of age, at 52Æ4% in woman and 60Æ0% in men.

Hypogonadism was more frequent in patients with beta TM

compared to those with beta TI (P < 0Æ001) or those with E-beta (P = 0Æ006) (Table I) among participants greater than 10 years old and after adjustment for age, gender, and serum

ferritin levels. In addition to diagnosis, both age and serum

ferritin concentration at the time of the study were associated

with the development of hypogonadism. Specifically, the odds

of hypogonadism increased by 13Æ9% for every 5 years of age (P < 0Æ001) and by 11Æ8% for every 1000 lg/l increase in ferritin concentration (P = 0Æ04).

The age of menarche was 15Æ7 ± 3Æ9 years for the overall group. For females who experienced normal puberty, menar-

che occurred at 13Æ3 ± 1Æ0 years (the youngest being 11 years old). Girls with delayed puberty reached spontaneous menar-

che at 17Æ2 ± 1Æ2 years. Among those with a diagnosis of

hypogonadism, menses were induced at 17Æ1 ± 4Æ0 years. Hypogonadal females were treated with either standard

postmenopausal gonadal steroid replacement regimens (i.e. a

combination of premarin and provera) or a birth control

given not for contraception but as a form of gonadal

replacement. Twenty-five percent of hypogonadal females

were untreated at the time of the study. Among hypogonadal

males, 36Æ2% were treated with either testosterone, given intramuscularly or transdermally by patch or gel, 7Æ25% were treated with HCG and 56Æ5% remained untreated. Of those treated, 27% had serum testosterone concentrations below

the normal range indicating inadequate gonadal steroid

replacement. The mean testosterone concentration among

adequately replaced hypogonadal males was 23Æ3 ± 8Æ2 nmol/l (range 12Æ2–38Æ2 nmol/l) vs. 3Æ9 ± 3Æ5 nmol/l (range 0Æ35– 11Æ2 nmol/l) among poorly replaced patients. The type of steroid replacement made no difference on serum testosterone

concentrations.

Growth hormone deficiency was the second most frequent

endocrinopathy and was reported in 9Æ6% of the study participants, with a mean age at diagnosis of 14Æ3 years (range 5–28 years). Most (90Æ3%) of the growth hormone-deficient patients (8Æ5% of all study participants) had been treated with growth hormone. There were no adults currently receiving

growth hormone replacement.

Diabetes mellitus (DM) was seen only in TM, with a

frequency of 14Æ1% and a mean age at diagnosis of 22 years, range 10–45 years. Of the diabetics, 29Æ5% were treated with an oral hypoglycemic and 47Æ1% with insulin. Analysis of fasting blood glucose concentrations in the non-diabetic, otherwise

asymptomatic, participants identified 37 patients (13Æ9%) with impaired glucose tolerance (mean age 26 years, range 8Æ8– 47Æ5 years) and four additional patients (1Æ5%) with diabetes. In the group of patients who were Hepatitis C positive, 35Æ7% had either a diagnosis of diabetes or an abnormal fasting blood

glucose (either in the diabetic or impaired glucose tolerance

range), compared to 16Æ3% of patients without Hepatitis C (P = 0Æ0001, chi-square test). The overall prevalence of hypo- thyroidism was 8Æ7% (mean age at diagnosis 23Æ8 years, range 10–44 years). Hypoparathyroidism occurred only in TM (2Æ1% prevalence; mean age at diagnosis, 19 years, range 12–

30 years).

An increase of 5 years in age was associated with a 13Æ6% increase in the risk of hypothyroidism and 14% increase in the

odds of DM using multivariate analysis. In the same analysis,

the serum ferritin concentration at the time of the study was

not a predictor of the above-mentioned endocrinopathies.

However, the results differed when factors that affected the

development of multiple combined endocrinopathies were

considered. Patients with increasing serum ferritin concentra-

tions (P = 0Æ05) and those with beta TM had higher rates of multiple endocrinopathies compared to beta TI, E-b and a hemoglobinopathies (P < 0Æ0001). Finally, increasing serum ferritin concentrations (P = 0Æ002), the presence of hypo- gonadism (P = 0Æ01) and having beta TM (P = 0Æ05) were all

Endocrinopathies and Vitamin D Abnormalities in Thalassemia

ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 146, 546–556 551

Age (years)

H e ig

h t (c

m )

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

100

105

110

115

120

125

130

135

140

145

150

155

160

165

170

175

180

185

190

–5·00

% Z

–4·00

% Z

–3·00

% Z

1 –2·33

% Z

3 –1·88

% Z

10 –1·28

% Z

25 –0·67

% Z

50 0·00

% Z

75 0·67

% Z

90 1·28

% Z 97 1·88 % Z

●

● ●

●

● ●

●

● ●

●

● ●

●

● ●

●

● ●

●

● ●

●

■

■ ■

■

■ ■

■

■ ■

■

■ ■

■

H e ig

h t (c

m )

W e ig

h t (k

g )

W e ig

h t (k

g )

Age (years) 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

100

105

110

115

120

125

130

135

140

145

150

155

160

165

170

175

180

185

190

–5·00

% Z

–4·00

% Z

–3·00

% Z

1 –2·33

% Z

3 –1·88

% Z

10 –1·28

% Z

25 –0·67

% Z

50 0·00

% Z

75 0·67

% Z

90 1·28

% Z

97 1·88

% Z

●

● ●

●

● ●

●

● ● ●

● ●

●

● ●

●

● ●

●

● ●

●

● ●

●

● ●

●

● ●

■

■ ■

■

■ ■

■

■ ■

■

■ ■

■

■ ■

■

Age (years) 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

100

–5·00

% Z

–4·00

% Z

–3·00

% Z

1 –2·33

% Z

3 –1·88

% Z

10 –1·28

% Z

25 –0·67

% Z

50 0·00

% Z

75 0·67

% Z

90 1·28

% Z

97 1·88

% Z

●

● ●

●

● ●

●

● ● ●

●

● ●

●

● ●

●

● ●

●

● ●

●

● ●

●

● ●

●

● ●

●

● ●

●

Age (years) 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

100

–5·00

% Z

–4·00

% Z

–3·00

% Z

1 –2·33

% Z

3 –1·88

% Z

10 –1·28

% Z

25 –0·67

% Z

50 0·00

% Z

75 0·67

% Z

90 1·28

% Z

97 1·88

% Z

● ●

●

● ●

●

● ●

●

● ●

●

● ●

●

● ● ●

●

(A) (B)

M. G. Vogiatzi et al

552 ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 146, 546–556

associated with higher fasting glucose levels in multivariate

analysis.

Calcium metabolism and vitamin D

Measurements of serum 25 vitamin D concentrations revealed

that 12Æ0% of the subjects were vitamin D deficient and 69Æ8% had insufficient levels. There were no differences in the

prevalence of abnormal 25 vitamin D concentrations according

to the thalassemia syndrome. Twenty-five vitamin D concen-

trations were lower among adolescents (47 ± 20 nmol/l in the

11- to 19-year old group vs. 62 ± 21 nmol/l in the 6- to

10-year old group and 58Æ4 ± 31 nmol/l among the 20 years+ group; P = 0Æ0014) and among Asians (P = 0Æ02). As expected, 25 vitamin D was higher during the summer months

(P = 0Æ0001). Patients with vitamin D deficiency had decreased urinary calcium excretion (P = 0Æ0004).

Total daily calcium intake (dietary and supplementation)

was adequate across all thalassemia syndromes and age groups.

Dietary calcium intake was lower among older subjects

(P < 0Æ0001) and Asians (P = 0Æ018) and did not differ according to the type of thalassemia syndrome. Calcium

supplementation was prescribed more frequently in older

patients (P = 0Æ004) and in beta TM (P = 0Æ03), most likely

reflecting efforts to treat patients with low bone mass. The total

calcium intake in beta TM was 1721 ± 924 mg/d.

PTH levels decreased with increasing vitamin D concentra-

tions (P < 0Æ001), in a pattern similar to that described in the general population. Hypoparathyroidism was rare, and was

seen in only five patients with beta TM. However, patients with

beta TM had lower PTH concentrations compared to those

with beta TI for the same 25 vitamin D levels (P = 0Æ02; Fig 3). This finding could not be attributed to differences in calcium

intake, because total calcium intake did not significantly vary

according to the thalassemia diagnosis or vitamin D sufficiency

(Fig 3). Hypercalciuria was present in 29% of patients with

beta TM compared to none with beta TI (P = 0Æ002; Table II). Among patients with beta TM, the incidence of hypercalciuria

was increased in vitamin D sufficient subjects compared to

those with vitamin D deficiency (OR = 3Æ2, P = 0Æ007).

Discussion

We have defined differences in the rates of growth, endocrine-

and calcium-related abnormalities in the various thalassemia

syndromes in North America with current therapy. The

results of this study indicate that patients with beta TM

constitute a high risk group, as they manifested higher rates of

Fig 1. Plots of height, bone age and weight of study participants. A: Height of male participants versus calendar age (d) and bone age ( ). B: Height

of female participants versus calendar age (d) and bone age ( ). Individuals are connected by dashed lines. C: Weight of male participants versus

calendar age. D: Weight of female participants versus calendar age.

Male IG

F 1 (

m cg

/l)

100

1000 (A) (B)

Female

0 10 20 30 40 50 0 10 20 30 40 50

IG F

B P

3 (

m g /l)

Age (years)

Hypogonadal Not hypogonadal

Fig 2. IGF1 and IGFBP3 versus age stratified by gender and hypogonadal status. +: hypogonadal, o: not hypogonadal. IGF1 concentrations below the

normal range were found in 57Æ7% of children aged 6–11 years, 56Æ8% of those 12–19 years and 83Æ3% of subjects older than 20 years. All children aged 6–11 years had normal IGFBP3 levels. IGFBP3 concentrations below the normal range were found in 12Æ5% of children aged 12–19 years and 58Æ3% of subjects older than 20 years.

Endocrinopathies and Vitamin D Abnormalities in Thalassemia

ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 146, 546–556 553

hypogonadism, combined multiple endocrinopathies and

worse fasting hyperglycemia than individuals with other

thalassemia syndromes. All of these conditions were found

to correlate with higher ferritin concentrations. In addition,

patients with beta TM had subtle abnormalities in PTH

secretion and high rates of hypercalciuria, which appear as

newly-appreciated complications of this particular syndrome.

Hypogonadism remains the most common endocrinopathy

(De Sanctis et al, 2004; Borgna-Pignatti et al, 2005; Shalitin

et al, 2005) and is frequently under-treated. Among hypogo-

nadal girls, the age of menarche was delayed to 17 years.

Approximately half of the males were untreated, and of those

who were receiving gonadal steroid replacement, 27% had low

serum testosterone concentrations. We have previously

reported a high incidence of low bone mass and fractures in

this population and identified hypogonadism as a strong

independent predictor of low bone mass (Vogiatzi et al, 2009).

In light of the above, we advocate a more timely and aggressive

hormone replacement therapy. The most beneficial gonadal

steroid regimen for the treatment of low bone mass in

thalassemia has yet to be established, however, and needs to be

considered in future prospective studies.

Height was only mildly affected in the overall group. Weight

and BMI were not decreased relative to height, suggesting that

nutritional factors are unlikely to adversely affect stature in this

cohort. Nevertheless, approximately 25% of children and

adults present with growth failure, which occurs independent

of the thalassemia syndrome, serum IGF1 and IGFBP3 levels

and serum ferritin concentration. Similar recent studies report

that short stature occurs in approximately 30–35% of beta TM

patients (De Sanctis et al, 2004; Shalitin et al, 2005; Skordis

et al, 2006). Rates of growth failure were similar among

children and adults. This implies that the current transfusion

and chelation therapies, which were introduced approximately

25 years ago, have not significantly improved growth in

thalassemia.

Both normal and low vitamin D concentrations have been

reported previously in thalassemia (Voskaridou et al, 2001;

Table II. Calcium metabolism data and preva-

lence of hypercalciuria according to vitamin D

sufficiency status and diagnosis*.

Mean ± SD or N (%)

25 vitamin D

< 27 nmol/l

25 vitamin D

= 27–75 nmol/l

25 vitamin D

> 75 nmol/l

All patients N = 43 N = 251 N = 66

Intact PTH (ng/l) 48Æ67 ± 26Æ46 38Æ80 ± 20Æ70 28Æ61 ± 14Æ13 1,25 vitamin D (pmol/l) 97Æ86 ± 40Æ32 116Æ43 ± 42Æ15 134Æ86 ± 53Æ67 Serum Ca (mmol/l) 2Æ25 ± 0Æ14 2Æ28 ± 0Æ14 2Æ31 ± 0Æ14 Total Ca intake (g/d) 1Æ47 ± 0Æ95 1Æ63 ± 0Æ91 1Æ66 ± 0Æ93 % Hypercalciuria

(>4 mg/kg/24 h)

5/24 (21%) 27/130 (21%) 20/42 (48%)

Beta TM N = 27 N = 150 N = 54

Intact PTH (ng/l) 43Æ44 ± 21Æ75 33Æ16 ± 18Æ70 27Æ63 ± 14Æ60 1,25 vitamin D (pmol/l) 92Æ70 ± 40Æ40 112Æ21 ± 44Æ30 130Æ40 ± 56Æ10 Serum Ca (mmol/l) 2Æ27 ± 0Æ14 2Æ29 ± 0Æ15 2Æ30 ± 0Æ13 Total Ca intake (g/d) 1Æ73 ± 0Æ91 1Æ71 ± 0Æ87 1Æ78 ± 0Æ97 % Hypercalciuria

(>4 mg/kg/24 h)

5/21 (24%) 27/115 (23%) 20/40 (50%)

Beta TI N = 5 N = 31 N = 5

Intact PTH (ng/l) 48Æ40 ± 17Æ76 42Æ58 ± 21Æ50 39Æ00 ± 6Æ24 1, 25 vitamin D (pmol/l) 75Æ61 ± 34Æ20 117Æ40 ± 41Æ70 139Æ18 ± 31Æ30 Serum Ca (mmol/l) 2Æ23 ± 0Æ11 2Æ29 ± 0Æ13 2Æ30 ± 0Æ21 Total Ca intake (g/d) 1Æ85 ± 1Æ05 1Æ66 ± 0Æ98 1Æ17 ± 0Æ67 % Hypercalciuria (>4 mg/kg/24 h) 0/3 (0%) 0/15 (0%) 0/2 (0%)

*Totals for some measures are less than the total number of participants due to missing data.

80 P

a ra

th yr

o id

h o rm

o n e (

n g /l)

25 vit D (nmol/l)

beta TM beta TI

T o ta

l c a lc

iu m

in ta

ke (

g /d

)

25 vit D (nmol/l)

beta TM beta TI

<27 >7527–75 <27 >7527–75

Fig 3. Parathyroid hormone and total Calcium intake according to

vitamin D sufficiency and diagnosis. Parathyroid hormone is signifi-

cantly lower in beta TM compared to beta TI (P = 0Æ02), and decreases with vitamin D (P < 0Æ001). Total Calcium intakes does not signifi- cantly vary by diagnosis (P = 0Æ76) or vitamin D sufficiency (P = 0Æ88).

M. G. Vogiatzi et al

554 ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 146, 546–556

Napoli et al, 2006). In this study, 12% of the participants had

25 vitamin D levels below 27 nmol/l and 82% below 75 nmol/l

regardless of the thalassemia syndrome. The 25 vitamin D

concentrations used as cut-off points to define vitamin D

deficiency and sufficiency have recently increased (Holick,

2007). Depending on the definition, the prevalence of vitamin

D deficiency has been reported to range in the U.S. from 15%

to 80% (Holick, 2007). A current review from the National

Health and Nutrition Examination Survey (NHANES) III, a

cross-sectional survey administered to a representative sample

of U.S. civilians, documented that the prevalence of vitamin D

deficiency was 8Æ6% in 2005–2006 vs. 12% in this study using the standard of 25 nmol/l (Ginde et al, 2009). According to

same study, 70–80% of the population had 25 vitamin D levels

below 75 nmol/l (Ginde et al, 2009), which is close to the

prevalence that we detected. Compared to children and adults

with thalassemia, our adolescents had lower mean 25 vitamin

D concentration at 47 nmol/l. The recently reported mean

concentration among U.S. adolescents was considerably higher,

at approximately 60–75 nmol/l (Ginde et al, 2009; Saintonge

et al, 2009). All together, our data indicate that patients with

thalassemia in the U.S. have rates of vitamin D abnormalities

slightly higher than the general population with the exception

of adolescents, which represent a high-risk group for low

vitamin D concentrations. Adolescence is a critical period for

optimal bone accrual. Vitamin D abnormalities may contrib-

ute, therefore, to the low bone mass accrual that is seen among

adolescents with thalassemia (Vogiatzi et al, 2009).

Overt hypoparathyroidism is rare in well-chelated thalassemia

patients, while the presence of subtle abnormalities in PTH

secretion is not well established (Chern & Lin, 2002; Even et al,

2007). In the current study, we observed lower PTH concentra-

tions in beta TM, after correcting for vitamin D concentrations

and calcium intake, indicative of subclinical hypoparathyroid-

ism. Of interest, patients with beta TM had high rates of

hypercalciuria that worsened with increasing vitamin D con-

centrations. This is a new finding. Whether it is related to a renal

tubular defect (Smolkin et al, 2008) and/or a contributing effect

of subtle PTH dysfunction on calcium excretion cannot be

determined at this point. The optimal vitamin D concentrations

that are required for bone accrual and heath while minimizing

hypercalciuria, have not been established in thalassemia. Finally,

the presence and extent of renal tubular damage, the long-term

consequences of hypercalciuria (i.e. increased risk for nephro-

lithiasis) as well as the development of appropriate treatment

strategies need to be addressed in future studies.

Although the present study reports on a large number of

patients with all thalassemia syndromes, the cross-sectional

design limits our ability to determine the aetiology of described

endocrine and calcium abnormalities beyond the extent of

determining certain associations. Determination of serum

ferritin was the only methodology used to assess iron overload

in this study. Since serum ferritin is a poor marker in heavily

overloaded individuals, it is possible that the role of iron on

the development of various endocrinopathies was under-

estimated in this study. Regardless of these limitations, this

study provides evidence that endocrinopathies in thalassemia

are frequent and start early in life. We have identified patients

with beta TM as a particularly vulnerable group for multiple

endocrinopathies and adolescents as prone to vitamin D

abnormalities, regardless of their genotype. Our findings can

be used to establish evidence-based practise guidelines for

regular endocrine evaluations starting in childhood as part of

the comprehensive care of these patients. Given that

life expectancy in thalassemia has improved significantly

with current transfusion and chelation, we need to better

understand the long-term implications of these endocrine

complications on the welfare of the ageing patient.

Acknowledgements

Supported by a cooperative agreement with the National

Heart, Lung, and Blood Institute, National Institutes of Health

(U01-HL-65232 to Children’s Hospital of Philadelphia, U01-

HL-65233 to University Health Network Toronto General

Hospital, U01-HL-65239 to Children’s Hospital and Research

Center at Oakland, U01-HL-65244 to Weill Medical College of

Cornell University, U01-HL-65260 to Children’s Hospital

Boston, and U01-HL-65238 to New England Research Insti-

tutes). The study was also supported, in part, at Weill Medical

College of Cornell University grant K08 HL088231 awarded to

Maria G. Vogiatzi, at Children’s Hospital of Philadelphia by

NIH-NCRR grant UL1-RR024134, and at Children’s Hospital

Boston by NIH-NCRR grant M01-RR02172 and NIH grant

5K24HL004184-08 to Ellis Neufeld.

References

Al-Rimawi, H.S., Jallad, M.F., Amarin, Z.O. & Al Sakaan, R. (2006)

Pubertal evaluation of adolescent boys with b-thalassemia major and delayed puberty. Fertility and Sterility, 86, 886–890.

Borgna-Pignatti, C., Cappellini, M.D., De Stefano, P., Del Vecchio, G.C.,

Forni, G.L., Gamberini, M.R., Ghilardi, R., Origa, R., Piga, A., Romeo,

M.A., Zhao, H. & Cnaan, A. (2005) Survival and complications of

thalasemia. Annals of the New York Academy of Sciences, 1054, 40–47.

Brittenham, G.M., Griffith, P.M., Nienhuis, A.W., McLaren, C.E.,

Young, N.S., Tucker, E.E., Allen, C.J., Farrell, D.E. & Harris, J.W.

(1994) Efficacy of deferoxamine in preventing complications of iron

overload in patiens with beta thalassemia major. The New England

Journal of Medicine, 331, 567–573.

Calleja, E., Shen, J.Y., Lesser, M., Grady, R.W., New, M.I. & Giardina,

P.J. (1998) Survival and morbidity in transfusion dependent thal-

assemia patients on subcutaneous desferrioxamine chelation: nearly

two decades of experience. Annals of the New York Academy of

Sciences, 850, 469–470.

Caruso-Nicoletti, M., De Sanctis, V., Raiola, G., Skordis, N., Mancuso,

M., Coco, M. & Wonke, B. (2004) No difference in pubertal growth

and final height between treated hypogonadal and non-hypogonadal

thalassemic patients. Hormone Research, 62, 17–22.

Chern, J.P. & Lin, K.H. (2002) Hypoparathyroidism in transfusion-

dependent patients with b-Thalassemia. Journal of Pediatric Hema- tology Oncology, 24, 291–293.

Endocrinopathies and Vitamin D Abnormalities in Thalassemia

ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 146, 546–556 555

Cunningham, M.J., Maclin, E.A., Neufeld, E.J. & Cohen, A.R. for the

Thalassemia Clinical Research Network. (2004) Complcations of

b-thalassemia major in North America. Blood 104, 34–39. De Sanctis, V., Katz, M., Vullo, C., Bagni, B., Ughi, M. & Wonke, B.

(1994) Effect of different treatment regimes on linear growth and

final height in beta-thalassaemia major. Clinical Endocrinology, 40,

791–798.

De Sanctis, V., Eleftheriou, A. & Malaventura, C. (2004) Thalassemia

International Federation Study Group on growth and endocrine

complications in thalassemia. Prevalence of endocrine complications

and short stature in patients with thalassemia major: a multicenter

study by the Thalassemia International Federation (TIF). Pediatric

Endocrinology Reviews, 2(Suppl. 2), 249–255.

Even, L., Bader, T. & Hochberg, Z. (2007) Nocturnal calcium, phos-

phorus and parathyroid hormone in the diagnosis of concealed and

subclinical hypopathyroidsim. European Journal of Endocrinology,

156, 113–116.

Gamberini, M.R., Fortini, M., De Sanctis, V., Gilli, G. & Testa, M.R.

(2004) Diabetes mellitus and impaired glucose tolerance in thalas-

semia major: incidence, prevalence, risk factors and survival in

patients followed in the Ferrara Center. Pediatric Endocrinology

Reviews, 2( S2), 285–291.

Ginde, A.A., Liu, M.C. & Camargo, C.A. (2009) Demographic differ-

ences and trends in vitamin D insufficiency in the US population,

1998–2004. Archives of Internal Medicine, 169, 626–632.

Greulich, W.W. & Pyle, S.I. (1959) Radiographic Atlas of Skeletal

Development of the Hand and Wrist, 2nd edn. Stanford University

Press, Stanford, CA.

Holick, M.F. (2007) Vitamin D deficiency. The New England Journal of

Medicine, 357, 266–281.

Jensen, C.E., Tuck, S.M., Old, J., Morris, R.W., Yardumian, A.,

De Sanctis, V., Hoffbrand, A.V. & Wonke, B. (1997) Incidence of

endocrine complications and clinical disease severity related to

genotype analysis and iron overload in patients with B-thalassemia.

European Journal of Haematology, 59, 76–81.

Napoli, N., Carmina, E., Bucchieri, S., Sferrazza, C., Rini, G.B. & Di

Fede, G. (2006) Low serum levels of 25-hydroxy vitamin D in adults

affected by thalasemia major or intermedia. Bone, 38, 888–892.

Pollack, R.D., Rachmilewitz, E., Blumenfeld, A., Idelson, M. & Goldfarb,

A.W. (2000) Bone mineral metabolism in adults with B-thalassemia

major and intermedia. British Journal of Haematology, 111, 902–907.

Roth, C., Pekrun, A., Bartz, M., Jarry, H., Eber, S., Lakomek, M. &

Schroter, W. (1997) Short stature and failure of pubertal develop-

ment in thalassemia major: evidence of hypothalamic neurosecre-

tory dysfunction of growth hormone secretion and defective

pituitary gonadotropin secretion. European Journal of Pediatrics,

156, 777–783.

Saintonge, S., Bang, H. & Gerber, L.M. (2009) Implications on a new

definition of vitamin D deficiency in a multiracial US adolescent

population: The National Health and Nutrition Examination Survey

III. Pediatrics, 123, 797–803.

Shalitin, S., Carmi, D., Weintrob, N., Phillip, M., Miskin, H., Kornr-

eich, L., Zilber, R., Yaniv, I. & Tamary, H. (2005) Serum ferritin level

as a predictor of impaired growth and puberty in thalassemia major

patients. European Journal of Haematology, 74, 93–100.

Skordis, N., Michaelidou, M., Savva, S.C., Ioannou, Y., Rousounides,

A., Kleanthous, M., Skordos, G. & Christou, S. (2006) The impact of

genotype on endocrine complications in thalassemia major. Euro-

pean Journal of Haematology, 77, 150–156.

Smolkin, V., Halevy, R., Levin, C., Mines, M., Sakran, W., Ilia, K. &

Koren, A. (2008) Renal function in children with b-thalassemia major and thalassemia intermedia. Pediatric Nephrology, 23, 1847–

1851.

Vogiatzi, M.A., Macklin, E.A., Fung, E.B., Vichinsky, E., Oliveri, N.,

Kwiatkowski, J.L., Cohen, A., Neufeld, E. & Giardina, P.J. (2006)

Prevalence of fractures among the Thalassemia syndromes in North

America. Bone, 38, 571–575.

Vogiatzi, M.A., Macklin, E.A., Fung, E.B., Cheung, A.M., Vichinsky, E.,

Oliveri, N., Kirby, M., Kwiatkowski, J.L., Cunningham, M., Holm, I.,

Lane, J., Schneider, R., Fleisher, M., Grady, R.W., Peterson, C. &

Giardina, P.J. for the Thalassemia Clinical Research Network (2009)

Bone disease in thalassemia: a frequent and still unresolved problem.

Journal of Bone and Mineral Research, 24, 543–557.

Voskaridou, E., Kyrtsonis, M.C., Tempos, E., Skordili, M., Theodor-

opoulos, I., Bergele, A., Diamanti, E., Kalovidoulis, A., Loutradi, A.

& Loukopoulos, D. (2001) Bone resorption is increased in young

adults with thalassemia major. British Journal of Haematology, 112,

36–41.

Appendix 1

This work was performed through the Thalassemia Clinical

Research Network (TCRN). The authors would like to thank

the patients who volunteered their time to participate in this

study. The following TCRN sites and investigators also

contributed to the study (listed in alphabetical order):

Childrens Hospital, Boston: Ellis Neufeld, MD, PhD, Principal

Investigator, Melody Cunningham, MD, Co-Principal Inves-

tigator; Childrens Hospital of Philadelphia: Alan R. Cohen,

MD, Principal Investigator, Janet L. Kwiatkowski, MD,

Co-Principal Investigator, Catherine S. Manno, MD, Coinves-

tigator, Marie Martin, RN, Nurse Coordinator, Debra Hill-

man, Regulatory Affairs Coordinator, Gail M. Jackson, CDT,

Nutrition Assessment Program Coordinator, Maria J. Hen-

wood-Storto, DO, Endocrinologist; Shannon H. Fourtner,

MD, Endocrinologist; Childrens Hospital & Research Center

Oakland: Elliott Vichinsky, MD, Principal Investigator, Dru

Foote, NP, Study Coordinator, Eun-Ha Pang, Study Coordi-

nator, Zahra Pakbaz, MD, CCD, Certified Clinical Densitom-

etrist, Selma Holden, Study Coordinator; Toronto General

Hospital: Nancy Olivieri, MD, Principal Investigator; U.T.

Southwestern Medical Center: Charles T. Quinn, MD, Assis-

tant Professor of Pediatrics, Elizabeth Shull, BSN, RN, Clinical

Research Associate; Weill Medical College of Cornell Univer-

sity: Patricia J. Giardina, MD, Principal Investigator, Robert

W. Grady, PhD, Co-Investigator, Jeffrey E. Mait and Dorothy

Kleinert, NP, MPH, MA, Study Coordinators, Irina Cha-

ikhoutdinov and Gladys Cintron, Data Coordinators, Sylvia

Hom, DXA Technician, Hospital for Special Surgery; National

Heart, Lung, and Blood Institute: Charles Peterson, MD,

Project Officer; Data Coordinating Center, New England

Research Institutes: Sonja McKinlay, PhD, Principal Investi-

gator, Felicia Trachtenberg, PhD, Statistician, Haddy Jallow,

MS, MPH, Project Coordinator.

M. G. Vogiatzi et al

556 ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 146, 546–556