Original Article

Journal of Child Neurology

1-6
Assessment of Cognitive Function in ª The Author(s) 2014
Reprints and permission:
sagepub.com/journalsPermissions.nav
Children With Beta-Thalassemia Major: DOI: 10.1177/0883073814550827
jcn.sagepub.com
A Cross-Sectional Study

Nelly Raafat, MD1, Usama El Safy, MD2, Nahed Khater, MD2,

Tamer Hassan, MD2, Basheir Hassan, MD2, Ahmed Siam, MD2,
Amira Youssef, MD1, and Amany El Shabrawy, MD1

Abstract
Multiple risk factors contribute to cognitive impairment in children with b-thalassemia major. For a more refined understanding of
this issue, we attempted to evaluate cognitive function in b-thalassemia major patients and identify the relationship between
possible cognitive dysfunction and the following: demography, transfusion and chelation characteristics, iron overload, and disease
complications. We studied 100 b-thalassemia major children and 100 healthy controls who matched well in terms of age, sex, and
socioeconomic status. All participants underwent psychometric assessment using Wechsler Intelligence Scale for Children–Third
Edition, Arabic version. The mean Full-Scale IQ and Performance IQ of patients were significantly lower than those of controls,
whereas no significant difference was found for Verbal IQ. No significant relationship existed between IQ and any of the assessed
parameters. We concluded that Performance IQ, not Verbal IQ, was significantly affected in b-thalassemia major patients, but
there was no clear association between IQ and any of the parameters.

Keywords
thalassemia, cognitive, IQ

Received May 02, 2014. Received revised August 07, 2014. Accepted for publication August 14, 2014.

b-Thalassemia major, also known as Cooley’s anemia, is a
chronic, genetically determined hematologic disorder character-
ized by ineffective erythropoiesis, peripheral hemolysis, and
severe anemia.1 It is the most common chronic hemolytic anemia
in Egypt (85.1%), and its carrier rate has been estimated at 9% to
10.2% from an examination of 1000 normal random subjects from
different geographic areas of the country.2 Thalassemic patients
require regular red cell transfusion to eliminate anemia complica-
tions and compensatory bone marrow expansion.3
However, transfusions, combined with excessive iron
absorption, lead to iron deposition over various organs, princi-
pally the heart, liver, and endocrine glands. Such deposition
may ultimately lead to the death of the affected patient if left
untreated.4
Chronic hypoxic state, iron overload due to chronic blood
transfusion, and toxicity due to chelating agents may be associ-
ated with brain dysfunction in b-thalassemia major patients.5
In most cases, neurologic involvement in b-thalassemia
major patients does not initially present relevant signs and
symptoms (ie, subclinical); they are presented only during neu-
rophysiological and neuropsychological evaluation.6
Economou et al7 demonstrate subclinical involvement of
central and peripheral neural pathways in b-thalassemia major
patients and recommend regular neurophysiological tests, par-
ticularly intellectual monitoring using the Wechsler Intelli-
gence Scale, for early detection of any kind of intellectual
dysfunction in young b-thalassemia major patients.
Neuropsychological tests are safe and reliable for diagnosis
of cognitive impairment in b-thalassemia major patients, and
they may even facilitate early diagnosis.8 Wechsler Intelli-
gence Scale for Children–Third Edition is the most widely used
test for intelligence for school-age children and adolescents.9
We aimed at evaluating the cognitive function in b-thalasse-
mia major patients and identifying the relationship between
possible cognitive dysfunction and the following parameters:
demography, transfusion and chelation characteristics, iron
overload, and disease complications of the patients.
1
Department of Psychiatry, Zagazig University, Zagazig, Egypt
2
Department of Pediatrics, Zagazig University, Zagazig, Egypt
Corresponding Author:
Tamer Hassan, MD, Pediatrics department, Zagazig University, Zagazig 44111,
Egypt.
Email: dr.tamerhassan@yahoo.com

Downloaded from jcn.sagepub.com at Selcuk Universitesi on December 29, 2014

2 Journal of Child Neurology
Subjects and Methods
A cross-sectional study was conducted in the outpatient clinic
of the Pediatric Hematology Unit of Zagazig University Hospi-
tals during October 2012 to December 2013. One hundred chil-
dren with b-thalassemia major (60 males and 40 females), who
were in the age group of 6 to 16 years besides being free from
any other chronic medical illness, were consecutively enrolled
during their regular follow-up visits.
One hundred healthy controls, matching the patients in age,
sex, educational level, parental education, and socioeconomic
level, were also included in the study. It was ensured from their
past medical history and clinical examination that they and
their first-degree relatives never had any chronic disease,
including thalassemia.
Half of the patients and controls were aged 6 to 10 years and
half were aged 11 to 16 years.
All the patients were subjected to regular transfusion of
packed red cells at 3- to 4-week intervals to enable them main-
tain to their hemoglobin concentration at more than 9 mgm/dL
level. They were also receiving desferrioxamine subcutaneous
at a dose of 50 mg/kg/d alone or in combination with deferi-
prone oral at a dose of 75 mg/kg/d. Others received deferiprone
alone at a dose of 75 mg/kg/d. Compliance was calculated as
the percentage of medication doses taken divided by the num-
ber of prescribed doses.10
1. Collection of full medical history and complete data on
transfusion and chelation regimens.
2. Clinical examination with special emphasis on disease-
related complications and chelation therapy-related side
effects.
3. Routine laboratory and imaging investigations for tha-
lassemic patients according to international standards,
including complete blood picture, serum ferritin, liver
and kidney functions, blood glucose level, serum cal-
cium and phosphorus, T3, T4, TSH and sex hormones
and echocardiography.
All the patients and controls were subjected to psychometric
assessment using the Arabic version of the Wechsler Intelli-
gence Scale for Children–Third Edition,11 which provided the
output of their Verbal and Performance subtests and a com-
bined Full-Scale IQ test. The assessment included 6 Verbal
subtests and 6 Performance subtests, as detailed below:
Verbal IQ
This was derived from the scores obtained in the following
subtests: information, digit span, vocabulary, arithmetic, com-
prehension, and similarities. The information subtest was a test
of general knowledge, including geography and literature. The
digit span subtest required the child to repeat strings of digits
recited by the examiner. The vocabulary and arithmetic subt-
ests were general measures of the child’s vocabulary and arith-
metic skills. The comprehension subtest required the child to
solve practical problems and explain the meaning of simple
Downloaded from jcn.sagepub.com at Selcuk Universitesi on December 29, 2014
proverbs. The similarities subtest expected the child to describe
similarities between pairs of different items of the same cate-
gory, say apples and oranges, both of which belong to the cate-
gory of fruits.
Performance IQ
This was derived from the scores obtained in the following
subtests: picture completion, picture arrangement, block
design, object assembly, coding, and mazes. In the picture
completion subtest, the child was required to complete pictures
with missing elements. The picture arrangement subtest con-
sisted in arranging pictures in such an order as to tell a story.
The block design subtest required the child to use blocks for
making specific designs. The object assembly subtest expected
the child to put together pieces in such a way as to construct an
object. In the coding subtest, the child had to make pairs from a
series of shapes or numbers. The mazes subtest expected the
child to solve progressively more complex maze puzzles.
Scores on the Performance subtests were based not merely on
correct answers but also on the speed of response.
Administration Time
The administration time for the test was approximately 60 to 90
minutes. The child was allowed to complete the test in 2 sepa-
rate sessions.
Interpretation of IQ Score
The IQ was graded based on the following guidelines.
 130 and higher: very superior
 120-129: superior
 110-119: high average
 90-109: average
 80-89: low average
 70-79: borderline
 69 and lower: extremely low
Statistical Analysis
The data were checked, entered, and analyzed using SPSS ver-
sion 11 (SPSS Inc, Chicago, IL). Results were expressed as
mean + standard deviation for quantitative variables and as
number and percentage for qualitative ones. The significant
differences in means for IQ between patients and controls and
between males and females were analyzed by unpaired Student
t test. The relationship between IQ and disease complications
were also analyzed by unpaired Student t test. The relationship
between IQ and type of chelation were investigated using anal-
ysis of variance (F test). Correlation between IQ and other
parameters was investigated using Pearson coefficient of corre-
lation. P values .05 qualify as significant results and those
.001 as highly significant results.
Raafat et al 3

Ethics Table 1. Transfusion and Chelation Data and Disease Complications

of Patients.
This study was conducted in accordance with the ethical stan-
dards of the Helsinki Declaration of 1964, as revised in 2000, Patients (n ¼ 100)
and was approved by the institutional review board. Informed
Onset of blood transfusion 1.44 + 1.17 (0.5-5)
consent was obtained from all study participants. (y), mean + SD (range)
Onset of chelation therapy 3.72 + 1.87 (2-5)
(y), mean + SD (range)
Results Chelation type
No chelation 16 (16)
The mean age of patients—60 males and 40 females—is 10.1
Desferrioxamine 50 (50)
+ 3 years and that of controls—50 males and 50 females— Deferiprone 14 (14)
10.3 + 2.9 years. The patients and controls match well in terms Combined 20 (20)
of their age, sex, educational level, parental education, and (desferrioxamine and
socioeconomic status. The mean age of transfusion onset is deferiprone)
1.44 + 1.17 years and that of chelation onset 3.72 + 1.87 Compliance to chelation (%),
years. Only 84% of the patients received chelation therapy, mean + SD (range)
Desferrioxamine 64 + 24 (40-100)
50% on desferrioxamine, 14% on deferiprone, and 20% on
Deferiprone 81 + 24 (40-100)
combined desferrioxamine and deferiprone. Details regarding Combined 74 + 15 (40-100)
compliance with chelation therapy and disease complications (desferrioxamine and
are presented in Table 1. Details regarding educational level, deferiprone)
parental education, and socioeconomic status of patients and Disease complications, n (%)
controls are presented in Table 2. Growth retardation 26 (26)
The mean Full-Scale IQ (100.2 + 16.1) and Performance Hypogonadism 22 (22)
Hypoparathyroidism 6 (6)
IQ (83.9 + 14.7) of patients (P < .001) are significantly lower
Hypothyroidism 4 (4)
than those of controls (113.1 + 17.3 and 107.1 + 18), whereas Diabetes mellitus 4 (4)
there is no significant difference between patients and controls Cardiac complications 6 (6)
with regard to Verbal IQ (115.1 + 18 and 118.3 + 17, respec-
tively; P > .05; Figure 1). Abbreviation: SD, standard deviation.
Significant difference exists, between the patients and the
controls, in Performance IQ subtests (P < .001), but no such Table 2. Educational Level, Parental Education and Socioeconomic
difference exists in Verbal IQ subtests (P > .05; Tables 3 and Status of Patients and Controls.
4, respectively). Patients (n ¼ 100) Controls (n ¼ 100)
The distribution of patients and controls by grades of IQ is
shown in Table 5. Education level
No significant difference exists between male and female Illiterate 4 4
patients in Verbal IQ, Performance IQ, and Full-Scale IQ subt- Read and write only 4 4
Primary school 60 60
ests (P value .34, .71, and .6, respectively; Figure 2).
Preparatory school 22 22
No significant correlation exists between IQ (Verbal, Secondary school 10 10
Performance, and Full-Scale) and any of the following para- Father education level
meters: age, onset of transfusion, onset of chelation, chelation Illiterate 13 12
compliance, and serum ferritin levels (P > .05; Table 6). Also, Read and write only 17 15
no significant relationship exists between IQ and the type of Primary school 20 21
chelation therapy (P > .05; Figure 3). Preparatory school 14 13
Secondary school 10 11
Similarly, no significant relationship exists between IQ and
University or higher 26 28
any of the disease complications (P > .05; Table 7). Mother education level
Illiterate 19 21
Read and write only 16 13
Discussion Primary school 24 23
Preparatory school 17 18
Several previous workers demonstrated the involvement of ner-
Secondary school 9 9
vous system in b-thalassemia major patients. They attribute the University or higher 15 16
neurologic complications to various factors, such as chronic Socioeconomic level
hypoxia, bone marrow expansion, iron overload, and desfer- Low class 60 55
rioxamine neurotoxicity.6 Lower middle class 32 35
In our study, the mean Full-Scale IQ and Performance IQ Upper middle class 8 10
are significantly lower in patients than those in controls, High class 0 0
whereas no such difference exists in Verbal IQ between Abbreviation: SD, standard deviation.

Downloaded from jcn.sagepub.com at Selcuk Universitesi on December 29, 2014

4 Journal of Child Neurology

neuropsychological tests. Economou et al,7 using Wechsler

Intelligence Scale for Children–Third Edition, reported that
36.4% of b-thalassemia major patients had abnormal IQ. In
a similar Egyptian study,13 covering 100 b-thalassemia
major patients, statistically significant differences can be
seen between patients and controls in Full-Scale IQ, Verbal
IQ, and Performance IQ (respectively 81.5 and 74.1 Full-
Scale IQ, 79.9 and 73.8 Performance IQ, and 85.7 and
78.2 Verbal IQ). Their mean Performance IQ is lower than
mean Verbal IQ, but the difference is not statistically signif-
icant, as in our study.
The best explanation for lower mean Performance IQ in
Figure 1. IQ of patients and controls. The mean Full-Scale IQ and our study, as compared to that of Verbal IQ, comes from the
Performance IQ are significantly lower in patients than those in con- study of Ai et al14 on 171 Chinese children. They found that
trols, whereas in the case of Verbal IQ, no significant difference exists children with low Hb levels had significantly lower scores
between patients and controls. in Performance IQ but not in Verbal IQ. They attribute this
to the possibility that low Hb could have affected the portions
Table 3. Verbal IQ Subtests in Patients and Controls. of the brain that are associated with Performance IQ compo-
nents during the crucial stage of development in early child-
Patients Controls hood. This explanation can be applied to b-thalassemia
Verbal IQ (n ¼ 100), (n ¼ 100), t P major patients who suffered, early in their lives, from severe
subtest mean + SD mean + SD value value
anemia.
Information 10.4 + 3.6 10.5 + 3.7 0.19 .84 On the contrary, Logothetis et al,15 in their earlier work,
Digit span 10.7 + 3.9 10.9 + 4.1 0.35 .72 found that the IQ of patients with b-thalassemia major did not
Vocabulary 10.2 + 3.3 10.2 + 3.5 0.0 .99 deviate significantly from the expected norms. Using the Ravin
Arithmetic 9.5 + 3.2 9.9 + 3.3 0.87 .38 test, Karimi et al16 found no significant difference in IQ
Comprehension 10.3 + 3.5 11.1 + 3.4 1.63 .1
between b-thalassemia major patients and controls. Also,
Similarities 10.9 + 4.1 11.5 + 4.3 1.009 .3
Khairkar et al17 reported normal IQ in b-thalassemia major
Abbreviation: SD, standard deviation. patients.
Our results show no significant relationship between IQ
(whether Full-Scale IQ, Verbal IQ, or Performance IQ) and any
Table 4. Performance IQ Subtests in Patients and Controls.
of the following parameters: gender, age, onset of transfusion,
Patients Controls onset of chelation, chelation compliance, serum ferritin levels,
Performance (n ¼ 100), (n ¼ 100), t P type of chelation therapy, and disease complications. These
IQ subtest mean + SD mean + SD value value results agree with those of Economou et al,7 who found no cor-
relation between abnormal IQ in b-thalassemia major patients
Picture completion 7.0 + 3.6 10.1 + 3.8 5.92 <.001
Picture arrangement 6.5 + 3.4 9.0 + 3.6 5.04 <.001 and any of the assessed parameters, which included age, sex,
Block design 6.5 + 3.3 9.2 + 3.5 5.61 <.001 and serum ferritin levels.
Object assembly 6.8 + 3.2 9.9 + 3.6 6.43 <.001 Zafeiriou et al,6 in their review on neurologic complications
Coding 5.8 + 2.7 8.5 + 3.5 6.1 <.001 in b-thalassemia, reported that the neuropsychological studies
Mazes 6.0 + 2.8 8.5 + 3.3 5.77 <.001 available revealed a considerably high prevalence of abnormal
Abbreviation: SD, standard deviation. IQ, not correlating, however, to factors such as hypoxia or iron
overload. They proposed that factors associated with severe
chronic illness, rather than the disease per se, could be respon-
patients and controls. Our results partly match with those of sible for these findings. Such factors include regular school
other studies that document impairment of Full-Scale IQ absence due to transfusions and frequent hospitalizations,
(including both Verbal and Performance components) in physical and social restrictions resulting from the disease and
b-thalassemia major patients. Duman et al,8 evaluating cog- its treatment, abnormal mental state due to the awareness of
nitive function in 20 children with b-thalassemia major and being chronically ill, and, last, the overly protective family atti-
in 21 healthy controls, found that Full-Scale IQ, Verbal IQ, tude that leads to restricted initiative and psychosocial develop-
and Performance IQ (P < .05) were significantly lower in ment. Their proposal can be applied to explain the abnormal IQ
the patients. Monastero et al,12 in their study to assess cog- in our study that was not correlated to any of the assessed
nitive function in 46 b-thalassemia major patients and 46 parameters.
controls of matching age, sex, and education, found that the Also, Monastero et al12 reported no relationship between
b-thalassemia major patients, particularly those showing cognitive performances and signs of deferoxamine toxicity,
signs of hemosiderosis, were significantly impaired on all deferoxamine dosage, and levels of hemoglobin and serum

Downloaded from jcn.sagepub.com at Selcuk Universitesi on December 29, 2014

Raafat et al 5

Table 5. Distribution of Patients and Controls, Based on IQ Grades.

Very superior Superior High average Average Low average Borderline Extremely low

Verbal IQ
Patients 24 24 16 30 2 4 0
Controls 25 24 17 30 4 0 0
Performance IQ
Patients 0 0 4 34 30 18 14
Controls 8 20 20 44 6 2 0
Full-Scale IQ
Patients 4 4 20 44 20 4 4
Controls 12 23 26 34 4 1 0

Table 7. Relationship Between IQ and Disease Complications.

IQ,
mean + SD
n (range) t P

Growth retardation 1.39 .16

Yes 26 103.9 + 17.7
(67-130)
No 74 98.8 + 15.2
(62-133)
Hypogonadism 0.18 .85
Yes 22 100.7 + 15.9
(67-119)
Figure 2. IQ of patients in relation to sex. No significant difference No 78 100.1 + 16.1
exists between male and female patients as regards Verbal IQ, Per- (62-133)
formance IQ, and Full-Scale IQ. Diabetes mellitus
Yes 4 89.5 + 25.9 1.36 .17
(67-112)
Table 6. Correlation Between IQ and Other Parameters. No 96 100.6 + 15.5
(62-133)
Verbal IQ Performance IQ Full-Scale IQ Hypothyroidism 1.89 .06
Yes 4 85.5 + 21.3
r P r P r P (67-104)
No 96 100.8 + 15.6
Age (y) 0.17 >.05 0.121 >.05 0.02 >.05 (62-133)
Onset of transfusion (y) 0.17 >.05 0.11 >.05 0.14 >.05 Hypoparathyroidism 0.34 .73
Onset of chelation (y) 0.14 >.05 0.02 >.05 0.11 >.05 Yes 6 98 + 24
Compliance (%) 0.07 >.05 0.11 >.05 0.01 >.05 (67-115)
Serum ferritin (ng/mL) 0.14 >.05 0.09 >.05 0.14 >.05 No 94 100.3 + 15.5
(62-133)
Cardiac 1.23 .21
Yes 6 108 + 6.9
(103-112)
No 94 99.7 + 16.3
(62-133)
Abbreviation: SD, standard deviation.

ferritin. On the contrary, Logothetis et al15 found lower IQ

Figure 3. IQ of patients in relation to chelation type. No significant
relationship exists between IQ and type of chelation therapy (F ¼ 0.58,
P ¼ .62).
scores in b-thalassemia major patients with severe complica-
tions and less vigorous transfusions.
We conclude that Performance IQ, not Verbal IQ, is signifi-
cantly affected in b-thalassemia major patients, but no clear
association exists between IQ and any of the following para-
meters: gender, onset of transfusion, onset of chelation, chelation
compliance, serum ferritin levels, type of chelation therapy, and
disease complications. Cognitive assessment is feasible and

Downloaded from jcn.sagepub.com at Selcuk Universitesi on December 29, 2014

6 Journal of Child Neurology
should form part of routine comprehensive care of b-thalassemia
major patients.
Acknowledgments
The authors thank all the participants of this study for their unstinted
cooperation.
Author Contributions
NR, UES, NK, and TH selected the research theme and designed
the framework for study. TH, AS, and BH collected the clinical
data of thalassemic patients. NR, AY, and AES performed psycho-
metric assessment for all the study participants. All the authors
were involved in analyzing the data and drafting the paper. TH
did the final review of the manuscript and submitted it for
publication.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to
the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship,
and/or publication of this article.
Ethical Approval
The present study was conducted in accordance with the ethical stan-
dards of the Helsinki Declaration of 1964, as revised in 2000, and was
approved by the institutional review board. Informed consent was
obtained from all the study participants and their caregivers.
References
1. Galanello R, Origa R. Review: beta thalassemia. Orphanet J Rare
Dis. 2010;5:11.
2. El-Beshlawy A, Kaddah N, Rageb L, et al. Thalassemia preva-
lence and status in Egypt. Pediatr Res. 1999;45:760-760.
3. Fosburg M, Nathan D. Treatment of Cooley’s anemia. Blood.
1990;76:435-444.
4. Modell B, Khan M, Darlison M. Survival in beta-thalassemia
major in the UK: data from the UK Thalassaemia Register.
Lancet. 2000;355:2051-2052.
Downloaded from jcn.sagepub.com at Selcuk Universitesi on December 29, 2014
5. Sinniah D, Vignaendra V, Ahmad K. Neurological complications
of b-thalassemia major. Arch Dis Child. 1977;52:977-979.
6. Zafeiriou DI, Economou M, Athanasiou-Metaxa M. Neurologi-
cal complications in b-thalassemia. Brain Dev. 2006;28:
477-481.
7. Economou M, Zafeiriou D, Kontopoulos E, et al. Neurophysiolo-
gic and intellectual evaluation of beta thalassemia patients. Brain
Dev. 2006;28:14-18.
8. Duman O, Arayici S, Fettahoglu C, et al. Neurocognitive function
in patients with b-thalassemia major. Pediatr Int. 2011;53:
519-523.
9. Wechsler D. Wechsler Intelligence Scale for Children, Third
Edition. San Antonio, TX: Psychological Corporation; 1991.
10. Tanner MA, Galanello R, Dessi C, et al. A randomized, placebo-
controlled, double-blind trial of the effect of combined therapy
with deferoxamine and deferiprone on myocardial iron in thalas-
semia major using cardiovascular magnetic resonance. Circula-
tion. 2007;115:1876-1884.
11. Ismail ME, Melika LK. Wechsler Intelligence Scale for Children:
Arabic Manual. 7th ed. Cairo, Egypt: El-Nahda Press; 1999.
12. Monastero R1, Monastero G, Ciaccio C, et al. Cognitive deficits
in beta-thalassemia major. Acta Neurol Scand. 2000;102:
162-168.
13. Sabry N, Salama KH. Cognitive abilities, mood changes and
adaptive functioning in children with b thalassaemia. Curr
Psychiatry. 2009;16:244-254.
14. Ai Y, Zhao SR, Zhou G, et al. Hemoglobin status associate with
Performance IQ but not Verbal IQ in Chinese pre-school children.
Pediatr Int. 2012;54:669-675.
15. Logothetis J, Haritos-Fatouros M, Constantoulakis M, et al.
Intelligence and behavioral patterns in patients with Cooley’s
anemia (homozygous beta-thalassemia); a study based on 138
consecutive cases. Pediatrics. 1971;48:740-744.
16. Karimi M, Yarmohammadi H, Cappellini MD. Analysis of intel-
ligence quotient in patients with homozygous beta-thalassemia.
Saudi Med J. 2006;27:982-985.
17. Khairkar P, Malhotra S, Marwaha R. Growing up with the fami-
lies of b-thalassaemia major using an accelerated longitudinal
design. Indian J Med Res. 2010;132:428-437.