Comparison of baseline characteristics between the two study groups
Table 1 compares the distribution of demographics and clinical characteristics of two groups. There were no significant differences in age, gender, duration of illness, and uses of antipsychotics, benzodiazepines, anxiolytics, and selective serotonin reuptake inhibitors (SSRIs) between the two groups. The uses of serotonin-norepinephrine reuptake inhibitors (SNRIs) or trazodone were higher in the non-mirtazapine group than those in the mirtazapine group (P <0.01). There were no differences between the two groups with respect to baseline level of thyroid function measurements (Table 1). On the other hand, baseline free T4 levels were negatively associated with risk of hypothyroxinemia, with adjusted RR= 0.88 (95% CI: 0.86-0.90) for 1 pmol/L increment in baseline free T4.
Analysis of association between mirtazapine use and occurrence of hypothyroxinemia
The rate of hypothyroxinemia in the mirtazapine group (33/88, 37.5%) was higher than in the non-mirtazapine group (26/132, 19.7%). After adjusting for age, gender, duration of treatment, baseline free T4 level, use of other antidepressants (SSRIs, SNRIs or trazodone), use of non-antidepressant medications and other potential confounders, patients who received mirtazapine as an adjunctive antidepressant were more likely to develop hypothyroxinemia compared with those who did not receive mirtazapine (adjusted RR: 1.64, 95% CI: 1.31-1.78) (Table 2).
Analysis of association between mirtazapine use and other thyroid dysfunctions
In the mirtazapine group, the reduction in T4 was 15.63 nmol/L, and was 3.22 pmol/L in free T4, which were significantly greater than those in non-mirtazapine group (t =-3.053, P =0.003, andt =-2.861, P =0.005, for ΔT4 and Δ free T4, respectively) (Table 3). In the multiple linear regression model, mirtazapine use was statistically significantly associated with greater reduction in free T4 level [(Δ free T4(2–1))] after adjusting for confounders, with adjusted B -1.14 (95% CI: -1.88~ -0.41, P =0.002). The occurrence of subclinical and overt hypothyroidism was similar in the two groups (Table 3).