Psychopharmacology in the Medical Setting

Initiating Treatment

The appropriate use of psychotropic medications starts with as precise a formulation of the diagnosis as possible. The use of an antidepressant alone for a depressed college student presenting to an Emergency Department (ED) might be appropriate if the diagnosis is of a major depressive episode; less pertinent if the diagnosis is of an adjustment disorder; and seriously inadequate and quite possibly harmful if the diagnosis is of a bipolar disorder or a substance use disorder. As a rule, the “ready, fire … aim” approach should be avoided and it is best to defer pharmacologic treatment until a good working diagnosis can be reached. The establishment of a psychiatric diagnosis, however, often requires longitudinal assessment of course and treatment response. Many symptoms of psychiatric disorders may be obscured by co-occurring medical conditions, substance use, or inaccurate patient reports. In acute clinical situations, however, it is often not possible to defer the implementation of psychotropic medications until a diagnosis is fully clarified. In this context, it is crucial to document probable and differential diagnosis, to outline the rationale for selecting a particular treatment over others, and to indicate the kind of information needed to achieve greater diagnostic certainty. When a disorder appears in a sub-syndromal form, such as minor depression, the rationale and goals for proceeding with psychopharmacologic treatment should be well defined.

The identification of target symptoms plays an integral role in establishing a pre-treatment baseline and in later efforts to monitor the success of treatment. These symptoms might include aggression, insomnia, anhedonia, delusions of reference, hallucinations, or the frequency and intensity of suicidal longings. The identification of target symptoms—particularly clusters of target symptoms—serves to focus attention on the symptoms revealing greatest danger, disability, and distress to the patient while also informing the patient about the core symptoms of his or her illness and the specific goals for which psychotropic medications have been recommended.

For some conditions, clinician-rated instruments, such as the Brief Psychiatric Rating Scale, the Hamilton Rating Scale for Depression, the Hamilton Rating Scale for Anxiety, and the Young Mania Scale, provide useful, well-studied templates for the serial assessment of relevant symptoms. In addition, the patient and family or other caregivers can be recruited in formal efforts to monitor progress. In the case of episodic or complex presentations, the use of daily mood charts, patient-rating scales such as the Quick Inventory for Depressive Symptomatology, or sleep logs reveal temporal patterns (e.g., rapid mood cycling) and associations (e.g., to menstrual cycle or medication changes) that are not apparent cross-sectionally during office or bedside visits.

Along with the assessment of target symptoms, evaluation of current levels of function and subsequent changes with treatment relevant to quality of life are an integral part of good psychopharmacologic practice. Thus, for an outpatient, the clinician might query about improvement in work or school function, family and other social relationships, and use of leisure time. For an inpatient, progress in the level of independence and reduction in the overall degree of anguish can help confirm the adequacy of treatment, because lack of improvement along these dimensions directs attention to residual symptoms or to problems not initially apparent.

Since the last edition of this handbook, the Global Assessment of Function scale and its single composite score that pertains to quality of life has fallen out of favor. In its place, some clinicians use the Clinical Global Impression (CGI) scale. This scale, widely used in clinical trials, rates severity of disease on a scale of 1 to 7 (from normal to most extreme) and improvement on a scale of 1 to 7 (from very much improved to very much worse) and provides a simple quantitative means to document overall treatment outcome. The Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5) Disability Study Group recommends that practitioners use the World Health Organization Disability Assessment Schedule (WHODAS 2.0) to measure disability for routine clinical use. The WHODAS 2.0 is applicable to patients with any health condition. WHODAS 2.0 evaluates the patient's ability to perform activities in six domains of functioning (e.g., self-care) and uses these to calculate a score representing global disability.

The understanding that a patient's psychiatric condition is influenced by psychosocial factors does not imply that psychotropic medications should be withheld. A major depression evolving in the setting of a spouse's chronic illness or panic attacks emerging in the weeks following the break-up of a significant relationship may well be as severe and as responsive to pharmacologic treatment as the same disorders that developed in other patients without similar precipitants. Vague referrals to counseling do not constitute sufficient treatment under such circumstances and may be viewed by patients as dismissive. If psychotherapy is recommended and pharmacotherapy deferred, the referral to psychotherapy—whether, for example, to individual cognitive therapy, to couples therapy, or to a pain management group (i.e., behavioral treatment)—must be viewed with the same deliberateness as the prescription of medication, and it should include a plan for follow-up. If substantial progress is not made within a clinically appropriate time frame (e.g., no more than 8 to 12 weeks for a moderately severe episode of major depression), then the adequacy of the psychotherapy should be re-evaluated. Severity, chronicity, and risk of recurrence of symptoms are often more relevant in determining the need for pharmacotherapy of psychiatric conditions than the presence of aggravating life circumstances or a caricatured description of illness as chemical or reactive.

Reciprocally, the expanding range of safe and well-tolerated psychotropic agents, such as the selective serotonin re-uptake inhibitors (SSRIs) and other newer antidepressants, does not alter the imperative to explore the use of non-medication interventions whenever medications are considered. An assessment of a patient for psychiatric medications should include an equally careful evaluation for other targeted interventions instead of, or in addition to, pharmacotherapy or psychotherapy. Often uniquely helpful are judicious referrals to parenting classes; elder care; Alcoholics Anonymous, Narcotics Anonymous, and Al-Anon; vocational assessment and rehabilitation; support groups for persons who are bereaved, who are going through separation, and who have medical disorders (e.g., epilepsy, human immunodeficiency virus [HIV] infection); and psychiatric self-help groups, such as those sponsored by the Depression and Bipolar Support Alliance (DBSA).

Patient education and informed consent are important legal and ethical imperatives that are also critical to the success of a course of treatment with psychotropic medications. If the capacity of the patient to make his or her own decisions fluctuates or is questionable, the clinician should obtain the patient's permission to include family or other patient-appointed persons in important treatment decisions. When a patient clearly lacks the capacity to make such decisions, formal legal mechanisms for substituted judgment should be used. Such mechanisms, however, in no way diminish the importance of educating a patient about medications and target symptoms to the fullest extent possible. When one presents recommendations to a patient about medications, information about diagnosis, target symptoms, treatment options, and anticipated means of follow-up should be included, as well as the medication's name, class, and dosing instructions. Side effects that are common (e.g., dry mouth, nausea, tremor, drowsiness, sexual dysfunction, weight gain) should be reviewed together with side effects that are uncommon but require immediate attention, such as a dystonia on an antipsychotic, painful and prolonged erection on trazodone, or a rapidly progressing rash on lamotrigine. Patients should be specifically cautioned about the risks of abrupt discontinuation of a psychotropic drug. Dietary and drug restrictions must be clearly described and, particularly in the case of preventing a hypertensive crisis with monoamine oxidase inhibitors (MAOIs), should be provided in written form as well.

Prescribers have a duty to disclose to patients the information necessary for them to make informed decisions about treatment recommendations. Some prescribers have mixed feelings about the disclosure of side effect information (particularly about serious adverse reactions) to patients. They argue that providing such information may result in unnecessary anxiety and perhaps non-adherence to the medication. The consequence of such an approach carries a violation to patient's civil rights, and bears serious liability and responsibility if an adverse effect occurs and the patient has not been informed. Downplaying potential adverse events inherent in a particular medicine to mislead patients to give consent, may set the ground for a malpractice lawsuit. In the context of urgent, life-threatening conditions (such as acute mania), counseling about some potential adverse effects, particularly those not anticipated in the foreseeable future (e.g., tardive dyskinesia or perinatal risks), can be deferred until greater clinical stability is achieved and the risks and benefits of longer-term treatment can be meaningfully addressed.

All too often omitted in discussions preceding the initiation of medications, are clear information about the anticipated time course of response (whether to anticipate improvement in hours or weeks), the anticipated length of treatment, and the ready availability of strategies to address side effects, or lack of efficacy. Psychopharmacology decisions possess great meaning to patients and the goal of treatment is not simply to recommend or prescribe “the right” medicine for a given diagnosis. Patients may view medicine as a magic cure, a method of mind control, a poisonous substance, or a gift (especially if the prescriber gives it as a sample). Exploring a patient's reluctance to initiate treatment might elicit a variety of concerns, including the fear that medication will be stigmatizing, might engender physical or psychological dependence, might be “mind-altering” or personality transforming, might mask a problem rather than treat it, might imply consignment to life-long treatment, or reflect a narrow therapeutic philosophy. The faithfulness of a patient to a recommended course of treatment is invariably strengthened by a physician's dedicated efforts to elicit and address misgivings and potential misunderstandings at the outset. Referral to relevant websites and written materials on diagnosis and treatment are usually welcomed by patients and family members as a source of more detailed information, particularly when longer-term treatment is anticipated.

Selecting and Administering Medication

For many common psychiatric disorders, there exist at least several agents within a single class that are known to have roughly equivalent efficacy. Decisions regarding choice of a particular medication for a patient should give considerable weight to previous treatment response and the current feasibility of the medication in terms of cost, tolerability, and complexity of dosing. Anticholinergic, hypotensive, and sedative effects of drugs must be considered carefully, particularly when prescribing medications to elderly patients or patients who are medically frail.

Knowledge of a patient's genetic background may help clinicians provide a personalized medicine strategy by predicting both drug response and risk for adverse events. Pharmacogenetics is becoming more impactful to clinical practice and promises improved accuracy when selecting a medicine for a specific individual. Within psychiatry, studies have found genetic variations associated with altered treatment response/efficacy and increased side effect risk. Genetic testing for such variations may help identify which patients are more or less likely to respond to psychotropics and which are likely to experience an increased side effect burden. Large randomized controlled trials are needed to further substantiate the utility of genetic testing in psychiatry.

Once a drug is chosen, the goal should be to achieve a full trial with adequate dosages and an adequate duration of treatment. Inadequate dosing and duration count are among the principal factors in treatment failure for patients with accurate diagnoses. In the service of decisions regarding a patient's care weeks or months later, it is crucial to document whether a trial of medication succeeded, failed, or was abbreviated because of clinical deterioration, medication side effects, poor adherence to treatment, or drug abuse. Suffering is unnecessarily prolonged and resources are poorly used when medications that were previously ineffective have been tried again because of inadequate documentation of failure or when medications that could have been effective are avoided because previous trials of those medications had not been flagged as having been incomplete.

Medication dosages should be adjusted to determine the lowest effective dosage and the simplest regimen. There is significant variability among individual patients with respect to response, blood levels, the expression of side effects, and the development of toxicity, such that the recommended dosage ranges provide only a general guide. Documentation of a patient's response to a particular dosage becomes a meaningful reference point for future treatment. As a rule, elderly patients should be started on lower dosages than younger patients, and the interval between dosage changes should be longer because the time to achieve steady-state levels is often prolonged. In the elderly, often there is also prolonged storage of medication and active metabolites in body tissues. Nevertheless, the goal of reaching an effective dosage must be pursued with equal determination in the elderly as in younger patients.

For patients with chronic psychiatric conditions, exacerbation of symptoms might prompt increases in the dosages of medications or addition of other medications. So too, for patients presenting acutely with severe disorders, medication dosages may be titrated up more rapidly than usual or combined with other psychotropic medications at an early point such that the lowest effective dosage and simplest regimen is likely to be unclear. Under these circumstances, re-evaluation for cautious reduction of dosage when an appropriate interval of stability has followed should be routine. When a patient's care is likely to be transferred to another clinician or another setting, such as a chronic care facility or a community health center, it is essential that such a plan be communicated to the accepting clinical staff to avoid committing a patient to long-term treatment with dosages or regimens that are excessive.

The attentive management of side effects plays an important role in developing a therapeutic alliance and improving the quality of life for a patient who may be on psychopharmacologic treatment for months or years. Although some adverse events require immediate discontinuation of the drug (e.g., serotonin syndrome or neuroleptic malignant syndrome [NMS]), most can be addressed initially with a dosage reduction, modification in the timing or by dividing doses, taking the medication with or without food, a change in the preparation of medication (e.g., from valproic acid to divalproex sodium), or guidance about sleep hygiene, exercise, or diet (e.g., caffeine, fluids, or fiber). When such measures prove unhelpful in addressing a side effect that is causing distress or that poses a safety risk, other measures must be considered, such as prescribing benztropine for extrapyramidal symptoms (EPS) on high-potency antipsychotic medicine or bupropion for sexual dysfunction on SSRIs, or replacing the offending medication with a more tolerable agent. For side effects that are likely to be transient and not dangerous, a patient's understanding that a variety of straightforward strategies are available in the case of persistence or worsening may be enough to help the patient endure the side effects until they subside.

Sometimes the best recommendation for a psychopharmacologist to offer is to “stay the course” and to counsel against making a medicine change. Though difficult in practice, it is best to avoid responding to short-term crises with long-term changes in medication. The decision to discontinue a successful antidepressant and substitute another in the setting of despair and insomnia following a traumatic event, offers the patient the prospect of benefit from the new medication weeks hence, while currently depriving the patient of active treatment known to have been effective. Although it may seem tempting to respond proportionally to a patient's marked distress with a fundamental change in established treatment, exacerbations that are thought likely to be transient are most reasonably addressed with interventions that are short term and focused coupled with adequate follow-up.

Approach to Treatment Failure

Lack of improvement, clinical worsening, or the emergence of unexpected symptoms require a concerted re-evaluation of diagnosis, dosage, drugs, and disruptions. Some patients require higher-than-usual doses of medicine or augmentation strategies involving combined-medication regimens.

Discontinuing Medications

The discontinuation of psychotropic drugs must be carried out with as much care as their initiation. For patients on complicated regimens of psychotropic medications, periodic review for dosage reduction and potential discontinuation must be standard. Because data providing guidelines for drug discontinuation are scarce, the process is often empirical. Successful discontinuation, therefore, relies heavily on a good knowledge of a patient's history, together with adequate follow-up.

Assessment of a patient for discontinuing a drug involves appreciating the short-term risks of rebound and withdrawal as well as the long-term risks of relapse and recurrence. Rebound effects are the transient return of symptoms for which a medication has been prescribed (e.g., insomnia or anxiety), and withdrawal effects are the development of new symptoms characteristic of abrupt cessation of the medication, such as muscle spasms, delirium, or seizures following discontinuation of high-dosage benzodiazepine; hot flashes, nausea, unusual shock-like sensations, or malaise following discontinuation of an antidepressant.

To make sound decisions regarding the re-instatement of medications, it is essential to distinguish rebound and withdrawal effects from relapse. Relapse is typically a persistent rather than self-limited state associated with a more delayed onset, and the re-emergence of clinically significant symptoms of the underlying illness in the absence of (or sometimes despite the continuation of) active treatment. The return of daily panic attacks after a remission of several months and an exacerbation of psychosis requiring hospitalization after 2 years of exclusively outpatient treatment are examples of relapse.

For disorders that can occur episodically, such as major depression, the term relapse refers more precisely to the recrudescence of symptoms during an initial period of remission, whereas the additional term recurrence refers to return of symptoms following a defined period of full remission (at least 4 to 6 months) on or off continued treatment. In the case of recurrence, the re-appearing symptoms are conceptualized as denoting a new episode rather than a continuation of the one previously treated.

In parallel with the concepts of relapse and recurrence, continuation of treatment refers to the ongoing use of medication prescribed to consolidate a remission of symptoms brought about by an initial (acute) phase of treatment to prevent relapse. Maintenance treatment refers to a more extended course of medication thereafter aimed at preventing recurrences and is reserved for patients with an illness characterized by chronicity, past recurrences, or particular severity. For major depression, acute treatment is typically in the range of 6 to 12 weeks, whereas continuation treatment extends 4 to 6 months beyond that point, and maintenance treatment may extend a further 1 to 5 years or more depending on the clinical context. Although antidepressants appear to be more effective than placebo during long-term treatment, the number of controlled antidepressant trials focusing on treatment of depression beyond the first year remains limited.

A taper of medications over 48 to 72 hours is typically adequate to minimize the risk of rebound or withdrawal. With respect to relapse or recurrence, however, patients at risk may well benefit from a more protracted, carefully monitored taper of medications. This allows rapid re-instatement of full-dosage treatment at the early signs of worsening to avert a more serious escalation. Analyses of discontinuation of lithium and antipsychotic agents suggest that a too-rapid cessation can, in fact, increase the risk of relapse when compared with a more gradual taper. Findings such as these suggest that for elective discontinuation of psychotropic medications, a taper lasting at least 2 to 4 weeks should be considered.

With patients for whom the consequences of relapse are likely to be severe (e.g., most patients with bipolar and psychotic disorders), an extended taper with dosage reductions of no more than 25% at intervals of no less than 4 to 6 weeks is likely to be a more prudent course. For patients who have anxiety disorders and are maintained on high-potency benzodiazepines, the introduction of a targeted course of therapy (e.g., a cognitive-behavioral panic disorder group) in preparation for a drug taper is likely to further reduce the risks of relapse.

Far from being an afterthought, decisions regarding the timing and pace of drug discontinuation should be regarded as an integral part of psychopharmacologic management and remain an important topic for further study.

Absorption

Factors that influence drug absorption are generally of less importance in determining the pharmacokinetic properties of psychiatric medications than factors influencing subsequent drug disposition (e.g., drug metabolism). The term absorption refers to processes that generally pertain to orally (rather than parenterally) administered drugs, for which alterations in gastrointestinal (GI) drug absorption can affect the rate (time to reach maximum concentration) or the extent of absorption, or both. The extent or completeness of absorption, also known as the fractional absorption , is measured as the area under the curve (AUC) when plasma concentration is plotted against time. The bioavailability of an oral dose of drug refers, in turn, to the fractional absorption for orally compared with intravenously (IV) administered drug. If an agent is reported to have a 90% bioavailability (e.g., lorazepam), this indicates that the extent of absorption of an orally administered dose is nearly that of an IV-administered dose, although the rate of absorption may well be slower for the oral dose.

Because the upper part of the small intestine is the primary site of drug absorption through passive membrane diffusion and filtration and both passive and active transport processes, factors that speed gastric emptying (e.g., metoclopramide) or diminish intestinal motility (e.g., opioids or cannabis) can facilitate greater contact with, and absorption from, the mucosal surface into the systemic circulation, potentially increasing plasma drug concentrations. Conversely, bulk laxatives, such as psyllium, magnesium-based antacids, lactulose, kaolin-pectin, and cholestyramine, can bind to drugs, forming complexes that pass unabsorbed through the GI lumen.

Changes in gastric pH associated with food or other drugs alter the non-polar, un-ionized fraction of drug available for absorption. In the case of drugs that are very weak acids or bases, however, the extent of ionization is relatively invariant under physiologic conditions. Properties of the preparation administered (e.g., tablet, capsule, or liquid) can also influence the rate or extent of absorption and, for an increasing number of medications (e.g., bupropion, lithium, most stimulant medicines, quetiapine, and venlafaxine—to name a few), preparations intended for slow release are available.

The local action of enzymes in the GI tract (e.g., monoamine oxidase [MAO]; cytochrome P450, CYP3A4) may be responsible for metabolism of drug before absorption. This is of critical relevance to the emergence of hypertensive crises that occur when excessive quantities of the dietary pressor tyramine are systemically absorbed in the setting of irreversible inhibition of the MAO isoenzymes for which tyramine is a substrate.

Following gut absorption, but before entry into the systemic circulation, many psychotropic drugs are subject to first-pass liver metabolism. Therefore, conditions that affect hepatic metabolism of drug (e.g., primary liver disease) or conditions that impede portal circulation (e.g., congestive heart failure) are likely to increase the fraction of drug available for distribution for the majority of psychotropic drugs, thereby contributing to clinically significant increases in plasma levels of drug.

There has been increasing focus on the role of the drug transporter P-glycoprotein (Pgp) in drug absorption. While the tissue distribution of Pgp influences the effect of psychotropics and the interaction potential for drugs such as risperidone, nortriptyline, and citalopram at the interface between the blood and central nervous system (CNS), Pgp is also found in other areas of the body such as the intestines, which are a major site for drug absorption into the body. The Pgps found in the gut have not been as extensively studied; however, it is well known that the expression of Pgp in other tissues can be induced and inhibited by other drugs. It is thought that some interactions, mainly seen with the antiepileptic drugs (AEDs), previously assumed to be a result of CYP450 alterations, instead may actually be mediated by the modulation of the Pgp activity at the point of drug absorption or distribution. The capacity of St. John's wort to lower blood levels of several critical medications (e.g., cyclosporine, indinavir) is hypothesized to be related to an effect of the botanical agent on this transport system.

Distribution

Drugs distribute to tissues through the systemic circulation. The amount of drug ultimately reaching receptor sites in tissues is determined by a variety of factors, including the concentration of free (unbound) drug in plasma, regional blood flow, and physiochemical properties of drug (e.g., lipophilicity or structural characteristics). For entrance into the CNS, penetration across the blood–brain barrier is required. Fat-soluble drugs (e.g., benzodiazepines, antipsychotics, cyclic antidepressants) distribute more widely in the body than water-soluble drugs (e.g., lithium), which distribute through a smaller volume of distribution. Changes with age, typically including an increase in the ratio of body fat to lean body mass, therefore, result in a net greater volume of lipophilic drug distribution and potentially greater accumulation of drug in adipose tissue in older than in younger patients. A similar potential exists for female compared with male patients because of their generally higher ratio of adipose tissue to lean body mass.

In general, psychotropic drugs have relatively high affinities for plasma proteins (some to albumin but others, such as antidepressants, to α ₁ -acid glycoproteins and lipoproteins). Most psychotropic drugs are more than 80% protein-bound. A drug is considered highly protein-bound if more than 90% exists in bound form in plasma. Fluoxetine, aripiprazole, and diazepam are examples of the many psychotropic drugs that are highly protein-bound. In contrast, venlafaxine, lithium, topiramate, zonisamide, gabapentin, pregabalin, levomilnacipran, and memantine are examples of drugs with minimal protein binding and therefore minimal risk of participating in drug–drug interactions related to protein binding.

A reversible equilibrium exists between bound and unbound drug. Only the unbound fraction exerts pharmacologic effects. Competition by two or more drugs for protein-binding sites often results in displacement of a previously bound drug, which, in the free state, becomes pharmacologically active. Similarly, reduced concentrations of plasma proteins in a severely malnourished patient or a patient with a disease that is associated with markedly lowered serum proteins (e.g., liver disease, the nephrotic syndrome) may be associated with an increase in the fraction of unbound drug potentially available for activity at relevant receptor sites. Under most circumstances, the net changes in plasma concentration of active drug are, in fact, quite small because the unbound drug is available for redistribution to other tissues and for metabolism and excretion, thereby offsetting the initial rise in plasma levels. It is important to be aware, however, that clinically significant consequences can develop when protein-binding interactions alter the unbound fraction of previously highly protein-bound drugs that have a low therapeutic index (e.g., warfarin). For these drugs, relatively small variations in plasma level may be associated with serious untoward effects.

Metabolism

Most drugs undergo several types of biotransformation, and many psychotropic drug interactions of clinical significance are based on interference with this process. Metabolism refers to the biotransformation of a drug to another form, a process that is usually enzyme-mediated and results in a metabolite that might or might not be pharmacologically active and might or might not be subject to further biotransformations before eventual excretion. A growing understanding of hepatic enzymes, and especially the rapidly emerging characterization of the CYP isoenzymes and other enzyme systems including the uridine-diphosphate glucuronosyltransferases (UGTs) and flavin-containing mono-oxygenases (FMOs), has significantly advanced a rational understanding and prediction of drug interactions and individual variation in drug responses.

Phase I reactions include oxidation (e.g., hydroxylation, dealkylation), reduction (e.g., nitro reduction), and hydrolysis, metabolic reactions typically resulting in intermediate metabolites that are then subject to phase II reactions, including conjugation (e.g., glucuronide, sulfate) and acetylation. Phase II reactions typically yield highly polar, water-soluble metabolites suitable for renal excretion. Most psychotropic drugs undergo both phase I and phase II metabolic reactions. Notable exceptions are lithium and gabapentin, which are not subject to hepatic metabolism, and a subset of the benzodiazepines (lorazepam, oxazepam, temazepam), which undergo only phase II reactions and are therefore especially appropriate when benzodiazepines are used in the context of concurrent medications, advanced age, or disease states in which alterations of hepatic metabolism is likely to be substantial.

The synthesis or activity of hepatic microsomal enzymes is affected by metabolic inhibitors and inducers, as well as distinct genetic polymorphisms (stably inherited traits). Table 38-1 lists enzyme inducers and inhibitors common in clinical settings. These should serve as red flags that beckon further scrutiny for potential drug–drug interactions when they are found on a patient's medication list. In some circumstances an inhibitor (e.g., grapefruit juice) or an inducer (e.g., a cruciferous vegetable, such as Brussels sprouts) is a drug but it may be another ingested substance.

Inhibitors impede the metabolism of a concurrently administered drug, producing a rise in its plasma level, whereas inducers enhance the metabolism of another drug, resulting in a decline in its plasma levels. Although inhibition is usually immediate, induction, which requires enhanced synthesis of the metabolic enzyme, is typically a more gradual process. A fall in plasma levels of a substrate might not be apparent for days to weeks following introduction of the inducer. This is particularly important to keep in mind when a patient's care is being transferred to another setting where clinical deterioration may be the first sign that drug levels have declined. Reciprocally, an elevation in plasma drug concentrations could reflect the previous discontinuation of an inducing factor (e.g., cigarette smoking, carbamazepine) just as it could reflect the introduction of an inhibitor (e.g., fluoxetine, valproic acid).

Although the CYP isoenzymes represent only one of the numerous enzyme systems responsible for drug metabolism, they are responsible for metabolizing, at least in part, more than 80% of all prescribed drugs. The capacity of many of the SSRIs to inhibit CYP isoenzymes fueled great interest in the pattern of interaction of psychotropic and other drugs with these enzymes in the understanding and prediction of drug–drug interactions in psychopharmacology. The CYP isoenzymes represent a family of more than 30 related heme-containing enzymes, largely located in the endoplasmic reticulum of hepatocytes (but also present elsewhere, including gut and brain), which mediate oxidative metabolism of a wide variety of drugs as well as endogenous substances, including prostaglandins, fatty acids, and steroids. The majority of antidepressant and antipsychotic drugs are metabolized by or inhibited by one or more of these isoenzymes. Table 38-2 summarizes the interactions of psychiatric and non-psychiatric drugs with a subset of isoenzymes that have been increasingly well characterized. In addition to the numerous publications in which these interactions are cited, several relevant websites are regularly updated, including www.drug-interactions.com . The relevance of these and other interactions is highlighted in a later section of this chapter, in which clinically important drug–drug interactions are reviewed.

Within the group of CYP isoenzymes, there appears to be a polymodal distribution of metabolic activity in the population with respect to certain isoenzymes (including CYP2C19 and 2D6). Most people are normal (extensive) metabolizers with respect to the activity of these isoenzymes. A smaller number are poor metabolizers, with deficient activity of the isoenzyme. Probably very much smaller numbers are ultra-rapid metabolizers, who have more than normal activity of the enzyme, and intermediate metabolizers, who fall between extensive and poor metabolizers. Persons who are poor metabolizers with respect to a particular CYP isoenzyme are expected to have higher plasma concentrations of a drug that is metabolized by that isoenzyme, thereby potentially being more sensitive to or requiring lower dosages of that drug than a patient with normal activity of that enzyme. These patients might also have higher-than-usual plasma levels of metabolites of the drug that are produced through other metabolic pathways that are not altered by the polymorphism, thereby potentially incurring pharmacologic activity or adverse effects related to these alternative metabolites.

Studies on genetic polymorphisms affecting the CYP system suggest ethnic differences. Approximately 15% to 20% of Asian Americans and African Americans appear to be poor metabolizers with respect to CYP 2C19 compared with 3% to 5% of whites. Conversely, the proportion of frankly poor metabolizers with respect to CYP 2D6 appears to be higher among white (approx. 5% to 10%) than among Asian and African Americans (approx. 1% to 3%). As our understanding of the clinical relevance of genetic polymorphisms in psychopharmacology expands, commercial genotyping tests for polymorphisms of potential relevance to drug metabolism will likely become commonplace. For the use of certain drugs, notably carbamazepine, the US Food and Drug Administration (FDA) recommends genotyping Asians for the HLA B*1502 allele owing to data implicating the allele as a marker for carbamazepine-induced Stevens–Johnson syndrome and toxic epidermal necrolysis in Han Chinese. Future study of genetic polymorphisms and their relevance to the prediction of drug response promises new ways to compensate for a gene defect (pharmacodynamic genetic variations) or to adjust medication dosage based on the rate at which a patient metabolizes medication (pharmacokinetic genetic variations).

Excretion

Because most antidepressant, anxiolytic, and antipsychotic medications are largely eliminated by hepatic metabolism, factors that affect renal excretion (glomerular filtration, tubular re-absorption, and active tubular secretion) are generally far less important to the pharmacokinetics of these drugs than to lithium, for which such factors can have clinically significant consequences. Conditions resulting in sodium deficiency (e.g., dehydration, sodium restriction, use of thiazide diuretics) are likely to result in increased proximal tubular re-absorption of lithium, resulting in increased lithium levels and potential toxicity. Lithium levels and clinical status must be monitored especially closely in the setting of vomiting, diarrhea, excessive evaporative losses, or polyuria. Factors, such as aging, that are associated with reduced renal blood flow and glomerular filtration rate (GFR) also reduce lithium excretion. For this reason, as well as for their reduced volume of distribution for lithium because of the relative loss of total body water with aging, elderly, patients typically require lower lithium dosages than younger patients, and a low starting dosage (i.e., 150 to 300 mg/day) is often prudent. Apparently separate from pharmacokinetic effects, however, elderly patients may also be more sensitive to the neurotoxic effects of lithium even at low therapeutic levels. On the other hand, factors associated with an increased GFR, particularly pregnancy, can produce an increase in lithium clearance and a fall in lithium levels.

For other medications, renal excretion can sometimes be exploited in the treatment of a drug overdose. Acidification of the urine by ascorbic acid, ammonium chloride, or methenamine mandelate increases the rate of excretion of weak bases, such as the amphetamines and phencyclidine (PCP). Therefore, such measures may be important in the emergency management of a patient with severe phencyclidine or amphetamine intoxication. Conversely, alkalinization of the urine by administration of sodium bicarbonate or acetazolamide can hasten the excretion of weak acids including long-acting barbiturates, such as phenobarbital.

Mildly to moderately impaired renal function does not typically prompt routine changes in the dosage or dosing intervals of psychotropic medications other than lithium. In patients with severe impairment of kidney function, however, there may be accumulation of metabolites and, to a lesser extent, of the parent compound across repeated doses. An increase in the dosing interval and possible reduction in drug dosage should therefore be considered in this setting, particularly in the case of chronically administered agents with active metabolites.

Renal excretion is only one contribution to the elimination half-life , a pharmacokinetic construct that refers to the time required for the plasma concentration of a drug to be reduced by one half. The elimination phase (also referred to as the β-phase) reflects all processes that contribute to drug removal, including renal excretion, hepatic metabolism, and, to a much lesser extent, other factors (e.g., loss of drug in sweat or biliary secretions) potentially affecting drug clearance (the volume of blood or plasma cleared of drug per unit time). For the majority of drugs, whose elimination follows first-order kinetics (i.e., their rate of elimination is proportional to the amount of drug in the body rather than equal to a constant amount), steady-state drug levels are reached in four to five elimination half-lives, whereas, on discontinuation, almost all drug is out of the body within five half-lives.

For drugs that are administered for their single-dose effects (e.g., an as-needed benzodiazepine or antipsychotic) rather than for long-term effects of repeated administration (e.g., antidepressants), the duration of action of the drug depends not only on the elimination half-life but also often more critically on the initial phase of drug redistribution from the systemic circulation to other tissues, such as muscle and fat.