According to prior studies, the choice of initial antidepressant medication may be influenced by patient characteristics such as older age, illness severity, prior suicide attempts, and presence of a comorbid anxiety disorder . The question of how to proceed with the next step after initially unsuccessful AD treatment is crucial. In our study, switching to a different antidepressant drug was the most common strategy used to treat MDD patients with an inadequate response to first-line treatment (39.6%).
The clinicians’ preference for switching to a different antidepressant has previously been described . Currently, there is not enough clinical evidence to make a clear recommendation about the best next-step treatment in non-responders. However, changing to another class of antidepressant is a favoured option, although the change to another SSRI or to a tricyclic agent could also be used [19, 20]. The lower number of drugs involved with switching strategies could avoid potential interactions and assure treatment adherence . Such advantages of switching over the two other strategies may explain its preference among psychiatrists.
Despite the lack of clear evidence-based recommendations, we observed that some clinical characteristics seem to have been considered by psychiatrists when they decided to adopt a new pharmacological approach. Thus, an augmentation strategy seemed to be the preferential therapeutic option for patients with a worse clinical profile. According to the multivariate analysis in our study, patients with more previous depressive episodes and/or psychotic symptoms were more likely to receive augmentation therapy. No randomized controlled trials have directly compared switching and augmentation strategies. The most valuable evidence comes from the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial, a four-level study that evaluated different strategies in non-responders . The results of a retrospective analysis comparing switching and augmentation strategies in patients participating in the STAR*D trial has recently been published . Such analysis showed that patients who complete initial treatment of 12 weeks or more and have a partial response with residual mild depressive symptoms may benefit more from augmentation than switching.
The STAR*D trial also showed that patients with longer depressive episodes were less likely to achieve remission . Guidelines drawn up by scientific associations on the management of MDD patients recommend that, when remission is not achieved with an AD after 6 to 8 weeks of treatment at an adequate dose, this should be changed [5, 20, 23]. The timing of adopting a new pharmacological approach in our study was greater than recommended: mean duration of AD treatment at baseline was 12.0 weeks (SD 11.1). The variability observed by psychiatrists in their clinical attitude towards patients with major depression has previously been described at primary care level [24, 25]. A recent publication by Chang et al. found little active management among MDD patients treated in primary care centres in the USA . General practitioners were not more likely to adjust therapy, even when feedback regarding their patients´ symptoms indicated an inadequate response .
The fact that approximately one half of patients do not respond to first-line treatment with antidepressants constitutes a major concern for psychiatrists in daily clinical practice [12, 27].
Although widely used and usually safe, the efficacy of even the most widely prescribed combinations of antidepressants has not been established by properly controlled clinical trials . Among patients who did not achieve remission with initial citalopram treatment in the STAR*D trial, approximately one-third of patients augmenting treatment and one-quarter of patients switching antidepressant achieved subsequent remission .
Lithium is one of the oldest agents used to improve clinical outcomes in patients with inadequate responses from antidepressant trials . However, in our study only 5 patients received lithium (8.3% among the total of augmentation strategies). The adverse event profile, low serum therapeutic index, and long-term risks of thyroid and renal compromise associated with this drug might explain its poor uptake in our sample . Among the different augmentation strategies, augmentation with atypical antipsychotics has been the most studied approach over recent years. The use of adjunctive atypical antipsychotic drugs is the strategy with the largest evidence for AD non-responders with a pooled remission rate of approximately 47% . However, the potential benefits of these medications must be weighed up against their known acute and long-term risks, such as extrapyramidal symptoms and metabolic side effects . In refractory MDD studies, quetiapine extended release 150 mg/d and 300 mg/d had significantly higher risks for sedation: the numbers needed to treat to harm (NNTH) were 9 and 7, respectively . Aripiprazole adjunctive therapy to antidepressants resulted in a significantly higher risk for akathisia relative to placebo (NNTH of 5, 95%CI 4–7) . In addition, fluoxetine plus olanzapine resulted in significantly greater weight gain and increase in total cholesterol compared with fluoxetine monotherapy .
The present manuscript has several limitations. First, our study analyzed a total of 273 patients and probably this small sample size could have limited the finding of additional associations. Second, we only analyzed outpatients receiving care from psychiatrists. Therefore, our findings may neither generalize to mild depressive patients, mainly managed in a primary care setting, nor the inpatient population. Third, comorbidity, ethnicity, and adverse events associated with the first antidepressant were not assessed, which may also be relevant factors affecting MDD management.