An algorithm to determine. antimicrobial therapy based

Transcription

1 7 An algorithm to determine cost-savings of targeting antimicrobial therapy based on the results of rapid diagnostic testing J Clin Microbiol. 2003: 41 (10): JJ Oosterheert, MJM Bonten, E Buskens, MME Schneider, IM Hoepelman

2 Chapter 71 Abstract A rapid diagnosis of pneumococcal pneumonia may allow earlier use of narrow-spectrum antimicrobial therapy. It is however unknown whether rapid diagnostic testing in patients hospitalised with community-acquired pneumonia (CAP) is cost-saving. Therefore, an algorithm to calculate costs associated with diagnosis and treatment of CAP was formulated. Subsequently, the algorithm was applied to clinical data of one hundred and twenty two consecutively admitted patients with CAP, from whom sputum samples were Gram stained and urine was tested for Streptococcus pneumoniae antigen. Costs of initial antimicrobial therapy, personnel and materials were measured. When compared to the most expensive empiric regimen, rapid diagnostic testing would result in a cost saving per patient (PP) of 3.51 for GS and of 8.11 for urinary pneumococcal antigen testing. When compared to the cheapest regimen Gram staining would increase cost by 2.25 PP and urinary antigen testing by PP. In our setting the use of rapid diagnostic testing would not be cost saving. Cost savings depend however on price differences of different antibiotic choices and the proportion of evaluable and positive samples. 116

3 Chapter 7 Introduction In addition to therapeutic efficacy, rising costs have become a major concern in treating patients with serious infections. An important factor associated with high costs in treating these patients is the use of unnecessary broad spectrum antibiotics. Therefore, apart from microbiological and therapeutic considerations, from an economical perspective, strategies to decrease the unnecessary use of these agents are wanted. In an attempt to cover all suspected pathogens, initial antibiotic therapy is often broad spectrum. Results of microbiological investigations can help in targeting antimicrobial therapy to isolated pathogens, an approach known as streamlining (1). However, results from diagnostic procedures such as microbiological cultures or serological tests, have a delay of days to weeks and are, therefore, not suitable to guide therapy in an early stage of the disease. Other diagnostic procedures, however, yield almost instantaneous results and could be potentially useful in guiding initial antimicrobial therapy. In this analysis, we evaluate potential cost-savings associated with the use of rapid diagnostic tests to guide initial antimicrobial therapy in patients hospitalised with community-acquired pneumonia (CAP). As the causative micro-organism cannot be predicted from clinical, laboratory or radiological findings (2;3;4), initial antimicrobial therapy is mostly empiric covering different potential pathogens. Among different pathogens, S. pneumoniae is the most prevalent causative micro-organism, and found in up to 40% of episodes (5-7). Especially in areas with low resistance rates, S. pneumoniae can be adequately treated with narrow spectrum antibiotics, such as penicillin G or amoxicillin instead of more broad spectrum agents like cefriaxone both with or without a macrolide, or levofloxacin (8). Diagnostic procedures which can be used in the diagnostic workup of CAP that provide results within minutes are sputum Gram staining and antigen testing for pneumococci in urine. Advantages of sputum Gram staining include its wide availability and low costs. However, adequate sputum samples cannot always be obtained, either because there is no sputum production or because samples are not adequate for evaluation. Furthermore, sensitivity and specificity are unknown, 117

4 Chapter 71 some bacteria cannot be identified and a uniform definition of a positive stain does not exist. (9;10) For these reasons, the use of sputum Gram stain is controversial: its use is recommended by the Infectious Diseases Society of America (IDSA), but not by the American Thoracic Society (ATS) (11;12) Another way to rapidly diagnose pneumococcal pneumonia is urinary antigen testing. An immunochromatographic test, the NOW S. pneumoniae urinary antigen test (Binax, Inc. Portland, Maine) detects the C polysaccharide wall antigen common to all S. pneumoniae strains (13), with results being available within 15 minutes. Preliminary results suggest that this method has an % specificity and 74% sensitivity (14). Another report showed a lower specificity: test results could also be positive in patients who are nasopharyngeal carriers of pneumococci. (15) A rapidly established diagnosis of pneumococcal pneumonia may result in cheaper empiric antibiotics. However, it is unknown whether the potential cost savings outweigh the costs for personnel and materials. Therefore, we developed a simple algorithm to assess the potential costs and savings associated with rapid diagnostic testing for pneumococcal pneumonia using sputum Gram stain or a urinary pneumococcal antigen test and evaluated cost-savings in 122 consecutively admitted patients with CAP. Patients and methods Patients and setting The study was approved by the local ethics committee and all patients provided informed consent to participate. The University Medical Centre Utrecht is a 1042 bed tertiary care hospital. The department of medicine consists of 2 general internal medicine wards and 1 ward for acute medicine and infectious diseases. Together they accounted for 3036 admissions in All patients hospitalised with CAP between November 2000 and November 2002 on internal medicine wards in the University Medical Centre Utrecht with severe CAP (Fine class IV, V (16) or fulfilling the criteria for severe communityacquired pneumonia as defined by the American Thoracic Society (11) ) were included. Patients which needed mechanical ventilation in an intensive care unit were not included. Initial therapy, age and 118

5 Chapter 7 severity of pneumonia as defined by Fine (16) were documented. CAP was defined as a new or progressive infiltrate on chest X-ray and two or more of the following criteria: cough, production of purulent sputum, rectal temperature above 38 C or below 36 C, auscultatory findings consistent with pneumonia, leucocytosis (>10,000/mm) or CRP > 3x upper limit of normal. Patients with cystic fibrosis, neutropenic patients (<0,5 x 10 9 neutrophils / l) and patients in which another infection needed treatment, were excluded. All patients were encouraged to provide a sputum and urine sample, however, we only used samples provided within 24h of hospitalisation in the analysis. Microbiological assessment Sputum samples were evaluated in the microbiological laboratory and Gram stained according to standard techniques. Sputum samples were considered evaluable if no more than 10 squamous epithelial cells and more than 20 neutrophils per low power field were visible, and were considered positive for pneumococci when >10 Gram positive cocci per LPF were present as the predominant organism. To identify pneumococcal urinary antigen we used the NOW Streptococcus pneumoniae urinary antigen test, provided by Binax Inc., Portland. Serology samples for Chlamydia pneumoniae, C. psittaci, Legionella pneumophila and Mycoplasma pneumoniae, blood cultures and sputum cultures were obtained and evaluated according to standard procedures. In addition, to identify L. pneumophila, we used a urinary antigen test (Binax NOW ). Cost assessment Antimicrobial costs were based on the actual cost prices of the antibiotics paid by the department of Clinical Pharmacy of the University Medical Centre in Utrecht. We calculated the potential cost-reduction when therapy would be streamlined based on the results of rapid diagnostic tests before culture results become available. Therefore, only antimicrobial costs for the first three days of therapy were calculated. Duration of preparation and handling of medication were measured twice for all relevant antibiotics. Average costs per antibiotic and per combination of antibiotics for three days 119

6 Chapter 71 were calculated, which included costs for personnel (nurses wages for the time used for preparation and administration) and used materials (needles, syringes, antibiotics, intravenous solutions etc.), as described previously (17). Diagnostic costs (i.e. costs for preparation and examining Gram stains and performing urinary antigen test) were based on the hospital s tarif system, which includes wages for personnel and costs for material. To evaluate the potential cost-reduction, we formulated an algorithm. The algorithm to analyse cost reduction In the algorithm for cost reduction it was assumed that antimicrobial therapy would be streamlined based on a positive urinary pneumococcal antigen test or a positive Gram stain. Only samples obtained on the first day of hospitalisation and of sufficient quality for microbiological analysis were used. Patients without a positive test, either because sampling was not performed at all or not performed within 24h hospitalisation, or because test results were negative, received broadspectrum therapy. The difference in total costs between targeted narrow-spectrum therapy based on positive diagnostic tests and empiric and broad-spectrum therapy was defined as cost reduction, which could be expressed as: Cost Reduction = Costs for empirical therapy (Costs for targeted therapy in patients with positive test results + Costs for empirical therapy in patients with negative test results + Costs of diagnostic procedures) In formula: Cost Reduction = N p * C Emp { P Ev * P Pos * C Targ + [N p (P Ev * P Pos )] * C Emp + C Dx * N Tests } With N p = number of patients; C Emp = costs of empirical therapy; P Ev = % adequate samples, P Pos = % positive adequate samples; C Targ = costs of targeted therapy; ΔP = Price difference between empiric therapy and targeted therapy; C Dx = diagnostic costs of the test; N tests = number of tests that can be performed. Resolution of this equation results in: Cost Reduction = (N p * P Ev * P Pos * ΔP) - (N tests * C Dx ) {1} 120

7 Chapter 7 Sensitivity analysis A sensitivity analysis was performed by calculating the cost outcomes when varying the amount of evaluable samples (P Ev ), the amount of samples positive for pneumococci (P Pos ) and the price difference between broad spectrum and targeted antibiotics (ΔP). Results Patients One hundred and twenty two consecutive patients (84 male) admitted with CAP were evaluated (N p =122). The mean age of the population was (Standard deviation (SD) years, range 28-96) with a mean Fine score of (SD 28.17, range ). 67 (54.9%) fell in PSI risk class IV, 27 patients in risk class V (22.1) and 26 patients (21.7%) fulfilled the ATS-criteria for severe communityacquired pneumonia (11). Initial therapy consisted of amoxicillin / clavulanic acid in 68 (56%) patients, in twelve patients in combination with a macrolide, of ceftriaxone in 48 (39%) patients, in five patients in combination with a macrolide. One patient was switched from amoxicillin / clavulanic acid to erythromycin plus rifampicine as soon as a urinary antigen test indicated Legionella pneumophila infecion, 1 patient received trimethoprim / sulfamethoxazol and 1 patient was treated with ciprofloxacin. Twenty-eight (23%) patients had received prior antibiotic treatment before hospital admission. Ultimately, a causative agent for CAP was identified in 54 of 122 (44%) patients (Table 1). In another 12 (10%) patients a positive urinary antigen test was the only indicator of S. pneumoniae infection. Sputum samples of 52 patients (43%) (N tests = 52) were Gram stained during the first day of hospitalisation. Of these 52 Gram stains, 23 (19% of all patients) were evaluable (P Ev = 0.19), and in ten samples Gram positive cocci could be identified. However, Gram positive cocci were considered the predominant microorganism in only seven of these seven samples (7/52, P Pos =0.13). In one sample another predominant micro-organism was identified and in two samples multiple pathogens were present. Eighty-five patients (70.0%) (N tests = 85, P Ev = 0.70) provided urine samples, 23 of which 121

8 Chapter 71 (23/85, P Pos = 0.27) were positive for pneumococcal antigen. Microbiological cause Total no. of patiens (%) Detection with S. pneumoniae 27 (22.1%) (12 (9.8%) of wich had only positive urinary antigen test) Sputum culture Blood culture Urinary antigen test 9 (7.4%) 8 (6.6%) 23 (18.9%) H. influenzae 3 (2.5%) S. aureus 8 (6.6%) K. pneumoniae 3 (2.5%) M. catharralis 2 (1.6%) Sputum culture Sputum culture Blood culture Sputum culture Sputum culture 8 (6.6%) 2 (1.6%) 2 (1.6%) C. pneumoniae 12 (9.8%) L. pneumophila 3 (2.5%) Serology Urinary antigen test 3 (2.5%) 2 (1.6%) M. pneumoniae 2 (1.6%) Serology 2 (1.6%) E. coli 5 (4.1%) Sputum culture Blood culture 4 (3.3%) 1 (0.8%) Citrobacter freudii, corynebacterium, Streptcoccus oralis, Enterobacter cloacae, P. aeruginosa 1 (0.8%) No microbiological cause 68 (55.4%) Multiple pathogens 13 (10.7%) Table 1 Ultimate microbiological outcome. Cost calculations Costs per dosage of antibiotic in our hospital ranged from 0.80 for penicillin G to for ceftriaxone. Average material costs (needles, syringes, infusion fluids etc.) per dosage were Average time for preparing and dispensing antibiotics were 4 minutes 25 seconds (ranging from 4 minutes to 4 minutes 50 seconds) per dosage, which would mean an average nurses wage of 0.89 per dosage prepared. When including the number of dosages per day and preparation 122

9 Chapter 7 and handling costs amoxicillin was cheapest ( per day) and penicillin G, given 6 times daily, was most expensive ( per day). Calculated costs of combinations of therapy per day were for amoxicillin / clavulanic acid combined with erythromycin and for ceftriaxone combined with erythromycin. (Table 2) The costs for performing a sputum Gram stain were 2.42 ( 0.78 material costs and 1.64 personnel costs). The costs for performing a urinary antigen test for pneumococcal pneumonia were ( material costs and 5.59 wages for personnel). Antibiotic Dosis / frequency Total preparation costs for 3 days* Drug-cost for 3 days Total cost for 3 days Costs per patient per day Penicillin G 1 milj U q 4h Amoxicillin 1000 mg q 8h Amoxicillin / clavulanic acid 1200 mg q 8h Ceftriaxone 2000 mg q 24h Augmentin / erythromycin 1200 mg q 8h / 1000 mg q 8h Ceftriaxone / erythromycin 2000 mg q 24h / 1000 mg q 8h Augmentin / azithromycin 1200 mg q 8h/ 500 mg q 24h orally Ceftriaxone / azithromycin 2000 mg q 24h / 500 mg q 24h orally Average costs of antibiotics instituted Table 2 Antibiotic preparation and total costs for different antibiotic regimens in euro. * preparation costs were measured: average preparation plus administration time was 4 25 per dosage, wich means nurses wages of 0.89 per dosage. average material costs (needles, syringes, intravenous solutions) were 7.51 per dosage prepared. Total preparation costs per dosage were on average. 123

10 Chapter 71 Sputum Gram stain Using the data of our patient population and formula {1} results in the following algorithm for cost-reduction: (122 * 0.19 * 0.13 * ΔP) (52 * 2.42) = When targeted therapy consists of amoxicillin, given 3 times daily ( ) and recommended initial therapy would consist of the most expensive empiric regimen (ceftriaxone and erythromycin, ΔP = ) cost reduction would be for the total population ( 3.51 per patient). However, when targeted therapy is compared to the actual average costs for initial therapy as prescribed in our population (ΔP = 33.55) Gram staining would cost ( 0.20 per patient) When targeted therapy with Penicillin G, given 6 times daily ( ), is compared to initial therapy with the most expensive empiric regimen (ceftriaxone and erythromycin, ΔP = ) Gram staining would result in a total cost saving of ( 1.98 per patient), but when compared to the cheapest regimen (amoxicillin / clavulanic acid, ΔP = ), Gram staining would cost ( 2.25 per patient). When compared to the average costs for therapy as prescribed in our patient population (ΔP = ) Gram staining would cost ( 1.74 per patient). Urinary pneumococcal antigen testing For urinary testing, formula {1} results in the following algorithm for cost reduction (122 * 0.70 * 0.27 * ΔP) (85 * 21.39) = When targeted therapy consists of amoxicillin ( ), given 3 times daily and recommended initial therapy would consist of most expensive empiric regimen (ceftriaxone and erythromycin, ΔP = ) cost reduction is ( per patient). However, when targeted therapy is compared to the actual average costs for initial therapy as prescribed in our population, (ΔP = 33.55) urine antigen testing would cost ( 8.56 per patient) When therapy with Penicillin G, given 6 times daily ( ), is compared to the most expensive empiric regimen (ceftriaxone and erythromycin, ΔP = ) urinary antigen testing would result 124

11 Chapter 7 in a cost saving of ( 8.11 per patient). In contrast, when compared to the cheapest regimen (amoxicillin / clavulanic acid, ΔP =-49.50), urinary antigen testing would cost ( per patient). When compared to the average costs for initial therapy ( ΔP= 28.57), urinary antigen testing would cost ( per patient) The cost calculations are displayed in table 3. Variable Gram stain Urinary antigen test for pneumococci Number of patients N p Proportion Evaluable samples P Ev Proportion Positive samples P Pos Price difference* P When Pencillin G as targeted therapy compared to Ceftriaxone / erythromycin Amoxicillin / clavulanic acid Average antibiotic costs When Amoxicillin as targeted therapy compared to Ceftriaxone / erythromycin Average antibiotic costs Number of tests performed N tests Costs for diagnostic procedure C Dx Cost reduction* When Penicillin G as targeted therapy compared to Ceftriaxone / erythromycin Amoxicillin / clavulanic acid Average antibiotic costs When Amoxicillin as targeted therapy compared to Ceftriaxone / erythromycin Average antibiotic costs Table 3 Cost calculations Formula used: Cost Reduction = (N p * P Ev * P Pos * P) - (N tests * C Dx ) * in per patient 125

12 Chapter 71 Sensitivity analysis Evidently, when the prevalence of samples positive for pneumocci is high (P pos ), when more sputum samples are evaluable (P Ev ) or when price differences between initial and targeted therapy are greater (ΔP), cost outcome of performing rapid diagnostic tests will be influenced. As is clear from the algorithm, the influence of P pos, P Ev and ΔP are of equal importance on cost-outcome. A sensitivity analysis was performed by calculating the cost outcomes when varying these parameters. The associations between the price differences in targeted and non-targeted therapy, the proportion of evaluable and positive sputum samples and the resulting cost per patient per day are depicted in Figure 1. For example, when targeted therapy with amoxicillin is compared with the most expensive empiric regimen in our setting, 8.1% of the patients needs to have a positive urinary antigen test to reduce costs. Figure 1 Sensitivity analysis for sputum Gram stain Legend: X-axis shows the variation in price difference between broad-spectrum therapy (BST) and targeted therapy (NST). Negative figures mean that NST is more costly than BST. Y-axis shows the cost reduction per patient per day Z-axis shows the proportion of patients with positive test results (Pev*Ppos). The numbers vary from 2.5% %. 126

13 Chapter 7 Discussion We have formulated an algorithm to calculate potential cost-savings when using rapid diagnostic testing to target empirical antimicrobial therapy for CAP. The use of Gram staining and urine antigen tests appeared not to reduce health-care associated costs in our situation. The cost reduction is influenced by price differences between targeted therapy and non-targeted therapy and the proportion of positive test results. The algorithm provides a means to determine potential costsavings in any given setting and can also be applied when new rapid diagnostic tests are evaluated. The lack of cost reduction in our setting is explained by the small amount of patients (19%) which were able to provide a useful sputum sample, the small amount of samples (Gram stain 13%; urinary antigen 27%) being positive for pneumococci and the small price-difference of narrow spectrum and broad spectrum therapy. The cost reduction would increase when more samples were evaluable and positive. Reported percentages of adequate and positive samples have ranged from 24-39%, depending on the time interval between admittance and processing of samples and supervision during collection. (18) In addition, when the cost difference of broad spectrum and targeted therapy increases, cost reduction also increases. In settings where empirical therapy is more expensive, streamlining may have a higher impact on cost reduction. In our hospital, recommended empirical treatment for patients hospitalised with CAP consists of monotherapy with a β-lactam agent. Addition of a macrolide is not recommended, unless pneumonia is severe, needing admission to an intensive care unit, or when a strong suspicion of atypical pneumonia exists. (8) Cost reduction will also increase if broad-spectrum therapy would be associated with extra costs, for example due to adverse events. Several scenarios will result in a lower cost reduction than estimated in our study. The possibility of false positive results ((14), for example when staphylococci, although a rare case for communityacquired pneumonia, are falsely identified as streptococci(19) and the inability of Gram stains and urinary antigen tests to identify atypical (co-)infections could result in inappropiate antimicrobial therapy and lower clinical cure rates. Furthermore, when test results have no impact on treatment decisions, cost savings will never be achieved. 127

14 Chapter 71 The low impact of microbiological investigations on treatment decisions is noted by several authors. (18;20) And evidently, cost reduction will decrease when cost difference of broad spectrum and targeted therapy decreases, for example as a result of once daily dosing regimens of broad-spectrum therapy, instead of regimens containing penicillin G six times daily. This study was designed to investigate the cost-benefit of streamlining initial antibiotic therapy when using rapid diagnostic tests for pneumococcal pneumonia. Because of this perspective, we did not take in account other possible disadvantages of using unnecessary broad spectrum therapy, such as antimicrobial effectiveness or long term effects of antibiotics on antimicrobial resistance. Whether targeted therapy is more effective than broad-spectrum therapy is yet unclear. Recent analyses have suggested that initial therapy with a β-lactam and a macrolide antibiotic increases survival in CAP, even when pneumococci are causative micro-organisms (21;22;23;24;25;26). From this point of view, early recognition of a causative micro-organism would not be beneficial. However, these studies are retrospective and possible subject to prescription bias, showed inconsistencies in reported outcomes and provided no data whether targeted therapy based on the results of microbiological investigations influenced patient outcomes. (27;28) In addition, unnecessary use of broad spectrum antibacterial agents enhances induction of antimicrobial resistance. In theory, financial investment in methods allowing rapid streamlining of antibiotic therapy may outweigh future costs associated with treatment of less suspectible micro-organisms. In conclusion, we showed that using sputum Gram stain or urinary antigen test to streamline initial therapy in patients hospitalised with CAP, would not be associated with cost savings in our setting. However, clinical efficacy of different antibiotics and long-term effects on antimicrobial susceptibility were not included. Moreover, differences in costs of empirical treatment, and the proportion of evaluable and positive tests may lead to different amounts of cost-reduction. Our algorithm is an easy tool to calculate such cost-reduction. 128

15 Chapter 7 Reference List (1) Kuijper EJ, Van Der MJ, De Jong MD, Speelman P, Dankert J. Usefulness of gram stain for diagnosis of lower respiratory tract infection or urinary tract infection and as an aid in guiding treatment. Eur J Clin Microbiol Infect Dis 2003; 22(4): (2) Farr BM, Kaiser DL, Harrison BDW, Connolly CK. Prediction of microbial aetiology at admission to hospitall for pneumonia from the presenting clinical features. Thorax 1989; 44: (3) Macfarlane JT, Miller AC, Roderick Smith WH, Morris AH, Rose DH. Comparative radiographic features of community-acquired legionnaires disease, pneumococcal pneumonia, mycoplasma pneumonia and psittacosis. Thorax 1984; 39: (4) Woodhead MA, Macfarlane JT, American Thoracic Society. Comparative clinical and laboratory features of legionella with pneumococcal and mycoplasma pneumonias. Br J Dis Chest 1987; 81: (5) Ruiz M, Ewig S, Marcos MA, Martinez JA, Arancibia F, Mensa J et al. Etiology of community acquired pneumonia: impact of age, comorbidity and severity. Am J Resp Crit Care Med 1999; 160: (6) Lieberman D, Schlaeffer F, Boldur I, Lieberman D, Horowitz S, Friedman MG et al. Multiple pathogens in adult patients admitted with community-acquired pneumonia: a one year prospective study of 346 consecutive patients. Thorax 1996; 51: (7) Lim WS, Macfarlane JT, Boswell TCJ, Harrison TG, Rose D, Leinonen M et al. Study of community acquired pneumonia aetiology (SCAPA) in adults admitted to hospital: implications for management guidelines. Thorax 2001; 56: (8) Kasteren MEE van, Wijnands WJ, Stobbering EE, Janknegt R, Meer JW van der. Optimization of the antibiotics policy in the Netherlands. II. SWAB guidelines for the antimicrobial therapy of pneumonia in patients at home and as nosocomial infections. The Netherlands Antibiotic Policy Foundation. Ned Tijdschr Geneeskd 1998; 142(17): (9) Reed WW, Byrd GS, Gates RH Jr, et al. Sputum Gram s stain in community acquired pneumonia: A meta analysis. West J Med 1996; 165:197. (10) Smith PR. What diagnostic tests are needed for community-acquired pneumonia. Medical Clinics of North America 2001; 85(6):

16 Chapter 71 (11) American Thoracic Society. Guidelines for the management of adults with community acquired pneumonia. Am J Crit Care Med 2001; 163: (12) Bartlett JG, Dowell SF, Mandell LA, File Jr TM, Musher DM, Fine MJ. Practice Guidelines for the management of community-acquired pneumonia in adults. Infectious Diseases Society of America. Clin Infect Dis 2000; 31(2): (13) Skov Sorensen UB, Henrichsen J. Cross-reactions between pneumococci and other streptococci due to C polysaccaride and F antigen. J Clin Microbiol 1987; 25: (14) Murdoch DR, Laing RT, Mills GD, Karalus NC, Town GI, Mirrett S et al. Evaluation of a rapid immunochromatographic test for detection of Streptococcus pneumoniae antigen in urine samples from adults with community-acquired pneumonia. Journal of Clinical Microbiology 2001; 39(10): (15) Dowell SF, Garman RL, Liu G, Levine OS, Yang YH. Evaluation of Binax NOW, an assay for the detection of pneumococcal antigen in urine samples, performed among pediatric patients. Clin Infect Dis 2001; 32(5): (16) Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify lowrisk patients with community acquired pneumonia. N Engl J Med 1997; 336: (17) Hoepelman IM, Rozenberg-Arska M, Verhoef J. Comparison of once daily ceftriaxone with gentamicin plus cefuroxime for treatment of serious bacterial infections. Lancet 1988; 1(8598): (18) Ewig S, Schlochtermeier M, Goke N, Niederman MS. Applying sputum as a diagnostic tool in pneumonia: limited yield, minimal impact on treatment decisions. Chest 2002; 121(5): (19) Dominguez J, Gali N, Blanco S, Pedroso P, Prat C, Matas L et al. Detection of streptococcus pneumoniae antigen by a rapid immunochromatographic assay in urine samples. Chest 2001; 119(1): (20) Waterer GW, Jennings SG, Wunderink RG. The impact of blood cultures on antibiotic therapy in pneumococcal pneumonia. Chest 1999; 116: (21) Waterer GW, Somes GW, Wunderink RG. Monotherapy may be suboptimal for severe pneumococcal pneumonia. Arch Intern Med 2001; 161: (22) Gleason PP, Meehan TP, Fine JM, Galusha DH, Fine MJ. Associations between initial antimicrobial therapy and medical outcomes for 130

17 Chapter 7 hospitalized elderly patients with pneumonia. Arch Intern Med 1999; 159: (23) Stahl JE, Barza M, DesJardin J, Martin R, Eckman MH. Effect of macrolides as part of initial empiric therapy on length of stay in patients hospitalized with community-acquired pnuemonia. Arch Intern Med 1999; 159: (24) Houck PM, MacLehose RF, Niederman MS, Lowery JK. Empiric antibiotic therapy and mortality among medicare pneumonia inpatients in 10 western states. Chest 2001; 119: (25) Mufson MA, Stanek RJ. Bacteriemic pneumococcal pneumonia in one American city: a 20-year longitudinal study, Am J Med 1999; 107(1A):34S-43S. (26) Dudas V, Hopefl A, Jacobs R, Guglielmo BJ. Antimicrobial selection for hospitalized patients with presumed community-acquired pneumonia: a survey of nonteaching US community hospitals. Ann Pharmacother 2000; 34: (27) Macfarlane J. Severe pneumonia and a second antibiotic. Lancet 2002; 359(9313): (28) Dowell SF. The best treatment for pneumonia. New clues but no definitive answers. Arch Intern Med 1999; 159: