Bollotte A, Vial T, Bricca P, Bernard C, Broussolle C, Sève P
Service de médecine interne, hôpital de la Croix-Rousse, 1, place de l'Hôpital, 69317 Lyon, France.
Centre régional de pharmacovigilance et d'information sur les médicaments, 69424 Lyon cedex 3, France.
Rev Med Interne. 2014 Dec;35(12):779-89. doi: 10.1016/j.revmed.2014.05.009. Epub 2014 Jun 5.
Drug-induced immune haemolytic anemia occurs in one case per million and can be fatal. Our aim was to describe the main characteristics and the type of drug involved.
Cases were retrospectively identified using spontaneous notifications collected by our pharmacovigilance centre and the results of immuno-hematological investigations performed by the laboratory of French blood establishment of Lyon between 2000 and 2012. Inclusion criteria were: an immune (positive direct antiglobulin test), hemolytic, anemia (haemoglobin <100 g/L), with at least a plausible causal relationship with drug exposure according to the French method of imputability or the presence of drug-dependent antibodies, and exclusion of other causes of hemolysis.
Ten cases (5 men and 5 women, median age 54.4 years) were identified. Causal drugs were ambroxol, beta-interferon, cefotetan, ceftriaxone, loratadine, oxacillin, oxaliplatine, piperacillin-tazobactam, pristinamycin, and quinine. The median time to onset of anemia after starting the culprit drug was 6 days (2 hours to 16 days). The median nadir of hemoglobin was 57.9 g/L (range: 34-78). The direct antiglobulin test was positive in 8 patients: IgG only (n=4), IgG and complement (n=3), and IgA (n=1). Drug-induced immune haemolytic anemia was considered as definite in 5 cases with positive drug-induced antibodies, probable in 4 cases negative for the detection of drug-induced antibodies but with plausible or likely causal relationship with drug exposure, and probable with an autoimmune mechanism in 1 case.
The diagnosis of DIIHA is often difficult because of the similarities with autoimmune haemolytic anemia and the inconstant sensitivity of immunologic tests that sometimes required repetitive assessment.
药物性免疫性溶血性贫血的发病率为百万分之一，且可能致命。我们的目的是描述其主要特征及相关药物类型。
通过我们的药物警戒中心收集的自发报告以及2000年至2012年间里昂法国血液机构实验室进行的免疫血液学调查结果，对病例进行回顾性识别。纳入标准为：免疫性（直接抗球蛋白试验阳性）、溶血性贫血（血红蛋白<100 g/L），根据法国因果关系判定方法，与药物暴露至少存在合理的因果关系，或存在药物依赖性抗体，且排除其他溶血原因。
共识别出10例病例（5男5女，中位年龄54.4岁）。相关药物为氨溴索、β-干扰素、头孢替坦、头孢曲松、氯雷他定、苯唑西林、奥沙利铂、哌拉西林-他唑巴坦、利福霉素和奎宁。开始使用致病药物后至贫血发作的中位时间为6天（2小时至16天）。血红蛋白的中位最低点为57.9 g/L（范围：34 - 78）。8例患者直接抗球蛋白试验呈阳性：仅IgG阳性（n = 4）、IgG和补体阳性（n = 3）、IgA阳性（n = 1）。5例药物诱导抗体阳性的病例被认为药物性免疫性溶血性贫血确诊，4例药物诱导抗体检测阴性但与药物暴露存在合理或可能因果关系的病例被认为可能，1例具有自身免疫机制的病例被认为可能。
由于药物性免疫性溶血性贫血与自身免疫性溶血性贫血相似，且免疫检测的敏感性不稳定，有时需要重复评估，因此其诊断往往困难。
