There is a progressive functional decline in many organ systems with advancing age. Table 11–1 reviews some common physiologic changes associated with aging, with an emphasis on those changes that can affect pharmacotherapy. For more detailed information, readers are referred to excellent reviews.11,12
Age-associated physiologic changes may cause reductions in functional reserve capacity (i.e., ability to respond to physiologic challenges or stresses) and the ability to preserve homeostasis, thus making the elderly susceptible to decompensation in stressful situations.11–13 To deal with physiologic challenges or stresses, older individuals may require up to 95% of their remaining reserve capacity.11,12 The cardiovascular, musculoskeletal, and central nervous systems appear to be most affected.12 Examples of homeostatic mechanisms that may become impaired include postural or gait stability, orthostatic blood pressure responses, thermoregulation, cognitive reserve, and bowel and bladder function. An event resulting in functional impairment may involve an insult for which the body cannot compensate, and relatively small stresses may result in major morbidity and mortality.11–13
A number of age-related physiologic changes occur that could affect drug pharmacokinetics and pharmacodynamics (see Table 11–1). Unfortunately, data on the pharmacokinetics and pharmacodynamics of individual drugs commonly used in older adults are limited. This information gap may improve with implementation of Food and Drug Administration guidelines calling for pharmacokinetic studies by pharmaceutical companies for new molecular entities likely to be used in older adults.14
Table 11–2 and the following discussion summarize what is known about the effect of aging on each of the four major facets of pharmacokinetics.13,15 Of interest, when multivariate population pharmacokinetic analyses are conducted, age by itself seldom is a significant predictor of individual pharmacokinetic parameters (e.g., clearance). Aging-associated changes in drug absorption, distribution, metabolism, and elimination are more important predictors of altered pharmacokinetics than is aging, per se.
TABLE 11-2 Age-Related Changes in Drug Pharmacokinetics |Favorite Table|Download (.pdf)
TABLE 11-2 Age-Related Changes in Drug Pharmacokinetics
|Pharmacokinetic Phase||Pharmacokinetic Parameters|
|Gastrointestinal absorption||Unchanged passive diffusion and no change in bioavailability for most drugs|
|↓ Active transport and ↓ bioavailability for some drugs|
|↓ First-pass extraction and ↑ bioavailability for some drugs|
|Distribution||↓ Volume of distribution and ↑ plasma concentration of water-soluble drugs|
|↑ Volume of distribution and ↑ terminal disposition half-life (t1/2) for fat-soluble drugs|
|↑ or ↓ Free fraction of highly plasma protein-bound drugs|
|Hepatic metabolism||↓ Clearance and ↑ t1/2 for some oxidatively metabolized drugs|
|↓ Clearance and ↑ t1/2 for drugs with high hepatic extraction ratios|
|Renal excretion||↓ Clearance and ↑ t1/2 for renally eliminated drugs and active metabolites|
Most drugs are taken orally; thus, a number of age-related changes in gastrointestinal physiology could affect the absorption of medications. Fortunately, most drugs are absorbed via passive diffusion, and age-related physiologic changes appear to have little influence on drug bioavailability.16 A few drugs require active transport for absorption, so their bioavailability may be reduced (e.g., calcium in the setting of hypochlorhydria). However, there is evidence for a decreased first-pass effect on hepatic and/or gut wall metabolism that results in increased bioavailability and higher plasma concentrations of drugs such as propranolol and morphine.16 Increased drug bioavailability also may be seen with the concurrent ingestion of grapefruit juice. Constituents of this product inhibit cytochrome P450 (CYP450) isoenzyme CYP3A4, thus decreasing first-pass metabolism and resulting in exaggerated pharmacologic effects.17
The distribution of medications in the body depends on factors such as blood flow, plasma protein binding, and body composition, each of which may be altered with age. For example, the volume of distribution of water-soluble drugs is decreased, whereas lipophilic drugs exhibit an increased volume of distribution.13,15 Changes in the volume of distribution can have a direct impact on the amount of medication that must be given as a loading dose. The elderly may also exhibit differences in the distribution of drugs to their sites of action. For example, for an immunosuppressant that act with T-lymphocytes, the elderly had a mean 44% increase in the intracellular (T-lymphocyte)-to-whole blood concentration ratio of cyclosporine, compared with younger patients.18
P-glycoprotein, a member of the multidrug resistant (MDR)-associated protein family of efflux transporters, influences the transport of drugs across the blood—brain barrier. Studies using verapamil labeled with carbon-11 (a positron emitter) and positron emission tomography have demonstrated decreased P-glycoprotein activity in the blood—brain barrier with aging. As a result of this, the brain of aged individuals may be exposed to higher than normal levels of drugs and toxins.19
The two major plasma proteins to which medications can bind are albumin and α1-acid glycoprotein, and concentrations of these proteins may change with concurrent pathologies seen with increasing age.20 For acidic drugs such as naproxen, phenytoin, tolbutamide, and warfarin, decreased serum albumin may lead to an increase in free fraction. An increase in α1-acid glycoprotein induced by burns, cancer, inflammatory disease, or trauma may lead to a decreased free fraction of basic drugs such as lidocaine, propranolol, quinidine, and imipramine. In the absence of compromise in excretory pathways, these potential changes are unlikely to have any deleterious clinical effect. However, they may be important to consider when interpreting serum concentrations of these drugs because usually only total drug concentrations (sum of free and protein-bound drug) are reported.
The liver is the major organ responsible for drug metabolism, including phase I (oxidative) and phase II (conjugative) reactions.21 The most remarkable characteristic of hepatic function in older adults is the increase in interindividual variability compared with other age groups, a feature that may obscure true age-related changes.21 Data suggest that age-related declines in phase I metabolism more likely are the result of reduced hepatic volume than reduced hepatic enzymatic activity.22 Decreased phase I metabolism (e.g., hydroxylation, dealkylation) producing decreased drug clearance and increased terminal disposition half-life (t1/2) has been reported in the elderly for medications such as diazepam, piroxicam, theophylline, and quinidine. Phase II metabolism (e.g., glucuronidation, acetylation) of medications such as lorazepam and oxazepam appears to be relatively unaffected by advancing age. Hepatic enzyme induction (e.g., by rifampin, phenytoin) or inhibition (e.g., by fluoroquinolone and macrolide antimicrobials, cimetidine) does not appear to be affected by the aging process.21,23
Age-related decreases in hepatic blood flow can decrease significantly the metabolism of drugs with high hepatic extraction ratios, such as imipramine, lidocaine, morphine, and propranolol.21 The effect of aging on polymorphic drug metabolism has not been well studied. Advancing age reduces the metabolism of CYP450 isozyme 2D6 substrates by approximately 20%.24,25 Other available data suggest that advancing age has no significant effect on acetylation or on CYP450 isozymes 2C9- or 3A4-mediated metabolism.26–28 A single-point blood sampling method for evaluating CYP450 isoenzyme CYP3A4 activity in older adults has been described.29 A number of potential confounding factors, including race, gender, frailty, smoking, diet, and drug—drug interactions, may significantly affect hepatic metabolism in older adults.21
Frailty has proved a difficult condition to define and thus on which to conduct research. Recently, an objective definition of frailty has been proposed and is being widely used. Patients are considered frail if they have three or more of the following conditions: sarcopenia, weakness (measured by grip strength), self-reported exhaustion, low activity level, and slow walking speed.30 The activities of plasma cholinesterase, acetylsalicylate esterase, paraoxonase, and phenylacetate esterase are all significantly lower in frail elders compared with fit elders and young subjects.30 Acetaminophen clearance is significantly lower in frail versus fit elders, and similar findings have been noted for metoclopramide. Lastly, interindividual variability in the disposition parameters of antipyrine and theophylline is significantly greater in frail versus fit elderly women.30
Renal excretion is the primary route of elimination for many drugs. Although age-related reductions in glomerular filtration are well documented, as many as one third of “normal” older adult subjects may have no reduction as measured by creatinine clearance.13,15 Moreover, emerging information suggests that renal tubular secretion may not decline in proportion to other renal processes.31 The estimation of creatinine clearance, although not entirely accurate in individual patients, can serve as a useful screening approximation. Cockcroft and Gault32 created one of the most commonly used equations for adults with stable renal function whose actual weight is within 30% of ideal body weight:
where age is given in years, actual body weight in kilograms, and serum creatinine concentration in milligrams per deciliter. The resulting creatinine clearance is in units of mL/min. For women, multiply this result by 0.85.
When serum creatinine is expressed in μmol/L, creatinine clearance in units of mL/min can be calculated by the following equation:
The Modified Diet in Renal Disease33 equation has become more widely used for estimation of glomerular filtration rate. However, dosing guidelines for medications that primarily are renally cleared are still based on estimated creatinine clearance determined using the Cockcroft and Gault equation. MDRD results frequently can be used if the results are “deadjusted” for body surface area. In the future, use of another protein, cystatin C, a low—molecular-mass protein that is produced by all nucleated cells, is freely filtered at the glomerulus, and is not secreted by the renal tubules, may prove to be superior to the use of creatinine. This could be the case especially with coexisting conditions such as cachexia, sedentary lifestyle, malnutrition, and hepatic disease, because creatinine clearance is a poor predictor of glomerular filtration in the presence of these conditions.34
Medications whose excretion is primarily renal and for which there is evidence of age-related reduction in renal and total body clearance include (but are not limited to) amantadine, aminoglycosides, atenolol, captopril, cimetidine, digoxin, lithium, and vancomycin. Some hepatically metabolized medications can yield active, primarily renally excreted metabolites, such as N-acetylprocainamide, normeperidine, and morphine-6-glucuronide, which can accumulate with advancing age because of reduced renal function.
Sidebar: Clinical Controversy
When using the Cockcroft and Gault equation to estimate creatinine clearance in older adults, some clinicians round the value up to 1 if the patient's serum creatinine concentration is <1. Rounding the serum creatinine concentration may provide an underestimation of creatinine clearance and result in improper dose adjustment of renally eliminated medications. It is important to realize that the equation is merely an estimate, and attempts should be made to determine creatinine clearance accurately when use of certain medications (e.g., metformin) is being contemplated.
There is some evidence of altered drug response or “sensitivity” in older adults. Four possible mechanisms have been suggested: (a) changes in receptor numbers, (b) changes in receptor affinity, (c) postreceptor alterations, and (d) age-related impairment of homeostatic mechanisms.13,35 For example, muscarinic, parathyroid hormone, β-adrenergic, α1-adrenergic, and μ-opioid receptors exhibit reduced density with increasing age.13,34 Evidence from epidemiologic and experimental studies suggests that, independent of pharmacokinetic alterations, older adults are more sensitive to the central nervous system effects of benzodiazepines.13,35 Older adults exhibit a greater analgesic responsiveness to opioids compared with their younger counterparts, even when pharmacokinetic parameters are similar in the two groups.13,35 In addition, older adults demonstrate an enhanced responsiveness to anticoagulants such as warfarin and heparin as well as to thrombolytic therapy but not to the direct thrombin inhibitor ximelagatran.13,35,36 In contrast, older adults exhibit decreased responsiveness to certain drugs (e.g., β-agonists/antagonists).13,35 Reflex tachycardia, seen commonly with vasodilator therapy, often is blunted in older adults, perhaps because of dampened baroreceptor function. For some drugs (e.g., calcium channel blockers), both enhanced responsiveness (as demonstrated by greater reduction in blood pressure) and decreased responsiveness (as demonstrated by reduced atrioventricular nodal blockade) can occur simultaneously in older adults.13,35