European Journal of Internal Medicine 54 (2018) 76–80

Contents lists available at ScienceDirect

European Journal of Internal Medicine

journal homepage: www.elsevier.com/locate/ejim

Review Article

β-Thalassemia heterozygote state detrimentally affects health expectation T

a a b b
Luciano Graffeo , Angela Vitrano , Salvatore Scondotto , Gabriella Dardanoni ,
Walter Sebastiano Pollina Addariob, Antonino Giambonaa, Massimiliano Saccoa,
Rosario Di Maggioa, Disma Rendaa, Federico Taorminac, Andrea Triveric, Massimo Attanasioc,
⁎
Christian Gluudd, Aurelio Maggioa,
a
Campus of Hematology Franco and Piera Cutino, AOOR Villa Sofia-V. Cervello, Palermo, Italy
b
Dipartimento Attività Sanitarie e Osservatorio Epidemiologico Assessorato Salute Regione Sicilia, Palermo, Italy
c
Dipartimento di Scienze Economiche, Aziendali e Statistiche, Università degli Studi di Palermo, Palermo, Italy
d
The Copenhagen Trial Unit, Centre for Clinical Intervention Research, Department 7812, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark

A R T I C LE I N FO A B S T R A C T

Keywords: Background: Thalassemia minor (Tm) individuals, are generally considered healthy. However, the prognosis of
β-Thalassemia carrier state Tm individuals has not been extensively studied.
Mortality The aim of this study was to evaluate the prognosis of Tm versus controls without β-thalassemia carrier state.
Heterozygote Methods: A total of 26,006 individuals seeking thalassemia screening at the AOOR Villa Sofia-V. Cervello,
Health expectation
Palermo (Italy) were retrospectively studied. Logistic penalised regression model was used to estimate risk of
Thalassemia minor
potential complications and survival techniques were used to study mortality.
Results: We identified a total of 4943 Tm and 21,063 controls. Tm was associated with significantly higher risks
of hospitalisation for cirrhosis (OR 1·94, 95% CI 1·30 to 2·90, p = 0·001), kidney disorders (OR 2·11, 95% CI 1·27
to 3·51, p = 0·004), cholelithiatis (OR 1·39, 95% CI 1·08 to 1·79, p = 0·010), and mood disorders (OR 2·08, 95%
CI 1·15 to 3·75, p = 0·015). No statistically difference in life expectancy between thalassemia minor and control
group was found (HR 1·090, 95% CI 0·777 to 1·555, p < 0·590; log-rank test p = .426).
Conclusion: This study shows that Tm affects the prognosis of Tm carriers regarding health expectation.
Probably, iron overload and anaemia for several years may be at the basis of these effects.

1. Introduction
β-thalassemia is a hereditary blood disorder decreasing hae-
moglobin synthesis. β-thalassemia is classified into three phenotypes,
depending on the severity of symptoms: 1) thalassemia major (TM)
(also known as Cooley's anaemia); 2) thalassemia intermedia (TI); and
3) β-trait state or thalassemia minor (Tm) [1]. In TM, signs and
symptoms appear within the first two years of life and patients develop
life-threatening anaemia (haemoglobin(Hb) < 7·0 g/dl) [1]. TM pa-
tients are transfusion-dependent the rest of their life. Over time, chronic
blood transfusions can lead to severe body iron accumulation that, in
spite of chelation treatments, may result in liver, cardiac, and en-
docrinological complications [2]. In TI, the signs and symptoms appear
after the first two year of life [1]. Affected individuals have mild to
moderate anaemia (Hb 7·0 to 9·0 g/dl) and they may need occasionally
blood transfusions [3]. The clinical consequences may include en-
docrinopathies, bone disease, thromboembolism, pulmonary hy-
pertension, cerebrovascular and neuronal damage, liver fibrosis or
cirrhosis, and increased risk of hepatocellular carcinoma [3, 4]. Tm
individuals are usually considered healthy and they do not need blood
transfusions because of this condition. They are even informed that
their heterozygote carrier state provides advantages, like protective
effects against malaria and coronary heart disease [5]. Moreover, Tm
individuals will not know if they are carriers unless they have a specific
blood test to detect this condition [6]. Tm has been described as a
potential risk factor for liver, kidney, metabolic, cardiovascular, neu-
rological, and vascular complications, although the findings remain
controversial [7–29]. As the prognosis of Tm individuals could be af-
fected by long-term enhanced iron absorption and by negative effects of
long-standing anaemia, the topic needs further study [8, 15, 29, 30].

Abbreviations: AIC, Akaike Information Criterion; CVD, Cardio-vascular disease; DHSEO, Department Health Services and Epidemiological Observatory; Hb, haemoglobulin; HDRS,
Hamilton Depression Rating scores; Hosp, hospitalization; ICD-9-CM, International Classification of Diseases 9th revision; IQR, Interquartile range; LIC, liver iron concentration; LPRM,
Logistic penalised regression model; OR, Odds ratio; TI, Thalassemia Intermedia; TM, Thalassemia major; Tm, Thalassemia minor
⁎
Corresponding author.
E-mail addresses: md.amaggio@gmail.com, aurelio.maggio@ospedaliriunitipalermo.it (A. Maggio).

https://doi.org/10.1016/j.ejim.2018.06.009
Received 17 January 2018; Received in revised form 10 May 2018; Accepted 7 June 2018
Available online 19 June 2018
0953-6205/ © 2018 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
L. Graffeo et al. European Journal of Internal Medicine 54 (2018) 76–80

Table 1 Table 2
Causes of hospital admissions for different complications captured as diag- Main characteristics of studied groups.
nosis codes of the hospital discharge records (SDO), according to the
Total Tm Control
International Classification of Diseases 9th revision (ICD-9-CM).
ICD-9-CM diagnosis codes Description Demographics
N (%) 26,006 4943 (19·00) 21,063 (81·00)
427 Arrhythmia Sex, Females, n (%) 17,438 (67·00) 2869 (58·00) 14,569 (69.17)
574 Cholelithiasis Age, mean ± sd 37·00 ± 11·10 39·00 ± 12·40 36·60 ± 10·70
571 Cirrhosis Age_Test, mean ± sd 31·40 ± 10·30 32·20 ± 11·90 31·20 ± 9·80
250 Diabetes Death, n (%) 152 (0·58) 51 (1·03) 101 (0·48)
296 Mood disorders Hematology, mean ± sd
580–589 Kidney diseases RBC 4·80 ± 2·23 5·63 ± 0·78 4·65 ± 2·37
414 Ischaemic cardiomiopathy Hb 13·05 ± 2·72 11·75 ± 1·57 13·28 ± 2·82
MCV 81·61 ± 10·45 65·11 ± 10·54 84·60 ± 7·13
MCH 27·51 ± 4·26 21·00 ± 2·51 28·69 ± 3·36
The aim of this paper is to evaluate the prognosis regarding hospi- RDW 13·71 ± 4·04 15·44 ± 1·52 13·40 ± 4·27
talisations and mortality in Tm individuals compared to a control group HbA2 3·29 ± 2·05 5·23 ± 2·40 2·94 ± 1·77

without β-thalassemia carrier state.

All data are referred to follow-up.

2. Materials and methods small cohorts or case reports of Tm individuals [7–30]. Therefore, we
assessed the risk of hospital admissions for arrhythmia, cholelithiasis,
Data were retrospectively studied. They refer to all subjects, regis- cirrhosis, diabetes, mood disorders, kidney diseases, and ischaemic
tered in our database, at the Regional Centre for Thalassemia, Azienda cardiomyopathy for the Tm and Control groups.
Ospedali Riuniti Villa Sofia-V. Cervello, Palermo (Italy) from January Table 2 describes demographics and haematological parameters of
2000 to July 2014. In particular, subjects were sent by the family the Tm and Control groups.
doctors following a national screening program aiming at preventing Table 3 shows the causes in which a statistical significant difference
hemoglobinopathies. Thalassemia screening was performed according was found in hospital admissions between the Tm and the Control
to standard recommendations, described elsewhere [6]. The whole group. The risks of hospital admissions for cholelithiasis, cirrhosis,
cohort of subjects was divided into individuals positive for the test (Tm) mood disorders, and kidney diseases were statistically significant
and individuals negative for the test (without β-thalassemia carrier higher in Tm group compared to the control group (Table 3).
state, called “Controls”). We determined prognosis as hospital admis- The Tm group did not seem to differ from the Control group re-
sions for different complications, according to the ICD-9-CM [31] garding hospital admissions for arrhythmia, diabetes, and ischaemic
(Table 1), from January 2004 until December 2013. Hospital admis- cardiomyopathy. The risk of hospital admissions for cholelithiasis, cir-
sions and mortality data were, kindly, provided by Department for rhosis, mood disorders, and kidney diseases increases with age and
Health Activities and Epidemiological Observatory (DHAEO) of Sicilian
Region, and they refer to people who have had hospitalisation or were Table 3
dead in Sicily or in other parts of Italy since DHAEO can match, via Logistic penalised regression model for hospitalisation for cholelitiasis, cir-
identity fiscal Code, information entered on the National Register. In- rhosis, mood disorders, or kidney diseases.
formation on hospital admission and mortality before January 2004, for
Covariates OR 95% CI p-Value
all individuals included in the study, were not available at the DHAEO
because data were not recorded. Cholelithiasis – 574 code
Descriptive analyses were presented as mean ± standard deviation, Intercept 0·003 (0·002;0·004) 0.000
median with IQR, and percentages. Since the two studied groups had SEX 0·612 (0·473;0·792) 0·000
GROUP 1·393 (1·083;1·792) 0·010
very similar distributions of the follow up, Logistic Penalised HOSP 1·937 (1·540;2·436) < 0·001
Regression Model (LPRM) was used to estimate the risk to develop AGE 1·031 (1·022;1·040) < 0·001
different complications [32] instead of a hazard model. These models,
Cirrhosis - 571
one for each complication, include the effect of group, sex, age, and Intercept 0·000 (0·000;0·000) < 2e-16
interaction between sex and age. The variable group was set to zero for SEX 12·000 (3·230;44·581) 0·000
the control group and 1 for the Tm group. Moreover, the model was also GROUP 1·943 (1·298;2·906) 0·001
HOSP 2·849 (1·857;4·370) < 0·001
used for evaluating the interaction of a single complication compared to
AGE 1·067 (1·046;1·090) < 0·001
hospitalisation for other causes (Hosp), following the ICD-9-CM classi- SEX×AGE 0·976 (0·952;1·002) 0·068
fication [31]. The variable hospitalisation was set to 1 when the in-
Mood disorders - 296
dividual have had at least one hospitalisation for causes that differ from Intercept 0·000 (0·000;0·001) < 0·001
the complication under study, and was set to zero otherwise. SEX 0·626 (0·312;1·257) 0·187
The choice of the best model was based on penalised AIC (Akaike GROUP 2·076 (1·149;3·750) 0·015
Information Criterion) [32]. HOSP 2·105 (1·242;3·569) 0·005
AGE 1·035 (1·016;1·054) < 0·001
Moreover, Kaplan-Meier curves and Cox regression model were used
SEX×AGE 0·329 (0·074;1·459) 0·143
to investigate difference in survival between the two groups of interest.
Kidney diseases 580–589
Intercept 0·000 (0·000;0·000) < 2e-16
3. Results SEX 2·022 (1·224;3·340) 0·006
GROUP 2·111 (1·270;3·509) 0·004
HOSP 6·231 (3·116;12·459) < 0·001
Overall, 26,006 individuals participated. A total of 4943 were Tm
individuals; the controls numbered 21,063 individuals without β-tha- AGE: age of patient in years; GROUP: No-Tm = 0. Tm = 1; HOSP: 0 = no
lassemia carrier state. The median time from inclusion and until cen- hospitalisation or no hospitalisation differently from the studied complication,
soring was 2348 days (IQR, 1359–3613 days). 1 = hospitalisation for other complication differently from the studied com-
Table 1 describes causes of hospital admissions for different com- plication; SEX: F = 0, M = 1; SEX×AGE: interaction effect between sex and
plications. These were selected according to the previous papers on age. OR = odds ratio.

77
L. Graffeo et al.
Fig. 1. Logistic penalised regression model estimated profiles for cirrhosis, kidney diseases, cholelithiasis and disorder of mood. (A.1) and (A.2) show risk profiles of
developing cirrhosis during the course of life stratified for Group (Tm compared to Ctr), sex, and hospitalisation; (B.1) and (B.2) show risk profiles of developing kidney diseases
during the course of life stratified for Group (Tm compared to Ctr), sex, and hospitalisation; (C.1) and (C.2) show risk profiles of developing cholelithiasis during the course of
life stratified for group (Tm versus Ctr), sex, and hospitalisation; (D.1) and (D.2) show risk profiles of developing disorder of mood during the course of life stratified for group
(Tm versus Ctr), sex, and hospitalisation.
hospitalisation for other causes (Fig. 1; Table 3).
However, the probability of hospital admissions for cirrhosis and
kidney diseases was most common in males, cholelithiasis in females,
and no sex difference was identified for mood disorders (Fig. 1;
Table 3).
Results on mortality were reported on Table 4. No statistically dif-
ference was found for the risk of death of the Tm compared to the
control group (HR 1·090, 95% CI 0·777 to 1·555, p < 0·590; log-rank
test p = .426; Table 3 and Fig. 2.
4. Discussion
These findings suggest, in a large cohort of Italian individuals, that
Tm is a risk factor for health expectation. Particularly, the Tm state
increases significantly the risk of hospitalisation for cirrhosis, kidney
disorders, cholelithiatis, mood disorders in comparison the control
group without β-thalassemia carrier state. Moreover, the probability of
hospital admissions for cirrhosis, kidney diseases, cholelithiasis, and
mood disorders increases with age and hospitalisation for other causes.
Instead, the hazard rate for death from Cox regression model was
not statistically significant. The lack of significance could be due to lack
of power in the LPRM analysis as the number of deaths is few.
The statistically significant increased risk of hospitalisation for cir-
rhosis may be related to long-term enhanced iron absorption in the Tm
compared to the control group, probably due to ineffective ery-
thropoiesis and erythroid hyperplasia [15]. The association between
Table 4
Cox Regression Model for evaluating the risk of death in the two studied groups.
Mortality- Results from Cox regression model
Covariates HR 95%CI p-Value
SEX 1·254 (0·909;1·729) 0·170
GROUP 1·090 (0·777;1·555) 0·590
HR = hazard ratio; 95% CI = 95% confidence interval.
78
European Journal of Internal Medicine 54 (2018) 76–80
increased liver iron concentration (LIC) and liver fibrosis is widely re-
ported [33, 34]. Moreover, the iron threshold necessary to induce fi-
brosis is modulated by non‑iron-related factors such sex and age [34].
Different reports suggested the existence of a kidney disorders in Tm
[18, 22, 26, 27]. In a cohort of 50 children with Tm, Sadeghi-Bojd and
colleagues [22] demonstrated tubulopathy. It has been proposed that
renal tubular dysfunction in adults with Tm may be due to haemolysis,
decrease erythrocyte life span, tubular iron deposition, and oxidative
lipid peroxidation [18, 22, 26, 27]. The increased risk of hospitalisation
for kidney diseases shown in our present study supports that kidney
diseases is more common in Tm.
Borgna-Pignatti and colleagues suggested that Tm women have
significantly higher prevalence of cholethiasis [28]. The detection of
increased risk of hospitalisation for cholelithiasis in females in our
present larger cohort supports that cholelithiasis is most common in Tm
women. The higher prevalence of cholelithiasis in women without Tm
in comparison with men is well known [35]. The higher prevalence of
gallstones in women in comparison with men was attributed to ex-
posure to oestrogens and progesterone [35]. Therefore, the increased
prevalence of hospitalisation for cholelithiasis in Tm women suggests as
Tm has additive effect in increasing risk for cholelithiasis in women.
Moreover, the detection that the risk of hospitalisation for cholelithiasis
increases with age may support that mild haemolytic anaemia, due to
mild, ineffective erythropoiesis for years, may be at the basis of this
complications.
The increased risk of hospitalisation for mood disorders of 2·08
(95% CI 1·15 to 3·75, p = 0·015) in the Tm group, shown in the present
study, is in accordance with Keşkek and colleagues [9]. They found
higher Hamilton Depression Rating scores (HDRS) in Tm individuals
than in the healthy controls (p < 0·001) [9]. The cause of mood dis-
orders in Tm individuals is not clear.
Co-morbidities are common in patients with hospitalisation for
cholelithiasis, cirrhosis, disorder of mood, kidney disorders as it was
suggested by the high hospital admissions for other causes in these
cohorts of patients.
We did not observe increased risks for arrhythmia and ischaemic
L. Graffeo et al.
Fig. 2. Kaplan-Meier estimated survival curves to evaluate life expectancy thalassemia minor subjects versus the controls. The risk of death seems to be no statis-
tically different between the two groups (Tm vs Ctr). HR = 1·090, 95% CI 0·777 to 1·555; p-value from log-rank test = 0.426.
cardiomyopathy hospitalisations in Tm in comparison to controls
(Arrhythmia: Tm = 26 (0·53%) compared to controls = 59 (0·28%); OR
1·34, 95% CI 0·84 to 2·13. Ischaemic cardiomyopathy: Tm = 8 (0·16%)
compared to controls = 27 (0·13%); OR 1·26, 95% CI 0·57 to 2·78). The
role of Tm state in increasing the risk for cardio-vascular disease (CVD),
due to anaemia and pro-atherogenic state, is controversial [13, 14, 17,
19–21, 23]. Sanrnak and colleagues [8] reported that Tm was asso-
ciated with an increased risk of CVD (hazard ratio [95% confidence
interval] of 1·41 [1·01 to 1·95]), whereas a case-control study on 1363
people by Hashemi and colleagues [19] found no statistically sig-
nificant difference in coronary artery disease risk. Moreover, Tm in-
dividuals with newly diagnosed hypertension have a better overall
cardiovascular risk factors and metabolic profile [14]. Our present
study, in a larger cohort of Tm individuals, seems to suggest as Tm is
not associated with increased risk for arrhythmia and ischaemic car-
diomyopathy complications.
The associations between insulin resistance or diabetes and Tm are
controversial [36–39]. Indeed, while some authors described relation-
ship between Tm and insulin resistance or diabetes, related with high
level of C-reactive protein [36, 37], others were not able to show this
association [38, 39]. Our present study was not able to detect an in-
creased risk of hospitalisation for diabetes (Tm = 25 (0·51%) compared
to controls = 75 (0·36%; OR 1·42, 95% CI 0·90 to 2·22). However, be-
cause patients with insulin resistance or diabetes may not need hospital
admission, this risk could be underestimated using hospitalisation as an
indicator of this complication.
The meaning of the increased hospitalisations for cholelithiasis,
cirrhosis, mood disorders, and kidney diseases in Tm individuals could
be considered less important because of the absolute low prevalence of
these complications (Table 3). However, it has been estimated that
approximately 7·0% of the world population are carriers of thalassemia
or haemoglobinopathies [29]. Therefore, the increased hospitalisations
for these complications in Tm may likely play significant roles for
health expectancy, health cost implications, and survival. Indeed, just
the cost of admission for cirrhosis in these patients could potentially
justify intervention to prevent morbidity and to minimise the human
and economical burdens [40]. The possibility of preventing today iron
body burden complications by use of oral chelation could make pre-
vention of some of these complications possible [41].
The major limitations of the present study are that these findings
have been collected retrospectively, although in the largest cohort of
individuals to date. Furthermore, the number of hospitalisations and
deaths were few and we may therefore both have produced type I and
type II errors. Moreover, we did not publish a protocol and data
79
European Journal of Internal Medicine 54 (2018) 76–80
management plan for our study. Therefore, large prospective studies
have to be conducted to confirm and reexamine our findings.
5. Conclusions
In conclusion, this large retrospective study shows as Tm β-tha-
lassemia carrier state, differently from Controls, affects health ex-
pectation and suggests as the presence of iron overload and anaemia,
extended for years, may be at the basis of this condition.
Author contributions
L. G. assisted with data collection, data analyses and statistical in-
terpretation; A.V. assisted with statistical design, data analyses, statis-
tical interpretation, and wrote the manuscript; S. S., G. D., F. T., and A.
T. assisted with data collection; A. G., M. S., R. DM., and D. R. con-
tributed to review the manuscript for important intellectual content; M.
A. assisted with statistical design; C. G. assisted with statistical analyses
and reviewed the manuscript for important intellectual content; A. M.
designed research questions, assisted with medical interpretation, and
wrote the manuscript.
Conflicts of interest
The authors declare no competing financial interests.
Acknowledgments
This study was supported by the Foundation Franco and Piera
Cutino (148785).
References
[1] Modell B, Berdoukas V. The clinical approach to thalassaemia. Grune & Stratton;
1984. (Front Cover, - Medical - 453 pages).
[2] Cunningham MJ, Maclin EA, Neufeld EJ, Cohen AR, Thalassemia Clinical Research
Network. Complications of β-thalassemia major in North America. Blood
2004;104(1):34–9.
[3] Borgna-Pignatti C, Marsella M, Zanforlin N. The natural history of thalassemia in-
termedia. Ann N Y Acad Sci 2010;1202:214–20.
[4] Musallam KM, Cappellini MD, Wood JC, Taher AT. Iron overload in non-transfu-
sion-dependent thalassemia: a clinical perspective. Blood Rev 2012;26(Suppl.
1):S16–9.
[5] Karetti M, Yardumian A, Karetti D, Modell B. Informing carriers of beta-thalassemia:
giving the good news. Genet Test 2004;8(2):109–13.
[6] Giambona A, Damiani G, Vinciguerra M, et al. Incidence of haemoglobinopathies in
Sicily: the impact of screening and prenatal diagnosis. Int J Clin Pract
L. Graffeo et al.
2015;69:1129–38.
[7] Brown DW, Giles WH, Croft JB. Hematocrit and the risk of coronary heart disease
mortality. Am Heart J 2001;142:657–63.
[8] Sarnak MJ, Tighiouart H, Manjunath G, et al. Anemia as a risk factor for cardio-
vascular disease in the Atherosclerosis Risk in Communities (ARIC) study. J Am Coll
Cardiol 2002;40:27–33.
[9] Keşkek SO, Kirim S, Turhan A, Turhan FG. Depression in subjects with beta-tha-
lassemia minor. Ann Hematol 2013;92:1611–5.
[10] Singh AN, Maguire J. Bipolar affective disorder and thalassemia minor–a genetic
linkage? Br J Psychiatry 1988;152:581.
[11] Harada H, Nakajima T, Inazawa J, Abe T. Bipolar affective disorder associated with
beta-thalassemia minor. Biol Psychiatry 1995;37:477–9.
[12] Marvasti VE, Dastoori P, Karimi M. Is beta-thalassemia trait a risk factor for de-
veloping depression in young adults? Ann Hematol 2006;85:873–4.
[13] Maioli M, Pettinato S, Cherchi GM, et al. Plasma lipids in beta-thalassemia minor.
Atherosclerosis 1989;75:245–8.
[14] Triantafyllou AI, Vyssoulis GP, Karpanou EA, et al. Impact of β-thalassemia trait
carrier state on cardiovascular risk factors and metabolic profile in patients with
newly diagnosed hypertension. J Hum Hypertens 2014;28:328–32.
[15] Piperno A, Mariani R, Arosio C, et al. Haemochromatosis in patients with beta-
thalassaemia trait. Br J Haematol 2000;111:908–14.
[16] Liaw DC, Kotkiewicz A, Kender MA. Acute splenic infarct in beta-thalassemia
minor: a novel combination of heterozygous beta-globin mutations with latent
phenotypes and the clinical implications. Hemoglobin 2009;33:262–8.
[17] Brankovic-Sreckovic V, Milic Rasic V, Djordjevic V, Kuzmanovic M, Pavlovic S.
Arterial ischemic stroke in a child with beta-thalassemia trait and methylente-
trahydrofolate reductase mutation. J Child Neurol 2007;22(2):208–10.
[18] Kang JH, Park BR, Kim KS, et al. Beta- thalassemia minor is associated with IgA
nephropathy. Ann Lab Med 2013;33:153–5.
[19] Hashemi M, Shirzadi E, Talaei Z, et al. Effect of heterozygous beta-thalassaemia
trait on coronary atherosclerosis via coronary artery disease risk factors: a pre-
liminary study. Cardiovasc J Afr 2007 May-Jun;18(3):165–8.
[20] Lai ME, Vacquer S, Carta MP, et al. Evidence for a proatherogenic biochemical
phenotype in beta thalassemia minor and intermedia. Acta Haematol
2011;126:87–94.
[21] Kırım S, Keşkek ŞÖ, Turhan A, Saler T. Is β-thalassaemia minor associated with
metabolic disorder? Med Princ Pract 2014;23:421–5.
[22] Sadeghi-Bojd S, Hashemi M, Naderi M, Shikhani S. Kidney function tests in children
with beta-thalassemia minor in Zahedan, southeast of Iran. Iran J Kidney Dis
2011;5:201–3.
[23] Namazi MR. Minor thalassemia as a protective factor against cerebrovascular ac-
cidents. Med Hypotheses 2002;59:361–2.
[24] Karimi M, Bagheri MH, Tahmtan M, Shakibafard A, Rashid M. Prevalence of he-
patosplenomegaly in beta thalassemia minor subjects in Iran. Eur J Radiol
2009;69:120–2.
80
European Journal of Internal Medicine 54 (2018) 76–80
[25] Bahar A, Kashi Z, Sohrab M, Kosaryan M, Janbabai G. Relationship between beta-
globin gene carrier state and insulin resistance. J Diabetes Metab Disord
2012;11(1):22.
[26] Oktenli C, Bulucu F. Renal tubular dysfunction in a patient with beta-thalassemia
minor. Nephron 2002;92:222–3.
[27] Prabahar MR, Jain M, Chandrasekaran V, Indhumathi E, Soundararajan P. Renal
tubular dysfunction with nephrocalcinosis in a patient with beta thalassemia minor.
Saudi J Kidney Dis Transpl 2008;19:964–8.
[28] Borgna-Pignatti C, Rigon F, Merlo L, et al. Thalassemia minor, the Gilbert mutation,
and the risk of gallstones. Haematologica 2003;88:1106–9.
[29] Premawardhena A, Arambepola M, Katugaha N, Weatherall DJ. Is the beta tha-
lassaemia trait of clinical importance? Br J Haematol 2008;141(3):407–10.
[30] Fargion S, Sampietro M, Cappellini MD. Thalassemias and their interactions with
hemochromatosis. In: Barton JC, Edwards CQ, editors. Hemochromatosis.
Cambridge, UK: Cambridge University Press; 2000. p. 435–41.
[31] Organizzazione Mondiale della Sanità. ICD-10, international classification of dis-
eases. 9th revision, clinical modification. Istituto Poligrafico e Zecca dello Stato;
1997.
[32] Firth D. Bias reduction of maximum likelihood estimates. Biometrika
1993;80:27–38.
[33] Angelucci E, et al. Effects of iron overload and hepatitis C virus positivity in de-
termining progression of liver fibrosis in thalassemia following bone marrow
transplantation. Blood 2002;100:17–21.
[34] Loréal O, Deugnier Y, Moirand R, et al. Liver fibrosis in genetic hemochromatosis.
Respective roles of iron and non-iron-related factors in 127 homozygous patients. J
Hepatol 1992;16(1–2):122–7.
[35] Lee JY, Keane MG, Pereira S. Diagnosis and treatment of gallstone disease.
Practitioner 2015;259(1783):15–9. (2).
[36] Tong PC, Ng MC, Ho CS, et al. C-reactive protein and insulin resistance in subjects
with thalassemia minor and a family history of diabetes. Diabetes Care
2002;25(8):1480–1.
[37] Bahar A, Kashi Z, Sohrab M, Kosaryan M, Janbabai G. Relationship between beta-
globin gene carrier state and insulin resistance. J Diabetes Metab Disord
2012;11(1):22.
[38] Kırım S, Keşkek ŞÖ, Turhan A, Saler T. Is β-thalassaemia minor associated with
metabolic disorder? Med Princ Pract 2014;23(5):421–5.
[39] Gozashti MH, Hasanzadeh A, Mashrouteh M, et al. Prevalence of metabolic syn-
drome in patients with minor beta thalassemia and its related factors: a cross-sec-
tional study. J Diabetes Metab Disord 2014;13(1):108.
[40] Vitor S, Marinho RT, Gíria J, Velosa J. An observational study of the direct costs
related to hospital admissions, mortality and premature death associated with liver
disease in Portugal. BMC Res Notes 2016;9(1):62.
[41] Maggio A, Filosa A, Vitrano A, et al. Iron chelation therapy in thalassemia major: a
systematic review with meta-analyses of 1520 patients included onrandomized
clinical trials. Blood Cells Mol Dis 2011;47(3):166–75.