Altered Pharmacokinetics in the Elderly

ALTERED PHARMACOKINETICS IN THE ELDERLY

Elderly patients may also develop drug-related problems even when their medication is confined to a single agent or non-interacting multiple agents. This may relate to pharmacokinetic and pharmaco-dynamic changes associated with ageing. Such age-related physiological changes may alter the way in which the body handles medication, leading to changes in drug disposition in the elderly patient.

ABSORPTION

Following oral administration, most drugs dissolve in the stomach. Little absorption takes place here because of the small surface area and low pH, which means that drugs which are weak bases are in an ionised state. Absorption primarily takes place in the small intestine because of the large surface area and high pH, which favours the unionised state of most drugs. With increasing age, many changes occur in the gastrointestinal tract which should make the rate and extent of absorption less predictable, including a reduction in acid secretion in the stomach, decreased gastric emptying, diminished splanchnic blood flow and decreased gastrointestinal mobility (Geokas and Haverback, 1969; Evans et al., 1981; Greenblatt et al., 1982; Goldberg and Roberts, 1983; Montamat et al., 1989; Woodhouse, 1994). However, in practice, few drugs have significantly delayed rates of absorption (Greenblatt et al., 1982; Woodhouse, 1994). This is probably because potentially rate-limiting factors in the small intestine (such as surface area and lumi-nal pH) are not altered to a critical degree. A recent review by Cusack (2004) concludes that stud-ies performed in the last decade assessing extent of absorption by comparing area under the curve (AUC) after oral and IV administration and rate of absorp-tion using Tmax corroborate the prevailing opinion that ageing does not affect absorption to a significant degree.

Once drugs are absorbed from the gut, they enter the portal circulation and must pass through the liver before entering the systemic circulation. The bioavail-ability of most polar or water-soluble drugs is not affected by age because they are not highly extracted by the liver. For many lipophilic drugs, this first pass through the liver is accompanied by pronounced (sometimes over 90%) extraction with only 5%–10% of the dose reaching the systemic circulation. It is clear that a small change in hepatic function may result in a large increase in bioavailability in those drugs which undergo a high presystemic first-pass metabolism (Montamat et al., 1989; Woodhouse, 1994). For exam-ple, decreased presystemic extraction in the elderly may lead to increases in the bioavailability of propra-nolol (Castleden and George, 1979) and nifedipine (Robertson et al., 1988), but usually not to a clinically significant extent. The changes may be more marked, however, in the frail and hospitalised elderly (Wood-house, 1994). Bioavailability may also be regulated by presystemic extraction by small bowel cytochrome P-450 3A4 and by the extrusive action of P-glycoprotein on the surface of cells in the small bowel. Limited data do not support the effect of ageing on either process (Cusack, 2004).

DISTRIBUTION

Following absorption of a drug, the extent to which it is distributed within the body depends on body composition, plasma protein binding and blood flow.

Body Composition

With age, there is a decrease in lean body mass and body water and a corresponding increase in adipose tissue in relation to total body weight (Edelman and Leibman, 1959; Forbes and Reina, 1970; Novak, 1972). Adipose tissue increases from about 18% to 36% in men and from 33% to 45% in women (Novak, 1972). Therefore, the distribution of lipid-insoluble drugs such as paracetamol (Divoll et al., 1982) or ethanol (Vestal et al., 1977) may decrease in the elderly. This means that plasma concentra-tions per unit dose are higher. Lipid-soluble drugs such as diazepam are more widely distributed in the elderly and may have prolonged action and a ‘hang-over’ effect because of the longer elimination half-life (Macklon et al., 1980).

Protein Binding

Serum albumin levels decline with age, but in healthy elderly people this change is minimal. More marked reductions appear to relate to disease, immobility and poor nutrition rather than age itself (MacLennan et al., 1977; Campion, de Labrey and Glynn, 1988). This reduction may result in a decrease in the bind-ing capacity of weakly acidic drugs such as salicy-lates and phenytoin (Wallace and Verbeeck, 1987). Measurement of the plasma-free drug concentration (which will be increased under these circumstances) may be a better guide to the dose requirements than the total plasma concentration, particularly if the ther-apeutic ratio is low (Grandison and Boudinot, 2000). However, a raised free fraction will also result in an increased clearance allowing a new steady state to be achieved with regular dosing. Total plasma drug concentrations may then be lower, but free-drug concentrations will remain the same as these are deter-mined by hepatic or renal clearance of free drug. On the other hand, -l-acid glycoprotein increases with age, and basic drugs such as lignocaine display increased protein binding in elderly patients (Cusack et al., 1980).

Metabolism

Although some drugs are eliminated directly by the kidneys, many undergo metabolism in the liver first. Clearance of drugs by the liver depends on the activ-ity of the enzymes responsible for biotransformation and on blood flow, which determines the rate of delivery of the drug to the liver. For drugs that are metabolised relatively slowly by the liver (those with low intrinsic clearance), clearance is proportional to the rate of hepatic metabolism (Woodhouse, 1994). Hepatic mass decreases with age by 25%–35%, so the metabolism of such drugs may be reduced (Wood-house and James, 1990).

The metabolic pathways involved in the biotrans-formation of drugs may be divided into two phases (Williams, 1967). Phase 1 reactions comprise oxida-tive, reductive or hydrolytic processes which render the compound less lipophilic but can be fully or partly active. Products of phase 1 may then undergo phase 2 reactions which involve glucuronidation, sulphation or acetylation. The resulting conjugates are much more polar than the parent compound, usually have little or no pharmacological activity and are gener-ally excreted in the urine. Phase 1 oxidative drug metabolism may be reduced in the elderly (O’Malley et al., 1971), but phase 2 reactions are generally thought not to be altered, at least in fit elderly patients. However, in the frail elderly, in those who have suffered injury or have undergone surgery, enzyme activity may be significantly depressed, resulting in higher blood concentrations and an increased risk of adverse reactions (Woodhouse, 1994). In particular, a reduction in plasma aspirin esterase activity, paraceta-mol conjugation and metabolism of metoclopromide and theophylline have been reported in frail elderly patients (Wynne et al., 1990, 1993; Groen et al., 1993; Israel et al., 1993).

Metabolism of many drugs, such as the benzodi-azepines, may involve phase 1 followed by phase 2 reactions. Diazepam undergoes oxidative (phase 1) metabolism and its elimination is prolonged in the elderly (Belantuono et al., 1980). It is also partly converted to an active metabolite, desmethyl-diazepam, which has a half-life of up to 220 hours in elderly people. However, other benzodiazepines, such as lorazepam, undergo conjugation reactions in the liver, and their metabolism is unaltered by age. These compounds which do not give rise to active compounds may therefore be safer for elderly people to use than the other benzodiazepines (Williams and Lowenthal, 1992).

Age may not be the only factor that affects drug metabolism. Cigarette smoking, alcohol intake, dietary considerations, drugs, illnesses and caffeine intake may be equally important (Vestal et al., 1975; Montamat et al., 1989). In addition, hepatic blood flow rather than microsomal enzyme activity is the major determinant of total clearance of many drugs which have a very rapid rate of metabolism and consequently high extraction rates across the liver. Hepatic blood flow is 35% lower in healthy people over 65 years of age than in young people (Wynne et al., 1989). Reductions in systemic clear-ance of drugs with high hepatic extraction ratios (including presystemic clearance) have been reported in elderly people. Such drugs include propranolol (Castleden and George, 1979), clomethiazole (Nation et al., 1976) and morphine (Baillie et al., 1989), and the reduced clearance is compatible with a decline in liver blood flow.

RENAL EXCRETION

Most polar drugs or polar drug metabolites are elimi-nated by the kidney after filtration at the glomerulus. In addition, drugs such as the -lactam antibiotics are actively secreted in the proximal tubules. As part of normal ageing, both renal functional capac-ity and renal reserve diminish. The structural changes include a decrease in renal weight, thickening of the intrarenal vascular intima, a reduction in the number of glomeruli with increased sclerosis within those remaining and infiltration by chronic inflammatory cells and fibrosis in the stroma (Muhlberg and Platt, 1999). Altered renal tubular function may also lead to impaired handling of water, sodium and glucose in old age. There is a steady decline in the glomeru-lar filtration rate by approximately 8 ml/minute per decade (Rowe et al., 1976). By the age of 70, there-fore, a person may have a 40%–50% reduction in renal function (even in the absence of overt renal disease).

Drug elimination may be reduced even in patients with normal serum creatinine concentrations because creatinine production decreases with age. Many drugs which are dependent on the kidney for elimina-tion will accumulate to toxic levels if given in  the usual doses to elderly people. Examples include digoxin (Smith, 1973), atenolol (McAinsh, 1977) and amiloride (George, 1980). In addition, reduced clearance of active metabolites of certain drugs may increase the risk of toxicity particularly in very elderly patients. One example is morphine-6-glucuronide, the active metabolite of morphine (McQuay et al., 1990). Furthermore, many drugs themselves adversely affect renal function in the elderly, for example aminoglyco-sides, diuretics, NSAIDs and angiotensin-converting enzyme (ACE) inhibitors. In this way, age-dependent changes in renal function are responsible for altered pharmacokinetics in the elderly, but in many cases, the kidneys are the target for the ADRs produced by these changes (Muhlberg and Platt, 1999).

As drug elimination is correlated to creatinine clear-ance, estimating the creatinine clearance may be help-ful in deciding whether a dose reduction is necessary. A useful method that may be used at the bedside is the Cockcroft formula (Cockcroft and Gault, 1976):

The diagnostic value of age and creatinine clear-ance (calculated by the Cockcroft formula) for the prediction of potentially toxic drug plasma levels has been reviewed by Muhlberg and Platt (1999). They found that 256 geriatric patients with many differ-ent illnesses have been studied in 17 pharmacokinetic studies with 17 different drugs, including angiotensin-converting enzyme inhibitors, NSAIDs, antibiotics, beta-blockers, bronchodilators and benzodiazepines. Mathematical simulation and pharmacokinetic meth-ods were used to determine whether a dose reduction was necessary in elderly patients with a reduced crea-tinine clearance determined by the Cockcroft formula. For most drugs studied, elevated plasma levels at steady state could be correctly predicted when the creatinine clearance was < 40 ml/min, particularly when age was taken into account, suggesting that a dose reduction was necessary. This confirms the usefulness of the Cockcroft formula for clinical use in elderly patients taking drugs which are eliminated in the kidney and which are toxic at higher plasma concentrations. When using drugs with a low ther-apeutic ratio, estimation of the creatinine clearance helps determine the initial dose but, when possible, should be supplemented by therapeutic drug monitor-ing (Cusack, 2004).