Let’s face it, dealing with side effects is not high on the list of “things we like most about psychiatry.” In this issue’s interview, Mark Zimmerman describes a study showing just how unenthusiastic we tend to be in ferreting out our patients’ side effects. The bottom line of his study was that patients on antidepressants reported 20 times more side effects than were recorded by their psychiatrists. It’s not quite as bad as it sounds, though, because when the analysis was limited to the most frequent and bothersome side effects, patients reported only two to three times more side effects than their clinicians (Zimmerman M et al, J Clin Psychiatry 2010;71(4):484–490).
How common and bothersome are antidepressant side effects for our patients? It’s not the easiest question to answer. Zimmerman and colleagues used the self report Toronto Side Effects Scale to ascertain side effects. The survey begins with the question: “Within the last two weeks, have you had any of the symptoms listed below,” and then lists 32 potential symptoms. Some patients presumably checked off symptoms, such as “agitation” or “decreased sleep,” that were not side effects but rather symptoms of depression. On the other hand, other patients might have underreported true side effects, because the scale is quite long, at times uses medical jargon, and can be confusing to complete. These potential problems were noted by the authors in their discussion.
In 2004, another study of side effects was published, one that might be more clinically relevant in terms of providing more valid estimates of side effect prevalence from SSRIs (Hu XM et al, J Clin Psychiatry 2004;65(7):959–965). In this study, 401 patients who had received an SSRI for depression were telephoned. Rather than relying on a long paper and pencil questionnaire, the interviewers verbally discussed with patients a list of 17 potential side effects. Having a conversation allowed researchers to judge whether the symptoms were likely to be actual side effects vs aspects of underlying psychiatric or medical conditions.
In this study, the top five most commonly reported side effects were:
Sexual dysfunction, 34%
Dry mouth, 34%
When researchers asked patients to name the side effects they considered to be most bothersome, the top five looked a bit different:
Sexual dysfunction, 16.7%
Weight gain, 11.5%
In another study of the same patients, these researchers reported on a clever extension of their methodology. In an effort to give us some guidance on what to tell our patients as we hand them a prescription, they asked patients what they recalled their doctors having told them when treatment began. Those who recalled being told to stay on their medication for at least six months were three times more likely to continue their meds than those told to take them for less than six months. Also, discussing possible side effects before starting medication was associated with patients staying the course longer—which is reassuring for those of us who worry that we will scare patients away by talking about side effects (Bull SA et al, JAMA 2002;288(11):1403–1409).
While there are several side effect scales out there, they tend to be oriented more toward research and are therefore more thorough and longer than we need in the office. Therefore, I’ve created a simple antidepressant side effect scale, which is available on our website at www.thecarlatreport.com/issue/vol9-no5. You can print it out and have patients complete it in the waiting room, you can complete it with them during the visit, or you can simply keep it on your desk to remind you of the most common side effects to ask about. Obviously this scale has not been validated as a reliable research tool, but it may encourage you to keep better track of your patients’ side effects.
So, how should we manage the common antidepressant side effects? Obviously, the first technique on everyone’s list is to simply stop the offending agent. I’ll say that up front because I don’t want to have to keep repeating it in every section below. The problem is that sometimes the patient is doing so well on the medication that neither of you want to discontinue it. Alternatively, some medications cause such terrible discontinuation side effects (did someone say Effexor?) that stopping the med may cause more “side effects” than continuing it.
For those who don’t mind the possibility of chasing their tail here and there (ie, using more meds to treat the side effects of meds), here are some possible remedies for common adverse events.
Drowsiness can be a symptom of the depression or a symptom of the cure. If the patient is sleeping well but is still tired all day long, that is a clue to AD-induced fatigue. First, make sure your patient is taking the medication at night instead of in the morning. If that doesn’t work, try a generic psychostimulant such as methylphenidate or an amphetamine preparation. Modafinil (Provigil) at 200 mg/day to 400 mg/day is another option, and has been shown to be effective for treating residual fatigue in partially treated depressed patients in placebo-controlled trials (Fava M et al, Ann Clin Psychiatry 2007;19(3):153–159). You can also choose armodafinil (Nuvigil), which lasts a bit longer than modafinil but may cost more because it is not available as a generic.
2. Sexual Dysfunction
The first rule for managing AD-associated sexual dysfunction (ADASD) is to make sure that you’re not falsely blaming the drug. Sexual dysfunction is surprisingly common in the general population: In one large study in the U.S., 43% of women reported sexual dysfunction (22% low libido, 14% sexual arousal problems, and 7% sexual pain), vs 31% of men (21% premature ejaculation, 5% erectile dysfunction, 5% low libido) (Laumann EO et al, JAMA 1999;281(6):537–544). Other studies have found that untreated depression roughly doubles the rate of sexual dysfunction, and antidepressant treatment increases the prevalence further. (For a review, see Baldwin DS, Br Med Bull 2001;57(1):81–99). Therefore, at least half of your depressed patients already have sexual problems before they walk in your door.
Anecdotally, psychiatrists tend not to get very detailed sexual histories during an initial evaluation, but we probably should, particularly if we are about to prescribe a medication that can affect sexual functioning. Having this baseline data will help you and your patient sort out whether or not the medication is responsible for future sexual issues. In general, it is thought that depression per se is more likely to cause low libido and erectile dysfunction than delayed ejaculation or anorgasmia. On the other hand, serotonergic antidepressants can cause all four of these problems (Segraves RT, Urol Clin N Am 2007;34(4):574–579).
Aside from discontinuing the AD or a weekend drug holiday, here are some reasonably tried and true strategies.
Switch to a different antidepressant. This generally means switching from a serotonergic antidepressant to bupropion (Wellbutrin), mirtazapine (Remeron), or nefazodone (Serzone). Occasionally you’ll see an article implying that duloxetine (Cymbalta) is less likely to cause sexual side effects than SSRIs. But this is based on a comparison with paroxetine (Paxil), the SSRI that is generally acknowledged to be the most side effect prone.
If you’re interested in this issue, take a look at the paper combining the results of four placebo- and paroxetine-controlled studies of duloxetine for depression. These included a total of 1466 patients, all of whom took the ASEX scale of sexual functioning. A majority of the patients (59.3%) entered the study with ASEX-defined sexual dysfunction. The authors properly focused on the 475 patients who did not have sexual dysfunction at baseline, and found that after eight weeks, rates of sexual dysfunction were as follows: paroxetine, 61.4%; duloxetine, 46.4%; and placebo, 28.8%. This tells us that paroxetine is more sexually deadening than duloxetine, but duloxetine doesn’t exactly do any favors to anyone’s love life. Furthermore, in those patients who continued in 34 week long extension trials, there were no statistically significant differences among any of the treatments: (paroxetine, 45.5%; duloxetine, 39.4%; and placebo, 35.3%) (Delgado PL et al, J Clin Psychiatry 2005;66(6):686–692).
Apologies for being so long-winded with this aside on duloxetine, but I think it’s an interesting study, and says something about Lilly as a company. On the one hand, they deserve kudos for going the extra mile and doing large scale studies comparing duloxetine with an active antidepressant, rather than simply comparing it with placebo. On the other hand, they chose the worst tolerated SSRI as their comparison, knowing that the studies would eventually provide some fodder for claims of superiority. But then, on the third hand, they reported their data fairly and completely. Bottom line: duloxetine causes about the same rates of sexual dysfunction as all of its competitors, except paroxetine.
Add an antidote. You could try any of the following medications as an antidote to the sexual dysfunction caused by an antidepressant.
PDE-5 Inhibitors. PDE-5 inhibitors (Viagra, Levitra, and Cialis) are robustly effective for males with AD induced erectile dysfunction. In the largest study, 142 men with AD associated erectile dysfunction were randomly assigned to sildenafil (Viagra), 50 mg to 100 mg vs placebo; sildenafil significantly improved erectile dysfunction and overall sexual satisfaction (Fava M et al, J Clin Psychiatry 2006;67(2):240–246). Sildenafil appears to help ADASD in women as well as men, though this is based on a smaller study, in which 98 women with ADASD were randomly assigned to sildenafil (mean dose, 90 mg) vs placebo. After eight weeks, those in the treatment group reported more improvement in orgasm and sexual satisfaction than the placebo group (Nurnberg GH et al, JAMA 2008;300(4):395–404).
Bupropion. While bupropion clearly causes less sexual dysfunction than SSRIs when used as monotherapy, bupropion as an antidote seems to have marginal effectiveness at best. The first controlled study randomized 30 patients to bupropion SR 150 mg/day or placebo for three weeks, and could find no benefit of adjunctive bupropion (Masand P et al, Am J Psychiatry 2001;158(5):805–807). Three years later, a different group of researchers tried again but with a higher dose, randomly assigning 42 patients with ADASD to either adjunctive bupropion SR 300 mg/day or placebo. After four weeks, bupropion produced generally better scores on the sexual functioning scale used, but the difference was statistically significant on only one of four subscales (Clayton A et al, J Clin Psychiatry 2004;65(1):62–67). Bottom line: You can try bupropion SR as an antidote, but use the 300 mg dose and cross your fingers.
Buspirone. In most review papers you read, buspirone (BuSpar) is listed as one of the few “evidence-based” antidotes to ADASD. But when you actually dive into this “evidence,” you find some case reports and a single controlled trial. And the controlled trial showed alarmingly unimpressive results. Basically, this was an afterthought of a different study, one that was originally designed to test buspirone as an augmenting agent for patients who failed an SSRI. Buspirone did not separate from placebo for depressive symptoms, but the authors tried to salvage what they could by doing a retrospective analysis to see if buspirone was helpful for sexual side effects. This analysis included only the 47 patients (out of the original 119) who had ADASD when they entered the trial. Of the 47 patients, 27 were randomized to buspirone and 20 to placebo; the average buspirone dose was 47 mg/day. For the 21 men, buspirone did not significantly improve sexual functioning, but for the 26 women there was a statistically significant improvement (Landen M et al, J Clin Psychopharmacol 1999;19(3):268–271).
3. Weight Gain
A recent paper in the new open access Journal of Obesity does a good job reviewing the literature on psychotropic drugs, obesity, and potential treatments (Nihalani N et al, J Obes 2011; Vol 2011, available free online at http://bit.ly/ hibZdr). Among the more modern antidepressants, weight gain is most likely to be a problem with paroxetine or mirtazapine, but we’ve all seen patients gain weight on all other antidepressants, including the SSRIs and SNRIs. In the current era of atypicals being approved for either adjunctive depression treatment (quetiapine (Seroquel XR) and aripiprazole (Abilify)), treatment resistant depression (combined olanzapine and fluoxetine (Symbyax)), and bipolar depression (quetiapine XR), more patients with depression are being exposed to frank obesity generators. (Aripiprazole is fairly weight neutral, but has still caused a pound or two of weight gain in controlled trials).
“Lifestyle interventions” are always the first line treatment—including dietary counseling and admonitions to exercise more. This usually helps to a degree, but often enough, the discussion will turn to medication. Here are some things to try.
Switch to bupropion, or add bupropion augmentation. Bupropion is the only antidepressant that actually causes weight loss.
Add a psychostimulant. While stimulants such as methylphenidate and amphetamine are not FDA approved for weight, they are clearly effective—with the disadvantages of the risk of abuse, and side effects such as insomnia or anxiety.
Try an FDA approved weight loss pill.
- Sibutramine (Meridia), a mixed serotonin and noradrenergic reuptake inhibitor that acts by increasing the sensation of satiety, has been removed from the market in the U.S. because of cardiovascular risks. It was not a great choice for some of our patients anyway, as there was a risk of serotonin syndrome when it was combined with a serotonergic agent.
- Orlistat (Xenical) (120 mg three times daily with meals) blocks lipases and therefore decreases fat absorption through the GI tract. Controlled studies have shown an average of 10% body weight loss (Hollander PA et al, Diabetes Care 1998;21(8):1288–1294). The side effects of diarrhea and fatty stools make it unappealing for many, but on the other hand it can be safely combined with any antidepressant.
Try an off-label treatment
- Bupropion + naltrexone, which was packaged into a single pill called Contrave, has been shown to be effective for weight loss in large controlled trials (Greenway FL et al, Lancet 376(9741):595–605). The drug was ultimately rejected because it increased pulse and blood pressure slightly, potentially raising the risk of heart disease. Many believed the FDA was too cautious, considering that both of these medications are already in widespread use for other indications. If you want to prescribe a Contrave-like formulation anyway, studies found that the dose with the best benefit/risk ratio was naltrexone 16 mg + bupropion 200 mg twice daily. Since naltrexone comes in only a 50 mg pill, your patients will have to try breaking it into thirds to get close to 16 mg; bupropion is readily available in 200 mg strength.
- Metformin (Glucophage), a medication usually used first-line in diabetes, has consistently produced modest weight loss in obese patients taking atypical antipsychotics. (For a meta-analysis, see Praharaj SK, et al, Br J Clin Pharmaco 2011;71(3):377– 382). A commonly used dose is 750 mg/day. Side effects include uncommon diarrhea and rare lactic acidosis.
- Topiramate (Topamax) dosed at 100 mg/day to 150 mg/day works on food cravings, but beware of its cognitive side effects, which have earned it the nickname “Dopamax” among savvy psychopharmacologists.
4. Dry Mouth
This turns out to be more than a minor nuisance, since it can lead to dental problems due to a decrease in the usual antibacterial activity of saliva. There are over the counter liquid saliva products (eg, “Saliva Substitute” and “Salivant”) which are twice daily mouth rinses that can relieve the sensation of dryness temporarily. Biotene makes an OTC toothpaste and a mouthwash targeting dry mouth. Make sure your patients know about the higher risk of dental caries and encourage frequent dental checkups.
5. Other Side Effects
We didn’t have room in this print issue to cover some of the other bothersome side effects, such as apathy syndrome, excessive sweating, and constipation, but you can visit our website for detailed descriptions and recommendations on these issues. Please see www.thecarlatreport.com/issue/vol9-no5 for these and to see the TCPR side effects scale.