Background: Prior studies have shown that paroxetine and fluoxetine may exacerbate opioid-induced respiratory depression. The mechanism is not clear; moreover, little is known whether other selective serotonin reuptake inhibitors (SSRIs) may cause similar effects.
Objectives: To compare the risk of opioid overdose across different SSRIs when added to oxycodone therapy.
Methods: Using three US databases (IBM MarketScan, Medicaid, and fee-for-service Medicare; spanning 2000-2020), we examined the risk of emergency room visits or hospitalizations for opioid overdose in adults who initiated SSRIs while on oxycodone therapy. Patients were required to have at least 180 days of continuous enrollment prior to SSRI initiation (index date) and an active oxycodone prescription at the time of initiation. Patients were allocated to one of 4 treatment groups (fluoxetine, paroxetine, citalopram/escitalopram, sertraline) based on the SSRI initiated and followed from the day after initiation until outcome occurrence, change in therapy, death, or end of data/enrollment (primary analysis). Propensity scores (PS) were estimated as a function of baseline covariates and used to calculate matching weights. Weighted Cox proportional hazards regression was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for 3 treatment groups relative to sertraline within each database and in the pooled data, stratified on database. Secondary analyses included a 60-day intention-to-treat (ITT) analysis and alternative outcome definitions that (1) excluded heroin, opium, and methadone overdoses, or (2) added codes indicative of respiratory problems.
Results: The study cohort included 136,449 fluoxetine, 92,254 paroxetine, 342,944 citalopram/escitalopram, and 237,318 sertraline initiators with 154, 145, 147, and 126 outcomes, respectively, in the primary weighted analysis. The weighted HR of opioid overdose relative to sertraline was 1.20 (95% CI, 1.00 - 1.45) for fluoxetine, 1.18 (95% CI, 0.96 - 1.44) for paroxetine, and 1.16 (95% CI, 1.00 - 1.36) for citalopram/escitalopram. Intention-to-treat 60-day analysis yielded similar results for fluoxetine (HR 1.20; 95% CI 1.01 - 1.43), but higher estimates for paroxetine (HR 1.29; 95% CI 1.07 - 1.54) and citalopram/escitalopram (HR 1.22; 95% CI 1.06 - 1.41). The results were consistent across sensitivity analyses with alternative outcome definitions.
Conclusions: There appears to be no substantial difference in the risk of opioid overdose across individual SSRIs in patients on oxycodone therapy. Limitations include short duration of concomitant treatment, missing outcomes for patients that died prior to hospital admission, and missing information on illicit drug use.
