Psychiatrists, neurologists, primary care physicians, physician assistants, psychiatric nurses, social workers and other mental health care professionals. Continuing education credit is available for most specialties. To determine if this article meets the requirements of your specialty, please contact your state licensing board.

In the absence of disease, a drug taken orally is absorbed through the mucosa of the small intestine and stomach (Due to copyright restrictions, we cannot publish this figure online. Please see our print publication for this figure -- Ed.). From there, it travels through the system of portal veins to the liver. Some proportion of the drug passes straight through the liver to the systemic circulation. For many psychotropics, a much larger proportion of the drug is metabolized in the liver and intestinal wall before entering the circulation, a process known as first-pass metabolism.

The drug or metabolite then enters the circulation. It is either distributed to target organs or storage sites, or it is cleared from the body through hepatic metabolism and/or renal excretion. Most psychotropics are too lipophilic for direct excretion by the kidney; they are metabolized in the liver into more hydrophilic compounds before being excreted. Among the current psychotropic drugs, there are four notable exceptions: lithium (Eskalith, Lithobid) and gabapentin (Neurontin), which are completely renally excreted, and amantadine (Symadine, Symmetrel) and topiramate (Topamax), which are primarily renally excreted. These drugs undergo little or no hepatic metabolism.

For drugs that are metabolized by the liver, the first phase of metabolic processing (phase I) involves oxidation, reduction or hydrolysis. Oxidation via cytochrome P450 (CYP450) pathways is by far the most common and important of these processes, particularly for psychotropic drugs. Phase I reactions usually result in loss of pharmacological activity. Phase II metabolic reactions involve conjugation, wherein metabolites are transformed into water-soluble compounds suitable for excretion. A covalent bond is formed between a functional group on the drug and an endogenous compound like glucuronic acid, sulfate, glutathione, an amino acid or acetate. Of these reactions, glucuronidation is the most common. Most conjugates are inactive and are rapidly excreted via the kidney.

From the standpoint of drug metabolism, acute hepatitis, chronic hepatitis and cirrhosis represent a spectrum of severity of liver disease. In acute hepatitis, there is usually no need to modify drug dosing, since drug metabolism is only minimally changed, and the change is transient. In chronic hepatitis, depending on the severity of hepatocyte injury, drug doses may need to be modified. In cirrhosis, hepatocytes are destroyed, and if the disease is severe, drug doses will require significant modification (Bauer, 2001).

There is no laboratory test in clinical use that measures the liver's ability to metabolize drugs. Severity of liver disease provides an indirect index, however, and this can be approximated using the Child-Pugh scoring system (Pugh et al., 1973; Riley and Bhatti, 2001). Table 1 shows the five elements constituting the Child-Pugh score: serum total bilirubin, serum albumin, prothrombin time, ascites and grade of hepatic encephalopathy. Each element is scored as 1 (normal), 2 (abnormal) or 3 (severely abnormal). A patient with normal liver function would score 5 or 6, and a patient with severe cirrhosis could score as high as 15. For a patient with a Child-Pugh score of 8 to 9 (the higher end of Child-Pugh class B), the initial dose for any drug primarily metabolized by the liver (60% or more) should be decreased by 25%. For a patient with a Child-Pugh score ≥10 (Child-Pugh class C), the initial dose should be decreased by 50% (Bauer, 2001).

Subsequent doses can be reduced either by lowering the dose itself (as with the initial dose) or by increasing the dosing interval. If a medication is available in a limited number of formulations (e.g., only orally, as a capsule), the interval will be increased. If there is more flexibility in dosing and/or route of administration, the choice will depend upon whether it is more important for that medication to maintain a constant serum level (so that the dose is reduced but the interval kept the same) or to have the usual peak and trough concentrations (in which case, the dose is kept the same but the interval is prolonged).

Psychotropic medications can themselves be a cause of liver injury. More commonly, however, medications are associated with benign elevations in liver function tests. The following guidelines may help to distinguish problematic from benign changes in the values of these tests:

First-pass metabolism. As would be expected, significant liver dysfunction adversely affects metabolism and hepatic clearance of drugs. In cirrhosis, blood may be shunted directly from the portal circulation to the systemic circulation through collateral channels, bypassing the liver and thereby reducing the extent of first-pass metabolism. Oral bioavailability of certain drugs may thus be increased significantly in patients with cirrhosis.

Some drugs passing through the liver are removed almost completely (high clearance), others only to a small extent (low clearance). High-clearance drugs are metabolized almost as fast as they can reach the liver; the rate of metabolism depends mostly on the rate of hepatic blood flow. In the nondiseased system, only a relatively small fraction of a high-clearance drug even reaches the systemic circulation. When hepatic clearance is reduced in liver disease, a larger fraction of an acutely administered, oral dose of the drug passes into the systemic circulation, and the peak plasma concentration as well as the half-life are increased. Examples of high-clearance drugs include haloperidol (Haldol), paroxetine (Paxil), sertraline (Zoloft), nefazodone (Serzone), venlafaxine (Effexor), tricyclic antidepressants, triazolam (Halcion) and midazolam (Versed). For low-clearance drugs, first-pass metabolism is relatively small. When hepatic clearance is reduced in liver disease, for these drugs only the half-life is increased. Examples of low-clearance drugs include diazepam (Valium), trazodone (Desyrel) and chlorpromazine (Thorazine) (Greenblatt et al., 1998).

Protein binding. Drugs in the circulation may travel as complexes bound to serum proteins such as albumin and α₁-acid glycoprotein. Bound drugs are largely inactive; they do not bind to target receptors, are not metabolized and are not excreted. Both hepatic and renal disease are associated with reduced serum protein levels, and reduced levels can have implications with regard to the proportion of a drug circulating in the unbound versus bound form (i.e., the free fraction).

There is a common misconception that when protein levels are reduced, there is increased drug action because of an increased concentration of the active drug at the target receptor. In fact, what is increased is the free fraction of the drug, as shown in the Figure. Since the unbound drug is equally available for metabolism, excretion and target receptor binding, it does not follow that lowered protein levels increase drug activity, because the drug does not preferentially go to target receptors. In fact, when serum protein (e.g., albumin) levels decrease, the total plasma concentration of most drugs decreases because drug clearance increases. Eventually, after equilibration, the free drug concentration (but not the free fraction) returns to its baseline level, as long as the drug dose is not changed.

What protein-binding changes do influence is the interpretation of measured drug levels. When protein levels are low and the free fraction of a drug is high (Figure), the measured concentration of the drug will underestimate the amount of drug potentially acting at the target receptor. In this situation, the patient could develop symptoms of toxicity at what appears to be a therapeutic drug level. For many psychotropic drugs (e.g., carbamazepine [Atretol, Tegretol] and valproate or divalproex sodium [Depakote]), it is possible to obtain free drug levels to circumvent this problem, but free drug levels must be specifically ordered.

Metabolism and hepatic clearance. In general, liver diseases affect phase I metabolic processes such as oxidation significantly more than they do the phase II process of glucuronidation. Most psychotropics are metabolized by phase I processes. For certain classes of psychotropics, a choice exists among drugs depending upon whether they undergo phase I or phase II processing. For example, most benzodiazepines (e.g., alprazolam [Xanax], chlordiazepoxide [Librium], diazepam and triazolam) are metabolized by phase I processes, while lorazepam (Ativan), oxazepam (Serax) and temazepam (Restoril) are metabolized by phase II processes. One of the latter drugs would be a better choice for a patient with liver disease.

Although changes in hepatic metabolism are roughly proportional to severity of liver disease, individual drug levels are unpredictable because different enzyme systems are variably affected in hepatic insufficiency. The following guidelines may be useful in prescribing psychotropic medication for a liver-impaired patient:

Table 2 displays available information regarding the use of selected psychotropics in hepatic insufficiency.

Renal insufficiency involves a functional loss of nephrons. In acute disease, such as that produced by a sudden drop in renal perfusion or medication toxicity, changes may be transient and are reversible with dialysis and supportive care. Chronic disease involves permanent loss of nephrons and, in its more severe forms, requires maintenance dialysis (Bauer, 2001).

As renal insufficiency develops, waste products such as blood urea nitrogen (BUN) and creatinine accumulate in the body. Although serum levels will reflect this accumulation, they are poorly correlated with degree of renal dysfunction because of confounding variables such as rate of creatinine production from muscle tissue. For this reason, creatinine clearance (as discussed below) is preferred as a measure of the efficiency of kidney function. A creatinine clearance value less than 80 mL/min is considered evidence of renal insufficiency.

Renal dysfunction has both pharmacodynamic and pharmacokinetic effects. Pharmacodynamic effects usually take the form of increased receptor sensitivities, as is the case with alprazolam in the renally impaired patient (Ateshkadi, 1998).

Pharmacokinetic effects may be significant, particularly for water-soluble drug metabolites and for several parent drugs in routine psychiatric use: amantadine, gabapentin, lithium and topiramate. Renal insufficiency delays clearance and may result in accumulation of these compounds, as discussed more fully below. In addition, renal insufficiency adversely affects the following aspects of oral drug disposition.

Absorption. Renal insufficiency can be associated with decreased absorption of certain drugs from the small intestine. It is thought that this might be the result of the gastric-alkalinizing effects of ammonia generated from excess urea (Anderson et al., 1976). Also, routine administration of antacid medications to renal patients may reduce gastric absorption of certain psychotropic medications (Aronoff et al., 1999).

Distribution. The volume of distribution for water-soluble or protein-bound drugs increases with edema or ascites -- conditions more prevalent among patients with renal insufficiency. Where the volume of distribution is increased, normal doses of a drug could result in unacceptably low plasma levels. At the other extreme, the volume of distribution for water-soluble or protein-bound drugs decreases with dehydration or muscle wasting. Where the volume of distribution is decreased, normal doses of a drug could result in unacceptably high plasma levels (Aronoff et al., 1999).

Protein binding. In renal failure as in hepatic failure, protein binding is reduced, and this reduction may be clinically significant for highly protein-bound drugs such as acidic drugs. As previously discussed, reduced protein binding does not affect drug activity, but does influence interpretation of measured drug levels. For many drugs used clinically, it is possible to obtain free drug levels to circumvent this problem.

First-pass metabolism. Renal dysfunction may be associated with reduced first-pass metabolism but generally has complex effects on this aspect of drug disposition. Drug bioavailability varies more among patients with abnormal renal function than among those with normal renal function (Aronoff et al., 1999).

Hepatic clearance. To some degree, renal disease may influence hepatic clearance of drugs. Renal failure may be associated with decreased oxidation via the CYP 2D6 isoenzyme because of the presence of a circulating 2D6 inhibitor (Ateshkadi, 1998). In addition, deacetylation, acetylation, hydroxylation, O-demethylation, N-demethylation, conjugation and sulfoxidation may be reduced in renal failure. Conjugation reactions, which are reversible, may actually be reversed, such that inactive metabolites are converted back to parent compounds (Ateshkadi, 1998).

Renal clearance (excretion). In renal disease, the reduction in renal excretion of a drug is a function of both the fraction of that drug removed by the kidneys and the degree of renal insufficiency. For drugs cleared primarily by renal excretion (e.g., lithium), even a mild degree of renal insufficiency can be problematic if the drugs were administered using conventional doses and schedules. The same is true for renally excreted active metabolites of certain drugs, such as hydroxy-bupropion, hydroxy-risperidone and hydroxy-venlafaxine (Jacobson et al., 2002; Pollock, 1998).

Adequate data are available to support specific dosing guidelines in renal insufficiency and failure for only a small number of renally excreted psychotropics, listed in Table 3. To use the guidelines, the patient's degree of renal insufficiency must be known. An inverse measure of the degree of renal insufficiency is provided by the glomerular filtration rate (GFR), which is approximated by the creatinine clearance (Cl_cr). The latter value is best measured using a 24-hour urine collection, but where this is not feasible, the Cl_cr may be calculated using the Cockcroft-Gault equation (Jacobson et al., 2002):

It should be noted that the Cockcroft-Gault formula has several caveats. It becomes increasingly inaccurate at the extremes of weight and serum creatinine and in the setting of an unstable creatinine. Some patients (e.g., children, burn patients or elderly patients with muscle wasting) may require the use of other formulas (Peterson and Bates, 2002). Pharmacy consultation is advised in these cases.

For psychotropics and active metabolites that are partially hepatically cleared and partially renally excreted, dosing guidelines are potentially more complicated. To simplify dosing for such drugs, the rule of two-thirds was proposed: for patients with renal insufficiency, two-thirds of the dose for patients with normal renal function is used (Brater, 1985, as cited in Levy, 1990). In addition, serum concentrations of drugs are assayed periodically, particularly for drugs with a narrow therapeutic index (Levy, 1990).

Dialysis considerations. During a dialysis session, the equilibrium in drug concentration between the systemic circulation and the periphery is disrupted, so that there may be an initial lowering of the plasma drug level, followed by a rebound rise after dialysis as the drug redistributes from the periphery to the circulation (Aronoff et al., 1999). Drugs with a narrow therapeutic index are thus best avoided in the dialysis population. In addition, dialysis patients experience significant fluid shifts and frequently become dehydrated. For these reasons, the risk of neuroleptic malignant syndrome is higher in this population. Even more common and problematic is hypotension, especially orthostasis, which is often experienced immediately post-dialysis. Psychotropic drugs with significant orthostatic side effects should be avoided if possible in dialysis patients.

Dialysis methods in current use include hemodialysis, peritoneal dialysis and continuous renal replacement therapy. Little is known from systematic study about drug removal by the latter two techniques. During hemodialysis, a drug is more likely to be removed if it is water-soluble, has a low molecular weight and has a low plasma protein binding value (Ateshkadi, 1998). Since most psychotropics are highly fat-soluble (lipophilic), they are not dialyzed in their parent form. Glucuronidates and other polar metabolites may be removed by dialysis. If a therapeutic drug is removed by dialysis, it must be supplemented post-dialysis.

Drugs such as lithium and gabapentin circulate unchanged in the body between dialysis sessions and are removed during dialysis (Levy, 1990). These drugs are then dosed at the end of a dialysis session, as noted below. Other drugs may undergo hepatic biotransformation to simple compounds like water and carbon dioxide or to more complex metabolites. Since metabolites may be active, as noted earlier, the lack of ability to excrete them between dialysis sessions may be problematic. Moreover, metabolites that are not effectively removed by dialysis could theoretically accumulate in the body to extremely high levels. For these reasons, drugs with renally excreted active metabolites that are not effectively removed by dialysis should be avoided in the dialysis patient.

Maintenance doses of lithium and gabapentin are routinely given three times weekly (tiw) at the end of each dialysis session. Dosing guidelines for various levels of renal insufficiency are shown in Table 3. Lithium is usually given at a low dose (300 mg to 600 mg) after dialysis. For lithium, when a serum level needs to be checked, this is done two to three hours after the dose. Usual target dosing guidelines are followed. On initial dosing, levels are checked after each dialysis session; subsequently, levels are checked monthly (Levy, 1990).

Careful attention to psychotropic selection, dosing and dose titration in patients with hepatic or renal impairment can help avoid adverse drug events, which are commonly seen in these populations (Peterson and Bates, 2002). To supplement the guidelines outlined above, readers may find the following references useful in clinical practice.

Dr. Jacobson is assistant professor of psychiatry at Tufts University School of Medicine and chief of the Consultation-Liaison Service for New England Medical Center. She is co-author, along with Ronald W. Pies, M.D., and David J. Greenblatt, M.D., of Handbook of Geriatric Psychopharmacology (American Psychiatric Publishing, 2002).

Anderson RJ, Gambertoglio JG, Schrier RW (1976), Clinical Use of Drugs in Renal Failure. Springfield, Ill. Thomas.

Aronoff GR, Berns JS, Brier M, Bennett WM, eds. (1999), Drug Prescribing in Renal Failure: Dosing Guidelines for Adults, 4th ed. Philadelphia: American College of Physicians.

Ateshkadi A (1998), Principles of drug therapy in renal failure. In: Primer on Kidney Diseases, 2nd ed., Greenberg A, Cheung AK, Coffman TM, eds. San Diego: Academic Press, pp298-306.

Greenblatt DJ, von Moltke LL, Shader RI (1998), Pharmacokinetics of psychotropic drugs. In: Geriatric Psychopharmacology, Nelson JC, ed. New York: Marcel Dekker, pp27-41.

Jacobson SA, Pies RW, Greenblatt DJ (2002), Handbook of Geriatric Psychopharmacology. Washington, D.C.: American Psychiatric Publishing Inc.

Levy NB (1990), Psychopharmacology in patients with renal failure. Int J Psychiatry Med 20(4):325-334.

Peterson JF, Bates DW (2002), Automated selection of drugs and drug dose in patients with renal insufficiency. Medscape Pharmacists 3(1). Available at: www.medscape.com/viewarticle/429055_print. Accessed Sept. 3.

Pollock BG (1998), Psychotropic drugs and the aging patient. Geriatrics 53(suppl 1):S20-S24.

Pugh RN, Murray-Lyon IM, Dawson JL et al. (1973), Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 60(8):646-649.

Riley TR 3rd, Bhatti AM (2001), Preventive strategies in chronic liver disease: part II. Cirrhosis. Am Fam Physician 64(10):1735-1740.

CME, Inc. is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. CME, Inc. designates this educational activity for a maximum of 1.5 credit hours in category 1 credit toward the AMA Physician's Recognition Award. Each physician should claim only those hours of credit that he/she actually spent in the educational activity.

Psychopharmacology: Prescribing for Patients With Hepatic or Renal Dysfunction

Educational Objectives:

Psychiatric Times - Category 1 Credit