The Journal of International Medical Research

2010; 38: 1374 1380

Update on the Incidence of Metamizole

Sodium-induced Blood Dyscrasias in
Poland
GW BASAK, J DROZD-SOKOOWSKA

AND

W WIKTOR-JEDRZEJCZAK

Department of Haematology, Oncology and Internal Diseases, The Medical University of

Warsaw, Warsaw, Poland
Metamizole sodium (metamizole) is a
popular non-opioid analgesic and a
common non-prescription product in
Poland. Controversy exists regarding the
level of risk of agranulocytosis or aplastic
anaemia associated with its use. Two
previous
pharmacovigilance
studies
conducted in Poland found the risk was
low. Twenty-four of the 25 haematology
centres that provide specialist care for the
30 million adults in Poland participated in
this prospective 12-month study. Twentyone cases of agranulocytosis, 48 of aplastic

anaemia, 15 of neutropenia and 11 of

pancytopenia were reported. Of these cases,
three (two agranulocytosis; one aplastic
anaemia) were judged as being possibly
related to metamizole. Crude estimates of
the rate of agranulocytosis and aplastic
anaemia associated with metamizole were
0.16 and 0.08 cases/million person-days of
use, respectively. Ongoing national safety
surveillance in Poland shows that, despite
the possibility of drug-induced blood
dyscrasias with metamizole, the risk is very
low.

KEY WORDS: METAMIZOLE SODIUM; ANALGESIC; AGRANULOCYTOSIS; APLASTIC

BLOOD DYSCRASIAS; PROSPECTIVE PHARMACOVIGILANCE

Introduction
Agranulocytosis and other blood dyscrasias,
such as aplastic anaemia, are rare but
potentially life-threatening adverse reactions
to many drugs.1,2 Over 100 drugs have been
implicated in the development of blood
dyscrasias, including some antithyroid drugs
(e.g. carbimazole and methimazole),
antibiotics (e.g. -lactam and cotrimoxazole),
antiplatelet agents (e.g. ticlodipine),
antipsychotics (e.g. clozapine), antiepileptics
(e.g. carbamazepine) and non-steroidal antiinflammatory drugs (e.g. indomethacin,
phenylbutazone and metamizole sodium

ANAEMIA;

[metamizole]).1 The observed reactions are

idiosyncratic and are not dose-dependent.1,2
Metamizole is an effective antipyretic and
analgesic drug3,4 but there is controversy
regarding the exact level of risk of
haematological toxicity associated with its
use.5 7 This is reflected in the notable
differences in regulatory policies of different
countries. While metamizole is prohibited in
the USA, UK, Denmark and Sweden, it is
widely available in Spain, Africa, Poland,
Bulgaria, Russia, India and Latin and South
America.3,5 Differences in the methods used
in the various pharmacovigilance studies

1374

Downloaded from imr.sagepub.com by guest on January 5, 2015

GW Basak, J Drozd-Sokoowska, W Wiktor-Jedrzejczak

Polish data on metamizole-induced blood dyscrasias
and concurrent drug use probably account
for the widely differing estimates of the risk
of blood dyscrasias associated with
metamizole.6 9
Although a process for spontaneous
reporting exists, the culture of adverse-event
reporting is not well established in Poland.
Consequently, little information is available
on the safety of medicinal products used in
the general population. Active safety
surveillance on metamizole was, however,
initiated in Poland in 1997, following
divergent reports on its safety and its
widespread use. Previous studies, covering
the periods 1997 2001 and 2002 2003,
examined the frequency of metamizoleinduced blood dyscrasias, specifically
agranulocytosis and aplastic anaemia, that
occurred at six haematology centres across
Poland.4,8 The purpose of the present study
was to confirm previous surveillance
findings and provide more comprehensive
data on the possible association between
metamizole use and the occurrence of blood
dyscrasias from almost the entire adult
population of Poland.

Patients and methods

STUDY DESIGN AND CENTRES
Haematology centres in Poland receive
primary and secondary referrals for serious
blood dyscrasias from their surrounding
regions. This prospective safety surveillance
study was performed between 1 April 2006
and 31 March 2007, and collected
information from 24 of the 25 centres that
provide haematology care for the entire
Polish adult population (approximately 30
million people; adults defined as 16 years
old for the purposes of this study). Six of
these centres had previously participated in
similar studies evaluating the frequencies of
agranulocytosis and aplastic anaemia
following metamizole use.4,8 One small
centre based in Gorzw Wielkopolski did not
participate because of insufficient staff.

PATIENTS
Prior to the start of the study, the diagnostic
criteria for different blood dyscrasias were
defined by the National Consultant in
Haematology (Table 1) and a physician was
assigned in each participating centre to

TABLE 1:
Diagnostic criteria for blood dyscrasias defined by the National Consultant in
Haematology, Poland, for a prospective safety surveillance study of metamizole sodium
use performed between 1 April 2006 and 31 March 2007
Blood dyscrasia

Definition

Agranulocytosis

Granulocyte count < 0.5 109/la

Patient does not meet the criteria for aplastic anaemia
Granulocyte count < 1.5 109/l, platelets < 100 g/l, reticulocytes < 0.5%a
Bone marrow biopsy: fat-rich with haematopoietic cells occupying < 25% of
the marrow space
Granulocyte count < 1.5 109/la
Patient does not meet the criteria for aplastic anaemia
Granulocyte count < 1.5 109/l, platelets < 100 g/l, haemoglobin level
< 90 g/la
Patient does not meet the criteria for aplastic anaemia

Aplastic anaemia

Neutropenia
Pancytopenia

aTwo

measurements from separate blood samples.

1375

Downloaded from imr.sagepub.com by guest on January 5, 2015

GW Basak, J Drozd-Sokoowska, W Wiktor-Jedrzejczak

Polish data on metamizole-induced blood dyscrasias
collect and report data. Irrespective of
previous therapy, all patients were screened
for agranulocytosis, aplastic anaemia,
neutropenia and pancytopenia. Those on
cancer chemotherapy, radiation therapy or
immunosuppressive drugs, and those with
diseases that may have caused blood
dyscrasias (e.g. tumour infiltration of the
bone marrow) were excluded from the
analysis. At diagnosis of the blood dyscrasia,
demography and medical history including
any medication taken in the previous 3
months were obtained, including the use of
metamizole in the 8 weeks prior to diagnosis.
All cases in which metamizole had been used
were reported within 15 days to the Centre
for the Monitoring of Drug Side-effects of the
Pharmaceutical Institute, in Warsaw.

ASSESSMENTS
The frequency of each of the blood dyscrasias
potentially judged to be associated with the
use of metamizole was calculated by
dividing the total adult population
(approximately 30 million people) by the
number of reports of each disorder.
Polpharma company internal sales data on
tablets sold during the study period and the
defined daily dose of 3 g (i.e. 6 500 mg
tablets) were used to calculate the number of
person-days of use, on the assumption that
the entire adult population had been
exposed to the tablet formulation.

Results
During the 12 months of the study, 21 cases
of agranulocytosis were reported and the
incidence was calculated to be 0.7 cases per
million adults per year. Metamizole had
been used in the previous 8 weeks in only
two cases. Among the 48 cases of aplastic
anaemia (1.6 cases per million adults per
year), metamizole had been previously used
in only one case. Among the 15 cases of

neutropenia (0.5 cases per million adults per

year) and 11 cases of pancytopenia (0.37
cases per million adults per year)
metamizole had been used in one case of
neutropenia and one case of pancytopenia.
Of the five cases of blood dyscrasia
identified in which metamizole had been
used, two were in male and three were in
female patients. Their ages ranged from 16
to 82 years, and all had been receiving
concurrent medication at the time of the
reports. Four of the patients recovered fully
but one patient with agranulocytosis died. In
three of the five cases (two cases of
agranulocytosis and one of aplastic
anaemia), metamizole causality was
categorized as possibly related to therapy.
The two patients with agranulocytosis were
also receiving long-term therapy with
carbamazepine (an agent that has also been
reported to be associated with blood
dyscrasias1) and the patient with aplastic
anaemia was a chronic glue-sniffer. The two
remaining reports were categorized as
unrelated to metamizole therapy.
The five cases of blood dyscrasia identified
in which metamizole had been used are now
discussed.

CASE 1: AGRANULOCYTOSIS
An 82-year old man with trigeminal
neuralgia was admitted to hospital with
fever and dysphagia resulting from fungal
mucositis. He had previously taken
carbamazepine chronically (300 mg/day),
ketoprofen, diclofenac and a few tablets of
metamizole during the previous 8 weeks. The
patient died of septic shock 5 weeks after
admission to hospital and the event was
judged by the reporting physician to be
possibly related to metamizole.

CASE 2: AGRANULOCYTOSIS
A 62-year old woman with cervical spinal

1376

Downloaded from imr.sagepub.com by guest on January 5, 2015

GW Basak, J Drozd-Sokoowska, W Wiktor-Jedrzejczak

Polish data on metamizole-induced blood dyscrasias
neuralgia was admitted to hospital because
of pharyngitis and fever. She had taken
carbamazepine (800 mg/day) for 24 days
and metamizole for several preceding days
(15 500 mg tablets in total). Normalization
of her peripheral blood cell count was
achieved following antibacterial, antifungal
and
haematopoietic
therapy.
Her
agranulocytosis was judged to be possibly
related to metamizole.

CASE 3: APLASTIC ANAEMIA

500 mg tablets) because of gastrointestinal

tract infection, headache and fever.
Investigations led to a diagnosis of coeliac
disease. After administration of folic acid
and vitamin B12, her blood count returned to
within normal values. Pancytopenia was
judged to be unrelated to metamizole in this
patient.

OVERALL FREQUENCY OF
METAMIZOLE-RELATED BLOOD
DYSCRASIAS

A 21-year old man was admitted to hospital

because of fatigue lasting > 1 month. Four
weeks before admission he had an upper
respiratory tract infection, and as a
symptomatic treatment took metamizole
(two tablets, 500 mg each) and paracetamol
(two tablets, 500 mg each). He was a chronic
glue-sniffer, therefore exposure to toxic
substances could not be ruled out as the
main cause of the blood dyscrasia. While
highly unlikely, the blood dyscrasia was
judged to be possibly related to metamizole.

A total of 75 475 180 tablets of 500 mg

metamizole were sold during the study
period, equivalent to 12 579 196 person-days
of use. Thus, assuming that the study
involved almost 100% of the adult
population of Poland, the three reports of
possible metamizole-related blood dyscrasias
resulted in a crude estimate of 0.16 cases of
agranulocytosis per million person-days of
use and of 0.08 cases of aplastic anaemia per
million person-days of use.

CASE 4: NEUTROPENIA

The observed overall frequency of aplastic

anaemia (1.6 cases per million adults per
year) was similar to that reported in a
previous Polish study (1.8 per million adults
per year) and within internationally
accepted values (0.7 4.1 per million adults
per
year).2,4
The
frequency
of
agranulocytosis (0.7 cases per million adults
per year) was lower than in a previous
prospective study from Poland (1.07 per
million adults per year)4 and lower than
data from several international studies (1.1
4.9 per million individuals per year).2 It was
similar to data from Thailand (0.7 per
million individuals per year)9 and from Latin
America (0.4 per million individuals per
year).7 Although one of the 25 Polish
haematology centres did not wish to
contribute to the study, its exclusion is

Neutropenia and vitamin B12 deficiency were

reported in a 25-year old woman admitted to
hospital with fever, headaches, malaise and
fatigue. Within the previous 4 days she had
taken two tablets of metamizole (500 mg
each), three tablets of ibuprofen (200 mg
each) and three tablets of paracetamol (500
mg
each).
Following
vitamin
B12
supplementation, her blood count returned
to within normal values. The neutropenia
was judged to be unrelated to metamizole.

CASE 5: PANCYTOPENIA
Pancytopenia was reported in a 16-year old
female with gluten intolerance. Three weeks
before admission to hospital she had taken
metamizole (two 500 mg tablets), ibuprofen
(five 200 mg tablets) and paracetamol (two

Discussion

1377

Downloaded from imr.sagepub.com by guest on January 5, 2015

GW Basak, J Drozd-Sokoowska, W Wiktor-Jedrzejczak

Polish data on metamizole-induced blood dyscrasias
unlikely to have affected the overall findings
since it was a small centre covering a small
catchment area.
None of the reports of neutropenia or
pancytopenia following the use of
metamizole was judged to be treatmentrelated. Two cases of agranulocytosis in
patients
also
receiving
regular
carbamazepine therapy were, however,
judged to be possibly related to metamizole.
Relating
these
reports
to
drug
consumption data over the same period, a
crude estimate of the incidence of
agranulocytosis associated with the use of
metamizole in Poland was calculated as 0.16
cases per million person-days of use. This
value is similar to data from a previous study
conducted in Poland (0.2 cases per million
person-days of use)8 and data from several
international studies (0.3 4.0 per million
persons-days of use),6,10 but much lower than
data from Sweden (one case in 1400
individuals).11 The one case of aplastic
anaemia in the present study that was
judged possibly related to metamizole was
reported in a patient with chronic substance
abuse (glue-sniffing). As a crude estimate,
this equates to 0.08 cases per million persondays of use, a value lower than observed
previously in data from Poland (0.25 cases
per million persons-days of use).4
Poland is one of several countries where
metamizole is frequently used (over 75
million tablets per year). In Latin America,
metamizole one of the most widely used
drugs7 and in this region its association with
agranulocytosis appears to be minimal.7
Although differences in the methods used in
the various pharmacovigilance studies as
well as differences in age range, gender,
underlying conditions and concurrent drug
use of the patients may be responsible for
some of the disparity, some authors have
suggested that genetic polymorphisms may

be more prevalent in some ethnic groups

than in others.7 For example, data from a
Bulgarian study have suggested that the
histocompatibility
antigen
(human
lymphocyte antigen [HLA]) A24 is involved
in the predisposition to metamizole-induced
agranulocytosis.12 An association between
HLA-B27 and HLA-B38 and increased risk of
agranulocytosis with clozapine has also been
reported.13 Patients who receive clozapine
need to have their full blood counts
monitored regularly and, if the total white
cell and/or neutrophil counts indicate
agranulocytosis,
clozapine
must
be
terminated.14 Testing the involvement of a
genetic link to this, however, would require
DNA sampling and assessment of each
patient. While further study into ethnic
group differences and concomitant drug use
is warranted, a pragmatic solution is to
maintain pharmacovigilance.
In conclusion, these data show that,
despite the possibility of drug-induced blood
dyscrasias with the use of metamizole, these
blood disorders are rare events and the drug
may be considered relatively safe. Until
adverse event reporting in Poland becomes a
more reliable source of safety data, however,
metamizole pharmacovigilance should be
continued. Consideration should also be
given
to
introducing
similar
pharmacovigilance
programmes
for
carbamazepine and other agents potentially
associated with blood dyscrasias.

Acknowledgements
The authors wish to thank the following
individuals for their participation in this
study (all based in Poland): Piotr
Centkowski, Institute of Haematology and
Transfusion Medicine, Warsaw; Waldemar
Tomczak, Department of Haematooncology
and Bone Marrow Transplantation, Medical
University of Lublin, Lublin; Anna

1378

Downloaded from imr.sagepub.com by guest on January 5, 2015

GW Basak, J Drozd-Sokoowska, W Wiktor-Jedrzejczak

Polish data on metamizole-induced blood dyscrasias
Szmigielska-Kaplon,
Department
of
Haematology, Medical University of Lodz,
Lodz; Malgorzata Pawlowska, Department of
Haematology, Medical University of
Bialystok, Bialystok; Teresa BorysewiczCzajka, Department of Internal Diseases and
Haematology, Military Institute of Medicine,
Warsaw; Mariusz Janczarski, Department of
Haematology and Oncology, Central
Hospital of the Ministry of Internal Affairs,
Warsaw;
Malgorzata
Wojciechowska,
Haematology Ward, Voivodship Specialized
Hospital, Olsztyn; Malgorzata Calbecka,
Haematology Ward, Specialized Hospital,
Torun;
Jolanta
Starzak-Gwozdz,
Haematology Ward, Voivodship Specialized
Hospital, Rzeszow; Dorota Krochmalczyk,
Department of Haematology, Collegium
Medicum, Jagiellonian University, Cracow;
Andrzej Zdunczyk, Haematology Ward,
Ludwik Rydygier Voivodship Specialized
Hospital, Cracow; Marek Kielbinski,
Department of Haematology, Medical
University of Wroclaw, Wroclaw; Aleksandra
Holowiecka-Goral,
Department
of
Haematology, Silesian Medical University,
Katowice; Marcin Wojtowicz, Haematology
Ward, Voivodship Hospital, Opole; Maria

Bieniaszewska, Department of Haematology

Medical University of Gdansk, Gdansk;
Wojciech Pietras, Department of Paediatric
Haematology, Medical University of Wroclaw,
Wroclaw; Joanna Rupa, Department of
Haematology, Medical University of Poznan,
Poznan; Agnieszka Kasprzyk, Haematology
Ward, J. Strus Hospital, Poznan; Maria
Nowakowska-Domagala,
Haematology
Ward, Swietokrzyskie Oncology Centre,
Kielce; Grazyna Gadomska, Haematology
Ward, Jan Biziel Voivodship Hospital,
Bydgoszcz; Piotr Adamek, Ward of Internal
Diseases, Military Hospital, Szczecin; Barbara
Zdziarska, Department of Haematology,
Pomeranian Medical University, Szczecin;
Andrzej Lange, Low-Silesian Centre for
Cellular Transplantation, Wroclaw. The
authors also wish to acknowledge the
editorial help provided by Dr Sue Daly and
Dr Yvonne Lis during the preparation of the
manuscript. The study was supported by
Polpharma Pharmaceuticals, Starogard
Gdansk, Poland

Conflicts of interest
The authors had no conflicts of interest to
declare in relation to this article.

Received for publication 17 February 2010 Accepted subject to revision 4 March 2010
Revised accepted 8 June 2010
Copyright 2010 Field House Publishing LLP
References
1 Garbe E: Non-chemotherapy drug induced
agranulocytosis. Expert Opin Drug Saf 2007; 6:
323 335.
2 Kaufman DW, Kelly JP, Issaragrisil S, et al:
Relative incidence of agranulocytosis and
aplastic anemia. Am J Hematol 2006; 81: 65
67.
3 Ramacciotti AS, Soares BG, Atallah AN:
Dipyrone for acute primary headaches.
Cochrane Database Syst Rev 2007, 2: CD004842.
4 Maj S, Centkowski P: A prospective study of the
incidence of agranulocytosis and aplastic
anemia associated with the oral use of
metamizole sodium in Poland. Med Sci Monit
2004; 10: PI93 PI95.

5 Arellano F, Sacristn JA: Metamizole:

reassessment of its therapeutic role. Eur J Clin
Pharmacol 1990; 38: 617 619.
6 Ibez L, Vidal X, Ballarn E, et al:
Agranulocytosis associated with dipyrone
(metamizol). Eur J Clin Pharmacol 2005; 60: 821
829.
7 Hamerschlak N, Maluf E, Biasi Cavalcanti A, et
al:
Incidence
and
risk
factors
for
agranulocytosis in Latin American countries
the Latin Study: a multicenter study. Eur J Clin
Pharmacol 2008; 64: 921 929.
8 Maj S, Lis Y: The incidence of metamizole
sodium-induced agranulocytosis in Poland. J
Int Med Res 2002; 30: 488 495.
9 Shapiro S, Issaragrisil S, Kaufman DW, et al:

1379

Downloaded from imr.sagepub.com by guest on January 5, 2015

GW Basak, J Drozd-Sokoowska, W Wiktor-Jedrzejczak

Polish data on metamizole-induced blood dyscrasias

Agranulocytosis in Bangkok, Thailand: a

predominantly drug-induced disease with an
unusually low incidence. Aplastic Anemia
Study Group. Am J Trop Med Hyg 1999; 60: 573
577.
10 Risks of agranulocytosis and aplastic anemia.
A first report of their relation to drug use with
special
reference
to
analgesics.
The
International Agranulocytosis and Aplastic
Anemia Study. JAMA 1986; 256: 1749 1757.
11 Hedenmalm K, Spigstet O: Agranulocytosis and
other blood dyscrasias associated with dipyrone
(metamizole). Eur J Clin Pharmacol 2002; 58: 265

274.
12 Vlahov V, Bacrecheva N, Tontcheva D, et al:
Genetic factors and risk of agranulocytosis from
metamizol. Pharmacogenetics 1996; 6: 67 72.
13 Andrs E, Zimmer J, Affenberger S, et al:
Idiosyncratic drug-induced agranulocytosis:
update of an old disorder. Eur J Intern Med 2006;
17: 529 535.
14 Essali A, Al-Haj Haasan N, Li C, et al: Clozapine
versus typical neuroleptic medication for
schizophrenia. Cochrane Database Syst Rev 2009;
1: CD000059.

Authors address for correspondence

Dr Wieslaw Wiktor-Jedrzejczak
Department of Haematology, Oncology and Internal Diseases, The Medical University of
Warsaw, 1A Banacha Str., 02-097 Warsaw, Poland.
E-mail: ylis@btinternet.com

1380

Downloaded from imr.sagepub.com by guest on January 5, 2015