Infect Dis Clin N Am 18 (2004) 669689

Current use for old antibacterial agents:

polymyxins, rifampin, and aminoglycosides
Donald Kaye, MD*
Department of Medicine, Drexel University, College of Medicine,
3300 Henry Avenue, Philadelphia, PA 19129, USA

The polymyxins, rifampin, and the aminoglycosides may be considered

special-use antibacterial agents. They are all old agents and are rarely
considered the drugs of choice for common bacterial infections. The
polymyxins are important because of the recent emergence of multidrugresistant (MDR) bacteria, such as strains of Pseudomonas aeruginosa, which
are susceptible only to the polymyxins. Rifampin is only considered in
nonmycobacterial infections where its role is limited and sometimes
controversial. The aminoglycosides are discussed as to their current role
in nonmycobacterial systemic infections with an emphasis on the use of
single-daily dosage.

Polymyxins (polymyxin B and colistin)

The two parenteral polymyxins that have been available are polymyxin B
and polymyxin E (colistin, formulated as colistimethate sodium). They are
cyclic cationic polypeptide detergents with molecular weights of greater than
or equal to 1000 kDa. Polymyxin B was initially isolated from Bacillus
polymyxa, and colistin was isolated from Bacillus colistinus. The sulfomethylated formulation of colistin (colistimethate sodium) must be hydrolyzed
to be active as an antibiotic. Hydrolysis occurs both at body temperature
and in in vitro testing systems [1].
The polymyxins were discovered in 1947. They were frequently used to
treat serious infections caused by gram-negative bacilli from 1962 until
aminoglycosides active against strains of P aeruginosa (eg, gentamicin) came
into common use after the mid to late 1960s.The polymyxins fell into disuse
by 1980 because of their nephrotoxicity and subsequently have been mainly
* 1535 Sweet Briar Road, Gladwyne, PA 19035.
E-mail address: donjank@aol.com
0891-5520/04/$ - see front matter 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.idc.2004.04.007

670

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

reserved for topical and oral use [15]. The recent increasing emergence of
P aeruginosa and other gram-negative bacilli resistant to all other
antimicrobial agents, however, has resulted in the occasional need for use
of an injectable polymyxin. Colistimethate sodium is less active in vitro and
less nephrotoxic than polymyxin B [6].
These agents have been used to treat patients with bacteremia, pneumonia, and other systemic infections caused by gram-negative bacilli [1]. In
recent years, colistimethate sodium has been the most commonly used form
of polymyxin for parenteral therapy. Colistimethate has also been used for
nebulization therapy in patients with cystic brosis.
Colistin has been available as colistin sulfate for use topically and orally
and as colistimethate sodium for intramuscular and intravenous use.
Polymyxin B sulfate is available for topical use (discussed elsewhere in this
issue). There is also a parenteral preparation that can be given intramuscularly and intravenously; it has been used intrathecally and intraventricularly for central nervous system infections.

Mechanism of action and antimicrobial activity

The polymyxins are surface-active amphipathic agents containing both
lipophilic and lipophobic groups. They penetrate into cell membranes and
interact with phospholipids in the membranes, leading to permeability
changes that quickly disrupt the membranes leading to cell death. They are
rapidly bactericidal in a concentration-dependent manner and have a postantibiotic eect [6,7]. They also bind to the lipid A portion of endotoxin or
lipopolysaccharide, and in animal studies block many of the biologic eects
of endotoxin [8]. Resistance of gram-negative bacteria to the polymyxins
is most likely related to a change in architecture of the membrane [9]. The
activity of the polymyxins is diminished in the presence of high concentrations of divalent cations, such as calcium and magnesium [9].
The polymyxins are active against commonly isolated gram-negative
aerobic bacilli with the exception of Proteus spp. Proteus spp are generally
very resistant to the polymyxins. The polymyxins also have poor activity
against Providencia, Burkholderia, and Serratia spp [5,10]. When the sulfate
salt is used for testing, most gram-negative bacilli are inhibited by 1 lg/mL;
P aeruginosa strains are more resistant, but most strains are inhibited by
2 lg/mL [5]. Strains with a minimal inhibitory concentration (MIC) greater
than 4 lg/mL are generally dened as resistant; some consider strains requiring 4 lg/mL for inhibition to be resistant [5,9].
It is important to recognize that the sulfate forms of colistin and
polymyxin B (usually used for in vitro testing) are much more active in
vitro than the sulfomethyl form, often a fourfold to eightfold dierence [9].
Gram-positive organisms and most anaerobes are resistant to these agents.
The polymyxins have retained activity against many MDR gram-negative

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

671

bacilli, such as MDR P aeruginosa and Acinetobacter baumannii. There is

complete cross-resistance between the polymyxins.
Studies of in vitro synergy have been limited to relatively few organisms.
Using colistin-susceptible strains of MDR bacteria, synergism has been
demonstrated between colistin and ceftazidime, colistin and rifampin, and
colistin and trimethoprim-sulfamethoxazole [7,11].
Pharmacokinetics and pharmacodynamics
The basic and clinical evaluation of the polymyxins took place many
years ago, and the pharmacokinetic and pharmacodynamic information
available is limited. Review articles often have internally inconsistent data
concerning pharmacokinetics, especially blood levels [12].
None of the polymyxins are absorbed when given orally. According to
the old literature, following intramuscular injection of 2 to 4 mg/kg polymyxin B, peak serum levels of 1 to 8 lg/mL were achieved. There are little or
no reliable data on serum levels after intravenous administration of polymyxin B. According to the old literature, when colistimethate, 2.5 mg/kg,
was given intramuscularly to adults, a peak serum level of 5 to 7 lg/mL was
achieved. When colistimethate was given intravenously, the peak was about
20 lg/mL. The package insert for colistimethate shows peak levels of about
8 lg/mL and 20 lg/mL, respectively, following intramuscular and intravenous administration of 150 mg. The excretion of both polymyxin B and
colistin is primarily by glomerular ltration. There is a 12- to 24-hour lag
period after the initial dose of polymyxin B (but not colistimethate) before
signicant amounts appear in the urine. With subsequent doses, urine levels
of both polymyxin B and colistimethate exceed 15 lg/mL for at least 6 hours
after injection. The half-life of polymyxin B in serum is about 4.5 to 6 hours
and that of colistimethate is about 3 hours [13]; half-lives are prolonged
when the agents are given to patients with renal insuciency. Accumulation
of both drugs occurs with repeated dosing. Distribution to the cerebrospinal
uid, biliary tract, pleural uid, and joint uid is poor. In one report,
however, the peak cerebrospinal uid levels following the intravenous
administration of colistimethate were 25% of the peak serum levels [14]. The
polymyxins are poorly dialyzed.
Toxicity
Hypersensitivity is unusual. There is dose-related nephrotoxicity, damaging the convoluted tubular epithelium; this is usually reversible after
discontinuing the drug [12,15]. There is also dose-related reversible
neurotoxicity, which can be manifested by neuromuscular blockade; this
can result in muscle weakness and even apnea [12,15]. Neuromuscular
blockade is most likely to occur with drug overdosage and in patients with
renal insuciency or who are receiving curariform drugs. With respiratory

672

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

paralysis, support of respiratory function is required until the eects of the

polymyxin wear o. Aminoglycosides may potentiate the neurotoxic eects.
Perioral paresthesias, paresthesias of the tongue and extremities, peripheral
neuropathy, and other neurotoxic side eects are not uncommon [13].
Toxicity is less likely to occur with colistimethate than with polymyxin B.

Clinical indications
Polymyxin B is very painful when given intramuscularly, and this route of
administration should be avoided. If intramuscular injection is necessary,
the usual intramuscular dose is 2.5 to 3 mg/kg/d in divided doses every 4 to
6 hours. The intravenous dose is 1.5 to 2.5 mg/kg/d by continuous
intravenous infusion or in divided doses every 12 hours over a period of
60 to 90 minutes. For treatment of gram-negative bacillary meningitis,
polymyxin B has been given intrathecally in doses of 5 to 10 mg/d for the
initial 3 days of therapy and then every other day. There is a recent report of
successful intraventricular use [16].
The recommended dosage of colistimethate in patients with normal renal
function is 2.5 to 5 mg/kg (depending on the severity of infection) each day
intramuscularly or intravenously in two to four divided doses. The dose
should be based on ideal body weight and should not exceed 300 mg/d.
Doses must be decreased in patients with impaired renal function to avoid
drug accumulation and toxicity. When given intravenously, colistimethate
is administered over 3 to 5 minutes. Administration of doses as high as
8 mg/kg/d has been reported recently [17]. The question of use of colistimethate as a single daily dose theoretically to lower the risk of toxicity has been
raised but clinical validation is lacking [7].
Colistin sulfate has been used orally for intestinal decontamination. The
oral preparation is not available in the United States.
Inhalation therapy with aerosolized colistimethate has been used with
varying results to treat colonization or infection of the bronchial system
in patients with cystic brosis, especially with MDR P aeruginosa [18,19].
Systemic blood levels are not achieved with inhalation therapy.
Most of the literature on the use of the polymyxins for parenteral therapy
is old. In recent years, colistimethate has been used parenterally to treat
systemic infections caused by MDR gram-negative bacilli, mainly ventilatorassociated pneumonia [1,3,14,20]. In contrast with the old experience, recent
reports claim little in the way of nephrotoxicity or neurotoxicity [1,3,17,20].
More modern-day experience is necessary, however, before colistimethate
can be considered a relatively nontoxic systemic agent. Modern day reports
of use of polymyxin B have been sparse [21].
Recent observations have indicated emergence of resistance of P
aeruginosa to the polymyxins, probably in relationship to the use of
nebulized colistimethate in patients with cystic brosis [6,22].

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

673

Until more clinical data become available, parenteral polymyxin B and

colistimethate should be reserved for use only when no other less toxic or
potentially more eective drug can be used.

Rifampin
Rifampin was rst synthesized in 1965 from a fermentation product of
Streptomyces mediterranei. It is the 3-4-methyl-piperazinyl-iminomethyl
derivative of rifamycin. It and other rifamycin derivatives (such as rifabutin)
are valuable drugs for treatment of mycobacterial infections. This section
deals only with rifampin and its use as an agent against nonmycobacterial
organisms.
Rifampin is unusual in its ability readily to enter cells and have
antibacterial activity at intracellular sites. It also has a broad range of
antibacterial activity, but cannot be used as a single therapeutic agent
because of rapid emergence of resistance. For this reason, there are very
limited and special circumstances in which rifampin is used for nonmycobacterial infections.

Mechanism of action and antimicrobial activity

Rifampin acts by suppressing initiation of RNA synthesis by inhibiting
DNA-dependant RNA polymerase after binding to the beta subunit of
the enzyme [23]. This interferes with protein synthesis by preventing chain
initiation. Its activity is usually bactericidal, but may be bacteriostatic
depending on the organism and the concentration of drug. In general most
gram-positive cocci (including staphylococci and streptococci) are highly
susceptible (MICs \1 lg/mL) to rifampin; enterococci, which are only
moderately sensitive, are an important exception [24,25].
Meningococci, gonococci, and Hemophilus inuenzae are susceptible to
rifampin. Most gram-negative bacilli, however, are resistant. Rifampin is
highly active against Legionella spp [26].
Microorganisms inhibited by less than or equal to 1 lg/mL are
considered to be susceptible to rifampin, those inhibited by 2 lg/mL are
intermediate, and those greater than or equal to 4 lg/mL are resistant.
Resistance to rifampin often develops by mutations in the gene (rpoB) that
encodes the beta subunit of DNA-dependant RNA polymerase; in addition,
resistance can be mediated by alterations in membrane permeability [27].
The eect of adding rifampin to other antibiotics is a dicult and
complicated issue, because the results are unpredictable. When rifampin is
added to other bactericidal agents, the eect may be synergistic, additive,
indierent, or antagonistic depending on the drugs, their concentrations, the
organisms being studied, and the model. Examples of some of the results
follow.

674

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

Addition of rifampin to penicillinase-resistant penicillins in vitro inhibited the bactericidal activity of the penicillins against Staphylococcus
aureus [28]. In a study in the rabbit model of S aureus endocarditis, the
combination of a penicillinase-resistant penicillin and rifampin was synergistic for some strains and indierent or antagonistic for other strains [29].
The addition of rifampin to vancomycin or to vancomycin plus gentamicin
improved the eectiveness of therapy in a rabbit model of Staphylococcus
epidermidis endocarditis [30]. Similarly, substantial improvements in cure
rates for S epidermidis prosthetic valve endocarditis were achieved by adding
rifampin, gentamicin, or both to vancomycin [31].
Experimental meningitis studies with limited number of strains of
pneumococci have demonstrated that addition of rifampin to vancomycin
or ceftriaxone may have diering eects on the rate of killing of pneumococci
in cerebrospinal uid. One study showed a benecial eect [32]. In contrast,
another study demonstrated no increase in killing rate against pneumococci
when rifampin was added to vancomycin [33]. Interestingly, an in vitro study
showed that the addition of rifampin to ceftriaxone resulted in a marked
decrease in killing rates of ceftriaxone-susceptible pneumococci compared
with ceftriaxone alone [34].
Synergism has been demonstrated between colistin and rifampin in vitro
against strains of Stenotrophomonas maltophilia [11].
Pharmacokinetics and pharmacodynamics
Rifampin is available orally and for intravenous administration. It is well
absorbed when given orally with peak serum concentrations of 7 to 10 lg/
mL following a dose of 600 mg. Food interferes with absorption and it
should be administered with an empty stomach. Following intravenous
administration of 300 or 600 mg over 30 minutes, mean peak serum levels of
9 or 17.5 lg/mL, respectively, are achieved.
Rifampin is 80% protein bound. It is highly lipid soluble and is widely
distributed into body uids including saliva, bile, pleural uid, and
cerebrospinal uid. It also penetrates into tissues, such as liver, lung,
prostate, and bone. The cerebrospinal uid concentrations achieved with
inamed meninges are 10% to 20% of simultaneous serum concentrations.
Given the lipophilic nature of rifampin, it penetrates relatively well across
the blood-brain barrier, even in the absence of meningeal inammation.
Rifampin is excreted mainly by the liver. It is desacetylated in the liver to
an active metabolite and excreted in the bile. There is signicant enterohepatic circulation. The half-life in serum is 2 to 5 hours but becomes
shorter with repeated dosing because of enhanced biliary excretion related
to rifampins property of inducing P-450 enzymes in the liver. Most of the
drug is eventually excreted in the stool. Blood levels are elevated in the
presence of hepatic insuciency. The presence of renal insuciency does not
have much of an eect on serum levels. With doses of over 450 mg there is

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

675

a disproportionate rise in blood levels related to saturation of the excretory

capacity of the liver for rifampin.
Rifampin readily enters phagocytic cells and can kill microorganisms in
the cells. Although there is relatively little information available about the
pharmacodynamics of rifampin per se, this information is of little use
clinically. This is because rifampin cannot be used therapeutically as a sole
agent because of the rapid emergence of resistance [35]. The pharmacodynamics of rifampin are unusual; for example, there was a paradoxical
decrease in bacterial killing rate of pneumococci in spinal uid of rabbits at
doses of 20 mg/kg compared with 10 mg/kg [36]; this may have been the
result of the need for some protein synthesis for bacterial killing. Rifampin
has a very long postantibiotic eect [37].
Toxicity
With administration of rifampin, the urine and sweat may have an orange
tinge and soft contact lenses may be stained. Rash may occur in up to 5%
of patients, but the ulike syndrome associated with long-term intermittent
use is not associated with the short courses used for nonmycobacterial
infections. Gastrointestinal side eects occur in up to 5% of patients (eg,
nausea, vomiting, diarrhea, heartburn). Abnormalities in liver chemistries
are common but frank hepatitis is uncommon (\1%), especially with the
short courses used for nonmycobacterial infections.
The major problem with the use of rifampin is its potent induction of
hepatic microsomal enzymes with consequent multiple drug interactions.
Some of the drugs with altered metabolism related to rifampin are oral
contraceptives, cyclosporine, digoxin, uconazole, sulfonylureas, theophylline, thyroxine, warfarin, many of the agents used in patients with HIV, and
many others [38,39].
Clinical indications
Rifampin can be administered either orally or intravenously. It should
be given orally 1 hour before a meal. The only Food and Drug Administration (FDA) approved uses for rifampin are for therapy of mycobacterial
infections and for prophylaxis of meningococcal infection.
The only situation in which rifampin can be considered for use as a sole
antibacterial agent is in prophylaxis of infection. It is indicated for
prophylaxis of invasive meningococcal infection. It has also been used for
prevention of H inuenzae infection in children by administration to family
members who were close contacts of patients with serious infections caused
by these organisms. The principle of chemoprophylaxis is to eradicate the
organism from the nasopharynx. The widespread use of H inuenzae vaccine
has decreased the need for use for prevention of H inuenzae infection.
Resistance to rifampin was recently reported in 3% of isolates of
Neisseria meningitidis in the United States [40]. Rifampin is administered

676

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

in a dose of 10 mg/kg (not to exceed 600 mg) in patients greater than or

equal to 1 month of age, twice a day for 2 days for prevention of
meningococcal disease; this generally has an ecacy of more than 90%.
Rifampin has several shortcomings, however, including nasopharyngeal
eradication rates of only about 70% to 80% in some studies; adverse eects;
necessity for multiple doses over 2 days; and emergence of resistant
organisms (up to 10%27% of N meningitidis isolates). Ciprooxacin may
be more eective in eradicating meningococci [41]. To protect children
under 4 years of age who have been incompletely immunized against
H inuenzae type B, all household contacts who have been in contact with
a case of severe infection should receive rifampin prophylaxis. The dose in
adults is 600 mg once daily for 4 days.
Rifampin used alone intermittently has also been shown to decrease the
frequency of infections of the shunt and catheter in patients undergoing
hemodialysis and peritoneal dialysis [42,43].
The situations for which rifampin has been used in combination with
other antibacterial agents for treatment of nonmycobacterial infections are
limited. The major uses have been in staphylococcal infections especially on
foreign bodies (eg, prosthetic valve endocarditis); meningitis caused by
penicillin- or cephalosporin-resistant S pneumoniae; for infections caused by
Brucella spp; and for infection caused by Legionella spp [44]. Rifampin has
also been used alone or in combinations for therapy of some other infections
that are not discussed here, such as Q fever, chancroid, bartonellosis, and
Rhodococcus infections [44].

Staphylococcal infection
Clinical studies in serious S aureus infection treated with a combination
of rifampin and penicillinase-resistant penicillin versus the penicillin alone
have been inconclusive in terms of a benecial eect of the rifampin.
Although it is common practice to add rifampin to an antistaphylococcal
regimen when the patient is not responding, reports of better response after
addition of rifampin are dicult to evaluate.
Substantial improvements in cure rates for S epidermidis prosthetic valve
endocarditis have been reported by adding rifampin, gentamicin, or both to
vancomycin [31]. The dose of rifampin recommended by the American
Heart Association in this situation is 300 mg orally every 8 hours [45]. In
a study of right-sided S aureus endocarditis, patients were treated successfully orally with ciprooxacin plus rifampin [46]. Use of a uoroquinolone
for S aureus infection, however, is questionable.
Because of the excellent penetration of rifampin into tissues and spaces,
there is a general enthusiasm for including rifampin in regimens when
treating staphylococcal infections in the presence of abscesses or foreign
bodies. In animal models of S aureus osteomyelitis, there is a benet of
adding rifampin to the regimen [47], but clinical studies have not been

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

677

convincing. In the presence of a foreign body, however, it is reasonable to

add rifampin [48,49].
Rifampin alone or in combination with another drug (eg, trimethoprimsulfamethoxazole) has been used to eradicate nasal carriage of S aureus
including methicillin-resistant strains. These combinations have also been
used for oral treatment of infections from nonlife-threatening methicillinresistant strains.
Pneumococcal meningitis
Despite high in vitro susceptibility against Streptococcus pneumoniae,
rifampin should never be used alone to treat pneumococcal meningitis.
When the infecting organism is highly resistant to ceftriaxone and
cefotaxime, but is rifampin susceptible, combination therapy with vancomycin plus ceftriaxone plus rifampin might be eective [32]. In these
circumstances, vancomycin concentrations should be maximized to ensure
adequate central nervous system penetration. Steroid therapy can reduce
antimicrobial penetration into the cerebrospinal uid by decreasing inammation; in animal models, steroid therapy has been associated with
failure of cephalosporin or vancomycin monotherapy, although this has not
been observed in all experimental studies. The combination of these agents
with rifampin was an eective regimen, however, even when given concomitantly with steroids [33,50,51].
There are no reliable clinical data concerning results of the addition of
rifampin to vancomycin or a third-generation cephalosporin in patients with
pneumococcal meningitis. Because there are conicting results of the
benets of rifampin in in vitro and animal model studies, it has been
suggested that rifampin be used for patients in whom the pneumococcal
isolate is sensitive to rifampin in vitro and who are not responding
appropriately to standard therapy (discussed elsewhere in this issue).
Brucellosis
Rifampin in combination with doxycycline has been used successfully to
treat brucellosis [52,53].
Legionellosis
Legionella pneumophila and micdadei are highly sensitive to rifampin in
vitro, and it has been suggested that rifampin be added to a macrolide in
treatment of infections that are responding poorly to the macrolides alone
[54]. There are no convincing clinical data, however, to support this.

Aminoglycosides
The aminoglycosides are a class of bactericidal antibiotics characterized
by the presence of a six-carbon aminocyclitol ring, covalently bonded to

678

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

multiple amino sugar groups. Aminoglycosides were once widely used as

primary agents in the therapy of gram-negative bacillary infections.
Because of their toxicity and the availability of newer eective agents,
however, their use has signicantly diminished. The current use of systemic
aminoglycosides in adults is primarily as companion drugs to either
broaden coverage against gram-negative aerobic bacilli or to provide
synergistic killing against gram-positive cocci or certain gram-negative
bacilli.
The major change in the use of the aminoglycosides has been a trend
toward single-daily dosage. FDA-approved dosing regimens for aminoglycoside antibiotics require division of the total daily dose into multiple doses in
individuals with normal renal function. Improvements in the understanding
of the pharmacodynamics of aminoglycosides and mechanisms of toxicity,
however, have prompted the evaluation and use of once-daily dosing
regimens. Gentamicin, the most widely administered of the aminoglycosides
and the most studied, is used as the prototype for discussion.
Mechanism of action and antimicrobial activity
The aminoglycosides act in part by impairing bacterial protein synthesis
through irreversible binding to the 30S subunit of the bacterial ribosome.
They are rapidly bactericidal against a broad range of aerobic gramnegative bacilli (including P aeruginosa), but lack activity against anaerobes.
Although there is activity against some gram-positive aerobic cocci, it is
unreliable. The aminoglycosides are generally used against gram-positive
cocci, such as enterococci and staphylococci, only in combination with cellwall active antibiotics to achieve synergistic bactericidal activity.
Resistance to the aminoglycosides in normally susceptible bacteria is
usually by means of bacterial elaboration of aminoglycoside-inactivating
enzymes. These enzymes are widely distributed in both gram-positive and
gram-negative bacteria, and usually are plasmid-mediated [55]. Perhaps the
most clinically signicant resistance occurs in enterococci where the
synergistic killing activity of gentamicin is required for a high cure rate of
endocarditis. Although all enterococci are intrinsically resistant to low
concentrations of aminoglycosides, the clinical problem arises with highlevel gentamicin-resistant enterococci [56]. High-level gentamicin-resistant
enterococci are highly resistant to all other aminoglycosides in clinical use
in the United States, with the possible exception of streptomycin, and do
not demonstrate synergistic killing of enterococci when aminoglycosides
are combined with penicillin or vancomycin [57]. Detection of highlevel gentamicin resistance requires either special susceptibility wells or
screening plates with high concentrations of gentamicin or streptomycin
(eg, 500 lg/mL) (discussed elsewhere in this issue).
When aminoglycosides are combined with cell-wall active agents, such as
b-lactam drugs or vancomycin, there is a synergistic killing eect against

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

679

susceptible bacteria (aerobic gram-negative bacilli and gram-positive cocci).

This phenomenon has been demonstrated with multiple organisms and in
multiple animal models [58]. This activity has been best demonstrated to be
important clinically in infective endocarditis caused by enterococci and in S
epidermidis endocarditis on a prosthetic valve. It has been suggested that
contribution to this activity may also be important in serious P aeruginosa
infections, and in infections caused by gram-negative bacilli in neutropenic
patients.
Pharmacokinetics and pharmacodynamics
Aminoglycosides are poorly absorbed orally. After intravenous infusion
of 1.7 mg/kg gentamicin (every 8-hour dosage), peak serum levels (Cmax) are
4 to 10 lg/mL. After infusion of 5 mg/kg (once-daily dosage), peak levels are
about 20 lg/mL serum. The half-life of all of the aminoglycosides is about 2
to 3 hours. Protein binding is low (\10%) and because the agents are water
soluble, they are distributed in the intravascular space and into interstitial
uid. The drugs diuse into synovial, pleural, and peritoneal uids, but
penetration into the cerebrospinal uid and into bile is poor.
In general, in the absence of large eusions and edema, the volume of
distribution is low. Increases in the volume of distribution tend to decrease
the Cmax and area under the serum concentration-time curve (AUC), and
increases in clearance tend to decrease the AUC. For example, the volume
of distribution tends to be elevated and peak serum levels decreased in
patients with large eusions, fever, burns, or congestive heart failure and in
critically ill patients [59,60]. Excretion of aminoglycosides is primarily by
glomerular ltration; clearance is decreased with renal insuciency and increased in children, in pregnancy, and in patients with cystic brosis.
The pharmacodynamics of aminoglycosides has been extensively studied
(discussed elsewhere in this issue). Bacterial killing by aminoglycosides in
vitro is concentration dependent [6164]. In vivo, higher doses of the drug
not only increase the rate of reduction of bacteria, but also the length of
time of drug exposure to bactericidal concentrations as measured by the
AUC. In animal studies, aminoglycoside ecacy is correlated with the ratio
of Cmax to the MIC of the infecting organism [65] and also with the AUC
[66]. These pharmacodynamic properties stand in contrast to those of blactam antibiotics, whose ecacy correlates best with time above MIC [66].
In animal studies, eective dosing regimens for concentration-dependent
antibiotics require that either the 24-hour AUC/MIC be 80 to 100 against
gram-negative bacilli or the Cmax/MIC of the causative pathogen be 8 to 10
[63].
Some clinical studies have demonstrated a relationship between peak
aminoglycoside concentrations and response to therapy [67,68]. For
concentration-dependent drugs, such as aminoglycosides, giving the total
daily dose as a single dose every 24 hours rather than giving smaller divided

680

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

doses maximizes the Cmax and possibly allows for comparable or better
ecacy at greater convenience and lower cost. A Cmax/MIC of greater than
or equal to 10 seems to be a reasonable goal [63,68].
Aminoglycoside antibiotics have a signicant postantibiotic eect for
staphylococci and gram-negative bacilli, meaning that suppression of
bacterial growth persists despite concentrations of antibiotic below the
MIC [66,6971]. This permits longer intervals before another dose of
aminoglycoside is given. The postantibiotic eect tends to be longer in vivo
than in vitro even in neutropenic animals [72]. The postantibiotic eect is
extended by higher doses of aminoglycosides [65,71] and concurrent administration of a cell-wall active antibiotic [72].
Enhanced phagocytosis of aminoglycoside-exposed bacteria by host
leukocytes has also been observed in vitro [73]. This has been referred to
as postantibiotic leukocyte enhancement. The absence of postantibiotic
leukocyte enhancement may help explain the decreased ecacy of oncedaily dosing that has been described in neutropenic animals as compared
with nonneutropenic animals [71].
The term adaptive resistance refers to the transient reduction in rate of
bacterial killing by an antibiotic following pre-exposure to that drug [74].
Adaptive resistance to aminoglycoside agents has been observed principally
in P aeruginosa, but also in other gram-negative bacilli [75]. Once-daily
dosing may circumvent the possibility of adaptive resistance through the
provision of a drug-free interval.
Toxicity
The well known major toxic side eects of the aminoglycosides are for
the renal proximal convoluted tubules, the cochlear and vestibular bodies of
the ear, and for the neuromuscular junction. The nephrotoxicity is usually
reversible as is the neuromuscular blockade [76].
The uptake of aminoglycosides by renal cortical cells seems to be
a saturable process [76]. Once proximal tubular cells are saturated with
aminoglycoside, the higher peak serum concentrations seen with once-daily
dosing should not cause greater intracellular accumulation of drug than is
seen with multiple lower doses. In fact, renal cortical accumulation of
gentamicin in rodents is lower following once-daily dosing regimens than
after multiple-daily doses, and greater increases in serum creatinine are seen
with multiple-daily dosing regimens [77].
Correlation of increased renal cortical accumulation of aminoglycosides
with dosing frequency rather than peak serum concentrations has been
demonstrated in a cohort of human subjects undergoing nephrectomy for
cancer [78]. Increased nephrotoxicity has been observed with elevated trough
levels; with prolonged use; in patients with diabetes mellitus; with concurrent
use of vancomycin, loop diuretics, cyclosporine, and cisplatin; and when the
drug is administered at time of rest as opposed to activity [7982].

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

681

There have been many clinical studies of nephrotoxicity of aminoglycosides. A recent review [63] compiled a summary of the results of metaanalyses of once-daily dosing of aminoglycosides versus multiple-daily
dosing. Although some of these reported a decrease in nephrotoxicity with
once-daily dosing, most of the meta-analyses demonstrated no signicant
decrease in nephrotoxicity.
Aminoglycoside therapy may result in both vestibular and cochlear
toxicity. In rats, (but not in guinea pigs) gentamicin uptake into perilymph,
endolymph, and inner ear tissues seems to be a saturable process, as in the
kidney [83]. It is unknown if gentamicin uptake is a saturable process in
humans, and it is unclear whether aminoglycoside accumulation in the inner
ear predicts ototoxicity [83].
Risk of ototoxicity has been associated with elevated trough levels of
aminoglycoside, although some authors have found no association between
drug levels and ototoxicity after controlling for age [84]. Nonetheless, the
suggestion that aminoglycoside uptake into inner ear uids is saturable has
raised the possibility that once-daily dosing of aminoglycosides may decrease ototoxicity. Most published meta-analyses, however, have reported no
dierence in ototoxicity between once- and multiple-daily dosing [63].
Clinical use
The indications for parenteral use of aminoglycosides have not changed
in many years. Their use in mycobacterial infections and in infections caused
by unusual pathogens, such as Yersinia pestis, Brucella spp, and Francisella
tularensis, is beyond the scope of this article.
Outside of these uses aminoglycosides are generally reserved for serious
gram-negative bacillary infections, almost always in combination with other
antimicrobial agents (usually b-lactam antibiotics). They are included to
broaden coverage against resistant gram-negative bacilli, such as P
aeruginosa, and to provide potential synergistic bactericidal activity against
these organisms. They are also used in combination with cell-wall active
antibiotics to provide synergistic bactericidal activity in treatment of serious
gram-positive coccal infections, such as staphylococcal, enterococcal, and
streptococcal endocarditis [85].
The major change in use of aminoglycosides has been the consideration of
use of single-daily dosage versus divided-daily doses. For example, the daily
dose of gentamicin is usually 5 mg/kg. It can be given as 1.7 mg/kg
intramuscularly or intravenously every 8 hours (FDA approved) or as 5 mg/
kg in one single-daily intravenous dose over 30 to 60 minutes. Once-daily
dosing should be performed in a manner that ensures a high Cmax/MIC ratio
for the pathogen being treated and ensures that trough levels are low enough
to minimize toxicity and permit the loss of adaptive resistance. Higher rates of
bactericidal action result in lower residual bacterial counts and longer intervals
before signicant regrowth occurs. Maximizing serum concentrations of

682

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

aminoglycosides by using the single large dose maximizes the rate and extent
of bactericidal activity.
The possibility of increased ecacy with once-daily aminoglycoside
dosing has been evaluated in a multitude of randomized clinical trials. In
general, these individual trials have lacked sucient power to determine
whether observed dierences in ecacy between once-daily and multipledaily dosage regimens were caused by chance alone. Meta-analysis has been
used to synthesize the data contained in these multiple studies. Many of
these [8694] were reviewed in a recent publication [63]. Most of these metaanalyses demonstrated an improvement (albeit usually small) in the clinical
ecacy of once-daily aminoglycoside dosing.
Overall, existing clinical data suggest that once-daily administration of
aminoglycosides may be more ecacious and less nephrotoxic, or at least
not more nephrotoxic, than multiple-daily dosing. Giving the total 24-hour
dose as a single dose, rather than in smaller divided doses, and using
extended dosing intervals has now become the standard in many clinical
settings [63]. This strategy may be especially appropriate for treatment of
infections caused by borderline susceptible pathogens (eg, P aeruginosa),
with MICs that are close to the breakpoint [95].
Dosages used in clinical trials of once-daily administration of gentamicin,
netilmicin, and tobramycin have varied from 4 to 7 mg/kg/d [95]. A starting
dose of 5 mg/kg has been suggested for adults 30 to 60 years of age with
doses of 6 mg/kg for adults below 30 years and 4 mg/kg for those over 60
[63]. It has been pointed out that, because of patient-to-patient variability,
4 mg/kg dosing does not always ensure suciently high peak concentrations
to provide optimal activity against such organisms as P aeruginosa [95]. The
use of 7 mg/kg dosing of gentamicin in 2148 adult inpatients was associated
with only 27 incidents of nephrotoxicity and three of vestibulotoxicity;
however, the median duration of therapy was only 3 days [95]. It has been
suggested that once-daily doses should be based on FDA-approved
maximum daily doses of aminoglycoside antibiotics (ie, 5 mg/kg/d for
gentamicin) [96].
Once-daily dosage should not be used with creatinine clearances of less
than 20 mL/min. Two approaches have been advanced for initial dosing of
once-daily aminoglycosides in individuals with mild to moderate renal
impairment: reduction of drug dosage, and increase in dosing interval. Prins
et al [97] suggested that gentamicin and tobramycin be administered at
a dose of 4 mg/kg/d when estimated creatinine clearance is greater than
80 mL/min, and that the daily dose be decreased when renal function is
compromised. Nicolau et al [95] have suggested that a xed 7 mg/kg dose be
administered, and that the dosing interval be extended beyond 24 hours in
individuals with impaired renal function. Recommendations on dosing with
renal insuciency are discussed in more detail elsewhere in this issue. When
renal function is decreased, it is important subsequently to monitor aminoglycoside serum levels.

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

683

The doses of aminoglycosides used for synergistic bactericidal activity in

the treatment of gram-positive coccal endocarditis have been lower than
those used for serious gram-negative infections (target serum peak levels
34 mg/L). Furthermore, although a signicant postantibiotic eect has been
demonstrated against enterococci with aminoglycosideb-lactam combinations in vitro, no postantibiotic eect was demonstrated in an animal model
of enterococcal endocarditis [85]. Some have found multiple-daily dosing
regimens to be superior to once-daily regimens in the treatment of experimental enterococcal endocarditis [98,99].
Once-daily gentamicin dosing for streptococcal endocarditis was used in
a clinical trial, comparing 4 weeks of ceftriaxone with 2 weeks of ceftriaxone
combined with gentamicin; the two regimens were equivalent [100]. At
present, the use of once-daily dosing of aminoglycosides for the treatment of
gram-positive coccal endocarditis remains controversial, and further data
are needed before such regimens can be recommended.
Aminoglycoside antibiotics are commonly used in the treatment of
pulmonary exacerbations in individuals with cystic brosis and P aeruginosa
colonization. Antibiotic treatment of such exacerbations is typically
prolonged, and often occurs in the home setting; once-daily dosing for
home therapy has obvious advantages. Individuals with cystic brosis clear
aminoglycosides at an increased rate. Higher doses of aminoglycosides (eg,
tobramycin at doses up to 15 mg/kg/d) once daily have been used
successfully, with little or no increase in nephrotoxicity [101105], although
transient ototoxicity has been associated with these doses [102].
Several dierent methods of monitoring once-daily dosing aminoglycoside have been suggested. Some authors have suggested monitoring trough
concentrations only, with dose adjustments for troughs greater than 2 lg/
mL. This approach does not allow clinicians to appreciate underdosage of
drug, however, and trough levels of 2 lg/mL are too high and indicate
reduced clearance with once-daily dosing.
Another proposed approach to monitoring of once-daily dosage is with
a drug level taken 6 to 14 hours after infusion and adjusting the doses
according to a nomogram [93,95]. Suboptimal peaks are not recognized by
this method. A third approach measures two levels (at 0.51 hour and 614
hours) and calculates the AUC [106,107].
It has been proposed that aminoglycoside levels need not be measured
with once-daily dosing in patients with creatinine clearance greater than
60 mL/min who receive less than 5 days of therapy [95,96]. Levels should
be checked in patients at increased risk for toxicity or clinical failure,
however, such as the elderly, those receiving other nephrotoxic drugs,
those with severe infections, or those expected to require longer than 5
days of treatment. In patients on long-term therapy, aminoglycoside levels
should be measured weekly as long as the serum creatinine is stable
[95,96].

684

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

Summary
This article reviews three classes of antibacterial agents that are uncommonly used in bacterial infections (other than mycobacterial infections)
and can be thought of as special-use agents. The polymyxins are reserved for
gram-negative bacilli that have become resistant to virtually all other classes
of drugs. Rifampin is used therapeutically, mainly as a companion drug in
treatment of refractory gram-positive coccal infections especially involving
foreign bodies. The aminoglycosides are used mainly as companion drugs
for the therapy of resistant gram-negative bacillary infection and for grampositive coccal endocarditis. The major change in use of aminoglycosides
has been a shift to once-daily dosing in many situations.

References
[1] Levin AS, Barone AA, Penco J, et al. Intravenous colistin as therapy for nosocomial
infections caused by multidrug-resistant Pseudomonas aeruginosa and Acinetobacter
baumannii. Clin Infect Dis 1999;28:100811.
[2] Wolinsky E, Hines JD. Neurotoxic and nephrotoxic eects of colistin in patients with
renal disease. N Engl J Med 1962;266:75962.
[3] Garnacho-Montero J, Ortiz-Leyba C, Jimenez-Jimenez FJ, et al. Treatment of multidrugresistant Acinetobacter baumannii ventilator-associated pneumonia (VAP) with intravenous colistin: a comparison with imipenem-susceptible VAP. Clin Infect Dis 2003;36:
11118.
[4] Koch-Weser J, Sidel VW, Federman EB, et al. Adverse eects of sodium colistimethate:
manifestations and specic reaction rates during 317 courses of therapy. Ann Intern Med
1970;72:85768.
[5] Gales AC, Reis AO, Jones RN. Contemporary assessment of antimicrobial susceptibility
testing methods for polymyxin B and colistin: review of available interpretative criteria
and quality control guidelines. J Clin Microbiol 2001;39:18390.
[6] Li J, Turnidge J, Milne R, et al. In vitro pharmacodynamic properties of colistin and
colistin methanesulfonate against Pseudomonas aeruginosa isolates from patients with
cystic brosis. Antimicrob Agents Chemother 2001;45:7815.
[7] Gunderson BW, Ibrahim KH, Hovde LB, et al. Synergistic activity of colistin and
ceftazidime against multiantibiotic-resistant Pseudomonas aeruginosa in an in vitro
pharmacodynamic model. Antimicrob Agents Chemother 2003;47:9059.
[8] From AH, Fong JS, Good RA. Polymyxin B sulfate modication of bacterial endotoxin:
eects on the development of endotoxin shock in dogs. Infect Immun 1979;23:6604.
[9] Hermsen ED, Sullivan CJ, Rotschafer JC. Polymyxins: pharmacology, pharmacokinetics,
pharmacodynamics, and clinical applications. Infect Dis Clin North Am 2003;17:54562.
[10] Catchpole CR, Andrews JM, Brenwald N, et al. A reassessment of the in-vitro activity of
colistin sulphomethate sodium. J Antimicrob Chemother 1997;39:25560.
[11] Giamarellos-Bourboulis EJ, Karnesis L, Giamarellou H. Synergy of colistin with
rifampin and trimethoprim/sulfamethoxazole on multidrug-resistant Stenotrophomonas
maltophilia. Diagn Microbiol Infect Dis 2002;44:25963.
[12] Evans ME, Feola DJ, Rapp RP. Polymyxin B sulfate and colistin: old antibiotics for
emerging multiresistant gram-negative bacteria. Ann Pharmacother 1999;33:9607.
[13] Reed MD, Stern RC, ORiordan MA, et al. The pharmacokinetics of colistin in patients
with cystic brosis. J Clin Pharmacol 2001;41:64554.

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

685

[14] Jimenez-Mejias ME, Pichardo-Guerrero C, Marquez-Rivas FJ, et al. Cerebrospinal uid

penetration and pharmacokinetic/pharmacodynamic parameters of intravenously administered colistin in a case of multidrug-resistant Acinetobacter baumannii meningitis. Eur J
Clin Microbiol Infect Dis 2002;21:2124.
[15] Bosso JA, Liptak CA, Seilheimer DK, et al. Toxicity of colistin in cystic brosis patients.
Drug Intelligence and Pharmacology: Annals of Pharmacotherapy 1991;25:116870.
[16] Segal-Maurer S, Mariano N, Qavi A, et al. Successful treatment of ceftazidime-resistant
Klebsiella pneumoniae ventriculitis with intravenous meropenem and intraventricular
polymyxin B: case report and review. Clin Infect Dis 1999;28:11348.
[17] Beringer P. The clinical use of colistin in patients with cystic brosis. Curr Opin Pulm
Med 2001;7:43440.
[18] Hodson ME, Gallagher CG, Govan JR. A randomised clinical trial of nebulised
tobramycin or colistin in cystic brosis. Eur Respir J 2002;20:65864.
[19] Hamer DH. Treatment of nosocomial pneumonia and tracheobronchitis caused by
multidrug-resistant Pseudomonas aeruginosa with aerosolized colistin. Am J Respir Crit
Care Med 2000;162:32830.
[20] Conway SP, Pond MN, Watson A, et al. Intravenous colistin sulphomethate in acute
respiratory exacerbations in adult patients with cystic brosis. Thorax 1997;52:98793.
[21] Ouderkirk JP, Nord JA, Turett GS, et al. Polymyxin B nephrotoxicity and ecacy against
nosocomial infections caused by multiresistant gram-negative bacteria. Antimicrob
Agents Chemother 2003;47:265962.
[22] Livermore DM. Multiple mechanisms of antimicrobial resistance in Pseudomonas
aeruginosa: our worst nightmare? Clin Infect Dis 2002;34:63440.
[23] Wehrli W, Knusel F, Schmid K, et al. Interaction of rifamycin with bacterial RNA
polymerase. Proc Natl Acad Sci U S A 1968;61:66773.
[24] Thornsberry C, Hill BC, Swenson JM, et al. Rifampin: spectrum of antibacterial activity.
Rev Infect Dis 1983;5(Suppl 3):S4127.
[25] Pohlod DJ, Saravolatz LD, Somerville MM. In-vitro susceptibility of gram-positive cocci
to LY146032 teicoplanin, sodium fusidate, vancomycin, and rifampicin. J Antimicrob
Chemother 1987;20:197202.
[26] Schulin T, Wennersten CB, Ferraro MJ, et al. Susceptibilities of Legionella spp. to newer
antimicrobials in vitro. Antimicrob Agents Chemother 1998;42:15203.
[27] Klugman KP, Madhi SA. Emergence of drug resistance: impact on bacterial meningitis.
Infect Dis Clin North Am 1999;13:63746.
[28] Watanakunakorn C, Tisone JC. Antagonism between nafcillin or oxacillin and rifampin
against Staphylococcus aureus. Antimicrob Agents Chemother 1982;22:9202.
[29] Zak O, Scheld WM, Sande MA. Rifampin in experimental endocarditis due to
Staphylococcus aureus in rabbits. Rev Infect Dis 1983;5(Suppl 3):S48190.
[30] Kobasa WD, Kaye KL, Shapiro T, et al. Therapy for experimental endocarditis due to
Staphylococcus epidermidis. Rev Infect Dis 1983;5(Suppl 3):S5337.
[31] Karchmer AW, Archer GL, Dismukes WE. Staphylococcus epidermidis causing prosthetic
valve endocarditis: microbiologic and clinical observations as guides to therapy. Ann
Intern Med 1983;98:44755.
[32] Friedland IR, Paris M, Ehrett S, et al. Evaluation of antimicrobial regimens for treatment
of experimental penicillin- and cephalosporin-resistant pneumococcal meningitis. Antimicrob Agents Chemother 1993;37:16306.
[33] Martinez-Lacasa J, Cabellos C, Martos A, et al. Experimental study of the ecacy of
vancomycin, rifampicin and dexamethasone in the therapy of pneumococcal meningitis.
J Antimicrob Chemother 2002;49:50713.
[34] Fitoussi F, Doit C, Geslin P, et al. Killing activities of trovaoxacin alone and in
combination with beta-lactam agents, rifampin, or vancomycin against Streptococcus
pneumoniae isolates with various susceptibilities to extended-spectrum cephalosporins at

686

[35]
[36]
[37]
[38]
[39]
[40]

[41]

[42]
[43]

[44]
[45]

[46]

[47]

[48]
[49]
[50]
[51]

[52]

[53]

[54]
[55]

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

concentrations clinically achievable in cerebrospinal uid. Antimicrob Agents Chemother
1999;43:23725.
Spratt BG. Resistance to antibiotics mediated by target alterations. Science 1994;264:
38893.
Nau R, Kaye K, Sachdeva M, et al. Rifampin for therapy of experimental pneumococcal
meningitis in rabbits. Antimicrob Agents Chemother 1994;38:11869.
Bundtzen RW, Gerber AU, Cohn DL, et al. Postantibiotic suppression of bacterial
growth. Rev Infect Dis 1981;3:2837.
Niemi M, Backman JT, Fromm MF, et al. Pharmacokinetic interactions with rifampicin:
clinical relevance. Clin Pharmacokinet 2003;42:81950.
Finch CK, Chrisman CR, Baciewicz AM, et al. Rifampin and rifabutin drug interactions:
an update. Arch Intern Med 2002;162:98592.
Rosenstein NE, Stocker SA, Popovic T, et al. Antimicrobial resistance of Neisseria
meningitidis in the United States, 1997. The Active Bacterial Core Surveillance (ABCs)
Team. Clin Infect Dis 2000;30:2123.
Cuevas LE, Kazembe P, Mughogho GK, et al. Eradication of nasopharyngeal carriage of
Neisseria meningitidis in children and adults in rural Africa: a comparison of ciprooxacin
and rifampicin. J Infect Dis 1995;171:72831.
Yu VL, Goetz A, Wagener M, et al. Staphylococcus aureus nasal carriage and infection in
patients on hemodialysis: ecacy of antibiotic prophylaxis. N Engl J Med 1986;315:916.
Zimmerman SW, Ahrens E, Johnson CA, et al. Randomized controlled trial of
prophylactic rifampin for peritoneal dialysis-related infections. Am J Kidney Dis 1991;
18:22531.
Morris AB, Brown RB, Sands M. Use of rifampin in nonstaphylococcal, nonmycobacterial disease. Antimicrob Agents Chemother 1993;37:17.
Wilson WR, Karchmer AW, Dajani AS, et al. Antibiotic treatment of adults with
infective endocarditis due to streptococci, enterococci, staphylococci, and HACEK microorganisms. American Heart Association. JAMA 1995;274:170613.
Heldman AW, Hartert TV, Ray SC, et al. Oral antibiotic treatment of right-sided
staphylococcal endocarditis in injection drug users: prospective randomized comparison
with parenteral therapy. Am J Med 1996;101:6876.
Norden CW. Experimental chronic staphylococcal osteomyelitis in rabbits: treatment
with rifampin alone and in combination with other antimicrobial agents. Rev Infect Dis
1983;5(Suppl 3):S4914.
Norden CW, Fierer J, Bryant RE. Chronic staphylococcal osteomyelitis: treatment with
regimens containing rifampin. Rev Infect Dis 1983;5(Suppl 3):S495501.
Widmer AF, Gaechter A, Ochsner PE, et al. Antimicrobial treatment of orthopedic
implant-related infections with rifampin combinations. Clin Infect Dis 1992;14:12513.
Friedland IR, McCracken GH Jr. Management of infections caused by antibioticresistant Streptococcus pneumoniae. N Engl J Med 1994;331:37782.
Paris MM, Hickey SM, Uscher MI, et al. Eect of dexamethasone on therapy of
experimental penicillin- and cephalosporin-resistant pneumococcal meningitis. Antimicrob Agents Chemother 1994;38:13204.
Akova M, Uzun O, Akalin HE, et al. Quinolones in treatment of human brucellosis:
comparative trial of ooxacin-rifampin versus doxycycline-rifampin. Antimicrob Agents
Chemother 1993;37:18314.
Ariza J, Gudiol F, Pallares R, et al. Treatment of human brucellosis with doxycycline plus
rifampin or doxycycline plus streptomycin: a randomized, double-blind study. Ann Intern
Med 1992;117:2530.
Roig J, Domingo C, Morera J. Legionnaires disease. Chest 1994;105:181725.
Davies J, Wright GD. Bacterial resistance to aminoglycoside antibiotics. Trends
Microbiol 1997;5:23440.

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

687

[56] Moellering RC Jr. The Garrod Lecture. The enterococcus: a classic example of the impact
of antimicrobial resistance on therapeutic options. J Antimicrob Chemother 1991;28:112.
[57] Krogstad DJ, Korfhagen TR, Moellering RC Jr, et al. Aminoglycoside-inactivating
enzymes in clinical isolates of Streptococcus faecalis: an explanation for resistance to antibiotic synergism. J Clin Invest 1978;62:4806.
[58] Fantin B, Carbon C. In vivo antibiotic synergism: contribution of animal models.
Antimicrob Agents Chemother 1992;36:90712.
[59] Buijk SE, Mouton JW, Gyssens IC, et al. Experience with a once-daily dosing program of
aminoglycosides in critically ill patients. Intensive Care Med 2002;28:93642.
[60] Tod M, Lortholary O, Seytre D, et al. Population pharmacokinetic study of amikacin
administered once or twice daily to febrile, severely neutropenic adults. Antimicrob
Agents Chemother 1998;42:84956.
[61] Drusano GL. Human pharmacodynamics of beta-lactams, aminoglycosides and their
combination. Scand J Infect Dis Suppl 1990;74:23548.
[62] Lacy MK, Nicolau DP, Nightingale CH, et al. The pharmacodynamics of aminoglycosides. Clin Infect Dis 1998;27:237.
[63] Turnidge J. Pharmacodynamics and dosing of aminoglycosides. Infect Dis Clin North
Am 2003;17:50328.
[64] Maglio D, Nightingale CH, Nicolau DP. Extended interval aminoglycoside dosing: from
concept to clinic. Int J Antimicrob Agents 2002;19:3418.
[65] Craig WA, Redington J, Ebert SC. Pharmacodynamics of amikacin in vitro and in mouse
thigh and lung infections. J Antimicrob Chemother 1991;27(Suppl C):2940.
[66] Vogelman B, Gudmundsson S, Leggett J, et al. Correlation of antimicrobial
pharmacokinetic parameters with therapeutic ecacy in an animal model. J Infect Dis
1988;158:83147.
[67] Moore RD, Lietman PS, Smith CR. Clinical response to aminoglycoside therapy:
importance of the ratio of peak concentration to minimal inhibitory concentration.
J Infect Dis 1987;155:939.
[68] Kashuba AD, Nafziger AN, Drusano GL, et al. Optimizing aminoglycoside therapy for
nosocomial pneumonia caused by gram-negative bacteria. Antimicrob Agents Chemother
1999;43:6239.
[69] Craig WA, Vogelman B. The postantibiotic eect. Ann Intern Med 1987;106:9002.
[70] Craig WA. Post-antibiotic eects in experimental infection models: relationship to invitro phenomena and to treatment of infections in man. J Antimicrob Chemother 1993;
31(Suppl D):14958.
[71] Kapusnik JE, Hackbarth CJ, Chambers HF, et al. Single, large, daily dosing versus
intermittent dosing of tobramycin for treating experimental pseudomonas pneumonia. J
Infect Dis 1988;158:712.
[72] Gudmundsson S, Einarsson S, Erlendsdottir H, et al. The post-antibiotic eect of
antimicrobial combinations in a neutropenic murine thigh infection model. J Antimicrob
Chemother 1993;31(Suppl D):17791.
[73] McDonald PJ, Wetherall BL, Pruul H. Postantibiotic leukocyte enhancement: increased
susceptibility of bacteria pretreated with antibiotics to activity of leukocytes. Rev Infect
Dis 1981;3:3844.
[74] Barclay ML, Begg EJ. Aminoglycoside adaptive resistance: importance for eective
dosage regimens. Drugs 2001;61:71321.
[75] Daikos GL, Lolans VT, Jackson GG. First-exposure adaptive resistance to aminoglycoside antibiotics in vivo with meaning for optimal clinical use. Antimicrob Agents
Chemother 1991;35:11723.
[76] Swan SK. Aminoglycoside nephrotoxicity. Semin Nephrol 1997;17:2733.
[77] Bennett WM, Plamp CE, Gilbert DN, et al. The inuence of dosage regimen on
experimental gentamicin nephrotoxicity: dissociation of peak serum levels from renal
failure. J Infect Dis 1979;140:57680.

688

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

[78] Verpooten GA, Giuliano RA, Verbist L, et al. Once-daily dosing decreases renal
accumulation of gentamicin and netilmicin. Clin Pharmacol Ther 1989;45:227.
[79] Rybak MJ, Abate BJ, Kang SL, et al. Prospective evaluation of the eect of an
aminoglycoside dosing regimen on rates of observed nephrotoxicity and ototoxicity.
Antimicrob Agents Chemother 1999;43:154955.
[80] Beauchamp D, Labrecque G. Aminoglycoside nephrotoxicity: do time and frequency of
administration matter? Curr Opin Crit Care 2001;7:4018.
[81] Baciewicz AM, Sokos DR, Cowan RI. Aminoglycoside-associated nephrotoxicity in the
elderly. Ann Pharmacother 2003;37:1826.
[82] Rougier F, Claude D, Maurin M, et al. Aminoglycoside nephrotoxicity: modeling,
simulation, and control. Antimicrob Agents Chemother 2003;47:10106.
[83] Tran Ba Huy P, Bernard P, Schacht J. Kinetics of gentamicin uptake and release in the
rat: comparison of inner ear tissues and uids with other organs. J Clin Invest 1986;77:
1492500.
[84] Gatell JM, Ferran F, Araujo V, et al. Univariate and multivariate analyses of risk factors
predisposing to auditory toxicity in patients receiving aminoglycosides. Antimicrob
Agents Chemother 1987;31:13837.
[85] Graham JC, Gould FK. Role of aminoglycosides in the treatment of bacterial
endocarditis. J Antimicrob Chemother 2002;49:43744.
[86] Galloe AM, Graudal N, Christensen HR, et al. Aminoglycosides: single or multiple daily
dosing? A meta-analysis on ecacy and safety. Eur J Clin Pharmacol 1995;48:3943.
[87] Hatala R, Dinh T, Cook DJ. Once-daily aminoglycoside dosing in immunocompetent
adults: a meta-analysis. Ann Intern Med 1996;124:71725.
[88] Barza M, Ioannidis JP, Cappelleri JC, et al. Single or multiple daily doses of aminoglycosides: a meta-analysis. BMJ 1996;312:33845.
[89] Ferriols-Lisart R, Alos-Alminana M. Eectiveness and safety of once-daily aminoglycosides: a meta-analysis. Am J Health Syst Pharm 1996;53:114150.
[90] Freeman CD, Strayer AH. Mega-analysis of meta-analysis: an examination of metaanalysis with an emphasis on once-daily aminoglycoside comparative trials. Pharmacotherapy 1996;16:1093102.
[91] Hatala R, Dinh TT, Cook DJ. Single daily dosing of aminoglycosides in immunocompromised adults: a systematic review. Clin Infect Dis 1997;24:8105.
[92] Ali MZ, Goetz MB. A meta-analysis of the relative ecacy and toxicity of single daily
dosing versus multiple daily dosing of aminoglycosides. Clin Infect Dis 1997;24:796809.
[93] Bailey TC, Little JR, Littenberg B, et al. A meta-analysis of extended-interval dosing
versus multiple daily dosing of aminoglycosides. Clin Infect Dis 1997;24:78695.
[94] Munckhof WJ, Grayson ML, Turnidge JD. A meta-analysis of studies on the safety and
ecacy of aminoglycosides given either once daily or as divided doses. J Antimicrob
Chemother 1996;37:64563.
[95] Nicolau DP, Freeman CD, Belliveau PP, et al. Experience with a once-daily aminoglycoside program administered to 2,184 adult patients. Antimicrob Agents Chemother
1995;39:6505.
[96] Urban AW, Craig WA. Daily dosage of aminoglycosides. Curr Clin Top Infect Dis 1997;
17:23655.
[97] Prins JM, Weverling GJ, de Blok K, et al. Validation and nephrotoxicity of a simplied
once-daily aminoglycoside dosing schedule and guidelines for monitoring therapy.
Antimicrob Agents Chemother 1996;40:24949.
[98] Fantin B, Carbon C. Importance of the aminoglycoside dosing regimen in the penicillinnetilmicin combination for treatment of Enterococcus faecalis-induced experimental endocarditis. Antimicrob Agents Chemother 1990;34:238791.
[99] Marangos MN, Nicolau DP, Quintiliani R, et al. Inuence of gentamicin dosing interval
on the ecacy of penicillin-containing regimens in experimental Enterococcus faecalis
endocarditis. J Antimicrob Chemother 1997;39:51922.

D. Kaye / Infect Dis Clin N Am 18 (2004) 669689

689

[100] Sexton DJ, Tenenbaum MJ, Wilson WR, et al. Ceftriaxone once daily for four weeks
compared with ceftriaxone plus gentamicin once daily for two weeks for treatment of
endocarditis due to penicillin-susceptible streptococci. Endocarditis Treatment Consortium Group. Clin Infect Dis 1998;27:14704.
[101] Vic P, Ategbo S, Turck D, et al. Ecacy, tolerance, and pharmacokinetics of once daily
tobramycin for pseudomonas exacerbations in cystic brosis. Arch Dis Child 1998;78:
5369.
[102] Bragonier R, Brown NM. The pharmacokinetics and toxicity of once-daily tobramycin
therapy in children with cystic brosis. J Antimicrob Chemother 1998;42:1036.
[103] Banerjee D, Stableforth D. The treatment of respiratory pseudomonas infection in cystic
brosis: what drug and which way? Drugs 2000;60:105364.
[104] Beringer PM, Vinks AA, Jellie RW, et al. Pharmacokinetics of tobramycin in adults
with cystic brosis: implications for once-daily administration. Antimicrob Agents
Chemother 2000;44:80913.
[105] Byl B, Baran D, Jacobs F, et al. Serum pharmacokinetics and sputum penetration of
amikacin 30 mg/kg once daily and of ceftazidime 200 mg/kg/day as a continuous infusion
in cystic brosis patients. J Antimicrob Chemother 2001;48:3257.
[106] Barclay ML, Duull SB, Begg EJ, et al. Experience of once-daily aminoglycoside dosing
using a target area under the concentration-time curve. Aust N Z J Med 1995;25:2305.
[107] Paterson DL, Robson JM, Wagener MM, et al. Monitoring of serum aminoglycoside
levels with once-daily dosing. Pathology 1998;30:28994.