(227) Risk assessment of decreased platelet count in patients prescribed antidepressants using the MID-NET in Japan

Background: Regarding Selective serotonin reuptake inhibitors (SSRIs), Serotonin and noradrenaline reuptake inhibitors (SNRIs), and vortioxetine hydrobromide, which are antidepressants, the status of issuance of precautions for decreased platelet count differs among the drugs in Japan. Although it would be valuable to evaluate the risk of decreased platelet count, it may be difficult to evaluate the causal relationship between the suspected drug and the adverse drug reaction based on individual case reports for each drug.

Objectives: To evaluate the risk of decreased platelet count for SSRIs, SNRIs, and vortioxetine hydrobromide in patients with depression, in comparison to that for paroxetine hydrochloride hydrate for which precaution for decreased platelet count is made in the Clinically Significant Adverse Reactions (CSAR) section in a package insert.

Methods: We conducted a new user cohort study using the Medical Information Database NETwork (MID-NET®) in Japan. We identified patients who were newly prescribed any of the SSRIs (escitalopram oxalate, sertraline hydrochloride, paroxetine hydrochloride hydrate, fluvoxamine maleate), SNRIs (duloxetine hydrochloride, venlafaxine hydrochloride, milnacipran hydrochloride), and vortioxetine hydrobromide between January 2009 and March 2021. In patients with a diagnosis of depression who were newly prescribed antidepressants, the adjusted hazard ratios weighted by the inverse of the propensity score in comparison to the paroxetine group were estimated using the Cox proportional hazard model for each combination of the exposure groups and comparator group to evaluate the risk of decreased platelet count during the prescription period.

Results: A number of patients who were newly prescribed antidepressants and not met the exclusion criteria was 10,470. Hazard ratios (HRs) could not be estimated for the vortioxetine group since no occurrence of outcomes was observed. Among drugs which no precaution for decreased platelet count is made in the CSAR in package inserts, the point estimate of the adjusted HR was higher than 1 for the sertraline group only (the adjusted HR of 1.23 (95% confidential intervals 0.78−1.94)). A similar trend was observed in sensitivity analyses. For the other groups, no significant increase in risk compared to the paroxetine group was observed.

Conclusions: It was suggested that the risk of decreased platelet count for sertraline hydrochloride is possibly comparable to that for paroxetine hydrochloride hydrate. Based on the results and individual case reports, the Ministry of Health, Labour and Welfare and the Pharmaceuticals and Medical Devices Agency concluded that the Precautions in a package insert should be revised for sertraline hydrochloride.