Anaesthesiology Intensive Therapy, 2009,XLI,3; 141-145

Safety of metamizol and paracetamol for acute pain treatment

*Maciej Żukowski, Katarzyna Kotfis

Department of Anaesthesiology and Intensive Therapy, Pomeranian Medical University in Szczecin

Postoperative analgesia should be adjusted to current needs of a patient. Non-opioid agents are recommended, wherever possible: both non-steroidal anti-inflammatory drugs (NSAIDs), metamizole or paracetamol may be useful for treatment of acute pain.

The use of metamizol is associated with such complications as bone marrow damage (agranulocytosis, aplastic anaemia), chronic interstitial nephritis, gastro-intestinal disturbances, etc. Paracetamol, currently also intravenous, is likely to cause hepatocyte damage, renal necrosis, as well as vomiting, diarrhoea and skin reactions.

Opinions about metamizol are controversial. In some countries, metamizol was withdrawn already in the 70ties of the previous century; in others, it is still widely used. 

According to the Uppsala Monitoring Centre, WHO Collaborating Centre for International Drug Monitoring, the number of adverse effects registered in the years 1978-2009 (March) was 14441 for metamizol and 67581 for paracetamol.

Modern multimodal analgesia should be based on a good combination of analgesics. Both metamizol and paracetamol may be used for such a purpose, yet in the lowest effective doses, within the shortest needed time and once evident contraindications have been considered. Safety of both drugs is several times higher than that of commonly used NSAIDs. 

Pain management is the main element of perioperative therapy and its efficacy is essential for the quality of care perceived by patients. The treatment of choice for acute pain is multimodal, balanced analgesia based on the combination of several drugs of different mechanisms of action. The combined use of non-opioid and/or non-steroidal anti-inflammatory drugs  (NSAIDs) with opioids or local anaesthetics is recommended by the Polish Chapter of International Association for Study of Pain as it provides higher analgesic efficacy with reduced adverse effects [1]. Non-opioid analgesics, including paracetamol and metamizol, are well-recognized agents of multimodal analgesia in the treatment of postoperative pain, enabling the reduction of opioid doses by 40-50% [2].

Classification and pharmacological data

Metamizol and paracetamol are difficult to classify as both agents are characterized by poor or lack of anti-inflammatory effects, are weak COX-1 and COX-2 inhibitors, hence cannot be classified as NSAIDs; unfortunately, such a factual mistake can still be found in some reports or even textbooks of pharmacology [3, 4].

Metamizol is a pirazolone derivative of analgesic and antipyretic action, yet without an anti-inflammatory component. Although the drug has been available since 1922, its mechanism of action is not fully known. Inhibition of the activity of COX in the CNS, which diminishes the synthesis of prostaglandins has been suggested [5]. The literature reports contain several hypotheses explaining the analgesic efficacy of metamizol, including inhibition of the COX-3 isoenzyme and decrease in prostaglandin synthesis in the spinal posterior horns [6, 7].  Additionally, the drug exerts spasmolytic effects in spastic conditions of the urinary and biliary tract.

Likewise, paracetamol has analgesic and antipyretic action without anti-inflammatory effects. Its mechanism of action is not fully explained. Antipyretic effects of paracetamol are likely to be caused by inhibition of prostaglandin synthesis in the hypothalamus [8]. The attempts to explain its analgesic action are based on several hypotheses, which implicate the inhibition of COX-2 activity or effects on COX-3 in CNS; however, the existence of this latter isoenzyme is the subject of much controversy. Another potential mechanism of action of paracetamol is its effect on the antinociceptive system by stimulation of the activity of descending serotoninergic pathways. Moreover, the influence on peripheral opioid receptors is considered [4].

Metamizol is hydrolyzed in the gastrointestinal tract to 4-methylaminoantipirine (4-MAA) and is absorbed in that form; its bioavailability is over 80%. Hepatic enzymes metabolize metamizol to 4-aminoantipirine (AA) and 4-formylaminoantipirine (FAA); in turn, AA is acethylated to 4-acetylaminoantipirine (AAA). All metabolites of metamizol show the biological activity, are responsible for its analgesic action and permeate into maternal milk. The metabolites bind with plasma proteins in about 60%, hence the displacement of other compounds from these bonds responsible for numerous drug interactions is not a relevant problem; 65-70% of active metabolites of metamizol are excreted with urine. The elimination of 4-MAA is prolonged by 22% after multiple doses and by 33% in the elderly [5, 9].  

Paracetamol is absorbed in the small intestine; its bioavailability after oral administration is 70-90% due to the effect of first transition; 5-10% of paracetamol bind with plasma proteins and its volume of distribution is high. As a non-ionized substance soluble in fats, it readily penetrates the blood-brain barrier. It is mainly metabolized in the liver, 63% is bound to glucoronate and 34% to sulphate, which leads to the formation of water-soluble metabolites. Less than 5% of the therapeutic dose of paracetamol is metabolized by cytochrome P-450 (mainly CYP2E1) to a highly reactive indirect metabolite N-acethyl-p-benzoquinone imine (NAPQI), which is inactivated after binding with glutathione. About 1% of the drug is excreted with urine in its intact form [8]. According to literature data, symptoms of metamizol overdosage are observed at the doses of 5-30 g whereas of  paracetamol at 7.5-30 g or 150 mg kg-1 day-1.

Adverse side effects

The metamizol-related side effects include gastrointestinal disorders (nausea, vomiting, stomach pain, diarrhoea), headache and dizziness, hepatic and renal dysfunctions (chronic interstitial nephritis with renal papillary necrosis), hypersensitivity skin reactions – rash, urticaria, erythema, exfoliating dermatitis and toxic-diffuse skin necrosis, as well as exacerbation of bronchial asthma, anaphylactic shock. Metamizol inhibits the excretion of water and sodium, which may cause peripheral oedema [10, 11].

Adverse effects following paracetamol are rare – vomiting and nausea <1/100 individuals, urticaria, erythema, dermatitis  <1/1000 whereas thrombocytopenia, leucocytosis, agranulocytosis, and enlargement of the liver <1/10000 individuals [8].

Skin reactions, such as the Lyell or Stevens-Johnson syndrome, are equally rare after paracetamol and metamizol although their course is quiet severe. The pseudo-allergic reactions affect mainly patients with mucosal inflammations, in whom any non-opioid analgesic is likely to induce aspirin asthma [3].

According to WHO, the overall number of adverse side effects (number of events) attributable to metamizol was 14, including 182 cases in Poland. The data reported by the Uppsala Monitoring Centre, WHO Collaborating Centre for International Drug Monitoring covered the period between 1978 and March 2009. The respective number for paracetamol was 67581, including 19 cases in Poland. The number of metamizol-implicated deaths, irrespective of their causes, was 832 during 31 years, including 2 in Poland; 354 cases (2 in Poland) – associated with oral administration while 194 (0 in Poland) with intravenous route. For paracetamol, the respective numbers were as follows: in total, 19cases, with 12940 following oral administration and 56 after intravenous injection [12]. It is difficult, however, to assess this report explicitly as it does not provide information on the general number of preparations sold during that period and covers only the cases notified to the Monitoring Centre.


Myelotoxicity of metamizol has been disputed for years. In Sweden, metamizol was banned in 1974, in the USA in 1977. Similar steps were taken by other 30 countries including Japan, Australia, Iran and some European Union countries. In many countries, e.g. Spain, Mexico, India, Egypt, Brazil, Poland, Russia, Turkey, Bulgaria and Germany, metamizol, usually as an OTC drug, has been widely used and still the incidence of agranulocytosis in these countries is casuistic. According to some Polish studies, the incidence of aplastic anaemia in the general population is incidental [13]. In Sweden, the ban was abolished in 1996 and re-introduced in 1999 after the studies conducted by Backstrom and co-workers [14] evidencing 10 cases of agranulocytosis during 3 years of metamizol use and assessing the risk of metamizol-induced agranulocytosis to be 1/31000 of hospitalized patients and 1/1400 of those treated outside hospital. In patients treated in hospitals, the risk remained low, yet the mean dose of metamizol, 2.7 g used in non-hospitalized individuals was higher than the recommended one and, as even the authors admit, the drug was used for more than 7 days, often more than 21 days.

The recent analyses demonstrate that earlier findings concerning the incidence of metamizol-induced agranulocytosis and aplastic anaemia, which led to the ban of metamizol in many countries, were overestimated. Although metamizol is considered the drug potentially inducing agranulocytosis, the estimated risk of this complication shows high geographic variability, which is likely to reflect genetic differences of the populations assessed in epidemiological studies. Hedenmalm and Spigset [15] reported the incidence of 1 case per 1 431 prescriptions in Sweden. Ibanez and co-workers [16] found that in Spain, the total risk of agranulocytosis following metamizol at usual doses and for a short period is extremely low, i.e. 0.56 cases/1inhabitants/1 year. Moreover, they observed that the risk increased with prolongation of metamizol administration yet disappeared 10 days after its final dose. In cases of long-term use, the risk of fatal agranulocytosis increases twentyfold [16].

To account for such discrepancies, a large, prospective, multi-centre study should be carried out in the countries in which metamizol is routinely used for acute pain management. Until then, metamizol should be administered over the periods <7 days; in cases of prolonged treatment, peripheral blood dyscrasias should probably be monitored. However, in the discussion of metamizol-induced agranulocytosis, the risk of other adverse effects and life-threatening complications (gastrointestinal bleeding, renal failure, severe skin reactions) of alternative analgesics should be considered, i.e. NSAIDs. In general, despite geographic differences, the maximum incidence of metamizol-induced agranulocytosis is 5/1 000000 inhabitants compared to 0.25 cases when the drug was not used [17]. These data should be contrasted with the risk of death due to gastrointestinal haemorrhage caused by non-steroidal anti-inflammatory drugs, which is 300 times higher.

According to the Polish Ministry of Health, “considering the metamizol therapy-related cases of agranulocytosis observed, metamizol is recommended only when other analgesics are ineffective or contraindicated” and “the use of the drug in high doses or its prolonged administration increases the risk of agranulocytosis, hence the use should not exceed 7 days”.  It appears there are no conclusive data to consider metamizol a second-line medication, yet its administration shorter than 7 days seems justified.

Mechanisms of toxicity

  • Hepatotoxicity of paracetamol

The available literature lacks data indicating that therapeutic doses of paracetamol induce liver damage. The dose of 150 mg kg-1 or 5-30 g is considered toxic. Liver damage develops after paracetamol overdose and filling of the basic elimination pathway. The depletion of glutathione stores leads to accumulation of n-acethylbenzoquinone imine (NAPQI), one of its metabolites, which results in hepatocyte damage – histopathological findings show centrolobular necrosis. Chronic hepatic damage is rare [18].

  • Nephrotoxicity of paracetamol

Hepatotoxicity of paracetamol has been widely described whereas sparse data concern kidney damage induced by this drug. Nephrotoxicity, however, is equally relevant, affecting about 1-2% of patients after paracetamol overdose yet at the doses lower than those inducing hepatotoxicity and later, i.e. during 2-5 days after use. Kidney damage may be independent of hepatic damage. The implicated mechanisms of paracetamol-induced nephrotoxicity include isoenzymes of oxidase of cytochrome P-450 (CYP2E1) present in kidneys as well as synthetase of prostaglandins or N-deacethylase. Paradoxically, administration of glutathione, the basis of detoxication during post-paracetamol hepatic damage therapy, may lead to formation of nephrotoxic glutathione compounds resulting in activation of caspases and liposomal enzymes initiating apoptosis, which eventually results in renal necrosis and dysfunction. The histopathological picture shows acute necrosis of renal tubules [19].

  • Myelotoxicity of metamizol

The mechanism of metamizol-induced agranulocytosis has not been fully elucidated; moreover, the risk factors of this damage are unknown. The bone marrow damage may occur even after a single administration of the drug. Agranulocytosis is likely to develop due to idiosyncrasies. As a pro-drug, metamizol is hydrolyzed to active metabolites, including 4-methyloaminophenazone which binds with the lysine site at neutrophils, hence a new antigen determinant is formed, which may cause damage to neutrophils and progenitor cells in the bone marrow [20]. Metamizol-induced agranulocytosis is a hypersensitivity reaction; once the antigenic determinant occurs, the reaction is independent of the drug dose used. However, higher doses or prolonged use may potentially induce hypersensitivity reactions [21]. The possible risk factors may also include geographic differences, genetic and/or local environmental cofactors.

2008 guidelines of the Polish Chapter of International Association for Study of Pain regarding postoperative use of paracetamol and metamizol

Both paracetamol and metamizol are recommended as agents of multimodal analgesia. In the guidelines of the Polish Chapter of International Association for Study of Pain, both are included as an analgesic component in all four categories of procedures, namely those with low (VAS <4), average (VAS >4 with pain lasting <3 days), considerable (VAS >4 with pain longer than 3 days) and extensive  (VAS >6 and pain sensations lasting more than 7 days) tissue trauma.

Paracetamol is recommended for pre-emptive analgesia in the dose 1.0 g iv or 1.0-2.0 g per  rectum as well as for postoperative analgesia. The drug ought to be administered intravenously 30 min before completion of the surgical procedure or orally (as quickly as possible). The analgesic efficacy of equivalent doses is higher following intravenous than oral administration due to higher maximal concentration, quicker onset of action and stronger CNS effects. The absorption of the drug after oral administration in the postoperative period is patient-dependent and highly variable, which is caused by delayed emptying of the stomach induced by paracetamol alone and simultaneous use of opioids. The recommended dose – 1 g of paracetamol 4 times a day should be reduced to 2-3 administrations daily in patients with liver failure. The recommended monitoring includes: pain severity (VAS), sedation, respiratory rate and adverse effects.

Likewise, metamizol may be used for pre-emptive analgesia, 1-2.5 g in the slow intravenous infusion and for postoperative analgesia, 0.5-1.0 g every 4 to 6 h in the slow (about 30 min) intravenous infusion [1, 3]. It should be remembered that sudden pressure drops during administration of metamizol are caused by too quick intravenous infusions.

  • Comparison of efficacy of paracetamol and metamizol

Equipotent analgesic efficacy of 1 g of intravenous paracetamol and metamizol was demonstrated during postoperative pain treatment following retinal detachment repairs or breast surgeries (total or subtotal mastectomy) [22, 23]. Kampe and colleagues [24] comparing both drugs  found no differences in the incidence of adverse effects other than hypotension, patients` satisfaction and additional opioid use. Only 5 patients in the metamizol group developed hypotension. According to Grundmann and co-workers [25], who compared equivalent doses of paracetamol and metamizol in patients undergoing lumbar microdiscectomy, metamizol was superior to paracetamol, parecoxib and placebo.  

  • Safety of analgesics – comparative assessment

The comparative assessment of non-opioid analgesics in epidemiological studies covering the period of 25 years was to determine the cumulative mortality dependent on adverse effects of aspirin, diclofenac, paracetamol and metamizol. The mortality rates related to agranulocytosis, aplastic anaemia, anaphylaxis and severe upper airway complications were analysed. The estimated mortality related to  this complications was 185 cases/100for aspirin, 592/100 000 000 for diclofenac, 20/100 000 000 for paracetamol and 25/100 000 000 for metamizol and mainly concerned gastrointestinal complications. To equal the risk of metamizol-induced agranulocytosis and that of aspirin or diclofenac, the anticipated risk of death for metamizol should be >300 cases/100 000 000,i.e. tenfold higher than currently. According to the authors of that analysis, legal regulations in individual countries limiting or banning the use of metamizol are ungrounded.  Moreover, the analysis indicates that the metamizol-related risk is similar to that of paracetamol, erroneously considered much safer [26].

The risk of metamizol-induced agranulocytosis should be assessed in comparison to other potentially fatal complications related to analgesics, a high proportion of which are NSAIDs. The incidence of gastrointestinal haemorrhage is estimated at 400 cases/1000 000/1 year, 38% of which are associated with NSAIDs [27]. The use of paracetamol and metamizol is not related to the risk of gastrointestinal haemorrhage, thus both drugs are safe alternatives in patients with contraindications for non-steroidal analgesic agents.

Furthermore, paracetamol is considered safe during pregnancy; it does no increase the risk of spontaneous abortion, has no teratogenic effects and does not cause foetus malformations. Contrary to paracetamol, metamizol is currently contraindicated during pregnancy and breast-feeding. However, its teratogenic action has been demonstrated only in animal studies. The prospective study carried out by Bar-Oz and colleagues [28] involving 108 women using metamizol in the first trimester of pregnancy, did not demonstrate metamizol-related increased risk of foetal malformations or higher incidences of spontaneous abortions. For methodological reasons, prospective, randomized studies on the safety of metamizol in pregnancy cannot be conducted in humans and therefore the teratogenic effects of the drugs may only be assessed in retrospective or observational studies.  


It should be strongly emphasized that multimodal analgesic therapy is a suitable combination of analgesics of various mechanisms of action to increase the efficacy and reduce the incidence of adverse side effects. Both non-opioid analgesics in question, paracetamol and metamizol, are recognized pain management agents for pre-emptive as well as postoperative analgesia. Both are well known in Poland, no advertising campaigns or recommendations are required. The objective of our review was to stress the safety of their use in the Polish population. The key issue is to use them for what they were intended; metamizol should not be used for more than 7 postoperative days and the recommended doses should not be exceeded. The additional asset of metamizol is its spasmolytic action useful in contraction pains. In some selected groups of patients, e.g. with the risk of gastrointestinal haemorrhage or contraindications for NSAIDs both agents are a safe and effective alternative.

The safety of metamizol and paracetamol is comparable and several times higher than that of non-steroidal anti-inflammatory drugs used for postoperative pain management. Obviously, the drugs should be administered at the lowest effective doses, over the shortest possible time; absolute contraindications should be considered.



1.  Dobrogowski J, Mayzner-Zawadzka E, Drobnik L, Kusza K, Woron J, Wordliczek J: Usmierzanie bólu pooperacyjnego – zalecenia 2008; Ból 2008; 9: 9-22.

2.  Dobrogowski J, Przeklasa-Muszynska A, Woron J, Wordliczek J: Zasady kojarzenia leków wbólu. Medycyna Paliatywna w2007; 1: 6-15.

3.  Diener H: Leczenie bólu. Zespoly bólowe – metody postepowania. Urban & Partner, Wroclaw, 2005; 294-296.

4.  Rezende RM, Franca DS, Menezes GB, WGP dos Reis, Bakhle YS, Francischi JN: Different mechanisms underlie the analgesic actions of paracetamol and dipyrone in amodel of inflammmatory pain. Br J Pharmacol 2008; 153: 760-768.

5.  Hinz B, Cheremina O, Buchmakov J, Renner B: Dipyrone elicits substantial inhibition of peripheral cyclooxygenanses in humans: new insights into the pharmacology of an old analgesic; FASEB J 2007; 21: 2343-2351.

6.  Arellano F, Sacristan JA: Metamizole: reassessment of its therapeutic role. Eur J Clin Pharmacol 1990; 38: 617-619.

7.  Pierre SC, Schmidt R, Brennels C, Michaelis M: Inhibition of Cyclooxygenanses by Dipyrone. Br J Pharmacol 2007; 151: 494-503.

8.  Peck T: Pharmacology for Anaesthesia and Intensive Care, 2nd Edition; Greenweich Medical Media Ltd. 2003.

9.  Heinemeyer G, Gramm HJ, Roots I, Dennhardt R, Singen W: The kinetics of metamizol and its metabolites in critical-care patients with acute renal dysfunction. Eur J Clin Pharmacol 1993; 45: 445-450.

10. Woron J, Wordliczek J, Filipczak-Bryniarska I, Dobrogowski J: Powiklania farmakoterapii bólu. Anestezjologia i2008; 2: 177-184.

11. Woron J, Porebski G, Filipczak-Bryniarska I, Wordliczek J, Bisagew, Dobrogowski J, Przeklasa-Muszynska A, Salakowski A: Monitorowanie niepozadanych dzialan leków stosowanych wbólu. Problemy Terapii Monitorowanej 2008; 19: 17-22.  

12. Uppsala Monitoring Centre, WHO Collaborating Centre for International Drug Monitoring, 2009.

13. Maj S, Centkowski P: Astudy of the incidence of agranulocytosis and aplastic anemia associated with the oral use of metamizole sodium in Poland. Med Sci Monit 2004; 10: PI93-95.

14. Backstrom M, Hauml S, Mjorndal T, Dahlqvist R: Utilization pattern of metamizole in northern Sweden and risk estimates of agranulocytosis. Pharmacoepidemiol Drug Saf 2002; 11: 239-245.

15. Hedenmalm K, Spigset O: Agranulocytosis and other blood dyscrasias associated with dipyrone (metamizole). Eur J Clin Pharmacol 2002; 58: 265-274.

16.Ibanez L, Vidal X, Ballarin E, Laporte JR: Agranulocytosis associated with dipyrone (metamizol). Eur J Clin Pharmacol 2005; 60: 821-829.

17. Hamerschlak N, Maluf E, Biasi Cavalcanti A, Avezum Junior A, Eluf-Neto J, Passeto Falcao R, Lorand-Metze IG, Goldenberg D, Leite Santana C, de Oliveira Werneck Rodrigues D, Nascimento da Motta Passos L, Oliveira de Miranda Coelho E, Tostes Pintao MC, Moraes de Souza H, Borbolla JR, Pasquini R: Incidence and risk factors for aganulocytosis in Latin American countries – the Latin Study: astudy. Eur J Clin Pharmacol 2008; 64: 921-929.

18. Chun LJ, Tong MJ, Bussutil RW: Acetaminophen hepatotoxicity and acute liver failure. J Clin Gastroenterology 2009; 43: 342-349.

19. Mazer M, Perrone J: Acetaminophen-induced nephrotoxicity. Pathophysiology, clinical manifestations and management. J Med Toxicol 2008; 4: 2-6.

20. Garcia-Martinez JM, Vara JAF, Lastres P, Bernabeu C, Betes PO, Martin-Perez J: Effect of metamizol on promyelocytic and terminally differentiaited granulocytic cells. Comparative analysis with acetylosalicylic acid and diclofenac. Biochem Pharmacol 2003; 65: 209-217.

21. Andres E, Maloisel F: Idiosyncratic drug-induced agranulocytosis or acute neutropenia. Curr Opin Hematol 2008; 15: 15-21.

22. Landwehr S, Kiencke P, Giesecke T, Eggert D, Thumann G, Kampe S: Abetween iv paracetamol and iv metamizol for postoperative analgesia after retinal surgery. Curr Med Res Opin 2005; 21: 1569-1575.

23. Ohnesorge H, Bein B, Hanss R, Francksen H, Mayer L, Scholz J, Tonner PH: Paracetamol versus metamizol in the treatment of postoperative pain after breast surgery: acontrolled trial. Eur J Anaest 2009; 26: 648-653.

24. Kampe S, Warm M, Landwehr S, Dagtekin O, Haussmann S, Paul M, Pilgram B, Kiencke P: Clinical equivalence of iv paracetamol compared to iv dipyrone for postoperative analgesia after surgery for breast cancer. Curr Med Res Opin 2006; 22: 1949-1954.

25. Grundmann U, Wörnle C, Biedler A, Kreuer S, Wrobel M, Wilhelm W: The efficacy of the non-opioid analgesics parecoxib, paracetamol and metamizol for postoperative pain relief after lumbar microdiscectomy.  Anesth Analg 2006; 103: 217-222.

26. Andrade S, Martinez C, Walker AM: Comparative safety evaluation of non-narcotic analgesics. J Clin Epidemiol 1998, 51: 1357-1365.

27. Lanas A, Serrano P, Bajador E, Fuentes J, Sainz R: Risk of upper gastrointestinal bleeding associated with non-aspirin cardiovascular drugs, analgesics and nonsteroidal anti-inflammatory drugs. Eur J Gastroent Hepat 2003; 15:173-178.

28. Bar-Oz B, Clementi M, Di Giantonio E, Greenberg R, Beer M, Merlob P, Arnon J, Ornoy A, Zimmerman DM, Berkovitch M: Metamizol (dipyrone, optalgin) in pregnancy, is it safe? Acomparative study. Eur J Obstet Gynecol Reprod Biol 2005; 119: 176-179.



*Maciej Żukowski
Klinika Anestezjologii i Intensywnej Terapii
Pomorskiej AM w Szczecinie
ul. Unii Lubelskiej 1, 71-242 Szczecin
tel.: 0-91 425 33 78

Received: 20.07.2009
Accepted: 20.08.2009