Ceftriaxone

Study objective: To determine the cost-effectiveness of sequential IV to oral gatifloxacin therapy vs IV ceftriaxone with or without IV erythromycin to oral clarithromycin therapy to treat community-acquired pneumonia (CAP) patients requiring hospitalization.

Patients: Two hundred eighty-three patients enrolled in a randomized, double-blind, clinical trial were eligible for inclusion in the cost-effectiveness analysis.

Methods: Data collected included patient demographics, clinical and microbiological outcomes, length of stay (LOS), and antibiotic-related LOS (LOSAR). Costs evaluated include drug acquisition (level 1); plus costs of preparation, dispensing, and administration, treating adverse events, and clinical failures (level 2); plus hospital per diem costs (level 3). Robustness of economic findings was tested using sensitivity analyses.

Results: Two hundred three patients were clinically and economically evaluable (98 receiving gatifloxacin and 105 receiving ceftriaxone). IV erythromycin was administered to 35 patients in the ceftriaxone-treated group. Oral conversion was achieved in 98% of patients in each group. Clinical cure and microbiological eradication rates did not differ statistically (98% and 97% with gatifloxacin vs 92% and 92% with ceftriaxone, respectively). Overall, neither geometric mean LOS nor LOSAR differed significantly (4.2 days and 4.1 days with gatifloxacin vs 4.9 days and 4.9 days with ceftriaxone, respectively). Treatment failures in the ceftriaxone group contributed to a mean incremental increase in LOSAR of 1.09 days and increased mean cost per patient. The geometric mean costs per patient (level 3) were $5,109 for gatifloxacin and $6,164 for ceftriaxone (p = 0.011). The cost-effectiveness ratios (mean cost per expected success) were $5,236:1 and $7,047:1 for gatifloxacin and ceftriaxone, respectively.

Conclusions: Gatifloxacin monotherapy for CAP patients requiring hospitalization is clinically effective and provides an economic advantage compared to the regimen of ceftriaxone with or without erythromycin IV with a switch to oral clarithromycin. (CHEST 2001; 119:1439-1448)

Key words: antibiotics; community-acquired pneumonia; cost-effectiveness

Abbreviations: ATS = American Thoracic Society; CAP = community-acquired pneumonia; CEA = cost-effectiveness analysis; CER = cost-effectiveness ratio; IDSA = Infectious Diseases Society of America; LOS = length of stay; LOSAR = antibiotic-related length of stay

Community-acquired pneumonia (CAP) is the most common cause of death from infectious diseases and sixth-leading cause of death in the United States.[1-4] A recent evaluation of CAP found an annual incidence of 5.6 million cases, of which 1.1 million patients (20%) required hospitalization.[5] Among those hospitalized, mortality ranges from 2 to 21%, increasing to 20 to 53% among those with severe disease.[6] In the United States, CAP is estimated to be responsible for 64 million days of decreased activity, 39 million days of bed rest, and 10 million lost work days per year; annual costs attributed to CAP are significant at approximately $23 billion.[2,7] Inpatient treatment alone accounts for $6 to $8 billion, predominantly due to per diem room fee and therefore linked directly to length of stay (LOS).[6,8]

Initial antibiotic therapy for CAP is empiric, pending pathogen identification. Appropriate empiric selection is based on anticipated pathogens and should targets both typical organisms (Streptococcus pneumoniae, Haemophilus influenzae, Moraxella catarrhalis) and atypical organisms (Mycoplasma pneumoniae, Legionella pneumophila, Chlamydia pneumoniae).[1,4] However, in as many as 30 to 60% of CAP patients, diagnostic testing does not yield an identifiable pathogen, further highlighting the importance of appropriate empiric antimicrobial selection.[6] Treatment guidelines for the management of CAP have been developed by several organizations, including the American Thoracic Society (ATS), the Infectious Diseases Society of America (IDSA), the Canadian Thoracic Society, and the British Thoracic Society.[2,3,9-11] The 1993 ATS guidelines suggest a second-generation or third-generation cephalosporin (cefuroxime, cefotaxime, ceftriaxone) or a [Beta]-lactam/[Beta]-lactamase inhibitor, with or without a macrolide.[8] The 1998 IDSA guidelines recommended treating patients with CAP requiring hospitalization with a [Beta]-lactam with or without a macrolide, or monotherapy with a fluoroquinolone with enhanced S pneumoniae activity.[2] The most recent 2000 IDSA guidelines have changed this recommendation to an extended-spectrum cephalosporin plus a macrolide or monotherapy with an antipneumococcal fluoroquinolone.[3] For more severe eases, the preferred regimens remain cefotaxime, ceftriaxone, or a [Beta]-lactam/[Beta]-lactamase inhibitor plus erythromycin, azithromycin, or a fluoroquinolone.[2,3]

Gatifloxacin is a recently approved 8-methoxy fluoroquinolone with a broad spectrum of activity encompassing Gram-positive, Gram-negative, and atypical respiratory pathogens.[12-14] In a randomized, double-blind clinical trial IV gatifloxacin followed by oral gatifloxacin vs IV ceftriaxone with or without IV erythromycin followed by oral clarithromycin were recently compared for the treatment of CAP patients requiring hospitalization.[15] Although the clinical success rates appeared to differ (97% with gatifloxacin and 91% with ceftriaxone), statistical significance was not achieved, perhaps limited by sample size (n = 205 evaluable patients). Among patients with an identified pathogen, the presumed or documented cure rates were 97% for patients treated with gatifloxacin vs 92% for those who received ceftriaxone, consistent with the clinical outcomes.[15]

The drug-acquisition costs of fluoroquinolones may exceed those of traditional agents ([Beta]-lactams, macrolides) used for the treatment of CAP. However, antibiotic costs are only a small percentage of the overall costs of treating CAP.[16-18] The availability of highly bioequivalent oral formulations, and the potential for earlier switch to oral therapy leading to decreased length of hospitalization, suggest the fluoroquinolones may actually result in substantial cost savings.[19,20] The predominant cost driver for hospital admission is the LOS; therefore, any treatment option that potentially decreases LOS may result in significant reduction of hospital costs.

We subjected the data from the above-described clinical study to a cost-effectiveness analysis (CEA), from the hospital perspective, to ascertain whether one regimen was more cost-effective than the alternative. Resource utilization information was collected prospectively during the clinical trial to allow for economic analysis. Data collection included information regarding LOC, ICU LOS, study medication use, subsequent antibiotic use for treatment failures, concomitant medications, treatment of adverse events, and other hospital procedures.

MATERIALS AND METHODS

All clinically evaluable patients enrolled in the trial were eligible for inclusion in this pharmacoeconomic analysis. Criteria for clinical trial inclusion included age [is greater than or equal to] 18 years, enrollment [is less than] 24 h after hospitalization, and clinical, laboratory, and radiologic evidence of CAP. Patients were excluded if more than one dose of systemic antibiotic (or a combination of systemic antimicrobials) was administered within 7 days of enrollment, if concomitant systemic antimicrobial therapy was necessary during the study period, if the investigator believed that [is greater than] 14 days of therapy would be required, or if expected survival was [is less than] 72 h.[15] Other exclusion criteria included hypersensitivity to any of the study medications, renal insufficiency, or clinically significant hepatic disease. A complete description of the inclusion and exclusion criteria along with the clinical and bacteriologic results has been published.[15]

Patients were randomized 1:1 in a double-blind fashion to receive either IV gatifloxacin, 400 mg qd, or IV ceftriaxone, 1 g or 2 g q24h (dosage per the discretion of the prescribing physician). IV erythromycin, 500 mg or 1 g qid, was added to the ceftriaxone regimen, again at the discretion of the investigator, to provide atypical organism coverage. When clinically appropriate and after receiving at least 2 days of IV study medication, patients could be converted to oral gatifloxacin or clarithromycin therapy per their randomized grouping. A final assessment of treatment outcome was made at the posttherapy visit (day 7 to day 14). The following clinical end points were used: cure (all acute signs and symptoms of pneumonia were resolved or improved to a level such that no further antimicrobial therapy was required); failure (signs or symptoms relevant to the original infection persisted or progressed after at least 3 days of therapy, change in antimicrobial therapy was necessary, or patient died due to pneumonia); or indeterminate (events precluded classification as cured or failed).[15]

Data from the 205 clinically evaluable patients, made available by the study sponsor, were used to formulate the independent economic evaluation. To be considered economically evaluable, the following criteria were employed: a clinically evaluable patient and all pertinent resource use data captured. A CEA using a decision tree was designed to characterize, measure, and compare outcomes and potential economic differences between treatment groups. Sensitivity analysis was used to assess the robustness of the results and to determine whether the economic decision changes when index variables are altered within reasonable ranges.

Resource Utilization

A database (Excel Version 7.0, 95; Microsoft; Redmond, WA) was established to assimilate the information necessary for the economic analyses. The economic evaluation period for each patient begins on the day that study-drug treatment was initiated. The number of days of hospitalization associated with antibiotic treatment of infection is known as the antibiotic-related length of stay (LOSAR).[21] The LOSAR for each patient begins at initiation of study-antibiotic administration and ends at discontinuation of all antibiotic therapy or at hospital discharge, whichever occurs first. Also included within the LOSAR is any additional LOS associated with (1) treatment of adverse events resulting from use of the study antibiotic or (2) subsequent antimicrobial therapy for treatment failures. This additional LOS was calculated as the duration of the subsequent treatment, in days, up to but not exceeding the duration of hospitalization, and was added to the LOS during which the study drug was received. Final follow-up study assessment was made 21 to 28 days after the end of treatment; therefore, patients who were outliers for LOS ([is greater than] 28 days) had their LOSAR capped at 28 days.[8]

The costs of treating economically significant adverse events (defined as those that required treatment) caused by study treatment or of unknown etiology were extracted from the summary and clinical response evaluation pages of the relevant case-report forms. Information regarding concomitant-medication administration was presented to an investigator blinded to the treatment regimen for assessment of economic significance.

Resource Cost

Pharmacoeconomic analyses can be classified into three levels of costs.[21] Level 1 (pharmacy perspective) considers only the acquisition price of the study medication. The manufacturer provided the usual contract price for gatifloxacin. Drug-acquisition costs for the comparator agents were estimated using national benchmarking data from 1996 to 1998 (S. Bhavnani, PharmD; personal communication; December 1999). Level 2 adds all costs directly related to antibiotic use and infection treatment, exclusive of the hospital per diem. Antibiotic-related items include medication preparation, dispensing, and administration, therapeutic drug monitoring, treatment of adverse events, and secondary treatment for failures. Medication preparation and administration was priced at an average figure of $2 (US dollars) for oral administration and $8 per IV dose, as indexed to 2000 costs.[22-25]

Level 3 costs include all level 2 items plus per diem and other hospital costs incurred during treatment.[21] The LOSAR, as defined previously, was used to quantify the duration of hospital stay directly related to the treatment of infection.

The daily cost of occupying a hospital bed is highly variable and depends on the type of unit (intensive care or general), the levels of technology and services provided, and geographic location. To determine an appropriate hospital per diem for the treatment of CAP, the following methodology was used, as described by RTI Health Solutions, part of Research Triangle Institute, Research Triangle Park, NC (T. Bell, MHA; personal communication; January 2000). A database was obtained that contained national data pertaining to patients hospitalized for the treatment of CAP (HCIA-Sachs; Baltimore, MD). A subpopulation was extracted from this database to match the clinical trial study population according to patient demographics, diagnoses, comorbidities, resource use, and charges associated with resource use. Patients in the clinical trial were categorized according to the ATS guidelines as having mild-to-moderate or severe CAP.[9] A regression analysis was performed on each stratum to derive an equation to estimate the cost of providing care for a patient hospitalized with CAP. The natural logarithm of the total charges was the dependent variable. Independent variables included the non-ICU LOS and the ICU LOS, use of mechanical ventilation, presence of comorbidities (diabetes mellitus, heart failure, or COPD), age, gender, death during study period, and duration of supplemental oxygen use. This provided a total charge estimate for ICU and non-ICU hospital days for each population. Diagnosis-related group-specific cost-to-charge ratios were used to convert charges reported in the HCIA-Sachs database (www. HCIA.com) to cost values. These ratios have been developed by US Health Care Financing Administration (www.hcfa.gov) to provide a national conversion factor between the charge billed for services and their costs. Discounting was not necessary, as costs were incurred and outcomes occurred during the same time period.

Analytic Plan

A CEA from the perspective of the institution was performed. Costs of outpatient visits and subsequent out-of-hospital treatments are not incorporated.

A decision tree categorized each case as a treatment success or failure according to the clinical investigators' assessment at the posttherapy test-of-cure visit.[15] There were no indeterminate outcomes. Sensitivity analysis was employed to test the robustness of the results by varying key variables over a reasonable range. The following data were varied: drug-acquisition price [+ or -] 25%, hospital per diem to bracket the lowest per-day cost of $1,135 and the highest of $1,719.84 by [+ or -] $250, for a range of $885 to $1,970, and the probability of success from 45 to 95%.

Statistical Analysis

Statistical analysis was performed using software (SYSTAT 8.0; SYSTAT; Evanston, IL) on a personal computer. The probability of a type-1 error of 0.05 was used to determine statistical significance. Categorical data were compared using [chi square] or Fisher's Exact Tests. Comparisons of LOS, LOSAR, ICU LOS, and per-patient mean costs associated with the two regimens were made using Kruskal-Wallis nonparametric one-way analysis of variance. The natural logarithm (geometric mean) and coefficient of variation of LOS and costs were used as the data distribution was right skewed.

RESULTS

Clinical Trial Summary

A detailed account of the patient demographic data and the clinical and bacteriologic results can be found in the clinical trial manuscript.[15] The initial study included 283 patients enrolled at 45 North American sites; 141 patients were treated with gatifloxacin and 142 patients were treated with ceftriaxone. Patient demographics were similar between groups as was the distribution of prognostic factors (eg, age, comorbidities; Table 1). According to the criteria of the protocol and the ATS guidelines, 73% patients (206 of 283) of patients were categorized as having severe pneumonia.

All patients randomized to the gatifloxacin arm received 400 mg IV with switch to gatifloxacin, 400' mg po qd, except two patients who received IV gatifloxacin only. In the ceftriaxone treatment group, 96 patients were administered 1 g IV q24h and the remainder received 2 g IV q24h. IV erythromycin was prescribed concomitantly to 56 of the ceftriax-one-treated patients (39%).

The most commonly described adverse events involved the GI system and were more frequent in the ceftriaxone arm of the study. Adverse effects reported included diarrhea (13% ceftriaxone vs 6% gatifloxacin), dyspepsia (6% ceftriaxone vs 3% gatifloxacin), and vomiting (6% ceftriaxone vs 2% gatifloxacin). Study-drug treatment was discontinued in 12 patients receiving ceftriaxone and 11 patients in the gatifloxacin arm of the study because of an adverse event or abnormal laboratory finding. Of these, nine ceftriaxone and seven gatifloxacin discontinuations were deemed related to study-drug administration.[15]

Two hundred five patients (99 receiving gatifloxacin and 106 receiving ceftriaxone) were considered clinically evaluable. The clinical success rates were 97% for gatifloxacin and 91% for ceftriaxone (95% confidence interval, -- 2.5 to 17.6%; Table 2) Thirteen clinical failures (3 with gatifloxacin and 10 with ceftriaxone) occurred; they were most frequently attributed to persistent, worsening, or new signs and symptoms of pneumonia. The clinical cure rates for patients with severe CAP (96% with gatifloxacin vs 90% with ceftriaxone) were similar to overall cure rates.[15]

Among clinically evaluable patients, 49% were also microbiologically evaluable. Microbiological eradication rates were 97% in the gatifloxacin treatment group and 92% for the ceftriaxone arm of the study. Among patients with severe CAP and a microbiologically evaluable outcome, the cure rates were 97% and 91%, respectively, in the gatifloxacin and ceftriaxone treatment arms of the study.[15]

Economic Analysis

The database comprising the information collected for all enrolled patients was made available by Bristol-Myers Squibb (Plainsboro, NJ) for the economic analysis. Two patients considered clinically evaluable were deemed economically nonevaluable; one patient (gatifloxacin treated) was nonevaluable due to insufficient detail regarding subsequent antibiotic treatment, the other patient (ceftriaxone treated) was nonevaluable for incomplete description of treatment course. Both patients considered economically nonevaluable were clinical failures, resulting in 11 treatment failures (2 with gatifloxacin and 9 with ceftriaxone). Thus, the total cohort for the economic analysis included 203 subjects; 98 subjects received gatifloxacin and 105 subjects received ceftriaxone.

Clinical failure occurred in two patients receiving gatifloxacin; however, this did not result in additional antibiotics, hospital prolongation or readmission, or additional LOSAR days. Both patients were classified as having severe pneumonia; however, neither patient was treated in the ICU. There were nine treatment failures in the ceftriaxone arm of the study; four patients required readmission and additional antimicrobial treatment. These treatment failures added a geometric mean of 1.09 incremental days per patient to the LOSAR in the ceftriaxone group (Table 3). Seven of these nine patients were classified as having severe pneumonia; two patients were treated in the ICU and ultimately died. Table 3 also illustrates the geometric mean (per patient) data regarding study-drug treatment duration, LOSAR, and LOS. Mean duration of IV study-drug administration was the same with either antimicrobial regimen at [is less than] 4 days.

Only one patient was considered to have an economically significant adverse event. This patient's study drug therapy was discontinued, but no further antimicrobial therapy was required.

More patients were admitted to the ICU in the ceftriaxone treatment group[11] than in the gatifloxacin treatment group.[3] Exclusion of patients treated in the ICU does not change the results significantly. No significant difference was found in the ICU LOS between groups; however, the geometric mean ICU LOS was 2 days longer for patients treated with ceftriaxone (Table 4). The geometric mean LOSAR of those patients who required admission to the ICU for some portion of their hospitalization did not differ statistically between groups (7.0 days with gatifloxacin and 7.6 days with ceftriaxone). Interestingly, the LOSAR of this small group of patients was 2 to 3 days longer than the geometric mean LOSAR of the overall group.

A multivariate logistic regression analysis of predictors for ICU admission was performed. The independent variables included in the analysis were age ([is less than] 65 years or [is greater than or equal to] 65 years), CAP severity (severe or mild to moderate), presence of comorbidities, presence of hypoxemia (room air oxygen saturation [is less than] 90%, [PO.sub.2] [is less than] 60 mm Hg, [PO.sub.2]/fraction of inspired oxygen ratio [is less than] 250 mm Hg), bilateral or multilobar involvement, and microbiological evaluability. More than 50% of patients were [is greater than or equal to] 65 years old and had at least one comorbidity. All patients admitted to the ICU (n = 14) were categorized as having severe pneumonia; 71% of them were microbiologically evaluable. Of the 15 pathogens isolated, in the ICU patients, the most frequently identified were S aureus (28.6%), M catarrhalis (21%), and S pneumoniae (14.3%). A number of other organisms were isolated on a single occasion: C pneumoniae, Pseudomonas putida, H influenzae, H parainfluenzae, M pneumoniae, and L pneumophila. Four patients had two or more pathogens isolated. Of these independent variables, microbiological pathogen isolation was associated with an odds ratio of 2.87 (p = 0.083) for ICU admission.

The majority of patients enrolled were classified as having severe CAP. Among patients with severe CAP (71 receiving gatifloxacin and 79 receiving ceftriaxone), the geometric mean LOS was similar between treatment groups: 5.0 days for the gatifloxacin group and 5.3 days for the ceftriaxone group (Table 4). The geometric mean LOSAR of all patients evaluated (4.1 days in the gatifloxacin group vs 4.9 days in the ceftriaxone group) or those with severe CAP (4.5 days vs 5.2 days, respectively) was not significantly different between treatment groups and appeared consistent between those with differing classifications of severity. Patients were categorized as having severe pneumonia according to the protocol criteria based on the ATS guidelines.[9,15] When the patients were retrospectively categorized according to the Fine criteria, the proportion considered to have severe pneumonia dropped to 40% (class IV, 33%; class V, 7%).[15]

The geometric mean cost per patient for each treatment regimen is shown in Table 5. The drug costs used for the level 1 analysis were $20 per 400-mg IV gatifloxacin dose, $5.83 per 400-mg po gatifloxacin dose, $24 per 1-g ceftriaxone dose, $1.20 per 500-mg IV erythromycin dose, and $2.76 per 500-mg clarithromycin dose. Gatifloxacin prevailed at each cost level (p = 0.011) when comparing level 3 costs ($5,109 for gatifloxacin and $6,164 for ceftriaxone). The economic analysis was repeated removing the data of those patients treated in the ICU. Gatifloxacin remained the dominant cost-effective regimen with a mean level 3 cost of $5,038 vs $5,739 for ceftriaxone (p = 0.049; Fig 1).

[ILLUSTRATION OMITTED]

Results of the overall level 3 economic analysis are presented in the decision tree (Fig 2). The decision tree incorporates the actual probabilities of success or failure. Each branch terminates in the resulting geometric mean cost and LOSAR for each possible outcome. A successful outcome predominantly resulted in a shorter LOSAR and a lower cost compared to unsuccessful outcomes. Since a CEA includes both costs and an outcome measure (success or failure), an economic summary statistic is necessary. This is indicated using a cost-effectiveness ratio (CER) calculated by dividing the mean cost per patient by the probability of success, resulting in the mean cost per expected cure. The regimen with the lowest CER is the cost-effective treatment. Based on the results of the decision analysis, the CER for gatifloxacin is $5,236:1 compared to $7,047:1 for ceftriaxone, a difference of $1,811 per successful outcome in favor of gatifloxacin.

[ILLUSTRATION OMITTED]

Of the 105 patients treated with ceftriaxone, 35 patients (33%) also received IV erythromycin. Although this group is too small to detect statistically significant differences, it was noted that four of the nine ceftriaxone-regimen treatment failures occurred among patients who received both IV ceftriaxone plus IV erythromycin. One of these treatment failures resulted in a hospital readmission; none occurred in ICU-treated patients. Three of the 11 patients randomized to receive ceftriaxone who were treated in the ICU received IV erythromycin in combination. Table 5 depicts the level 1 to level 3 costs by treatment regimen. Level 1 and level 2 costs for ceftriaxone alone are lower than ceftriaxone plus erythromycin. This is not unexpected, as one would predict that an antimicrobial regimen comprised of two agents would have greater drug-acquisition, preparation, dispensing, and administration costs compared to a single-agent regimen. The level 3 costs for ceftriaxone alone exceed those for ceftriaxone plus erythromycin. However, the level 3 cost analysis incorporates the hospital per diem cost; therefore, this finding can be explained by the different LOSAR for treatment failures between these regimens: 12 days for ceftriaxone alone and 8 days for ceftriaxone plus erythromycin.

Sensitivity Analysis

Drug-acquisition costs were varied [+ or -] 25%, but the overall economic decision was not affected. The cost of treatment with gatifloxacin was consistently lower for all permutations. Hospital per diem costs were varied between $885/d and $1,970/d for the LOSAR analysis. Again, the economic decision was not altered.

Results obtained by varying the clinical success rate of each drug independently between 45% and 95% are shown in Figure 3. The expected cost in thousands of US dollars is depicted along the y-axis and the probability of success along the x-axis. The lower line demonstrates the expected cost of treatment with gatifloxacin as the probability of success decreases moving from left to right. The line is relatively flat because the mean cost of a treatment failure ($5,752) was similar to the mean cost of a successful outcome ($5,162). Conversely, the probability of a successful outcome with a ceftriaxone-containing regimen increases moving from left to right along the top axis. The expected cost per patient treated with ceftriaxone decreases sharply as the probability of success increases, reflecting the higher cost per treatment failure ($12,663) found in this cohort. A point of intersection of the two lines would reflect the breakpoint at which the cost-effectiveness decision changes. Gatifloxacin remained the cost-effective alternative over the entire range of probabilities tested.

[ILLUSTRATION OMITTED]

DISCUSSION

We conducted a CEA from the hospital perspective in order to assess the difference in economic outcomes between regimens used for the treatment of CAP patients requiring hospitalization. A CEA is a useful methodology for the consideration of costs (resources used) and consequences (outcomes) when comparing competing medications or treatment modalities. In this analysis, the geometric mean cost per patient treated with gatifloxacin was $1,055 less than a ceftriaxone-containing regimen: $5,109 vs $6,164. Gatifloxacin was the less costly option at all cost levels (levels 1, 2, and 3). Despite using sensitivity analyses over a wide range of probabilities of success, we were unable to change the economic decision.

For level 1 and usually in level 2 cost analyses, the drug-acquisition cost is the predominant cost driver. In the treatment of pneumonia, like other infectious diseases, the cost of hospitalization is the key cost factor.[21] Thus, the cost of treating CAP requiring hospitalization is directly linked to the LOS. We found a trend toward a longer LOS and LOSAR in the ceftriaxone-treated patients. The 1-day increase in the LOS and LOSAR was driven by the four clinical failures who required readmission to the hospital and administration of further antimicrobial therapy. The two clinical failures among those treated with gatifloxacin did not result in rehospitalization. One treatment failure occurred in an elderly female patient with multiple underlying medical conditions who died of pneumonia on day 8 of IV antimicrobial therapy. The second failure was a middle-aged man infected with S pneumoniae. He received 2 days of IV gatifloxacin therapy while hospitalized and an additional 9 days of oral therapy following discharge home. At the test-of-cure visit, the patient received a diagnosis of bronchitis and was prescribed a 5-day course of antibiotic treatment as an outpatient. The mean LOSAR for these failures was 5 days, and the mean cost per patient was $5,752. These results were very similar to the geometric mean LOSAR and level 3 cost (4 days; $5,162) for successfully treated gatifloxacin recipients.

Our economic analysis was an independent evaluation of the data prospectively collected during a randomized, double-blind, controlled clinical trial,[15] The scope of the information gathered is therefore limited by the need for defined enrollment criteria with this study design; however, there were no exclusion criteria that would render the findings not applicable to a wide range of patients.

The 1993 ATS guidelines proposed a definition of severe CAP to aid the clinician with empiric treatment selection and to identify patients potentially requiring ICU admission.[9] Ewig et al[26] conducted a study to validate the ATS criteria for severe CAP. These authors chose to define severe pneumonia as admission to the ICU. According to their definition, 84% of patients admitted to the hospital with CAP during a single respiratory season had nonsevere pneumonia and 16% had severe pneumonia. When the same cohort was categorized according to the ATS guidelines, 68% of patients admitted to the hospital with CAP who met at least one ATS criterion for severe pneumonia did not require admission to the ICU.[26] Thus, the ATS guidelines had high sensitivity (98%) but low specificity and positive predictive value (32% and 24%, respectively).[26] In the clinical trial on which our CEA is based, [is greater than] 70% of patients enrolled were categorized as having severe pneumonia according to protocol criteria based on the ATS guidelines (Table 1), and the admission to ICU rate among these patients was only 9%.[15] When the patients from the gatifloxacin vs ceftriaxone trial were retrospectively categorized according to the Fine criteria, the proportion considered to have severe pneumonia dropped to 40% (class IV, 33%; class V, 7%).[15,27] These findings support the conclusion of Ewig et al[26] that further validation trials are warranted and modification of the ATS definition of severe CAP be considered. Therefore, although it appeared that a large proportion of the patients enrolled in the clinical trial had severe pneumonia, the analysis applies primarily to patients not treated in the ICU.

There was an apparent discrepancy in the number of patients admitted to the ICU from each group: 11 ceftriaxone-treated patients compared to 3 gatifloxacin-treated patients. It should be noted that 9 of the 11 ceftriaxone recipients and all 3 of the gatifloxacin-treated subjects were admitted to the ICU before or on the day of study enrollment. Therefore, this difference was not a treatment group effect but rather a function of the study randomization. When we excluded those patients treated in the ICU and repeated the economic analysis, gatifloxacin consistently remained the cost-effective alternative (p = 0.049), although the overall treatment cost difference decreased to approximately $700.

Current published guidelines from the ATS for adults hospitalized with CAP recommend a second-generation or third-generation cephalosporin or a [Beta]-lactam/[Beta]-lactamase inhibitor with or without a macrolide.[9] The 2000 IDSA guidelines suggest cefotaxime or ceftriaxone with a macrolide, or a [Beta]-lactam/[Beta]-lactamase inhibitor plus a macrolide, or an antipneumococcal fluoroquinolone.[3] These new guidelines provide for improved coverage of atypical organisms based on the findings of three studies.[28-30] In the first study, of 3,000 patients with pneumonia, those treated with a [Beta]-lactam plus a macrolide had reduced mortality compared with the group that received a [Beta]-lactam alone.[28] In another study, involving 13,000 Medicare patients, initial therapy with a nonpseudomonal third-generation cephalosporin plus a macrolide, a second-generation cephalosporin plus a macrolide, or fluoroquinolone monotherapy was associated with 26%, 29%, and 39% lower mortality, respectively, than the reference cohort treated with a nonpseudomonal third-generation cephalosporin alone.[29] The third study included 100 patients hospitalized with CAP.[30] Patients treated with a macrolide within the first 24 h of hospital admission had a shorter LOS (2.8 days) than those who did not (5.3 days; p = 0.01). The effect diminished as the interval before administering macrolides increased. The authors concluded that the use of macrolides as part of the initial therapeutic regimen may be associated with a shorter LOS.[30] Each of these studies evaluated clinical outcomes postulating a correlation with atypical pathogen coverage that has not yet been demonstrated with serologic evidence.

In our analysis, the level 1 and level 2 costs for the group receiving a [Beta]-lactam plus a macrolide were higher than the monotherapy groups. However, the level 3 costs were lower for the ceftriaxone plus erythromycin group compared to the ceftriaxone-alone group. This difference is due to the occurrence of treatment failures in the [Beta]-lactam monotherapy group who required rehospitalization and increased the associated LOSAR. We found a similar LOSAR for a successful outcome between those treated with ceftriaxone alone vs ceftriaxone plus erythromycin. The difference between the groups was in the LOSAR of the treatment failures. There were five failures with a geometric mean LOSAR of 12 days in the ceftriaxone-alone group compared to four failures with a geometric mean LOSAR of 8 days in the ceftriaxone plus erythromycin group (Fig 2). There were two deaths in the ceftriaxone-alone treated patients and none among those who received erythromycin IV initially. The number of patients included in these groups is too small to derive a statistical inference; however, the findings do support the studies[20-30] described suggesting an increased importance of providing atypical pathogen coverage to patients hospitalized for CAP.

The use of fluoroquinolone monotherapy for CAP avoids the need to decide whether atypical pathogen coverage is necessary. Fluoroquinolone monotherapy for CAP with agents such as levofloxacin, sparfloxacin, or trovafloxacin has been shown to be effective.[31,32] In the large study of CAP by File et al[31] comparing IV and/or oral levofloxacin monotherapy vs IV ceftriaxone and/or oral cefuroxime axetil with or without erythromycin or doxycycline, a statistically significant difference in clinical success rate in favor of levofloxacin was found. The clinical success rate at 5 to 7 days after end of treatment was 96% for levofloxacin and 90% for ceftriaxone and/or cefuroxime axetil (95% confidence interval, -10.7 to -1.3). The clinical trial on which our CEA is based was not powered to demonstrate a statistical difference in clinical outcomes, but did show a trend toward improved clinical outcome with gatifloxacin monotherapy (98% vs 91%).

There is evidence to support fluoroquinolone monotherapy as a safe and effective treatment alternative for CAP.[31,32] Rittenhouse et al[16] conducted a cost analysis of the clinical trial of IV and/or oral levofloxacin vs IV ceftriaxone and/or oral cefuroxime axetil and found levofloxacin therapy conferred a statistically significant lower total cost. Our analysis offers further evidence that fluoroquinolone monotherapy, in this case gatifloxacin, represents a cost-effective alternative over the "standard" regimen of IV ceftriaxone with or without an IV macrolide. With the plethora of safe and effective treatment options available for CAP, selecting a regimen that offers an economic advantage is an increasingly important consideration.

REFERENCES

[1] Bartlett JG, Mundy LM. Community-acquired pneumonia. N Engl J Med 1995; 333:1518-1524

[2] Bartlett JG, Breiman RF, Mandell LA, et al. Community-acquired pneumonia in adults: guidelines for management. Clin Infect Dis 1998; 26:811-838

[3] Bartlett JG, Dowell SF, Mandell LA, et al. Practice guidelines for the management of community-acquired pneumonia in adults. Clin Infect Dis 2000; 31:347-382

[4] Marston BJ, Plouffe JF, File TM, et al. Incidence of community-acquired pneumonia requiring hospitalizations: results of a population-based active surveillance study in Ohio. Arch Intern Med 1997; 157:1709-1718

[5] Niederman MS, McCombs JS, Unger AN, et al. The cost of treating Community-acquired pneumonia. Clin Ther 1998; 20:820-837

[6] Marrie TJ. Community-acquired pneumonia: epidemiology, etiology, treatment. Infect Dis Clin North Am 1998; 12:723-740

[7] Fine MJ, Stone RA, Singer DE, et al. Processes and outcomes of care for patients with community-acquired pneumonia: results from the pneumonia patient outcomes research team (PORT) cohort study. Arch Intern Med 1999; 159:970-980

[8] McCormick D, Fine MJ, Coley CM, et al. Variation in length of hospital stay in patients with community-acquired pneumonia: are shorter stays associated with worse medical outcomes? Am J Med 1999; 107:5-12

[9] American Thoracic Society. Guidelines for the initial management of adults with community-acquired pneumonia: diagnosis, assessment of severity, and initial antimicrobial therapy. Am Rev Respir Dis 1993; 148:1418-1426

[10] Mandell LA, Marrie TJ, Grossman RF, et al. Canadian guidelines for the initial management of community-acquired pneumonia: an evidence-based update by the Canadian Infectious Diseases Society and the Canadian Thoracic Society. Clin Infect Dis 2000; 31:383-421

[11] The British Thoracic Society. Guidelines for the management of community-acquired pneumonia in adults admitted to hospital. Br J Hosp Med 1993; 49:346-350

[12] Bauerfeind A. Comparison of the antibacterial activities of the quinolones Bay 12-8039, gatifloxacin (AM 1155), trovafloxacin, clinafloxacin, levofloxacin and ciprofloxacin. J Antimicrob Chemother 1997; 40:639-651

[13] Jones RN, Beach ML, Pfaller MA, et al. Antimicrobial activity of gatifloxacin tested against 1676 strains of ciprofloxacin-resistant Gram-positive cocci isolated from patient infections in North and South America. Diagn Microbiol Infect Dis 1998; 32:247-252

[14] Fukuda H, Hori S, Hiramatus K. Antibacterial activity of gatifloxacin (AM-1155, CG5501, BMS-206584); a newly developed fluoroquinolone against sequentially acquired quinolone-resistant mutants and the norA transformant of Staphylococcus aureus. Antimicrob Agents Chemother 1998; 42: 1917-1922

[15] Fogarty C, Dowell ME, Ellison WT, et al. Treating community-acquired pneumonia in hospitalized patients: gatifloxacin vs. ceftriaxone/clarithromycin. J Respir Dis 1999; 20(suppl A):S60-S69

[16] Rittenhouse BE, Stinnett AA, Dulisse B, et al. Evaluating the costs of levofloxacin and ceftriaxone in inpatient adults with community-acquired pneumonia. Pharm Ther 1999; 24:169-179

[17] Walters DJ, Solomkin JS, Paladino JA. Cost effectiveness of ciprofloxacin plus metronidazole Versus imipenem-cilastatin in the treatment of intra-abdominal infections. Pharmacoeconomics 1999; 16:551-561

[18] McKinnon PS, Paladino JA, Grayson ML, et al. Cost-effectiveness of ampicillin/sulbactam versus imipenem/cilastatin in the treatment of limb-threatening foot infections in diabetic patients. Clin Infect Dis 1997; 24:57-63

[19] Partsch DJ, Paladino JA. Cost-effectiveness comparison of sequential ofloxacin versus standard switch therapy. Ann Pharmacother 1997; 31:1137-1145

[20] Jensen KM, Paladino JA. Cost-effectiveness of abbreviating the duration of intravenous antibacterial therapy with oral fluoroquinolones. Pharmacoeconomics 1997; 11:64-74

[21] Paladino JA, Fell RE. Pharmacoeconomic analysis of cefmenoxime dual individualization in the treatment of nosocomial pneumonia. Ann Pharmacother 1994; 28:384-389

[22] Smith CF, Amen RJ. Comparison of seven methods of preparing and administering small-volume injections. Am J Hosp Pharm 1988; 45:1896-1901

[23] Foran RM, Brett JL, Wulf PH. Evaluating the cost impact of intravenous antibiotic dosing frequencies. DICP Ann Pharmacother 1991; 25:546-552

[24] Hatoum HT. Microcost analysis of inpatient dispensing and administration of oral solids. Am J Hosp Pharm 1990; 47:800-805

[25] Cunha BA. Intravenous-to-oral antibiotic switch therapy. Postgrad Med 1997; 101:111-128

[26] Ewig S, Ruiz M, Mensa J, et al. Severe community-acquired pneumonia: assessment of severity criteria. Am J Respir Crit Care 1998; 158:1102-1108

[27] Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med 1997; 336:243-250

[28] Guglielmo BJ, Dudas V, Tran S, et al. Treatment outcomes associated with community-acquired pneumonia in US hospitals: a 3,000 patient survey. In: Program and abstracts of the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy, September 28 to October 1, 1997; Toronto, Ontario, Canada. Washington DC: American Society for Microbiology, 1997; abstract K-146

[29] Stahl JE, Barza J, DesJardin J, et al. Effect of macrolides as part of initial empiric therapy on length of stay in patients hospitalized with community-acquired pneumonia. Arch Intern Med 1999; 159:2576-2580

[30] Gleason PP, Meehan TP, Fine MJ, et al. Associations between initial antimicrobial therapy and medical outcomes for hospitalized elderly patients with pneumonia. Arch Intern Med 1999; 159:2562-2572

[31] File TM Jr, Segreti J, Dunbar L, et al. A multicenter, randomized study comparing the efficacy and safety of intravenous and/or oral levofloxacin versus ceftriaxone and/or cefuroxime axetil in treatment of adults with community-acquired pneumonia. Antimicrob Agents Chemother 1997; 41:1965-1972

[32] Hooper DC. New uses for new and old quinolones and the challenge of resistance. Clin Infect Dis 2000; 30:243-254

(*) From the Clinical Pharmacokinetics Laboratory (Drs. Dresser and Paladino), State University of New York at Buffalo, Division of Pulmonary and Critical Care Medicine, Winthrop-University Hospital (Dr. Niederman), State University of New York at Stony Brook, Stony Brook, NY.

Dr. Dresser has received funding for educational meeting attendance or speaker bureaus from Bayer, Bristol-Myers Squibb, Ortho-McNeil, and Roche. Dr. Paladino has recently received funding for research grants, consultant activities, or participation in speakers bureaus from Aventis, Bayer, Bristol-Myers Squibb, Dura, Merck, Ortho-McNeil, Pfizer, and Pharmacia. Dr. Niederman has recently received funding for research grants, consultant activities, or participation in speakers bureaus from Aventis, Bayer, Bristol-Myers Squibb, Dura, Glaxo, Ortho-McNeil, Pfizer, Pharmacia, Roche, and Zeneca.

Manuscript received June 29, 2000; revision accepted November 29, 2000.

Correspondence to: Joseph A. Paladino, PharmD, University at Buffalo School of Pharmacy, Department of Pharmacy Practice, 313 Hochstetter Hall, Buffalo, NY 14260; e-mail: drjoepal@hotmail.com

COPYRIGHT 2001 American College of Chest Physicians
COPYRIGHT 2001 Gale Group