The most commonly prescribed class of antidepressants today are the selective serotonin reuptake inhibitors (SSRIs), the so-called "Prozac family" of medications, of which Prozac was the prototype, but which also includes Zoloft, Paxil, Luvox, Celexa, and Lexapro. They are widely prescribed because they are very safe and generally well-tolerated, and they are highly effective at improving a person's mood and reducing symptoms of tearfulness, hopelessness, irritability and also anxiety, including panic attacks, obsessions, and compulsions.
Because they all share the same mechanism of action--increasing the bioavailability of the neurotransmitter serotonin--they are all remarkably similar with respect to how they help, as well as in regard to their side effect profiles. This is because, by and large, it is the increase in serotonin itself that is responsible for both the beneficial effects on the brain as well as the bothersome physical effects that occur elsewhere in the body, such as the GI tract. Various types of serotonin receptors are widespread throughout the body and wherever serotonin acts upon one of these receptors, you get an effect, sometimes a desired effect, but also more often than not, a side effect. In this respect the SSRIs are remarkably similar; differences are usually a matter of degree. Where these medications tend to differ more so is in relation to their effect at other, non-serotonergic, receptor sites, such as acetylcholine or histamine receptors, which is what accounts for the differences in their side effect profiles and hence, their tolerability for a given patient.
Practitioners are fond of saying that the choice of SSRI is an empiric one: starting an SSRI for the first time, making dosage adjustments, resuming therapy and/or switching to a new and different medication in this class is essentially a trial-and-error process, they explain, but that doesn't mean that it is completely arbitrary or haphazard (although some prescribers tend to want to approach it that way). Educated guesses can be made before a medication trial is undertaken, if not regarding how useful a medication's effect on an individual's mood will be, then at least with regard to how well or how poorly it will be tolerated. This mindfulness on the part of the prescriber, as well as the person agreeing to subject himself to a trial of this type of medication, can avoid wasted time and an unpleasant experience.
So, how best to choose among this family of medications? Well of course your doctor will be the one making the final recommendation, but with regard to treatment with an SSRI there is never only one option for a patient (unless she's tried them all). This article explains what considerations should go into that decision-making process; there is no reason why, armed with a little information, you cannot be an active participant in that process. It is you who will be taking the medication, after all. Why not help your doctor save you some trouble and/or wasted effort, if you can?
Assuming this is the best class of medication from which to choose, there are two main considerations: 1) selecting a medication that is likely to work and 2) selecting the medication that will best be tolerated by the patient. Because the SSRIs are more-or-less equally likely to work, the ultimate decision is more often based on the second consideration.
Selecting an SSRI that is likely to work. Assuming that the diagnosis is correct and that medication management is indicated in the first place and that the target symptoms in question indicate that an SSRI would be the best initial option (agitated, irritable depressions and/or anxiety states), I have already stated that there is little to predict beforehand whether one or another SSRI will ultimately have the desired effect; unless there is prior history to indicate one way or another, they all start out with the same likelihood of helping. Although in general we can expect that certain symptoms will be significantly reduced on these medications as a person's threshold for tearfulness or irritability or feeling overwhelmed is raised with increasing serotonin levels, in terms of overall efficacy--in terms of remission--it is never possible to predict with 100% accuracy whether a given compound will "work" on a person's depression. Likewise, while panic attacks can always be stopped with these medications if taken properly, residual, sometimes unacceptably high levels of generalized anxiety may fail to completely resolve despite treatment with one or another SSRI, necessitating a switch of medication.
The lesson here is that, while some people respond better to one SSRI over another, to a large extent this is unpredictable. Differences in efficacy are idiosyncratic; that is to say, they depend largely upon the individual who is taking them, and they do so for unknown biological reasons. And so we try one medication to see if it works, and we ensure that a good therapeutic trial is completed before trying something else, if necessary. However, what I have found in my practice is that switching from one SSRI to another is not usually because of a lack of effectiveness, but instead due to tolerability issues. By anticipating these tolerability issues (which is more methodical than anticipating efficacy), time and frustration can be saved.
Selecting the SSRI that will best be tolerated by the patient. While the reasons for differences among individuals in terms of the efficacy derived from medications in this class is a mystery, differences in tolerability can often be explained, and even predicted, by the relative affinity of one SSRI versus another for certain receptor types. If you reviewed the complete prescribing information for any of the following medications, you would be left with the impression that all of them can cause all of the same potential side effects, which is true, but the clinical reality is that some of them are much, much more likely to cause certain side effects than the others. All you need to know as a patient is that one SSRI tends to cause more diarrhea or constipation or dry mouth than another, or that a given SSRI tends to cause more weight gain or more sexual dysfunction than the others, etc.
Likewise, if you read the pages and pages devoted to each medication in the Physician's Desk Reference (PDR) , you would come to realize that nearly every side effect known to man has been reported at one time or another or associated in some way or another with every known medication, rendering an encyclopedic book like the PDR practically useless to the individual who just wants to know which side effects are likely, not which ones are possible. Of the 1001 potential side effects listed for any given medication, there are only a handful of truly common problems.
When I first started routinely prescribing these medications in the mid-nineties there were only four available SSRIs in the United States, but even then I had no idea how Prozac would be different for a patient than Zoloft, or Paxil versus Luvox. Now, after having prescribed these medications for more than a decade, although they all share the same general side effect profile, it has become clear to me which SSRIs are more likely to induce which unacceptable side effects, and to what degree. (Remember that, often, it is simply a matter of degree to which the medications differ.) For example, while they all induce sexual dysfunction, some are simply worse offenders in this regard. Similarly, while they all can cause weight gain, some unequivocally have a greater propensity to do so. What follow are some general profiles, what you might expect from treatment with one SSRI versus another.
All of the SSRIs tend to cause diarrhea, headaches, queasiness, restless legs/body, sexual dysfunction (delayed orgasm, decreased libido), sweating/hot flashes, lethargy, yawning (not related to feeling tired), waking up in the middle of the night, jaw clenching, and weight gain. Sounds terrible, but any list of potential side effects does. The fact is that many of these side effects, if they present themselves at all, are mild and resolve over time. Given that the above-listed are all common SSRI side effects, the side effects mentioned below are not being highlighted because they are in any way exclusive to the medication in question. Rather, the medication in question is simply especially prone to inducing those particular side effects. Given that they are all equally likely to help, carefully consider which unwanted effects you would most like to avoid.
The following blurbs are not at all meant to be comprehensive. They are not even highlights; they are highlights of the differences of each medication with respect to its sister medications. The last thing I'll say here is that there are always exceptions, of course, and you may be that one rare person who actually gets more diarrhea on Paxil than on Zoloft, but otherwise these guidelines will prove true, time and again.
Prozac (fluoxetine). Still the most famous, in part because it was the first, Prozac truly revolutionized the pharmacologic treatment of depressive and anxious mood disorders when it was released in 1987. It was far safer than the first-generation antidepressants available up to then, particularly in overdose, and induced far fewer side effects.
One of the things that distinguishes Prozac from the other SSRIs that followed is that it has an exceptionally long half-life. A drug's half-life is a measure of how long it takes the body to metabolize it away. The longer the half-life, the longer the drug remains in circulation in its original form. Once drug levels have reached steady-state, it takes five half-lives to clear 97% of the drug. The half-life of most SSRIs is approximately one day, so stopping them means they will be cleared by the body in about one week. Fluoxetine, on the other hand, has a half-life of 4-6 days with chronic dosing, and its active metabolite norfluoxetine has a half-life of 1-2 weeks. Compared with most other drugs, this is extraordinarily long. The practical implications are that, because a drug's half-life also affects how quickly it builds up to steady-state levels in the body with regular dosing, 1) it takes longer for Prozac to reach therapeutic levels, possibly delaying the onset of action even more so than what is typical for these types of medications (typically around two weeks) and even more significantly, 2) when you stop taking Prozac, it can take 5 or more weeks to fully exit your system, although the clinical effects of the medicine will begin to wane long before that. One advantage to this is that Prozac withdrawal is not seen; you can abruptly stop taking the medication and it tapers itself out of your system (abruptly stopping any psychotropic medication is generally frowned upon, however). One disadvantage would be a delay in switching to a different medication or starting a new medication that cannot be co-administered with Prozac, such as the MAOIs (monoamine oxidase inhibitors).
Prozac is also known to be among the more "activating" of SSRIs. It tends to cause jitteriness and motor restlessness (akathesia) more so than some of the others. Prozac may not be the best choice if you are already troubled with restless legs syndrome or the general inability to sit still for very long periods of time due to motor restlessness.
Prozac is also famous for what is affectionately called the Prozac "poop-out." It is not at all uncommon for any SSRI, after a time, to not work as vigorously as it did initially, or even to stop working altogether. Usually a simple dose adjustment is all that is needed to jump start the treatment, but sometimes a medication switch becomes necessary. I have certainly seen many cases of "Prozac poop-out," although it is difficult to say based on anecdotal evidence alone whether Prozac is truly more prone to this disappointment than the other SSRIs. Why a medication stops working or stops working as well over time is a complicated issue that may have a lot to do with extra-pharmaceutical factors as well.
Fluoxetine (the generic preparation) is also marketed in the
Prozac Weekly. When pharmaceutical companies lose their patent on brand-name medications and other companies are allowed to sell generic preparations of the drug, the original company will often rebrand the medication in order to hold on to some of the market share for their product. This was the case with Sarafem. A more aggressive strategy involves obtaining a new patent for a medication by either slightly tweaking the molecule (so that legally it is now a "new" compound) or else revamping the delivery system by pairing it with a different vehicle or placing it in a new type of capsule. Although the pharmaceutical companies spend many hundreds of millions of dollars developing and marketing these novel preparations, many times it amounts to nothing more than a marketing ploy. The most common scheme is to take a drug and place it in an extended-release capsule.
Prozac Weekly is just that: 90mg of good old fluoxetine in a fancy, enteric-coated capsule that dissolves more slowly and thus releases the medication more slowly, but that's not what allows for once-weekly dosing. It is fluoxetine's exceptionally long half-life that facilitates once-weekly dosing, and there's nothing new about the half-life. As a psychiatrist, I never recommend once-weekly dosing of any anti-depressant; the only reason to dose once a week instead of every day is apparently a matter of convenience, although this is arguable, since it may actually be more challenging to remember to take your medication if you only take it once a week, as opposed to every day. When Prozac Weekly was first released, the manufacturer touted the fact that once-weekly dosing would allow individuals to feel "less daily dependence on the need for medication" that many patients find stigmatizing, but this does not change the fact that individuals who are clinically depressed do, in fact, require daily levels of the medication that the long half-life allows, even when it is not being taken every day. Marketing ploys aside, my own preference is to be completely upfront with my patients and to educate them regarding the exact nature of their need for treatment, whatever that may be, and to encourage destigmatization in other, healthier and more honest ways. The last thing I would ever want to do as a psychiatrist is to downplay the physical nature of a patient's major mood disorder and lull that person into a false sense of security by promoting the idea that his illness is not so serious because he "only needs to take medication once a week." Too many patients who truly need medication to function and remain safe already underestimate their need for treatment and eventually disregard their regimens, often with disastrous results. But, if I ever were to recommend once-weekly dosing, a patient could just as easily take generic fluoxetine once a week and save a lot of money.
Zoloft (sertraline). Zoloft was the next SSRI to be developed, and it was just as effective across the board and just as well tolerated as its predecessor, with subtle differences that varied from individual to individual. In my experience, Zoloft is not as overtly activating as Prozac, although akathesia is common enough. Zoloft is also one of the SSRIs that is especially prone to produce diarrhea. And again, while the other SSRIs also frequently cause loose stools--at least initially--the diarrhea that Zoloft brings on is often more pronounced, and often endures for a longer period of time, than the others.
Unlike Prozac, which is often started at the target dose of 20mg, Zoloft is started at 25mg, a subtherapeutic dose for the vast majority of people, thus it requires more of a titration to reach the recommended dose, which could eventually be as high as 200mg for some. With each adjustment, side effects like diarrhea can reappear even after they had previously resolved (although this effect becomes attenuated over time as well; in general, the more dosage is tolerated, the less increasing that dosage further will renew old side effects).
Plenty of patients have noticed weight gain on Zoloft, but in my experience it is not the worst in this regard (see Lexapro, and especially Paxil).
Paxil (paroxetine). Paxil is the most potent inhibitor of serotonin reuptake in vitro. Clinically, I have found it to be the most aggravating in terms of sexual dysfunction. Many patients report complete anorgasmia on even starting doses (10mg or less). This makes low-dose Paxil an excellent treatment for premature ejaculation, but something of a problem in treating depression and anxiety, particularly at the higher doses recommended for those conditions.
Paxil is unusual among the SSRIs in that it is highly anticholinergic; that is, it blocks one subtype of receptor that is normally bound by the neurotransmitter acetylcholine. The result is side effects such as dry mouth, constipation, and blurred vision for close-up objects (e.g., text on a page). Anticholinergic effects can also include a subtle impairment in short-term memory. For this reason, Paxil is usually a poor choice for an elderly person, who may already be experiencing significant problems with constipation, presbyopia and noticeable short-term memory loss.
And while all of the SSRIs can potentially lead to weight gain, Paxil is one that I consider notorious for doing so. I avoid it in patients for whom this is a particular concern.
Unlike Prozac, Paxil is also one of the SSRIs that is associated with a peculiar discontinuation syndrome. I call it a "discontinuation" syndrome, as opposed to "withdrawal" purposely, to avoid the notion that Paxil is in some way addicting. In the classic sense it is not. However, because of its particular neurochemistry, abruptly stopping this medication or abruptly lowering the dosage--or even for some people taking it a few hours later than scheduled--can lead to dizziness, flu-like symptoms and paresthesias that feel like tiny, lightening-fast "electric shocks" in the brain. These paresthetic jolts do not represent seizure activity and they are more uncomfortable than painful; they are not harmful and they eventually subside, but they are very unpleasant.
SSRI withdrawal paresthesias (also commonly experienced upon the abrupt discontinuation of Luvox and Effexor, which has an SSRI component) are exacerbated by rapidly turning or moving the head, which can bring one on, and by stimulants. Conversely, they are reduced in intensity or blocked entirely by mild tranquilizers and do not wake people up from sleep. The way to avoid them is to avoid abruptly stopping the medication, which is always ill-advised. The recommendation is always to lower the dose gradually over several days under a doctor's supervision before stopping the medication entirely. If you are going out of town you want to take extra precautions not to leave your Paxil behind, and you want to avoid getting caught between refills for the same reason.
With all of its extra tolerability issues, Paxil is often clinically superior to the other SSRIs in certain conditions that require a truly potent serotonergic medication. I have seen it outperform the other SSRIs, even at maximal doses, in generalized anxiety disorder and obsessive-compulsive disorder in particular. Paxil definitely has its place.
Luvox (fluvoxamine). Luvox is not as well-recognized among laypersons, due in large part to the fact that it was originally approved and marketed for the treatment of obsessive-compulsive disorder, which is not as common as other anxiety disorders and major depression. But as repeatedly stated above, all of these medications are useful for all of the same types of depressive and anxious mood disorders.
Like Paxil, Luvox is associated with a pronounced "withdrawal" syndrome if not gradually tapered before discontinuation. Like Zoloft, Luvox typically has to be titrated from a starting dose to a significantly larger final dose and is more likely to cause GI disturbances than the others, either loose stools or alternately, constipation. Weight gain is another not uncommon issue with Luvox.
Celexa/Lexapro (citalopram/escitalopram). Whereas Paxil is the most potent selective serotonin reuptake inhibitor, Celexa and Lexapro are the most selective, referring to their relative effect on serotonin receptors versus other receptor types; i.e., among SSRIs, their action is the most specific to serotonin receptors. Lexapro (escitalopram) is the left-handed stereoisomer of Celexa (citalopram). Escitalopram is the biologically active molecule; the right-handed enantiomer is not only inert with regard to blocking the reuptake of serotonin, but actually competes with the left-handed enantiomer, so that by removing it from the mix, both potency and efficacy are increased. Lexapro can therefore be thought of as a "purified" form of Celexa that works better on a milligram-per-milligram basis because the active component is not competing with the inactive component for receptor sites.
If Paxil has the worst sexual dysfunction, Lexapro typically has the least. However, like Paxil, Lexapro is notorious in my book for causing significant weight gain. I have also noticed (and this of course is purely anecdotal) more "rebound" symptoms between doses in anxious patients taking Lexapro. I have had several cases in which panic attacks and symptoms of social anxiety seemed to do worse at various times of the day despite consistent dosing and even with divided dosing.
Effexor (venlafaxine). Effexor is actually a dual-acting agent: it was the first commercially available (non-tricyclic) serotonin-norepinephrine reuptake inhibitor, or SNRI. SNRIs can be advantageous over SSRIs based on their dual mechanism of action and indeed have been shown to be somewhat more efficacious with respect to depression. They may also have a more rapid onset of action, which I have certainly seen with Effexor in some cases.
I include Effexor in this discussion because at low starting doses, Effexor is essentially an SSRI. The affinity of venlafaxine for serotonin reuptake is many orders of magnitude greater than that for norepinephrine and noradrenergic reuptake does not occur until higher doses are reached. Effexor's side effect profile is very similar to the other SSRIs, with significant sexual dysfunction and a marked discontinuation syndrome. In my experience, I have seen a lot more temperature dysregulation (sweating and hot flashes) in patients taking Effexor, but no weight gain, which is rare with Effexor.
Other considerations that come into play in choosing one SSRI over another may be practical: cost, for example, or whether or not the medication is supplied in liquid form, for those patients (children, usually) who are unable to swallow a capsule. Most of the SSRIs are also available in extended release (XR) or controlled-release (CR) preparations, which, although their development may have been prompted by corporate financial interests, can nonetheless be very useful at reducing immediate side effects by prolonging absorption and simplifying a regimen to once-a-day that you might otherwise be taking in divided doses throughout the day.
Family History. One other point bears mentioning in any discussion regarding psychotropic medication selection. Importantly, if you have a blood relative--particularly a first-degree relative, such as a parent, sibling or child--who responded favorably to a particular medication, you have a statistically higher probability of responding favorably to that medication yourself, so inasmuch as selecting a psychotropic medication can be a trial-and-error process, be sure to mention that to your prescribing doctor if you know about it.
Tip: Ask for samples. If you are trying a medication for the first time, your doctor will often be able to provide you with free samples or a coupon for a free month's supply.