Once-Daily Cefepime Versus Ceftriaxone for Nursing HomeAcquired Pneumonia

Joseph A. Paladino, PharmD, w David A. Eubanks, MD,z Martin H. Adelman, PhD, and Jerome J. Schentag, PharmD w

OBJECTIVES: To compare once-daily intramuscular cefepime with ceftriaxone controls. DESIGN: Double-blind study. SETTING: Six skilled nursing facilities. PARTICIPANTS: Residents aged 60 and older with nursing homeacquired pneumonia. INTERVENTION: Cultures were obtained, and patients were randomized to cefepime or ceftriaxone 1 g intramuscularly every 24 hours. MEASUREMENTS: Clinical success: cure or improvement. Cure was dened as complete resolution of all symptoms and signs of pneumonia or a return to the patients baseline state. Improvement was dened as clear improvement but incomplete resolution of all pretherapy symptoms or signs or incomplete return to the patients usual baseline status. Safety and pharmacoeconomics were also assessed. RESULTS: Sixty-nine patients were randomized; 61 were evaluable: (32 to cefepime, 29 ceftriaxone). Patients were predominately female (76%). They had a mean age standard deviation of 856, with a mean 5.81.9 comorbidities; they had age-appropriate renal dysfunction, with a mean estimated creatinine clearance of 357 mL/min. Clinical success occurred in 78% of cefepime- and 66% of ceftriaxone-treated patients (P 5.39). Fifty-seven patients (93%) were switched to oral antibiotics after 3 days. Antibiotic-related adverse events occurred in 5% of patients. Seven patients (11.5%) were hospitalized. The overall mortality rate was 8%. Mean antibiotic costs were $11740 for cefepime- and $21568 for ceftriaxone-treated patients (Po.001). Cost-effectiveness analysis of total costs showed
From the CPL Associates, LLC, Amherst, New York; wSchool of Pharmacy and Pharmaceutical Sciences and zSchool of Medicine, University at Buffalo, State University of New York, Buffalo, New York; and Elder Medical Services, Buffalo, New York. Presented in part at the 42nd Annual Meeting of the Infectious Diseases Society of America, Boston, Massachusetts, October 1, 2004, Abstract 288, and the 45th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, DC, December 16, 2005, Abstract O-188. Address correspondence to Dr. Joseph Paladino, CPL Associates, LLC, 3980 Sheridan Drive, Suite 501, Buffalo, NY 14226. E-mail: paladino@cplassociates.com DOI: 10.1111/j.1532-5415.2007.01152.x

that cefepime would cost $597 and ceftriaxone $1,709 per expected successfully treated patient. One- and two-way sensitivity analyses using a generic price for ceftriaxone and improving its comparative efcacy revealed that the results were robust. CONCLUSIONS: Once-daily cefepime was a cost-effective alternative to ceftriaxone for the treatment of elderly nursing home residents who developed pneumonia and did not require hospitalization. J Am Geriatr Soc 55:651657, 2007. Key words: NHAP; intramuscular cephalosporins; pharmacoeconomics

ursing homeacquired pneumonia (NHAP) is a significant cause of morbidity, mortality, and healthcare resource consumption for residents of long-term care facilities (LTCFs).1,2 Frequently isolated bacteria are similar to those found in community-acquired pneumonia (CAP): Streptococcus pneumoniae, Haemophilus inuenzae, Staphylococcus aureus (although frequently methicillinresistant), Moraxella catarrhalis, and other aerobic gram-negative bacilli;24 Chlamydia pneumoniae and Legionella pneumophila are rarely implicated, although outbreaks can occur in individual LTCFs.5 There is increasing support for treating NHAP without hospitalization.2,68 Intramuscular (IM) cephalosporins are often administered to residents of LTCFs who develop NHAP not amenable to oral treatment but not requiring hospitalization.9,10 Ceftriaxone is frequently used, because its broad spectrum of activity covers most of the pathogens commonly found in NHAP, with the exception of methicillin-resistant S. aureus (MRSA). Ceftriaxones long serum half-life supports once-daily dosing in most patients,11 and its efcacy in treating lower respiratory tract infections has been well established. Ceftriaxone given 1 g every 24 hours resulted in clinical efcacy rates between 73% and 95%, with bacteriologic eradication rates of 93% to 100%.12 Cefepime has excellent in vitro activity against grampositive organisms associated with NHAP (excluding MRSA) with minimum inhibitory concentrations (MICs)

JAGS 55:651657, 2007 r 2007, Copyright the Authors Journal compilation r 2007, The American Geriatrics Society

0002-8614/07/$15.00

652

PALADINO ET AL.

MAY 2007VOL. 55, NO. 5

JAGS

equivalent to ceftriaxone.1315 Cefepime is also active against gram-negative pathogens associated with NHAP, including several that are resistant to ceftriaxone.14 The efcacy of cefepime 2 g every 12 hours was compared with that of ceftriaxone 1 g every 12 hours for the intravenous treatment of 115 hospitalized patients with CAP. Success rates (clinical, microbiological) were similar in cefepime(95%, 100%) and ceftriaxone-treated patients (98%, 97%).14 Although cefepime has a half-life of 2.3 hours in healthy adults,16 as creatinine clearance (ClCr) decreases below 60 mL/min, its half-life progressively increases from 4.9 to 10.5 hours, enabling once-daily dosing for all but the most difcult to treat infections.1618 The IM route of administration of cefepime and ceftriaxone is well tolerated.1,1921 IM cefepime is at least as safe and perhaps less painful on injection than IM ceftriaxone.19 The cost to LTCFs for 1 g of ceftriaxone was three times that of 1 g of cefepime during the study. With equivalent gram-positive and superior gram-negative activity, cefepime may provide a cost-effective alternative to ceftriaxone. The study objective was to assess and compare the efcacy, safety, and pharmacoeconomics of once-daily IM cefepime with those of ceftriaxone for the treatment of elderly LTCF residents with NHAP.

METHODS Study Design and Protocol This was a double-blind randomized trial of residents of LTCF for 30 days or longer who were aged 60 and older and developed NHAP not amenable to oral treatment but not requiring hospitalization. Inclusion criteria were a new inltrate on chest roentgenogram compatible with the diagnosis of pneumonia, estimated ClCr of 60 mL/min or less, and presence of at least one of the following major criteria (cough, hyperthermia ( 381C), hypothermia ( 4 o361C), or purulent sputum) or two of the following minor criteria (dyspnea, pleuritic chest pain, altered mental status, signs of pulmonary consolidation on physical examination, or white blood cell count (WBC) 4 12,000 neutrophils/mm3). ClCr was estimated initially from serum creatinine values22,23 obtained within 6 months of enrollment. Exclusion criteria included allergy to beta-lactams, receipt of an antibiotic for more than 24 hours unless no improvement was evident, an existing infection due to a pathogen known to be resistant to either study drug, expected mortality within 48 hours, immunosuppression (WBC o1,000 neutrophils/mm3), hypotension or respiratory distress requiring ventilatory support, prior limited

treatment or supportive care only directives, suspected Pseudomonas aeruginosa or MRSA infections, primary lung cancer or another malignancy metastatic to the lungs, bronchial obstruction or a history of postobstructive pneumonia, or a requirement for renal dialysis. The University at Buffalo Health Sciences institutional review board (UB HS-IRB) approved the protocol and informed consent form. After each patient or his or her legally recognized representative signed informed consent, pulse oximetry was measured, and blood chemistry, hematology, blood cultures, and whenever possible a sputum sample for Gram stain and culture were obtained. Patients were allowed to receive up to 24 hours of IM antibiotics before the rst dose of study drug. Patients were randomized to receive 1 g IM every 24 hours of cefepime (Elan Pharmaceuticals, San Diego, CA) or ceftriaxone (Roche Pharmaceuticals, Nutley, NJ). To ensure that the volume of each injection was less than 3 mL and to maintain blinding, all doses were split and administered with lidocaine. After 3 days, patients with objective evidence of improvement could be switched to oral antibiotics (Table 1) to complete a 10- to 14-day course. The preferred oral antibiotics9 were amoxicillin/clavulanate acid 875 mg twice daily or 500 mg thrice daily or levooxacin 250 mg once daily (500 mg if estimated ClCr 5 50 60 mL/min). Assessment for oral conversion was repeated daily until the patient met criteria or completed therapy. The test-of-cure (TOC) visit occurred 7 to 14 days after the completion of antibiotic treatment, at which time the physical examination and laboratory tests were repeated. Blood cultures were repeated if previously positive, and a sputum sample was obtained if possible.

Clinical Response Patients were evaluated for safety daily until the TOC and for clinical response at the TOC visit. Success included cure or improvement; cure was dened as complete resolution of all symptoms and signs of pneumonia or a return to the patients baseline state. Improvement was dened as clear improvement but incomplete resolution of all pretherapy symptoms or signs or incomplete return to the patients baseline status. Failure was dened as persistence or progression of symptoms and signs of pneumonia or any of the following: development of new pulmonary or extrapulmonary clinical ndings consistent with active infection, persistence or progression of radiographic abnormalities, death due to pneumonia, inability to complete study antibiotic treatment because of adverse events, treatment required with another antimicrobial regimen, or hospitalization for

Table 1. Criteria for Switch to Oral Antibiotics9

General Improvement in symptoms and signs Afebrile for !16 hours Respiratory rate No acute cardiac or other life-threatening event Able to take oral medications Specific Decreased cough, decreased shortness of breath, increased breath sounds, increased oxygenation 4361C to o381C 20 breaths per minute In the rst 3 days of treatment No vomiting or diarrhea

JAGS

MAY 2007VOL. 55, NO. 5

IM CEFEPIME VERSUS CEFTRIAXONE FOR NHAP

653

progression of pneumonia. Indeterminate was dened as the presence of extenuating circumstances that precluded classication as success or failure, including missed doses of study drug within the rst 4 days and death due to noninfectious causes.

Microbiological Response Patients with a positive sputum or blood culture were reevaluated for their microbiological status on the nal day of parenteral therapy and at the TOC visit, if clinically necessary and possible. Eradication was dened as documented elimination of the causative organism, whereas presumed eradication was dened as the absence of sputum for culture, because the patient had improved clinically or there was no clinical reason for obtaining follow-up blood cultures. Persistence was dened as failure to eradicate the original causative organism. Superinfection, colonization, and reinfection were also considered. Indeterminate was applied to cases in which data were not available or were accompanied by confounding factors preventing clear assessment. Pharmacokinetics/Pharmacodynamics To determine whether 1 gram IM every 24 hours of cefepime would provide adequate antibacterial coverage to the patients, a microbiological assay measuring serum inhibitory titers (SITs) was employed on an initial group of enrolled patients. Steady-state trough serum samples (i.e., obtained immediately predose) were inoculated with a standard bacterium (Escherichia coli #99180, susceptible to ceftriaxone and cefepime, with a MIC o0.06 mg/mL), at 351C for 18 hours. The SIT is dened as the highest dilution of the serum that inhibits visible growth.24 Pharmacoeconomic Analysis A cost-effectiveness analysis was performed from the provider perspective. Standard costs were used for acquisition of study and subsequent antibiotics; ancillary materials required for dose preparation, dispensing, and administration; emergency department visits; hospitalizations; and other pertinent resources. The primary economic endpoint, reecting the reimbursement issues facing LTCFs, was a comparison of the mean cost of antibiotic therapy, including cefepime, ceftriaxone, and subsequent antibiotics for switch therapy or to treat failures of study antibiotic. The costs applied were what the nursing homes paid at the time of the study (cefepime 1 g, $17; ceftriaxone 1 g, $51; LTCF prices to CPL Associates effective April 22, 2004). For patients transported to the emergency department, a cost of $350 was included, and for those hospitalized, a mean per diem of $1,046 was applied.25 Sensitivity analysis was rst applied by decreasing the price of ceftriaxone to $27 to reect its recent availability as a generic product. Two-way sensitivity analysis was then run using the lower ceftriaxone price and altering success and failure rates to the threshold point.26 Cost-effectiveness ratios, consisting of the total cost of care and clinical success rate, were also determined.27,28 Analysis of the incremental cost-effectiveness ratio was planned to assess the relative value of a more-expensive alternative producing superior results.

Statistical Analysis Sample size was determined for the primary economic endpoint using an alpha of 0.05, a beta of 0.02, and an estimate of 25% difference in costs between groups. With a 20% coefcient of variation, a sample size of 22 patients per group was needed. Assuming a 10% nonevaluability rate, a sample of 25 patients enrolled per group would provide an adequate number of patients. All statistical tests were run as two-sided, with the probability of a Type I error less than .05 considered statistically significant. Demographics were compared using the chi-square or Fischer exact test for categorical items and two-way analysis of variance for continuous variables. Clinical response was assessed using chi-square or Fischer exact test, and costs were analyzed using the t test. It was not anticipated that it would be possible to demonstrate clinical differences with the small numbers of patients in this study, and the likely outcome was expected to be noninferiority even if the study population had been much larger. Any patient who received a dose of study drug was included in the safety analysis. RESULTS The study was conducted initially at a 200-bed skilled nursing facility (SNF) in Western New York. To facilitate an adequate rate of patient enrollment, the study was gradually expanded to other SNFs. Patients were enrolled from six SNFs in Western New York ranging in size from 120 to 200 beds. A total of 181 patients were screened for possible enrollment in the study. A resident not meeting inclusion criteria or having an exclusion criterion produced the majority of screening failures, including living in the SNF for less than 30 days (n 5 21), have received more than 24 hours of antibiotic (n 5 16), beta-lactam allergy (n 5 9), ambiguous chest x-ray (n 5 7), younger than 60 (n 5 4), and other reasons (n 5 23). The remainder of the screening failures were due to resident or family refusing consent (n 5 16), patient requiring immediate hospital admission (n 5 11), or the family deciding to limit treatment to comfort care only (n 5 5). Sixty-nine patients were enrolled and randomized between March 27, 2002, and April 7, 2005. There was disagreement regarding the chest x-ray of one patient, and because the radiologists nal report did not support pneumonia, the patient was deemed not evaluable. One patient was withdrawn at family request, ve patients missed doses of study antibiotic, and one patient grew MRSA from the initial sputum culture. Thus, there were 61 evaluable patients: 32 who received cefepime and 29 who received ceftriaxone. Demographics of the clinically evaluable study population are shown in Table 2. The patients were predominately elderly women with multiple comorbidities and age-appropriate renal dysfunction. There were no signicant differences in patient characteristics except for a slightly slower ClCr in the ceftriaxone group (P 5.04). All subsequent data, with the exception of the safety assessment, pertain only to the 61 evaluable patients. Seventeen patients received prestudy IM cephalosporin. There were 12 such patients in the ceftriaxone-treated group; one received a prestudy dose of ceftriaxone 1 g and 11 received one to two doses of cefotaxime 1 g every 12 hours. There were ve such patients in the cefepime-treated

654

PALADINO ET AL.

MAY 2007VOL. 55, NO. 5

JAGS

Table 2. Demographic Characteristics of the Evaluable Study Population

Characteristic Age, meanSD (range) Female, n (%) Weight, kg, meanSD (range) Comorbidities Estimated creatinine clearance, mL/min, meanSD (range)
SD 5 standard deviation.

Cefepime Group (n 5 32) 856 (6897) 24 (75) 6418 (38114) 5.61.8 367 (2055)

Ceftriaxone Group (n 5 29) 866 (7198) 22 (76) 6417 (3699) 6.12 328 (2050)

P-value .46 4.99 .80 .29 .04

group; each received one to two doses of cefotaxime 1 gram every 12 hours. The patients universally complained of pain on injection with cefotaxime but not with ceftriaxone or cefepime. Prestudy antibiotics were not included in data analyses.

two cefepime-treated patients (6.2%) and three ceftriaxone-treated patients (10.3%).

Serum Inhibitory Titers Trough serum samples were obtained from nine patients. The mean SIT was 1:64 (range 1:41:128). The results were sufcient for the UB HS-IRB to allow the study to proceed. Postblinding, it was determined that the samples were obtained from six patients who received cefepime and three patients who received ceftriaxone. Cefepime and ceftriaxone each produced a mean trough SIT of 1:64. Clinical Response Clinical success was achieved in 78% of patients receiving cefepime and 66% of those receiving ceftriaxone (P 5.39) (Table 3). Patients received IM study cephalosporin for a mean of 3 days (range 15 days for each); 57 (93%) were switched to oral antibiotics. Levooxacin (n 5 45) was the most frequently used oral antibiotic, followed by cefuroxime (n 5 7), cefprozil (n 5 3), and amoxicillin/clavulanate (n 5 2). Most patients experienced mild to no discomfort at the site of IM injection of ceftriaxone or cefepime; if present, it abated quickly. One patient with a history of diabetes mellitus had high blood glucose while receiving ceftriaxone. Other drug-related adverse events occurred rarely and only with the oral antibiotics. One patient with a history of seizures experienced one while receiving levooxacin. Another patient developed Clostridium difficileassociated diarrhea while receiving amoxicillin/clavulanate. Seven patients (11.5%) were hospitalized: two cefepime-treated patients (6.2%) for 10 days total and ve ceftriaxone-treated patients (17.2%) for 25 days total. Five patients (8%) died as a consequence of their pneumonia or its sequelae:
Table 3. Clinical Response of Evaluable Patients
Treatment Group (n) Cefepime (32) Ceftriaxone (29) Cure Improvement Failure Indeterminate n 11 6 14 13 7 8 0 2

Bacteriological Results Baseline sputum or blood cultures were positive for six patients (9.8%), three in each group. In the cefepime-treated group, a coagulase-negative staphylococcus (considered a contaminant) grew from one of two blood cultures from one patient who responded to the antibiotic. The sputum culture of another patient grew H. inuenzae and M. Catarrhalis, which responded to cefepime. The sputum culture of the third patient grew H. inuenzae. She responded initially to cefepime and after 3 days was switched to oral levooxacin 250 mg for 7 additional days. One week later, she deteriorated, with increased cough, shortness of breath, and general malaise, for which she was started on a macrolide. In the ceftriaxone-treated group, the initial sputum culture from one patient revealed colonization with P. aeruginosa, with the Gram stain showing less than 25 WBCs/ low-power eld, although after initial favorable response to ceftriaxone and switch to oral levooxacin, the organism persisted on repeat sputum culture. The patient then received a second course of levooxacin. Another patient grew E. coli and methicillin-sensitive S. aureus from a sputum culture, which responded to ceftriaxone. A third patient grew H. inuenzae and later MRSA from a sputum culture; after a variable response, he was subsequently changed to a course of trimethoprim-sulfamethoxazole. Pharmacoeconomic Results Mean per patient costs for the study IM cephalosporins, oral antibiotics, and hospitalization are shown in Table 4. The costs of those who failed treatment were many times greater than those of patients who were successfully treated, largely because of the costs of hospital care (Figure 1). Individual cost-effectiveness ratios of total costs indicated that cefepime would cost $597 and ceftriaxone $1,709 per expected successfully treated patient. Because the ceftriaxone group had no clinical advantage and higher costs, cefepime was the economically dominant choice, and an incremental cost-effectiveness analysis was unnecessary. Sensitivity Analyses Ceftriaxone was available during the study only as a branded product. It is now available as a generic, with a lower

JAGS

MAY 2007VOL. 55, NO. 5

IM CEFEPIME VERSUS CEFTRIAXONE FOR NHAP

655

Table 4. Cost per Evaluable Patient ($)

Cefepime Group (n 5 32) Variable Intramuscular cephalosporin Oral antibiotics Total antibiotics Hospital Total 51 6637 11740 3491,311 4661,295 Ceftriaxone Group (n 5 29) P-value 151 6348 21568 9142,174 1,1282,156 o.001 .79 o.001 .23 .16

MeanStandard Deviation

price. Therefore, the cost was reanalyzed using the new price of $27 per g. This reduced the mean cost of IM ceftriaxone regimens from $151 to $80 and reduced the total antibiotic cost from $215 to $143, although the cost remained higher than that of cefepime treatment. The mean total cost for the ceftriaxone-treated patients was reduced from $1,128 to $1,057, which did not approach the $466 cost for cefepime treatment. A two-way sensitivity analysis was then performed using the generic ceftriaxone price and changing clinical success rates in the decision tree to the threshold point, whereby the economic decision would no longer apply. The rst analysis increased the efcacy rate of ceftriaxone while holding cefepimes stable. In order for the mean total cost of generic ceftriaxone to be less than that of cefepime, ceftriaxones efcacy rate would have to increase 22% (to 88%). The second analysis decreased the efcacy rate of cefepime while holding ceftriaxones stable. In order for the mean total cost of generic ceftriaxone to be less than that of cefepime, cefepimes efcacy rate would have to decrease 38% (to 40%). The third analysis employed an iterative process whereby the efcacy rate of cefepime decreased as the efcacy rate of ceftriaxone increased. In order for the mean total cost of generic ceftriaxone to be less than that of
Success 0.78 $115 $32

cefepime, cefepimes efcacy rate would have to decrease 15% (to 63%), whereas ceftriaxones efcacy rate would have to increase 15% (to 81%). Each of the sensitivity analyses showed the base results to be robust.

Cefepime $466 NHAP

Failure 0.22

$1,720 $2,513

Success Ceftriaxone $1,057 0.66

$137 $16

Failure 0.34

$2,806 $3,058

DISCUSSION The Food and Drug Administrationapproved product labeling for cefepime lists the recommended dosage schedule for moderate to severe pneumonia in patients with normal renal function as 1 to 2 g every 12 hours,29 although if the patients ClCr is 30 to 60 mL/min, the recommended dosing schedule is 1 to 2 g every 24 hours. To ensure a study population of patients with a ClCr less than 60 mL/min, only patients aged 60 and older were eligible. Simple calculation of estimated ClCr22,23 reveals that, for patients aged 60 and older, men with a serum creatinine of 1.1 mg/dL or greater and women with a serum creatinine of 1 mg/dL or greater will have an estimated ClCr 60 mL/min or less (Table 5). This study provides clinical evidence that the dosing recommendation is appropriate. The microbiological assay tests, SIT, run with actual study patients serum obtained at the time when the antibiotic concentrations would be at their lowest, demonstrated that both study antibiotics were able to inhibit the growth of the test bacterium. Despite not being a common pathogen in NHAP, E. coli was chosen as the bacterium for the in vitro test, because it was equally susceptible to ceftriaxone and cefepime. Thus, it was established that both study antibiotics, dosed at 1 g IM every 24 hours, provided potent antibacterial coverage throughout the dosing interval. Using the IM route for administration ensures effective delivery of the antibiotics. Although the antibiotics were well tolerated, patients and staff welcomed the switch to an oral antibiotic. It has previously been demonstrated that
Table 5. Minimum Serum Creatinine (mg/dL) Necessary for Creatinine Clearance (ClCr) 60 mL/min
Age 50 60 70 80 90 Male 1.3 1.1 1 0.9 0.7 Female 1.1 1 0.9 0.7 0.6

Figure 1. Decision tree of one-way sensitivity analysis using a generic price for ceftriaxone. The open diamond represents a choice node, where the randomization scheme determined the selection of whether a patient received cefepime or ceftriaxone. The open circles represent chance nodes, meaning that whether a patient ended up in the successfully treated or failed treated paths could not be predetermined, chosen, or controlled. The values under each terminal branch indicate the probability of a patient following that path. The data in the terminal boxes represent the mean coststandard deviation for patients with that outcome. The per-patient mean total cost for choosing cefepime or ceftriaxone appears under the respective antibiotic names. NHAP 5 nursing homeacquired pneumonia.

ClCrmale 5 (1140.9 (age))/serum creatinine. ClCrfemale 5 Clcrmale(0.84).

656

PALADINO ET AL.

MAY 2007VOL. 55, NO. 5

JAGS

early conversion to oral therapy is possible if patients respond to parenteral treatment, even if they are in an acute care facility, are septic, and are receiving intravenous antibiotics.30 Conducting a double-blind randomized, clinical trial in a LTCF presented many challenges. LTCFs are populated with many individuals who have dementia or are otherwise incapable of providing informed consent. Combined with the altered mental status often associated with pneumonia, the availability of local family members willing to give consent proved a crucial element, because time was a factor. Moreover, professional and support staff responsibilities do not typically include supporting interventional research studies. Logistical arrangements with radiology, laboratory, and pharmacy affected the decision to conduct the study at particular facilities. To complete this study, it was necessary to increase the number of participating LTCFs gradually, as well as extend the enrollment period through a third year. Treating patients in their residence is preferable to admitting them to a hospital. Acute care facilities are not intended, designed, or staffed to provide the daily functional support7 so necessary for the quality of life of incapacitated individuals. If the hospitalization extends beyond a specific time period, residents may be faced with losing their nursing home bed. But it is more than a bed, it is their homeFreplete with a known environment, familiar staff, and friends close by. The ability to deliver potent therapy while residents remain in their LTCF is an important medical service.

Schering-Plough, Wyeth, and ViroPharma. Dr. Schentag has received research support from Bayer, CUBIST, Elan, Oscient, Pzer, Sano-Aventis, and honoraria for lectures and/or consulting from AstraZeneca, Bayer, Bristol, CUBIST, Elan, GlaxoSmithKline, King, Merck, Ortho-McNeil, Pzer, Pharmacia, Sano-Aventis, Schering-Plough, Wyeth, ViroPharma, Oscient, and Nektar. Author Contributions: Paladino, Adelman, and Schentag had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Paladino. Acquisition of data: Paladino and Eubanks. Analysis and interpretation of data: Paladino, Eubanks, Adelman, and Schentag. Drafting of the manuscript: Paladino. Critical revision of the manuscript for important intellectual content: Eubanks, Adelman, and Schentag. Statistical analysis: Adelman. Obtained funding: Paladino. Administrative, technical, or material support: Eubanks, Adelman, and Schentag. Study supervision: Paladino and Schentag. Sponsors Role: Elan Pharmaceuticals reviewed the protocol but had no role in the study design; the collection, analysis, or interpretation of the results; the decision to publish the ndings; or the writing of the manuscript.

REFERENCES
1. Phillips SL, Branaman-Phillips J. The use of intramuscular cefoperazone versus intramuscular ceftriaxone in patients with nursing home-acquired pneumonia. J Am Geriatr Soc 1993;41:10711074. 2. Medina-Walpole AM, Katz PR. Nursing home-acquired pneumonia. J Am Geriatr Soc 1999;47:10051015. 3. Marrie TJ, Slayter KL. Nursing home-acquired pneumonia. Treatment options. Drugs Aging 1996;8:338348. 4. Mylotte JM, Ksiazek S, Bentley DW. Rational approach to the antibiotic treatment of pneumonia in the elderly. Drugs Aging 1994;4:2133. 5. Strausbaugh LJ, Sukumar SR, Joseph CL. Infectious disease outbreaks in nursing homes: An unappreciated hazard for frail elderly persons. Clin Infect Dis 2003;36:870876. 6. Zimmer JG, Hall WJ. Nursing home-acquired pneumonia: Avoiding the hospital. J Am Geriatr Soc 1997;45:380381. 7. Fried TR, Gillick MR, Lipsitz LA. Short-term functional outcomes of longterm care residents with pneumonia treated with and without hospital transfer. J Am Geriatr Soc 1997;45:302306. 8. Thompson RS, Hall NK, Szpiech M. Hospitalization and mortality rates for nursing home-acquired pneumonia. J Fam Pract 1999;48:291293. 9. Naughton BJ, Mylotte JM. Treatment guideline for nursing home-acquired pneumonia based on community practice. J Am Geriatr Soc 2000;48:8288. 10. Mylotte JM. Nursing home-acquired pneumonia. Clin Infect Dis 2002;35: 12051211. 11. Geny F, Costa P, Bressolle F et al. Ceftriaxone pharmacokinetics in elderly subjects and penetration into epididymis. Biopharm Drug Dispos 1993;14: 161169. 12. Lamb HM, Ormrod D, Scott LJ et al. Ceftriaxone. An update of its use in the management of community-acquired and nosocomial infections. Drugs 2002;62:10411089. 13. Wynd MA, Paladino JA. Cefepime. A fourth-generation parenteral cephalosporin. Ann Pharmacother 1996;30:14141424. 14. Zervos M, Nelson M. Cefepime versus ceftriaxone for empiric treatment of hospitalized patients with community-acquired pneumonia. The Cefepime Study Group. Antimicrob Agents Chemother 1998;42:729733. 15. Chapman TM, Perry CM. Cefepime. A review of its use in the management of hospitalized patients with pneumonia. Am J Respir Med 2003;2:75107. 16. Barbhaiya RH, Knupp CA, Forgue ST et al. Disposition of the cephalosporin cefepime in normal and renally impaired subjects. Drug Metab Dispos 1991;19:6873. 17. Barbhaiya RH, Knupp CA, Forgue ST et al. Pharmacokinetics of cefepime in subjects with renal insufciency. Clin Pharmacol Ther 1990;48: 268276. 18. Barbhaiya RH, Knupp CA, Pittman KA. Effects of age and gender on pharmacokinetics of cefepime. Antimicrob Agents Chemother 1992;36: 11811185.

CONCLUSION Cefepime and ceftriaxone, each dosed 1 g IM every 24 hours, produced acceptable clinical success rates with minimal adverse events in this population of patients with estimated ClCr of 60 mL/min or less. These antibiotics were well tolerated, in contrast to prestudy cefotaxime, which consistently produced much pain on injection. The mortality rates were within the range of the 7% to 19% cited for residents treated in nursing homes.10 Pharmacoeconomic analysis revealed that cefepime can save a LTCF approximately $100 per patient in antibiotic costs alone with a trend toward a reduction in hospital costs. Sensitivity analysis showed that, despite generic-priced ceftriaxone, cefepime would continue to cost the LTCF less in antibiotic purchases. When the effect of changing efcacy rates was assessed, it was found that there would have to be a dramatic decline in the performance of cefepime or a dramatic improvement in the performance of ceftriaxone to change the economic decision. Once-daily cefepime was a safe, efcacious, and cost-effective alternative to ceftriaxone for the treatment of elderly LTCF residents who develop NHAP and do not require hospitalization. ACKNOWLEDGMENTS Financial Disclosure: This study was supported by an un restricted research grant from Elan Pharmaceuticals. Dr. Paladino has received research support from Bayer, CU BIST, Elan, Pzer, Sano-Aventis, Oscient, and honoraria for lectures and/or consulting from AstraZeneca, Bayer, Bristol, CUBIST, Elan, GlaxoSmithKline, King, Merck, Ortho-McNeil, Pzer, Pharmacia, Sano-Aventis, Oscient,

JAGS

MAY 2007VOL. 55, NO. 5

IM CEFEPIME VERSUS CEFTRIAXONE FOR NHAP

657

19. Gubbelmans HL, Maesen FP. Cefepime versus ceftriaxone: A tolerance study by intramuscular administration. J Drug Ther Res 1990;15:124126. 20. Rybak M. The pharmacokinetic prole of a new generation of parenteral cephalosporin. Am J Med 1996;100:39S44S. 21. Barbhaiya RH, Knupp CA, Tenney J et al. Safety, tolerance, and pharmacokinetics of cefepime administered intramuscularly to healthy subjects. J Clin Pharmacol 1990;30:900910. 22. Paladino JA, Kapfer JA, DiBona JR. Bedside estimation of creatinine clearance. Hosp Formul 1986;21:709715. 23. Canaday BR, Poe TE, Sawyer WT et al. Fractional adjustment of predicted creatinine clearance in females. Am J Hosp Pharm 1984;41:18421843. 24. Burgess DS, Hastings RW, Hardin TC. Pharmacokinetics and pharmacodynamics of cefepime administered by intermittent and continuous infusion. Clin Ther 2000;22:6675.

25. The DRG Handbook. Comparative Clinical and Financial Benchmarks. Evanston, IL: Solucient Publications, 2006. 26. Weinstein MC, Fineberg HV, Elstein AS et al. Probabilities and Clinical Decisions. Clinical Decision Analysis. Philadelphia: W.B. Saunders Co., 1980, p 62. 27. Karlsson G, Johannesson M. The decision rules of cost-effectiveness analysis. Pharmacoeconomics 1996;9:113120. 28. Freund DA, Dittus RS. Principles of pharmacoeconomic analysis of drug therapy. Pharmacoeconomics 1992;1:2031. 29. Maxipime (cefepime) package insert. San Diego, CA: Elan Pharmaceuticals, 2003. 30. Paladino JA, Sperry HE, Backes JM et al. Clinical and economic evaluation of oral ciprooxacin after an abbreviated course of intravenous antibiotics. Am J Med 1991;91:462470.