Oseltamivir Population Pharmacokinetics in the Ferret: Model Application for Pharmacokinetic/Pharmacodynamic Study Design.

The ferret is a suitable small animal model for preclinical evaluation of efficacy of antiviral drugs against various influenza strains, including highly pathogenic H5N1 viruses. Rigorous pharmacokinetics/pharmacodynamics (PK/PD) assessment of ferret data has not been conducted, perhaps due to insufficient information on oseltamivir PK. Here, based on PK data from several studies on both uninfected and influenza-infected groups (i.e., with influenza A viruses of H5N1 and H3N2 subtypes and an influenza B virus) and several types of anesthesia we developed a population PK model for the active compound oseltamivir carboxylate (OC) in the ferret. The ferret OC population PK model incorporated delayed first-order input, two-compartment distribution, and first-order elimination to successfully describe OC PK. Influenza infection did not affect model parameters, but anesthesia did. The conclusion that OC PK was not influenced by influenza infection must be viewed with caution because the influenza infections in the studies included here resulted in mild clinical symptoms in terms of temperature, body weight, and activity scores. Monte Carlo simulations were used to determine that administration of a 5.08 mg/kg dose of oseltamivir phosphate to ferret every 12 h for 5 days results in the same median OC area under the plasma concentration-time curve 0-12 h (i.e., 3220 mg h/mL) as that observed in humans during steady state at the approved dose of 75 mg twice daily for 5 days. Modeling indicated that PK variability for OC in the ferret model is high, and can be affected by anesthesia. Therefore, for proper interpretation of PK/PD data, sparse PK sampling to allow the OC PK determination in individual animals is important. Another consideration in appropriate design of PK/PD studies is achieving an influenza infection with pronounced clinical symptoms and efficient virus replication, which will allow adequate evaluation of drug effects.

Optimal sampling of antipsychotic medicines: a pharmacometric approach for clinical practice.

AIM: To determine optimal sampling strategies to allow the calculation of clinical pharmacokinetic parameters for selected antipsychotic medicines using a pharmacometric approach. METHODS: This study utilized previous population pharmacokinetic parameters of the antipsychotic medicines aripiprazole, clozapine, olanzapine, perphenazine, quetiapine, risperidone (including 9-OH risperidone) and ziprasidone. d-optimality was utilized to identify time points which accurately predicted the pharmacokinetic parameters (and expected error) of each drug at steady-state. A standard two stage population approach (STS) with MAP-Bayesian estimation was used to compare area under the concentration-time curves (AUC) generated from sparse optimal time points and rich extensive data. Monte Carlo Simulation (MCS) was used to simulate 1000 patients with population variability in pharmacokinetic parameters. Forward stepwise regression analysis was used to determine the most predictive time points of the AUC for each drug at steady-state. RESULTS: Three optimal sampling times were identified for each antipsychotic medicine. For aripiprazole, clozapine, olanzapine, perphenazine, risperidone, 9-OH risperidone, quetiapine and ziprasidone the CV% of the apparent clearance using optimal sampling strategies were 19.5, 8.6, 9.5, 13.5, 12.9, 10.0, 16.0 and 10.7, respectively. Using the MCS and linear regression approach to predict AUC, the recommended sampling windows were 16.5-17.5 h, 10-11 h, 23-24 h, 19-20 h, 16.5-17.5 h, 22.5-23.5 h, 5-6 h and 5.5-6.5 h, respectively. CONCLUSION: This analysis provides important sampling information for future population pharmacokinetic studies and clinical studies investigating the pharmacokinetics of antipsychotic medicines.

Pharmacodynamic variability beyond that explained by MICs.

Monte Carlo simulations (MCS) present a powerful tool to evaluate candidate regimens by determining the probability of target attainment. Although these assessments have traditionally incorporated variability in pharmacokinetic (PK) parameters and MICs, consideration of interstrain pharmacodynamic (PD) variability has been neglected. A population PK/PD model was developed for doripenem using murine thigh infection data based on 20 bacterial strains. PK data were fit to a linear two-compartment model with first-order input and elimination processes and an absorption lag time from a separate site (r(2) > 0.96). PK parameters were utilized to simulate free-drug profiles for various regimens in PD studies, from which the percentage of the dosing interval for which free-drug concentrations exceed the MIC of the targeted strain (%fT>MIC) was calculated. Doripenem PD was excellently described with Hill-type models (r(2) > 0.98); significant differences between mean PD estimates determined using a two-stage approach versus population analyses were not observed (P > 0.05); however, the variance in 50% effective concentration (EC50) and maximum effect (Emax) among strains was much greater using the two-stage approach. Even using the population approach, interstrain variability in EC50 (coefficient of variation expressed as a percentage [CV%] = 29.2%) and H (CV% = 46.1%) parameters was substantive, while the variability in Emax (CV% = 19.7%) was modest. This resulted in extensive variability in the range of %fT>MIC targets associated with stasis to those associated with a 2-log10 reduction in bacterial burden (CV% ∼ 50%). It appears that MCS, based on the assumption that PD variability is due to MIC alone, underestimates variability and may consequently underestimate treatment failures.

Front-loaded linezolid regimens result in increased killing and suppression of the accessory gene regulator system of Staphylococcus aureus.

Front loading is a strategy used to optimize the pharmacodynamic profile of an antibiotic through the administration of high doses early in therapy for a short duration. Our aims were to evaluate the impact of front loading of linezolid regimens on bacterial killing and suppression of resistance and on RNAIII, the effector molecule of the accessory gene regulator system (encoded by agr) in methicillin-resistant Staphylococcus aureus (MRSA). Time-killing experiments over 48 h were utilized for linezolid against four strains of MRSA: USA100, USA300, USA400, and ATCC 29213. A hollow-fiber infection model simulated traditional and front-loaded human therapeutic regimens of linezolid versus USA300 at 10(6) CFU/ml over 240 h. Over 48 h in time-kill experiments, linezolid displayed bacteriostatic activity, with reductions of >1 log(10) CFU/ml for all strains. Front-loaded regimens that were administered over 5 days, 1,200 mg every 12 h (q12h) (total, 10 doses) and 2,400 mg q12h (total, 10 doses) followed by 300 mg q12h thereafter, resulted in sustained bactericidal activity, with reductions of the area under the CFU curve of -6.15 and -6.03, respectively, reaching undetectable limits at the 10-day study endpoint. All regimens displayed a reduction in RNAIII relative expression at 24 h and 240 h compared with that of the growth control. Monte Carlo simulations predicted a <1.27× increase in the fractional decreases in platelets for all front-loaded regimens versus the 600 mg q12h regimen, except for the highest-dose front-loaded regimen. Front-loading strategies for linezolid are promising and may be of utility in severe MRSA infections, where early aggressive therapy is necessary.

Outcomes and costs associated with a history of vancomycin exposure in patients with MRSA-related complicated bacteremia and infective endocarditis.

BACKGROUND: Methicillin-resistant Staphylococcus aureus (MRSA) is the primary cause of complicated bacteremia (CB) and infective endocarditis (IE). Studies have compared the costs of treatment with vancomycin to those of other agents, as well as the efficacy and tolerability of these treatments. However, a literature search found no published studies of the effects of vancomycin exposure on outcomes and hospital costs in patients with CB or IE due to MRSA. OBJECTIVE: The aim of this study was to determine whether there is a quantitative relationship between the duration of vancomycin treatment or cumulative vancomycin exposure and outcomes or costs in patient with CB or IE due to MRSA. METHODS: Electronic medical records of confirmed cases of MRSA-related CB or IE from July 1, 2006, to June 30, 2008, were retrospectively reviewed to identify patients with a history of vancomycin exposure or no vancomycin exposure. Those who received vancomycin were stratified by the amount of drug administered or the duration of treatment to determine the relationship between treatment and outcomes. Data collected included demographic information, treatment information, attributable mortality, MIC data, and hospital costs. Classification and regression tree analysis (CART) was used to determine whether a history of vancomycin exposure was associated with treatment failure, attributable mortality, or both. The Mann-Whitney U test and the Fisher exact test were used for univariate analyses, and logistic regression was used for multivariate modeling. RESULTS: Data from 50 patients were evaluated (CB, 32; IE, 18). Overall rates of failure and attributable mortality were 32% and 16%, respectively. No significant differences were observed between the variables and costs. The CART break points for failure were ≥18.75 g and ≥14 days of vancomycin treatment in the previous 3 years; for attributable mortality, the CART break points were ≥45 g and ≥31 days. In the final multivariate model for failure, ≥18.75 g and ≥14 days of vancomycin treatment in the previous 3 years were predictors of failure (both, P = 0.002). Acute Physiology and Chronic Health Evaluation (APACHE) II score (P = 0.04), ≥45 g (P = 0.002), and ≥31 days of treatment (P = 0.002) in the previous 3 years were predictors of attributable mortality after adjustment for all covariates. CONCLUSIONS: Using the present model, cumulative vancomycin amount and duration were associated with attributable mortality and clinical failure but not with costs.

Application of pharmacokinetic-pharmacodynamic modeling and the justification of a novel fusidic acid dosing regimen: raising Lazarus from the dead.

Perhaps the most crucial step in the clinical development of an antimicrobial agent is the selection of a dosing regimen. Such decisions impact not only the success of a program but also the well being of individual patients, the emergence of resistance, and society as a whole. For fusidic acid, the selection of a dosing regimen for the treatment of patients with acute bacterial skin and skin-structure infection (ABSSSI) was based on the integration of knowledge gained from human population pharmacokinetic, in vitro infection, and mathematical models. The overarching goal of these studies was to identify a dosing regimen that would maximize the probabilities of positive clinical outcomes and limit the emergence of bacterial resistance during therapy. Novel dosing regimens identified included 1500 mg twice daily on day 1 followed by 600 mg twice daily for 10-14 days, a regimen that was subsequently found to be effective in a phase 2 clinical study of patients with ABSSSI. Herein, we review the data supporting the use of this novel fusidic acid dosing regimen, which will undergo further clinical evaluation in phase 3 clinical trials.

Pharmacokinetics-pharmacodynamics of gatifloxacin in a lethal murine Bacillus anthracis inhalation infection model.

We determined the pharmacokinetic-pharmacodynamic (PK-PD) measure most predictive of gatifloxacin efficacy and the magnitude of this measure necessary for survival in a murine Bacillus anthracis inhalation infection model. We then used population pharmacokinetic models for gatifloxacin and simulation to identify dosing regimens with high probabilities of attaining exposures likely to be efficacious in adults and children. In this work, 6- to 8-week-old nonneutropenic female BALB/c mice received aerosol challenges of 50 to 75 50% lethal doses of B. anthracis (Ames strain, for which the gatifloxacin MIC is 0.125 mg/liter). Gatifloxacin was administered at 6- or 8-h intervals beginning 24 h postchallenge for 21 days, and dosing was designed to produce profiles mimicking fractionated concentration-time profiles for humans. Mice were evaluated daily for survival. Hill-type models were fitted to survival data. To identify potentially effective dosing regimens, adult and pediatric population pharmacokinetic models for gatifloxacin and Monte Carlo simulation were used to generate 5,000 individual patient exposure estimates. The ratio of the area under the concentration-time curve from 0 to 24 h (AUC(0-24)) to the MIC of the drug for the organism (AUC(0-24)/MIC ratio) was the PK-PD measure most predictive of survival (R(2) = 0.96). The 50% effective dose (ED(50)) and the ED(90) and ED(99) corresponded to AUC(0-24)/MIC ratios of 11.5, 15.8, and 30, respectively, where the maximum effect was 97% survival. Simulation results indicate that a daily gatifloxacin dose of 400 mg for adults and 10 mg/kg of body weight for children gives a 100% probability of attaining the PK-PD target (ED(99)). Sensitivity analyses suggest that the probability of PK-PD target attainment in adults and children is not affected by increases in MICs for strains of B. anthracis to levels as high as 0.5 mg/liter.

Assessment of eptifibatide dosing in renal impairment before and after in-service education provided by pharmacists.

BACKGROUND: Anticoagulant and antithrombotic agents are frequently cited as sources of medication errors. Several factors increase the risk of receiving excess dosing of glycoprotein IIb/IIIa inhibitors in the management of acute coronary syndrome (ACS), including older age, female gender, elevated serum creatinine, a history of diabetes mellitus, and a history of heart failure. In June 2003, the manufacturer of eptifibatide released a recommendation adjusting infusion rate downward to 1 mcg per kg per minute for eptifibatide in patients with renal impairment, defined as an estimated creatinine clearance (CrCl) < 50 ml per minute. Eptifibatide is known to accumulate in patients with renal impairment, thereby increasing hemorrhagic risk. OBJECTIVE: To assess the impact of education on physician adherence to the renal dosing recommendation for eptifibatide at 2 academic medical centers. The primary outcome measure was the proportion of patients with renal impairment dosed appropriately with eptifibatide before and after in-service education provided by a clinical pharmacist. Secondary outcome measures included the difference in the improvement in dosing adherence between the 2 sites and the influence of patient variables on the incidence of bleeding events. METHODS: This prospective study was conducted in patients with renal impairment who received eptifibatide for the medical management of unstable angina (UA) or non-ST-elevation myocardial infarction (NSTEMI) or for the interventional management of chronic stable angina, UA, NSTEMI, or ST-elevation myocardial infarction (STEMI, not a Food and Drug Administration-approved use). Patient data were assessed at 2 tertiary care teaching institutions between June 2003 and December 2005. The preeducation phase for the sites ran from June 2003 through April 2005 for Site A and from June 2003 through May 2005 for Site B. The posteducation phase ran from May 2005 through December 2005 for Site A and from June 2005 through December 2005 for Site B. At site A, a 1-hour educational seminar on ACS management strategies was employed, in which 5 minutes focused on adherence of prescribers to the guideline for renal dosing recommendations for eptifibatide. This tutorial was accomplished through (1) an in-service provided by 1 clinical pharmacist to the cardiology department, and (2) handouts containing the renal dosing recommendations for eptifibatide along with dosing for other medications used to manage ACS. The intervention at Site B involved an eptifibatide-focused seminar presented to cardiologists by a clinical pharmacist, 10 minutes of which was devoted to renal dosing recommendations that included (1) a summary of literature supporting the infusion rate reduction in patients with renal impairment and (2) the specific updated dosing recommendation for eptifibatide. The data collected in retrospective chart review included patient demographics, baseline laboratory values, and risk factors for bleeding. An appropriate eptifibatide dose was defined as a physician order for a continuous infusion of 1 mcg per kg per minute in patients with an estimated CrCl < 50 ml per minute. RESULTS: A total of 148 patients with renal impairment who received eptifibatide were evaluated (106 in the preeducation phase and 42 in the posteducation phase). A significant increase in the adherence rate for eptifibatide dosing in patients with renal impairment was observed from 36.8% in the preeducation phase to 69.0% in the posteducation phase (P < 0.001) for the 2 sites combined. The incidence of major and minor bleeding was 16.7% in the preeducation phase and 14.3% in the posteducation phase (P = 0.742). When bleeding incidence was stratified by the appropriateness of infusion, the incidence of major and minor bleeding was also similar for appropriate dosing (1 mcg per kg per minute, 16.4%) versus inappropriate dosing (2 mcg per kg per minute, 15.7%; P = 0.916). CONCLUSION: This educational intervention provided by a clinical pharmacist was associated with improved prescriber adherence to dosing recommendations for eptifibatide in patients with renal impairment. Improved adherence to the dosing guideline and administration of an appropriate infusion rate were not associated with reduction in either minor or major bleeding events.

The clinical impact of 1:1 conversion from Neoral to a generic cyclosporine (Gengraf) in renal transplant recipients with stable graft function.

The introduction of cyclosporine (CYA) to the immunosuppressive armamentarium has had a significant effect on graft survival. An improvement in the formulation from the oil-based to a microemulsion-based form has resulted in better absorption and more predictable CYA bioavailability. Since the introduction of the first microemulsion form (Neoral), several bioequivalent formulations are now available and are switched in a 1:1 fashion at pharmacies to curtail costs. The purpose of our study was to study the effect of a 1:1 switch from Neoral to Gengraf on CYA trough levels and serum creatinine (SRC) in renal transplant recipients with stable graft function. Eighty-two renal transplant recipients with stable graft function were enrolled in the study, and of these, 73 were switched to Gengraf, whereas nine remained on Neoral. The 13 patients switched to Gengraf required a dosage change after the mean CYA trough levels changed from 234 +/- 96 ng/mL at baseline to 289 +/- 102 ng/mL (p < 0.05) at 2 wk. With the adjustments in dosage, the levels approached the baseline trough concentrations (239 +/- 151 ng/dL). The nine patients who remained on Neoral had no change in the CYA levels or SCR. Nearly 20% of patients who switched to a bioequivalent CYA preparation required a dose adjustment to return to pre-conversion CYA trough levels. Our study raises serious concerns regarding the switchability of generic CYA for Neoral without careful follow-up therapeutic drug monitoring.

Evaluating ciprofloxacin dosing for Pseudomonas aeruginosa infection by using clinical outcome-based Monte Carlo simulations.

Pseudomonas aeruginosa causes serious infections whose outcome is highly dependent on antimicrobial therapy. The goal of this study was to predict the relative efficacies of three ciprofloxacin dosing regimens for P. aeruginosa infection using clinical outcome-based Monte Carlo simulations (MCS) with "real patient" demographics, pharmacokinetics, MICs, and pharmacodynamics (PDs). Each cohort consisted of 1,000 simulated study subjects. Three ciprofloxacin dosing regimens were studied, including (i) the recommended standard dose of 400 mg given intravenously (i.v.) every 12 h (q12h), (ii) the recommended high dose of 400 mg i.v. q8h, and (iii) a novel, PD-targeted regimen to attain a fAUC/MIC value of >86. Probability of target attainment (PTA) and probability of cure (POC) were determined for each regimen. POC with the standard dose was at least 0.90 if pathogen MICs were < or =0.25 microg/ml but only 0.59 or 0.27 if MICs were 0.5 or 1 microg/ml, respectively. Predicted cure rates in these MIC categories were significantly higher at 0.72 and 0.40 with the high dose and 0.91 and 0.72 with the PD-targeted regimen(P < 0.0001). Analyses based on the local susceptibility profile produced PTA and POC estimates of 0.44 and 0.74 with the standard ciprofloxacin dose, 0.58 and 0.81 with the high dose, and 0.84 and 0.93 with the PD-targeted regimen, respectively. In conclusion, as demonstrated by clinical outcome-based MCSs, the highest recommended ciprofloxacin dose of 400 mg i.v. q8h should be used in the treatment of P. aeruginosa infection to improve PD target attainment and clinical cure. However, even this appears ineffective if pathogen MICs are 1 mug/ml, warranting the consideration of a lower MIC breakpoint, < or =0.5 microg/ml.

Application of an in vitro infection model and simulation for reevaluation of fluoroquinolone breakpoints for Salmonella enterica serotype typhi.

Salmonella enterica serotype Typhi and nontyphoidal Salmonella remain major causes of morbidity and mortality worldwide. Ampicillin, trimethoprim-sulfamethoxazole, and chloramphenicol no longer provide reliable coverage of Salmonella, and fluoroquinoloes have emerged as first-line treatment options. Due to mounting evidence of decreased in vitro susceptibility and diminished clinical response to fluoroquinolone therapy, it has been suggested that the NCCLS breakpoints for the salmonellae be reevaluated. We utilized an in vitro infection model to determine which pharmacokinetic-pharmacodynamic (PK-PD) measure was most closely linked to fluoroquinolone activity against salmonellae and the magnitude that was predictive of efficacy. Monte Carlo simulation was utilized to determine the probability of attaining potential susceptibility breakpoints for three fluoroquinolones. The free-drug area under the concentration-time curve from 0 to 24 h/MIC ratio was the PK-PD measure most predictive of efficacy, and a ratio of 105 corresponded to 90% of maximal activity. Simulation results suggested susceptible breakpoints of 0.12 microg/ml for ciprofloxacin and gatifloxacin and 0.25 microg/ml for levofloxacin. These proposed breakpoints correspond to the MIC separating the wild-type susceptible organism population from those strains possessing single-step mutations in the quinolone resistance-determining region. These results that integrate PK-PD measures and fluoroquinolone MIC distributions in the genetic context of examined Salmonella isolates clearly demonstrate that the prudent use of a lower susceptibility breakpoint minimizes the probability of clinical failure or delayed response in fluoroquinolone-treated patients.

Effect of fluoroquinolone expenditures on susceptibility of Pseudomonas aeruginosa to ciprofloxacin in U.S. hospitals.

The effect of fluoroquinolone use on the susceptibility of Pseudomonas aeruginosa to fluoroquinolones in U.S. hospitals was studied. Benchmarking surveys were sent annually to pharmacists practicing in U.S. hospitals from 1993 to 1999. Data collected included hospital characteristics, antimicrobial expenditures and use, antimicrobial stewardship activities, and bacterial susceptibilities. Antimicrobial expenditures were normalized for the number of occupied beds (OBs) per year. General linear modeling and repeated-measures mixed-effects modeling were used to determine factors predictive of P. aeruginosa susceptibility to fluoroquinolones. A total of 174 hospitals provided data for fluoroquinolone expenditures and susceptibility of P. aeruginosa; the median number of years of data was 3 (range, 1-6), representing 416 hospital years. Community hospitals contributed a majority of the data. Median fluoroquinolone expenditures increased gradually from $230 per OB in 1993 to $400 per OB in 1998. A 55% increase to $620 per OB occurred in 1999, largely because of increased spending on levofloxacin. Susceptibility to ciprofloxacin was commonly used to assess fluoroquinolone susceptibility. The median susceptibility of P. aeruginosa to ciprofloxacin decreased from 84% to 71%. Increasing expenditures for ofloxacin and levofloxacin, but not ciprofloxacin, were associated with decreasing P. aeruginosa susceptibility to ciprofloxacin. In the final multivariable model, each study year after 1993 and every increase in ofloxacin expenditure of $100 per OB were associated with decreases in P. aeruginosa susceptibility. Data from a benchmarking survey of U.S. hospitals for 1993-1999 revealed increases in levofloxacin expenditures, total fluoroquinolone expenditures, expenditures for nonfluoroquinolone antipseudomonal antimicrobials, and total antimicrobial expenditures in 1999. Increases in expenditures for levofloxacin and ofloxacin were associated with a significant decrease in P. aeruginosa susceptibility to ciprofloxacin.

Characterization of the onset and consequences of pneumonia due to fluoroquinolone-susceptible or -resistant Pseudomonas aeruginosa.

OBJECTIVES: This study was conducted to identify and compare the microbiological and clinical outcomes among hospitalized adults with pneumonia caused by fluoroquinolone-susceptible or -resistant strains of Pseudomonas aeruginosa. Antibiotic regimens used prior to, as well as those used to treat, the infections were characterized. PATIENTS AND METHODS: This non-randomized multicentre study included 100 consecutively identified patients with pneumonia caused by fluoroquinolone-susceptible (n = 50) or fluoroquinolone-resistant (n = 50) strains of P. aeruginosa. Medical records were examined for demographic, clinical and treatment variables including antibiotics received in the 30 days before the index respiratory or blood culture; AUICs were calculated for each patient using reported or derived MICs. Multivariate logistic and linear regressions were used to identify factors associated with successful clinical and microbiological outcomes. RESULTS: The study population was primarily elderly, frequently in a critical care unit, with low serum albumin and with a high probability of failure and mortality. Patients with pneumonia caused by fluoroquinolone-resistant P. aeruginosa were more likely to have received antibiotics within 7 days before the infection (P = 0.027); the antibiotic regimen was more likely to be of a weak potency (mean AUIC of 58 versus 169, P = 0.001) and to include levofloxacin (P < 0.0001) than what was administered to patients who became infected with a fluoroquinolone-susceptible strain. Regardless of susceptibility, a mean of between 2 and 3 weeks of directed antibiotic therapy was administered to each patient. CONCLUSIONS: Pneumonia caused by fluoroquinolone-resistant P. aeruginosa is frequently associated with prior exposure to levofloxacin. Treatment of P. aeruginosa pneumonia is difficult and usually consists of combination regimens with multiple modifications.

Cost-effectiveness of IV-to-oral switch therapy: azithromycin vs cefuroxime with or without erythromycin for the treatment of community-acquired pneumonia.

STUDY OBJECTIVE: To conduct a cost-effectiveness analysis of IV-to-oral regimens of azithromycin vs cefuroxime with or without erythromycin in the treatment of patients hospitalized with community-acquired pneumonia (CAP). PATIENTS: Of the 268 evaluable patients enrolled into a randomized, multicenter clinical trial of adults, 266 patients had sufficient data to be included in this cost-effectiveness analysis. One hundred thirty-six patients received azithromycin, and 130 patients received cefuroxime with or without erythromycin. METHODS: A pharmacoeconomic analysis from the hospital provider perspective was conducted. Health-care resource utilization was extracted from the clinical database and converted to national reference costs. Decision analysis was used to structure and characterize outcomes. Sensitivity analyses were performed, and statistics were applied to the cost-effectiveness ratios. RESULTS: The clinical success and adverse event rates and antibiotic-related length of stay were 78%, 11.8%, and 5.8 days for the azithromycin group and 75%, 20.7%, and 6.4 days for the group receiving cefuroxime with or without erythromycin, respectively. Geometric mean treatment costs were 4,104 US dollars (95% confidence interval [CI], 3,874 to 4,334 US dollars) for the azithromycin group, and 4,578 US dollars (95% CI, 4,319 to 4,837 US dollars) for the group receiving cefuroxime with or without erythromycin (p = 0.06). The cost-effectiveness ratios were 5,265 US dollars per expected cure for the azithromycin group, and 6,145 US dollars per expected cure for group receiving cefuroxime with or without erythromycin (p = 0.05). CONCLUSIONS: Despite a higher per-dose purchase price, overall costs with azithromycin tended to be lower due to decreased duration of therapy, lower preparation and administration costs, and reduced hospital length of stay. As empiric therapy, azithromycin monotherapy was cost-effective compared to cefuroxime with or without erythromycin for patients hospitalized with CAP who have no underlying cardiopulmonary disease, and no risk factors for either drug-resistant pneumococci or enteric Gram-negative pathogens.