Mechanisms of toxicity
The mechanisms associated with common adverse reactions to antimicrobials include dose-related toxicity that occurs in a certain fraction of patients when a critical plasma concentration or total dose is exceeded, and toxicity that is unpredictable and mediated through allergic or idiosyncratic mechanisms. For example, certain classes of drugs such as the aminoglycosides are associated with dose-related toxicity.
In contrast, the major toxicity of the penicillins and cephalosporins is due to allergic reactions. These differences are explained in part by the relative ability of specific drugs to inhibit enzymatic pathways in the host versus their stimulation of specific immune response. Not included in these lists is mention of the subtle adverse effects of a number of antibiotics on the host immune response. Antibiotics can reduce the efficacy of the host response to microbial pathogens.
They can diminish chemotaxis, phagocytosis, neutrophil- and macrophage-mediated microbial killing, lymphocyte transformation, delayed-hypersensitivity reactions, and production of antibody. For example, doxycycline decreases chemotaxis, phagocytosis, lymphocyte transformation, delayed-hypersensitivity reactions, and production of antibody. The antimicrobial effects on immune responses observed in vitro may or may not be clinically relevant; however, these data reinforce the concept that antimicrobials have potential to produce deleterious effects that are related to their pharmacologic effects independent of their actions on bacteria or viruses.
| Table Adverse Effects of Common Antimicrobial Agents | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
A consideration of the relative toxicities of different antimicrobials in relation to their efficacy is critical to the appropriate choice of antimicrobials. If two antibiotics have equivalent efficacy, the less toxic antibiotic should be chosen. The therapeutic index of an antimicrobial compares the plasma concentration of drug at which host toxicity appears with that which is effective in the treatment of the infectious disease. This ratio varies with the properties of the different drugs and the amount of antibiotic necessary to inhibit the organism in vivo at the site of infection. It may also vary with host factors that alter susceptibility to drug toxicity. For example, gentamicin and nafcillin both exhibit in vitro activity against staphylococci. However, the therapeutic index of gentamicin in most patients is quite narrow, whereas that of nafcillin is relatively wide. Accordingly, nafcillin or other penicillinase-resistant penicillins are the treatment of choice for staphylococcal infections (in subjects not allergic to penicillin). Similarly, because of their wide therapeutic indices, the penicillins and the cephalosporins are the preferred agents in the treatment of most serious infections caused by susceptible organisms in which bactericidal therapy is preferred. The picture changes dramatically in the case of penicillin allergy, when microgram amounts of these drugs may lead to fatal anaphylaxis. The one notable exception to this principle is the finding that anaphylactic reactions to penicillin do not occur on exposure to the monobactams (e.g., aztreonam). Otherwise, an anaphylactic reaction to one penicillin usually precludes administration of any other β-lactam antibiotic. Appropriate substitution therapy in the event of penicillin allergy is discussed in the section on specific antimicrobial drugs. When an untoward event occurs during antimicrobial therapy, the potential deleterious role of the drug must always be considered. For example, patients receiving parenteral amphotericin B should have a baseline potassium, magnesium, blood urea nitrogen, and creatinine concentration, urinalysis, and peripheral blood count determined before therapy is initiated. These values should be checked frequently (e.g., every 3 days) so that appropriate adjustments of dosage can be made or the drug stopped in the event of serious toxicity. Such close monitoring is not necessary when less toxic antimicrobials are used, although in any patient receiving prolonged therapy it is wise to follow blood counts and renal function, to examine the skin for allergic rashes, and to measure body temperature to check for drug-induced fever.
Principles
When the drug used to treat disease (e.g., infection) causes a sign of the disease (e.g., fever), then the physician must be especially vigilant in following the patient.
Antimicrobial toxicity due to altered host factors
Elimination of antibiotics may be modified by genetic factors, concomitant treatment with other drugs, or disorders that alter normal elimination pathways. Furthermore, the change in endogenous flora that occurs as a result of antimicrobial treatment can lead to unfavorable reactions.
For example, any antimicrobial that affects aerobic bacteria can alter the normal gut flora, thereby promoting selection for anaerobic Clostridium difficile superinfection of the gastrointestinal lumen.
This organism produces a potent cytotoxin that can result in pseudomembranous colitis. The potential contribution of these factors to both therapeutic efficacy and drug toxicity must be taken into consideration when selecting antimicrobials and monitoring patients for the effects of treatment.
Toxicity in specific patient groups
The exposure of pregnant women and neonates to certain drugs poses several problems that may have serious consequences to the fetus or infant (see chapter 19, Drug Therapy in Pregnant and Breast-Feeding Women, and chapter 30, Drug Therapy in Pediatric Patients).
Some drugs readily cross the placental barrier and can produce toxicity in the fetus. Examples of fetal toxicity include dental staining or tooth malformation (e.g., tetracyclines), ototoxicity (e.g., aminoglycosides), arthropathy (e.g., quinolones), and displacement of bilirubin from serum albumin with the production of kernicterus at birth (e.g., sulfonamides). Accordingly, these drugs should not be used during pregnancy nor should they be given to neonates. The physician must also recognize that there is a considerable lack of knowledge on the safety of a variety of newer antimicrobials in the pregnant woman.
Therefore, it is always prudent to use these drugs with restraint and only when clearly necessary. Newborns, particularly if premature, have a relative deficiency of the hepatic enzyme glucuronyl transferase and reduced hepatic clearance of chloramphenicol. This leaves neonates susceptible to a potentially lethal syndrome characterized by flaccidity, ashen color, and cardiovascular collapse (the “gray baby” syndrome).
The use of chloramphenicol in neonates should be avoided if possible. In the rare situation when it must be given, the dose of chloramphenicol should be limited to no more than 25 mg/kg per day for premature babies and 50 mg/kg per day for full-term infants, and the concentrations in blood should be frequently monitored. Subjects with glucose-6-phosphate dehydrogenase deficiency may develop hemolysis when given drugs with oxidant activity. Antimicrobials to be avoided include nitrofurantoin, chloramphenicol, the sulfonamides, furazolidone, nalidixic acid, aminosalicylic acid, and primaquine.
Reactions secondary to drug interactions
Certain combinations of drugs may lead to inactivation or to an exaggeration of the effects of antimicrobials. The potential for interactions is increased in seriously ill patients who are receiving several different medications and in whom drug metabolism may be altered by disease. For example, concomitant administration of antacids containing calcium or magnesium, or the administration of ferrous sulfate, can prevent absorption of tetracycline and quinolones from the gut. Chloramphenicol inhibits the activity of certain hepatic enzymes, and it interferes with biotransformation of barbiturates, phenytoin, warfarin, and tolbutamide. Isoniazid with phenytoin and sulfonamides with sulfonylureas similarly compete for the same enzymatic inactivating systems. The practitioner must be aware of all drugs and dietary substances ingested by his or her patient and be able to anticipate the presence of clinically important drug interactions.
Principles
Whenever more than one drug is given to a patient, the potential for toxicity is increased through known or unknown interactions. Polypharmacy should be avoided unless the indications for it are compelling.
Reactions due to impaired excretion
Almost all antimicrobials are excreted to some extent by the kidney; some are cleared predominantly by the liver. Whenever renal or hepatic failure is present, the physician should be aware of necessary alterations of dosage in order to avoid dose-related toxicity. The degree of change in the regimen is determined by the potential of a compound to cause dose-related toxicity, the drug’s route of clearance, and the magnitude of the renal or hepatic impairment. Table 14-19 provides a guide to modifications of dose necessary for different classes of drugs in the presence of renal failure. The major principles on which these recommendations are made include the following:
- The concentration of drug in plasma after an initial dose is a function of the dose, the rate of absorption of the drug, the rate of redistribution to tissues, and the rate of excretion. If the rate of absorption is much faster than the distribution and excretion of the drug, then slow excretion in renal failure will not appreciably alter initial concentrations of drug in blood. This is the situation that prevails with most antibiotics with significant dose-related toxicity (e.g., aminoglycosides). Therefore, the initial dose (loading dose) of these drugs requires no modification in patients with renal failure.
- After the initial dose of drug is given, the clearance becomes the important determinant of the rate of decline of concentration in plasma. Subsequent dosing must be reduced to correspond to slowed elimination. These reductions of dose can be achieved by lengthening the interval between doses or by reducing the dose administered at a fixed interval.
These points can be illustrated by using gentamicin as an example. The drug is excreted by glomerular filtration and is handled in the kidney in a way similar to the handling of creatinine. Consequently, creatinine clearance can be used as a guide to adjustment of the maintenance dose. In some patients, especially elderly ones, serum creatinine concentrations may be within normal values even when there is significant renal insufficiency. The administration of gentamicin at dose intervals that are three times its half-life in the plasma usually is adequate to maintain concentrations in plasma such that the peak concentration is approximately 8 times the rough concentration. Gentamicin has a half-life that varies from 2 to 4 hours in subjects with normal renal function. Using these points, one suggested dosage regimen for gentamicin makes use of the serum creatinine as follows:
- Initial (loading) dose: Administer 1.7 to 2.0 mg/kg IM or intravenous to produce a peak plasma concentration of 7 to 8 mg/L in most patients.
- Subsequent maintenance doses: Administer a dose of 1.0 to 1.7 mg/kg every 8 hours (about three half-lives) if the creatinine concentration is 1 mg/dL or less (i.e., normal renal function). To extend the dose interval in patients with renal compromise, multiply the usual dose interval (8 hours) by the patient’s serum creatinine concentration (in mg/dL). Higher doses would be reserved for patients with serious infections outside the urinary tract, in which case one cannot take advantage of the concentrating ability of the kidney to provide high concentrations at the site of infection.
| Table Pharmacokinetics and Dosage of Antimicrobial Agents in Renal Failure | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The above dosing guidelines are not precise. Patients vary in the plasma concentrations that are achieved with a given loading dose of gentamicin (interpatient variation in volume of distribution). In some patients, the excretion of gentamicin may be more rapid than the creatinine clearance would suggest; in others, it is slower. When the serum creatinine concentration is above 3 mg/dL and the interval between doses is more than 24 hours, some patients may have lower-than-therapeutic concentrations for a substantial time.
The second approach is to administer the second and subsequent doses of gentamicin at the usual fixed interval of every 8 hours, reducing the amount of drug according to a formula based on the serum creatinine value: the total amount of gentamicin administered over a given period would be as indicated in the above formula, but the smaller doses given more frequently would ensure against subtherapeutic concentrations of the drug near the end of a dose interval. A more recent method for dosing gentamicin relies on once daily dosing in patients with normal renal function.
The rationale for this is based on three principles: 1) aminoglycosides exhibit concentration-dependent killing; 2) there is a long postantibiotic effect following exposure to high concentrations; and 3) there is less renal and inner ear accumulation with once-daily administration compared with conventional dosing. Once-daily dosing should not be used for patients with endocarditis, renal failure, cystic fibrosis, ascites, or >20% surface area burns. The once-daily dose for gentamicin and tobramycin is 6 mg/kg, whereas for amikacin it is 15 mg/kg. The serum concentration of the aminoglycoside should be measured 6 hours before the next dose. This level should be ≤1 mg/L for gentamicin and tobramycin and ≤5 mg/mL for amikacin to avoid accumulation.
Principles
Formulae are only approximations; individual patients vary in their distribution and clearance of antibiotics despite apparently comparable degrees of renal or hepatic impairment. Thus, in any patient with significant impairment in excretion or metabolism of an antimicrobial, drug concentrations in plasma should be measured, particularly if the agent has a low therapeutic index.
When patients with severe renal failure are given antimicrobials, consider whether they are undergoing dialysis, and if so, by what technique. Some drugs such as gentamicin are appreciably removed by hemodialysis, whereas their removal by peritoneal dialysis usually is quantitatively less and unpredictable.
Conversely, most penicillins and cephalosporins are not removed well by either method of dialysis, since no dialysis apparatus comes close to the penicillin-secreting function of the renal tubule. The physician should refer to standard texts or consult with infectious disease physicians or nephrologists to determine whether replacement dosing is indicated after dialysis. In the neonate, particularly a premature one, pathways of metabolism or excretion may not be fully developed and modification of dosages may be required.
Furthermore, some classes of drugs may cause toxicity in neonates but not in older persons, e.g., chloramphenicol and the tetracyclines. In such patients, impaired excretion or metabolic pathways may lead to extraordinarily high plasma concentrations of antibiotics, which may cause significant toxicity. It is critical to know both the manifestations of dose-related toxicity and the normal elimination pathways of drugs, so that required dosage modifications can be made when antimicrobial clearance is impaired. Adverse effects due to changes in endogenous flora Whenever the natural flora are suppressed by the administration of an antimicrobial, other organisms proliferate (e.g., C. difficile). The emergence or overgrowth of flora resistant to a given antibiotic may cause superinfections that are more severe than the original infection itself (e.g., enterococci that develop high-level aminoglycoside or vancomycin resistance, or methicillin-resistant staphylococci).
We now recognize that several commonly used antibiotics (e.g., clindamycin and many β-lactams) produce a pseudomembranous colitis that may also be fatal if the cause of the characteristically profuse diarrhea is unrecognized. Strains of C. difficile that produce a cytotoxin are most commonly associated with antibiotic-associated colitis. Changes in bowel flora as a result of antibiotic administration may lead to decreased absorption of vitamin K with resultant bleeding in patients already receiving oral anticoagulants. However, controversy remains as to whether antibiotic-induced killing of intestinal bacteria results in hypoprothrombinemia.
Apparently, the vitamin K produced by these bacteria is not responsible for synthesis of clotting factors. Perhaps multiple factors are involved in the vitamin K-responsive hypoprothrombinemia observed in patients receiving antimicrobials.
Prolonged administration of oral neomycin may cause malabsorption; whether this is due to an alteration in normal bowel flora or to direct mucosal toxicity is not known. Another consequence of altered flora is the rising incidence of severe superinfections caused by fungi, especially in patients with cancer who receive treatment with broad-spectrum antimicrobials.
Use of Combinations of Antimicrobials
Combinations of antimicrobials are frequently used to treat infection; rational combinations are chosen after carefully considering several important principles. For a specific organism or organisms causing infection, combinations of antibiotic may be synergistic, antagonistic, or indifferent. All too frequently, antibiotic combinations are used to provide “broad-spectrum” coverage in response to the physician’s insecurity rather than a true medical indication. In some situations, broad-spectrum coverage is appropriate when treating a mixed infection, a rapidly worsening infection from pathogens whose precise identification is in progress, or a life-threatening infection caused by an organism that responds best to two synergistic agents. Combination therapy beyond these situations is unwarranted.
Principles
Appropriate uses of antimicrobial combinations include necessary synergy against an infecting organism, initial empiric treatment of life-threatening infections, or treatment of mixed infections.
Synergy
Antimicrobials may exert synergistic effects if they work at two different sites, involving either the same or different metabolic pathways in an organism. Examples of synergistic combinations of antibiotics are the combined use of penicillins (or cephalosporins) with aminoglycosides for killing certain aerobic bacteria. The mechanism of this synergy has been elucidated best in enterococci.
The resistance of enterococci to aminoglycosides may be mediated by either of two mechanisms. One is the failure of the compounds to enter the microbial cell in amounts sufficient to bind to ribosomes and thereby to alter translation of the genetic code by ribosomal RNA. The other mechanism is seen when entry is adequate but ribosomal binding and altered translation do not occur.
In the former situation, the potential for synergy between penicillins, cephalosporins, or vancomycin and the aminoglycosides exists because the “cell-wall-active” compounds act to facilitate entry of the aminoglycoside into the bacterial cell. The same principles probably apply to the mechanism of synergy between other cell-wall-active agents and the aminoglycosides in their action against staphylococci or gram-negative rods. The major determining factors in these situations are some activity of the cell-wall-active antibiotic against the organism and ribosomal activity of the aminoglycoside.
Another example of synergism of antibiotics is seen when amphotericin B is combined with flucytosine against fungi. Much of the synergistic effect is related to facilitated entry of the companion drug by amphotericin B. Synergy potentially exists when two drugs are active against an organism at different points along the same vital biologic pathway. An example of this mechanism is the use of a dihydrofolate reductase inhibitor (trimethoprim or pyrimethamine) together with a sulfonamide. Both classes of drugs inhibit folate metabolism, which ultimately prevents the transfer of methyl groups used in purine synthesis.
When these drug combinations are used, a microbicidal action frequently results. Despite suggestive in vitro data, convincing clinical evidence that synergistic drug combinations are superior to single drugs exists only for treatment of enterococcal endocarditis and when the combination trimethoprim-sulfamethoxazole is used. Considerable in vitro data suggest that most antimicrobial combinations are associated with antagonism (e.g., rifampin plus penicillins for staphylococci) or indifference (e.g., chloramphenicol plus rifampin for selected Haemophilus influenzae strains). Therefore, unless there is clear need or documentation of synergism, combination therapy should not be used.
Extended antimicrobial spectrum
Another indication (often abused) for the use of combinations of antimicrobials is to provide broad-spectrum coverage in the early empiric treatment of presumed life-threatening infections such as bacterial septicemia. The use of a combination such as a penicillinase-resistant penicillin or a cephalosporin with an aminoglycoside is based on the presumption that almost all likely organisms will be treated.
The potential for abuse of this type of reasoning is that all too often it is applied in clinical situations either as a substitute for adequate collection of data or to treat the physician’s insecurity. Furthermore, a second type of abuse may be observed in patients who respond to treatment and who eventually are found to have clinical isolates that are affected by only one member of the combination. Despite this new information, antibiotic combinations are frequently continued to the ultimate disadvantage of the patient, using the faulty reasoning that one does not like to tamper with “success.” Besides the potential harm to the patient offered by continued application of toxic or inappropriate drugs, this type of approach may be particularly deleterious in the hospital setting, where selecting out flora resistant to multiple drugs can be lethal.
Principles
Broad-spectrum coverage with antibiotics should not be a substitute for adequate collection of data or adequate clinical judgment. In most cases, single drugs with a selective spectrum of activity will provide adequate coverage for a successful outcome and will avoid the potential for toxic interactions or the selection of a highly resistant flora. In life-threatening situations, broad-spectrum coverage is justified initially; once the pathogen is identified, selective single-drug treatment is highly preferable.
Prevention of resistance
Another indication for the use of a combination of antibiotics is to prevent the emergence of strains of bacteria that are resistant to one or more of the drugs employed. The reasoning behind such an approach is simply an application of probability statistics. If the probability of an event (i.e., spontaneous mutation to a resistant strain) is known for two different antibiotics given separately, then the probability that resistance to two drugs will occur simultaneously is the product of those two probabilities. This rationale has been applied most successfully to the treatment of tuberculosis, since the rates of development of resistance to single drugs are high. This same strategy has been applied successfully to treatment of human immunodeficiency virus infection.
Treatment of mixed infections
Combination therapy is appropriate for the treatment of mixed infections. Such combinations may be useful even when the doses chosen exhibit antagonism. Chloramphenicol may antagonize the activity of aminoglycosides; this is usually of no clinical consequence, unless host factors involved in eradication of the organism during bacteriostatic therapy have been substantially inhibited.
Principles
As with any combination of drugs, theory must be backed by clinical hypothesis testing. Beware of potential antagonistic combinations of antimicrobials when selecting therapy for mixed infections. The least “active” drug in a combination must be adequate to treat potential pathogens by itself in case of overlapping sensitivities, particularly if compromised host factors are of overriding importance in the determination of a successful outcome.
Treatment Schedules
Treatment schedules must be designed to ensure adequate delivery of active antimicrobials to the site of infection for a sufficient time to promote a cure. In treating infections that pose a serious threat to life, or that are in sites where high plasma concentrations of antimicrobials are necessary to ensure penetration of tissue (e.g., in the endocardium and brain), parenteral antibiotics are required initially. In less severe infections with highly sensitive organisms and excellent delivery of antimicrobials into tissues, oral administration of drugs can be employed (e.g., most urinary tract infections, streptococcal pharyngitis, and some patients with pneumonia).
The duration of therapy with antibiotics is determined by the amount of antibiotic that must be administered over a given time to effect a cure of an active infection and prevent a relapse of the infection. The precise duration of therapy required for different infectious diseases to meet these criteria is poorly documented. A few generalizations based on clinical observations can be made and applied to specific clinical situations. The critical points in these considerations of duration of treatment revolve around:
- the ability of the organism to resist the normal host’s defense mechanisms;
- the physical location of the infection and its accessibility to therapeutic concentrations of an antimicrobial;
- the primary activity of the antimicrobial against the organisms, as determined by minimal inhibitory concentration or minimal microbicidal concentration;
- the frequency of development of resistance of the organism to the drug.
At least 4 to 6 weeks of treatment of osteomyelitis and endocarditis are recommended, since treatment failure or relapse rates are unacceptably high (with considerable morbidity) with shorter periods of therapy.
Conversely, urinary tract infections caused by the same organisms usually can be eradicated within several days of treatment because of very high urinary concentrations of effective drugs.
| Table Some Examples of Appropriate Combinations of Antimicrobials | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Table Suggested Treatment Regimens for Defined Infections | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Infections with intracellular pathogens require prolonged therapy. For example, staphylococci are capable of survival intracellularly; this site provides a sanctuary from antibiotic activity. Relapse of staphylococcal infections is common unless treatment is prolonged.
The longest durations of treatment are used for mycobacterial infections. Mycobacterium leprae’s intracellular location and slow generation time (10 to 15 days) limit the ability of the sulfones to inhibit their replication.
Treatment of extrapulmonary tuberculosis must often be continued for 12 to 24 months, depending upon the severity of disease and the sensitivity of the organisms to drugs.
Treatment for infections caused by bacteria that are promptly killed by phagocytosis need not be long, unless endocarditis is present.
Pneumococcal pneumonia is a good example of this type of infection. Cures have been reported with a single dose of penicillin. However, the usual duration of treatment of pneumococcal pneumonia is a minimum of 5 days or until the patient has been afebrile for at least 3 days. Remember that the recommended duration of therapy is a minimum.
These schedules should be considered only as guidelines. It is quite possible that duration of antimicrobial therapy might need to be longer in a specific patient who is not responding promptly.
Factors Responsible for Failure of Treatment
Failure of treatment can result from a number of different factors, including the following:
- Antimicrobial agents that have poor in vitro activity against the infecting organism were selected.
- Active antimicrobials were not delivered to the site of infection because an inappropriate route of drug administration was selected, the wrong dosage was used, or abnormal pharmacokinetic variables in the patients went undetected.
- The location of the infection (as in endocarditis) was inaccessible.
- The host defenses were inadequate.
- Infection was not treated for a sufficient period to prevent relapse.
- Serious toxicity necessitated discontinuation of therapy.
- Antimicrobial resistance by the infecting organism developed.
- Superinfection occurred.
- There is a foreign body or abscess.
- The patient did not comply with the therapeutic regimen.
Much of the potential for failure of treatment can be obviated by basing the initial choice of antimicrobials on principles enumerated in the section on determinants of antimicrobial efficacy and by considering how host defenses can be altered in favor of eradication of the organism. Whenever host defense mechanisms are inadequate, the physician should try to choose antimicrobials that kill the infecting organism rather than merely inhibiting its growth.
The high relapse rate (approximately 50%) seen when patients with endocarditis are treated with bacteriostatic rather than bactericidal drugs is ample testimony to this principle, as are the poor results from inappropriate use of bacteriostatic agents in patients with severe neutropenia.
Mechanisms of Resistance to Antimicrobials
Organisms may develop resistance to antimicrobials in a number of different ways. Well-recognized mechanisms of bacterial resistance to antibiotics include enzymatic inhibition, changes in permeability in the outer and inner bacterial membranes, alterations in structure or production of new targets of the antimicrobials, promotion of efflux pump mechanisms, and selection of auxotrophs that escape drug effects. Use of antimicrobials encourages the selection of resistant strains. Therefore, indiscriminate use of antimicrobials selects for resistance and promotes infection that will be even more difficult to eradicate with antimicrobials.
The more common mechanisms of resistance associated with failure of treatment are enzymatic inhibition, alteration in membrane permeability, and alteration in drug targets. The enzymes known as β-lactamases are commonly produced by a variety of bacteria. They are responsible for cleavage of the β-lactam ring, thereby destroying the ability of susceptible β-lactam antibiotics to bind to bacterial penicillin-binding proteins on the inner bacterial membrane.
Other microorganisms develop resistance to antimicrobials by restricting their entry into the interior of the microbial cell. This mechanism is illustrated by the intrinsic resistance of enterococci to aminoglycosides. This type of resistance can be overcome by combining aminoglycoside therapy with penicillin. The penicillin facilitates entry of the aminoglycoside into bacteria.
Bacteria may be drug-resistant because they modify the target site for antibiotic action or binding. For example, some enterococci resist the action of aminoglycosides at the ribosomal level, despite facilitated entry of the drug by penicillins. Plasmid-bound antimicrobial resistance factors (R factors), possessed by aerobic gram-negative rods, are readily transmitted among various species of microorganisms within a hospital setting.
Monitoring Results of Treatment
The success or failure of treatment should be carefully assessed.
Useful endpoints to follow include the temperature curves, leukocyte counts, elevation of the erythrocyte sedimentation rate or C-reactive protein during chronic infections such as osteomyelitis, direct demonstration of the extent of tissue injury by radiography, abnormal elevations of concentrations of tissue-derived enzymes in the circulation, and many other tests.
Bacteriologic cultures and serologic tests should be repeated periodically. The intervals at which such tests should be obtained depend on the type, the severity, the site of the infection, the typical course when treatment is successful, and the interval at which most relapses occur.
Failure of treatment of an infection can be defined as failure to produce a clinical remission of disease or as relapse of the infection once treatment is stopped. When treatment does fail, the physician must once again base rational therapy on general principles, including the following:
- Reestablishing the microbial cause of infection and determining whether it is caused by the original or a new agent (superinfection)
- Redetermining the in vitro sensitivity of the isolated organism to the antibiotics
- Checking plasma concentration of antibiotics and their in vitro activity to determine whether they remain adequate in vivo, and to confirm that the patient is actually receiving the medication
- Making certain that treatment failure is not due to a failure on the part of the physician to identify correctable host factors that contribute to infection (e.g., failure to drain abscesses, remove foreign bodies and necrotic tissues)
- Excluding the possibility that persisting evidence of inflammation is not due to factors other than the infection (e.g., drug hypersensitivity) and phlebitis from intravenous catheters
Inappropriate Uses of Antimicrobials
Errors in the treatment of infectious diseases fall into two major categories: omission, in which treatment is indicated but none is given, and commission, in which treatment is given but it is inadequate or inappropriate. From statistics on the use of antibiotics in the United States, it is clear that most errors are of commission (see Table Errors of Commission in Antimicrobial Therapy).
Principles
To maximize the benefit/risk ratio of prescribing, constantly reexamine your reasons for treatment and reevaluate your knowledge of the potential hazards of the prescribed drugs.
| Table Errors of Commission in Antimicrobial Therapy | |
|