CME credit for this article is now expired. It appears here for your reference.
At the end of this article, readers should be able to:
1. Appreciate and analyze the pharmacodynamic and pharmacokinetic changes that occur as patients age.
2. Develop/modify appropriate treatment strategies for older adults based on pharmacodynamic and pharmacokinetic changes.
3. Assess signs of problems associated with medication pharmacodynamics and pharmacokinetics in older patients.
Heterogeneity is a hallmark of the geriatric population, and nowhere is this as evident as it is in geriatric pharmacotherapy. Variable changes in organ and receptor function with aging make it difficult, in fact, to formulate rules that apply to all patients in the geriatric population. Comorbid disease and coadministered medications, which generally increase with age, add to the complexity.
Certain determinants of drug response in younger patients continue into old age. This is true, for example, of genetic endowment related to cytochrome P-450 (CYP450) metabolizing capacity. Other processes, such as drug absorption and glucuronidation, are little affected by aging. As a general rule, however, all other pharmacological processes change with aging.
Basic pharmacokinetic processes
Absorption. As shown in Figure 1, an orally ingested drug passes from the stomach to the proximal small intestine, where most absorption takes place. Absorption was once thought to be a simple, passive process, but research over the past few decades has shown that enterocytes lining the intestinal villi express a variety of specific drug transporters to take in drugs and efflux mechanisms to extrude them back into the gut lumen. The enterocytes are thus able to regulate how much drug is passed on to the liver.
P-glycoproteins (Pgp) that are first encountered in the gut wall act as “pumps” to move absorbed drug back into the gut lumen, where it can be metabolized by resident CYP3A4 enzymes, reabsorbed, or passed along the gut for excretion. The action of the Pgp pump is shown schematically in Figure 2. This pump has a general efflux function in the liver, the brain, the kidneys, and other organs. The Pgp pump has specific drug substrates, inhibitors, and inducers, similar to CYP450 enzymes. Of particular interest to psychiatrists are drugs listed in Table 1.
As with CYP450 enzymes, genetic variations that result in expression and activity of the Pgp pump have been identified in the MDR1 gene that codes for Pgp.1 With aging, reduced Pgp pump function at the blood-brain barrier has been demonstrated in vivo.1,2 This results in a “leakier” blood-brain barrier that is more porous to drugs and toxins.
How quickly a drug is absorbed determines in part how quickly it takes effect. Among the elderly, the rate of absorption may be affected by reduced gastric motility or delayed emptying due to diseases (eg, diabetes) or drugs (eg, antacids). In the absence of significant disease, however, the extent of absorption is little affected in the elderly.3
Distribution. Drugs that pass from the GI tract to the liver travel by way of the portal circulation. In the liver, the drugs can travel unchanged into the general circulation by way of hepatic veins. Alternatively, drugs can be metabolized in the liver before entering the circulation, via first-pass metabolism. The extent of first-pass metabolism is determined by the activity of enzymes such as CYP450 and uridine 5'-diphosphate-glucuronosyltransferase (UGT). Any drug that is conjugated before leaving the liver for the general circulation can be excreted directly by the kidneys. A drug can also be removed by Pgp pump action in the liver, which results in the secretion of substrate drugs and toxins into bile, where they are removed as waste through the GI tract.
Pharmacokinetic processes up to this point are circumvented by intramuscular or intravenous administration of the drug. Avoiding first-pass metabolism renders drugs nearly 100% bioavailable, so that the dose required for the same effect is usually much smaller. Intravenous administration of drugs in geriatric patients should be approached with caution, because dangerous adverse effects can emerge quickly. In addition, intramuscular administration of drugs is not recommended for geriatric patients with small muscle mass, because this leads to erratic absorption and painful injection.
As shown in Figure 1, drugs that pass into the general circulation are distributed to target organs, peripheral storage sites in fat or muscle, the kidneys for elimination, or the liver for metabolism. Distribution to peripheral storage sites is significantly affected by aging, because as lean body mass decreases, there is a relative increase in fat stores. This is true even for thin elderly patients. These changes create a larger volume of distribution for fat-soluble drugs (including most psychotropics) and a smaller volume of distribution for water-soluble drugs, such as lithium.4 Because the elimination half-life of a drug is directly proportional to its volume of distribution, the practical significance of these changes is that most psychotropics remain in the body longer in geriatric patients.
Another result of the relative increase in body fat with aging is that highly lipophilic drugs, such as diazepam, are rapidly taken up by fat storage sites, so that drug concentration in the blood falls quickly below a minimum effective threshold. With one-time dosing, the duration of effect is short for diazepam. The drug is removed slowly from fat stores, and significant accumulation of the drug can occur with repeated dosing. Moreover, the drug may be released erratically, which can result in changing serum levels over time. For these and other reasons, diazepam is not recommended for the elderly.
On the other hand, the smaller volume of distribution for water-soluble drugs, such as lithium, is associated with a greater amount of drug in the circulation and a correspondingly greater amount available to the brain as the target organ. This is one reason why smaller doses of lithium are used in elderly patients. Drugs that reach the blood-brain barrier gain access to the brain by passive diffusion or by transporters. Those drugs that are substrates for the Pgp pump residing in capillary endothelial cells may be extruded back into the circulation, a process that can be blocked by Pgp pump inhibitors or facilitated by inducers.
Metabolism. Drug metabolism occurs by phase 1 (oxidation) and/or phase 2 (primarily glucuronidation) processes. The most important drug-metabolizing enzymes are the phase 1 CYP450 enzymes and the phase 2 UGT enzymes. In general, phase 1 oxidation processes are more affected by aging than are UGT processes. Given the choice in treating an elderly patient, a drug metabolized by glucuronidation (also known as “conjugation”) would be preferred to a drug metabolized by those CYP450 enzymes whose activity is reduced with aging. Among benzodiazepines, for example, lorazepam, oxazepam, and temazepam are preferred to other drugs because these drugs are simply conjugated and excreted. Diazepam and chlordiazepoxide—not recommended for the elderly—have several active metabolites, some with long half-lives (eg, desmethyldiazepam).
Three families of CYP enzymes are relevant to psychopharmacology: CYP1, CYP2, and CYP3. Current evidence suggests that within these families, 5 enzymes are primarily responsible for the metabolism of commonly used psychotropics. The activity of each CYP450 enzyme can be reduced by inhibitors or facilitated by inducers. Inhibitory effects are additive when more than one inhibitor is present. For most drugs metabolized by CYP450 enzymes, it is the parent drug that is active, and metabolism results in an inactive or less active metabolite. For certain drugs (eg, codeine, tramadol), the parent drug is a pro-drug with little activity, and metabolism to an active metabolite is necessary for drug effect. In these cases, inducers result in greater drug effect. Clinically relevant substrates, inhibitors, and inducers of CYP450 enzymes are listed in Table 2.
Genetic polymorphisms expressed as the absence or changed activity of certain CYP enzymes may result in marked differences in serum levels of psychotropic drugs among individuals administered the same dose. The following are CYP450 metabolizer phenotypes5:
• Poor (slow): little or no enzyme activity, leading to high levels of parent drug
• Intermediate: depending on the specific genotype, activity ranges from slightly more than the poor phenotype to slightly less than the extensive phenotype
• Extensive (average, rapid): normal enzyme activity but can be converted to a poor metabolizer by sufficient CYP enzyme inhibition
• Ultrasensitive (ultrarapid): parent drug is metabolized so rapidly that higher drug doses may be needed; in the case of a pro-drug, toxicity may be seen at usual therapeutic doses
Only 2 of the CYP enzymes important in psychotropic prescribing decline meaningfully with aging: CYP1A2 and CYP3A. Aging also brings a gradual reduction in liver mass, reduced blood flow, and reduced hepatic metabolic rate.4 Among other processes, these changes affect demethylation reactions involved in the conversion of tertiary amines (eg, amitriptyline) to secondary amines (eg, nortriptyline), which may cause the accumulation of more active and more toxic tertiary compounds.
Excretion. Most drugs and metabolites are excreted from the body by the kidneys at a rate determined by the glomerular filtration rate (GFR).3 For many individuals, this rate falls linearly with aging at a rate of about 1 mL/min/1.73 m2 of body surface area (the average for an adult), starting at around age 20 to 30. Thus, the GFR for such an individual at age 80 years is about half what it was at 30 years.
Drugs cleared solely by renal excretion—lithium and hydroxylated metabolites (eg, hydroxy-bupropion, hydroxy-venlafaxine, hydroxy-nortriptyline, and hydroxy-risperidone)—are most affected by this change in GFR. Reduced clearance of lithium has led to the recommendation that elderly patients be treated with a single nightly dose of short-acting lithium. This schedule allows the serum level to drift downward so that the kidneys have time to recover before the next dose.
Reduced GFR with aging is not a universal phenomenon; some elderly persons show no decline in renal function with aging. Nevertheless, in the elderly with reduced renal function, psychotropics should be used with great caution. The following recommendations apply:
• Optimize hydration
• Beware of coadministration of nephrotoxic drugs (eg, NSAIDs)
• Evaluate renal function to determine dosage
• Do not rely on serum creatinine value to gauge renal function
• Use laboratory-generated report of the GFR or calculate GFR using the Cockroft-Gault equation
Drug clearance. Clearance of a drug from the circulation is a function not only of renal excretion but also of hepatic metabolism. Although oxidative metabolism may be reduced in old age, a more significant factor is decreased hepatic blood flow, which may be as much as 50% decreased.3 This is the primary reason why drug dosing is reduced for hepatically metabolized drugs in elderly patients. The correlation of blood flow reduction with hepatic clearance is a poor one, however, and existing liver function tests are not a good indicator of drug metabolizing capacity.3
Clearance is the principal determinant of the plasma concentration of a drug at steady state. The two quantities are related by the following equation:
drug concentration at steady state = dosing rate/clearance
Reduced GFR and reduced hepatic blood flow result in reduced clearance in elderly patients compared with younger patients. This results in increased steady-state drug concentrations, with enhanced main effects and toxic effects. Reduced clearance can be managed with a smaller dose or a longer dosing interval, giving rise to the “start low and go slow” principle of prescribing for a geriatric patient.
Another reason to “go slow” with dosing in elderly patients is because of the time it takes a drug to reach steady state with repeated dosing. Steady state is the point at which the average drug concentration plateaus. Time to steady state is determined by the drug’s half-life:
steady state = 4.5 x half-life
In general, a drug’s half-life is increased in geriatric patients compared with younger patients, so that it takes longer to reach steady state. Changes in drug dosage are best made when steady state has been achieved, so that main and toxic effects can be observed before the dosage is escalated. The half-life of a drug can also be used to determine the time to washout:
washout = 4.5 x half-life
The time to washout can be helpful in determining when a drug is out of the system so that a new medication can be started or withdrawal effects can be expected.
Basic pharmacodynamic processes
Pharmacodynamic processes are set in motion when the drug reaches the target tissue. Effects can be presynaptic or postsynaptic, or involve enzyme inhibition. Reuptake inhibitors, such as SSRI antidepressants, are examples of drugs that act at presynaptic sites. Acetylcholinesterase inhibitors (donepezil, galantamine, and rivastigmine) are examples of enzyme inhibitors.
Most psychotropic drugs act at postsynaptic sites; some also interact with autoreceptors on the presynaptic neuron. In general, the mechanism for these drugs includes drug-receptor binding, signal transduction, and cellular response. Effects are mediated by variables such as receptor density, concentration of the drug at the receptor site, affinity of the drug for the receptor, allosteric modulation of drug binding, intrinsic activity of the drug (the degree to which it influences the receptor to generate a cellular response), the function of second messenger systems, and homeostatic processes that tend to counter drug effects. Different drugs have different effects at the receptor.
Another variable in drug binding that is particularly relevant to drugs such as antidepressants is that of time: acute effects may be different from chronic effects. This occurs because initial pharmacological effects at the receptor induce small changes that over time result in receptor adaptation.
In general, receptors for dopamine, norepinephrine, and serotonin are coupled to G-proteins as second messengers. The mechanisms by which drug-receptor binding ultimately triggers brain effects are complex. Signaling molecules other than G-proteins—Akt, glycogen synthase kinase-3, and β-arrestin 2—have been shown in recent years to be involved in the mechanism of action of antidepressants, lithium, and antipsychotics.6
Although elderly patients experience greater drug effects than younger patients, the reason is not always clear. The differences in drug responses can be attributed either to different baselines or to different sensitivities.7 Baseline differences in postural sway (the elderly have more sway than younger patients), for example, might account for the increased risk of falling with benzodiazepine use in the elderly. It may be that for those with a “normal” amount of postural sway (old or young) at baseline, these drugs are safe.
Another factor in differing drug responses among the elderly for certain psychotropic drugs is the reduced function of the Pgp pump with aging. The EC50 of a drug is the (effective) concentration of that drug that yields a half-maximal response. The value of EC50 for any drug is determined systemically, not in the CNS. When the elderly are found to have a lower EC50 than younger patients for a particular psychotropic drug, is that because of increased brain “sensitivity” to the drug or because of a relatively higher concentration of the drug in the CNS?
When increased sensitivity is actually demonstrated, it may be attributed to the fact that the elderly use an ever-greater proportion of reserves (eg, cognitive, motor) as they age, so that fewer reserves are available to offset a perturbation. For example, the elderly are at increased risk for orthostasis as an adverse drug effect.
Drug interactions are more frequent in elderly patients because more medications are taken. In addition, drug interactions may be more serious because of insufficient physiological reserves. When new medications are started or stopped in elderly patients, it is very important to take note of potential interactions with other drugs or foods.
Drug interactions may be pharmacokinetic or pharmacodynamic. Pharmacokinetic interactions may involve any of the phases of drug processing, although the most important and numerous relate to drug metabolism. An example of a drug interaction influencing drug absorption is the effect of an oral antacid on absorption of an antipsychotic or a benzodiazepine. The combination results in a slowing of the rate of absorption, and adverse effects are not seen as quickly. Interactions relating to drug metabolism are mediated largely by CYP450 and UGT enzyme systems. The various actions of the Pgp pump can interact with drug distribution. Introducing a low-sodium diet to a patient receiving lithium results in competition for excretion and may cause lithium levels to rise.
Pharmacodynamic interactions are numerous and involve both receptor binding effects (eg, antipsychotics given with dopamine agonists), complex remote effects (eg, SSRIs inhibiting platelet function when given with warfarin, causing significant bleeding), and additive effects (eg, MAOIs and SSRIs causing serotonin syndrome).
Sandra Jacobson, MD, has no disclosures to report.
James M. Ellison, MD, (peer/content reviewer) has disclosed that he received research support from Eli Lilly and Company.
1. Toornvliet R, van Berckel BN, Luurtsema G, et al. Effect of age on functional P-glycoprotein in the blood-brain barrier measured by use of (R)-[(11)C]verapamil and positron emission tomography. Clin Pharmacol Ther. 2006;79:540-548.
2. Bartels AL, Kortekaas R, Bart J, et al. Blood-brain barrier P-glycoprotein function decreases in specific brain regions with aging: a possible role in progressive neurodegeneration. Neurobiol Aging. 2009;30:1818-1824.
3. Greenblatt DJ, Sellers EM, Shader RI. Drug therapy: drug disposition in old age. N Engl J Med. 1982;306:1081-1088.
4. Thompson TL 2nd, Moran MG, Nies AS. Drug therapy: psychotropic drug use in the elderly: Part 1. N Engl J Med. 1983;308:134-138.
5. Wynn GH, Oesterheld JR, Cozza KL, Armstrong SC. Clinical Manual of Drug Interaction Principles for Medical Practice. Washington, DC: American Psychiatric Publishing, Inc; 2009.
6. Beaulieu JM, Gainetdinov RR, Caron MG. Akt/GSK3 signaling in the action of psychotropic drugs. Annu Rev Pharmacol Toxicol. 2009;49:327-347.
7. Bowie MW, Slattum PW. Pharmacodynamics in older adults: a review. Am J Geriatr Pharmacother. 2007;5:263-303.
8. Jacobson SA, Pies RW, Katz IR. Clinical Manual of Geriatric Psychopharmacology. Washington, DC: American Psychiatric Publishing, Inc; 2007.
9. Oesterheld J. P-glycoprotein (PGP) table—the effect of drugs and foods. http://www.genemedrx.com/PGPtable.php. Accessed December 18, 2012.
10. Flockhart DA. P450 Drug interaction table: abbreviated “clinically relevant” table. http://medicine.iupui.edu/clinpharm/ddis/ClinicalTable.aspx. Accessed December 18, 2012.