Rylie is a 20 yo woman in a first episode of major depression (severe, PHQ9-22). She started weekly psychotherapy and connected with her PCP to start escitalopram, titrated to 20mg daily. After 8 weeks of therapy + escitalopram 20mg, her depressive symptoms have worsened slightly (PHQ9-24).
Rylie learns that her father takes duloxetine for depression and switches from escitalopram 20mg to duloxetine 60mg daily (cross-tapered over 2 weeks) with ~40% improvement in her symptoms after 8 weeks. An additional 4-week trial of duloxetine 90mg does not lead to further improvement. The duloxetine dose is reduced to 60mg, and Rylie starts to walk for 45 minutes 3 times a week. Her depressive symptoms do not improve further.
Is adding bupropion to an SSRI/SNRI an effective depression augmentation strategy?
Bupropion is an attractive option for depression augmentation. The combination of an SSRI/SNRI and bupropion is safe and generally well-tolerated, and clinicians in primary care are comfortable prescribing it. Bupropion at 300mg daily can also help sexual side effects (in particular low libido) caused by an SSRI/SNRI.
A systematic review and meta-analysis published in JAMA in 2022 looked at whether antidepressant combinations (i.e. 2 antidepressants) were better than antidepressant monotherapy for the treatment of major depression. Antidepressant combinations that included bupropion were NOT associated with improved antidepressive efficacy compared with monotherapy (SMD 0.10; 95% CI -0.07 -0.27) (Hennsler, 2022).
Let’s look more closely at the evidence.
Open label trials of augmenting an SSRI with bupropion showed positive outcomes for depression dating back to the 1990s. By 2000, bupropion was the most widely chosen augmenting agent among clinicians for a patient with MDD who did not respond to an SSRI (33% of 801 surveyed clinicians) (Mischoulon, 2000).
The aforementioned 2022 meta-analysis included 7 studies that looked at an SSRI + bupropion in the treatment of MDD, either started together (combination treatment) or added sequentially after an inadequate response to the first antidepressant. The forest plot for the difference in effect size (expressed as the standard difference in means) between combination and monotherapy for these studies is shown below. Risk of bias (RoB) is a summary indicator of the risk of bias from each study. We’ll take a closer look at the 2 largest studies with a low risk of bias (Rush, Stewart). Neither of these studies found a benefit to the combination of bupropion plus an SSRI vs monotherapy with an SSRI.
The CO-MED study included patients (n=665) from 9 psychiatry clinics and 6 primary care clinics. Patients with moderately severe depression were randomized to 3 different antidepressant combinations that were flexibly dosed.
- escitalopram monotherapy (to 20mg) + placebo pill
- bupropion SR (to 200mg BID) + escitalopram (to 20mg/day)
- venlafaxine XR (to 300mg/day) + mirtazapine (to 45mg/day)
The first pill was started, unblinded to patients and treating physicians. The second pill (including placebo) was started a few weeks later, blinded to patients. Response and remission rates at 12 weeks and 7 months are shown below, with the y-axis indicating the percent of participants who achieved remission or response (Rush, 2011).
The 12-week remission rates for escitalopram + placebo vs bupropion + escitalopram were 38.8% and 38.9% (Rush, 2011).
Stewart et al looked at remission rates in a group of 245 adults with MDD who were randomized to double-blind treatment with 3 different flexibly-dosed antidepressant combinations.
- escitalopram (to 40mg) + placebo
- bupropion (to 450mg) + placebo
- escitalopram (to 40mg) + bupropion (to 450mg)
The titration to the max dose of both medications was attained by week 4, unless tolerability issues prevented dose increases or the patient’s depression was in remission.
The remission rates at 12 weeks were: 52% (escitalopram monotherapy), 34% (bupropion monotherapy), and 46% (escitalopram + bupropion combination). Combination therapy was superior to bupropion monotherapy at several time points, but not superior to escitalopram monotherapy at any time point. At 12 weeks, there was no difference in the remission rates between the escitalopram + bupropion combination group and either monotherapy group (Stewart, 2014).
In addition, two systematic reviews and network meta-analyses of augmentation medications for adults with major depression both concluded that bupropion was no more likely than placebo to result in depression response or remission (Zhou 2015, Nunez 2022).
To summarize, despite the widespread popularity of combining bupropion with an SSRI/SNRI , our existing studies suggest that this strategy is NOT effective for depression augmentation.
But this contradicts my clinical experience. I’ve seen bupropion help my patients when added to an SSRI.
Mine too! When we prescribe a medication and a patient gets better, both the clinician and the patient may think that the medication is responsible. Depression can also resolve without any specific treatment (ie the natural course of the illness), or due to the “non-medication” parts of the treatment plan (ie talking with a clinician, care from a supportive clinic, etc.). Placebo groups in randomized controlled trials attempt to bundle these non-drug variables and determine the specific efficacy of a drug. If the answer from multiple reasonably well-done RCTs is that a specific drug effect does not differ from placebo, we need to take it seriously, even if some textbooks and clinical guidelines have not yet been appropriately updated.
What about other combinations of antidepressants – like an SSRI + mirtazapine?
The evidence is less negative than with bupropion, but still mixed. In the 2022 JAMA meta-analysis, combinations of antidepressants that included a serotonin reuptake inhibitor (SSRI, SNRI, TCA) and an antagonist of presynaptic α2-autoreceptors (like mirtazapine) were superior to monotherapy with an antidepressant (Standard Mean Difference 0.37; 95% CI, 0.19-0.55) (Hennsler, 2022).
In contrast, the 2019 Cochrane meta-analysis asking “Are there effective medications for treating depression that does not improve with the first medicine used?” had more stringent inclusion criteria than the Hennsler study and found high quality evidence that mirtazapine augmentation does not work. This conclusion was based on data from one large (n=490) double blind placebo-controlled RCT of depression augmentation with mirtazapine 30mg (Davies, 2019).
Are there medications that should be used for depression augmentation instead of bupropion?
Augmentation strategies with some second generation antipsychotics, in particular aripiprazole and quetiapine, and with lithium have consistent or increasing levels of evidence to support their use. To learn more, check out the Nuñez 2022 meta-analysis or stay tuned for future PsychSnaps.
How do you put this all together?
The addition of bupropion to an SSRI/SNRI is likely to be well-tolerated and may improve sexual side effects caused by the serotonergic antidepressant. The majority of our current evidence suggests that adding bupropion to an SSRI/SNRI is NOT an effective augmentation strategy for major depression, despite remaining common clinical practice.
Back to Rylie: Rylie has severe depressive symptoms, no response to an adequate trial of escitalopram (20mg x8 weeks); a partial response to duloxetine (dose-optimized to 60mg daily); unimproved with regular exercise or psychotherapy.
Research suggests that bupropion augmentation is unlikely to be better than placebo augmentation. You recommend adding aripiprazole 2mg daily to the duloxetine and counsel her about the risk of akathisia (internal sense of restlessness) and weight gain. You plan to increase the aripiprazole dose to 5mg in 2-4 weeks if her depression has not improved.
Key Points
- Adding bupropion to an SSRI/SNRI is not an effective augmentation strategy for major depression.
- The data for mirtazapine as an augmentation agent for MDD is mixed.
- Aripiprazole, quetiapine, and lithium have high quality studies that support their use as augmentation agents for MDD.
Related PsychSnaps:
“How do you evaluate for bipolar disorder in depressed patients?” Emma Samelson-Jones, January, 2024.
“Does this patient’s history of binging and purging make it unsafe to start bupropion?” Emma Samelson-Jones, March, 2025.
References:
Davies, Philippa, et al. “Pharmacological interventions for treatment‐resistant depression in adults.” Cochrane Database of Systematic Reviews 12 (2019).
Gulrez, Gaurav, et al. “Bupropion as an augmenting agent in patients of depression with partial response.” Basic & clinical pharmacology & toxicology 110.3 (2012): 227-230.
Henssler, Jonathan, Tom Bschor, and Christopher Baethge. “Combining antidepressants in acute treatment of depression: a meta-analysis of 38 studies including 4511 patients.” The Canadian Journal of Psychiatry 61.1 (2016): 29-43.
Henssler, Jonathan, et al. “Combining antidepressants vs antidepressant monotherapy for treatment of patients with acute depression: a systematic review and meta-analysis.” JAMA psychiatry 79.4 (2022): 300-312.
Mischoulon, David, et al. “Strategies for managing depression refractory to selective serotonin reuptake inhibitor treatment: a survey of clinicians.” The Canadian Journal of Psychiatry 45.5 (2000): 476-481.
Nuñez, Nicolas A., et al. “Augmentation strategies for treatment resistant major depression: a systematic review and network meta-analysis.” Journal of affective disorders 302 (2022): 385-400.
Stewart, Jonathan W., et al. “Combination antidepressant therapy for major depressive disorder: speed and probability of remission.” Journal of psychiatric research 52 (2014): 7-14.
Rush, A. John, et al. “Combining medications to enhance depression outcomes (CO-MED): acute and long-term outcomes of a single-blind randomized study.” American Journal of Psychiatry 168.7 (2011): 689-701.
Zhou, Xinyu, et al. “Comparative efficacy, acceptability, and tolerability of augmentation agents in treatment-resistant depression: systematic review and network meta-analysis.” The Journal of clinical psychiatry 76.4 (2015): 3423.
Send a Comment