GME Logo

User login

User menu

Issue 45, Jan 2016
Share This
Does vortioxetine really have a low incidence of sexual dysfunction?

Jacobsen et al. Treatment-emergent sexual dysfunction in randomized trials of vortioxetine for major depressive disorder or generalized anxiety disorder: a pooled analysis. CNS Spectr. 2015 Nov 17:1-12. PubMed PMID: 26575433.

Antidepressant-induced sexual dysfunction is perhaps the most important adverse effect of serotonergic antidepressants in longer-term treatment. It is an important cause of distress, impairment, and non-adherence to treatment.

If one or more newer antidepressants is clearly shown to have a lower incidence of sexual dysfunction than the SSRIs, this would be a big deal, right?


Vortioxetine is an antidepressant approved in the United States and Europe for the treatment of major depressive disorder.

Based on its pharmacological profile, it was hypothesized to have a limited adverse effect on sexual functioning. But did this actually pan out?


Raw data from seven short-term clinical trials of vortioxetine were combined. Six of these clinical trials were for the treatment of major depressive disorder and one for generalized anxiety disorder. How did they choose these particular studies? These were all the studies in which sexual functioning had been assessed using one particular rating scale.

All of these studies had used a simple, commonly used questionnaire for sexual functioning called the Arizona Sexual Experiences Scale (ASEX). For a brief introduction to this scale, see On this scale, sexual dysfunction is defined as any one of the following:

1. Total ASEX score ≥19,

2. A score of ≥5 on any individual ASEX item, or

3. A score ≥4 on any three ASEX items

Only those persons who did not already have sexual dysfunction at baseline were included. Emergence of sexual dysfunction in these persons was called “Treatment-emergent sexual dysfunction” and was compared between vortioxetine and placebo. The studies included a total of about 3800 patients and about 28% of them did not have sexual dysfunction at baseline as defined by ASEX.

In the same studies, duloxetine 60 mg/day was included as a comparator.  


The incidence of treatment-emergent sexual dysfunction increased with the dose of vortioxetine. Note: generally if as A increases, B increases, this suggests that A really is related to B.

The incidences of treatment-emergent sexual dysfunction were as follows: placebo 32%, vortioxetine 5 mg 26%, vortioxetine 10 mg 35%, vortioxetine 15 mg 43%, and vortioxetine 20 mg 46%, duloxetine 48%.

Statistically, vortioxetine 5 mg was specifically shown to be “non-inferior” to placebo in terms of causing treatment-emergent sexual dysfunction.

Vortioxetine 10, 15, and 20 mg did not meet that higher criterion for being shown to be non-inferior to placebo. But these higher doses were not shown to cause more treatment-emergent sexual dysfunction than placebo either.  Put together, what this means is that we can’t be sure either way whether or not the incidence of treatment-emergent sexual dysfunction with vortioxetine 10, 15, and 20 mg is equal to that with placebo.

Persons receiving duloxetine 60 mg/day had greater treatment-emergent sexual dysfunction than persons receiving vortioxetine 5 mg/day or 10 mg/day . But that sexual dysfunction was not more than that with vortioxetine 15 mg/day or 20 mg/day.

A different way of looking at treatment-emergent sexual dysfunction is to look at worsening of sexual functioning in the persons who already had sexual dysfunction prior to starting in the study. These were over 70% of the patients. None of the doses of vortioxetine had statistically significantly greater worsening of sexual dysfunction than placebo, but duloxetine did.

Now, you may say to yourself that you are not familiar with the Arizona Sexual Experiences Scale and are not sure what the cut offs of this questionnaire really mean. In that case, a simpler (though less statistically powerful) way is to ask: what percentages of patients came in and when asked how they were doing, spontaneously reported that they had some form of treatment-emergent sexual dysfunction. The percentages of persons reporting at least one adverse event related to sexual dysfunction were: placebo 1%, vortioxetine 2.2%, duloxetine 5.7%. But keep in mind that these percentages refer to when the persons were not specifically asked about sexual dysfunction but spontaneously reported it. That method of eliciting sexual dysfunction is notorious for failing to identify many persons with sexual dysfunction. 


The authors concluded that in short-term clinical trials, rates of treatment-emergent sexual dysfunction were not statistically significantly different between persons on vortioxetine or placebo. 

Clinical Commentary

Given the numerical differences and failure of higher doses (15 mg/day and 20 mg/day) to meet non-inferiority criteria, I think that vortioxetine may cause more sexual dysfunction than placebo, at least at doses of 15 mg/day and 20 mg/day.

But the rates are low and demonstrably lower than those for duloxetine. 

Expand Content
Share this Section:
Are higher doses of SSRIs better than lower doses? If so, how high?

Jakubovski et al. Systematic Review and Meta-Analysis: Dose-Response Relationship of Selective Serotonin Reuptake Inhibitors in Major Depressive Disorder. Am J Psychiatry. 2015 Nov10:appiajp201515030331. [Epub ahead of print] PubMed PMID: 26552940.

It has never been clear whether there is a dose response relationship with SSRIs.

This is an everyday problem for clinicians who prescribe SSRIs. When a person does not respond adequately to an SSRI, should we increase the dose or should we move to other strategies? 


Previous meta-analyses suggested that when antidepressants are used for the treatment of major depressive disorder -- within the recommended therapeutic dose range -- higher doses may not be better than lower doses. That is, as long as the minimum therapeutic dose was given, there was no advantage of giving higher doses even within the accepted dose range.

This study aimed to systematically review all the literature on this topic to provide an up-to-date answer to this question. It used a different methodology and also included several clinical trials that were published after the previous meta-analyses were done. 


A search of the MEDLINE database and a comprehensive database of controlled clinical trials was done. Randomized, placebo-controlled clinical trials that looked at the efficacy of SSRIs for the short-term treatment of adults (19 to 60 years old) with major depressive disorder were identified. 


Forty studies involving over 10,000 participants were included in this analysis.

Using complex statistical methods, a small but statistically significantly greater response with higher doses was found. How much more? Comparing doses higher in the therapeutic range to doses lower within the therapeutic range, the odds ratio was 1.3. 

Because fluvoxamine may have a different dose-response relationship, a reanalysis was done after fluvoxamine was excluded; the results did not change.  

However, the higher doses of SSRIs were also associated with an increased likelihood of dropping out of the study due to side effects.


The authors concluded that higher doses of SSRIs appear slightly more effective for the treatment of major depressive disorder.

However, this benefit appears to plateau, i.e, not increase further, when a dose of about 50 mg of fluoxetine or its equivalent is reached. 

Clinical Commentary

Let’s keep in mind that this data applies to:

a) short-term treatment,

b) groups of patients, and

c) persons who had not already had an inadequate response to that SSRI.

Is it possible that higher doses may have a slight advantage in the short-term, but in the longer-term both lower and higher doses are equally efficacious?

Is it possible that individual persons may have a marked benefit from higher doses?

Lastly, perhaps the most important question is whether a person who has had an inadequate response to a lower dose of an SSRI would benefit from an increase in the dose. This question was beyond the purview of the present study. Meta-analyses of data on this question have yielded contradictory results.

Expand Content
Share this Section:
Major depressive disorder: Is “much better” good enough?

Judd et al. A new empirical definition of major depressive episode recovery and its positive impact on future course of illness. J Clin Psychiatry. 2015 Oct 27. [Epub ahead of print] PubMed PMID: 26580150. 

Despite availability and widespread use of antidepressant medications, outcomes for the treatment of major depressive disorder are not that great. What can we do about this?

Maybe one thing we can do is to be more aggressive in our treatment and aim for 100% resolution of symptoms?

Will that affect longer-term outcomes? Most studies we come across are of a relatively short duration. Would you like to know whether complete resolution of symptoms makes a difference over 31 years of follow up?


Aim: This is the first study to provide direct, head-to-head data on the predictive significance of two different levels of recovery from a major depressive episode.

Also, this study provides a more scientific answer to the question: when should we consider the major depressive episode to be over? 


From 1978 to 1981, participants with a major depressive episode were enrolled in a famous long-term study called the National Institute of Mental Health Collaborative Depression Study. This study has yielded many dozens of papers and a lot of what we know today about the depressive disorders.

This particular paper focuses on 322 of these persons who were followed for up to 31 years. These persons were divided into two groups depending on the extent of their recovery from the major depressive episode:

1. Eight consecutive weeks of being asymptomatic (“Asymptomatic recovery”)

2. Residual subsyndromal depressive symptoms.


61% of the participants had an asymptomatic recovery from their major depressive episode.

Those who had an asymptomatic recovery remained free of relapse or recurrence four times longer than those who had residual subsyndromal depressive symptoms – for a median of 135 versus 32 weeks. That’s a big difference!

These persons had had shorter and less severe major depressive episodes and less additional psychopathology when they entered the study.

During follow up over 10 or 20 years, they had less psychosocial dysfunction and less burden of depression.

Now, here’s a powerful fact: having an asymptomatic recovery from the initial major depressive episode predicted long term course better than any of the 18 other predictors evaluated and better than all 18 combined!

How long should this asymptomatic recovery last before we consider the episode to be over? In terms of long term outcomes, having had four consecutive weeks of asymptomatic recovery was almost as good as having had eight weeks.


Based on this data, four consecutive weeks of asymptomatic recovery should be taken to denote the end of a major depressive episode.

When the major depressive episode improves the episode should not be considered to have ended if residual symptoms are still present. 

Clinical Commentary

Two important points emerge from this paper:

1. I always hear people saying, “She’s doing much better on this medication.” We need to change that to something like, “She’s better, but not completely so. She still has the following symptoms.” The goal should be 100% improvement and we must continue to tweak the treatment to try to achieve that.

2. What this paper proposes regarding when a major depressive episode should be considered to have ended (four consecutive weeks of being asymptomatic) is very different from that in DSM-5. The DSM-5 states: “For an episode to be considered recurrent, there must be an interval of at least 2 consecutive months between separate episodes in which criteria are not met for a major depressive episode.”  Not asymptomatic. Just not meeting the criteria for a major depressive episode. 

Expand Content
Share this Section:
Does long-term treatment with lithium harm the kidney or not?

Clos et al. Long-term effect of lithium maintenance therapy on estimated glomerular filtration rate in patients with affective disorders: a population-based cohort study. Lancet Psychiatry. 2015 Dec;2(12):1075-83. PubMed PMID: 26453408.

Lithium remains an important treatment for bipolar disorder and its efficacy data is very good.

However, reluctance to prescribe lithium as a first line treatment is not due to doubts about its efficacy, but mainly due to concerns about its potential adverse effects.

Of these, perhaps the most serious concern is that of potential renal impairment and chronic renal failure.

The study discussed here has, therefore, received a lot of attention in the press and in psychiatry-related websites and magazines.


The effect of long-term lithium treatment on renal function has been debated for many years.

Glomerular Filtration Rate is a key measure of renal function. “eGFR” is estimated glomerular filtration rate which is calculated using serum creatinine and a few demographic variables.

This study aimed to:

1. Assess the effect of long-term lithium treatment on estimated glomerular filtration rate (eGFR) in patients with affective disorders, and

2. See if any predictors for a decrease in eGFR can be identified.


This was a population-based cohort study conducted in Scotland, UK.

It was done retrospectively using a health database.

Adult patients (18-65 years of age at baseline) were included.

Participants had recently started maintenance treatment with lithium or other medications for bipolar disorder (quetiapine, olanzapine, valproate)

Patients had to have taken lithium or one of the comparator medications for at least 6 months.

Patients who had taken lithium or one of the comparator medications previously or who had preexisting kidney disease were excluded.

Patients were followed up in the database for between 6 months and 12 years. 


305 persons who had taken lithium and 815 persons who had taken comparator medications were included.

The mean duration of exposure to lithium was 55 months (range 6 to 144 months).

The mean annual decline in eGFR (adjusted for age, sex, and baseline eGFR) was slightly larger with lithium but this difference was not statistically significantly different from that with one of the comparator medications, i.e., may have occurred by chance.

Who was more likely to have greater decline in eGFR? The predictors of decline in eGFR in this study were found to be:

  1. Increasing age
  2. Higher baseline eGFR
  3. Medical comorbidities
  4. Also taking a nephrotoxic medication
  5. Episodes of lithium toxicity

Comment: Other than lithium toxicity, these are probably predictors of decline in eGFR even in persons not taking lithium.

In this study, the duration of taking lithium and the mean serum lithium level were not found to predict decline in eGFR.


The authors concluded that there was no effect of lithium maintenance (in therapeutic range) on the rate of decline in eGFR.

They noted that their results, therefore, “contradict the idea that long-term lithium therapy is associated with nephrotoxicity in the absence of episodes of acute intoxication and that duration of therapy and cumulative dose are the major determinants of toxicity.”

Clinical Commentary

So, are you surprised by these results? Before we accept the conclusions of this study, let’s pause and look more carefully at how the study was done. Usually, in GME Research Review I only include articles whose conclusions I agree with and which have some action implication for clinical practice. But I decided to discuss this article both because it received so much press and because it illustrates how methodology affects conclusions.

We know from clinical experience that lithium does not cause renal impairment in a few months; it appears after several years of being on the lithium. Participants in this study were on lithium for a mean of 55 months (range 6 to 144 months). So, half of the patients had been on lithium for less than 5 years, many of them for less than one year. The authors did try to look at the effect of duration of treatment and did not find any effect. But they can only look at the effect of duration of treatment within the limitations of who was included in the study.

Another methodological problem is that there were only 305 persons on lithium.  When looking at an adverse event that occurs infrequently, it is hard to be sure without studying a larger sample. In other words, it is hard to identify uncommon events in prospective studies with small samples. Let’s keep in mind that the decline is eGFR was greater in persons taking lithium, though we cannot rule out the possibility that it was just due to chance.

Expand Content
Share this Section:
Lithium and Chronic Kidney Disease

Kessing et al. Use of Lithium and Anticonvulsants and the Rate of Chronic Kidney Disease: A Nationwide Population-Based Study. JAMA Psychiatry. 2015 Dec 1;72(12):1182-91. PubMed PMID: 26535805.

We discussed the limitations of one study about the effects of lithium on the kidney that was published in December 2015. In the same month, another study of the same subject published in a different journal and reached somewhat different conclusions.

But this study had some surprising results too!


Aim: To compare rates of chronic kidney disease in persons who took lithium, anticonvulsants, or other medications for bipolar disorders.


This study was done in Denmark and included the entire country. It is nice that researchers can do that in some countries that have various nationwide databases.

The following groups of individuals were studied:

1. Randomly selected sample of persons living in Denmark (n= 1.5 million)

2. All persons (n =10,591).with a diagnosis of either a single manic episode or of bipolar disorder (during a specified period).

3. All persons who took either lithium (n = 26,731) or anticonvulsants (n=420,959).

The outcome measures studied were:

1. Possible chronic kidney disease

2. Definite chronic kidney disease

3. End-stage chronic kidney disease (defined as being on long-term dialysis or requiring renal transplantation).


In both the total sample and in the subgroup of persons with bipolar disorder, use of lithium was associated with both an increased rate of having Possible chronic kidney disease and of having Definite chronic kidney disease. In persons who had taken 60 prescriptions or more of lithium, the hazard ratio for having Definite chronic kidney disease was increased by 2.5 to 3.7 times.

In the total sample, use of anticonvulsants, antipsychotics, or antidepressants was not associated with increased hazard of having chronic kidney disease. But in the subgroup of persons with bipolar disorder, use of anticonvulsants was associated with an increased rate of both Possible and Definite chronic kidney disease. No association was found between use of antipsychotics or antidepressants and having chronic kidney disease.

In the total sample, neither use of lithium nor use of any other class of medications was associated with increased rates of End-stage chronic kidney disease. In the subgroup of persons with bipolar disorder, use of lithium was not associated with an increased rate of End-stage chronic kidney disease, but use of anticonvulsants was. In persons who had taken 60 prescriptions or more of anticonvulsants, the hazard ratio for having End-stage chronic kidney disease was doubled. 


Longer-term treatment with lithium or anticonvulsants is associated with an increased rate of chronic kidney disease.

However, these authors concluded use of lithium was not associated with an increased rate of End-stage chronic kidney disease. 

Clinical Commentary

This study highlights an important point that persons with bipolar disorder appear to have a higher likelihood of developing chronic kidney disease irrespective of medications given.

But, use of lithium is associated with development of chronic kidney disease over and above the increased risk in persons with bipolar disorder.

Note that there were only 62 persons in the bipolar disorder subgroup who developed End-stage chronic kidney disease and not all of them, of course, had taken lithium. Therefore, I don’t agree with the authors that this study shows that while use of lithium is associated with chronic kidney disease, it is not associated with End-stage chronic kidney disease.

The association between use of anticonvulsants and development of chronic kidney disease is confusing. One possible explanation is that persons with early kidney disease or multiple risk factors for chronic kidney disease may be put on anticonvulsants rather than on lithium. 

Expand Content
Share this Section:

Rajnish Mago, MD
Associate Professor of Psychiatry, Thomas Jefferson University

GME Research Review is a monthly newsletter edited by Rajnish Mago, MD, who is Associate Professor of Psychiatry at Thomas Jefferson University and is author of "The Latest Antidepressants and Side Effects of Psychiatric Medications: Prevention, Assessment, and Management." Dr. Mago selects, summarizes, and provides a clinical commentary on the latest published research in psychiatry.

We are always carefully evaluating which research papers to discuss in GME Research Review. Have you come across a research paper published in the last 6 months that you think is clinically relevant? If you would like to ask Dr. Mago to consider analyzing it, please email him the citation at: [email protected]

To contact GME, email us at [email protected]

GME does not provide medical advice. The website and articles are intended for informational purposes only. They are not a substitute for professional medical advice, diagnosis or treatment. Never ignore professional medical advice in seeking treatment because of something you have read on the GME Website. If you think you may have a medical emergency, immediately call your doctor or dial 911.

Log In

Join GME For Free

Become a GME subscriber and gain full access to our extensive library of 700+ psychiatric medical education videos, free CME webcasts, latest research updates, and more. To sample our content, watch the featured videos below.
Join Today!