The characteristics that need to be considered when administering antibiotics include absorption (when dealing with oral antibiotics), volume of distribution, metabolism, and excretion. These factors determine the dose of each drug and the time interval of administration. To effectively clear a bacterial infection, serum levels of the antibiotic need to be maintained above the minimum inhibitory concentration for a significant period. For each pathogen, the minimum inhibitory concentration is determined by serially diluting the antibiotic into liquid medium containing 104 bacteria per millihter. Inoculated tubes are incubated overnight until broth without added antibiotic has become cloudy or turbid as a result of bacterial growth. The lowest concentration of antibiotic that prevents active bacterial growth — that is, the liquid media remains clear — constitutes the minimum inhibitory concentration.
Automated analyzers can now quickly determine, for individual pathogens, the minimum inhibitory concentrations for multiple antibiotics, and these data serve to guide the physicians choice of antibiotics. The mean bactericidal concentration is determined by taking each clear tube and inoculating a plate of solid medium with the solution. Plates are then incubated to allow colonies to form. The lowest concentration of antibiotic that blocks all growth of bacteria — that is, no colonies on solid medium — represents the mean bactericidal concentration. Successful cure of an infection depends on multiple host factors in addition to serum antibiotic concentration.
However, investigators have attempted to predict successful treatment by plotting serum antibiotic levels against time. Three parameters can be assessed: time above the minimum inhibitory concentration (T>minimum inhibitory concentration), ratio of the peak antibiotic concentration to the minimum inhibitory concentration (Cmax/minimum inhibitory concentration), and the ratio of the area under the curve to the minimum inhibitory concentration (area under the curve/minimum inhibitory concentration). Cure rates for β-lactam antibiotics are maximized by maintaining serum levels above the minimum inhibitory concentration for >50% of the time.
Peak antibiotic concentrations are of less importance for these antibiotics, and serum concentrations above 8 times the minimum inhibitory concentration are of no benefit other than to enhance penetration into less permeable body sites. Unlike β-lactam antibiotics, aminoglycosides and flu-oroquinolones demonstrate concentration-dependent killing. In vitro studies show that these antibiotics demonstrate greater killing the more their concentrations exceed the minimum inhibitory concentration. High peak levels of these antibiotics may be more effective than low peak levels at curing infections.
Therefore, for treatment with aminoglycosides and fluoroquinolones Cmax/minimum inhibitory concentration and area under the curve/minimum inhibitory concentration are more helpful for maximizing effectiveness. In the treatment of gram-negative bacteria, aminoglycosides have been suggested to achieve maximal effectiveness when Cmax/minimum inhibitory concentration is 10 to 12. For fluoroquinolones, best outcomes in community-acquired pneumonia may be achieved when the area under the curve/minimum inhibitory concentration is >34. To prevent the development of fluo-roquinolone resistance to S. pneumoniae, in vitro studies have suggested that area under the curve/minimum inhibitory concentration should be >50. For P. aeruginosa, an area under the curve/minimum inhibitory concentration of >200 is required. In vitro studies also demonstrate that aminoglycosides and fluoroquinolones demonstrate a post-antibiotic effect: when the antibiotic is removed, a delay in the recovery of bacterial growth occurs.
Gram-negative bacteria demonstrate a delay of 2 to 6 hours in the recovery of active growth after aminoglycosides and fluoroquinolones, but no delay after penicillins and cephalosporins. But penicillins and cephalosporins generally cause a 2-hour delay in the recovery of gram-positive organisms. Investigators suggest that antibiotics with a significant post-antibiotic effect can be dosed less frequently; those with no post-antibiotic effect should be administered by constant infusion. Although these in vitro effects suggest certain therapeutic approaches, it must be kept in mind that concentration-dependent killing and post-antibiotic effect are both in vitro phenomena, and treatment strategies based on these effects have not been substantiated by controlled human clinical trials.
About Antibiotic Dosing
- Absorption, volume of distribution, metabolism, and excretion all affect serum antibiotic levels.
- Mean inhibitory concentration is helpful in guiding antibiotic choice.
- To maximize success with β-lactam antibiotics, serum antibiotic levels should be above the minimum inhibitory concentration for at least 50% of the time (T>minimum inhibitory concentration > 50%).
- To maximize success with aminoglycosides and fluoroquinolones, high peak concentration, Cmax/minimum inhibitory concentration, and high area under the curve/minimum inhibitory concentration ratio are recommended.
- The clinical importance of concentration-dependent killing and post-antibiotic effect for aminoglycosides and fluoroquinolones remain to be proven by clinical trials.
Basic Strategies
For Antibiotic Therapy
The choice of antibiotics should be carefully considered. A step-by-step logical approach is helpful.
1. Decide Whether The Patient Has a Bacterial Infection
One test that has traditionally been used to differentiate an acute systemic bacterial infection from a viral illness is the peripheral white blood cell count. In patients with serious systemic bacterial infections, the peripheral white blood cell count may be elevated and may demonstrate an increased percentage of neutrophils. On occasion, less mature neutrophils such as band forms and, less commonly, metamyelocytes are observed on peripheral blood smear. Most viral infections fail to induce a neutrophil response. Viral infections, particularly Epstein-Barr virus, induce an increase in lymphocytes or monocytes (or both) and may induce the formation of atypical monocytes. Unfortunately, the peripheral white blood cell count is only a rough guideline, lacking both sensitivity and specificity. Recently, serum procalcitonin concentration has been found to be a far more accurate test for differentiating bacterial from viral infection. In response to bacterial infection, this precursor of calcitonin is synthesized and released into the serum by many organs of the body; production of interferon in response to viral infection inhibits its synthesis. The serum procalcitonin test may also be of prognostic value, serum procalcitonin levels being particularly high in severe sepsis.
2. Make a Reasonable Statistical Guess as to the Possible Pathogens
Based on the patient’s symptoms and signs, as well as on laboratory tests, the anatomic site of the possible infection can often be determined. For example, burning on urination, associated with pyuria on urinalysis, suggests a urinary tract infection. The organisms that cause uncomplicated urinary tract infection usually arise from the bowel flora. They include E. coli, Klebsiella, and Proteus. Antibiotic treatment needs to cover these potential pathogens. Later chapters review the pathogens commonly associated with infections at specific anatomic sites and the recommended antibiotic coverage for those pathogens.
3. Be aware of the Antibiotic Susceptibility Patterns in Your Hospital and Community
In patients that develop infection while in hospital (“nosocomial infection), empiric therapy needs to take into account the antibiotic susceptibility patterns of the flora associated with the hospital and the floor where the patient became ill. Many hospitals have a high incidence of methicillin-resistant Staphylococcus aureus and therefore empiric antibiotic treatment for a possible staphylococcal infection must include vancomycin, pending culture results. Other hospitals have a large percentage of Pseudomonas strains that are resistant to gentamicin, eliminating that antibiotic from consideration as empiric treatment of possible gram-negative sepsis. In many communities, individuals who have never been hospitalized are today presenting with soft-tissue infections caused by Community-acquired methicillin-resistant Staphylococcus aureus, and physicians in these communities must adjust their empiric antibiotic selection.
4. Take into Account Previous Antibiotic Treatment
The remarkable adaptability of bacteria makes it highly likely that a new pathogen will be resistant to previously administered antibiotics. If the onset of the new infection was preceded by a significant interval when antibiotics were not given, the resident flora may have recolonized with less resistant flora. However, the re-establishment of normal flora can take weeks, and patients in hospital are likely to recolonize with highly resistant hospital flora.
5. Take into Consideration Important Host Factors
- Penetration into the site of infection. For example, patients with bacterial meningitis should not be treated with antibiotics that fail to cross the blood-brain barrier (examples include lst-generation cephalosporins, gentamicin, and clindamycin).
- Peripheral white blood cell count. Patients with neutropenia have a high mortality rate from sepsis. Immediate broad-spectrum, high-dose intravenous antibiotic treatment is recommended as empiric therapy for these patients.
- Age and underlying diseases (hepatic and renal dysfunction). Elderly patients tend to metabolize and excrete antibiotics more slowly; longer dosing intervals are therefore often required. Agents with significant toxicity (such as aminoglycosides) should generally be avoided in elderly patients because they exhibit greater toxicity. Antibiotics metabolized primarily by the liver should generally be avoided or reduced in patients with significant cirrhosis. In patients with significant renal dysfunction, antibiotic doses need to be modified.
- Duration of hospitalization. Patients who have just arrived in the hospital tend to be colonized with community-acquired pathogens; patients who have been in the hospital for prolonged periods and have received several courses of antibiotics tend to be colonized with highly resistant bacteria and with fungi.
- Severity of the patient’s illness. The severely ill patient who is toxic and hypotensive requires broad-spectrum antibiotics; the patient who simply has a new fever without other serious systemic complaints can usually be observed off antibiotics.
6. Use the Fewest Drugs Possible
- Multiple drugs may lead to antagonism rather than synergy. Some regimens, such as penicillin and an aminoglycoside for Enterococcus, have been shown to result in synergy — that is, the combined effects are greater than simple addition of the MBCs of the two agents would suggest. In other instances, certain combinations have proved to be antagonistic. The use of rifampin combined with oxacillin is antagonistic in some strains of S. aureus, for example. Many combination regimens have not been completely studied, and the natural assumption that more antibiotics lead to more killing power often does not apply.
- Use of multiple antibiotics increases the risk of adverse reactions. Drug allergies are common. When a patient on more than one antibiotic develops an allergic reaction, all antibiotics become potential offenders, and these agents can no longer be used. In some instances, combination therapy can increase the risk of toxicity. The combination of gentamicin and vancomycin increases the risk of nephrotoxicity, for example.
- Use of multiple antibiotics often increases costs and the risk of administration errors. Administration of two or more intravenous antibiotics requires multiple intravenous reservoirs, lines, and pumps. Nurses and pharmacists must dispense each antibiotic dose, increasing labor costs. The more drugs a patient receives, the higher the probability of an administration error. Use of two or more drugs usually increases the acquisition costs.
- Use of multiple antibiotics increases the risk of infection with highly resistant organisms. Prolonged use of broad-spectrum antibiotic coverage increases the risk of infection with methicillin-resistant Staphylococcus aureus, vancomycin-resistant Enterococcus, multiresistant gram-negative bacilli, and fungi. When multiple antibiotics are used, the spectrum of bacteria killed increases. Killing most of the normal flora in the pharynx and gastrointestinal tract is harmful to the host. The normal flora compete for nutrients, occupy binding sites that could otherwise be used by pathogenic bacteria, and produce agents that inhibit the growth of competitors. Loss of the normal flora allows resistant pathogens to overgrow.
7. Switch to Narrower-Spectrum Antibiotic Coverage Within 3 Days
Within 3 days following the administration of antibiotics, sequential cultures of mouth flora reveal that the numbers and types of bacteria begin to change significantly. The normal flora die, and resistant gram-negative rods, gram-positive cocci, and fungi begin to predominate. The more quickly the selective pressures of broad-spectrum antibiotic coverage can be discontinued, the lower the risk of selecting for highly resistant pathogens. Broad coverage is reasonable as initial empiric therapy until cultures are available. By the 3rd day, the microbiology laboratory can generally identify the pathogen or pathogens, and a narrower-spectrum, specific antibiotic regimen can be initiated.
Table Classification of Antibiotics by Spectrum of Activity
| Narrow | Moderately Broad | Broad | Very Broad |
| Penicillin | Ampicillin | Ampicillin-sulbactam | Ticarcillin-clavulinate |
| Oxacillin/Nafcillin | Ticarcillin | Amoxicillin-clavulanate | Piperacillin-tazobactam |
| Cefazolin | Piperacillin | Ceftriaxone, | Cefepime |
| Cephalexin/Cephradine | Cefoxitin | Cefotaxime | Imipenem |
| Aztreonam | Cefotetan | Ceftizoxime | Meropenem |
| Aminoglycosides | Cefuroxime-axetil | Ceftazidime | Ertapenem |
| Vancomycin | Cefaclor | Cefixime | Gatifloxacin |
| Macrolides | Ciprofloxacin | Cefpodoxime proxetil | Moxifloxacin |
| Clindamycin | Azithromycin | Tetracycline | Tigecycline |
| Linezolid | Clarithromycin | Doxycycline | |
| Quinupristin/dalfopristin | Talithromycin | Chloramphenicol | |
| Daptomycin | Trimethoprim-sulfamethoxazole | Levofloxacin | |
| Metronidazole |
Despite the availability of culture results, clinicians too often continue the same empiric broad-spectrum antibiotic regimen, and that behavior is a critical factor in explaining subsequent infections with highly resistant superbugs.
Obey the 3-day rule. Continuing broad-spectrum antibiotics beyond 3 days drastically alters the host’s resident flora and selects for resistant organisms. After 3 days, streamline antibiotic coverage. Use narrower-spectrum antibiotics to treat the specific pathogens identified by culture and Gram stain.
8. All Else Being Equal, Choose The Least Expensive Drug
As is discussed in later chapters, more than one antibiotic regimen can often be used to successfully treat a specific infection. Given the strong economic forces driving medicine today, the physician needs to consider the cost of therapy whenever possible. Too often, new, more expensive antibiotics are chosen over older generic antibiotics that are equally effective. In this book, the review of each specific antibiotic tries to classify that antibiotic’s cost range to assist the clinician in making cost-effective decisions. However, in assessing cost, factoring in toxicity is also important. For example, the acquisition cost of gentamicin is low, but when blood-level monitoring, the requirement to closely follow blood urea nitrogen and serum creatinine, and the potential for an extended hospital stay because of nephrotoxicity are factored into the cost equation, gentamicin is often not cost-effective.
About the Steps Required to Design an Antibiotic Regimen
- Assess the probability of bacterial infection. (Antibiotics should be avoided in viral infections.)
- Be familiar with the pathogens primarily responsible for infection at each anatomic site.
- Be familiar with the bacterial flora in the local hospital and community.
- Take into account previous antibiotictreatment.
- Take into account the specific host factors (age, immune status, hepatic and renal function, duration of hospitalization, severity of illness).
- Use the minimum number and narrowest spectrum of antibiotics possible.
- Switch to a narrower-spectrum antibiotic regimen based on culture results.
- Take into account acquisition cost and the costs of toxicity.
Obey the 3-day rule. Continuing broad-spectrum antibiotics beyond 3 days drastically alters the host’s normal flora and selects for resistant organisms. After 3 days streamline the antibiotics. Use narrower-spectrum antibiotics to treat the specific pathogens identified by culture and Gram stain.
Colonization Versus Infection
Case 1
Following a motor vehicle accident, a 40-year-old man was admitted to the intensive care unit with 4 fractured ribs and a severe lung contusion on the right side. Chest X-ray demonstrated an infiltrate in the right lower lobe. Because of depressed mental status, this man required respiratory support. Initially, Gram stain of the sputum demonstrated few polymorphonuclear leukocytes and no organisms. On the third hospital day, this patient developed a fever to 103°F (39.5°C), and his peripheral white blood cell increased to 17,500 from 8000 (80% polymorphonuclear leukocytes, 15% band forms). A new Chest X-Ray demonstrated extension of the right lower lobe infiltrate. Gram stain of sputum revealed abundant polymorphonuclear leukocytes and 20 to 30 gram-positive cocci in clusters per high-power field. His sputum culture grew methicillin-sensitive S.aureus.Intravenouscefazolin (1.5g every8 hours) was initiated. He defervesced, and secretions from his endotracheal tube decreased over the next 3 days. On the fourth day, a repeat sputum sample was obtained. Gram stain revealed a moderate number of polymorphonuclear leukocytes and no organisms; however, culture grew E. coli resistant to cefazolin. The physician changed the antibiotic to intravenous cefepime (1 g every 8 hours).
Case 1 represents a very typical example of how antibiotics are misused. The initial therapy for a probable early S. aureus pneumonia was appropriate, and the patient responded (fever resolved, sputum production decreased, gram-positive cocci disappeared from the Gram stain, and S. aureus no longer grew on culture). However, because the sputum culture was positive for a resistant E. coli, the physician switched to a broader-spectrum antibiotic. The correct decision should have been to continue cefazolin. One of the most difficult and confusing issues for many physicians is the interpretation of culture results. Wound cultures and sputum cultures are often misinterpreted. Once a patient has been started on an antibiotic, the bacterial flora on the skin and in the mouth and sputum will change. Often these new organisms do not invade the host, but simply represent new flora that have colonized these anatomic sites. Too often, physicians try to eradicate the new flora by adding new more-powerful antibiotics. The result of this strategy is to select for organisms that are multiresistant.
The eventual outcome can be the selection of a bacterium that is resistant to all antibiotics. No definitive method exists for differentiating between colonization and true infection. However, several clinical findings are helpful in guiding the physician. Evidence supporting the onset of a new infection include a new fever or a change in fever pattern, a rise in the peripheral white blood cell with a increase in the percentage of polymorphonuclear leukocytes and band forms (left shift), Gram stain demonstrating an increased number of polymorphonuclear leukocytes in association with predominance of bacteria that are morphologically consistent with the culture results. In the absence of these findings, colonization is more likely, and the current antibiotic regimen should be continued.
About Differentiating Colonization from Infection
- Growth of resistant organisms is the rule in the patient on antibiotics.
- Antibiotics should be switched only on evidence of a new infection.
- Evidence for a new superinfection includes a) new fever or a worsening fever pattern, b) increased peripheral leukocyte count with left shift, c) increased inflammatory exudate at the original site of infection, d) increased polymorphonuclear leukocytes on Gram stain, and e) correlation between bacterial morphology and culture on Gram stain.