domonas aeruginosa are sensitive to gentamicin, tobramycin, netilmicin and amikacin. The dif- ferences between gentamicin and tobramycin are usually small ...
REVIEW
ARTICLES
Aminoglycoside therapy Current use and future prospects R. Janknegt ficient to block protein synthesis. The u p t a k e of Introduction The aminoglycosidic aminocyclitols, usually aminoglycosides in the bacterial cell is irreverscalled aminoglycosides, h a v e been in clinical use ible [2]. The entry into the cell is an active, oxygenfor over 45 years. The introduction of new agents with a broad antibacterial action, like third- dependent process. Therefore, aminoglycosides g e n e r a t i o n cephalosporins, aztreonam, imipe- are especially active against actively growing n e m and the fluoroquinolones has modified the bacteria and practically inactive under anaerobic conditions. An increased u p t a k e of aminoclinical applications of aminoglycosides. In this article the bacteriological, pharmaco- glycosides in combination with b e t a - l a c t a m antikinetic, clinical and toxicological aspects of biotics t h r o u g h disruption of the cell wall is aminoglycosides will be reviewed and initial re- believed to contribute to the synergistic activity sults with new dosage regimens for aminoglyco- of this combination [3]. ~ides will be discussed. Spectrum of activity Aminoglycosides h a v e a broad spectrum of acMechanism of action I m p o r t a n t advances h a v e been made in the tivity. Most Enterobacteriaceae, including Pseustudy of cellular events underlying the bacteri- domonas aeruginosa are sensitive to gentamicin, cidal action of aminoglycosides. Yet, our knowl- tobramycin, netilmicin and amikacin. The differences between gentamicin and t o b r a m y c i n are edge is still far from complete. It has been k n o w n for m a n y years t h a t amino- usually small, gentamicin is often more active glycosides bind irreversibly to the subunit 30 S of against Escherichia coli and Serratia, tobrabacterial ribosomes, causing a disruption of pro- mycin has a better in vitro activity against Pseutein synthesis, eventually resulting in cell death domonas aeruginosa [4]. A m i k a c i n is 1-4 times less active t h a n genta[1]. This mechanism, however, does not completely explain the rapid bactericidal effect. Re- micin against both Gram-positive and Gramcent studies have provided more insight into the negative bacteria [4]. Aminog]ycosides are less active against Gramevents preceding the bactericidal action of aminoglycosides. After binding to the cellular positive bacteria; staphylococci are moderately surface, a f e w aminoglycoside molecules enter susceptible, while streptococci are resistant to the cell t h r o u g h pore channels or by low-affinity aminoglycosides. Yet, streptomycin or gentat r a n s p o r t mechanisms, causing misreading of micin together with benzylpenicillin or vancoprotein elongation. mycin are standard t h e r a p y for enterococcal Some misread proteins are incorporated in the endocarditis, because of the synergistic effect of cell m e m b r a n e , somehow resulting in increased aminoglycosides [5]. N e i t h e r aminoglycoside has permeability. More drug molecules enter the activity against anaerobic bacteria. Streptocell, thereby increasing leakiness even further, mycin is the most active aminoglycoside against until the concentration of drug molecules is suf- Mycobacterium tuberculosis. It is currently re-
Janknegt R. Aminoglycoside therapy. Current use and future prospects. Pharm Weekbl [Sci] 1990;12(2):81-90.
Keywords Amikacin Aminoglycosides Drug administration schedule Gentamicins Netilmicin Pharmacokinetics Tobramycin Toxicology R. Janknegt: Department of Clinical Pharmacy, Maasland Hospital, P.O. Box 5500, 6130 MB Sittard, the Netherlands.
12(3) 1990
Abstract The microbiological, pharmacokinetic, toxicological and clinical aspects of aminoglycosides are reviewed. Aminoglycosides still have an important place in serious infections in neutropenic patients, endocarditis and Pseudomonas aeruginosa infections, all in combination with betadactams. Monotherapy (with streptomycin) is indicated in less common diseases like tularaemia and bubonic plague. Several experimental studies support a oncedaily dosing regimen for aminoglycosides (comparable or better efficacy with less ototoxicity and nephrotoxicity). Only a very limited number of prospective comparative studies have been performed, and much more data on efficacy, development of resistance and toxicity is needed before once-daily administration can be recommended. The choice of an aminoglycoside should be based primarily on the local sensitivity patterns and cost. Differences in ototoxicity and nephrotoxicity are usually minor. If the acquisition costs of amikacin decline, it is to be expected that amikacin will be the aminoglycoside of choice. Accepted 16 November 1989.
P h a r m a c e u t i s c h Weekblad Scientific edition
81
garded as a second-choice tuberculostatic drug, because of development of resistance and its nephrotoxicity and ototoxicity [5]. Amikacin may still be active against streptomycin-resistant strains of Mycobacterium tuberculosis [6] and could be an alternative for strepto~ mycin in multiple resistant Mycobacterium tuberculosis. Amikacin shows in vitro and in vivo strong activity against Mycobacterium avium complex. Its rapid bactericidal action is enhanced by clofazimine [7 8]. Amikacin is the only aminoglycoside with significant activity against Nocardia asteroides. Sulfonamides and minocycline are also very active against Nocardia [9 10]. Table I lists the organisms for which aminoglycosides are regarded as drug of first choice. Combination therapy with penicillins or cephalosporins is recommended in serious cases of life-threatening infections, especially in immunocompromised patients.
Factors influencing activity The antibacterial activity of aminoglycosides is hardly influenced by the size of the inoculum. The minimal bactericidal concentration is usually equal to or 1-2 dilution steps higher than the minimal inhibiting concentration (MIC). The killing effect is strongest during log phase growth [4]. Aminoglycosides are 8-128 times less active in the presence of 5-100 mg/1 calcium and/or magnesium and are also less active at low pH values and in urine. The mechanisms behind these inhibitory influences are unknown. It is postulated that the binding to the outer membrane and energy-dependent uptake are inhibited by calcium and magnesium ions [12]. The rate and intensity of the bactericidal activity of aminoglycosides is concentration-dependent. Complete and rapid killing is observed with higher concentrations [13]. Aminoglycosides show a significant postantibiotic effect, which increases in duration at higher dosages. The duration of the postantibiotic effect usually lies between 3 and 6 h. The in vitro postantibiotic effect may be augmented by the fact that exposed bacteria are more susceptible to phagocytosis and that sub-MIC serum levels still
Table 1 Micro-organisms for which aminoglycosides are regarded as drug of choice [11] Monotherapy Francisella tularensis Yersinia pestis Combination therapy Enterococcus faecalis Listeria monocytogenes Escherichia coli Brucella Pseudomonas aeruginosa Pseudomonas mallei
82
streptomycin or gentamicin streptomycin gentamicin; + benzylpenicillin or ampicillin/amoxicillin gentamicin; + ampicillin/amoxicillin + ampicillin/amoxicillin streptomycin + tetracycline + antipseudomonal beta-lactam streptomycin + tetracycline
have a certain growth-inhibiting effect [14]. The concentration-dependent bactericidal activity and postantibiotic effect have stimulated studies with different dosing regimes for aminoglycosides. Once-daily dosing of netilmicin compared with three times daily administration showed higher bactericidal titers in the once-daily group [15].
Synergy Aminoglycosides are often used in combination with beta-lactams, because of proposed synergism. A synergistic action of aminoglycosides with various beta-lactams is seen in vitro and in vivo against enterococci, against which aminoglycosides alone are completely inactive, and sometimes also against Ps. aeruginosa, Serratia marcescens, E. coli, Klebsiella spp. and Proteus mirabilis [5 16]. Amikacin is almost always the most synergistic aminoglycoside in in vitro studies. Synergy is often ( > 90%) observed with ceftazidime, ceftriaxone and aztreonam. The frequency of synergy is less with broadspectrum penicillins, like carbenicillin. Synergy is infrequent with imipenem or ciprofloxacin [17]. The importance of synergism should not be overestimated for clinical infections. In vivo synergy has only rarely been demonstrated. Often the value of a combination lies also in a broader spectrum and a slower development of resistance [18]. In vivo synergy has been claimed to be important in the treatment of infections in granulocytopenic patients [4], although double beta-lactam combinations appear to be as active, despite the fact that in vitro antagonism is sometimes observed between two beta-lactams [19]. Even very low gentamicin concentrations in the cerebrospinal fluid have a synergistic action against Haemophilus influenzae with amoxicillin and cefotaxime [20]. A combination of ceftazidime and amikacin was more effective than a combination of ceftazidime and pefloxacin in nosocomial pneumonia or bacteraemia in intensive-care patients [21].
Resistance Resistance to aminog]ycosides occurs through several mechanisms. Anaerobic bacteria are intrinsically resistant, because of the O2-dependent uptake. A decreased permeability or alterations in the ribosomal target site (enterococci) usually result in resistance to all aminoglycosides, but are infrequent [5 10 22]. The most important mechanism of resistance is the production of aminoglycoside-modifying enzymes. These usually plasmid-mediated enzymes inactivate aminoglycosides by acetylation at an amino group (acetyltransferase), by phosphorylation at a hydroxyl group (phosphotransferases) or by adenylation at a hydroxyl group (adenyltransferases). The sites of action of the enzymes are shown in Figure 1. The enzymes are not excreted extracellularly. Important differences exist in the susceptibility of aminoglycosides to enzymatic modification. Amikacin and to a lesser extent netil-
Pharmaceutisch Weekblad Scientific edition
12(3) 1990
I
AKanamyc~n
II
CHNH O
NH
c.aGentamJcln~
O
N~2H
III
!
HO/k~q,
CHNH
CHOH H H
. c/e---o~
patterns from various countries in Europe (mostly university centres). The decision on the most suitable standard aminoglycoside should not be based on literature data, but much more on the susceptibility pattern in the hospital in question. Amikacin is often reserved as a 'back-up' aminoglycoside. In some hospitals where gentamicin resistance was high, amikacin has been used as first line aminoglycoside. High level use of amikacin during 14 years has not led to an increase in resistance. In many instances the incidence of gentamicin and tobramycin resistance has even declined during amikacin use [4 24-26].
~
Pharmacokinetics The pharmacokinetic profile is very similar for all aminoglycosides. Because of their high poCHNH NH O larity, aminoglycosides are only minimally absorbed from the gut and penetrate the central Arn,k~:m O OH I~H nervous system poorly. However, therapeutic OH / NH HO J blood levels have occasionally been found in 0 CHOH patients receiving oral aminoglycosides for selective decontamination [27]. The main pharmacokinetic properties of aminoglycosides are shown OH in Table 3. Aminoglycosides are rapidly and completely Figure 1 absorbed after intramuscular administration, Chemical structure of aminoglycosides. AAC: side of action of acetylation peak levels are reached in 30-60 min. The rate of enzymes; AAD: site of action of adenylation enzymes; AHP: site of action of absorption was significantly slower in patients phosphorylation enzymes with spinal cord injury and in elderly or severely ill patients [27 29]. Aminoglycosides have a volmicin are more resistant to inactivation than ume of distribution of about 0.25 1/kg, which aptobramycin and gentamicin [5 10]. Important proximates the extracellular fluid compartment. geographic differences exist in the incidence of After intravenous administration an initial aminoglycoside-modifying enzymes and in the rapid distribution phase is observed with a halfpatterns of aminoglycoside resistance. In life of about 5-15 rain. Timing of peak level deterNorthern Europe the incidence (and difference) mination, 1 h after intramuscular adminisin aminoglycoside resistance is much lower than tration or 30 rain after an intravenous infusion, in Southern Europe, which parallels a greater is therefore critical. Aminoglycosides are only aminoglycoside use per patient day in Southern minimally bound to serum proteins. The volume Europe [22 23]. of distribution, on a l/kg basis, may be increased The incidence of gentamicin-resistant Pseudo- in patients below ideal body weight, and may be monas aeruginosa varied from 0.5% in Sweden to decreased in obese patients [30]. 54.5% in Italy. Also, the differences in resistance An increased volume of distribution is also in Southern and Northern Europe between gen- seen in patients with fluid overload and in neotamicin and amikacin were much larger in nates [5 27]. Aminoglycosides are excreted alSouthern Europe [23]. Table 2 shows resistance most entirely by glomerular filtration, the elim-
o/)c.
Q I
Table 2 Arninoglycoside resistance patterns (percentage resistance) in blood isolates [22]
Sweden Finland Federal Republic of Germany The Netherlands Switzerland United Kingdom Spain France Belgium Austria Greece Italy
12(3) 1990
Gentamicin
Tobramycin
Amikacin
Netilmicin
3.4 3.6 5.0 6.5 7.5 14 16 17 18 21 28 31
3.0 6.1 5.5 8.0 5.0 12 15 17 20 19 25 31
1.9 1.0 0 5.5 0 5.6 2.1 2.5 2 2.1 18 14
1.3 2.5 2.0 3.5 1.0 5.6 13 7.9 10 4.1 25 24
Pharmaceutisch Weekblad Scientific edition
83
Table
3
Pharmacokinetic properties [5 27 28]
Elimination half-life (h) Volume of distribution (1/kg) Plasma clearance (ml/min) Protein binding (%) Renal excretion (%) Maximum urinary concentration (mg/1) Elimination half-life ESRD* (h) Plasma clearance ESRD* (ml/min) Dialysis clearance (ml/min) Desired serum levels (mg/1) - peak - trough
Gentamicin
Tobramycin
Netilmicin
Amikacin
2 0.24 95 < 10 > 95 100-300 (1-2 mg/kg) 60
2 0.23 80 < 10 > 95 100-300 (1-2 mg/kg) 60
2 0.25 88 < 10 > 90 100-300 (1-2 mg/kg) 40
2-3 0.25 90 < 10 > 95 700-805 (5-7 mg/kg) 70
2 24
3 27
5 30-35
3 18
6-10 0.5-2
6-10 0.5-2
6-10 0.5-2
20-30 10
*ESRD: end-stage renal disease. ination half-lives are 2 h for gentamicin, tobramycin and netilmicin and 2-3 h four amikacin and kanamycin. The rate of elimination is increased in patients with cystic fibrosis [31] and decreased in neonates, severely ill patients [32 33] and of course in patients with disturbed renal function. However, also in patients with normal renal function, aminoglycoside half-lives may vary from 0.432.7 h [34]. The concentration in most body fluids is comparable to or slightly lower than the serum levels. Low levels are reached in sputum (30% of serum concentrations). Very high concentrations are reached in urine; over 85% of a dose is excreted in the urine within 24 h. Significant accumulation occurs in renal cortex tissue, which is an important determinant of aminoglycoside-induced nephrotoxicity. Because of the wide interpatient variability, which is probably less in the case of amikacin and the small therapeutic index of aminoglycosides, determination of peak and through levels are essential in aminoglycoside therapy. A high ratio of peak level/MIC is an important determinant of therapeutic success of aminoglycosides. When peak/MIC ratios are higher than 6-8, more patients are cured or improved than at lower peak/MIC ratios [27 35 36]. Therefore, high peaklevels, > 5 mg/1 in case of gentamicin, netilmicin and tobramycin and > 20 mg/1 with amikacin, are needed for successful therapy. It should be kept in mind that careful monitoring of serum levels of aminoglycosides will reduce, but by no means prevent the incidence of, ototoxicity and nephrotoxicity. Administration of aminoglycosides to patients with end-stage renal disease is dangerous. High peak levels are necessary for adequate efficacy, while high trough levels correlate with ototoxicity and nephrotoxicity [37]. Dose adjustments can be made by giving lower doses (with consequently lower peak levels), longer dosage intervals (with consequently longer periods between peak levels) or by a combination of both [28]. 84
The necessity of using aminoglycosides in dialysis patients must be outweighed against a relatively high risk of irreversible ototoxicity. Nephrotoxicity
Aminoglycoside-induced nephrotoxicity is one of the key problems of aminoglycoside therapy. Very high levels of aminoglycosides are reached in renal cortical tissue. After binding to the brush-border membrane, aminoglycosides are absorbed by pinocytosis and show a high affinity for phospholipid membranes. An aspecific increase in number and size of lysosomes is observed with a diminished lysosome stability. Aminoglycosides bind tightly to negatively charged phospholipids once transferred to the lysosomes. Lysosomal phospholipases are dependent on the presence of negative charges in lipid bilayers. The strong binding of aminoglycosides to these negative charges results in a marked decrease of phospholipase activity [38]. Mitochondrial alterations and changes in membrane permeability contribute to cell death [39 4 0 ] . Clinically, aminoglycoside-induced nephrotoxicity is characterized by enzymuria (brush-border enzymes and lysosomal enzymes), followed by glycosuria, aminoaciduria and proteinuria and potassium and magnesium wasting, causing polyuria and nephrogenic diabetes insipidus. Changes in glomerular filtration rate and creatinine clearance usually do not occur within 7-10 days. Therefore, serum creatinine is not a very reliable marker for early detection of aminoglycoside nephrotoxicity [39]. Nephrotox~ icity is not always fully reversible and recovery may take many months [5 28 39], corticosteroids might have a negative influence on recovery from nephrotoxicity [41]. Broad-spectrum penicillins like piperacillin might protect against nephrotoxicity; in an experimental study performed in rats piperacillin caused a significant reduction of cortical and medullar gentamicin levels, without effect on serum levels [42]. It should be noted that aminoglycosides may also be inactivated by penicillins. Van-
Pharmaceutisch Weekblad Scientific edition
12(3) 1990
comycin and endotoxins may increase nephrotoxicity [43 44]. The most important risk factors for aminoglycoside-induced nephrotoxicity include: duration of therapy, dosage, repeated courses, elderly patients, liver disease, renal disease, obese patients, nephrotoxic comedication and volume depletion [39 45 46]. Renal cortical uptake is a saturable process for gentamicin and netilmicin, the uptake of amikacin is linear from serum levels />5 mg/1 and the uptake of tobramycin is almost linear [47]. The cortical uptake of gentamicin and netilmicin is higher than that of amikacin or tobramycin. Less frequent dosing of aminoglycosides could reduce renal cortical uptake, because of the saturable uptake process. Indeed, once daily dosing in animals resulted in significantly lower renal cortical levels, compared to continuous infusion, especially for gentamicin and netilmicin [47 48]. Timing of renal tissue level determination is important. When nephrotoxicity proceeds, concentrations fall because of diminished binding to necrotic cells [38]. In experimental studies amikacin caused least lysosomal overloading and inhibition of lysosoreal enzymes [40 47]. In clinical studies, no single aminoglycoside has been shown to be clearly more or less nephrotoxic than others. Meyer reviewed 27 comparative clinical studies, involving 2 or more aminoglycosides [49]. A very wide incidence of nephrotoxicity was found for all aminoglycosides, depending on definitions, duration, methodology, co-medication, severity of illness etc. The 'ideal' clinical study comparing amikacin, gentamicin, netilmicin and tobramycin, should be prospective, randomized and double-blind. Serum levels should be strictly monitored; toxicity should be clearly defined; renal function (creatinine clearance) should be measured before, during and after therapy, and other possible causes of toxicity would have to be differentiated. Finally, the number of patients entered into the study would have to be sufficient to detect moderate difference between aminoglycosides. Such a study has not yet been done and given the limited financial interest of the pharmaceutical industry it seems unlikely that such a study will ever be performed.
able fashion. In the case of gentamicin the affinity for vestibular binding sites was greater than for cochlear binding sites. Amikacin bound better to cochlear binding sites, which is well in accordance with the clinically observed ototoxicity [5]. By binding to phospholipids and by inhibiting protein synthesis in cochlear and vestibular cells, the cells eventually degenerate. Relatively low levels of aminoglycosides are reached in the inner ear, but with a long elimination half-life of about.24-48 h [4 50]. The incidence of ototoxicity is highly variable and depends largely on the definitions of ototoxicity and the methodology. In a small population a loss of hearing of 15 dB ore more occurred in 44% of the netilmicin-treated patients to 23.5% in the amikacin-treated group [52]. Risk factors for developing ototoxicity include impaired renal function, total dose, duration of therapy, prior exposure to aminoglycosides, other ototoxic medication, high peak and/or through levels, preexisting ear disease and severely ill patients [5]. High-tone audiometry is a sensitive method for determining early ototoxic effects, but this is not routinely performed [50]. From experimental studies it seemed that netilmicin was the least ototoxic aminoglycoside, but this has not yet been fully established during comparative clinical studies [3-5], partly because most clinical studies were small-scale. Other variables, like the risk factors mentioned above, often outweigh the subtle differences between aminoglycosides. Just like stated in the section on nephrotoxicity the 'ideal' comparative study has not yet been performed and it is extremely difficult to obtain initial (high-tone) audiograms in severely ill patients, so it seems unlikely that the ototoxicity question will soon be settled.
Other side effects Aminoglycosides inhibit neuromuscular transmission by preventing acetylcholine release and reducing postjunctional sensitivity to acetylcholine [53]. Some experimental studies have shown that aminoglycosides potentiated the blocking effects of steroid nondepolarizing agents like pancuronium and atracurium and that this effect is even further enhanced by verapamil [54 55]. Netilmicin has a stronger blocking effect than gentamicin and tobramycin [53]. Although the Ototoxicity clinical relevance of this side effect is unclear, Ototoxicity is the second major side effect of the interaction with agents like pancuronium is aminoglycosides. Contrary to nephrotoxicity this a possible disadvantage for application in surgiside effect is largely irreversible. Two types of cal prophylaxis. ototoxicity should be distinguished: Hypersensitivity reactions are infrequent and - vestibular toxicity: initially nausea, vomiting, usually mild [4]. Other uncommon side effects insweating and vertigo, later ataxia and some- clude mental confusion, nausea and peripheral times nystagmus; neuritis [5]. - cochlear toxicity: tinnitus, later loss 'of hearing, initially in high frequencies. Place of aminoglycosides in therapy Vestibular toxicity is seen most with gentamicin. A combination of an aminoglycoside with a Auditory toxicity appears to occur most fre- beta-lactam antibiotic is still considered firstquently with amikacin [4 50], although results line treatment in febrile episodes in neutropenic from different studies are often contradictory. patients [4 5 10 36]. It is still unclear whether In an experimental study with guinea pigs, aminoglycosides will eventually be replaced by a gentamicin bound to cochlear and vestibular beta-lactam or by double beta-lactam or betastructures in the inner ear in a rapid and satur- lactam-quinolone combinations (Table 4).
12(3) 1990
Pharmaceutisch Weekblad Scientific edition
85
Table 4
Indications for aminoglycosides Monotherapy Tularaemia Bubonic plague
Combination therapy Febrile episodes in neutropenic patients (+ antipseudomonal beta-lactam) Septicaemia of unknown origin (+ beta-lactam) Endocarditis - streptococci (+ benzylpenicillin or amoxicillin) - staphylococci (+ flucloxacillin or vancomycin) Severe Pseudomonas aeruginosa infections (+ antipseudomonal 3-tactam)
The combination with beta-lactams is also important in the initial treatment of septicaemia until the causative pathogen is known, the synergistic combination with many beta-lactams is often used [4 5]. Streptomycin or gentamicin in combination with benzylpenicillin, amoxicillin or flucloxacillin is a standard therapy for endocarditis caused by enterococci or staphylococci [4 5 10]. Streptomycin (or gentamicin) monotherapy is first-line treatment in tularaemia or bubonic plague [11]. Aminoglycosides are still important in infections caused by Pseudomonas aeruginosa. Whether aminoglycosides will eventually be replaced by quinolones remains to be seen. Aminoglycosides have a limited position in the treatment of Gram-negative central nervous system infections (virtually no penetration into spinal fluid and into the brain); third-generation cephalosporins are becoming first-line treatment for these infections, although even low levels of gentamicin may have a synergistic effect with ampicillin or cefotaxime in meningitis [5 10 20]. Aminoglycosides do not have a major place in the treatment of respiratory tract infections, except when caused by beta-lactam-resistant organisms, because of their limited penetration into bronchial secretions and diminished activity at acid pH values [5]. The position of aminoglycosides in the treatment of complicated urinary tract infections will probably be taken over by quinolones, because of the lesser toxicity, higher tissue levels, oral application and wider spectrum (including many strains of Chlamydia tr.achomatis and Mycoplasma hominis) of the latter compounds. Table 5
Experimental data favouring once-daily administration Concentration-dependent killing Concentration-dependent postantibiotic effect First exposure effect Better bacteriological results in animal studies Less accumulation in renal cortex tissue Less phospholipidosis Less uptake into inner ear
86
Aminoglycosides have been an important treatment of bone and joint infections caused by Gram-negative bacteria. The long duration of the treatment, with concomitant nephrotoxicity and ototoxicity is an important drawback for these indications. Initial results with fiuoroquinolones, like ciprofloxacin, have been encouraging and it seems likely that quinolones or betalactam-quinolone combinations will replace aminoglycosides for this indication.
Once-daily administration Experimental studies Several studies in vitro in animals have strongly suggested that once-daily administration of aminoglycosides may offer advantages over multiple-daily doses. The results of these studies are summarized in Table 5. Several studies have shown that the rate of bactericidal activity of aminoglycosides is concentration-dependent, i.e. higher concentrations result in a faster and greater reduction in the number of bacteria [13 56]. Moreover, the postantibiotic effect of aminoglycosides is also concentration-dependent and increases significantly with higher concentrations. Transport of aminoglycosides through the bacterial outer-membrane is an energy-dependent process, followed by (also energy-dependent) active transport to the bacterial ribosome. If bacteria survive the first dose of an aminog]ycoside, they are less metabolically active for a given period; the second dose may be significantly less effective in reducing the number of bacteria, because of a reduced internalization by the bacteria. The lethality of the first dose is therefore very important in reducing this effect, which is called the first exposure effect. High intermittent doses offer theoretical advantages in this respect [57]. In animal studies comparing the efficacy of once-daily versus multiple dosing or continuous infusion, once daily dosing was as effective or more effective in reducing the number of bacteria [3 13 56 58], although once-daily dosing in monotherapy tended to be less effective in neutropenic animals. The uptake of aminoglycosides, especially gentamicin and netilmicin, into renal cortex tissues is concentration-dependent. Several studies have shown that once-daily administration of aminoglycosides reduces the concentration in renal tissue compared to three times daily administration or continuous infusion [38 48 58 59]. Phospholipidosis was also reduced in animals receiving once-daily netilmicin compared to trice-daily administration [38] and serumcreatinine levels were lower in rabbits receiving oncedaily tobramycin compared to three times daily [58]. Few data is known on the effects of once-daily administration of aminoglycosides on ototoxicity. Experimental studies suggested much higher levels in the inner ear after continuous infusion compared to bolus injections. The uptake of aminoglycosides in the inner ear appeared more effective at low, but sustained, plasma
Pharmaceutisch Weekblad Scientific edition
12(3) 1990
levels t h a n at high, but brief, p l a s m a concentrations [60]. Clinical studies
Several clinical studies comparing the efficacy of once-daily and multiple dosing have been performed, especially with netilmicin. The results of these, often preliminary, studies are shown in Table 6. The theoretical a d v a n t a g e s of once-daily administration, as discussed above, have not yet been fully confirmed in these clinical studies. It will require large n u m b e r s of patients to show equality or a difference in efficacy with once-daily regimens compared to the (quite effective) multiple-dose schedules now in use. Much is still unclear about the best dosing schedule for aminoglycosides. Optimal p e a k levels and t h r o u g h levels h a v e not yet been established with once-daily administration. Once-daily dosing of aminoglycoside is an attractive alternative for the currently used dosing schedules, but m u c h more data on efficacy, development of resistance and toxicity is needed before once-daily dosing will be the new 'golden s t a n d a r d ' for use of aminoglycosides.
Which aminoglycoside?
itial t h e r a p y in surgical patients [70]. W h e n culture results were known, t h e r a p y was changed to the cheapest, active aminoglycoside. The use of a m i k a c i n t u r n e d out to be the least expensive method (shorter follow-up courses), although differences were small. This A m e r i c a n study cannot be t r a n s l a t e d directly to other locations, because of differences in susceptibility p a t t e r n s and also .in cost calculation. It is to be expected t h a t the acquisition costs of tobramycin and a m i k a c i n will decline in the n e a r future, because of the a p p e a r a n c e of generic, or hospital-prepared, products. In t h a t case amikacin, which has no disadvantages in comparison with gentamicin, will probably be the aminoglycoside of choice.
Future prospects Only a limited n u m b e r of new aminoglycosides, like isepamicin and dactimicin are in development. It seems unlikely t h a t i m p o r t a n t advances (better efficacy or less toxicity) in the group of the aminoglycosides will be made in the n e a r future. It r e m a i n s to be seen w h e t h e r quinolones will t a k e over the place of aminoglycosides in the t r e a t m e n t of infections caused by G r a m - n e g a t i v e bacteria. It seems attractive to use high-dose aminoglycosides for a short period (2-3 days), followed by m o n o t h e r a p y or combination t h e r a p y of less toxic antimicrobials to reduce the nephrotoxicity and ototoxicity of aminoglycosides.
The choice between aminoglycosides is clear in several indications. Streptomycin is the drug of choice for bubonic plague and t u l a r a e m i a . In combination with benzylpenicillin, amoxicillin or flucloxacillin, streptomycin or gentamicin are the aminoglycosides for the t r e a t m e n t of endocarditis caused by Gram-positive cocci [4 5 10 11]. Acknowledgement I a m grateful to Prof. J.W.M. v a n der Meer, Although all these indications are i m p o r t a n t applications of aminoglycosides, they are only a D e p a r t m e n t of I n t e r n a l Medicine, University small fraction of the total use of aminoglycosides. Hospital Nijmegen, for his critical reviewing of The choice of an aminoglycoside should be this manuscript. based on its efficacy, toxicity and cost. The expected efficacy of an aminoglycoside largely de- References 1 Edson RS, Terrell CL. The aminoglycosides: streptopends on the sensitivity p a t t e r n in the hospital mycin, kanamycin, gentamicin, tobramycin, amikacin, in question. Gentamicin resistance is u n c o m m o n netilmicin and sisomicin. Mayo Clin Proc 1987;62: in non-university hospitals and in hospitals in 916-20. 2 Davis BD. The lethal action of aminoglycosides. J N o r t h e r n Europe. Antimicrob Chemother 1988;22:1-3. The better synergistic activity of a m i k a c i n is 3 Humbert G, Carbon C, Collatz E. Amineglycosides potentially an advantage, but large-scale trials (aminocyclitols). In: Peterson PK, Verhoef J, eds. confirming a superior activity are lacking. If reAntimicrobial agents annual 3. Amsterdam: Elsevier Science Publishers, 1988:1-14. sistance to g e n t a m i c i n (and usually also tobra4 Cunha BA. Aminoglycosides: current role in antimimycin) is high, a m i k a c i n is the aminoglycoside of crobial therapy. Pharmacotherapy 1988;8:334-50. choice. Firstqine use of a m i k a c i n has not led to 5 Pancoast SJ. Aminoglycosides antibiotics in clinical increased resistance to amikacin, while resistuse. Med Clin N Am 1988;72:581-612. 6 Hoffner SE, K~illenius G. Susceptibility of streptomyance to gentamicin and t o b r a m y c i n usually decin-resistant M. tuberculosis to amikacin. Eur J Clin clined. The choice of an aminoglycoside should Microbiol Infect Dis 1988;7:188-90. not be based p r i m a r i l y on nephrotoxicity or oto7 Gangm'adham PR, Kesavalu L, Rao PN, et al. Activity toxicity, because all aminoglycosides are potentiof amikacin against M. avium complex under sinmlated in vivo conditions. Antimicrob Agents Chemoally toxic and differences are small. The use of ther 1988;32:886-9. aminoglycosides in patients with renal function 8 Gangaradham PR, Perumal VK, Rao PN, Kesavula L, i m p a i r m e n t should be restricted to those Iseman MD. In vivo activity of amikacin alone or in patients in which their use is absolutely necesscombination with clofazimine or rifabutin or both against acute experimental M. avium complex infecary. The risk of nephrotoxicity and irreversible tions in beige mice. Antimicrob Agents Chemother ototoxicity is strongly increased in these 1988;32:1400-3. patients. 9 Wallace RJ, Steele LC, Sumter G, Smith JM. AntimiThe choice of an aminoglycoside depends crobial susceptibility patterns of N. asteroides. Antimicrob Agents Chemother 1988;32:1776-9. largely on economic factors. Gentamicin is by far the cheapest aminoglycoside and is therefore the 10 Siegenthaler WE, Bonetti A, Lfithy R. Aminoglycosides antibiotics in infectious diseases. Am J Med preferred drug in the majority of hospitals. 1986;80(Suppl 6B):2-14. Gladen made a cost comparison of when amika- 11 Anonymous. The choice of antimicrobial drugs. Med Lett Drugs Ther 1988;30:33-40. cin, gentamicin or t o b r a m y c i n were chosen as in12(3) 1990
P h a r m a c e u t i s c h Weekblad Scientific edition
87
16 Ristuccia AM, Cunha BA. An overview of amikacin. Ther Drug Monitor 1985;7:12-25. 17 Giamarellou H. Aminoglycosides plus beta-lactams against Gram-negative organisms. Am J Med 1986;80 (Suppl 6B):126-37. 18 Mich6a-Hamzepour M, Pech~re JC, Marchou B, Auckenthaler R. Combination therapy: a way to limit emergence of resistance. Am J Med 1986;80(Suppl 6B): 138-42. 19 Dejace P, Klastersky J. Comparative review of combination therapy: two beta-lactams versus beta-lactam plus aminoglycoside. Am J Med 1986:80(Suppl 6B): 29-38. 20 Bingen E, Lambert-Zechovsky N, Auyard Y, et al. Early synergistic killing activity at concentrations attainable in CSF of amoxicillin or cefotaxime and aminoglycosides against H. influenzae. Infection 1988; 16:121-5.
12 Blaser J, Lfithy R. Comparative study on antagonistic effects of low pH and cation supplementation on invitro activity of quinolones and aminoglycosides against P. aeruginosa. J Antimicrob Chemother 1988; 22:15-22. 13 Bakker-Woudenberg IA, Roosendaal R. Impact of dosage regimens on the efficacy of antibiotics in the immuno-compromised host. J Antimicrob Chemother 1988;21:145-7. 14 Isaksson B, Nilsson L, Maller R, SSr6n L. Post-antibiotic effect of aminoglycosides on Gram-negative bacteria evaluated by a new method. J Antimicrob Chemother 1988;22:22-33. 15 Van der Auwera P. Bactericidal activity, killing rate and post-antibiotic effect in the serum of patients with urinary tract infection receiving netilmicin 6 mg/kg once daily in comparison with 2 mg/kg t.i.d. J Drug Dev 1988;l(Suppl 3):25-33.
Table 6 Clinical studies with once-daily aminoglycosides
Netilmicin Gangrenous and perforated appendicitis Pe]vic inflammatory disease
Dose
Co-medication
Duration
A 1 x 4.5 mg/kg (n=20) B 3 x 1.5 mg/kg (n=20) intramuscular A 1 x 6.6 mg/kg (n=19)
metronidazole 3 x 30 mg/kg
7 days
ampicillin 2 x 2 g + tinidazole 1 x 800 mg
7 days
metronidazole 3 x 5 0 0 m g intravenous
8.7 days
B 3 x 2.2 mg/kg (n=19) intravenous infusion
Abdominal infections
Febrile neutropenic patients
Septicaemia
A I x 4.5 mg/kg (n=58) B 3 x 1.5 mg/kg (n=56) intramuscular/ intravenously A 7 0 kg 1 x 450 rag(n=31) B < 7 0 k g 3 x 100rag; > 7 0 kg 3 x 150 m g ( n = 3 3 ) intravenous A 1 x 6 mg/kg (n=20)
8.8 days
cefuroxime 3 d.d. 750 mg
7 days
ceftriaxone 2 g q.d.
8.3 +_ 0.2 days
-
6.5 days 6.6 days
B 3 x 2 mg/kg (n=20) Severe infections
A 1 x 4 0 mmol/1 65 A 7% r i s e in c r e a t i n i n e ~>30 mmol/1 (2 p a t i e n t s ) B 14% r i s e i n c r e a t i n i n e >~30 mmol/1 (5 p a t i e n t s ) A 15% 66 B 17% no difference b e t w e e n A a n d B 67
A 100% cure; 1 r e c u r r e n c e
B 100% cure; 0 r e c u r r e n c e
A h e a r i n g loss t>15 dB: < 8 k H z 0/19; 10-18 k H z 3/19 B h e a r i n g loss ~< 15 dB: < 8 k H z 2/19; 10-18 k H z 9/19
_
A 1 p a t i e n t /> 15 dB loss b o t h ears B
-
A2 B1 no d i f f e r e n c e b e t w e e n A a n d B; 5% a u d i t o r y ; 1% v e s t i b u l a r
A1
A 100% c u r e
B 100% c u r e
12(3) 1990
BCA h e a r i n g loss ~>15 dB;
< 8 k H z 0/20; 10-18 k H z 2/20 B h e a r i n g loss ~
