[Pharmacotherapy of hyperthyreosis--adverse drug reactions]

The antithyroid drugs mainly include thioimidazole (carbimazole, methimazole=thiamazole) and propylthiouracil. After absorption, carbimazole is rapidly metabolized to methimazole and thus switching between these two drugs should not be considered in case of side effects. Furthermore, in case of side effects, sometimes even cross reactions between thioimidazoles and propylthiouracil occur. Common and typical adverse reactions of antithyroid drugs include dose dependent hypothyroidism and thus thyroid function should be repeatedly checked while the patient is on antithyroid drugs. Furthermore, pruritus and rash may develop. In this case, one might try to switch from thioimidazoles to propylthiouracil or vice versa. Antithyroid drugs may cause mild dose dependent neutropenia or severe allergy-mediated agranulocytosis, which typically occurs during the first three months of treatment, has an incidence of 3 per 10,000 patients and cross reactivity between thioimidazoles to propylthiouracil may occur. Rarely, antithyroid drugs can cause aplastic anemia. Mainly propylthiouracil, but sometimes also methimazole may lead to an asymptomatic transient increase in liver enzymes or to severe, even lethal liver injury of cholestatic or hepatocellular pattern. Since propylthiouracil associated liver injury was observed increasingly among children and adolescent, it has been suggested to prefer thioimidazoles for these patients. Because of these potential serious adverse effects, physicians should advise patients to immediately seek medical help if they get a fever or sore throat or malaise, abdominal complaints or jaundice, respectively. Furthermore, arthralgias may develop in 1-5% of patients under both antithyroid drugs. Since arthralgias may be the first symptom of more serious immunologic side effects, it is recommended to stop the antithyroid drug in this case. Drug induced polyarthritis mainly develops during the first month of therapy, whereas ANCA-positive vasculitis is generally observed only after long term exposure to propylthiouracil or very rarely with the thioimidazoles. The teratogenic risk of the thioimidazoles is somewhat higher (Aplasia cutis congenita), that is why one generally recommends preferring propylthiouracil during pregnancy. During breast feeding both, thioimidazoles or propylthiouracil, may be administered. Nowadays, perchlorate is only used short term in case of latent hyperthyroidism before administering iodine-containing contrast agents. Therefore, the known side effects, which usually are only observed after long term treatment, are not an issue any more.

Toxicity of thienopyridines on human neutrophil granulocytes and lymphocytes

Thienopyridines can cause neutropenia and agranulocytosis. The aim of the current investigations was to compare cytotoxicity of ticlopidine, clopidogrel, clopidogrel carboxylate and prasugrel for human neutrophil granulocytes with the toxicity for lymphocytes and to investigate underlying mechanisms. For granulocytes, clopidogrel, ticlopidine, clopidogrel carboxylate and prasugrel were concentration-dependently toxic starting at 10μM. Cytotoxicity could be prevented by the myeloperoxidase inhibitor rutin, but not by the cytochrome P450 inhibitor ketoconazole. All compounds were also toxic for lymphocytes, but cytotoxicity started at 100μM and could not be prevented by rutin or ketoconazole. Granulocytes metabolized ticlopidine, clopidogrel, clopidogrel carboxylate and prasugrel, and metabolization was inhibited by rutin, but not by ketoconazole. Metabolism of these compounds by lymphocytes was much slower and could not be inhibited by ketoconazole or rutin. In neutrophils, all compounds investigated decreased the electrical potential across the inner mitochondrial membrane, were associated with cellular accumulation of ROS, mitochondrial loss of cytochrome c and induction of apoptosis starting at 10μM. All of these effects could be inhibited by rutin, but not by ketoconazole. Similar findings were obtained in lymphocytes; but compared to neutrophils, the effects were detectable only at higher concentrations and were not inhibited by rutin. In conclusion, ticlopidine, clopidogrel, clopidogrel carboxylate and prasugrel are toxic for both granulocytes and lymphocytes. In granulocytes, cytotoxicity is more accentuated than in lymphocytes and depends on metabolization by myeloperoxidase. These findings suggest a mitochondrial mechanism for cytotoxicity for both myeloperoxidase-associated metabolites and, at higher concentrations, also for the parent compounds.

Metamizole-associated adverse events: a systematic review and meta-analysis.

BACKGROUND Metamizole is used to treat pain in many parts of the world. Information on the safety profile of metamizole is scarce; no conclusive summary of the literature exists. OBJECTIVE To determine whether metamizole is clinically safe compared to placebo and other analgesics. METHODS We searched CENTRAL, MEDLINE, EMBASE, CINAHL, and several clinical trial registries. We screened the reference lists of included trials and previous systematic reviews. We included randomized controlled trials that compared the effects of metamizole, administered to adults in any form and for any indication, to other analgesics or to placebo. Two authors extracted data regarding trial design and size, indications for pain medication, patient characteristics, treatment regimens, and methodological characteristics. Adverse events (AEs), serious adverse events (SAEs), and dropouts were assessed. We conducted separate meta-analyses for each metamizole comparator, using standard inverse-variance random effects meta-analysis to pool the estimates across trials, reported as risk ratios (RRs). We calculated the DerSimonian and Laird variance estimate T2 to measure heterogeneity between trials. The pre-specified primary end point was any AE during the trial period. RESULTS Of the 696 potentially eligible trials, 79 trials including almost 4000 patients with short-term metamizole use of less than two weeks met our inclusion criteria. Fewer AEs were reported for metamizole compared to opioids, RR = 0.79 (confidence interval 0.79 to 0.96). We found no differences between metamizole and placebo, paracetamol and NSAIDs. Only a few SAEs were reported, with no difference between metamizole and other analgesics. No agranulocytosis or deaths were reported. Our results were limited by the mediocre overall quality of the reports. CONCLUSION For short-term use in the hospital setting, metamizole seems to be a safe choice when compared to other widely used analgesics. High-quality, adequately sized trials assessing the intermediate- and long-term safety of metamizole are needed.