Correlation between in vitro and in vivo activity of antimicrobial agents against gram-negative bacilli in a murine infection model

Treatment of experimental pneumonia due to penicillin-resistant Streptococcus pneumoniae in immunocompetent rats

A model of pneumonia due to Streptococcus pneumoniae resistant to penicillin was developed in immunocompetent Wistar rats and was used to evaluate the efficacies of different doses of penicillin, cefotaxime, cefpirome, and vancomycin. Adult Wistar rats were challenged by intratracheal inoculation with 3 x 10(9) CFU of one strain of S. pneumoniae resistant to penicillin (MICs of penicillin, cefotaxime, cefpirome, and vancomycin, 2, 1, 0.5, and 0.5 microg/ml, respectively) suspended in brain heart broth supplemented with 0.7% agar. The rats experienced a fatal pneumonia, dying within 5 days and with peak mortality (70 to 80%) occurring 48 to 72 h after infection, and the bacterial counts in the lungs persisted from 8.87 +/- 0.3 log10 CFU/g of lung at 24 h of the infection to 9.1 +/- 0.3 log10 CFU/g at 72 h. Four hours after infection the animals were randomized into the following treatment groups: (i) control without treatment, (ii) penicillin G at 100,000 IU/kg of body weight every 2 h, (iii) penicillin G at 250,000 IU/kg every 2 h, (iv) cefotaxime at 100 mg/kg every 2 h, (v) cefpirome at 200 mg/kg every 2 h, and (vi) vancomycin at 50 mg/kg every 8 h. Two different protocols were used for the therapeutic efficacy studies: four doses of beta-lactams and one dose of vancomycin or eight doses of beta-lactams and two doses of vancomycin. Results of the therapy for experimental pneumonia caused by penicillin-resistant S. pneumoniae showed that initially, all the antimicrobial agents tested had similar efficacies, but when we prolonged the treatment, higher doses of penicillin, cefotaxime, and cefpirome were more effective than penicillin at lower doses in decreasing the residual bacterial titers in the lungs. Also, when we extended the treatment, vancomycin was more efficacious than penicillin at lower doses but was less efficacious than higher doses of penicillin or cefpirome. The model that we have developed is simple and amenable for inducing pneumonia in immunocompetent rats and could be used to explore the pathophysiology and to evaluate optimal therapy of this infection in the immunocompetent host.

Use of a new mouse model of Acinetobacter baumannii pneumonia to evaluate the postantibiotic effect of imipenem

Acinetobacter baumannii is responsible for severe nosocomial pneumonia. To evaluate new therapeutic regimens for infections due to multiresistant strains and to study the pharmacodynamic properties of various antibiotics, we developed an experimental mouse model of acute A. baumannii pneumonia. C3H/HeN mice rendered transiently neutropenic were infected intratracheally with 5 x 10(6) CFU of A. baumannii. The mean log10 CFU/g of lung homogenate (+/- the standard deviation) were 9 +/- 0.9, 9.4 +/- 0.8, 8.6 +/- 1.2, and 7.7 +/- 1.4 on days 1, 2, 3, and 4 postinoculation. The lung pathology was characterized by pneumonitis with edema and a patchy distribution of hemorrhages in the peribronchovascular spaces of both lungs. Abscesses formed on days 3 and 4. Four days after inoculation, subacute pneumonitis characterized by alveolar macrophage proliferation and areas of fibrosis was observed. The cumulative mortality on day 4 was 85%. This new model was used to study the effects of 1, 2, or 3 50-mg/kg doses of imipenem. Imipenem concentrations in lungs were above the MIC for 2 h after the last dose. The in vivo postantibiotic effect (PAE) was determined during the 9-h period following the last dose; it decreased in duration with the number of doses: 9.6, 6.4, and 4 h after 1, 2, and 3 50-mg/kg doses, respectively. In contrast, no in vitro PAE was observed. This model offers a reproducible acute course of A. baumannii pneumonia. The presence of a prolonged in vivo PAE supports the currently recommended dosing intervals of imipenem for the treatment of human infections due to A. baumannii, i.e., 15 mg/kg three times a day.

Experience of once-daily aminoglycoside dosing using a target area under the concentration-time curve

BACKGROUND

Many centres are changing to once-daily aminoglycoside administration. However, proposed methods for this practice often have theoretical and practical difficulties. We have developed a method in which a target area under the concentration-time curve (AUC) is used instead of traditional peak and trough serum concentrations.

AIMS

To analyse our experience with the target AUC method in the first 100 courses of once-daily aminoglycoside administration in the Christchurch, New Zealand hospitals.

METHODS

Following a starting dose of 5-7 mg/kg, administered by 30-minute infusion, the AUC was calculated using two serum aminoglycoside concentrations taken at one and six-14 hours after the start of the infusion. Dose adjustment was made to correct for any difference between the calculated AUC and a target AUC (72-101 mg.1(-1).h). The method was assessed for practicality and precision in 100 courses of treatment. The incidence of aminoglycoside toxicity was documented.

RESULTS

The mean final dose of 6.68 mg/kg, and AUC of 92.8 mg.1(-1).h, were significantly different from the mean starting dose and AUC of 5.67 mg/kg and 74.0 mg.1(-1).h, respectively. The method appeared to be more precise than empirical dosing at achieving the target AUC even though the final recommended dose had more variability than the starting dose. Although the study was uncontrolled, observed nephrotoxicity (2%) and ototoxicity (up to 6.9%) were no greater than expected from the results of other studies. There were no deaths related to antibiotic failure.

CONCLUSIONS

The AUC method was practical, and more appropriate for once-daily dosing than the conventional method of aiming for target peak and trough concentrations. Dose adjustment can be made before the next dose.

Experimental Streptococcus pneumoniae infection in mice for studying correlation of in vitro and in vivo activities of penicillin against pneumococci with various susceptibilities to penicillin

The purpose of the study was to investigate the correlation of in vitro activity with the in vivo effect and the pharmacokinetics of penicillin in the treatment of infections with pneumococci with various susceptibilities to penicillin. We used 10 pneumococcal strains for which penicillin MICs ranged from 0.016 to 8 micrograms/ml. Time-kill curve experiments were performed with all strains. We found that the effect of penicillin in vitro is concentration independent, with a maximum effect at two to four times the MIC for penicillin-susceptible as well as penicillin-resistant pneumococci. The mouse peritonitis model with an inoculum of approximately 10(6) CFU, to which mucin was added, resulted in a reproducible lethal infection with the pneumococci. The 50% effective dose was determined for each strain, and we found a highly significant correlation between the log MIC and the log 50% effective dose of penicillin against these strains (P < 0.01). Furthermore, it was shown that the most important pharmacokinetic parameter determining the effect of penicillin in vivo was the time that the concentration of penicillin in serum was greater than the MIC.

Pharmacodynamic (kinetic) considerations in the treatment of moderately severe infections with cefotaxime

Information about the pharmacodynamics of beta-lactams has accumulated rapidly over the last 20 years, and their application to cefotaxime are discussed in this review. Application of pharmacodynamics requires an integration of the pharmacokinetic and in vitro properties of the agent. Cefotaxime is similar to other beta-lactams in that it has little concentration-dependent killing and produces no postantibiotic effect against Gram-negative bacteria. However, it has a microbiologically active metabolite, deascetylcefotaxime, which can show synergy, partial synergy, or an additive effect in combination with the parent drug. More than any other technique, animal models have been able to elucidate the pharmacokinetic parameters that predict efficacy in vivo. They have shown that for beta-lactams it is the time that levels exceed the minimum inhibitory concentration (MIC) that is the most important determinant of efficacy. For bacteria to have no postantibiotic effect, plasma levels need to exceed the MIC for the whole of the dosing interval to achieve maximum killing at the site of infection. When applying these concepts as the most stringent criteria for efficacy using pharmacokinetic values from young, healthy volunteers, it can be shown that organisms with MICs of < or = 0.03 microgram/ml for a 1-g dose and 0.06 microgram/ml for a 2-g dose to achieve optimum efficacy with 12-h dosing of cefotaxime. However, two clinical studies have demonstrated trough levels much greater than would be predicted from these pharmacokinetic values, as a result of the effects of decreased renal function accompanying sepsis and older age. These studies showed that organisms with MICs < or = 1 microgram/ml for a 1-g dose or 2 micrograms/ml for a 2-g 12-h dose were covered for the whole of the dosing interval. Thus, all strains of Enterobacteriaceae and pathogenic Neisseria spp. that lack resistance mechanisms to third-generation cephalosporins would be covered using 12-h dosing schedules.

Optimal adaptive control of pharmacodynamic effects with aminoglycoside antibiotics: a required approach for the future

Many factors are considered in designing optimal drug regimens, including dose, route of administration, and the frequency and rate of administration. It is evident that the first parameter to be considered in dosage regimen determinations is the pharmacokinetics of the drug in the treated patient, or more extensively in the corresponding specific population of patients characterized by a similar pathophysiological status, e.g. neutropenics, ICU patients, subjects with renal insufficiency, children, geriatrics... Therefore, over the last two decades, pharmacokinetic principles have widely, and almost exclusively influenced the optimal adaptative control of antibiotics with a small therapeutic index. The large number of available new drugs associated with medical progress, specially in considering the increase of surviving patients with severe diseases or precarious physiopathological status (e.g., neutropenia, AIDS, ICU patients ...) have demonstrated the limits of this approach and pointed out the necessity of the determination of suitable parameters reflecting the in vivo efficacy. Thus, the search for pharmacodynamic parameters that can be accurately related to pharmacokinetic profiles in patients becomes a major tool. The progress in measurement of antibiotics and the additional dimension of the antibiotic's efficacy evaluation based on the analysis of the time course of the antibacterial effects, have provided the basis for the coupled evaluation of pharmacokinetics and pharmacodynamics of antibiotics, which has been largely assessed by in vivo animal models. In an attempt to propose a methodology that can provide information directly applicable to patients to be treated, we have developed a human in vivo/ex vivo model of analysis of the pharmacodynamics of antibiotics. This model has been applied to several drugs and particularly to aminoglycosides. It allows one to have a more rational evaluation of the dose-concentration-effect relations for antibiotics in man. The pharmacodynamics of aminoglycosides is characterized by a strong concentration dependent bactericidal activity and a significant post-antibiotic effect against Gram-negative bacilli. Experimental animal studies confirmed by investigations in man have shown that in vivo efficacy is enhanced by increasing the Cmax and the AUC of these antibiotics regardless of the possible below-MIC value of the C min. Moreover, the in vivo efficacy seems to be the same whether the total dose of aminoglycoside is given as a once a day dosing or as several doses while the toxic risk appears to be attenuated by the first dosage schedule. These data also emphasize the need for the development of models and software for optimal adaptive control including the pharmacodynamic parameters of antibiotics.

Behavior of antibiotics during human necrotizing pancreatitis

The aim of the study was to verify whether antibiotics excreted by the normal pancreas are also excreted in human necrotizing pancreatitis, reaching the tissue sites of the infection. Twelve patients suffering from acute necrotizing pancreatitis were treated with imipenem-cilastatin (0.5 g), mezlocillin (2 g), gentamicin (0.08 g), amikacin (0.5 g), pefloxacin (0.4 g), and metronidazole (0.5 g). Serum and necrotic samples were collected simultaneously at different time intervals after parenteral drug administration by computed tomography-guided needle aspiration, intraoperatively, and from surgical drainages placed during surgery. Drug concentrations were determined by microbiological and high-performance liquid chromatography assays. All antibiotics reached the necrotic tissues, but with varying degrees of penetration, this being low for aminoglycosides (13%) and high in the case of pefloxacin (89%) and metronidazole (99%). The concentrations of pefloxacin (13.0 to 23 micrograms/g) and metronidazole (8.4 micrograms/g) in the necrotic samples were distinctly higher than the MICs for the organisms most commonly isolated in this disease; the concentrations in tissue of imipenem (3.35 micrograms/g) and mezlocillin (8.0 and 15.0 micrograms/g) did not always exceed the MICs for 90% of strains tested, whereas the aminoglycoside concentrations in necrotic tissue (0.5 microgram/g) were inadequate. Repeated administration of drugs (for 3, 7, 17, and 20 days) seems to enhance penetration of pefloxacin, imipenem, and metronidazole into necrotic pancreatic tissue. The choice of antibiotics in preventing infected necrosis during necrotizing pancreatitis should be based on their antimicrobial activity, penetration rate, persistence, and therapeutic concentrations in the necrotic pancreatic area. These requisites are provided by pefloxacin and metronidazole and to a variable extent by imipenem and mezlocillin.

Qualitative susceptibility tests versus quantitative MIC tests

Qualitative susceptibility categories show reasonable, but incomplete, correlation with therapeutic outcome. Studies using quantitative MIC tests have demonstrated that treatment failures within the susceptible category are associated with higher minimum inhibitory concentrations (MICs) than with therapeutic successes. Other trials have exhibited enhanced response for increasing ratios of a pharmacokinetic parameter to MIC (for example, peak level to MIC ratio for aminoglycosides). Dose-response studies in animal infection models also demonstrate an excellent correlation between the dose of drug required for a given response and the infecting organism MIC. These studies suggest that the use of quantitative MIC tests may enable more individualization of the therapeutic regimen, especially in regards to dose and dosing frequency, than provided by qualitative category susceptibility tests. However, there are only rare studies that have used MIC results or pharmacokinetic parameters to improve efficacy. Furthermore, these studies have not consistently documented enhanced clinical efficacy. MICs can also be used to reduce drug dosage and cost of antimicrobial therapy for very susceptible organisms. Additional studies are clearly needed to define the full potential of the quantitative MIC test result.

In vitro pharmacodynamic effects of concentration, pH, and growth phase on serum bactericidal activities of daptomycin and vancomycin

Clinical trials with daptomycin were halted in December 1990 because of treatment failures including two resistant Staphylococcus aureus strains. High protein binding of daptomycin (> 90%) and the lower-than-expected concentrations in serum with the dosage regimen of 3 mg/kg of body weight every 12 h may have contributed to these failures. To evaluate the effect that higher concentrations would have on bactericidal activity measured by time-kill curves, peak and trough concentrations were estimated for dosage regimens of 3, 5, and 10 mg/kg every 12 h. MICs, MBCs, and killing curves for daptomycin and vancomycin were performed by using the estimated concentrations with four S. aureus strains obtained from patients who failed daptomycin therapy for endocarditis. MICs and MBCs of daptomycin demonstrated a greater inoculum effect than those of vancomycin; MICs and MBCs of daptomycin increased three- to fourfold, but those of vancomycin increased only one- to twofold when the inoculum was increased from 5 x 10(5) to 5 x 10(7) CFU/ml. No pH-dependent effect on MICs or MBCs was seen. Strenuous experimental conditions were chosen: high inoculums (5 x 10(7) CFU/ml), extremes of pH (6.4, 7.4, and 8), and stationary and exponentially growing organisms; and all experiments completed in the presence of pooled human serum. Daptomycin exhibited concentration-dependent killing and statistically faster kill rates than vancomycin against stationary- or exponential-growth-phase organisms. A pH-dependent decrease in activity with daptomycin was also demonstrated. Daptomycin and vancomycin produced higher kill rates against exponentially growing organisms. A pH-dependent decrease in activity with daptomycin was also demonstrated. Daptomycin and vancomycin produced higher kill rates against exponentially growing organisms. The results indicate that the use of higher dosage regimens with compounds similar to daptomycin may be capable of overcoming the effects of pH, high inoculum, and protein binding.

Tests for bactericidal effects of antimicrobial agents: technical performance and clinical relevance

Bactericidal testing has been used for several decades as a guide for antimicrobial therapy of serious infections. Such testing is most frequently performed when bactericidal antimicrobial agent therapy is considered necessary (such as when treating infectious endocarditis or infection in an immunocompromised host). It has also been used to ensure that the infecting organism is killed by (not tolerant to) usually bactericidal compounds. However, few data are available to support the role of such tests in direct patient care. Several important variables affect the reproducibility of the test results; however, proposed reference methods are now available for performing the MBC test. With minor modifications, these can provide a standardized approach for laboratories that need to perform them. Currently, little evidence is available to support the routine use of such testing for the care of individual patients. However, testing of new (investigational) antimicrobial agents can be beneficial in determining their potential to provide bactericidal antimicrobial activity during clinical use. New methods to assess bactericidal activity are being developed, but as yet none have been rigorously tested in patient care settings; further, for most of these methods, little information is available as to which technical parameters affect their results. In clinical laboratories, all bactericidal tests must be performed with rigorously standardized techniques and adequate controls, bearing in mind the limitations of the currently available test procedures.