Abstract：
OBJECTIVE: Increased mortality in epilepsy due to infections (other than pneumonia) has been demonstrated. Small case series of people on antiepileptic drugs (AEDs) have described hypogammaglobulinaemia, which may predispose to infections. It is unclear whether hypogammaglobulinaemia is more frequent in people on AEDs, what AEDs it is associated with, or what clinical impact it has. In this population-based study, we aimed to determine whether AEDs were associated with hypogammaglobulinaemia, which AEDs were associated, and whether the associations may be causal.
METHODS: We conducted an unmatched case-control study using data linkage of routinely collected biochemistry, prescribing, and morbidity datasets in North-East Scotland from 2009-2021. Cases were participants with immunoglobulin levels less than the reference range. Controls were those with normal/high immunoglobulin levels. Logistic regression was used to investigate associations between AED exposure and any hypogammaglobulinaemia, adjusting for age, sex, and comorbidity. We also analysed low IgA, IgM, and IgG separately. We analysed \"any AED\" exposure and common individual drugs separately. Cumulative exposure data were used to determine whether an exposure-response relationship was present.
RESULTS: 18,666 cases and 127,157 controls were identified. Use of any AED was associated with increased risk of hypogammaglobulinaemia (adjusted odds ratio [aOR] 1.20 [95% CI: 1.15-1.25]). Phenytoin use was strongly associated with low IgA (aOR 5.90 [95% CI: 3.04, 10.43]). Carbamazepine and lamotrigine were also associated with low IgA. Apart from topiramate, which was associated with a non-significant decrease in odds of hypogammaglobulinaemia, there was a consistent increase in odds of hypogammaglobulinaemia across most AEDs studied. Panhypogammaglobulinaemia was associated with any AED use, carbamazepine, lamotrigine, gabapentin, and multiple AED use. There was evidence of an exposure-response relationship between any AED use and any hypogammaglobulinaemia, low IgA, and low IgG. Carbamazepine and probably lamotrigine also had an exposure-response relationship with any hypogammaglobulinaemia.
CONCLUSIONS: AEDs may increase hypogammaglobulinaemia risk. Specific classes of immunoglobulins are differentially affected, and the exposure-response analysis suggests this may be causal. Further work should investigate the clinical impact of these findings. Clinicians should check immunoglobulin levels if unusual or recurrent infections occur in patients treated with AEDs.
METHODS: We conducted an unmatched case-control study using data linkage of routinely collected biochemistry, prescribing, and morbidity datasets in North-East Scotland from 2009-2021. Cases were participants with immunoglobulin levels less than the reference range. Controls were those with normal/high immunoglobulin levels. Logistic regression was used to investigate associations between AED exposure and any hypogammaglobulinaemia, adjusting for age, sex, and comorbidity. We also analysed low IgA, IgM, and IgG separately. We analysed \"any AED\" exposure and common individual drugs separately. Cumulative exposure data were used to determine whether an exposure-response relationship was present.
RESULTS: 18,666 cases and 127,157 controls were identified. Use of any AED was associated with increased risk of hypogammaglobulinaemia (adjusted odds ratio [aOR] 1.20 [95% CI: 1.15-1.25]). Phenytoin use was strongly associated with low IgA (aOR 5.90 [95% CI: 3.04, 10.43]). Carbamazepine and lamotrigine were also associated with low IgA. Apart from topiramate, which was associated with a non-significant decrease in odds of hypogammaglobulinaemia, there was a consistent increase in odds of hypogammaglobulinaemia across most AEDs studied. Panhypogammaglobulinaemia was associated with any AED use, carbamazepine, lamotrigine, gabapentin, and multiple AED use. There was evidence of an exposure-response relationship between any AED use and any hypogammaglobulinaemia, low IgA, and low IgG. Carbamazepine and probably lamotrigine also had an exposure-response relationship with any hypogammaglobulinaemia.
CONCLUSIONS: AEDs may increase hypogammaglobulinaemia risk. Specific classes of immunoglobulins are differentially affected, and the exposure-response analysis suggests this may be causal. Further work should investigate the clinical impact of these findings. Clinicians should check immunoglobulin levels if unusual or recurrent infections occur in patients treated with AEDs.
摘要：
目的：已证实因感染（肺炎除外）引起的癫痫死亡率增加。使用抗癫痫药物（AEDs）的小病例系列描述了低丙种球蛋白血症，这可能会导致感染。目前尚不清楚低丙种球蛋白血症是否在AED患者中更常见，它与什么AED有关，或者它有什么临床影响。在这项以人群为基础的研究中,我们的目的是确定AED是否与低丙种球蛋白血症有关，哪些AED与之相关，以及这些关联是否可能是因果关系。
方法：我们使用常规收集的生物化学数据链接进行了一项不匹配的病例对照研究，开处方,和2009-2021年苏格兰东北部的发病率数据集。病例为免疫球蛋白水平低于参考范围的参与者。对照是免疫球蛋白水平正常/高的那些。Logistic回归用于调查AED暴露与任何低丙种球蛋白血症之间的关联。调整年龄，性别,和合并症。我们还分析了低IgA，IgM,和IgG分开。我们分别分析了“任何AED”暴露和常见的单个药物。累积暴露数据用于确定是否存在暴露-反应关系。
结果：确定了18,666例病例和127,157例对照。使用任何AED都与低丙种球蛋白血症的风险增加相关（调整后比值比[aOR]1.20[95％CI：1.15-1.25]）。苯妥英的使用与低IgA密切相关(aOR5.90[95%CI:3.04,10.43])。卡马西平和拉莫三嗪也与低IgA有关。除了托吡酯,这与低丙种球蛋白血症的几率没有显着降低有关，在大多数研究的AED中,低丙种球蛋白血症的几率持续增加.泛低球蛋白血症与任何AED使用有关，卡马西平,拉莫三嗪,加巴喷丁,和多个AED使用。有证据表明，任何使用AED和任何低丙种球蛋白血症之间都存在暴露-反应关系，低IgA,低IgG卡马西平和拉莫三嗪也与任何低丙种球蛋白血症都有暴露反应关系。
结论：AED可能增加低丙种球蛋白血症的风险。特定类别的免疫球蛋白受到差异影响，暴露-反应分析表明这可能是因果关系。进一步的工作应该调查这些发现的临床影响。临床医生应检查免疫球蛋白水平，如果在接受AED治疗的患者中发生异常或复发性感染。
方法：我们使用常规收集的生物化学数据链接进行了一项不匹配的病例对照研究，开处方,和2009-2021年苏格兰东北部的发病率数据集。病例为免疫球蛋白水平低于参考范围的参与者。对照是免疫球蛋白水平正常/高的那些。Logistic回归用于调查AED暴露与任何低丙种球蛋白血症之间的关联。调整年龄，性别,和合并症。我们还分析了低IgA，IgM,和IgG分开。我们分别分析了“任何AED”暴露和常见的单个药物。累积暴露数据用于确定是否存在暴露-反应关系。
结果：确定了18,666例病例和127,157例对照。使用任何AED都与低丙种球蛋白血症的风险增加相关（调整后比值比[aOR]1.20[95％CI：1.15-1.25]）。苯妥英的使用与低IgA密切相关(aOR5.90[95%CI:3.04,10.43])。卡马西平和拉莫三嗪也与低IgA有关。除了托吡酯,这与低丙种球蛋白血症的几率没有显着降低有关，在大多数研究的AED中,低丙种球蛋白血症的几率持续增加.泛低球蛋白血症与任何AED使用有关，卡马西平,拉莫三嗪,加巴喷丁,和多个AED使用。有证据表明，任何使用AED和任何低丙种球蛋白血症之间都存在暴露-反应关系，低IgA,低IgG卡马西平和拉莫三嗪也与任何低丙种球蛋白血症都有暴露反应关系。
结论：AED可能增加低丙种球蛋白血症的风险。特定类别的免疫球蛋白受到差异影响，暴露-反应分析表明这可能是因果关系。进一步的工作应该调查这些发现的临床影响。临床医生应检查免疫球蛋白水平，如果在接受AED治疗的患者中发生异常或复发性感染。
