Chapter 9. Psychopharmacologic Interventions

Most psychiatric disorders remain without well-established biological substrates; however, standardized diagnostic nosology has gained acceptance, primarily in the form of the Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV). As with other areas of medicine, there is no substitute for a careful diagnostic evaluation using externally validated diagnostic criteria. Whenever possible, the diagnostic evaluation should draw on a comprehensive database that includes family members, previous or concurrent providers, and other sources of information, and the evaluation should focus more on longitudinal data than just acute presentation.

An understanding of the basic pharmacokinetics and pharmacodynamics of psychotropic drugs is required for their safe and effective utilization. With the exception of lithium (which is an element), most psychotropic drugs are lipophilic polycyclic amines. These drugs interact with neurotransmitter-binding sites, which confer their psychotropic effects. For example, most antidepressants act by allosteric binding to catecholamine- or indolamine-uptake binding sites (or with the monoaminergic catalytic enzyme monoamine oxidase), which enhances the synaptic availability of monoamines such as norepinephrine and serotonin. These effects can produce direct antidepressant actions but also may produce side effects such as nausea or nervousness. These drugs also interact with muscarinic cholinergic-binding sites, producing significant side effects such as dry mouth, blurred vision, and constipation, and with histamine-binding sites, producing drowsiness and weight gain. The relative profile of receptor binding affinities will allow the clinician to predict both the beneficial effects and side effects of specific drugs.

Most psychotropic drugs are fairly rapidly and completely absorbed from oral and intramuscular sites and, because of their relatively high lipid solubility, readily cross the blood–brain barrier. Intramuscular administration yields rapid absorption and distribution and bypasses first-pass hepatic metabolism. Therefore, plasma levels are achieved much more rapidly. Intramuscular delivery of antipsychotic drugs (e.g., haloperidol, olanzapine, ziprasidone) or antianxiety agents (e.g., lorazepam) are reserved primarily for the acute management of agitated and psychotic patients. An exception to this is the use of haloperidol or fluphenazine decanoate; both drugs are provided in oil suspension for long-term treatment of psychotic conditions. They require injection every 2–4 weeks and are given to patients who have problems with treatment compliance.

Most psychotropic drugs have high levels of protein (e.g., α1 glycoproteins) binding. Notable exceptions include lithium and venlafaxine. Therefore, drug interactions could occur at the level of protein binding, in which case these drugs will displace (and be displaced by) other drugs with significant binding. This can result in unexpected toxicities; therefore, clinicians should take a careful drug history whenever they plan to prescribe a new drug to a patient.

With the exception of lithium, all psychotropic drugs are metabolized, at least in part, via cytochrome enzymes. After one or more metabolic steps, water-soluble products (e.g., glucuronides) are formed and eliminated via the kidney. Certain psychopharmacologic agents may induce or inhibit metabolism by cytochrome P-450 (CYP) enzymes. For example, car bamazepine can induce CYP 3A4, 2B1, and 2B2 enzymes, whereas certain serotonin selective reuptake inhibitor (SSRI) antidepressants may inhibit metabolism via CYP 2D6. These metabolic interactions must be considered when drugs are coadministered.

Most psychotropic drugs produce widely varying plasma levels, and monitoring may be helpful. However, plasma-level ranges are established only with certain tricyclic antidepressants (e.g., imipramine, desipramine, nortriptyline), with antipsychotics such as haloperidol and clozapine, and with lithium. Because of the narrow therapeutic index, plasma-level monitoring of lithium is part of accepted practice. With other drugs, where there are no established plasma level-response relationships, plasma-level monitoring can be used to check compliance or when severe side effects suggest abnormally increased levels. In addition, plasma levels may clarify a situation in which one drug may be increasing or decreasing the level of another. For example, an SSRI such as paroxetine may elevate plasma levels of other drugs that are metabolized by CYP 2D6 (e.g., haloperidol). Unexpected side effects could occur that would be clarified by a plasma-level measurement.

Pharmacokinetics can differ by population. For example, children have a higher relative percentage of hepatic mass and a greater-than-expected elimination of drugs by first-pass metabolism. The elderly often have relatively reduced hepatic clearance and protein binding, which increases the relative plasma levels of most psychotropic drugs. Dosage adjustments must be made for differences in kinetics, and these adjustments should be based on research data, when available.

Pregnancy poses its own set of problems. Exposure of women of reproductive potential to psychotropic drugs raises the possibility of birth defects or spontaneous abortion should pregnancy occur. All women of reproductive potential should be counseled about pregnancy and drug exposure (i.e., to avoid pregnancy or to inform the physician before becoming pregnant). Lithium, anticonvulsant mood-stabilizing agents (e.g., carbamazepine or divalproex), and benzodiazepines have been associated with increased rates of specific birth defects and, generally, should be avoided, especially in the first trimester. The effects of antidepressants and antipsychotics are less clear, except for the case of paroxetine. Drugs should be withdrawn if possible; however, drug withdrawal is sometimes impossible. A risk–benefit analysis should be undertaken in which the potential risks of drug exposure are weighed against the hazards of no-drug therapy. Because the first trimester is the most important period of organogenesis, efforts should be made to avoid drug exposure during this period. Another consideration is that drugs may be withdrawn prior to delivery and reinstated afterward. This will help prevent untoward reactions such as excessive sedation, withdrawal, or discontinuation reactions in the newborn.

Nursing mothers generally can take psychotropics. Most drugs are excreted into breast milk, but the levels tend to be very low. The absolute short-term and long-term risks are unknown. An appropriate risk–benefit discussion should be undertaken with the patient.

Clinical practice involves establishing and maintaining an effective working relationship with a patient—the therapeutic alliance. This process requires the therapist and the patient to establish mutually acceptable goals and to agree on a plan to achieve these objectives. This is especially important in psychiatry because the beneficial effect of a drug may be delayed and because some patients may not have insight into their problem. The challenge of psychopharmacologic management is to maintain an effective working relationship with patients in the face of such obstacles.

The effective pharmacotherapist will maintain a broad biopsychosocial view of the patient's problem and will plan accordingly. Medications will help to address identified symptoms, but patients exhibit a wide array of problems, including psychosocial stressors (e.g., family and work issues) and more purely psychological factors that put the patient at risk for further problems. Pharmacotherapy, therefore, shares significant elements with psychotherapy. The pharmacotherapist must establish rapport and a working relationship with the patient, based on the expertise of the therapist and on mutual trust. The patient will depend on the therapist's ability to perform a risk–benefit analysis and to provide an adequate rationale for all elements of treatment. A therapeutic trial of medication may then be undertaken; however, this process of education and negotiation with the patient is an ongoing part of treatment.

Analyzing Risks & Benefits

The risk–benefit analysis considers many factors. The most basic of these factors is the presence of any medical or psychiatric contraindications to the use of a particular drug, including the potential for drug interactions. For example, although bupropion is an effective antidepressant, it generally is not given to patients who have a history of seizure disorder or a current psychotic condition because the drug has the potential to aggravate these problems. However, bupropion may be an especially good choice for patients who have erectile dysfunction. Alternatively, the pharmacotherapist may determine that medications are not required at all, preferring a course of individual or family therapy. The risk–benefit analysis involves the process of maximizing the potential benefit while minimizing the possible harm of pharmacotherapy.

Managing Nonresponse

Lack of complete response to a given treatment is common. The management of nonresponse should begin before treatment is initiated. The clinician should inform the patient of the possibility of nonresponse and the steps that might be taken in such a circumstance. Patients must be warned of the possible delay in response and of the importance of continuing compliance in face of incomplete improvement. If the response to treatment is inadequate, the clinician should review the treatment plan with the patient and explain to him or her the rationale for each next step. This review should always include the instillation of realistic hope. Suicidal potential must be evaluated as well.

Managing Side Effects

The review, documentation, and management of side effects is an important part of pharmacologic treatment. The clinician should warn patients of the possibility of side effects and should review with them those that are common. A general rule of thumb is that any side effect that occurs in 5% or more of patients in clinical trials should be discussed. However, the clinician should address any serious adverse experience, even those that might be rare, and should warn the patient that other unusual effects may occur. For example, although hypothyroidism is an uncommon outcome of lithium treatment, patients should be warned of this possibility. The clinician should review and document side effects with each patient and undertake a plan to manage unacceptable effects. Most side effects will improve with time, but noncompliance may be the price of ongoing serious problems. Each patient has his or her own tolerance for side effects, and the patient's tolerance must be considered. For example, sexual side effects of drugs may be completely unimportant for some patients (especially those with no current sexual partner) but critical for others. Further, sexual side effects may be unimportant in one phase of treatment and monumental in others (e.g., when a new relationship is established).

Laboratory evaluations may be required to determine if certain serious reactions have occurred (e.g., hepatotoxicity or nephrotoxicity). Certain drugs require regular plasma-level monitoring to provide the patient with a margin of safety. Biweekly white blood counts are an absolute requirement for clozapine therapy, and annual or biannual thyroid-stimulating hormone and creatinine tests may be performed for patients on lithium. Proper medication management requires regularly scheduled laboratory evaluations for many patients.

Addressing Concurrent Alcohol or Drug Use

Another safety issue is the concurrent use of alcohol or drugs of abuse. The clinician must take a careful drug and alcohol history; however, ongoing evaluation of the patient's drug and alcohol use also is required. In most cases, very modest alcohol use is allowed while a patient is taking psychotropic drugs; however, patients must be warned of potential drug interactions with alcohol with a potentiation of the sedative effects of both the prescribed medication and the alcohol. Alcohol should not be used concomitantly with sedative–hypnotic agents such as benzodiazepines.

Evaluating Response

The clinician should target specific symptoms as indicators of response to treatment. In the case of psychosis, the presence and severity of hallucinations, delusions, or agitation can be effective indicators of early response; however, a more comprehensive view of the patient's condition is required for longer-term management. This involves an ongoing evaluation of a variety of symptomatic domains, including cognitive and mood symptoms and social impairments. As a result, symptoms such as occupational impairment may be important indicators of the effectiveness of pharmacotherapy. Whatever the condition, the goal of pharmacotherapy is to minimize or eliminate symptoms and to return the patient to his or her maximal level of functioning. To this end, the clinician should review and document in each patient the change in a broad range of symptoms.

Psychopharmacotherapy ultimately represents a human endeavor in which the challenge to the clinician is the competent synthesis of the scientific basis of pharmacology with the psychosocial skills of the psychiatrist. The most successful pharmacotherapists have a thorough knowledge about the nature of psychiatric disorders, the mechanisms of drug action, and the applications of drug therapies, but they also have remarkable interpersonal skills in the management of patients. This integrative skill set in many ways defines competent contemporary psychiatry.