Contents

Summary

Drug Disposition

Clin. Pharmacokinet. 17 (6): 385-395, 1989 0312-5963/89/00 12-0385/$05.50/0 © ADIS Press Limited All rights reserved. CPK003 CPK252

Clinical Pharmacokinetics of Antibacterial Drugs in the Elderly Implications for Selection and Dosage

Burt R. Meyers and Patricia Wilkinson Division of Infectious Diseases, and Pharmacy Department, Mount Sinai Medical Center, New York, New York, USA

Summary ......... ... ... ....... .................. ... ................ ....... ... .. ... ..... .... ........... ....... ................... .. ........ .. 385 I. Pharmacokinetics in the Elderly ........................... .. ....... .. .................................................. .. 386

1.1 Absorption ....................................... .. .................................................. .... ....... ..... ........ .... 386 1.2 Distribution .......................... ............................. .. ... .. ........... .... ....... ...... ..... ....... ..... .... ...... 386

2. Pharmacokinetics of Antibiotics .. ... .. ... .. ... .. .... ... ...... .............. .......... ...... ....... ...... ..... ....... ..... 387 2.1 Penicillins ........ .. .............. ..... ....... ... ....... ... ....... ... ................. ... ......... ................... .. ........... 387 2.2 Cephalosporins ...... ......... ...... ............ ..... ...... ... .... ................ ... ..... ....... ... ... .. ......... .. ........... 388 2.3 Monobactams .......... .... ....... ..... .............. ... ..... .. .. ... ..... .... ........... .... .... ......... .. .......... ... ....... 389 2.4 Carbapenems ............................... .. .................... ..... ... .. .......... ........ ........ ..... ........ ... ......... . 389 2.5 Aminoglycosides ....... ..... ......... ... .. ... ....... ...... ... ..... ... .................. ..... ...... ............ ..... ........... 389 2.6 Quinolones ......... ... .... .... ... ..... .... ........ ..... .. ...... ..... ....... .. .... ...... .......... ... ..... .... ........ .. .... .. .... 390 2.7 Glycopeptides ..... .... ...... .... ......... ... .. .... ........ ..... ..... ... .. ..... ......... .. ....... .. ... ... ..... .. ... ............. 390 2.8 Discussion ...... ..... ...................... .. .......... .... ............... ... ....... .. ......... ... .... .. ........... .. ............. 390

3. Renal Function ....... .. .. ... .... ... ........... .... .... ....... ..... ....... ........... .. ... ... ........ ....... ....... ....... .... ....... 391 4. Drug-Drug Interactions ...... .. .. .. .......... .............. .. .... ............ ... .. ........ ... ................................... 393 5. Therapeutic Considerations ................ .. ........ ............. ..... ........... .. ......... ... ... ............ .... ...... .... 393

A review of the clinical pharmacokinetics of antibiotics in the healthy elderly reveals that for most compounds a decrease occurs in renal clearance (associated with age-related decreases in renal function). as well as a prolonged ha/flife and increased area under the plasma concentration-time curve. These changes are amplified in the sick infected elderly. It is important that the treating physician be aware of the potential side-effects of anti­microbial agents. and whenever possible choose those which are associated with the least adverse effects. Individual patient variability. including underlying diseases and other pre­scribed medications. must be taken into account when dosage is selected.

!'I-Lactam compounds have a remarkable safety record: specifically in the elderly, their therapeutic/toxic ratio is much higher than that observed with aminoglycosides. Regimens for this class of drugs in the elderly should maintain antibiotic concentrations above the minimum inhibitory concentrations for maximum efficacy. In the treatment of elderly patients, it is suggested that dosage and interval be based on estimated or measured cre-

386 Clin. Pharmacakinet. 17 (6) 1989

atinine clearance. UsuallyJor drugs that are excreted primarily by the kidney (i.e. amino­glycosides. (3-lactams and quinalones). dosage intervals must be increased when there is an associated fall in creatinine clearance. The pharmacokinetic parameters suggest that as an alternative to increasing dosage interval the usual dose may be decreased. but fitrther studies are necessary for confirmation. .

The elderly over age 65 are a rapidly growing segment of the total population. It is projected that by the year 2030 17% of the population in the United States will be over age 65, compared with I I % noted in 1978 (Vestal 1978). In fact, the age group over 85 years in the United States is the fast­est growing sector of the total population (Alvarez et ai. 1988). Studies have revealed that the elderly receive 25% of all prescription drugs and spend more of their income on health care than younger populations (Everitt & A vorn 1986; Gibson et ai. 1977; Greenblatt et ai. 1982; Kovar 1977; Rowe 1977). They take a variety of medications with dif­ferent dosing intervals, and the potential for drug­drug interactions and ensuing toxicity is therefore great in this population. It is estimated that ad­verse effects may occur in up to 40% of elderly outpatients taking medication; 12 to 17% of those were admitted to hospital for adverse drug reac­tions (Alvarez et ai. 1988). The geriatric population also represents an increasing proportion of the total number of hospitalised patients: there are currently more beds in nursing homes than in hospitals (Al­varez et al. 1988).

The physiological changes associated with age­ing have already been described (Massoud 1984). This review focuses on age-related physiological changes and their impact on the pharmacokinetic parameters associated with antimicrobial admin­istration; alterations in the pharmacokinetics of antibiotics in both the healthy geriatric volunteer and the infected, hospitalised patient are analysed.

1. Pharmacokinetics in the Elderly 1.1 Absorption

A decrease in gastric acid secretion and an in­crease in gastric pH are associated with the ageing process (Massoud 1984). The absorption of drugs

that are dependent on increased acidity (e.g. sul­phonamides) may be decreased; on the other hand, drugs such as penicillin may have greater bioavail­ability secondary to the higher pH. Decreased pro­duction of pancreatic trypsin may lead to impaired breakdown of certain drugs, and a reduced number of absorptive cells coupled with a decreased surface area of the small bowel mucosae may also lessen drug absorption. The diminished splanchnic blood flow which has been described in this population may delay and decrease absorption (Massoud 1984). Gastric emptying time is often delayed in geriatric patients, and changes in motility of the stomach and small intestine may be significant for the bed­ridden and immobile (Gibaldi 1984). This delay in emptying may prolong the time before onset of antimicrobial effects, although bioavailability itself is probably not decreased in the elderly (Massoud 1984).

Studies have been carried out on the intramus­cular absorption of antimicrobial agents in the el­derly (Mayer et ai. 1986; Ripa et al. 1987a). A re­duction of physical activity may explain the decrease in absorption of parenterally admini­stered penicillin (Schmidt & Roholt 1966); it is also possible that regional blood flow, affected by mechanisms associated with ageing (such as atherosclerosis), may playa role.

1.2 Distribution

The distribution of an antibiotic is affected by a variety of factors including cardiac output, renal function, bodyweight and composition, serum al­bumin, disease states and concomitantly admini­stered drugs which may affect protein binding (Wallace & Verbeeck 1987).

The elderly have a body composition which dif­fers from that of younger groups. Total body water

Kinetics of Antibacterials in the Elderly

is decreased, and the ratio of body fat to total body water is increased (more so in females than in males), compared with younger individuals (Mas­soud 1984). A decreased lean body mass, coupled with a decrease in total body water, is associated with a decreased volume of distribution (V d) for water-soluble drugs (polar compounds). This may lead to higher concentrations of drugs which are distributed in water versus fat - for example, serum concentrations are increased in obese adults (Mey­ers 1980) and elderly patients (Zaske et al. 1982). For fat-soluble compounds (non-polar) the Vd will increase, and this may lead to a delayed onset of activity due to accumulation; adverse effects may occur after multiple doses.

A decrease in plasma proteins and protein bind­ing has been noted with age. The concentration of albumin decreases with increasing age in both the healthy and the sick elderly (Bender et al. 1975; Cammarata et al. 1967; Christopher et al. 1979; Dybkaer et al. 1981; Meyers et al. 1987a). The 2 major plasma proteins are albumin and ai-acid glycoprotein: the former binds mainly to acidic drugs, and the latter to basic drugs. Plasma protein binding changes are significant when a drug is more than 90% protein bound, or when Vd is low « 0.20 L/kg) [Gibaldi 1984; Massoud 1984]. With a decrease in albumin, a rise in the concentration of free drug would be observed in the case of a highly bound compound. A variety of illnesses and disease states produce a decrease in serum albu­min, such as malnutrition, hyperbilirubinaemia, hepatic disease, burns, surgery, trauma and renal insufficiency. Competition for receptors in the al­bumin molecule may lead to the displacement of administered compounds by metabolic byproducts such as uric acid and bilirubin; therefore, when these substances are elevated, an increase in the free drug fraction occurs (Wallace & Verbeeck 1987). Changes in albumin concentration will also alter the distribution and elimination of a com­pound. For drugs with a wide therapeutic index, the increased free fraction of a compound enables more drug to reach tissue sites (i.e., the cerebro­spinal fluid, joints and peritoneal fluid) to exert a beneficial effect. Normally, this free fraction would

387

Table I. Drugs primarily or partially metabolised by the liver

Isoniazid

Metronidazole Quinolones Rifampicin Macrolides

be excreted more rapidly via the kidney, but since there is an age-related decline in renal function this does not occur; the amount of free drug circulating is thus further increased. As a result of this rise in free fraction, the potential for toxicity of com­pounds with a low therapeutic index is enhanced.

In the case of drugs which are metabolised or eliminated by the liver (table I), a reduction in he­patic flow may decrease the clearance of those with high extraction ratios (Massoud 1984).

The pharmacokinetics of a variety of antimi­crobial classes in the elderly have been extensively reviewed by Ljungberg and Nilsson-Ehle (1987). In the following paragraphs we cover the literature since that publication.

2. Pharmacokinetics of Antibiotics 2.1 Penicillins

The pharmacokinetics of amoxicillin (single oral dose, 500mg) and ampicillin (intravenous dose of 500mg, repeated at I week) in 2 groups of healthy elderly subjects were studied by Sjovall et al. (1986), who observed a decrease in drug clearance. Foulds (1986) reported on the pharmacokinetics of am­picillin and sulbactam (both administered as 6 intravenous doses of 500mg every 8 hours) in el­derly patients (average age 70 years) undergoing colorectal surgery: the elimination half-life (tl;') in­creased from J to 2.3 hours for ampicillin and 1.6 hours for sulbactam; two days after surgery the tl;' of each drug was prolonged (ampicillin 2.3 hours, sulbactam 2 hours). When dicloxacillin was stud­ied, the percentage of unbound drug was found to be increased for healthy elderly subjects compared with a younger group (8.8 ± 1.0 vs 7.3 ± 0.8%, respectively) [Pacifici et al. 1987].

The tl;' of sulbenicillin after single intramuscu-

388

lar and multiple intravenous doses of 2g increased in plasma, due to decreased renal function in the elderly, but did not accumulate with multiple dos­ing, implying non-renal elimination (Ripa et al. I 987b). Meyers et al. (1987b) studied the phar­macokinetic properties ofmezlocillin in 12 healthy elderly subjects. Doses of mezlocillin 4 and 5g were administered intravenously to elderly volunteers; after a 5g dose the area under the plasma concen­tration-time curve (AVC), Vd and non-renal clear­ance (CLNR) were higher while the total clearance (CL) and renal clearance (CLR) were lower in the elderly. CLNR was increased in the elderly as a re­sult of decreased CLR, but this was not totally compensatory (Meyers et al. 1987b).

2.2 Cephalosporins

Luderer et al. (1984) documented a statistically significant increase in the free fraction of ceftriax­one in elderly volunteers compared with healthy young volunteers after a single intravenous infu­sion of Ig (15.7 vs 13.5%). The pharmacokinetics of sulbactam plus cefoperazone after long term administration were studied in 6 elderly patients who were seriously ill with intra-abdominal infec­tions, and who received sulbactam Ig and cefo­perazone 2g intravenously every 12 hours for 5 days. Initially, elimination was slower compared with healthy volunteers and there was increased pharmacokinetic variability. As patients recovered, the only parameter that changed significantly was a decrease in the t'/2 of sulbactam. Elimination of both drugs was slower in the acutely ill elderly sur­gical patients; therapeutic concentrations were found during the entire dosing interval, without evidence of accumulation (Schwartz et al. 1988).

Cefoperazone pharmacokinetics in ambulatory elderly volunteers compared with healthy young adults were studied by Meyers et al. (1987a). The subjects were 10 adults aged less than 30 years and 10 healthy elderly volunteers 65 to 78 years old. After a single intravenous dose of cefoperazone 2g, an increased AVC and t'/2 were noted in the elderly volunteers in comparison with the young adults; decreased CLR and CLT were also observed in the

Clin. Pharmacokinet. 17 (6) 1989

elderly. Matsuzaki et al. (1986) demonstrated the t'/2 of cefoperazone following intravenous infusion in elderly patients to be twice the normal value. Following a single intravenous dose of ceftizoxime Ig administered over 1 hour in young adults, the t'/2 of this drug was 1.3 hours and 76.4% was ex­creted in the first 6 hours; in patients aged 65 years old, the t'/2 was 2.15 hours, with 58.2% excreted in 7 hours (Horiuchi et al. 1986).

Single and multiple intravenous and intramus­cular doses of moxalactam 250 and 500mg were given to 19 male volunteers 60 years of age or older, with normal liver function tests and creatinine clearance (CLcR) ~ 60 mljmin; the plasma clear­ance correlated with measured and calculated CLcR, and the authors concluded that major dif­ferences in pharmacokinetics were related to de­creased renal function (Andritz et al. 1984). The pharmacokinetics of intramuscular cefotetan were studied in 10 healthy elderly males (age range 65 to 75) with normal liver function and CLCR > 80 mljmin. The age-related changes in clearance of this drug were small, and CL was significantly related to CLCR (Ripa et al. 1987a).

Sugihara et al. (1988) and Ludwig et al. (1988) have documented that the t'/2 in elderly patients is prolonged for both cefotaxime and its active me­tabolite, desacetylcefotaxime. Elderly patients were given intravenous cefotaxime Ig every 12 hours for 7 or more days by Sugihara et al.; t'/2 was prolonged by 32 to 55% compared with healthy volunteers. Vrinary excretion over 6 hours was similar to that for healthy volunteers, but only 29% of the dose was excreted in the first 2 hours in the elderly patients as opposed to 59.6% in the healthy volun­teers; it was higher in the elderly patients during the 2 to 6 hours following the dose, compensating for the slowed elimination in the first 2 hours. These authors also found that the ratio of the serum con­centration of desacetylcefotaxime to cefotaxime tended to increase slightly.

Ludwig et al. (1988) studied the pharmacokin­etics of cefotaxime and desacetylcefotaxime in age­stratified elderly patients after intravenous admin­istration of cefotaxime Ig. In the group of patients aged> 80 years, a significant difference was noted

Kinetics of Antibacterials in the Elderly

in elimination compared both with the other 2 groups (60 to 70 years old and 70 to 80 years old) and with previous studies in younger patients. The tl;' in the group aged 80 and over was almost twice that of the other groups: 2.S vs 1.S hours, respec­tively. The change in elimination of desacetylcefo­taxime was similar, but less pronounced. Ludwig et al. conclude that patients more than 80 years old may require less drug, but younger patients should receive the normal dose.

Ljungberg and Ehle (1988) studied the phar­macokinetics of ceftazidime in young and elderly healthy and ill patients. The pharmacokinetics after single and multiple doses of ceftazidime were com­pared in young healthy and elderly acutely ill men. All patients received intravenous ceftazidime 2g twice daily except for 3 patients over the age of 80 who received Ig twice daily. All patients had nor­mal renal function for their age. Many of the phar­macokinetic parameters of the drug were found to be prolonged in the elderly patients. Elimination half-life was 3.1h (elderly) vs 1.9h (young); AVC was 414.0 mg/L· h (elderly) vs 276.6 mg/L· h (young). The elderly patients had lower total and renal clearances, reduced urinary recovery over 12 hours and an enlarged volume of distribution at steady-state (V ss). These authors advise a reduc­tion in the dose of (J-Iactam antibiotics eliminated by the kidney in elderly patients with an acute bac­terial infection.

Single and multiple dose pharmacokinetics of ceftazidime were investigated by Ljungberg and Ehle (1988) in young and elderly acutely ill patients with normal renal function for their ages. Patients received ceftazidime 2g twice daily for 7 days ex­cept for 7 of the oldest patients in the oldest group, who received an initial 2g dose followed by Ig twice daily for the remainder of the course of therapy. The tl;' was approximately 2 hours in young and middle-aged patients, 2.73 hours in patients aged 60 to 79 years and 3.S4 hours in patients over 80 years old. The AVC was more than doubled in the oldest patients compared with those younger than 40 years. Elimination variables did not change with multiple dosing, but a small, significant increase in AVC was detected in the elderly. The authors rec-

389

om mend a dose reduction by SO% in patients over 70 years old.

Faulkner et al. (1988) studied the pharmacoki­netics of cefixime in young and elderly healthy volunteers. Subjects received I x 400mg oral tablet once daily for S days. AVC, t,/" mean residence time, average concentration, trough concentration and renal clearance values were statistically sig­nificantly higher in the elderly. The authors con­cluded that no dosage adjustment was necessary in the elderly except those with significant renal com­promise.

2.3 Monobactams

In a study of the pharmacokinetics of az­treonam after a single intravenous dose of Ig, de­creased clearances were related to decreased renal function (Creasey et al. 1985). The V d of az­treonam was essentially similar among both young and elderly (O.IS vs 0.16 L/kg, respectively) [Mattie 1988].

2.4 Carbapenems

The pharmacokinetics of single and multiple doses of imipenem and cilastatin (each SOOmg 4 times daily for 6 days) in healthy elderly men were studied by Toon et al. (1987) and Finch et al. (1986). Imipenem and cilastatin pharmacokinetics were similar in both elderly and young patients with mild renal failure. CL correlated with glomerular filtration rate (GFR) for both drugs (Toon et al. 1987), and the pharmacokinetics remained un­changed after multiple dosing (SOOmg for 4 doses). The elimination constant (ke), CL and AVC were significantly correlated with GFR but not with age (Finch et al. 1986).

2.S Aminoglycosides

In a study by Mayer et al. (1986), 12 institu­tionalised elderly patients (mean age 81) received intramuscular tobramycin 2 to 4 mg/kg/day; the peak serum concentration (Cmax) was slightly de­layed compared with studies performed on young

390

patients. Blood supply to the muscle and muscle mass can affect the rate of absorption of intra­muscular drugs. The authors conclude that either changing levels of activity or alternation of the in­jection site may be responsible for this delay.

The disposition of gentamicin (initial dose I to 1.5 mg/kg, subsequently adjusted pharmacokinet­ically to maintain the desired peak and trough con­centrations) was studied in 50 patients between the ages of65 and 90, with varying levels of renal func­tion. There was a correlation between gentamicin clearance and CLCR; the disposition of gentamicin is usually dependent on renal function but inde­pendent of age. Additionally, since the V d was the same, it was suggested that similar loading doses should be employed with a modification of the maintenance dose based on renal function, regard­less of age (Matzke et al. 1987).

2.6 Quinolones

The pharmacokinetics of a single oral dose of ciprofloxacin 250mg were compared in 10 young (ages 20 to 30) and 10 elderly (ages 60 to 73) healthy volunteers: times to Cmax (tmax) were equal and the 24-hour urinary recovery was similar in both groups. Cmax was greater for the elderly (1. 7 vs 1.2 mg/L, respectively), and the AUC corrected for bodyweight was 48% greater. The tlh increased, but not significantly (4.3 vs 3.7 hours, respectively). The authors concluded that a reduced Vd accounts for the differences seen in the elderly (Bayer et al. 1987).

Guay et al. (1988) investigated the effect of acute illness on the pharmacokinetics of oral ciproflox­acin in the elderly. 20 elderly patients received single and multiple dose regimens of 750mg every 12 hours. The pharmacokinetic parameters in the seriously ill elderly patients were consistent with those values derived from single dose studies in healthy and mildly ill elderly patients, but mark­edly different from previous observations in young adult volunteers.

Enoxacin 200mg orally twice daily for 5 days was studied in 23 elderly patients (age> 70 years) with urinary tract infections. No significant accu­mulation of the oxo-metabolite occurred, and the

Clin. Pharmacokinet. 17 (6) 1989

2-hour plasma concentration of enoxacin increased from a mean of 1.5 mg/L on day I to 2.8 mg/L on day 5 (Wise et al. 1987). The pharmacokinetics of this drug were further compared in healthy adults (Naber et al. 1986): mean Cmax in elderly patients after a dose of enoxacin 200mg was similar to that seen in healthy young volunteers following a 600mg dose. The tmax was not significantly different in the elderly. The fraction of drug excreted at 24 hours (mean 31.2%) was significantly less in elderly patients than young healthy volunteers (mean 71.6% over 48 hours) [Naber et al. 1986].

Dobbs et al. (1987) studied the effects of age on the pharmacokinetics of enoxacin. A single oral dose of enoxacin 600mg was administered to 10 young (18 to 45 years old) and 10 elderly (> 65 years) patients. The tmax was higher in the elderly, as was Cmax and AUC. Renal clearance was less in the elderly and urinary recovery of unchanged enoxacin was decreased (Dobbs et al. 1987).

2.7 Glycopeptides

The pharmacokinetics of a single intravenous dose of teicoplanin 6 mg/kg were studied by Ro­sina (1988) in 12 elderly patients with a moderate degree of renal impairment (mean CLCR 51.3 mlj h/kg); the Cmax was 45 mg/L and tlh was 107 hours compared with 70 hours in healthy volunteers (Buniva et al. 1986). Total teicoplanin clearance averaged 0.0106 L/h/kg; CLR was 40% of the total. There was a linear correlation between total and renal teicoplanin clearance and endogenous CLCR. The average total recovery of teicoplanin in the urine was 28% over 8 days (Rosina 1988).

2.8 Discussion

In the treatment of infections in the elderly, a variety of factors must be considered in the selec­tion of antimicrobial agents; most important are age-related variations in pharmacokinetics. Many agents, e.g. penicillins or cephalosporins, are often highly protein bound; decreased protein binding may be associated with factors other than the level of serum albumin. In patients who are infected or

Kinetics of Antibacterials in the Elderly

uraemic, or who have elevated serum bilirubin, these factors come into play by competition for al­bumin binding sites. Defects in the protein binding of acidic drugs, such as cephalosporins and peni­cillins, have been noted in patients who are uraemic. This can result in an increased free frac­tion of many drugs, especially organic acids. There can be marked increases in apparent volumes of distribution in uraemia with antimicrobials highly bound in serum proteins (Craig & Welling 1977). Geriatric patients take a variety of drugs for mul­tisystem problems; these compounds may compete for binding sites on serum albumin, thus increasing the amount of free drug available for distribution.

Changes in body composition which will also have an impact on drug distribution include re­duced lean body mass, increased body fat in rela­tion to bodyweight and decreased total body water. These alterations will produce a decrease in the V d of water-soluble drugs and an increase in that of lipid-soluble substances.

In a study of cefoperazone administered in doses of I and 2g the Vd was decreased in the elderly versus younger populations (Meyers 1987a). With moxalactam (Andritz et al. (984), however, the Vd was noted to be 17 to 39% greater in the elderly than in younger subjects. With regard to amino­glycosides, large patient variability in the Vd has been described (Zaske et al. 1982). In general, drugs of this class are poorly bound to serum albumin (0 to 35%), but are affected by a variety of other fac­tors: for example, serum concentrations are de­creased in fever and increased with obesity. Serum concentrations of gentamicin are usually higher in older age groups (Meyers 1980), and the smaller volume of extracellular water in most elderly in­dividuals will accentuate this effect; however, con­flicting results have been described for amikacin (Yasuhara et al. (982). Aminoglycosides are in gen­eral poorly distributed following intravenous administration. Lower concentrations are found in the non-obstructed biliary tree, cerebrospinal fluid, sputum and aqueous humor, and serum concen­trations in patients with increased extracellular fluid secondary to oedema, ascites and pleural effusion may also be lower.

Table II. Drugs excreted primarily by the renal route

Aminoglycosides Carbapenems

Cephalosporins· Cephamycins Monobactams Penicillins Quinolones Teicoplanin Vancomycin

391

a Exceptions are ceftriaxone and cefoperazone, which are ex­creted by the renal and biliary routes.

3. Renal Function

Renal excretion, by glomerular filtration either alone or together with tubular secretion, is the most important route of elimination for {3-lactam anti­biotics and aminoglycosides, accounting for 60 to 100% of the administered dose in 24 hours (table II). Two cephalosporins (ceftriaxone and cefoper­azone) are excreted primarily by biliary secretion, which is increased in patients with impaired renal function. In our study of cefoperazone it was ob­served that both CLR and CLNR were decreased in the elderly at both the 1 and 2g dose. However, CLNR, while still reduced, accounted for a greater percentage of CL in the elderly volunteer popula­tion than was the case for the younger age group; CL was decreased, suggesting that in the elderly CLNR is not fully compensatory (Meyers et al. 1987a).

The same observation was made with mezlo­cillin (Meyers et al. 1987b). While those authors noted that CL was decreased in the elderly volun­teers compared with young subjects, the CLNR was 74% of CL versus 22% in the younger age group; however, the increase in CLNR was not totally compensatory, since CL was still reduced in the elderly compared with the young. With ceftriax­one, clearance did not correlate with CLCR, sug­gesting that an increase in CLNR occurred (Nils­son-Ehle et al. 1985). The AUCs for mezlocillin and cefoperazone were also increased (Meyers et al. 1987a,b).

Changes in renal function and lack of compen-

392

satory increases in CLNR have been associated with a prolongation in the serum t'h. With a decrease in Vd, a higher plasma concentration and increased AVC are observed. Studies in infected elderly patients given ceftazidime (Ljungberg & Nilsson­Ehle 1984; Naber et al. 1983), cefotaxime (Jonsson & Walder 1986) or ceftizoxime (Horiuchi et al. 1986) reveal an increased t'h and AVC and de­creased CL and CLR compared with healthy el­derly volunteers. Studies of infected patients with decreased renal function who were being treated with cefoperazone revealed a prolonged t'h and el­evated peak and trough concentrations compared with noninfected patients (Sattler et al. 1986). It is possible that septicaemia and serious infection as­sociated with a decrease in hepatic blood flow may account for a lack of compensatory excretion by the hepatic pathway (Sattler et al. 1986). In a study of cefoperazone and sulbactam in seriously ill, in­fected elderly patients, the authors noted hyper­bilirubinaemia in 5 of 6 patients, but the free frac­tion was not measured. Others (Reitberg et al. 1984) have noted an increase in the free fraction of cef­menoxime (usually 77% protein bound) in patients who were critically ill.

An increase in AVC or Cmax and t'h should have favourable effects on antimicrobial activity, and it is believed that AVC is a good indicator of anti­bacterial action (Colaizzi et al. 1986). Other studies have suggested that the length of time for which the serum concentration is above the minimum in­hibitory concentration (MIC) for iJ-Iactam antibi­otics correlates with eradication of infection (Schentag et al. 1984). In the treatment of Gram­negative bacteraemia and other serious infections with aminoglycosides, successful therapeutic out­come correlated with the peak serum concentra­tions of the aminoglycoside (Moore et al. 1984; Wenk et al. 1984). Vogelman et al. (1988), using the neutropenic mouse thigh model, has shown that the time for which serum concentrations exceed the MIC is the most significant parameter for deter­mining iJ-Iactam efficacy; for aminoglycoside ac­tivity, the log of the AVC was the major para­meter. Drusano believes that the outcome can be optimised by increasing the time during which the

Clin. Phannacokinct. 17 (6) 1989

free drug concentration remains higher than the MIC (Drusano 1988). For maximum efficacy of iJ­lactam dosing in the elderly, regimens should be designed to maintain antibiotic concentrations above the MIC, since these compounds do not ex­hibit a post-antibiotic effect with Gram-negative organisms. This phenomenon has been described for aminoglycosides and may allow for less fre­quent administration. been noted for drugs with a short t,;" such as cefo­taxi me. In a study of acutely ill elderly patients a dose of cefotaxime Ig was given 12 hourly (Jons­son & Walder 1986); this is a significantly lower dose and longer interval than are used conven­tionally. An increase oft,/, in both serum and tissue and an increased serum AVC was noted in elderly patients compared with younger volunteers. The authors ascribed this observation to age-related im­paired renal function in the elderly. While their microbiological assay method could not distin­guish cefotaxime from its metabolite, and thereby may negate some of the pharmacokinetic data (Ljungberg et al. 1987), their patients responded to this drug regimen. Others have shown that the pharmacokinetics of ceftazidime were altered in acutely ill elderly patients: t'h was prolonged (2.7 vs 1.9 hours), AVC increased and CL reduced. Similarly, ceftriaxone pharmacokinetics in patients with acute bacterial infection revealed a reduction in CL and a prolongation of t,;, (Nilsson-Ehle et al. 1985). Vsing these studies as a background, and noting the favourable pharmacokinetics observed with antimicrobial agents, it is felt that trials should be undertaken using a lower maximum dose and an increased dosage interval in seriously ill elderly patients. Third-generation cephalosporins, carba­penems and quinolones would be ideally suited for these studies.

Ifaminoglycosides are used in the elderly it may be necessary to obtain serum concentrations to ad­equately determine dose and dosing intervals. Since a delay in tmax of up to 1.5 hours was noted for tobramycin, it seems prudent to determine Cmax

after this interval. Other parenteral compounds, such as moxalactam, may also have a delayed Cmax

(Andritz et al. 1984).

Kinetics of Antibacterials in the Elderly

Age-related physiological changes such as de­layed gastric emptying and decreased gastrointes­tinal motility in the elderly may affect the absorp­tion of orally administered compounds. There have been several recent studies of the fluoroquinolones in the elderly, including ciprofloxacin, enoxacin, ofloxacin and pefloxacin (Bayer et al. 1987; Dobbs et al. 1987; Dow & Frydman 1988; Graber et al. 1988; Humbert et al. 1985; Naber et al. 1986; Veys­sier et al. 1986; Wise et al. 1987). In these studies no delay in achieving Cmax was noted compared with younger subjects, but the Cmax and AUC were increased. The smaller Vd with the fluoroquino­lones observed in the elderly may explain these pharmacokinetics. Other factors affecting absorp­tion include congestive heart failure and concur­rent medication such as antacids containing mag­nesium and aluminium ions, which interact with the fluoroquinolones and decrease absorption.

Pharmacodynamic alterations (i.e. receptor sen­sitivity) are increased in the elderly; at any serum concentration a greater effect may be observed, es­pecially if the free drug fraction is increased. How­ever, the principal activity with antimicrobial agents is directed against offending micro-organisms and not the host cell, and so altered receptivity should be only a minor factor in any increase in adverse effects.

4. Drug-Drug Interactions

Drug-drug interactions are significant, and must be considered by prescribing physicians. Theo­phylline concentrations may be elevated in patients taking some of the fluoroquinolones (e.g. cipro­floxacin and enoxacin). While the exact mechan­ism of this phenomenon is unknown, it may be that metabolic byproducts prevent the metabolism of theophylline. Enoxacin appears to be the most potent inhibitor, and it has been suggested that the dose of theophylline be decreased by 50% when the 2 drugs are administered concurrently. Since the main route of excretion of ofloxacin is renal and it is metabolised to a smaller extent than other quinolones, this drug does not appear to be asso­ciated with theophylline interactions. Other ad-

393

verse effects of the quinolones include central nervous system changes such as restlessness, diz­ziness, lightheadedness and insomnia. It is advised that the elderly limit their intake of coffee, tea, chocolate or colas, since they contain caffeine which may not be metabolised when quinolones are given. The dose of ofloxacin should be reduced in elderly patients with decreased clearances.

5. Therapeutic Considerations

The number of residents (average age 82 years) in long-term care facilities is increasing annually. Infectious problems occur with a prevalence of 3 to 18% (Brand et al. 1986; Franson et al. 1986; Gar­ibaldi et al. 1981; Magnussen & Robb 1980; Setia et al. 1985), with 10% of patients receiving anti­biotics. Jones et al. (1987) looked at the appropri­ateness of antibiotic therapy for particular infec­tions in these facilities, using a previously developed scale. Antibiotic therapy is rated as either appro­priate, probably appropriate, inappropriate, unjus­tified, or record insufficient for categorisation. Apart from nitrofurantoin, cephalosporins received the lowest rating (27%); in most cases, this applied to oral cephalosporins used inappropriately to treat Gram-negative organisms, which are common in this setting and are often resistant to the com­monly used antibiotics such as ampicillin, amoxy­cillin, oral cephalosporins and tetracyclines. Aminoglycosides have been recommended in se­riously ill patients in a nursing home environment (Jones et al. 1987); these authors suggest that non­toxic antibiotics with a broad spectrum of activity against Gram-negative bacilli, such as aztreonam (parenteral) and the oral quinolones, may play a role (Jones et al. 1987). Ciprofloxacin is the only oral quinolone available for the treatment of sys­temic infections in the United States. Studies are necessary to compare the use of this agent with other parenteral regimens before recommendations can be given for their use as initial therapy in se­riously ill patients, but the quinolones do represent a breakthrough in terms of follow-up therapy for seriously ill patients with nosocomial pneumonia or osteomyelitis and for the treatment of compli-

394

cated urinary tract infection due to Gram-negative bacilli. Not uncommonly in the nursing home set­ting, gastrointestinal infections occur due to Sal­monella and Shigella (Anand et al. 1980; Baine et al. 1973; Gotoff et al. 1966). These agents are highly effective and may be used in the treatment of any endemic or epidemic episodes associated with these organisms.

References

Alvarez S, Shell CG, Woolley TW. Nosocomial infections in long­term facilities. Journal of Gerontology 43: M9-17, 1988

Anand CM, Finlayson Me, Grason JZ, et al. An institutional outbreak of salmonellosis due to lactose-fermenting Salmo­nella newport. American Journal of Clinical Pathology 74: 657-660. 1980

Andritz MH, Smith RP, Baltch AL, Griffin PE. Conroy JV, et al. Pharmacokinetics of moxalactam in elderly subjects. Antimi­crobial Agents and Chemotherapy 25: 33-36. 1984

Baine WB, Gangarosa EJ. Benett JV. Barker Jr WHo Institutional salmonellosis. Journal of Infectious Diseases 128: 357-360. 1973

Bayer A, Gajewska A. Stephens M, Stark JM. Pathy J. Phar­macokinetics of ciprofloxacin in the elderly. Respiration 51: 292-295, 1987

Bender AD, Post A, Meier JP, Higson JE, Reichard Jr G. Plasma protein binding of drugs as a function of age in adult human subjects. Journal of Pharmacology and Science 64: 1711-1713. 1975

Brand F, Katz PRo Beam TR. Boyce KE. Antibiotic use in the nursing home. Abstract. In Proceedings of the 43rd Annual Meeting of the American Geriatrics Society, p.14. American Geriatrics Society, Chicago, 1986

Buniva G. Cavenaghi L. Parenti F, Frigo FM. Pharmacokinetics of teicoplanin after single and repeated IV injections. Abstract No. 1248. Presented at the Twenty-Sixth Interscience Confer­ence on Antimicrobial Agents and Chemotherapy, September 28 1986, New Orleans, Louisiana

Cammarata RJ. Rodnan GP. Fennell RH. Serum anti-I'-globulin and antinuclear factors in the aged. Journal of the American Medical Association 199: 455-458, 1967

Christopher B. Vallanger R. Shepart A. A survey of hospital pre­scribing for the elderly. In Eroks & Stevenson (Eds) Drugs and the Elderly, pp. 321-328, University Park Press. Baltimore. 1979

Colaizzi PA, Coniglio AA. Poynor WJ, Vishniavsky N, et al. Comparative pharmacokinetics of two multiple-dose mezlo­cillin regimens in normal volunteers. Antimicrobial Agents and Chemotherapy 30: 675-678, 1986

Craig WA, Welling PG. Protein binding of antimicrobials: clinical pharmacokinetics and therapeutic implications. Journal of Clinical Pharmacokinetics 2: 252-268, 1977

Creasey W A, Platt TB, Frantz M. Sugerman AA. Pharmacokin­etics of aztreonam in elderly male volunteers. British Journal of Clinical Pharmacology 19: 233-237, 1985

Dobbs BR, Gazeley LR, Campbell AJ, Edwards IR. The effect of age on the pharmacokinetics of enoxacin. European Journal of Clinical Pharmacology 33: 101-104. 1987

Dow J, Frydman AM. Single- and multiple-dose pharmacokin­etics of pefloxacin in elderly patients. Reviews of Infectious Diseases 10 (Suppl. I): S107. 1988

Drusano GL. Role of pharmacokinetics in the outcome of infec­tions. Antimicrobial Agents and Chemotherapy 32: 289-297, 1988

Clin. Pharmacokinet. 17 (6) 1989

Dybkaer R. Lauritzen M. Krakauer R. Relative reference values for clinical chemical and haematological quantities in 'healthy' elderly people. Acta Medica Scandinavica 209: 1-9. 1981

Everitt DE, Avorn J. Drug prescribing for the elderly. Archives of Internal Medicine 146: 2393-2396. 1986

Faulkner RD, Bohaychuk W. Lanc RA. Haynes JD. Desjardine RE. et al. Pharmcokinetics of cefixime in the young and el­derly. Journal of Antimicrobial Chemotherapy 21: 787-794. 1988

Finch RG. Craddock e, Kelly J. Deaney NB. Pharmacokinetics of imipenem/cilastatin in elderly patients. Journal of Anti­microbial Agents and Chemotherapy 18 (Suppl. E): 103-107. 1986

Foulds G. Pharmacokinetics of sulbactam/ampicillin in humans. A review. Reviews of Infectious Diseases 8 (Suppl. 5): S503-S511. 1986

Franson TR. Duthie EH. Cooper JE. Van Oudenhoven G. Hoff­man RG. Prevalence survey of infections and their predispos­ing factors at a hospital-based nursing home care unit. Journal of the American Geriatric Society 34: 95-100. 1986

Garibaldi RA. Brodine S. Matsumiya S. Infections among patients in nursing homes: policies. prevalence. and problems. New England Journal of Medicine 305: 731-735. 1981

Gibaldi M. Gastrointestinal absorption - biologic considerations. In Biopharmaceutics and Clinical Pharmacokinetics. 3rd ed .. pp. 29. Lea & Febiger. Philadelphia. 1984

Gibson RM. Smith Mueller M. Fisher CR. Age differences in health care spending. fiscal year 1976. Social Security Bulletin 40: 3-14. 1977

Gotoff SP. Boring JR. Lepper MH. An epidemic of Salmonella St Paul infection in a convalescent home. American Journal of Medical Science 56: 16-22. 1966

Graber H. Ludwig E. Arr M. et al. Pharmacokinetics of ofloxacin in young and elderly patients. Reviews of Infectious Diseases 10 (Suppl. I): S 106. 1988

Greenblatt DJ. Sellers EM. Shader RI. Drug disposition in old age. New England Journal of Medicine 306: 1081-1088. 1982

Guay DR. Awni WM, Peterson PK. Obaid S. Stein D. et al. Single and multiple dose pharmacokinetics of oral ciprofloxacin in elderly patients. International Journal of Clinical Pharmacol­ogy Therapy and Toxicology 26: 279-284. 1988

Horiuchi I. Kawamoto T. Tokumo H. Nomura Y. Kajiyama G. et al. Therapeutic effect of ceftizoxime on infections in aged patients. Japanese Journal of Antibiotics 39: 1671-1680. 1986

Humbert. Borsa F. Couet W. et al. Pharmacokinetics ofofloxacin in healthy volunteers after oral administration. In Ishigami (Ed.) Recent advances in chemotherapy. pp. 1787-1988. University of Tokyo Press. Tokyo. 1985

Jones SR. Parker DF. Liebow ES. Kimbrough Re, Frear RS. Ap­propriateness of antibiotic therapy in long-term care facilities. American Journal of Medicine 83: 499-502, 1987

Jonsson M. Walder M. Pharmacokinetics of intravenous antibi­otics in acutely ill elderly patients. European Journal of Clinical Microbiology 5: 629-633. 1986

Kovar MG. Health of the elderly and use of health services. Pub­lic Health Reports 92: 9-19, 1977

Ljungberg B, Nilsson-Ehle I. Pharmacokinetics of ceftazidime in elderly patients and young volunteers. Scandinavian Journal of Infectious Diseases 16: 325-326. 1984

Ljungberg B, Nilsson-Ehle I. Pharmacokinetics of antimicrobial agents in the elderly. Reviews of Infectious Diseases 9: 250-264, 1987

Ljungberg B, Nilsson-Ehle I. Influence of age on the pharmaco­kinetics of ceftazidime in acutely ill adult patients. European Journal of Clinical Pharmacology 34: 173-178. 1988

Ljungberg B. Nilsson-Ehle I, Nordbring F. Pharmacokinetics of antibiotics in the elderly. European Journal of Clinical Micro­biology 6: 329. 1987

Kinetics of Antibacterials in the Elderly

Luderer JR, Patel IH, Kurkin J, Schneck DW. Age and ceftriax­one kinetics. Clinical Pharmacology and Therapeutics 35: 19-25, 1984

Ludwig E, Szekely E. Csiba A. Graber H. Pharmacokinetics of cefotaxime and desacetylcefotaxime in elderly patients. Drugs 35 (Suppl. 2): 51-56, 1988

Magnussen MH, Robb SS. Nosocomial infections in a long-term care facility. American Journal of Infection Control 8: 12-17, 1980

Massoud N. Pharmacokinetic considerations in geriatric patients. In Bennett et al. (Eds) Pharmacokinetic basis for drug treat­ment, chap. 15, pp. 283-310, Raven Press, New York, 1984

Matsuzaki Y, Kitahara Y. Hara N, et al. Clinical evaluation of cefoperazone in lower respiratory tract infections. Japanese Journal of Antibiotics 39: 2780-2794, 1986

Matzke GR, Jameson JJ. Halstenson CE. Gentamicin disposition in young and elderly patients with various degrees of renal function. Journal of Clinical Pharmacology 27: 216-220, 1987

Mayer PR, Brown CH. Carter RA, Welty TE, Millikan MD, et al. Intramuscular tobramycin pharmacokinetics in geriatric patients. Drug Intelligence and Clinical Pharmacology 20: 611-615, 1986

Meyers BR. Aminoglycosides. In Kagan (Ed.) Antimicrobial therapy, 3rd ed., pp. 56-76, WB Saunders, Philadelphia, 1980

Meyers BR, Mendelson MH, Deeter RG, Srulevitch-Chin E, Tor­tosa-Sarni M, et al. Pharmacokinetics of cefoperazone in am­bulatory elderly volunteers compared with young adults. Anti­microbial Agents and Chemotherapy 31: 925-929, 1987a

Meyers BR, Mendelson MH, Srulevitch-Chin E, Bradbury K, McMurdo L, et al. Pharmacokinetic properties of mezlocillin in ambulatory elderly subjects. Journal of Clinical Pharma­cology 27: 678-681, 1987b

Moore RD, Smith CR. Lietman PS. The association of amino­glycoside plasma levels with mortality in patients with Gram­negative bacteremia. Journal of Infectious Diseases 149: 443-448, 1984

Naber KG. Kees F, Grobecker H. Ceftazidime: pharmacokinetics in young volunteers versus elderly patients and therapeutic ef­ficacy with complicated urinary tract infections. Journal of Antimicrobial Chemotherapy 12 (Suppl): 41-45, 1983

Naber KG, Sorgel F, Gutzler F, et al. In vitro activity, phar­macokinetics, clinical safety and therapeutic efficacy of en­oxacin in the treatment of patients with complicated urinary tract infections. Infection 14 (Suppl. 3): S203-S208, 1986

Nilsson-Ehle I, Nord CE, Ursing B. Ceftriaxone: pharmacokin­etics and effect on the intestinal microflora in patients with acute bacterial infections. Scandinavian Journal of Infectious Diseases 17: 77-82, 1985

Pacifici GM, Viana A. Taddeucci-Brunelli G. et al. Plasma pro­tein binding of dicloxacillin: effects of age and disease. Inter­national Journal of Pharmacology. Therapeutics and Toxicol­ogy ll: 622-626, 1987

Reitberg DP, Cumbo TJ, Smith IL, et al. Effect of protein binding on cefmenoxime steady-state kinetics in critical patients. Jour­nal of Clinical Pharmacology and Therapeutics 35: 64-73, 1984

Ripa S, Mignini F, Patrizi J, et al. Sulbenicillin: pharmacokinetics and penetration into bronchial secretion in elderly patients. Chemioterapia 6: 277-281. 1987b

Ripa S, Mignini F, Prenna M. Pharmacokinetics of cefotetan in elderly subjects after intramuscular administration. Chemi­oterapia 6 (5): 59-63, 1987a

395

Rosina R. Villa G, Danese A, Cavenaghi L, Picardi L, et al. Phar­macokinetics of teicoplanin in the elderly. Journal of Anti­microbial Chemotherapy 21 (Suppl. A): 39-45, 1988

Rowe JW. Clinical research on aging: strategies and directions. New England Journal of Medicine 297: 1332-1336, 1977

Sattler FR, Colao DJ, Caputo GM, et al. Cefoperazone for em­piric therapy in patients with impaired renal failure. American Journal of Medicine 81: 229-236, 1986

Schmidt H, Roholt K. Penicillin serum concentrations in relation to exercise. Acta Pathologica et Microbiologica Scandinavica 68: 396-400, 1966

Schentag JJ, Smith IL, Swanson DJ, et al. Role for dual indivi­dualization with cefmenoxime. American Journal of Medicine 77 (Suppl. 6a): 43-50, 1984

Schwartz JI, Jauregui LE, Bachmann KA, Martin ME, Reitberg DP. Multiple-dose pharmacokinetics of intravenously admin­istered cefoperazone and sulbactam when given in combina­tion to infected, seriously ill, elderly patients. Antimicrobial Agents and Chemotherapy 32: 730-735, 1988

Setia U, Serventi I, Lorenz P. Nosocomial infections among patients in a long-term care facility: spectrum, prevalence, and risk factors. American Journal of Infection Control 13: 57-62, 1985

Sjovall J, Alvan G. Huitfeldt B. Intra- and inter-individual vari­ation in pharmacokinetics of intravenously infused amoxicil­lin and ampicillin to elderly volunteers. British Journal of Clinical Pharmacology 21: 171-181, 1986

Sugihara H, Yoneyama K, Sorimachi H, Kamo T, Imamura S, et al. Pharmacokinetics of cefotaxime in elderly patients. Drugs 35 (Supp!. 2): 93-96, 1988

Toon S, Hopkins KJ, Garstang FM. Aarons L, Rowland M. Phar­macokinetics of imipenem and cilastatin after their simultan­eous administration to the elderly. British Journal of Clinical Pharmacology 23: 143-149, 1987

Vestal RE. Drug use in the elderly: a review of problems and special considerations. Drugs 16: 358-382, 1978

Veyssier P, Fourtillan JB, Modai J, Couet W, Lefebvre MA, et a1. Pharmacocinetique de l'ofloxacine chez Ie sujet age (65 a 85 ans) a fonction n;,nale normale apres prise orale unique de 200mg, Pathology and Biology 34: 586-599, 1986

Vogelman B, Gudmundsson S, Turnidge J, Leggett J, Craig WA. In vivo postantibiotic effect in a thigh infection in neutropenic mice. Journal of Infectious Diseases 157: 287-298, 1988

Wallace SM, Verbeeck RK. Plasma protein binding of drugs in the elderly. Clinical Pharmacokinetics 12: 41-72. 1987

Wenk M, Vozeh S, Follath F. Serum level monitoring of anti­bacterial drugs. Clinical Pharmacokinetics 9: 475-492, 1984

Wise R, Baker SL, Misra M, Griggs D. The pharmacokinetics of enoxacin in elderly patients. Journal of Antimicrobial Chemo­therapy 19: 343-350, 1987

Yasuhara H, Kobayashi S, Sakamoto K. Kamijo K. Pharmaco­kinetics of amikacin and cephalothin in bedridden elderly patients. Journal of Clinical Pharmacology 22: 403-409, 1982

Zaske DE, Irvine P, Strand LM, Strate RG, Cipolle RJ, et al. Wide interpatient variations in gentamicin dose requirements for geriatric patients. Journal of the American Medical Asso­ciation 248: 3122-3126, 1982

Authors' address: Dr Burt R. Meyers, Box 1090, Mount Sinai Medical Center, One Gustav L. Levy Place, New York, NY 10029-6574, USA.