gms | German Medical Science

27. Jahrestagung der Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie (GAA)

Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie

26.11.2020, digital

Metamizol-induced agranulocytosis in Germany – case-control study based on secondary data of persons insured by compulsory sickness funds

Meeting Abstract

Suche in Medline nach

  • corresponding author presenting/speaker Eberhard Greiser - Universität Bremen, Bremen, Germany
  • Claudia Greiser - Epi.Consult GmbH, Musweiler, Germany

Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie e.V. (GAA). 27. Jahrestagung der Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie. sine loco [digital], 26.-26.11.2020. Düsseldorf: German Medical Science GMS Publishing House; 2020. Doc20gaa07

doi: 10.3205/20gaa07, urn:nbn:de:0183-20gaa073

Veröffentlicht: 23. November 2020

© 2020 Greiser et al.
Dieser Artikel ist ein Open-Access-Artikel und steht unter den Lizenzbedingungen der Creative Commons Attribution 4.0 License (Namensnennung). Lizenz-Angaben siehe http://creativecommons.org/licenses/by/4.0/.


Gliederung

Text

Summary: A case-control study has been performed, based on a population of 1,159,257 persons from 8 German health insurance funds, where secondary data were used. The aim of the study was to investigate the impact of metamizole on the development of agranulocytosis. The data base for the case-control study were derived after excluding all cases with a diagnosis of agranulocytosis and a diagnosis of any cancer or previous prescription of immunosuppressants the final multivariate regression could be performed on 267 men and women with agranulocytosis. The final results presented odds ratios (OR), differing with age: <14 years: 11.82 (95% CI 2.84–49.19) / 15–39 years: 2.54 (95% CI 1.05–6.13) / 40–59 years: 2.17 (96% CI 1.11–4.27) / 60–79 years: 2.64 (95% CI 1.64–4.25) / 80+ years: 2.30 (95% CI 1.50–3.51).

Introduction: For decades, agranulocytosis has been known as a possible side-effect of dipyrone. Several reports on the development of agranulocytosis after medication with metamizole have been published since 1960. Subsequently dipyrone has been withdrawn from the market in numerous countries, e.g. Australia in 1965, Norway, USA in 1977, Denmark in 1979, Saudi-Arabia 1980, Malaysia 1987, Pakistan 1988 [1]. A study [2] conducted in several countries evaluated the risk of agranulocytosis after medication of metamizole. This study described a risk increase of 23.7 (95% CI 8.7–64.4) for several regions, including Germany and Spain. The Drug Commission of the German Medical Association [3] reported in 2009 that between 1990 and 2010, about 300 cases of agranulocytosis had been reported after application of metamizole. The German Federal Institute for Drugs and Medicinal Products [4] informed the medical community in 2009 that “a possible side effect of treatment with metamizole could be agranulocytosis with possible death as consequence. However, because there are different studies on the prevalence of this side effect, no specific statement on the extent of risk can be made.”

The aim of this study was to analyze the risk of agranulocytosis due to medication with metamizole, based on secondary data from several German health insurance funds.

Materials and methods: Study population: Two case-control studies [5], [6] had been conducted in the vicinity of Cologne-Bonn airport between 2004 and 2010 to investigate the effects of environmental noise on persons exposed to noise. The first of these studies analyzed the effects of environmental noise on prescription of therapeutic drugs, the second one analyzed in addition effects of hospital-treated diseases. Eight compulsory sickness funds contributed data on all insured persons living in the study area. Data provided comprised: age, sex, insurance period, all prescriptions, issued by practicing physicians during the insurance period, all hospital discharge diagnoses (Table 1 [Tab. 1]). The data comprised 1,159,257 persons contributing 3,402,715 person years. This is an equivalent of 72.2% of the total population of the study region (Cologne city, and two adjacent counties (Rheinisch-Bergischer Kreis, Rhein-Sieg-Kreis). The data base developed in these studies provided data for this study.

For the purpose of this study, prescriptions for metamizole-containing drugs were abstracted, using the following: ATC codes M01BA07, N02BB02, N02BB52, N02BB72, R05XA07 [7]. Cases with a discharge diagnosis of agranulocytosis were abstracted from the discharge diagnoses database, containing an ICD10-code of D70.

Statistical analysis: Cases were defined as all persons with a discharge diagnosis of agranulocytosis. Factors included into the statistical analyses were: age, gender, length of insurance (years), number of prescriptions of metamizole-containing drugs. As confounding factors were included percentage of persons on social welfare per community or district of communities, percentage of persons aged 65 or older in nursing homes in communities. Logistic regressions were performed using the SAS-9.2 procedure Phreg. As preliminary analyses showed near identical results for both genders, final analyses were performed on both genders combined. As major risk factors for the development of agranulocytosis besides metamizole-containing drugs the application of immunosupressants as well as ionizing radiation are well known, cases were excluded from final analyses with prescriptions of immunosuppressants. As there were no data available on exposure to ionizing radiation, either due to diagnostic or therapeutic procedures, all cases with a discharge diagnoses of cancer of any kind were excluded, too. Thus the analyses comprised persons exposed to metamizole as the single agranulocytosis-inducing factor.

Results: The exclusion of cases with prescriptions of immunosuppressants and with any malignancy leads to a considerable decrease of available cases for the final analysis (Table 2 [Tab. 2]); as from 1,889 cases with agranulocytosis only 299 remain. This has a major effect on metamizole-induced agranulocytosis (Table 3 [Tab. 3]). The table shows results by broad age groups. There is a considerable difference regarding broad age groups: highest OR are found in the youngest age group (11.8 (CI 2.84–49.20)) with a sharp decline in older age groups. However, the mean of prescriptions in the youngest is lowest compared to the oldest age group (1.19 vs. 4.24). This is a clear indication that the high OR cannot be a result of dosage of metamizole.

As to be expected, the ORs in the analysis without exclusions show much larger ORs in all age groups, indicative of the assumed contribution of cancer-related exposure to ionizing radiation as well as of application of immunosuppressants. The incidence of agranulocytosis per 1 million person-years came to 0.054% in men and to 0.057% in women.

Strengths and limitations: The major strength of this study is the fact that there is no problem of response and that the major part of the population of the study region is included. However, there are several shortcomings: Individual factors regarding social status are missing. The percentage of persons on social welfare within the community or within districts of the community might be an indicator which is too broad for individual persons.

However, there is a structural factor in the German health system: Medications applied during hospital stays are not documented. Regarding metamizole, this certainly is a major problem, as there are rebate contracts of sickness funds with drug manufacturers, including metamizole-containing drugs. Thus, in many hospitals metamizole is the preferred analgesic. The application of metamizole in hospitals may lead to agranulocytosis, which will appear as a discharge diagnosis. This would in consequence lead to increased odd ratios in this study. If the amount of metamizole applied during hospital stays could be enumerated, the relative risks probably would be lower than ascertained with the data available.

Discussion: One of the first publications on the risk of agranulocytosis due to application of metamizole is the study of the International Agranulocytosis and Anemia Study Group [2] dating back to 1986. In this study, cases of agranulocytosis were ascertained from various regions: Germany (West-Berlin, Ulm), Spain (Barcelona), Israel, and Hungary (Budapest). The combined cases of agranulocytosis came to 221. 1,425 hospital controls were selected. Metamizole intake was recorded for the week before the on-set of symptoms of agranulocytosis (sore throat, fever, chills, headache). The authors reported an increased risk for metamizole for 27 cases from Germany and Spain (23.7 (95% CI 8.7–64.4), but not for Israel and Hungary (17 cases, OR 0.8 (95% CI 0.4–1.8)). Meta-analysis was still uncommon during the period of this publication. A meta-analytic combination of the published risks results in an OR of 13.89 (95% CI 9.59–18.21). These results are comparable to the publication of Ibanez et al. (2005) [8], where metamizole-intake was also assessed in the week preceding an index day, based on 30 cases and 9 controls. The resulting OR of 25.76 (95% CI 8.93–79.12) is within the margins of the 1986 study. Compared to the results of this study, the ORs of the above-mentioned two studies are extremely high. Another study, published in 2020 [9] as based on 630,285 persons on metamizole treatment and 390,830 matched controls. The resulting OR came to 3.03 (95% CI 2.49–3.69). As the authors did not exclude persons exposed to other risk factors for agranulocytosis than metamizole (e.g. immunosuppressants, ionizing radiation due to diagnostic or therapeutic procedures, especially in malignancies), these results seem to be rather low compared to the results of the present study.

Prescriptions for metamizole-containing drugs in Germany have been rising constantly over the last years from 1991 (15 million DDD) to 2018 (225 million DDD) [10].

Despite agranulocytosis as ascertained side-effect of metamizole, the official paper of German physicians – Deutsches Ärzteblatt – in 2020 is considering metamizole as indispensable [11].

Regarding possible deadly side-effects, two questions arise:

1.
How are the health systems of those countries operating, where metamizole has been banned for years?
2.
Are there drugs on the German market which are as effective to treat severe pain as metamizole?

The second question is easy to answer: The opioid tramadol is as effective as metamizole, but without deadly side-effects [1]. Suljevic et al. in a controlled clinical trial compared metamizole and tramadol on the effectiveness to control post-operative pain when given intravenously before operations [12]. They found that tramadol was significantly more effective than metamizole. It can be assumed that similar effects could be seen when treating other types of pain. An analysis on the comparability of different analgesics by Schug et al. came to similar results [13].

A change from metamizole to tramadol would in addition have another positive effect: The costs for DDD for metamizole presently come to € 1.42 per DDD, whereas one DDD of tramadol would only come to € 0.90 [2]. Hoffman et al. described the steady increase of metamizole prescriptions from 1991 to 2018 [10]: it started with 15 million DDD and increased to 225 DDD in 2018: Assuming the price per DDD of metamizole of € 1.42 and of tramadol of € 0.9, the total costs for metamizole would be € 319.5 million, whereas the costs for tramadol would be not more than € 202.5 million.

As the Drug Commission of the German Medical Association grossly underestimated the number of cases of metamizole-related agranulocytosis, it should be possible to derive a realistic estimate for the total German population, assuming that similar conditions would be plausible for the German population as those in the population of the present study: For the German population of 2018 (82.79 million) one could expect about 134,900 cases of agranulocytosis, of which about 45% (=about 60,000) would be due to metamizole.

Conclusion: The results of the present study should lead to a political decision to withdraw metamizole from the German market as soon as possible. Strong objections from drug manufacturers have to be anticipated.

Acknowledgements: The authors are greatly indebted to the members of the scientific advisory boards, who were convened to control and advise on two subsequent case-control studies, conducted in the area in the vicinity of the Cologne-Bonn airport. The aims of these studies were to analyze health effects due to exposure to environmental noise.

Members of the scientific advisory boards were: Prof. Dr. Ursula Ackermann-Liebrich, Basel; Dr. Wolfgang Babisch, Berlin; Dr. Mathias Basner, Cologne; Dr. Alexander Samel, Cologne; Prof. Dr. Karl-Heinz Jöckel, Essen; Prof. Dr. Heinz-Erich Wichmann, Munich; Prof. Dr. Wolfgang Hoffmann, Greifswald; Dr. Barbara Hoffmann, Essen.


References

1.
World Health Organisation. Pharmaceuticals – Restrictions in Use and Availability. Genf: WHO; 2005.
2.
The International Agranulocytosis and Aplastic Anemia Study. Risks of Agranulocytosis and Aplastic Anemia. A First Report Of Their Relation To Drug Use With Special Reference To Analgesics. JAMA. 1986;256:1749-1757.
3.
Agranulozytose nach Metamizol – sehr selten, aber häufiger als gedacht. Deutsches Ärzteblatt. 2011;108(33):A-1758.
4.
Bundesinstitut für Arzneimittel und Medizinprodukte. Metamizol (Novalgin, Berlosin, Novaminsulfon etc) – BfArM weist auf die Beachtung von Vorsichtsmaßnahmen, richtige Indikationsstellung und Warnhinweisen hin. 2009 May 28. Available from: https://www.bfarm.de/SharedDocs/Risikoinformationen/Pharmakovigilanz/DE/RI/2009/RI-metamizol.html Externer Link
5.
Greiser E, Janhsen K, Greiser C. Beeinträchtigung durch Fluglärm: Arzneimittelverbrauch als Indikator für gesundheitliche Beeinträchtigungen. Dessau-Roßlau: Umweltbundesamt; 2006.
6.
Greiser E, Greiser C. Risikofaktor nächtlicher Fluglärm. Abschlussbericht über eine Fall-Kontroll-Studie zu kardiovaskulären und psychischen Erkrankungen im Umfeld des Flughafens Köln-Bonn. Schriftenreihe Umwelt und Gesundheit 01/2010. Dessau-Roßlau: Umweltbundesamt; 2010.
7.
Wissenschaftliches Institut der AOK (WIdO). Anatomisch-therapeutisch-chemische Klassifikation mit Tagesdosen. Amtliche Fassung des ATC-Index mit DDD-Angaben für Deutschland im Jahre 2020. Berlin: WIdO; 2020.
8.
Ibanez L, Vidal X, Ballarin E, Laporte JR. Population-based drug-induced agranulocytosis. Arch Int Med. 2005;165:869-874.
9.
Klose S, Pflock R, König IR, Linder R, Schwaninger M. Metamizole and the risk of drug-induced agranulocytosisand neutropenia in statutory health insurance data. Naunyn Schmiederbergs Arch Pharmacol. 2020;393:681-690.
10.
Hoffmann F, Bantel C, Tilmann von R, Fosen F, Jobski K. Regional differences in prescribing patterns of metamizole in Germany based on data from 70 million persons. Int J Environ Res Public Health. 2020;17:3892.
11.
Metamizol und Agranulozytose. Deutsches Ärzteblatt 2020;117(4):A-142-A-145.
12.
Suljevic I, Hadzziavdic M, Suljevic O, Turan M. The preemptive effect of tramadol and metamizole on the intensity of postoperative pain. Med Glas (Zenica). 2020;17:285-289.
13.
Schug S, Pogatzki-Zahn E, Phillips, LD, Noyes Essex M, Xia F, Reader AJ, Pawinski R. Multi-criterie decision nalysisto develop an efficacy-safety profile of parenteral analgesics used in the treatment of postoperative pain. J Pain Res. 2020;13:1969-1977.
14.
Lohmüller J, Schröder M, Telschow C. Der GKV-Arzneimittelmarkt 2018: Trends und Marktsegmente. In: Schwabe U, Paffrath D, Ludwig WD, Klaube J, editors. Arzneiverordnungsreport 2019. Springer; 2019. p. 249-299.