RESEARCH LETTER

Subclinical autonomic dysfunction in patientswith beta-thalassemia

Elefterios Stamboulis • Nikoleta Vlachou • Konstantinos Voumvourakis • Athina Andrikopoulou •

Chrisa Arvaniti • Athanasios Tsivgoulis • Dimitrios Athanasiadis • Sotirios Tsiodras • Nikolaos Tentolouris •

Heleni Triantafyllidi • Marouso Drossou-Servou • Aphrodite Loutradi-Anagnostou • Georgios Tsivgoulis

Received: 22 July 2011 / Accepted: 8 November 2011 / Published online: 15 December 2011

� Springer-Verlag 2011

Abstract We electrophysiologically evaluated the auto-

nomic function (AF) in a consecutive series of patients

with beta-thalassemia and in normal individuals. Six

quantitative autonomic function tests (AFTs) were used:

tilt test, hand grip test and sympathetic skin response for

sympathetic function; R–R interval, inspiration–expiration

difference and 30/15 ratio for parasympathetic function.

The prevalence of impaired AF was higher in beta-thalas-

semia patients (13%, n = 5) than in control subjects (0%,

n = 0; p = 0.026). Subclinical autonomic dysfunction

appeared to be more prevalent in beta-thalassemia patients

compared to controls in our series. Further independent

validation of this finding is required in larger cohorts of

beta-thalassemia patients.

Keywords Autonomic dysfunction � Beta-thalassemia �Neuropathy � Cardiovascular reflexes

Introduction

Beta-thalassemia syndromes are a group of hereditary

disorders characterized by a genetic deficiency in the

synthesis of beta-globin chains. It has recently been shown

that cardiac autonomic function was impaired in a popu-

lation of beta-thalassemia patients with preclinical heart

disease [1]. Although there is increasing evidence indi-

cating that the peripheral nervous system may be affected

in beta-thalassemia [4], the ANS has not been systemati-

cally evaluated in patients with this hereditary hemato-

logical disorder. In view of the former considerations, we

aimed to electrophysiologically assess the autonomic

function in a consecutive series of patients with beta-thal-

assemia and in normal individuals.

E. Stamboulis � N. Vlachou � K. Voumvourakis �A. Andrikopoulou � C. Arvaniti � A. Tsivgoulis �G. Tsivgoulis (&)

Second Department of Neurology, Attikon Hospital,

University of Athens, Iras 39, 15344 Athens, Greece

e-mail: tsivgoulisgiorg@yahoo.gr

D. Athanasiadis

Vascular Unit, Third Department of Surgery,

Attikon Hospital, University of Athens, Athens, Greece

S. Tsiodras

Fourth Academic Department of Internal Medicine

and Infectious Diseases, Attikon University Hospital,

University of Athens Medical School, Athens, Greece

N. Tentolouris

First Department of Propaedeutic Medicine,

Athens University Medical School, Athens, Greece

H. Triantafyllidi

Second Department of Cardiology, Medical School,

Attikon Hospital, University of Athens, Athens, Greece

M. Drossou-Servou

Thalassemia Unit, General Hospital of Piraeus, Athens, Greece

A. Loutradi-Anagnostou

Prenatal Diagnosis Unit, Center for Thalassemia, Laikon General

Hospital, Athens, Greece

G. Tsivgoulis

Department of Neurology, Democritus University of Thrace,

University Hospital of Alexandroupolis, Alexandroupolis,

Greece

123

Clin Auton Res (2012) 22:147–150

DOI 10.1007/s10286-011-0154-2

Subjects and methods

Study population

We examined a consecutive series of patients with beta-

thalassemia diagnosed and followed up regularly at two

tertiary referral centers [4]. Patients with age over 60 years,

diabetes or other known possible causes of neuropathy

were excluded from the study. All included patients had

normal vitamin B12, folic acid, and serum protein elec-

trophoresis and immunofixation. They were also free of

cardiovascular disease as assessed by clinical history,

electrocardiography and echocardiography. No patient had

clinical evidence of visual or auditory dysfunction. The

control group consisted of healthy volunteers with no his-

tory of diabetes mellitus, or neurological or hematological

disorders. Fasting serum glucose was measured in cases

and controls. Patients with fasting serum glucose levels

[110 mg/dL (cutoff for impaired glucose tolerance) were

excluded from further evaluation.

Autonomic function testing (AFT)

All AFTs were performed in the laboratory under stan-

dardized conditions by the same two experienced investi-

gators who were blinded to the medical history of the

patients. All measurements were performed between 8.00

a.m. and 11.00 a.m. in a silent room at a temperature of

22–25�C. The subject lay on a tilt table in the supine

position and rested for at least 15 min before each inves-

tigation. Patients were not permitted to have food, alcohol,

caffeine or nicotine for at least 12 h before the study. All

medication was discontinued 12 h before the examination.

The sympathetic function was tested using the following

battery of tests: tilt test, hand grip test and sympathetic skin

response. The parasympathetic function was tested using

the R–R interval during deep breathing, the Valsalva ratio

and responses to standing. The methodology used by our

group for assessing autonomic dysfunction in our electro-

physiology laboratory has been previously described in

detail [3].

Nerve conduction studies

In all patients, the electrophysiological studies were per-

formed by the same investigator and standard procedures

were used as previously described [3, 4].

Statistical analyses

Comparisons between patients and controls were per-

formed using the v2 test, Fisher’s exact test, unpaired t test

and Mann–Whitney U test as indicated. The Statistical

Package for Social Sciences (SPSS Inc., version 10.0 for

Windows) was used for statistical analyses.

Results

We evaluated a consecutive series of 39 patients (19 men,

mean age 29.2 ± 8.5 years) with beta-thalassemia fulfill-

ing our inclusion criteria and 40 control subjects (12 men,

mean age 26.6 ± 8.0 years). A total of 34 patients had

homozygous beta-thalassemia and 5 had thalassemia

intermedia. The baseline characteristics of the study pop-

ulation are presented in Table 1.

The findings of AFTs in beta-thalassemia patients and

controls are summarized in Table 2. We documented three

patients with impaired sympathetic function in the thalas-

semia group (8%) and no subject with abnormal sympa-

thetic function in the control group (0%; p = 0.116 by

Fisher’s exact test). There were two patients with impaired

parasympathetic function in the thalassemia group (5%)

compared to none in the control group (0%; p = 0.241 by

Fisher’s exact test). The prevalence of impaired autonomic

function was higher in thalassemia patients (n = 5; 13%)

compared to controls (0%; p = 0.026 by Fisher’s exact

test).

Pathological nerve conduction studies indicating an

axonal sensorimotor neuropathy were more frequent in

patients than controls (49 vs. 0%; p \ 0.001 by Fisher’s

exact test; Table 2). After excluding patients with thalas-

semia intermedia (n = 5), we repeated all statistical anal-

yses and documented a higher prevalence of impaired

autonomic function in patients with thalassemia major

Table 1 Baseline characteristics in cases and controls

Characteristic Cases

(n = 39)

Controls

(n = 40)

p

Age, years (mean ± SD) 29.2 ± 8.5 26.6 ± 8.0 0.160

Male sex 49% 30% 0.110

Hematocrit % (mean ± SD) 28.7 ± 2.8 NA NA

Ferritin, ng/mL (median, IQR) 1686

(916, 2841)

NA NA

Intermediate phenotype 13% NA NA

HCV-Abs (?) % 18 NA NA

Splenectomy 54% NA NA

Transfusions (%) 87 NA NA

Duration of transfusions

(mean ± SD, years)

233 ± 5.6 NA NA

Desferrioxamine treatment (%) 74 NA NA

Duration of desferrioxamine

treatment (mean ± SD, years)

12.9 ± 5.6 NA NA

NA not applicable, IQR interquartile range

148 Clin Auton Res (2012) 22:147–150

123

(4/34; 12%) than controls (0/40, 0%; p = 0.040 by Fisher’s

exact test).

Beta-thalassemia patients with impaired (n = 5) and

normal (n = 34) autonomic function did not differ in terms

of age (p = 0.176), gender (p = 0.999), genotype (p =

0.517), ferritin (p = 0.736) and hematocrit (p = 0.512)

levels, splenectomy (p = 0.999), HCV-Abs (p = 0.999)

and duration of transfusion (p = 0.954) or desferrioxamine

(p = 0.441) treatment. Peripheral neuropathy tended to be

more prevalent (p = 0.182 by Fisher’s exact test) in thal-

assemia patients with impaired [80%; (4/5)] than with

normal [44%; (15/34)] autonomic function.

Discussion

Our study showed that autonomic function was subclini-

cally affected in approximately 1 out of 13 patients with

beta-thalassemia. However, the prevalence of impaired

parasympathetic and sympathetic function was 5 and 10%,

respectively, and this difference was not significant com-

pared to control individuals.

Our findings are in line with the results of a recent study

that evaluated autonomic functions of the heart using time

domain heart variability analysis in a group of beta-thal-

assemia patients with a preclinical stage of heart disease

[1]. The authors documented an impaired sympatho-vagal

activity and concluded that diminished heart rate variability

may be an underlying electrophysiological substrate

predisposing to cardiac dysfunction in patients with

beta-thalassemia. Similarly, QTc dispersion, which is a

measure of the duration of ventricular depolarization and

repolarization, was increased in a group of beta-thalasse-

mia patients without any clinical, electrocardiographic or

echocardiographic signs of cardiac disease [5]. Finally, a

recent study from Italy has suggested that intestinal

dysmotility may constitute the underlying factor for bile

stasis and pigment gallstone formation in patients with

beta-thalassemia major [2]. Interestingly, the investigators

documented an independent correlation between impaired

autonomic function and abnormal gallbladder and small

bowel motility in their study population.

Certain limitations of the present report need to be

acknowledged. First, the case–control design of the present

study does not allow us to establish a cause–effect rela-

tionship between autonomic dysfunction and beta-thalas-

semia major. Similarly, given the present sample size, we

should interpret with caution the trend toward a higher

prevalence of autonomic dysfunction in beta-thalassemia

patients with peripheral nervous system involvement. More

specifically, there were only five patients with abnormal

autonomic function tests. These findings indicate that the

degree of autonomic dysfunction in patients with beta-

thalassemia appears to be relatively mild and may be

associated with the presence of peripheral neuropathy.

Finally, subtle cardiac dysfunctions which may influence

cardiac responses to autonomic control are not necessarily

identified on the measures used (echocardiography and

electrocardiography) and therefore remain a plausible

explanation for the moderate autonomic impairment doc-

umented in patients with beta-thalassemia.

In conclusion, subclinical autonomic dysfunction

appears to be more prevalent in beta-thalassemia patients

compared to controls. Given the relative low prevalence of

impaired AF (13%) documented in our series, further

independent validation of this finding is required in larger

cohorts of beta-thalassemia patients.

Acknowledgments Dr Elefterios Stamboulis and Dr Georgios

Tsivgoulis had full access to all of the data in the study and take

responsibility for the integrity of the data and the accuracy of the data

analysis.

Conflict of interest There are no conflict of interest to disclose.

References

1. Gurses D, Ulger Z, Levent E, Aydinok Y, Ozyurek AR (2005)

Time domain heart rate variability analysis in patients with

thalassaemia major. Acta Cardiol 60:477–481

2. Portincasa P, Moschetta A, Berardino M, Di-Ciaula A, Vacca M,

Baldassarre G, Pietrapertosa A, Cammarota R, Tannoia N,

Palasciano G (2004) Impaired gallbladder motility and delayed

Table 2 Autonomic function tests and nerve conduction studies

(NCS) in cases and controls

Variable Cases

(n = 39)

Controls

(n = 40)

P*

Impaired sympathetic function,

% (n)

8 (3) 0 (0) 0.116

Tilt test (abnormal/normal) 2/36 0/37

Hand grip test (abnormal/normal) 0/39 0/1/39

SSR (abnormal/normal) 1/38 0/40

Impaired parasympathetic function,

% (n)

5 (2) 0 (0) 0.241

30/15 ratio (abnormal/borderline/

normal)

0/4/35 0/0/40

Valsalva ratio(abnormal/borderline/

normal)

1/14/24 0/0/40

I–E difference (abnormal/borderline/

normal)

1/10/28 0/1/39

Impaired autonomic function, % (n) 13 (5) 0 (0) 0.026

Abnormal NCS, % (n) 49 (19) 0 (0) \0.001

* Fisher’s exact test

Clin Auton Res (2012) 22:147–150 149

123

orocecal transit contribute to pigment gallstone and biliary sludge

formation in beta-thalassemia major adults. World J Gastroenterol

10:2383–2390

3. Stamboulis E, Katsaros N, Koutsis G, Iakovidou H,

Giannakopoulou A, Simintzi I (2006) Clinical and subclinical

autonomic dysfunction in chronic inflammatory demyelinating

polyradiculoneuropathy. Muscle Nerve 33:78–84

4. Stamboulis E, Vlachou N, Drossou-Servou M, Tsaftaridis P, Koutsis

G, Katsaros N, Economou-Petersen E, Loutradi-Anagnostou

A (2004) Axonal sensorimotor neuropathy in patients with beta-

thalassaemia. J Neurol Neurosurg Psychiatry 75:1483–1486

5. Ulger Z, Aydinok Y, Levent E, Ulger Z, Aydinok Y, Levent E

(2006) Evaluation of QT dispersion in beta thalassaemia major

patients. Am J Hematol 81:901–906

150 Clin Auton Res (2012) 22:147–150

123