Textbook of Psychiatry/Psychopharmacology

The use of psychotropic medicines to treat psychiatric illness has increased dramatically in recent times. Although the biological etiologies of most psychiatric disorders are still unclear, effective pharmacological treatments have been developed over the past 50 years that have become part of the standard of care in the treatment of most major psychiatric disorders. Psychiatric medications are part of the armamentarium of most practicing physicians, regardless of medical specialty. In the United States, although most severe types of mental illness are likely to be treated by psychiatrists, most prescriptions for psychotropics (e.g., anxiolytics and newer antidepressants) are written by non-psychiatrists.(Stagnitti, 2008) Psychiatric medications are consistently prominent in the list of the top 200 most commonly prescribed medications, and in the top 20 pharmaceuticals in terms of sales in the United States. From 2003-2007 antidepressants, as a class, topped all other therapeutic classes for the overall number of dispensed prescriptions in the U.S.(IMS Health, 2007) As in the treatment of all medical disorders, a thorough evaluation must precede psychiatric diagnosis and subsequent psychopharmacological treatment. A complete history should be obtained and the patient should be examined. Medical or neurological etiologies that may contribute to the presentation of psychiatric illness should be identified and addressed. Nearly 10% of patients presenting with a psychiatric complaint will turn out to have a medical problem as the primary cause.(Hall, Popkin et al. 1978) Active substance abuse, if present, should be treated before or at the same time that pharmacological therapies are initiated. The clinician should then decide if the condition requires medication treatment. Mild to moderate anxiety and depression generally respond equally well to supportive interventions or psychotherapy.(APA, 2004; Barkham and Hardy, 2001; Cuijpers, van Straten et al. 2009; King, Sibbald et al. 2000) On the other hand, if the psychiatric disorder or symptoms are severe, or if psychosis, mania, or dangerousness are present, then psychopharmacological treatments (and referral to a psychiatrist) are indicated. Although many primary care physicians may be quite comfortable with their ability to manage psychiatric illness, the amount of monitoring that is required to provide adequate follow-up should be taken into account before initiating treatment. When treating moderate to severe psychiatric illness, optimum therapy includes the use of concomitant psychotherapy in addition to pharmacotherapeutic measures.(APA, 2004; APA, 1998; APA, 2000; Keller, McCullough et al. 2000; Banerjee, Shamash et al. 1996; Reynolds, Frank et al. 1999; Katon, Von Korff et al. 1999; Miklowitz, 2008) Placebo-controlled randomized clinical trials, using strict exclusionary criteria when selecting subjects, have traditionally been used to study a psychiatric medication’s efficacy (i.e., the ability of the medication to treat the condition better than placebo under controlled conditions). For example, studies comparing an antidepressant to placebo may use an 8 week double-blind parallel design and include subjects with major depression but without any other medical or psychiatric co-morbidities. Response may be defined as a 50% improvement in a chosen outcome rating scale. These efficacy studies also provide the response data that pharmaceutical companies must submit to the Food and Drug Administration (FDA) to obtain indications for developed drugs. Effectiveness studies, on the other hand, are often larger, naturalistic studies that attempt to approximate "real world" conditions by studying patients who may have psychiatric and medical co-morbidities, and by relying on broader outcome measures for assessing response. These studies may compare outcomes of treatment with multiple medications. As such, effectiveness studies complement our understanding of drug efficacy.(Summerfelt and Meltzer, 1998) Recent National Institute of Mental Health (NIMH) sponsored effectiveness studies have the added benefit of funding from a neutral (non-pharmaceutical industry) source, thereby avoiding possible study design shortcomings or evaluator biases that may influence study results.(Heres, Davis et al. 2006; Osser, 2008)  These studies include (1) the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE),(Keefe, Bilder et al. 2007; Lieberman, Stroup et al. 2005) (2) the Sequenced Treatment Alternatives to Relieve Depression Study (STAR*D),(Rush, Trivedi et al. 2006; McGrath, Stewart et al. 2006; Nierenberg, Fava et al. 2006; Trivedi, Fava et al. 2006; Fava, Rush et al. 2006) (3) the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD),(Sachs, Nierenberg et al. 2007; Goldberg, Perlis et al. 2007; Miklowitz, Otto et al. 2007) (4) the Clinical Antipsychotic Trials of Intervention Effectiveness—Alzheimer’s Disease (CATIE-AD),(Schneider, Tariot et al. 2006; Sultzer, Davis et al. 2008) and (5) the National Institute of Alcohol Abuse and Alcoholism (NIAAA) sponsored Combined Pharmacotherapies and Behavioral Interventions Study (COMBINE).(Anton, O'Malley et al. 2006; Anton, Oroszi et al. 2008)  Findings from these studies are now influencing clinical psychiatric practice. In clinical practice, even after an appropriate diagnosis is made for an individual patient and the decision is made to use a medication from a particular pharmacotherapeutic class (e.g., an antidepressant for depression), multiple variables need to be considered prior to selecting a specific agent. The physician should take the following into account: (1) patient acuity and the need to address the most dangerous presenting symptoms (e.g., behavioral agitation, suicidality, catatonia, etc.) first, (2) the patient’s past treatment history, (3) pre-existing medical conditions in order to minimize any increase in medical risk, (4) possible medication interactions, (5) the time required for amelioration of symptoms, (6) a medication’s known side effect profile and how this may affect presenting symptoms, (7) the need to minimize the use of polytherapy, (8) possible pharmacogenetic factors and hereditary patterns of drug response and tolerance, and (9) financial cost-benefit considerations. The practicing physician should consider these issues prior to initiating treatment.(Ansari, Osser et al. 2009) Characteristics of the major classes of psychotropics and their use in adults are discussed below. Children and adolescents may tolerate or respond to these medications differently. The use of psychopharmacological therapies in these age groups is outside the scope of this chapter.

Antidepressants
Currently available antidepressants primarily affect the norepinephrine and serotonin (monoamine) neurotransmitter systems.(Nestler, Hyman et al. 2009) Norepinephrine systems originate primarily from the locus ceruleus (and lateral tegmental areas) and project widely to almost all areas of the brain and spinal cord. Serotonergic neurons reside in the raphe nuclei in the brainstem and diffusely make contact with all areas of the brain. Most antidepressants increase the available amount of norepinephrine and/or serotonin at the neuronal synapse by decreasing the reuptake of these neurotransmitters into the pre-synaptic cell. They do this by inhibiting the norepinephrine transporter and/or the serotonin transporter, or by decreasing the metabolism of these neurotransmitters. Other antidepressants have direct effects on monoamine receptors. Genetic polymorphisms of the norepinephrine and serotonin reuptake transporters (Kim, Lim et al. 2006) as well as polymorphisms of post-synaptic serotonin receptors(McMahon, Buervenich et al. 2006) have been associated with differences in responses to different antidepressants. Once synaptic changes have taken place with treatment, long-term adaptations in post-synaptic neurons and resultant changes in gene expression may then be responsible for alleviating depression.(Nestler, Hyman et al. 2009).

Tricyclic Antidepressants (TCAs)
Beginning with the introduction of imipramine in the late 1950’s,(Kuhn, 1958) tricyclic antidepressants were among the first classes of antidepressants developed. They share a tricyclic structure (two benzene rings on either side of a seven-member ring), exhibit variable degrees of norepinephrine and serotonin reuptake inhibition, and are antagonists at several other neurotransmitter receptors.(Hyman, Arana et al. 1995) Examples of commonly used TCAs include the tertiary amines imipramine, amitriptyline, clomipramine, and doxepin, and the secondary amines desipramine (metabolite of imipramine) and nortriptyline (metabolite of amitriptyline). All TCAs can cause the following adverse effects: (1) slowing of intra-cardiac conduction as measured by QRS and QTc prolongation, (2) anticholinergic effects such as dry mouth, urinary retention, and constipation due to muscarinic acetylcholine receptor antagonism, (3) orthostatic hypotension due to peripheral alpha-1-adrenergic antagonism, and (4) sedation and possible weight gain due to histamine (H1) receptor antagonism. For these reasons TCAs need to be started at low doses and increased gradually, giving the patient time to accommodate to these effects. Individual differences in both severity of side effects and therapeutic effects (along with differences in therapeutic serum levels)(Perry, Zeilmann et al. 1994) do exist among individual TCAs. There is some evidence to suggest that TCAs may have a particularly important role in the treatment of severe and psychotic depression.(Hamoda and Osser, 2008) The cardiac effects of TCAs have contributed to a reduction in their use over the past 20 years. Prolonged QT interval (measured as QTc when corrected for heart rate) may be associated with torsades de pointes, a potentially fatal ventricular arrhythmia (also see section on antipsychotics). All patients should have an ECG to rule out any existing conduction abnormalities prior to considering TCAs. Patients with recent myocardial infarctions should not initiate treatment with these antidepressants. Most importantly, depressed patients who are at risk for suicide and overdose may not be appropriate for treatment with TCAs. It should be noted that a 1-2 week supply of these medications can be fatal in overdose due to the risk of cardiac arrhythmias. Therefore, depending on the patient’s risk of suicide, clinicians may need to limit the number of tablets prescribed with each refill. This concern is significantly lessened with the use of newer antidepressants that are safer in overdose. TCAs are often used for their mild to moderate analgesic effects in the treatment of chronic pain syndromes.(Magni, 1991) These effects are independent of any effect on mood, with efficacy starting at lower doses, and with response seen earlier than when antidepressants are used for depression.(Magni, 1991; Max, Culnane et al. 1987; Onghena and Van Houdenhove, 1992)  The TCAs seem more effective for chronic pain than the selective serotonin reuptake inhibitor antidepressants (SSRIs, see below).(Ansari, 2000; Fishbain, 2000; Saarto and Wiffen, 2007)

Monoamine Oxidase Inhibitors (MAOIs)
Monoamine oxidase is an enzyme that acts to metabolize monoamines, both intracellularly and extracellularly. Its inhibition increases the amount of serotonin, norepinephrine and dopamine available for neurotransmission. The first MAOI, iproniazide, an anti-tuberculosis drug, was discovered in the 1950’s. Tranylcypromine, phenelzine, isocarboxazid, and more recently transdermal selegiline are MAOIs currently available in the United States for the treatment of depression. These antidepressants may be particularly effective for patients with atypical depression (i.e., depression characterized by hyperphagia and hypersomnia).(Quitkin, Stewart et al. 1993) Although serotonergic side effects—see SSRIs below—as well as orthostatic hypotension can occur with MAOIs, there are two other primary areas of concern that limit the use of these agents.(Lippman and Nash, 1990) First, dangerous interactions can occur with certain foods, such as aged cheeses and wines that contain biogenic amines such as tyramine. MAOIs can inhibit the metabolism of tyramine in the intestine, increasing its general circulation and ultimately leading to an increase in sympathetic outflow and an adrenergic ("hypertensive") crisis characterized by severe hypertension, headache, and increased risk of stroke and cerebral hemorrhage. Patients need to be advised regarding dietary restrictions before treatment. Also, to prevent hypertensive crisis, MAOIs cannot be combined with medications that have sympathomimetic properties such as some over the counter cold preparations, amphetamines, and epinephrine (which is often added to local anesthetics as a vasoconstrictor). Secondly, MAOIs if used concomitantly with serotonergic agents, such as SSRIs, may lead to "serotonin syndrome," a potentially fatal condition that is characterized by hyperreflexia, hyperthermia, and tachycardia, and may lead to delirium, seizures, coma and death.(Sternbach, 1991) A two week washout period is required when switching from SSRIs (or any other agents with serotonergic effects) to MAOIs, or vice versa. An exception is when the long half-life SSRI fluoxetine is being discontinued: a five week washout period is needed before starting an MAOI.(Boyer and Shannon, 2005) The treating physician should access appropriate online databases such as the drug-specific DRUG-REAX® System (www.micromedex.com/products/drugreax) or GeneMedRx (www.genemedrx.com) to rule out dangerous drug-drug interactions when considering the use of an MAOI.

Selective Serotonin Reuptake Inhibitors (SSRIs)
SSRIs are antidepressants with a more favorable side effect profile than TCAs and MAOIs and as such are used as first-line antidepressants. As their name implies, SSRIs inhibit the serotonin transporter from reuptaking serotonin at the neuronal synapse. Interestingly, polymorphisms at the promoter region of the serotonin transporter gene (SLC6A4) may influence response to SSRIs: the presence of the "short" form of the serotonin transporter gene may be associated with poor response to SSRIs, whereas the presence of the "long" allele may be associated with positive drug response (Malhotra, Murphy et al. 2004) and better tolerability.(Murphy, Hollander et al. 2004)  Recent data from the large NIMH STAR*D study, however, have failed to support the association between this polymorphism and drug response.(Kraft, Peters et al. 2007; Lekman, Paddock et al. 2008) Currently available SSRIs include fluoxetine, paroxetine, sertraline, fluvoxamine, citalopram and its S-enantiomer escitalopram. Possible mild early side effects (that can be minimized by starting the SSRI at a low dose and increasing the dose gradually) include gastro-intestinal upset, sweating, headaches, jitteriness or sedation. Continuation of these agents may be associated with reversible sexual side effects (i.e., delayed ejaculation, decreased libido, or erectile dysfunction) in 2-73% of treated patients (depending on how questions regarding sexual side effects are asked).(Montejo, Llorca et al. 2001) SSRIs differ in their propensities to inhibit hepatic cytochrome P450 enzymes (e.g., CYP1A2, CYP2C9, CYP2C19, CYP2D6, CYP3A4).(Ereshefsky, Jhee et al. 2005) Inhibition of hepatic enzymes may lead to decreased metabolism of substrate medications such as warfarin, metoprolol, tricyclic antidepressants, and antipsychotics. This may increase serum levels of these drugs and lead to increased risk of dangerous adverse effects such as bleeding, hypotension, cardiac arrhythmias, and parkinsonian effects, respectively. Among the SSRIs, citalopram and escitalopram are the least likely to inhibit the metabolism of other drugs and are therefore preferred in patients concomitantly treated with multiple other medications. Escitalopram is a more recently marketed, and still expensive, successor to generic citalopram, though it has no clinically significant efficacy advantage. The relatively benign side effect profiles of SSRIs and their ease of use have contributed to widespread use by clinicians who might not have been comfortable with using earlier antidepressants such as TCAs and MAOIs. In cases of atypical presentations of depression, or depression in the context of recent substance abuse, SSRIs are more readily used even before there is absolute clarity in diagnosis. Under these circumstances, many clinicians believe that the benefits of treatment may outweigh the risks. In the case of depressed patients with concomitant substance abuse, the possibility that some patients may not be able to maintain sobriety because of an underlying major depressive disorder has served as a rationale for beginning antidepressant therapy even when the patient is still actively using. Although this empirical "trial" of an SSRI (as an antidepressant with a relatively benign side effect profile), in situations where there is less than optimum diagnostic clarity, may be appropriate for some patients, the physician should be aware of at least two major areas of risk. First, all antidepressants can induce mania in the short-term, and overall mood instability in the long-term, in patients with a vulnerability to bipolar disorder. A clear family history should be obtained to investigate whether there is a genetic predisposition to bipolar disorder. Also, clinicians should be aware that younger depressed patients, who may go on later to exhibit manic symptoms, may be incorrectly diagnosed with unipolar depression when in fact they may have a bipolar diathesis. A pre-bipolar presentation of depression (O'Donovan, Garnham et al. 2008) should be suspected in patients with (1) a family history of bipolar depression, (2) a younger age of onset, (3) a family history of completed suicide, (4) past poor response to antidepressants, (5) a history of treatment-emergent agitation, irritability, or suicidality, and (6) a history of post-partum psychosis.(Chaudron and Pies 2003) Depressed patients with these characteristics may have bipolar rather than unipolar depression and therefore should not be reflexively started on an antidepressant.(Ghaemi, Ko et al. 2002; O'Donovan, Garnham et al. 2008; Phelps, 2008) Secondly, antidepressants have been associated with an increased risk of treatment-emergent suicidality—this occurs in about 4% of treated patients versus 2% on placebo—especially in children, adolescents and young adults as noted in the current package inserts of all antidepressants. It is still unclear if this risk is significant in adults over age 25. The reasons for this increase in suicidality are not clear, although increased agitation (e.g., akathisia) or activation as a side effect,(Harada, Sakamoto et al. 2008) or the possible emergence of "mixed" manic symptoms (mania combined with dysphoric mood) in depressed bipolar patients as noted above, may be responsible. Despite the concern that antidepressants may infrequently increase suicide risk, it should be noted that overall rates of suicide in the United States had actually decreased over a prior 15 year span probably due to the increasingly widespread use of SSRI antidepressants.(Grunebaum, Ellis et al. 2004) Nevertheless, the concern about treatment-emergent suicidality argues for the need for careful evaluation and diagnosis, increased discussion of risks and benefits of treatment with patients (and family when appropriate), and close monitoring of all patients beginning antidepressant therapy. Prescribing antidepressants when indicated, coupled with these steps, is more appropriate than withholding antidepressants in unipolar depressed patients who are more likely to benefit rather than come to harm from these treatments.(Bridge, Iyengar et al. 2007) Unfortunately, recent surveys have found that instead of the increase that was hoped for in the monitoring of patients undergoing antidepressant therapy,(Morrato, Libby et al. 2008) there has been an overall decrease in the use of antidepressants and a recent increase in the overall rates of suicide(Gibbons, Brown et al. 2007) since the "black box" warnings about treatment-emergent suicidality were issued.

Serotonin-Norepinephrine Reuptake Inhibitors (SNRIs)
The SNRIs, venlafaxine, desvenlafaxine and duloxetine, are dual action serotonergic and noradrenergic antidepressants that would be expected to have efficacy similar to TCAs though without anticholinergic, antihistaminic, hypotensive, or cardiac side effects. Venlafaxine is primarily serotonergic at lower doses and has a dual action only at higher doses.(Feighner 1999; Richelson 2003) Using venlafaxine at lower doses (i.e., less than 150 mg per day), therefore, should not be presumed to be any different from using an SSRI. At higher doses it can have a mild to moderate hypertensive effect,(Johnson, Whyte et al. 2006; Mbaya, Alam et al. 2007) although patients with effectively treated hypertension can tolerate venlafaxine without an increase in blood pressure(Feighner, 1995). Duloxetine, which exerts a dual action effect throughout its dose range (i.e., not only at higher doses as with venlafaxine),(Stahl and Grady, 2003) can also increase blood pressure, although the effect may be less pronounced and clinically insignificant.(Raskin, Goldstein et al. 2003; Wohlreich, Mallinckrodt et al. 2007) SNRIs, like TCAs, are more likely to induce mania in bipolar patients than SSRIs.(Leverich, Altshuler et al. 2006) Because low-dose TCAs have been shown to be modestly effective in the treatment of chronic pain syndromes, and SNRIs have a similar dual action, they have been proposed for the treatment of chronic pain symptoms as well. Duloxetine is currently the only antidepressant with an FDA indication for diabetic neuropathy and fibromyalgia. Although the more benign side effect profile of duloxetine may make it the preferred agent in a patient for whom the risks associated with a TCA are unacceptable, there is no evidence to suggest it would be more efficacious for the treatment of pain than the more cost-effective TCAs.

Antidepressants with Other Mechanisms
Bupropion is an antidepressant with a poorly understood mechanism of action. It is believed to exert its effect through dopamine reuptake inhibition although it is unclear why this mechanism alone should provide it with an antidepressant effect. Some data suggest that it may also exhibit norepinephrine reuptake inhibition.(Richelson, 2003; Rosenbaum, Arana et al. 2005) Bupropion has a different side effect profile than antidepressants that significantly affect the serotonergic systems. It is unlikely to cause sexual side effects or weight gain—two of the most common reasons for medication non-adherence in patients. However, bupropion can lower seizure threshold and is therefore contraindicated in patients who are seizure-prone (e.g., patients with a history of seizures or conditions that increase seizure risk, such as eating disorders, or active withdrawal from alcohol or benzodiazepines). The risk of seizure is dose dependent: this should be kept in mind when combining bupropion with CYP2D6 inhibitors such as paroxetine or fluoxetine that may increase bupropion serum levels. Among antidepressants, bupropion is least likely to cause mania in bipolar patients.(Leverich, Altshuler et al. 2006; Post, Altshuler et al. 2006) Mirtazapine increases both serotonin and norepinephrine at the neuronal synapse (and therefore like SNRIs has "dual actions") through mechanisms distinct from reuptake inhibition. It is an antagonist at alpha-2-adrenergic autoreceptors thereby increasing norepinephrine and serotonin release, and it blocks post-synaptic 5HT-2A, 5HT-2C, and 5HT-3 serotonin receptors.(Feighner, 1999) (Mianserin, an earlier analog of mirtazapine marketed in Europe, has a similar mechanism of action). Mirtazapine can improve appetite (likely through 5HT-3 and H1 antagonism) and sleep (through H1 antagonism). As expected, these immediate effects can be very beneficial in the treatment of the acutely depressed patient with poor oral intake and insomnia. Weight gain however can be a concern over the long run. Nefazodone is a post-synaptic 5HT2 antagonist with weak serotonin and norepinephrine reuptake inhibition.(DeVane, Grothe et al. 2002) Although nefazodone can improve sleep, and is neutral in regard to weight gain and less likely than SSRIs to cause sexual side effects, it is used much less often since it was found to produce rare (1 in 250,000 to 300,000 patient-years), but severe, hepatotoxicity.(Gelenberg, 2002)  Trazodone, a structurally similar antidepressant, is used primarily as a hypnotic (it proved to be too sedating for most patients at doses necessary for antidepressant effect). Trazodone can commonly cause orthostasis and should be used cautiously in the elderly. Priapism is a rare side effect that should be discussed with male patients before treatment.

Further Notes on the Clinical Use of Antidepressants
Clinical practice today emphasizes the use of newer ("second generation") antidepressants including bupropion, citalopram, duloxetine, escitalopram, fluoxetine, fluvoxamine, mirtazapine, nefazodone, paroxetine, sertraline, trazodone and venlafaxine. As discussed above, the older tricyclics and MAOIs are not first-line because of their greater toxicity and risk of harm from overdose. In a meta-analysis of 203 studies comparing the efficacy and side effects of these newer antidepressants, no substantial differences in effectiveness were found.(Gartlehner, Gaynes et al. 2008) The authors recommended that antidepressants be selected on the basis of differences in expected side effects and cost (i.e., – use generic products over brand items). Another review of 117 trials concluded that sertraline had the most favorable balance among benefits, side effects, and acquisition cost.(Cipriani, Furukawa et al. 2009) The STAR*D study (Sequenced Treatment Alternatives to Relieve Depression), sponsored by the NIMH was a study of medications for the treatment of major depression. It produced important insights into the optimum use of pharmacotherapy for this disorder. STAR*D started with almost 4,000 heterogeneous "real world" depressed patients, who were treated by "real world" clinicians such as primary care doctors. Patients agreed to have up to 4 medication trials with the goal of achieving remission from their depression. Each trial lasted up to 14 weeks. Patients started with citalopram for the first trial. If response was unsatisfactory, they could have a switch to one of three antidepressants, or an augmentation with one of two augmenting agents. For the third trial, there were other switches or augmentations available, and finally for those still depressed and still willing to undergo the fourth trial, there was the choice of an MAOI or a combination of venlafaxine and mirtazapine. The latter has been referred to informally as "rocket fuel" because of the four different neurotransmitter alterations that this combination is thought to induce.(McGrath, Stewart et al. 2006) Key findings from STAR*D include the following:
 * Citalopram did not work well if patients met the DSM-IV criteria for melancholic features.(McGrath, Khan et al. 2008)


 * The switches in the second trial (to another SSRI: sertraline, to bupropion, or to venlafaxine) had equal efficacy.


 * The augmentations in the second trial (buspirone--discussed in anxiolytic section below--or bupropion) worked equally well.


 * Nothing worked well in trials one or two if patients had significant anxiety symptoms along with their depression.(Fava, Rush et al. 2008) However, a recent study with adjunctive aripiprazole (an antipsychotic discussed below) added to an SSRI found good results in patients with depression mixed with anxiety, in a post-hoc analysis.(Trivedi, Thase et al. 2008)  This needs replication in a prospectively designed study.


 * In the third trial, switching to a tricyclic antidepressant worked fairly well. It might have worked better if clinicians had dosed it properly and used plasma levels to monitor adequacy of dosage.


 * Adding lithium (discussed in the mood stabilizer section below) did not work as well as adding triiodothyronine in the third trial, but lithium might have done better if clinicians had dosed it properly.

As a group, STAR*D subjects were not particularly interested in psychotherapeutic treatment for their depression. Psychotherapy was available as an option in the second treatment trial, but patients could elect to drop it from the option list and most did so.(Wisniewski, Fava et al. 2007) The modest remission rates seen in STAR*D may reflect that a major component of the improvement in depression seen in research and clinical settings comes from the non-specific, interpersonal supportive aspects of care including the therapeutic alliance. STAR*D patients might have been less susceptible to these benefits than other patients who are more invested in psychosocial treatments of their disorder. It is hoped that future studies will improve our ability to select the best treatments for each patient, psychopharmacological and psychotherapeutic, depending on their needs and preferences. Table 1 summarizes characteristics of commonly used antidepressants.(WHO, 2007; PDR, 2008; Hyman, Arana et al. 1995; Perry, Zeilmann et al. 1994; Rosenbaum, Arana et al. 2005; Stahl 2005; Taylor, Paton et al. 2007)
 * In trial 4, the MAOI did not do well compared to the venlafaxine/mirtazapine combination, but clinicians underdosed the MAOI. Unfortunately, for the few patients who improved from either treatment, early relapse was common.

'''TABLE 1. COMMONLY USED ANTIDEPRESSANTS'''


 * ® Generic and U.S. brand name(s).
 * Dosing should be adjusted downwards ("start low, go slow" strategy) for the elderly and/or the medically compromised.
 * Abbreviations: bid-(bis in die) twice a day; CYP-Cytochrome P450 enzyme; FDA-Food and Drug Administration; GAD-Generalized Anxiety Disorder; MAOI-Monoamine Oxidase Inhibitor; MAO-B-Monoamine Oxidase Inhibitor, B subtype; MDD-Major Depressive Disorder; MDE-Major Depressive Episode; mg-milligram; ng/mL-nanogram per milliliter; OCD-Obsessive Compulsive Disorder; PMDD-Pre-menstrual Dysphoric Disorder; po-(per os) orally; PTSD-Post-traumatic Stress Disorder; q-(quaque) every; qhs-(quaque hora somni) at bedtime; SNRI-Serotonin Norepinephrine Reuptake Inhibitor; SSRI-Selective Serotonin Reuptake Inhibitor; TCA-Tricyclic Antidepressant; WHO-World Health Organization.

Anxiolytics
The pharmacological treatment of anxiety symptoms is both simple and complicated. On the one hand, medications such as benzodiazepines and barbiturates can have a relatively immediate effect on distressing anxiety symptoms. On the other hand, the use of such medications carries the risk of cognitive impairment, physical dependence, as well as the risk of psychological dependence and inappropriate use in some patients. It is not clear that episodic anxiety that is associated with situational stressors should be treated with medications. Anxiety per se may be a normal response to distressing events and a signal that may enhance a person’s motivation to address these real-life events. As such, it may be better understood and addressed through psychotherapy rather than pharmacologically. Medical students and physicians should be aware of cultural (and managed care) pressures that push for "popping a pill" rather than addressing the underlying causes of the patient's anxiety and improving coping strategies. Anxiety disorders are characterized by persisting patterns of anxiety symptoms impairing functioning. Examples include panic disorder, social anxiety disorder, post-traumatic stress disorder, obsessive-compulsive disorder and generalized anxiety disorder. The first-line medication treatments for most anxiety disorders are SSRIs (or other antidepressants with serotonergic effects—see Table 1). A time period of several weeks may be necessary before clear response. During this time, anxiolytics with more immediate effects (e.g., benzodiazepines) may be used for early symptom control.

Benzodiazepines
Benzodiazepines were first developed in the 1960s and are now the most commonly used anxiolytics used in the United States. Diazepam, clonazepam, chlordiazepoxide, temazepam, oxazepam, lorazepam, and alprazolam are examples of benzodiazepines. Their mechanism of action is through their binding on ?-aminobutyric acid (GABA) receptors.(Nutt and Malizia 2001) GABA is the primary inhibitory neurotransmitter in the brain. Benzodiazepines bind to one type of GABA receptor (GABAA) thereby increasing the receptor’s affinity for GABA. Increased GABA effect then increases the frequency of chloride channel openings allowing this ion’s influx into the cell which in turn decreases normal cell firing. The benzodiazepine binding site is composed of multiple subunits; binding to the alpha-1 subunit may explain sedative effects of benzodiazepines whereas alpha-2 subunit binding may be needed for anxiolytics effects.(Nestler, Hyman et al. 2009) Benzodiazepines are associated with multiple adverse effects. They are sedating, can impair concentration, memory,(Buffett-Jerrott and Stewart, 2002) coordination, can lead to falls in the elderly(Wagner, Zhang et al. 2004) (especially at initiation of treatment and after dose increases), and can cause respiratory depression. The choice of which benzodiazepine to use is often based on the pharmacokinetics of each drug. Diazepam, chlordiazepoxide, and clonazepam have relatively long half-lives. Additionally, diazepam and chlordiazepoxide are significantly hepatically metabolized and have multiple active metabolites; the use of these medications in hepatically compromised patients is therefore problematic. In medically ill patients, and in the elderly, benzodiazepines with short half-lives, such as lorazepam and oxazepam, are preferred, especially when the risk of respiratory depression is a concern (e.g., patients with chronic obstructive pulmonary disease). Alprazolam has a shorter half-life than lorazepam and oxazepam. It is, however, also associated with significant rebound anxiety because of the rapid drop from peak serum level after each dose. Despite its current widespread use, alprazolam should generally be avoided in patients who may require frequent or daily administration of an anxiolytic drug. Perhaps the greatest drawback of benzodiazepines, however, is that they can lead to abuse and/or dependence. Physical dependence is characterized by increased tolerance to these drugs and the development of significant withdrawal symptoms upon discontinuation; this occurs with long term and/or high dose use of benzodiazepines and is not necessarily a sign of misuse or addiction (although patients should be made aware of the need for very gradual taper of these medications if used long term). Addiction, on the other hand, which may include elements of physical dependence, is characterized by maladaptive behavioral changes leading to medication misuse. Benzodiazepines (along with barbiturates discussed below) are controlled substances which should be prescribed judiciously and cautiously and only when adequate follow-up is available to ensure appropriate use. It should be noted that adequate follow-up is not often feasible in primary care settings. Benzodiazepines should generally be avoided in any patient with a history of substance or alcohol abuse: most benzodiazepine abuse or dependence occurs in these individuals. Admittedly, there are circumstances in which a patient with a history of substance abuse may require benzodiazepines (e.g., a patient with a severe debilitating panic disorder who has been refractory to all other non-benzodiazepine medications)—these circumstances, however, should be considered infrequent.

Barbiturates
Developed in the 1940’s and 50’s, barbiturates are now rarely used for the treatment of anxiety due to a higher risk of dependence and dangerousness in overdose when compared to benzodiazepines. Whereas benzodiazepine binding increases the receptor’s affinity for GABA and indirectly affects chloride channels, barbiturates (and alcohol), binding on a different site on GABAA receptors, can increase chloride influx into neurons even when GABA is not present. Chloral hydrate, a weak barbiturate, is still occasionally used in certain settings for the treatment of refractory insomnia or for sedation prior to anxiety provoking medical studies (e.g., MRI). Other barbiturates such as phenobarbital, pentobarbital, and butalbital, are still commonly used in other areas of medicine for treatment of conditions (e.g., seizure disorder, pain) other than anxiety disorders.

Medicines without Abuse Potential Used for the Treatment of Anxiety
Buspirone is a partial 5HT1A agonist (primarily on autoreceptors) causing decreased serotonin release from serotonergic neurons. It is not clear why this action would help decrease anxiety; any effect is presumed to be due to "downstream" adaptations over several weeks. Buspirone has no effect on GABA receptors and as such cannot immediately replace benzodiazepines. It has no immediate anxiolytic effects. However, it has no potential for abuse, and does not impair cognition or motor coordination. Side effects, however, may include headache, insomnia, jitteriness and nausea. It is indicated for the treatment of Generalized Anxiety Disorder. Propranolol is a beta-adrenergic antagonist. Although it is primarily used medically for its effect on heart rate and blood pressure, its "off-label" use in psychiatry is based on its ability to reduce overall sympathetic activation. It is particularly helpful in circumstances where a sympathetic reaction to an anxiety-provoking stimulus can occur, such as in instances of performance anxiety. Musicians, for example, may take a dose one hour prior to their appearance on stage. Propranolol can decrease somatic manifestations of anxiety such as tremulousness and tachycardia. It does not, however, help alleviate symptoms associated with generalized social phobia or generalized anxiety disorder. Propranolol should be avoided if the patient has congestive heart failure or significant asthma. Despite earlier concerns that beta-blockers may cause depression (Waal, 1967) this is not supported by more recent studies.(Ko, Hebert et al. 2002) Clonidine, initially developed as an antihypertensive, is an alpha-2-adrenergic autoreceptor agonist which serves to decrease sympathetic drive in the locus ceruleus. It can decrease hyperarousal in patients with post-traumatic stress disorder (Boehnlein and Kinzie, 2007) and other conditions associated with autonomic hyperactivity (e.g., rebound hyperactivity in opioid withdrawal states). Prazosin is an alpha-1-adrenergic receptor antagonist. Like clonidine, it is an antihypertensive which can decrease anxiety symptoms associated with post- traumatic states. It has no sedative properties but it can help decrease PTSD symptoms during the day and decrease associated sleep disturbances and nightmares at night.(Miller, 2008; Raskind, Peskind et al. 2007; Taylor, Lowe et al. 2006; Taylor, Martin et al. 2008) Hydroxyzine is an antihistamine with less affinity for muscarinic and alpha-1-adrenergic receptors than other antihistamines. Because it does not cause dependence and has no abuse potential, it is useful for treating anxiety symptoms in patients with a history of substance abuse. Also, it has been shown to have efficacy in the treatment of generalized anxiety disorder.(Llorca, Spadone et al. 2002; Lader and Scotto, 1998; Ferreri, and Hantouche et al. 1994)

Newer Hypnotics
Zolpidem, zaleplon, and eszopiclone (enantiomer of zopiclone which is available in Europe) are newly developed non-benzodiazepine hypnotics that bind to alpha-1 subunits on the benzodiazepine binding site on GABA receptors.(Sanger, 2004) These "z-drugs" cause sedation but lack anxiolytic effects despite some cross-reactivity with benzodiazepines. Although their abuse potential is purportedly less than benzodiazepines, they are not free from the risk of dependence and withdrawal symptoms upon discontinuation.(Cubala and Landowski, 2007; Liappas, Malitas et al. 2003; Sethi and Khandelwal, 2005; Victorri-Vigneau, Dailly et al. 2007) There is insufficient evidence that these hypnotics are either safer or more effective than benzodiazepines, yet, due to physician misconceptions, they have been widely used when more cost-effective treatments should be considered.(Siriwardena, Qureshi et al. 2006; Glass, Lanctot et al. 2005)  Zolpidem is now generic, but zaleplon and eszopiclone are still brand products and are expensive. Ramelteon is a new hypnotic that is a melatonin receptor agonist which may have modest efficacy in shortening sleep latency.(Sateia, Kirby-Long et al. 2008; Roth, Seiden et al. 2006) It is well-tolerated (Johnson, Suess et al. 2006) but again there are no studies to suggest it should be favored over more cost-effective alternatives. When considering treatment of chronic insomnia, physicians should not overlook the benefits that may be derived from nonpharmacological, e.g., behavioral, therapies.(Sivertsen, Omvik et al. 2006) Table 2 summarizes the characteristics of selected non-antidepressant medications for the treatment of anxiety and insomnia.(WHO 2007; PDR 2008; Hyman, Arana et al. 1995; Rosenbaum, Arana et al. 2005; Stahl, 2005; Taylor, Paton et al. 2007) Antidepressants used in the treatment of anxiety disorders are listed in Table 1.

'''TABLE 2. SELECTED NON-ANTIDEPRESSANT MEDICATIONS FOR THE TREATMENT OF ANXIETY AND INSOMNIA'''


 * ® Generic and U.S. brand name(s).
 * Dosing should be adjusted downwards ("start low, go slow" strategy) for the elderly and/or the medically compromised.
 * Abbreviations: bid-(bis in die) twice a day; CYP-Cytochrome P450 enzyme; DEA-Drug Enforcement Administration; FDA-Food and Drug Administration; IM-intramuscular; mg-milligram; po-(per os) orally; PTSD-Post-traumatic Stress Disorder; qhs-(quaque hora somni) at bedtime; qid-(quater in die) four times a day; tid-(ter in die) three times a day;  WHO-World Health Organization.

First Generation Antipsychotics (FGAs)
The first antipsychotic, chlorpromazine, was developed in the 1950s.(Meyer and Simpson 1997) Subsequently other antipsychotics were developed that share similarities in their mechanisms of action and in their side effect profiles. Chlorpromazine, thioridazine, perphenazine, molindone, thiothixene, pimozide, fluphenazine, and haloperidol are examples of these medications which are now characterized as "first generation antipsychotics" (FGAs). Alternative names for this class of medications include: "neuroleptics" (for their propensity to cause adverse neurological effects), "major tranquilizers" (as opposed to the later designated "minor tranquilizers" such as benzodiazepines and barbiturates), "typical" antipsychotics, and "conventional" antipsychotics. All first generation antipsychotics are believed to exert their antipsychotic effects through post-synaptic D2 dopamine receptor antagonism, thereby reducing the effect of endogenous dopamine released by presynaptic dopaminergic neurons.(Nestler, Hyman et al. 2009) Dopaminergic neurons originate from 3 distinct nuclei. One group of dopaminergic neurons projects from the ventral tegmental area of the midbrain to the nucleus accumbens, cingulate cortex and prefrontal cortex (the mesolimbic and mesocortical tracks); these affect emotions and cognition and as such are the targets for the therapeutic effects of antipsychotic drugs. Dopaminergic neurons also arise from the substantia nigra and project to the striatum (the nigrostriatal track); these are implicated in the neurological side effects of antipsychotics. And finally, hypothalamic dopaminergic neurons project to the pituitary gland and serve to regulate the release of prolactin (the tuberoinfundibular track); disruption of this system with D2 blockade can result in hyperprolactinemia associated with the use of antipsychotics. Overall, in terms of clinical use, it has been shown that the optimal D2 receptor occupancy level for maximizing antipsychotic effect while minimizing adverse effects is 60-70%.(Farde, Nordstrom et al. 1992) FGAs have traditionally been divided into low potency (e.g., chlorpromazine, thioridazine), mid-potency (e.g., perphenazine, molindone, thiothixene), and high potency (e.g., haloperidol, fluphenazine) antipsychotics, based on the number of milligrams of each drug needed to show comparable efficacy. For example, chlorpromazine 300 mg (low potency) may have the same therapeutic effect as perphenazine 24 mg (mid-potency) and haloperidol 6 mg (high potency). FGAs are often listed as a spectrum from low to high potency. The low potency antipsychotics usually exhibit TCA-like side effects such as anticholinergic, antihistaminic, and orthostatic effects (see section on antidepressants) but have a lower risk of causing acute neurological side effects such as acute muscle dystonias. At the other end of the spectrum, high potency FGAs have lower risks of TCA-like adverse effects but a much higher risk of causing acute dystonias. Mid-potency antipsychotics share all these side effects but less so than those of either pole. Physicians should become familiar with using at least one antipsychotic from each potency class in order to be able to match the side effect profile to the patient's pre-existing vulnerabilities. When dosing FGAs, it is important to consider that although in most efficacy studies the presumed therapeutic dose of haloperidol is 10 mg/day, the ideal dose may be much lower: haloperidol 2 mg/day in neuroleptic-naïve patients, and 4 mg in non-neuroleptic-naïve patients may be sufficient to produce a "neuroleptic threshold"—the dose at which cogwheel rigidity, a sign of sufficient D2 receptor blockade, first appears.(McEvoy, Stiller et al. 1986) The propensity of FGAs to cause neurological symptoms such as acute muscle dystonias, parkinsonism, akathisia, and tardive dyskinesia significantly limits their use in current practice. Acute dystonias, which are more likely to occur if the patient is young, male, has a history of substance abuse and/or a prior history of dystonias, is primarily seen in patients taking FGAs although it can also occur in patients with any antipsychotic with significant D2 receptor antagonism (see risperidone below). Use of anticholinergic medications, such as benztropine or diphenhydramine (or promethazine used outside the U.S.) can decrease the occurrence of early dystonias. Parkinsonism, characterized by bradykinesia, tremor, rigidity, and masked facies, can develop after 1-4 weeks of treatment with FGAs. Anticholinergic medications or a dopamine releasing agent such as amantadine may be helpful, although changing the antipsychotic may be required. Akathisia, which is an unpleasant subjective sense of inner restlessness relieved by movement, is also commonly seen in patients treated with FGAs. Identifying akathisia as a cause of agitation (or even worsening psychosis or suicidality) is important because treatment would include decreasing, rather than increasing, antipsychotic dose. Tardive dyskinesia (TD), a potentially irreversible syndrome of abnormal involuntary movements, can develop with extended use of antipsychotics, especially if high doses are used for long periods of time. Patients with prolonged antipsychotic treatment, a history of affective disorders, a history of parkinsonian side effects with initial antipsychotic treatment, as well as women and the elderly, are at a higher risk for developing TD. Once tardive dyskinesia has developed, withdrawal of the antipsychotic (especially if this is precipitous) may unmask worsened abnormal movements. Resumption of antipsychotic treatment may suppress these symptoms for a period of time, but progression of the underlying movement disorder may continue. Despite trials of multiple remedies, treatments for TD are generally only partially effective.(Soares-Weiser and Fernandez 2007) Neuroleptic malignant syndrome (NMS), is a poorly understood, rare, but potentially fatal complication of treatment with FGAs and other antipsychotics. NMS is characterized by a constellation of symptoms which may include delirium, lead-pipe rigidity, autonomic instability and high fevers. It can develop very early in the course of antipsychotic treatment. A high serum creatine phosphokinase (CPK) and elevated white blood cell count are supportive of the diagnosis of NMS. If NMS appears likely then the offending antipsychotic should be immediately discontinued. Medical hospitalization is necessary and treatment may include the use of a dopamine agonist (e.g., bromocriptine), a muscle relaxant (e.g., dantrolene), aggressive hydration, and the use of benzodiazepines if needed for behavioral agitation.(Hu and Frucht, 2007) Once the patient has been medically stabilized, the offending agent should be avoided. Rechallenge may be possible two weeks after all symptoms of NMS have abated, optimally with a low potency antipsychotic.

Second Generation Antipsychotics (SGAs)
Second generation antipsychotics, also known as atypical antipsychotics, are believed to exert their antipsychotic effects through a similar mechanism of action, but have profiles of receptor activity that produce different side effects than FGAs. The first SGA to be developed was clozapine, which was followed by the sequential introduction of risperidone, olanzapine, quetiapine, ziprasidone, and aripiprazole in the U.S. (and amisulpride and zotepine among others, in other countries). Whereas FGAs were known to (1) possibly worsen (or at least not improve) the negative symptoms of schizophrenia and (2) cause extrapyramidal symptoms including TD, SGAs were hoped to be more effective in treating negative symptoms and less likely to cause movement disorders. It is true that, by and large, SGAs do not worsen negative symptoms of schizophrenia and have a much lower risk of causing TD.(Correll, Leucht et al. 2004) However, questions regarding the differential effectiveness of SGAs as compared with FGAs, and their greater risks of inducing other, non-neurological, adverse effects have served to dampen the optimistic expectations initially associated with the medications. Nevertheless, in most of the world where they are available, SGAs are still considered to be the first line for treatment of psychotic disorders. Risperidone, one of the earliest SGAs to be developed, was released in 1994. It is similar to FGAs in that it is a potent D2 receptor antagonist, but like many other SGAs, it is also a post-synaptic serotonin 5HT2A antagonist. This is thought to mitigate the D2 receptor-mediated neurological side effects. At doses lower than 6 mg/day (i.e., at usual therapeutic doses), risperidone carries a low risk of causing EPS; at higher doses, D2 blockade effects predominate and the risk of EPS increases significantly. EPS are usually not present at risperidone 3 mg/day—a dose at which 72% of D2 receptors are occupied.(Nyberg, Eriksson et al. 1999) The optimal dose derived from clinical studies appears to be 3-6 mg daily.(Osser and Sigadel, 2001)  Although generally a well-tolerated antipsychotic, the side effects of risperidone include possible hypotension and hyperprolactinemia, and in children and adolescents it produces considerable weight gain.(Sikich, Frazier et al. 2008)  It is a hepatic CYP2D6 enzyme substrate and therefore its metabolism can be slowed by (1) inhibitors such as fluoxetine and paroxetine,(Spina, Scordo et al. 2003) or (2) the CYP2D6 variant gene for slow metabolism, which results in a less active form of the CYP2D6 enzyme (found more often in Chinese and other East Asian individuals).(Bertilsson, 1995)  Another gene variant that causes "poor" metabolism is more common in Caucasians and results in severe side effects. On the positive side, risperidone has a low to medium propensity to cause adverse metabolic effects in adults (see discussion below). It may also have a somewhat more rapid onset of action compared to other second generation antipsychotics.(Osser and Sigadel, 2001) The newly introduced (and therefore more costly) paliperidone, which is the major active metabolite of risperidone with similar efficacy and similar side effect profile, is not metabolized by CYP2D6 and is mostly renally excreted. Olanzapine was introduced in 1996. It has less affinity for D2 receptors than risperidone and a greater affinity for 5HT2A and 5HT2C serotonin receptors. Olanzapine also has significant antihistaminic and anticholinergic effects. Although it is an effective antipsychotic for the treatment of schizophrenia (especially at doses equal or greater than 15 mg/day),(Osser and Sigadel, 2001) it is (along with clozapine as discussed below) associated with a high risk of developing weight gain, insulin resistance, and hyperlipidemia (i.e., "metabolic syndrome"). Concern about the increased morbidity and mortality associated with the metabolic syndrome has lead to a reduction of the use of olanzapine in recent years. All side effects increase when olanzapine is used at higher than recommended doses (e.g., 40 mg/day vs. the package insert maximum dose of 20 mg/day) with very little additional antipsychotic benefit.(Kinon, Volavka et al. 2008) Liver transaminases can also become transiently elevated with olanzapine. Quetiapine shows weak affinity at both dopamine and 5HT2 serotonin receptors, but may have similar receptor occupancy to the more potent SGAs.(Seeman, 2002) It has alpha-adrenergic antagonism and antihistaminic effects, causing orthostasis and sedation, respectively. Quetiapine is less likely than olanzapine and clozapine, but more likely than most FGAs, risperidone and other SGAs, to cause metabolic side effects. Quetiapine, at low doses, is widely (and perhaps too readily) used in psychiatric practice for the treatment of insomnia and acute anxiety in a wide range of patients with personality and/or substance abuse disorders for whom benzodiazepine use may be problematic. This "off-label" use should only be carried out after a thoughtful review of risks, benefits, and alternative treatments, especially evidence-supported treatments, for patients' diagnosed conditions. Clinicians should be aware also of recent reports of abuse and "street value" for this medication.(Hanley and Kenna, 2008) Use of quetiapine for anxiety symptoms may be more appropriate in acute care settings such as during hospitalizations. Quetiapine’s effectiveness in psychotic disorders may be less than that of olanzapine and risperidone,(McCue, Waheed et al. 2006; Suzuki, Uchida et al. 2007) but it may have a stronger role in bipolar disorders (see section on mood stabilizers). Ziprasidone is an SGA with moderate D2 antagonism and significant 5HT2A antagonism (i.e., a high 5HT2A/D2 ratio). Although it is not clear if it is as effective as olanzapine and risperidone in the acute treatment of schizophrenia,(McCue, Waheed et al. 2006) it does not cause metabolic changes, and may even improve lipid profile, especially if the patient was previously on a weight gain-inducing agent.(Lieberman, Stroup et al. 2005) A major issue with using ziprasidone is the necessity of taking it with food, or it will not be well-absorbed. A 500 calorie meal is optimal with each of the twice daily doses.(Miceli, Glue et al. 2007) Ziprasidone has the potential to prolong QT more than other SGAs. Although a pre-treatment ECG is not required, those who are deemed, based on history or age, to be at higher cardiac risk would benefit from an ECG (and medical consultation if appropriate) before starting ziprasidone. Electrolyte disturbances such as hypomagnesemia and hypokalemia should be corrected. Other QT prolonging medications should not be used in combination with ziprasidone. Despite concerns regarding this effect, post-marketing studies (e.g., CATIE)(Lieberman, Stroup et al. 2005) did not show any clinically significant QT prolongation with ziprasidone use. Medical students and physicians should be aware that all antipsychotics (with the possible exception of aripiprazole discussed below) can affect cardiac conduction, potentially delaying conduction enough to lead to fatal arrhythmias. There is an association between the use of antipsychotics (as well as tricyclic antidepressants) and sudden death.(Ray, Chung et al. 2009; Ray, Meredith et al. 2004; Straus, Bleumink et al. 2004) As discussed in the section on antidepressants, prolonged QTc is associated with torsades de pointes, a potentially fatal arrhythmia. The QT interval includes both the QRS interval as well as the ST segment. Whereas TCAs and some FGAs with tricyclic structure (e.g., chlorpromazine) lengthen the QRS interval by interfering with sodium channels and depolarization, most other antipsychotics, including SGAs, can affect potassium channels and the repolarization phase.(Glassman and Bigger, 2001) Both effects would be reflected in the QT interval. Although it is not clear if QT prolongation is actually a reliable indicator of the risk of torsades, measuring this interval is the simplest way to estimate this risk.(Shah, 2005) In addition to ziprasidone, the FGAs thioridazine, mesoridazine, pimozide, and droperidol are among the antipsychotics with the highest propensity to prolong the QT interval.(Fayek, Kingsbury et al. 2001) Aripiprazole, in contrast to other SGAs, is a high affinity partial agonist at the D2 receptor.(Mamo, Graff et al. 2007) It is postulated that aripiprazole decreases overall dopamine effect(Stahl, 2008) in dopamine rich environments (e.g., in mesolimbic pathways--thereby ameliorating psychosis), and increases dopamine effect in dopamine depleted environments (e.g., in mesocortical pathways to the prefrontal cortex--thereby improving negative symptoms such as social withdrawal).(Stahl, 2008)  At therapeutic doses it highly saturates the targeted dopamine receptors and shows very slow dissociation from the receptors upon discontinuation.(Goff, 2008; Grunder, Fellows et al. 2008)  It also shows moderate 5HT2A and 5HT2C antagonism. On the other hand, it is free from anticholinergic and significant antihistaminic effects. More importantly, it does not appear to have significant cardiac or metabolic effects.(El-Sayeh, Morganti et al. 2006) Although it is less likely to cause EPS in general, it has been observed in practice to cause akathisia more readily than other SGAs. This side effect may be more common if the patient was recently on a strong D2 antagonist such as an FGA or risperidone and consequently has an up-regulated or hypersensitive population of D2 receptors.(Raja, 2007) Slow dose titration and/or combination with a benzodiazepine may be necessary to reduce the risk of akathisia. Aripiprazole at 15 mg/day may be more efficacious than 30 mg/day in schizophrenia, although full response may take longer than with a comparable dose of haloperidol.(Kane, Carson et al. 2002) Higher doses (e.g., 30 mg/day) may be more useful in treatment-resistant schizophrenia.(Kane, Meltzer et al. 2007)  Relapse rates may be somewhat higher with aripiprazole than with other SGAs.(Pigott, Carson et al. 2003) Clozapine, the first and in some respects the most impressive of the second generation antipsychotics, binds weakly at the D2 receptor (although with relatively greater net antagonism at D3 and D4 dopamine receptors) and has moderate affinity for 5HT2A and 5HT2C receptors. It is often effective when other antipsychotics are not,(Lewis, Barnes et al. 2006) and appears to have antisuicidal effects in patients with schizophrenia or schizoaffective disorder.(Meltzer, Alphs et al. 2003) However, because of the risk of agranulocytosis, it is reserved primarily for schizophrenic and schizoaffective patients who have failed to respond adequately to at least two other antipsychotics. Strict monitoring and initially weekly and then biweekly blood draws are required to monitor white cell count. The clinician should consult the package insert and strictly follow the monitoring guidelines. Clozapine should not be combined with other medications (e.g., carbamazepine) that may also cause leukopenia. Clozapine can also cause multiple other adverse effects, which include (but are not limited to) an increased risk of seizures, rare myocarditis, eosinophilia, anticholinergic and antihistaminic effects, orthostasis, weight gain and adverse metabolic effects.(Lamberti, Olson et al. 2006) Among the SGAs, clozapine and olanzapine are the most likely (and aripiprazole and ziprasidone are the least likely) to cause adverse metabolic effects. These would include weight gain, hyperglycemia and diabetes (with or without weight gain) and hyperlipidemia.(ADA, 2004) A 2-3 kilogram weight gain within the first 3 weeks of treatment often predicts the risk of significant weight gain over the long term.(Lipkovich, Citrome et al. 2006)  Decreased insulin secretion and increased triglycerides (i.e., the lipids most affected by SGAs)(Osser, Najarian et al. 1999) can also be seen within 1-2 weeks of treatment.(Chiu, Chen et al. 2006)  Treatment with the hypoglycemic medication metformin, especially if combined with lifestyle changes, may reduce antipsychotic-induced weight gain.(Baptista, Rangel et al. 2007; Wu, Zhao et al. 2008) Prior to starting olanzapine or clozapine, measurements of baseline weight, serum glucose, and lipid profile should be obtained. If the patient has pre-existing diabetes, other antispychotics should be considered. Once treatment is initiated, serum glucose and weight should be monitored and if glucose levels become elevated, a glucose tolerance test—which can predict up to 96% of patients who would develop diabetes—should be done.(van Winkel, De Hert et al. 2006) If metabolic problems do arise during treatment, switching to another antipsychotic should be considered. Clozapine may cause orthostatic hypotension. Patients who are elderly, have cardiac histories, or who are taking antihypertensives are at higher risk for this side effect. Clozapine should be increased gradually after treatment is initiated (starting at 12.5 mg/day and increasing the dose by 25 mg daily as tolerated). Usually, patients adjust and become tolerant to the hypotensive effects of this medication. However, this tolerance may not last longer than 48 hours. If a patient discontinues clozapine therapy for more than 48 hours, treatment should be restarted with a 12.5 mg dose. After that, the dose may be more quickly raised to the previous dose as tolerated. It is important for the physician who may be admitting a psychiatric patient to the medical or surgical ward of the hospital to stop and think before continuing clozapine at its prior dose: recent compliance needs to be verified first. Given the complicated nature of clozapine treatment, the clinician should refer to a more in depth discussion of this drug before use.(Phansalkar and Osser, 2009; Phansalkar and Osser. 2009)

Long-Acting Injectable Antipsychotics
In the United States, 3 antipsychotics are available for long-acting (i.e., depot) intramuscular administration: haloperidol decanoate, fluphenazine decanoate, and long-acting injectable risperidone. Olanzapine long-acting injection is expected to be approved soon. These long-acting formulations are options for patients who are frequently non-adherent to the prescribed oral medication.(Olfson, Marcus et al. 2007) A brief trial of the antipsychotic in oral form is first prescribed to assess patients' response to, and tolerance of, the selected agent. Every four week injections of haloperidol decanoate or biweekly injections of long-acting fluphenazine or risperidone are then continued while the oral agent is gradually tapered. Four to five repeated injection cycles of the selected antipsychotic may be necessary to achieve steady state before oral medications should be completely withdrawn.(Osser and Sigadel ,2001) Patients who adhere poorly to oral medications in real-world public sector settings are generally non-adherent to depot antipsychotics as well.(Olfson, Marcus et al. 2007)  These formulations seem to work best in research subjects and in other populations of relatively cooperative and less treatment-resistant patients. On the other hand, once steady state is achieved, depots do have the advantage that if the patient discontinues treatment, the antipsychotic effect can continue for up to several months after the last received dose.

Antipsychotics for Behavioral Control
Both SGAs and FGAs are used in psychiatric practice to treat behavioral agitation. In acutely psychotic and/or manic patients, FGAs, such as oral or intramuscular haloperidol (often combined with lorazepam and/or benztropine to decrease the risk of acute dystonias), remain the mainstay of treatment.(Ansari, Osser et al. 2009, Osser and Sigadel, 2001) Newer antipsychotics, such as olanzapine, ziprasidone, and aripiprazole are also available in short-acting intramuscular form but they are expensive and seem to have no advantage when compared with the combination therapy noted above.(Satterthwaite, Wolf et al. 2008)  When considering antipsychotics for behavioral agitation, clinicians should be advised not to use (1) intramuscular droperidol due to high risk of QT prolongation, (2) intramuscular chlorpromazine due to risk of severe hypotension, (3) intramuscular ziprasidone if the patient is taking other medications that can also prolong QT, or (4) intramuscular olanzapine in combination with lorazepam or other benzodiazepines due to the risk of hypotension.(Zacher and Roche-Desilets, 2005) The use of antipsychotics for the treatment of behavioral agitation in elderly patients with dementia is problematic both in terms of effectiveness and tolerability. First, in terms of effect, they do not appear to provide more than minimal benefit in targeting symptoms of agitation, and SGAs may not be different from placebo in this regard.(Yury and Fisher, 2007) The NIMH-sponsored CATIE-AD study, which studied the effectiveness of olanzapine, quetiapine and risperidone in the treatment of symptoms of psychosis, aggression and agitation in patients with Alzheimer's disease, also found that even when these symptoms did improve with treatment, the antipsychotic did not improve overall functioning.(Sultzer, Davis et al. 2008)  Furthermore, any improvement in specific symptoms was offset by adverse effects and led to overall discontinuation rates (when both efficacy and tolerability were considered) that were no different from placebo.(Schneider, Tariot et al. 2006)  Secondly, antipsychotics have been found to be associated with an increased risk of stroke in patients with dementia and an overall increased risk of adverse medical events and death in this population.(Gill, Rochon et al. 2005; Herrmann and Lanctot, 2005; Rochon, Normand et al. 2008; Schneider, Dagerman et al. 2005)  Both FGAs and SGAs appear to increase the risk of death in patients with dementia.(Schneeweiss, Setoguchi et al. 2007; Wang, Schneeweiss et al. 2005) High-potency FGAs are also often used in the treatment of delirium in hospitalized patients.(Lonergan, Britton et al. 2007) Medical students and physicians should be aware that although antipsychotics may treat the secondary manifestations of delirium, such as behavioral agitation and/or hallucinations, they do not treat the underlying condition. Delirium is a medical condition that is treated by addressing the underlying medical cause.

Further Notes on the Clinical Use of Antipsychotics
All antipsychotics are indicated for the treatment of schizophrenia and are considered reasonably safe and effective for this debilitating disorder. However, there has been much debate about whether there are efficacy differences among these medications, or whether the side effect differences, which are considerable, should be the primary basis for selecting a medication for a particular patient. A meta-analysis of 78 head-to-head comparisons in the literature through 2007 concluded that the efficacy differences are small, but there was some superiority to olanzapine and risperidone, when compared with aripiprazole, quetiapine, and ziprasidone. A problem with this meta-analysis, however, was that almost all of the studies were industry-sponsored. Such studies invariably find outcomes in favor of the sponsor’s product, and olanzapine and risperidone have sponsored the largest number of studies. Another meta-analysis focused on 150 studies that directly compared FGAs with SGAs.(Leucht, Corves et al. 2009) The authors found that clozapine was clearly superior to the others especially for positive symptoms of hallucinations and delusions. Olanzapine and risperidone were superior to the others, but with a small effect size. The others did not differ in efficacy. The side effect profiles differed markedly, with no pattern to the differences. The authors recommended abandoning the terms "FGA" and "SGA" as irrelevant to efficacy or side effects.

Many clinicians put more reliance on the relatively few comparison studies that were independently funded, such as the CATIE (Clinical Antipsychotics Trials of Intervention Effectiveness) and TEOSS (Treatment of Early-Onset Schizophrenia Spectrum Disorders) studies.(Lieberman, Stroup et al. 2005; Sikich, Frazier et al. 2008) Both were funded by the U.S. National Institute of Mental Health. These studies prospectively compared FGAs (perphenazine, molindone) and SGAs (clozapine, olanzapine, risperidone, quetiapine, ziprasidone) and found generally no differences in effectiveness except that clozapine was superior. There were no differences in the ability to improve impaired cognition, despite prior claims for SGA superiority from studies sponsored by the SGA pharmaceutical firms.(Keefe, Bilder et al. 2007) Some experts have interpreted CATIE as showing olanzapine to be superior to the other non-clozapine antipsychotics, but this seems likely to be due to peculiar results with the cohort of patients who were on olanzapine prior to entering the CATIE study. These patients (22% of the sample) were randomly assigned to either continue on olanzapine or be switched to one of the other options in CATIE (perphenazine, risperidone, quetiapine, or ziprasidone). The patients who were assigned to remain on olanzapine did better than those who were abruptly switched to any of the other options.(Essock, Covell et al. 2006) By contrast, the patients who entered the study on risperidone (the second largest group with 19%) showed no advantage to staying on risperidone compared to switching to another agent. Notably, there was no advantage to switching to olanzapine. Hence, the superiority of olanzapine seen in CATIE may be due to the study having a large sample of patients who had been previously stabilized on olanzapine and who clearly responded only to olanzapine or who may have been more prone to a withdrawal-induced exacerbation when taken off of olanzapine. Since olanzapine has a very unfavorable side effect profile with its tendency to promote weight gain, insulin resistance, and the metabolic syndrome, this would appear to make it undesirable as a first-line choice even if it does have slightly superior efficacy. Antipsychotics are also being used in a wide variety of mood and anxiety disorders as augmentations when antidepressants produce unsatisfactory results, and sometimes they are used as primary treatments for these disorders. Recent data indicating that SGAs and FGAs are associated with double to triple the rate of death from sudden cardiac arrest (presumably from electrophysiological effects related to QT prolongation) suggest that these agents should not be first-line treatments in these clinical situations.(Ray, Chung et al. 2009) However, antipsychotics are powerful and important options in the treatment of schizophrenia and severe bipolar disorders and these new cardiac concerns should not deter clinicians for prescribing them appropriately for these patients. Obtaining a baseline ECG, and if abnormal obtaining another after dosage has been optimized, is a prudent risk-management approach given this new data. Table 3 summarizes the characteristics of commonly used antipsychotics.(WHO, 2007; PDR, 2008; Hyman, Arana et al. 1995; Rosenbaum, Arana et al. 2005; Stahl, 2005; Taylor, Paton et al. 2007)

'''TABLE 3. COMMONLY USED ANTIPSYCHOTICS'''


 * ® Generic and U.S. brand name(s).
 * Dosing should be adjusted downwards ("start low, go slow" strategy) for the elderly and/or the medically compromised.
 * Abbreviations: ANC-Absolute Neutrophil Count; bid-(bis in die) twice a day; CYP-Cytochrome P450 enzyme; ECG-Electrocardiogram; EPS-Extra-pyramidal Symptoms; FGA-First Generation Antipsychotics; IM-intramuscular; MDD-Major Depressive Disorder; mg-milligram; ng/mL-nanogram per milliliter; po-(per os) orally; q-(quaque) every; qhs-(quaque hora somni) at bedtime; SGA-Second Generation Antipsychotics; TCA-Tricyclic Antidepressants; WBC-White Blood Cell; WHO-World Health Organization.

Mood Stabilizers
What is a mood stabilizer? Although there is no generally accepted definition, a mood stabilizer can be defined as a medication that can treat either phase of bipolar disorder while not inducing or worsening the other phase. More conservatively, however, a mood stabilizer can be defined as an agent that can both treat and prevent both manic and depressive episodes. By this definition only lithium qualifies as a true mood stabilizer.(Bauer and Mitchner, 2004) Lithium (as a salt) has been used as a homeopathic treatment for gout and other disorders since the 1800’s. Its calming effect on animals, and subsequently on manic patients, was first described in the 1940s.(Cade, 1949) In the brain, lithium inhibits inositol phosphatases that dephosphorylate inositol phosphates that are generated by the stimulation of G proteins in neuronal membranes activated by a neurotransmitter. This inhibition interferes with inositol regeneration and leads to its depletion in neurons, ultimately leading to decreased neuronal activity.(Nestler, Hyman et al. 2009) Lithium also inhibits protein kinases, glycogen synthase kinase-3beta, and adenylyl cyclase,(Bachmann, Schloesser et al. 2005; Lenox and Hahn 2000) and may increase the uptake of the excitatory neurotransmitter glutamate thereby reducing glutamate activity at the neuronal synapse.(Shaldubina, Agam et al. 2001)  Lithium also appears to have neuroprotective properties and may promote neurogenesis.(Chuang, 2005; Chen and Manji, 2006; Bearden, Thompson et al. 2007; Nunes, Forlenza et al. 2007; Fornai, Longone et al. 2008) Lithium is effective in both manic and depressive episodes associated with bipolar disorder, as well as for long-term maintenance. It is also the only mood stabilizer with anti-suicidal effect.(Baldessarini, Tondo et al. 1999; Cipriani, Pretty et al. 2005) Lithium works particularly well in patients who have a strong family history of bipolar disorder.(Alda, 1999) A target therapeutic serum level of 0.6-0.75mEq/L is recommended for the treatment of bipolar depression and prophylaxis against depressive relapses.(Kleindienst, Severus et al. 2007; Kleindienst, Severus et al. 2005; Severus, Kleindienst et al. 2008) Serum levels of 0.75-1.2mEQ/L may be more effective for the treatment of mania. Serum levels higher than 1.2mEq/L are associated with significant lithium toxicity. Lithium side effects usually increase with higher serum doses but can occur at any dose. These may include nausea, vomiting, diarrhea, tremor, thirst, polyuria, acne, weight gain, and a benign leukocytosis. Over the long run, lithium can cause hypothyroidism in up to 20% of patients(Johnston and Eagles 1999) (which can be treated with thyroid hormone replacement), and worsening renal function in 20% of patients (Lepkifker, Sverdlik et al. 2004) (which usually necessitates lithium discontinuation). Because of the many complexities of lithium use, access to references such as the Lithium Encyclopedia for Clinical Practice is recommended.(Jefferson, Greist et al. 1987) Physicians should be aware that the anti-manic effects of lithium (and other mood stabilizers, such as valproate and carbamazepine, discussed below) may not be achieved until 7-10 days after a therapeutic dose has been established. In the interim, sedative medications such as antipsychotics and benzodiazepines may be needed when the patient is acutely manic. Once the patient is stabilized, these adjunctive medications can often be tapered and lithium continued as monotherapy. Valproate (along with carbamazepine and lamotrigine discussed below) is an anticonvulsant with mood stabilizing properties. It is postulated that it exerts its effect via enhancement of GABA transmission.(Johannessen, 2000) Despite decades of clinical experience with lithium, valproate has become the most widely used mood stabilizer in the United States. This is primarily due to its ease of use and effectiveness in the treatment of mania. Valproate is not very effective in the treatment of bipolar depression, but has efficacy in the treatment of manic and mixed episodes (Bowden, Brugger et al. 1994; Freeman, Clothier et al. 1992) Although serum levels of 50-125 mcg/mL are generally considered to be within the therapeutic range (a range based on anticonvulsant data), the best results in acute mania may occur with levels of greater than 90 mcg/mL.(Allen, Hirschfeld et al. 2006)  Side effects in adults include liver enzyme elevations (usually benign and transient, but not always), weight gain, and possible thrombocytopenia and platelet dysfunction. Because of the latter, bleeding time should be measured prior to surgery even if the platelet count is normal.(De Berardis, Campanella et al. 2003; Gerstner, Teich et al. 2006) Valproate is highly protein-bound: concurrent use with warfarin can displace and increase the free fraction of warfarin and increase prothrombin time. The use of valproate is problematic in women of child bearing age. It is associated with a high risk of teratogenic effects (i.e., neural tube defects).(Cohen, 2007; Viguera, Koukopoulos et al. 2007) Valproate may also play a role in the development of polycystic ovary syndrome.(Joffe, Cohen et al. 2006; O'Donovan, Kusumakar et al. 2002) Many clinicians recommend the use of alternative medications, such as lithium in young bipolar women. Lithium can cause Ebstein's anomaly, an anomaly in the fetal tricuspid valve, at up to 3 times the baseline risk. Previously, this risk was thought to be much higher. For a more complete discussion of pregnancy risks of psychiatric medications, see the 2008 Practice Bulletin of the American College of Obstetrics and Gynecology.(ACOG, 2008) Carbamazepine is an anticonvulsant that can enhance Na+ channel inactivation, thereby blocking action potentials and repetitive neuronal firing. It is also thought to inhibit a process known as kindling—a process whereby repeated subthreshold electrical stimuli can lead to the development of spontaneous seizures. Hypothetically, subthreshold environmental stimuli or prior manias can similarly kindle the development and frequency of further manias.(Rosenbaum, Arana et al. 2005) Carbamazepine has efficacy in the treatment of manic (Weisler, Kalali et al. 2004; Weisler, Keck et al. 2005), but probably not depressive,(Ansari and Osser, 2009) episodes associated with bipolar disorder. Serum levels of 4-12 mcg/mL are considered to be therapeutic. Side effects such as dizziness, ataxia and gastrointestinal symptoms prohibit the use of loading strategies. Thrombocytopenia, leukopenia, hyponatremia, and dangerous rash may also develop with carbamazepine therapy. Another factor that significantly limits treatment with carbamazepine, especially in severe mania when concurrent antipsychotics may be necessary, is its propensity to induce multiple hepatic enzymes (e.g., CYP1A2, CYP2C9, CYP2C19, CYP3A4). It can therefore increase the metabolism of other concurrently administered drugs and render them less effective. (Notably, the antiepileptic drugs phenobarbital, phenytoin, and primidone also have similarly broad hepatic enzyme induction capacities.)(Perucca 2006) Teratogenic effects of carbamazepine are comparable in severity to those of valproate.(Cohen, 2007; Viguera, Koukopoulos et al. 2007) Oxcarbazepine, an analog of carbamazepine, may also have efficacy in the treatment of acute mania.(Ghaemi, Berv et al. 2003; Pratoomsri, Yatham et al. 2006) Serum levels do not need to be checked, and there is less enzyme induction with oxcarbazepine, thereby reducing the risk of drug-drug interactions. Hyponatremia, however, remains a concern.(Ortenzi, Paggi et al. 2008) Lamotrigine is an anticonvulsant that may inhibit the release of the excitatory amino acid glutamate, but its mechanism of action is not fully known. In bipolar disorder it is often used (but not FDA approved) for the treatment of bipolar depression. However, 4 out of 5 studies failed to show separation from placebo.(Calabrese, Bowden et al. 1999; Calabrese, Huffman et al. 2008) Nevertheless, it appears to be effective as maintenance therapy for depressive episodes in bipolar disorder.(Bowden, Calabrese et al. 2003; Calabrese, Bowden et al. 2003)  Although lamotrigine is generally well-tolerated, there is a 0.1% risk of dangerous rash (i.e., toxic epidermal necrolysis—Stevens-Johnson syndrome). Gradual titration is required to decrease the risk of rash. If rash develops, lamotrigine should be discontinued. Lamotrigine so far seems relatively safe in pregnancy.(ACOG, 2008)

Second Generation Antipsychotics
All SGAs have been found to have efficacy in the treatment of mania.(Janicak, 2006) However, among the SGAs, only quetiapine has been shown to have clear efficacy in the treatment of bipolar depression.(Calabrese, Keck et al. 2005; Thase, Macfadden et al. 2006)  Olanzapine and aripiprazole have been found effective, as monotherapy treatments, for prevention of manic episodes, and quetiapine has unpublished data suggesting it may prevent both manic and depressive episodes. It may soon present a challenge to lithium as the mood stabilizer with the best evidence base. This remains to be seen. There is evidence that the SGAs olanzapine, quetiapine, ziprasidone, and aripiprazole, when added to lithium or valproate, can increase maintenance efficacy for the manic phase. In the case of quetiapine, the depressive phase was also helped.

Newer Anticonvulsants
Newer anticonvulsants such as topiramate, gabapentin, pregabalin, tiagabine, zonisamide and levetiracetam, which are generally thought to exert their therapeutic effects by enhancing GABA transmission, may also be effective for the treatment of bipolar disorder.(Johannessen and Landmark, 2008) However, when used, they should be considered to be adjunctive treatments only (e.g., to decrease concurrent anxiety); the evidence base is insufficient to recommend their use as primary agents for the treatment of mood disorder symptoms.(Anand, Bukhari et al. 2005; Grunze, Langosch et al. 2003; Grunze, Normann et al. 2001; Keck, Strawn et al. 2006; Macdonald and Young, 2002; Pande, Crockatt et al. 2000; Vieta, Goikolea et al. 2003; Vieta, Manuel Goikolea et al. 2006; Vieta, Sanchez-Moreno et al. 2003; Yatham, Kusumakar et al. 2002; Young, Geddes et al. 2006; Young, Geddes et al. 2006)

Further Notes on the Clinical Approach to Bipolar Patients Mood stabilization is frequently difficult to achieve in bipolar disorder. Although the goal is to use as few medications as possible and rely on mood stabilizers whenever possible, it is common that more complex psychopharmacology regimens are required. The American Psychiatric Association's recently released updated Practice Guidelines for the Treatment of Bipolar Disorder contains a comprehensive review of the current knowledge base on the psychopharmacology of this disorder.(APA, 2009) The Systematic Treatment Enhancement Program – Bipolar Disorder (STEP-BD) is a publically funded project designed to add to our understanding of how to best treat this disorder. The program enrolled 4,360 bipolar patients who are being followed longitudinally at 15 sites. Some of these patients agree to enter controlled studies of a variety of psychosocial and psychopharmacological interventions. Among the significant findings to date are the following: Table 4 summarizes the characteristics of commonly used mood stabilizing medications.(Hyman, Arana et al. 1995; Rosenbaum, Arana et al. 2005; WHO, 2007; PDR, 2008; Stahl, 2005; Taylor, Paton et al. 2007)
 * Psychotherapy is effective for bipolar depression but it is a slow process. It takes 169 days vs. 279 days in the control group.(Miklowitz, Otto et al. 2007)
 * Antidepressants (bupropion, paroxetine) are not more effective than placebo for bipolar depression (24% for the antidepressants vs. 27% for the placebo in a 6-month trial). The antidepressants did not induce more switches to mania (10% vs. 11%), but the patients who participated in this study were probably at very low risk for switching.(Sachs, Nierenberg et al. 2007)
 * Other STEP-BD data did show that use of antidepressants was associated with more manic symptoms.(Goldberg, Perlis et al. 2007)
 * 262 suicide attempts and 8 completed suicides have occurred in this patient sample over a 6-year period. Lithium seemed to offer no protective effect, contrary to data from other studies strongly suggesting that lithium helps lower suicide risk in bipolar patients.  However, the patient sample clearly had a very low risk of suicidal behaviors so it was not the best population to demonstrate lithium’s possible benefit on this symptom.(Marangell, Dennehy et al. 2008)

'''TABLE 4. COMMONLY USED MOOD STABILIZING MEDICATIONS'''


 * ® Generic and U.S. brand name(s).
 * Dosing should be adjusted downwards ("start low, go slow" strategy) for the elderly and/or the medically compromised.
 * Abbreviations: ACE-Angiotensin Converter Enzyme; bid-(bis in die) twice a day; CBC-Complete Blood Count; CYP-Cytochrome P450 enzyme; ECG-Electrocardiogram; kg-kilogram; LFT-Liver Function Tests; mg-milligram; NSAIDS-Non-steroidal Anti-inflammatory Drugs; tid-(ter in die) three times a day; TSH-Thyroid Stimulating Hormone; po-(per os) orally; WHO-World Health Organization.

Stimulants and Other ADHD Medicines
The use of psychotropics for the treatment of attention-deficit/hyperactivity disorder (ADHD) in children and adolescents is beyond the scope of this chapter. In adults, the diagnosis of ADHD is controversial but guidelines have been developed.(Gibbins and Weiss, 2007) Diagnosis of adult ADHD is predicated on plausible historical evidence of childhood onset.(APA, DSM, 2000)  This is often difficult to establish retrospectively, and when earlier ADHD symptoms are suspected, it is difficult to rule out other etiologies for these symptoms (e.g., family stressors, childhood depression, learning disorders, etc.). Nevertheless, there are adults with undiagnosed ADHD, many of whom have other comorbid psychiatric illnesses, who continue to suffer chronic symptoms through adulthood and may benefit from treatment. Others may have had a clear history and diagnosis of ADHD in childhood and as adults may need to have pharmacological treatments considered or resumed.

Stimulants
Stimulants are the most effective and the first-line treatment for non-substance-abusing patients with ADHD. Amphetamine-like stimulants are sympathomimetic amines that likely enhance norepinephrine and dopaminergic transmission. They may disrupt the presynaptic storage of these transmitters and enhance their release—in both the ascending reticular activating system as well as in the regulation of "top-down" cortical-thalamic-striatal circuits.(Nestler, Hyman et al. 2009) Methylphenidate, dextroamphetamine and amphetamine salts are examples of psychostimulants used in the treatment of ADHD. Assuming correct diagnosis and adequate dose, stimulants' beneficial effects on attentional symptoms, impulsivity and hyperactivity are immediate and subside with medication clearance. Short half-life formulations need to be administered multiple times during the day, but not near bedtime. Stimulant side effects include decreased appetite, insomnia, and anxiety, necessitating gradual dose titration to improve tolerability. Blood pressure and heart rate can also increase with stimulant administration so patients with cardiac disease may not be good candidates for treatment with stimulants. Possible growth retardation and the development of transient tics, although of concern in children, are not likely to be problematic in adults. Chronic stimulant use (or overdose) can lead to psychosis; in susceptible individuals, increased psychosis can be seen after just one dose.(Curran, Byrappa et al. 2004) Medication interactions of note include that stimulants should not be combined with MAOIs. The major concern regarding the use of stimulants, however, is the risk of abuse and dependence. This seems related to the stimulants' ability to increase dopaminergic effects in the reward and reinforcement circuitry in the nucleus accumbens. Euphoria, tolerance, and addictive behaviors may develop in susceptible individuals. In the United States, therefore, amphetamine-like stimulants are highly regulated; they are "Schedule II drugs"-- which indicates that the Drug Enforcement Administration (DEA) designates them as being in the highest risk category for controlled substances that have an established therapeutic use. The risks of addiction and misuse have led some clinicians to be wary of using stimulants even when treatment with these medications is otherwise medically indicated. However, if the diagnosis of ADHD is accurate, these medications should not be avoided in patients who do not have a history of substance abuse. A clear risk and benefit assessment is necessary. Appropriate monitoring and supervision may decrease the risk of abuse. Recent data suggest no increase in risk of subsequent abuse of stimulants when children and adolescents with ADHD are treated with stimulants.(Biederman, Monuteaux et al. 2008) Notably, there was no decrease in risk either, i.e., no protective effect from treatment. This is controversial however as ADHD is considered to be a risk factor for substance abuse; other data regarding treatment of ADHD as a way to decrease the risk of substance abuse is mixed.(Wilens, 2004; Wilens, Faraone et al. 2003) Stimulants have also been historically used in the treatment of anergic medically ill, mildly depressed, often elderly patients. Response can often be noted in a matter of days. There is no evidence however that these medications are effective antidepressants in other patients with major depression.(Satel and Nelson, 1989)

Non-Stimulant Medicines for ADHD
Atomoxetine is a norepinephrine and dopamine reuptake inhibitor (Bymaster, Katner et al. 2002) which has shown efficacy in, and has been primarily marketed for, the treatment of ADHD.(Michelson, Adler et al. 2003) As might be expected by its mechanism of action, it may also have antidepressant effects but there are no published data to support its use in the treatment of depression. Unlike stimulants which can rapidly improve ADHD symptoms, atomoxetine requires several weeks of treatment before response occurs. Response is generally less robust than with stimulants. Atomoxetine may cause increases in blood pressure, insomnia and possible weight loss. It is not associated with abuse or dependence. In adults, other agents with noradrenergic and/or dopaminergic effects may be helpful in the treatment of ADHD symptoms, although again response is generally weaker than that expected from stimulants.(Meszaros, Czobor et al. 2007) These include bupropion (Wilens, Spencer et al. 2001), tricyclic antidepressants (especially the more noradrenergic desipramine and nortriptyline), (Higgins, 1999; Prince, Wilens et al. 2000; Wilens, Biederman et al. 1996) venlafaxine (Popper, 1997), modafinil (a wakefulness-promoting drug with unknown mechanism of action)(Biederman, Swanson et al. 2006) and clonidine.(Connor, Fletcher et al. 1999)  SSRIs and antipsychotics are not effective in the treatment of ADHD. Table 5 summarizes characteristics of selected ADHD medications.(Hyman, Arana et al. 1995; Rosenbaum, Arana et al. 2005; PDR, 2008; Stahl, 2005; Taylor, Paton et al. 2007) Antidepressants used in the treatment of ADHD are listed in Table 1.

'''TABLE 5. SELECTED ADHD MEDICATIONS'''


 * ® Generic and U.S. brand name(s).
 * Dosing should be adjusted downwards ("start low, go slow" strategy) for the elderly and/or the medically compromised.
 * Abbreviations: ADHD-Attention Deficit/Hyperactivity Disorder; bid-(bis in die) twice a day; CYP-Cytochrome P450 enzyme; mg-milligram; po-(per os) orally; tid-(ter in die) three times a day; q-(quaque) every.

Treatments for substance abuse/dependence
The past few decades have seen a dramatic increase in the number of pharmacological options available for the treatment of substance abuse disorders. Pharmacotherapeutic treatments are now available for the treatment of opioid dependence, alcohol dependence and nicotine dependence. Detailed algorithms for the use of pharmacotherapy in addiction disorders may be found at the website of the International Psychopharmacology Algorithm Project (www.ipap.org). The medical treatment of withdrawal states that emerge upon substance discontinuation are not covered in this chapter, but have been reviewed elsewhere.(Rosenbaum, Arana et al. 2005; Taylor, Paton et al. 2007) In treating patients with substance abuse, the beneficial effects of psychosocial interventions should not be overlooked.(Dutra, Stathopoulou et al. 2008)

Medications for Opioid Dependence
Methadone, a synthetic opioid mu-receptor agonist, is a long-acting analgesic that has shown efficacy in maintenance therapy for patients with a history of opioid dependence. Although methadone (at relatively low doses) can be prescribed as an analgesic by individual physicians in the United States, methadone for the treatment of heroin dependence can only be dispensed by centers registered and authorized to do so by regulatory agencies. The methadone dose is gradually increased over many months in patients attending these centers until a dose (of usually 90-120 mg/day or higher) is reached that stops cravings for illicit opiates and stops drug seeking behaviors.(Faggiano, Vigna-Taglianti et al. 2003) Methadone can cause respiratory depression (especially in patients who are not tolerant to opioids), additive CNS effects with concurrent use of other sedatives, QT prolongation(Ehret, Voide et al. 2006) and constipation. Levomethadylacetate (LAAM) is similar to methadone except with a longer half-life requiring less than daily administration, but it has been discontinued due to reports of it causing torsades de pointes arrhythmias. Physicians should be aware that patients maintained on high dose methadone who are admitted to the medical/surgical wards of hospitals for unrelated medical care are likely to need to continue their daily dose of methadone. However, high doses should never be administered without independent confirmation with the methadone center administering this drug to confirm the actual dose that the patient has been receiving prior to admission. Even 3-4 days of methadone discontinuation may significantly reduce patients' tolerance to the respiratory effects of this drug. To decrease the risk of death from respiratory depression, a single dose of methadone should never exceed 20 mg when independent confirmation of higher doses is not possible. Subsequent dose increments can then be added as necessary and as tolerated. Buprenorphine is an opioid mu-receptor partial agonist (with very high affinity for this receptor) that is used as an alternative to methadone for maintenance therapy in opioid dependence.(Fudala, Bridge et al. 2003) In the United States it can be prescribed in an office-based setting (for example with weekly counseling and weekly dispensing)(Fiellin, Pantalon et al. 2006) without requiring daily administration in a methadone center. Buprenorphine is less dangerous than methadone in overdose with a lower risk of respiratory depression. Concurrent use of benzodiazepines or alcohol however significantly increases the risk of death from respiratory depression(Megarbane, Hreiche et al. 2006; Kintz 2001); therefore patients with a history of polysubstance abuse may not be good candidates for buprenorphine maintenance therapy. When used in outpatient treatment buprenorphine is combined with the opioid antagonist naloxone and administered sublingually. In sublingual form the buprenorphine is absorbed while the naloxone is not. When absorbed through the GI tract, naloxone undergoes extensive first-pass liver metabolism decreasing its systemic availability. Naloxone is added to discourage intravenous abuse of this medication: if this combination is misused intravenously, the naloxone effect predominates and blocks any opioid effect. Buprenorphine may also be beneficial in chronic pain patients who are at risk of opioid dependence; higher doses may be needed when buprenorphine is used as an analgesic.

Medications for Alcohol Dependence
Disulfiram, one of the earliest treatments developed for addictive disorders, acts by producing unpleasant physical effects if alcohol is concurrently consumed. It disrupts ethanol metabolism by irreversibly inhibiting aldehyde dehydrogenase, thereby leading to a significant accumulation of ethanol metabolite acetaldehyde which is associated with severely unpleasant adverse effects (and cardiac stress). Although there is no evidence that it helps maintain abstinence over the long run, it may be useful as a disincentive to ethanol use in the short term.(Suh, Pettinati et al. 2006) It retains its effect on aldehyde dehydrogenase for up to 2 weeks, so even if the patient stops taking disulfiram and plans to drink, there may be time to reconsider and enlist other supportive mechanisms to maintain sobriety before it loses effectiveness. Ultimately however, most patients who wish to drink do so by discontinuing disulfiram, and many drink while still on it, placing themselves at severe risk. Therefore, like all pharmacotherapies for ethanol dependence, external supports (such as family supervision of medication adherence) and nonpharmacological therapies (such as ongoing counseling and behavioral therapies) are needed for continued effectiveness.(Lingford-Hughes, Welch et al. 2004; Hughes and Cook, 1997) Notably, however, a recent randomized comparison of disulfiram, acamprosate and naltrexone (discussed below) in 243 patients, all of whom received brief cognitive-behavioral psychotherapy, showed disulfiram to be more advantageous than the other agents.(Laaksonen, Koski-Jannes et al. 2008) Patients who are beginning disulfiram treatment should be informed of possible medication interactions and the need for avoidance of alcohol in foods (e.g., sauces) and topical preparations (e.g., perfumes). Disulfiram is not recommended for patients with cardiac disease, significant liver disease, peripheral neuropathy or psychosis. Acamprosate may increase the number of abstinence days and decrease overall alcohol consumption long-term in alcohol dependent patients.(Boothby and Doering, 2005; Sass, Soyka et al. 1996; Whitworth, Fischer et al. 1996; Mann, Lehert et al. 2004; Kranzler and Van Kirk, 2001) Its mechanism of action is unclear although it is thought to involve the enhancement of GABA transmission and possibly the antagonism of the excitatory neurotransmitter glutamate.(Littleton and Zieglgansberger, 2003)  It is generally well-tolerated, with mild GI symptoms as the most commonly seen adverse effects. It is renally excreted and may be administered to patients with liver disease. Recent evidence from a large multicenter study, however, has shed doubt on the effectiveness of acamprosate (Anton, O'Malley et al. 2006)—see below. Naltrexone is an opioid receptor antagonist. Alcohol can increase the release of endogenous opioids in the brain which may contribute to its euphoric effects. Naltrexone may reduce this opioid-mediated aspect of alcohol's reinforcing properties, and modestly reduce alcohol use in dependent patients.(Anton, 2008; Srisurapanont and Jarusuraisin, 2005) It appears to be most beneficial in severe alcoholics.(Pettinati, O'Brien et al. 2006)  A long-acting (i.e., every 4 weeks) injectable preparation is also available.(Garbutt, Kranzler et al. 2005; O'Malley, Garbutt et al. 2007)  Naltrexone may cause mild GI symptoms and an infrequent transaminitis that requires monitoring. Patients on naltrexone must not be given opiates for pain management: overdose and death can result from the high opiate doses needed to override the effect of naltrexone. Some studies have suggested superior efficacy of naltrexone as compared to acamprosate.(Anton, O'Malley et al. 2006; Morley, Teesson et al. 2006; Rubio, Jimenez-Arriero et al. 2001) The recent U.S. government-sponsored COMBINE study which compared naltrexone vs. acamprosate vs. the combination of the two, all combined with medical management (i.e., brief meetings with a healthcare provider modeled on a primary care setting), found naltrexone to be more effective than acamprosate. It also found that the meetings with a healthcare provider increased the likelihood of abstinence.(Anton, O'Malley et al. 2006) It should be noted that the dose of naltrexone used was twice the usual maximum dose (100 mg/day vs. 50 mg/day). Individuals with a specific polymorphism of the mu-opioid receptor gene (OPRM1), i.e., individuals with an Asp40 allele—coding for a receptor with increased beta-endorphin binding and activity(Bond, LaForge et al. 1998)—may be more likely to respond to naltrexone.(Anton, Oroszi et al. 2008)

Off-label use of naltrexone may also be considered in opioid dependent patients—where higher doses are usually needed—if there are significant external supports and motivation to ensure adherence to this medication.(Kirchmayer, Davoli et al. 2003) Highly motivated addicted physicians and other professionals are examples of patients who may benefit from naltrexone treatment for opiate dependency.(Ling and Wesson, 1984; Washton, Gold et al. 1984)

Other Medications for Alcohol Dependence
There is recent evidence to support the use of the anticonvulsant topiramate for the treatment of alcohol dependence.(Johnson, Ait-Daoud et al. 2003; Johnson, Rosenthal et al. 2007) Ondansetron, an antiemetic with 5HT3 serotonin receptor antagonist activity, has also emerged as an agent with possible efficacy for this indication.(Johnson, Ait-Daoud et al. 2000; Johnson, Roache et al. 2000)  The antidepressant mirtazapine is also a 5HT3 antagonist but has not been studied for this indication.

Medications for Nicotine Dependence
Nicotine replacement therapy is used to decrease withdrawal symptoms during smoking tapering and cessation and can double the odds of quitting.(Silagy, Lancaster et al. 2004) Nicotine replacement can be delivered transdermally via a patch, or by gum, oral inhaler, nasal spray or dissolving lozenge. All modes of delivery are likely to be effective(Silagy, Lancaster et al. 2004) Actual dosing and duration of treatment vary slightly for each formulation, although all nicotine replacement treatments involve setting a target date for smoking cessation followed by a gradual taper of the nicotine over 2-3 months. In a review of 88 trials, success rates on 6-12 month follow-up averaged 16% vs. 10% on placebo.(Silagy, Mant et al. 2000) Patients should therefore be encouraged to make repeated efforts to quit. Caution should be used in patients with a history of cardiac disease, especially when using the nicotine patch (avoid the patch if there is a history of serious arrhythmias, angina or immediately post-MI). Patients should not smoke at all while wearing the transdermal nicotine patch. Nausea and headaches can occur frequently with nicotine replacement therapy. Bupropion, an antidepressant with possible dopaminergic effects (see section on antidepressants), is also efficacious for smoking cessation.(Jorenby, Leischow et al. 1999) Bupropion should be started two weeks before the target stop date and then continued for at least 3 months. The addition of nicotine replacement therapy to bupropion can increase the chances of abstinence compared to the use of either drug alone.(Jorenby, Leischow et al. 1999) Varenicline is an alpha4beta2 nicotinic acetylcholine receptor partial agonist, with high affinity for this receptor. It is the latest advance in nicotine addiction treatment. It may have effectiveness that is comparable to, or greater than that of bupropion for smoking cessation.(Gonzales, Rennard et al. 2006; Jorenby, Hays et al. 2006; Tonstad, Tonnesen et al. 2006) Although varenicline appears to be generally well-tolerated, treatment-emergent mood changes and psychosis have been reported in susceptible patients.(Freedman, 2007; Kohen and Kremen, 2007)  Because of its cost it is often restricted by pharmacy benefit managers to patients who have failed nicotine replacement and bupropion therapy. Table 6 summarizes the characteristics of medications used for substance abuse/dependence disorders.(WHO, 2007; PDR, 2008; Hyman, Arana et al. 1995; Rosenbaum, Arana et al. 2005; Stahl, 2005; Taylor, Paton et al. 2007)

'''TABLE 6. MEDICATIONS FOR SUBSTANCE ABUSE/DEPENDENCE'''


 * ® Generic and U.S. brand name(s).
 * Dosing should be adjusted downwards ("start low, go slow" strategy) for the elderly and/or the medically compromised.
 * Abbreviations: bid-(bis in die) twice a day; CYP-Cytochrome P450 enzyme; FDA-Food and Drug Administration; GABA-Gamma-Aminobutyric Acid; IM-intramuscular; LFT-Liver Function Tests; MDD-Major Depressive Disorder; MDE-Major Depressive Episode; mg-milligram; po-(per os) orally; q-(quaque) every; TD-transdermally; tid-(ter in die) three times a day; WHO-World Health Organization.

Conclusion
Over the last five decades, multiple medications have become available for the treatment of patients with psychiatric disorders. TCAs, MAOIs, SSRIs, SNRIs and others antidepressants have expanded current treatment options for depressive and anxiety disorders. Anxiolytics, including benzodiazepines and non-dependence-producing alternatives, are available for the treatment of severe anxiety disorders. First and second generation antipsychotics with different receptor profiles and side effect profiles have expanded the choices for patients with psychotic disorders. Lithium and other medications with mood stabilizing properties are available for use in patients with bipolar disorder. New formulations of stimulants and non-stimulant agents can be used in adults with attention-deficit/hyperactivity disorder. Finally, pharmacological therapies for the treatment of substance abuse and dependence disorders have been greatly expanded in recent years. Medical students and physicians should become familiar with these medications and obtain some facility in using them. As always, the science and art of medicine comprise the ability to appropriately and carefully apply that which is learned in textbooks to a specific patient. In the clinical setting, pharmacotherapeutic treatments should be used judiciously: the risks and benefits of treatments should be considered so that every effort is made to "first do no harm." One should employ the strategy of using one medication at a time, so as to have the opportunity to know what is actually working and not working. The goal is to provide relief and lessen suffering, preferably in the most evidence-based and cost-effective manner possible. Finally, students and clinicians should keep in mind that for pharmacotherapeutic interventions to be successful there must also be appropriate psychosocial support and treatment. Only then can safe, effective, and comprehensive treatment be provided.

Recommended Textbooks
Ansari A, Osser DN, Lai LS, Schoenfeld PM, Potts KC: Pharmacological approach to the psychiatric inpatient, in Ovsiew F, Munich RL (eds), Principles of Inpatient Psychiatry. Philadelphia, PA: Lippincott Williams & Wilkins, 2009

Nestler EJ, Hyman SE, Malenka RC: Molecular Neuropharmacology: A Foundation for Clinical Neuroscience, Second Edition. New York: McGraw-Hill Companies, Inc., 2009

Taylor D, Paton C, Kerwin R: The South London and Maudsley NHS Foundation Trust Oxleas NHS Foundation Trust Prescribing Guidelines, 9th Edition. London: Informa Healthcare, Telephone House, 2007

Schatzberg AF, Cole JO, DeBattista C: Manual of Clinical Psychopharmacology, Sixth Edition. Washington, DC: American Psychiatric Publishing, Inc., 2007

Janicak PG, Davis JM, et al.: Principles and Practice of Psychopharmacotherapy. Philadelphia, PA: Lippincott Williams & Wilkins, 2006

Rosenbaum JF, Arana GW, et al.: Handbook of Psychiatric Drug Therapy, Fifth Edition. Philadelphia, PA: Lippincott Williams & Wilkins, 2005