Skip to main content

Lower Doses Best for Most Second-Generation Antidepressants

June 21, 2019

By Will Boggs MD

NEW YORK—The optimal doses for the most commonly used second-generation antidepressants are at the lower range of their licensed doses, according to a new systematic review and meta-analysis.

"The big challenge is not only to prescribe the right antidepressant to each patient, but also to find the best dosage for each individual, optimizing efficacy and reducing side effects," Dr. Andrea Cipriani of the University of Oxford, U.K., told Reuters Health by email.

While second-generation antidepressants are first-line options in the pharmacological management of major depression, there is considerable uncertainty about their optimal target doses, and current guidelines provide conflicting recommendations.

Dr. Cipriani and colleagues used data from 77 studies that examined various fixed doses of second-generation antidepressants in more than 19,000 participants. The drugs included five selective serotonin-reuptake inhibitors (SSRIs: citalopram, escitalopram, fluoxetine, paroxetine and sertraline), venlafaxine and mirtazapine.

For all SSRIs, the relative risk (RR) for efficacy increased from 1.0 for placebo to 1.24 for 20 mg fluoxetine equivalents and 1.27 for 40 mg fluoxetine equivalents before showing a flat to decreasing trend through higher doses.

On the other hand, the RR for dropouts due to adverse effects increased from 1.0 for placebo to 1.94 at 40 mg fluoxetine equivalents and to 3.73 at the upper end of the licensed range (80 mg fluoxetine equivalents), the team reports in The Lancet Psychiatry, online June 6.

There was little heterogeneity among the SSRIs in their dose-outcome relationships.

For venlafaxine, the efficacy increased fairly steeply up to 75-150 mg and more modestly with higher doses, while for mirtazapine, the efficacy increased up to a dose of 30 mg and then decreased at higher doses.

For both drugs, dropouts due to adverse effects increased steeply with increasing doses, resulting in a dose-acceptability curve that yielded an optimum balance at approximately 20-40 mg of fluoxetine equivalents in both cases.

"Our study shows that, on average, the lower range of the licensed dose of the most commonly used second-generation antidepressants achieves the optimal balance between efficacy, tolerability, and acceptability in the acute treatment of adults with major depression," Dr. Cipriani said.

"In the U.K., the NICE guidelines state that there is no dose dependency within the therapeutic range of SSRIs; in the U.S., guidelines from the American Psychiatric Association recommend slow titration up to the maximum tolerated dose," he said. "This new evidence suggests neither of these methods is correct."

"These figures come with a caveat concerning the manner in which fixed-dose trials are designed," writes Dr. Fredrik Hieronymus of the University of Gothenburg, in Sweden, in a linked editorial. "As (the authors) state, the common use of rapid or no titration can lead to an increase in early dropouts, confounding efficacy with tolerability because participants tend to improve over time. Flexible-dose trials, which commonly use doses above those found to be optimal by (these researchers), have been reported to yield larger drug-placebo differences than fixed-dose trials, suggesting that optimization of the dose on the basis of the patient's need and tolerability, and occasionally using quite high doses when doing so appears tolerable, might be beneficial."

"In summary, one possible reading of the article is that medium doses are optimal," he writes. "It is difficult to imagine a confounder that would inflate the efficacy of medium (as compared with low or high) doses and converging evidence (from dose-escalation and fixed-dose studies) discourages us from routinely using high dose. Thus, regularly aiming for a dose close to the upper boundaries identified by (the authors) as optimal, while being attentive to intolerable side-effects, might be the optimal strategy for most patients."

Dr. Ymkje Anna de Vries of University Medical Center Groningen, in the Netherlands, who has studied antidepressant treatment, said she was surprised by the finding "that higher doses of venlafaxine are only a little more effective than lower doses."

"Given that venlafaxine also becomes a noradrenaline-reuptake inhibitor at higher doses, I would have expected to see a clearer dose-response relationship for venlafaxine," she told Reuters Health by email. "This might be related to the small number of trials for venlafaxine, though, so I think we need more research before we can be sure about the dose-response relationship for venlafaxine and other non-SSRI antidepressants."

"In the case of SSRIs, more isn't always better and might actually be worse," said Dr. de Vries, who was not involved in the new work. "The findings suggest that clinicians should generally prescribe relatively low doses of SSRIs (e.g., 20-40 mg of fluoxetine) and should not routinely increase the dose beyond that, as this is associated with increases in the number of people quitting the medication due to side effects and no clear increase in efficacy."

The study had no commercial funding. Three of Dr. Cipriani's coauthors report ties to antidepressant manufacturers.


Lancet Psychiatry 2019.

(c) Copyright Thomson Reuters 2019. Click For Restrictions -

Back to Top