Lagerberg Tyra, Sjölander Arvid, Gibbons Robert D, Quinn Patrick D, D'Onofrio Brian M, Hellner Clara, Lichtenstein Paul, Fazel Seena, Chang Zheng
Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.
Departments of Medicine and Public Health Sciences, Center for Health Statistics, University of Chicago, Chicago, IL, United States.
Front Psychiatry. 2022 Nov 9;13:1012650. doi: 10.3389/fpsyt.2022.1012650. eCollection 2022.
Using other central nervous system (CNS) medications in combination with selective serotonin reuptake inhibitor (SSRI) treatment is common. Despite this, there is limited evidence on the impact on suicidal behavior of combining specific medications. We aim to provide evidence on signals for suicidal behavior risk when initiating CNS drugs during and outside of SSRI treatment.
Using a linkage of Swedish national registers, we identified a national cohort of SSRI users aged 6-59 years residing in Sweden 2006-2013. We used a two-stage Bayesian Poisson model to estimate the incidence rate ratio (IRR) of suicidal behavior in periods up to 90 days before and after a CNS drug initiation during SSRI treatment, while accounting for multiple testing. For comparison, and to assess whether there were interactions between SSRIs and other CNS drugs, we also estimated the IRR of initiating the CNS drug without SSRI treatment.
We identified 53 common CNS drugs initiated during SSRI treatment, dispensed to 262,721 individuals. We found 20 CNS drugs with statistically significant IRRs. Of these, two showed a greater risk of suicidal behavior after versus before initiating the CNS drug (alprazolam, IRR = 1.39; flunitrazepam, IRR = 1.83). We found several novel signals of drugs that were statistically significantly associated with a reduction in the suicidal behavior risk. We did not find evidence of harmful interactions between SSRIs and the selected CNS drugs.
Several of the detected signals for reduced risk correspond to drugs where there is previous evidence of benefit for antidepressant augmentation (e.g., olanzapine, quetiapine, lithium, buspirone, and mirtazapine). Novel signals of reduced suicidal behavior risk, including for lamotrigine, valproic acid, risperidone, and melatonin, warrant further investigation.
将其他中枢神经系统（CNS）药物与选择性5-羟色胺再摄取抑制剂（SSRI）联合使用是常见的治疗方式。尽管如此，关于特定药物联合使用对自杀行为影响的证据有限。我们旨在提供在SSRI治疗期间及之外开始使用中枢神经系统药物时自杀行为风险信号的证据。
利用瑞典国家登记数据库的链接，我们确定了2006年至2013年居住在瑞典的6至59岁SSRI使用者的全国队列。我们使用两阶段贝叶斯泊松模型来估计在SSRI治疗期间开始使用中枢神经系统药物前90天及之后一段时间内自杀行为的发病率比（IRR），同时考虑多重检验。为了进行比较，并评估SSRI与其他中枢神经系统药物之间是否存在相互作用，我们还估计了未进行SSRI治疗时开始使用中枢神经系统药物的IRR。
我们确定了在SSRI治疗期间开始使用的53种常见中枢神经系统药物，这些药物被配发给262,721人。我们发现20种中枢神经系统药物的IRR具有统计学意义。其中，两种药物在开始使用中枢神经系统药物后比之前显示出自杀行为风险更高（阿普唑仑，IRR = 1.39；氟硝西泮，IRR = 1.83）。我们发现了几种与自杀行为风险降低有统计学显著关联的新信号药物。我们没有发现SSRI与所选中枢神经系统药物之间存在有害相互作用的证据。
检测到的几个风险降低信号对应的药物，之前有证据表明其对增强抗抑郁作用有益（例如奥氮平、喹硫平、锂盐、丁螺环酮和米氮平）。自杀行为风险降低的新信号，包括拉莫三嗪、丙戊酸、利培酮和褪黑素，值得进一步研究。
