Department of Pharmacology, University of Bologna, Bologna, Italy.
Drug Saf. 2012 Apr 1;35(4):315-23. doi: 10.2165/11596510-000000000-00000.
: The risk of myocardial infarction, macular oedema and bone fractures associated with thiazolidinediones (TZDs) has been extensively investigated.
: The aim of the study was to verify if the analysis of a large spontaneous reporting database could generate early signals on these adverse drug reactions (ADRs) associated with TZDs.
: A case/non-case study, restricted to antidiabetic drugs, was performed on spontaneous reports of ADRs (2005-2008) in the US FDA Adverse Event Reporting System (AERS). The method was applied to TZDs, both as a drug class and as single agents. The reporting odds ratio (ROR) with 95% CI was calculated as a measure of disproportionality in the whole dataset and in a quarter-by-quarter analysis.
: TZD use was registered in 49 589 out of 301 950 drug-reaction pairs (16%), with significant disproportionality for myocardial infarction (ROR 4.71; 95% CI 4.40, 5.05), macular oedema (3.88; 2.79, 5.39) and bone fractures (1.73; 1.53, 1.96). Separate analysis of the two TZDs showed that only rosiglitazone was associated with myocardial infarction (7.86; 7.34, 8.34) and macular oedema (5.55; 3.94, 7.79), whereas pioglitazone was associated with multiple site fractures (2.00; 1.70, 2.35), in particular upper and lower limb and pelvic fractures. The quarter-by-quarter analysis identified disproportionality for myocardial infarction (3.13; 2.38, 4.10) and bone fractures since January-March 2005 (2.70; 1.04, 2.78).
: The frequency of reporting of myocardial infarction, macular oedema and fractures was significantly higher for TZDs in comparison with other antidiabetic drugs, with large intraclass differences. Both myocardial infarction and bone fracture signals appeared before major publications on these safety issues.
噻唑烷二酮类药物（TZDs）与心肌梗死、黄斑水肿和骨折的风险已被广泛研究。
本研究旨在验证分析大型自发报告数据库是否可以对 TZDs 相关的这些药物不良反应（ADR）产生早期信号。
在 2005 年至 2008 年期间，在美国食品和药物管理局不良事件报告系统（AERS）中，对自发报告的 ADR（2005-2008 年）进行了病例/非病例研究，仅限于抗糖尿病药物。该方法适用于 TZDs，无论是作为一种药物类别还是单一药物。报告比值比（ROR）及其 95%置信区间被计算为整个数据集和逐季分析中比例失调的衡量标准。
在 301950 对药物反应对中，有 49589 对与 TZD 有关（16%），心肌梗死（ROR 4.71；95%CI 4.40，5.05）、黄斑水肿（3.88；2.79，5.39）和骨折（1.73；1.53，1.96）存在显著的比例失调。对两种 TZD 的单独分析表明，只有罗格列酮与心肌梗死（7.86；7.34，8.34）和黄斑水肿（5.55；3.94，7.79）有关，而吡格列酮与多处骨折（2.00；1.70，2.35）有关，特别是上肢、下肢和骨盆骨折。逐季分析表明，自 2005 年 1 月至 3 月以来，心肌梗死（3.13；2.38，4.10）和骨折（2.70；1.04，2.78）的报告比例失调。
与其他抗糖尿病药物相比，TZDs 的心肌梗死、黄斑水肿和骨折报告频率明显更高，且类内差异较大。心肌梗死和骨折的信号都出现在这些安全性问题的主要出版物之前。
