A Cost-Effectiveness Analysis of Bacterial Endocarditis Prophylaxis for Febrile Children Who Have Cardiac Lesions and Undergo Urinary Catheterization in the Emergency Department
* Section of Pediatric Emergency Medicine, Baylor College of Medicine, Houston, Texas
Section of Health Services Research, Department of Biostatistics, University of Texas M.D. Anderson Cancer Center, Houston, Texas
Sections of Pediatric Emergency Medicine and Pediatric Infectious Disease, Baylor College of Medicine, Houston, Texas
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective. To prevent bacterial endocarditis (BE) in those at risk, the American Heart Association recommends antibiotics for patients who have a known urinary tract infection and are about to undergo urinary catheterization (UC). In young children who have cardiac lesions and undergo UC for fever without a source, the problem with prophylaxis only in the presence of infected urine is that the presence of urinary tract infection is unknown before testing. This study was conducted to determine the cost-effectiveness of BE prophylaxis before UC in febrile children aged 0–24 months with moderate-risk cardiac lesions.
Methods. We evaluated the cost-effectiveness of BE prophylaxis compared with no prophylaxis from the societal perspective. Clinical outcomes were based on BE incidence and quality-adjusted life years (QALYs). Probabilities were derived from the medical literature. Costs were derived from national and local sources in US dollars for the reference year 2000, using a discount rate of 3%.
Results. On the basis of the analysis, prophylaxis prevents 7 BE cases per 1 million children treated. When antibiotic-associated deaths were included, the no-prophylaxis strategy was more effective and less costly than the prophylaxis strategy. When antibiotic-associated deaths were excluded, amoxicillin cost $10 million per QALY gained and $70 million per case prevented. For vancomycin, it was $13 million per QALY gained and $95 million per case prevented. The results were robust to variations in the prophylactic efficacy of antibiotics, incidence of bacteremia after UC, incidence of BE after bacteremia, and costs associated with BE prophylaxis and treatment.
Conclusion. In the emergency department, BE prophylaxis before UC in febrile children who are aged 0 to 24 months and have moderate-risk cardiac lesions is not a cost-effective use of health care resources.
Key Words: bacterial endocarditis • antibiotic prophylaxis • fever • urinary tract infections • cost-effectiveness analysis
Abbreviations: BE, bacterial endocarditis • AHA, American Heart Association • UTI, urinary tract infection • UC, urinary catheterization • ED, emergency department • QALE, quality-adjusted life expectancy • CHF, congestive heart failure • QALY, quality-adjusted life year • HRQL, health-related quality of life
Bacterial endocarditis (BE) is a rare and serious disease in children, with an incidence of 0.30/100 000 children per year and a mortality of 11.6%.1–6 Given the severity of the illness, efforts are made to prevent its occurrence.
In 1997, the American Heart Association (AHA) made recommendations for BE prevention on the basis of the available literature as well as group consensus.7 Because the genitourinary tract is second only to the oral cavity as a portal of entry for organisms that cause endocarditis, sterilization of the urinary tract with antimicrobial therapy is recommended by the AHA before instrumentation in the presence of known urinary tract infection (UTI). Sterilization is not recommended before urinary catheterization (UC) in uninfected tissue.
In young children who have cardiac lesions and undergo UC for fever without a source, the problem with prophylaxis only in the presence of infected urine is that the presence of UTI is unknown before testing. The American Academy of Pediatrics recommends the use of UC or suprapubic aspiration to obtain urinary culture specimens from young children who are not toilet trained; bag specimens are not acceptable.8 However, in 1 study, the success rate of obtaining urine with initial suprapubic aspiration was only 46%, compared with 100% success rate for UC.9 For this reason, UC is the procedure of choice for obtaining the urine specimens needed for UTI diagnosis. As UC is an invasive method of specimen collection, antibiotic prophylaxis is often given to such patients (although the frequency of antibiotic prophylaxis could not be ascertained from the literature).
In considering the use of antibiotic prophylaxis in this setting, it is important to note that the organisms that typically cause endocarditis are relatively infrequent causes of febrile UTI in young children. In addition, the use of antibiotic prophylaxis before UC carries health risks and economic costs. It can cause side effects, such as allergic reactions, and can sterilize the urine such that the UTI-causing organism is not identified. It is also likely to increase the time a patient spends in the emergency department (ED), creating significant opportunity costs. Finally, the antibiotics used for prophylaxis increase medical costs.
Given the questionable benefits and the possible health risks and economic costs of antibiotic prophylaxis with UC, the best course of action is unclear for febrile infants with cardiac lesions that confer a moderate risk of BE. The purpose of this study was to determine the relative costs and effectiveness of UC with and without antibiotic prophylaxis for these patients. We undertook a decision analysis to determine the most cost-effective course of action.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Hypothetical Patients
The hypothetical patients for whom the analysis was constructed are aged 0 to 24 months, have moderate-risk cardiac lesions requiring BE prophylaxis, present to the ED with fever, and require urine collection to evaluate the possibility of an underlying UTI. We chose patients with moderate-risk lesions, because there was insufficient information from the medical literature to conduct the analysis on patients with high-risk lesions. According to the AHA guidelines, moderate-risk cardiac lesions include most congenital cardiac malformations (eg, ventricular septal defects), acquired valvular dysfunction (eg, rheumatic heart disease), hypertrophic cardiomyopathy, and mitral valve prolapse with valvular regurgitation and/or thickened leaflets.7
Cost-Effectiveness Analysis
We performed a cost-effectiveness analysis from the societal perspective. The model adheres to the reference case scenario recommended by the Panel on Cost-Effectiveness in Health and Medicine.10 Cost-effectiveness was estimated for 2 outcomes: 1) quality-adjusted life expectancy (QALE) and 2) BE incidence.
Decision Analysis Model
A decision analysis model was developed to represent the choices available to the ED physician when obtaining urine specimens by UC from the hypothetical patient group. A diagram of the decision tree is represented in Fig 1. We considered 2 strategies: UC with and UC without prophylactic antibiotics.
|
Each decision branch includes chance events with their probabilities for occurrence. Those who undergo UC with antibiotics are subject to adverse reactions, which can be mild, severe, or fatal. For both strategies, if urine is collected by UC, then there is a chance that there are bacteria in the urine, that the bacteria are BE pathogens, that the pathogenic bacteria cause bacteremia, and, finally, that endocarditis develops. For estimating QALE, the sequelae of endocarditis are included in the model. That is, if endocarditis develops and the patient survives, then he or she is treated in the hospital and may develop congestive heart failure (CHF) or, alternatively, require surgery for acute decompensation.
Clinical Assumptions
We made the following 6 clinical assumptions in conducting the cost-effectiveness analysis: 1) prophylaxis before UC prevents all BE by preventing bacteremia; 2) amoxicillin and vancomycin are equally effective in preventing bacteremia; 3) in the presence of bacteremia with organisms that cause endocarditis, the incidence of BE is constant, no matter the cause for the bacteremia or the type of moderate-risk cardiac lesion; 4) in the absence of bacteremia or in the presence of organisms not typically associated with endocarditis, BE does not occur; 5) there is no increased risk of bacteremia or BE with contaminated urine specimens; and 6) bacteremia occurs immediately after instrumentation11 and is followed immediately by bacterial seeding of the endocardium.12,13
Probabilities
A MEDLINE search of articles published between 1966 and 2001 was conducted to determine relevant probabilities for use in the decision analysis. Table 1 contains the probabilities and ranges used in the sensitivity analysis. When 1 study was used to determine the probability, 95% confidence intervals were calculated from the available data. When >1 study was used to determine the probability, a weighted mean was calculated and the minimum and maximum from the studies were taken to represent the range of values.
|
We determined the prevalence of UTI in febrile children under the age of 2 years from 3 studies. One of these studies examined infants under the age of 1 year,14 whereas the other 2 examined children under the age of 2 years.15,16 On the basis of these studies, we chose an average prevalence of 3.9% with a range of 3.3% to 5.3%.
Although any bacterium can cause BE, some organisms are more commonly associated with BE than others.1–3,17,18 The prevalence of BE-causing organisms among the UTI-causing organisms was derived from 2 studies of febrile children with UTIs.19,20 On the basis of these studies, the prevalence of BE-causing organisms in UTI (Escherichia coli, Enterococcus species, and Staphylococcus aureus) was calculated to be 3.4%, with a range of 0% to 100% (because any organism can potentially cause BE).
The prophylactic efficacy of antibiotics in preventing bacteremia after genitourinary procedures was determined from 1 clinical trial and 2 decision analyses. The randomized, controlled trial determined the efficacy of mezlocillin for transurethral prostatectomy.21 The 2 decision analyses were published in the early 1980s and assessed the effectiveness of predental antibiotic prophylaxis for patients with mitral valve prolapse.22,23 On the basis of these studies, the efficacy was estimated to be 89%, with a range of 0% to 100%, reflecting the uncertainty of the value.
The incidence of bacteremia after UC for adults with infected urine was estimated from 2 studies to be 23.1% (range: 14.3%–26.3%).24,25 Although there are no studies of the incidence of bacteremia after UC in children, the incidence should be similar in adults and children.
The incidence of endocarditis in children with rheumatic heart disease after bacteremia from tooth extractions has been estimated from studies done before antibiotics were available to be 1.1%, and 2.2% in those who were known to be bacteremic.26 There are no published data describing the incidence of BE in patients with high-risk lesions after bacteremia.
We determined BE outcomes from 4 studies. The average mortality was estimated to be 11.6% with a range of 0.0% to 13.5%.1–3,18 For those who survived, the rate of decompensation requiring surgery was estimated to be 18.6% with a range of 0% to 25%.1–3,18 Only 1 study reported the incidence of CHF attributable to BE, which was 27.1% (95% confidence interval: 14.5–39.7) in those who did not undergo surgical intervention.2 There was no information available about the average duration of survival in children who recover from BE with sequelae (CHF), but the average survival in adults is 6.2 years.23
Mild reactions, including urticaria, occur in 1% (range: 0.7–10.0) of penicillin courses27 but rarely with vancomycin28 or gentamicin.29 Vancomycin is associated with "red man" syndrome, although giving the infusion over 1 hour minimizes the incidence of this side effect. Anaphylactic reactions occur in 0.03% and death occurs in 0.002% of penicillin courses; these are virtually nonexistent with vancomycin and gentamicin.
QALE
QALE was calculated as quality-adjusted life years (QALYs) from the health-related quality of life (HRQL) and life expectancy data and discounted using a rate of 3%. HRQL scores were taken from the Years of Healthy Life Measure.30 For unspecified endocarditis, the HRQL score is 0.58 (0.29–0.84). The average time spent with endocarditis is 6 weeks. It is assumed that if the patient recovers fully, then he or she will return to baseline quality of life. The HRQL score for mitral valve disorder was taken to be representative of moderate-risk lesions, at 0.81. Although there are other cardiac lesions in this group, the decision was made to select mitral valve disorder as representative of this group, with a range reflective of all moderate-risk lesions (0.72–0.92). The HRQL score for CHF is 0.40 (0.17–0.55), and the average survival is 6.2 years. The average life expectancy for children aged 0 to 2 years was 76.5 years in 1999.31
Costs of Intervention
Costs of the intervention (antibiotic prophylaxis) and its adverse effects and the costs averted by giving prophylactic antibiotics are included in the analysis. The estimated costs are for 2000 and are presented in Table 2. Costs for UC, such as the cost of the catheter and the cost of performing the catheterization, were not included in the analysis because the costs are the same for the antibiotic and no-antibiotic groups.
|
According to the 1997 AHA recommendations for the prevention of BE in moderate-risk lesions, the prophylactic antibiotic options for patients depend on whether the patient is allergic to penicillin.7 For those who are not allergic, the recommendation is amoxicillin 50 mg/kg orally 1 hour before the procedure or ampicillin 50 mg/kg intramuscularly or intravenously completed within 30 minutes of starting the procedure. For the purposes of this analysis, only amoxicillin is considered because the use of oral antibiotics is preferable to the use of intravenous antibiotics. For those who are allergic to penicillin, the recommendation is vancomycin 20 mg/kg intravenously over 1 to 2 hours completed within 30 minutes of starting the procedure. Medication doses are based on per-kilogram doses for a 10-kg child, as this weight represents the average for all children between ages 0 and 24 months. Costs were derived from the 2001 Drug Topics Red Book and deflated to 2000,32 using the consumer price index of prescription drugs for urban consumers.33
The cost of antibiotic delivery includes the cost of giving the antibiotic as well as the time incurred while giving the antibiotics. The costs for intravenous placement and drug administration were derived from the hourly wage ($22.27)34 and 26% fringe for nursing care (total $28.06 per hour). The opportunity cost to the parent was taken as the number of hours of work missed while waiting for antibiotic delivery. An average hourly earning of $15.80 was taken from the US Department of Labor Bureau of Labor Statistics for 2000.34 The opportunity costs for time in the ED were derived from the average costs of ED visits in 1993, which was inflated to $300 for 2000.35 The costs of adverse events were derived from the medications and time required to treat the mild or anaphylactic reactions in the ED. The cost of medical care preceding death from anaphylaxis in the ED was assumed to be $2000.
Costs Averted
The costs averted by preventing BE include the costs of treating BE and its potential complications of CHF and cardiac surgery. The monetary cost of death from BE was assumed to be the same as the cost of BE hospitalization. Hospital costs for endocarditis, mitral valve replacement, and CHF were derived from Healthcare Cost and Utilization Project data for 2000 using a cost-to-charge ratio of 0.75.36 The average duration of BE illness in pediatric patients is 4 to 6 weeks, requiring 6 weeks of intravenous antibiotics.37 The average length of stay for mitral valve replacement is an additional 12.2 days. Patients who recover with sequelae (CHF) require an average of 4 outpatient visits and 2 hospital stays per year, with an average of 6.2 years of survival. The outpatient visit costs were derived from Medicaid charges for 2000. Transportation costs are based on a total average travel distance of 20 miles, at $0.31 per mile. Parental opportunity costs were based on a total time (including waiting time and time with the doctor) of 3 hours each visit.
Analysis
The software package DATA by TreeAge was used to construct the decision tree and calculate the relevant outcomes. Outcomes were calculated separately for patients who are penicillin tolerant and penicillin allergic. Strategy-specific outcomes included the development of BE, QALE, and costs of procedures and medications. Incremental cost-effectiveness ratios were calculated for the 2 primary outcomes: QALE and BE incidence. Sensitivity analyses were conducted to examine whether variations in the data would change the conclusions of the model.
Sensitivity analysis was conducted by varying study costs and probabilities. Uncertain probabilities that were varied in the sensitivity analysis include the prevalence of bacteria causing UTI and BE, the prophylactic efficacy of antibiotics in preventing bacteremia, the incidence of bacteremia after UC, and the incidence of BE after bacteremia. All costs were varied from $0 to twice the point estimate for sensitivity analysis. Uncertain costs that were varied included the ED opportunity costs; the parental opportunity costs; and the cost of endocarditis hospitalization, endocarditis surgery, and caring for CHF. QALE was varied for BE, no BE, and CHF. The discount rate was also varied between 0% and 5% for economic costs and clinical outcomes.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cost-Effectiveness Analysis
Tables 3 and 4 present the estimated discounted costs and outcomes for each strategy. When antibiotic-associated deaths were included in the analysis, the no-antibiotic strategy dominated the antibiotic strategy. In that case, antibiotics are both more costly and associated with a lower QALE as a result of antibiotic-associated deaths. When antibiotic-associated deaths were excluded, antibiotic prophylaxis was both more effective and more costly than no-antibiotic prophylaxis, independent of the antibiotic regimen used. The antibiotic strategy would prevent 7 cases of BE per 1 million children treated, with an incremental effectiveness of only 0.00005 QALYs. For patients who do not receive antibiotics, the total cost would be $1.47. For patients who receive amoxicillin, the total cost would be $495.26, with an incremental cost of $493.79. For patients who receive vancomycin, the total cost would be $667.63, with an incremental cost of $666.16. The incremental cost-effectiveness ratio of amoxicillin was $10 million per QALY saved and $70 million per BE case prevented. The incremental cost-effectiveness ratio of vancomycin was $13 million per QALY gained and $95 million per BE case prevented.
|
|
Sensitivity Analysis
Results of the cost-effectiveness analysis were sensitive to the inclusion of antibiotic-associated deaths. When amoxicillin is used for prophylaxis, population-based data indicate that the risk of a fatal reaction is 20 per million treated, whereas the risk of death from endocarditis is 1 per million without antibiotic prophylaxis. With these anaphylactic deaths, the antibiotic strategy was both less effective and more costly than the no-antibiotic strategy. Without any anaphylaxis deaths, the antibiotic strategy was more effective and more costly than the no-antibiotic strategy, at a cost of $10 million per QALY. The threshold probability of anaphylactic death for the effectiveness of antibiotics was 0.0000023. Below this value, the use of antibiotics would be more effective than no antibiotics. Above this value, the use of antibiotics would be less effective than no antibiotics. Given the dominance of the no-antibiotic strategy in the presence of antibiotic-related deaths, the results are reported separately for the strategies in the presence and absence of such deaths. This is prudent, as the probability of antibiotic-related deaths was derived from population data in which the access to emergency intervention would be less rapid than in the ED. Although there are no data available from the literature, it is likely that the incidence of antibiotic-related death approaches 0 when the patient is given antibiotics in the ED.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The results of this study indicate that antibiotic prophylaxis for UC of febrile children who are aged 0 to 2 years and have moderate-risk cardiac lesions is not a cost-effective use of health care resources. This is true for the regimen using amoxicillin and for the regimen using vancomycin. Both regimens have incremental cost-effectiveness ratios in the millions of dollars per QALY gained and per case prevented.
By varying the prevalence of UTIs to 100%, we found that the use of prophylactic antibiotics was not cost-effective even for those with known infected urine (as recommended by the AHA). In this case, the use of antibiotics prevents 178 BE cases per million children treated, at a cost per prevented case of $2.6 million for amoxicillin and $3.6 million for vancomycin. The cost per QALY gained is $311 507 for amoxicillin and $427 966 for vancomycin. Although the effectiveness of the strategy is better than in those with fever alone, the use of antibiotics still is not cost-effective.
As a randomized, clinical trial is not feasible, decision analysis is the best approach to assessing the efficacy of antibiotic prophylaxis for preventing BE. Two decision analyses have examined the efficacy of antibiotic prophylaxis in preventing BE in people who have mitral valve prolapse and undergo dental procedures.22,23 Both studies indicated that antibiotic prophylaxis was not cost-effective. In the analysis by Clemens and Ransohoff,23 the cost per prevented case using oral penicillin was >$2.6 million (1981 dollars). This is consistent with the high cost per prevented case in our study.
One of the weaknesses of the current study is the reliance on data from the literature, much of which pertains to the adult rather than to the pediatric population. However, the outcomes of the analysis did not change when these values were varied in sensitivity analyses. This indicates that although the values may not reflect accurately those of the pediatric population, more accurate values would not have changed the outcome of the cost-effectiveness analysis.
The analysis was also robust to the incidence of BE with bacteremia. This is important because the point estimate for this incidence came from only 1 study, conducted on a population that was different from the one considered for this study. In addition, it is important because it indicates that the findings for this analysis might pertain to the patients who have high-risk cardiac lesions and presumably would have a higher risk for endocarditis precipitation from bacteremia. That is, even if a lesion were susceptible to BE 100% of the time in the presence of bacteremia, the most cost-effective strategy would be no-antibiotic prophylaxis.
Another weakness of the study relates to the heterogeneity of the patient population with moderate-risk lesions. Separate analyses could have been conducted for each moderate-risk lesion, as the quality of life should be different for patients who live with each type of lesion. However, such analyses would make the analysis too complex and are unlikely to yield results different from those found in this analysis. Therefore, the quality of life with mitral valve disorder was taken to be representative of the group. In the sensitivity analysis, this value was varied to represent the other moderate-risk lesions, and the variation had no impact on the outcome of the cost-effectiveness analysis.
The cost of the antibiotic strategy was influenced primarily by the opportunity costs of extra time in the ED for delivering the antibiotics. These opportunity costs were based on the average cost per patient who could not be seen in the ED because of the extra time spent in the ED by the patient receiving antibiotic prophylaxis. Given the costs of emergency care delivery, these costs are high, at $300 per average visit forgone. However, if these opportunity costs are dropped from the analysis, then the antibiotic strategy still is less cost-effective than the no-antibiotic strategy, no matter the antibiotic used. The incremental cost-effectiveness ratio for amoxicillin is approximately $1 million per QALY saved, rather than $10 million when the ED opportunity cost of antibiotic delivery is included. The incremental cost-effectiveness ratio for vancomycin is $1.3 million per QALY saved, rather than $13 million when the ED opportunity cost of antibiotic delivery is included. Therefore, even if the opportunity cost of ED time had been excluded from the analysis, the antibiotic strategy still would not have been cost-effective.
The dollar value assigned to parental opportunity costs for their time was derived from the average hourly wage lost. The use of this measure might not reflect accurately such opportunity costs, as parents might value their leisure time more than their work time. However, the value was varied from 0 to 2 times the average wage without an impact on the outcome of the cost-effectiveness analysis.
These study findings indicate that the use of antibiotic prophylaxis is not cost-effective in the prevention of BE during UC of febrile children who are aged 0 to 24 months and have moderate-risk cardiac lesions. Therefore, it seems that despite the theoretical advantage of prophylactically administering antibiotics to prevent endocarditis from instrumentation during catheterization, there are no data to support this practice.
|
ACKNOWLEDGMENTS |
|---|
We thank Walter Pagel for critique of an earlier version of this manuscript.
|
FOOTNOTES |
|---|
Received for publication Feb 14, 2003; Accepted Aug 4, 2003.
Reprint requests to (A.C.C.) Texas Children’s Hospital; 6621 Fannin St, MC 1-1841; Houston, TX 77030. E-mail:accavine{at}texaschildrenshospital.org
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
- Moy RJ, George RH, de Giovanni JV, Silove ED. Improving survival in bacterial endocarditis.
Arch Dis Child. 1986;61
:394
–399
[Abstract/Free Full Text] - Awadallah SM, Kavey RW, Byrum CJ, et al. The changing pattern of infective endocarditis in childhood. Am J Cardiol. 1991;68 :90 –94[CrossRef][Web of Science][Medline]
- Sholler GF, Hawker RE, Celermajer JM. Infective endocarditis in childhood. Pediatr Cardiol. 1986;6 :183 –186[CrossRef][Web of Science][Medline]
- Hogevik H, Olaison L, Andersson R, et al. Epidemiologic aspects of infective endocarditis in an urban population. Medicine. 1995;74 :324 –339[CrossRef][Medline]
- Van der Meer JTM, Thompson J, Valkenburg HA, Michel MF. Epidemiology of bacterial endocarditis in The Netherlands. I. Patient characteristics.
Arch Intern Med. 1992;152
:1863
–1868
[Abstract/Free Full Text] - Schollin J, Bjarke B, Wesstrom G. Infective endocarditis in Swedish children. I. Incidence, etiology, underlying factors and port of entry of infection. Acta Paediatr Scand. 1986;75 :993 –998[Web of Science][Medline]
- Dajani AS, Taubert KA, Wilson W, et al. Prevention of bacterial endocarditis: recommendations by the American Heart Association.
Circulation. 1997;96
:358
–366
[Abstract/Free Full Text] - AAP Committee on Quality Improvement Subcommittee on Urinary Tract Infection. Practice parameter: the diagnosis, treatment, and evaluation of the initial urinary tract infection in febrile infants and young children.
Pediatrics. 1999;103
:843
–852
[Abstract/Free Full Text] - Pollack CV, Pollack ES, Andrew ME. Suprapubic bladder aspiration versus urethral catheterization in ill infants: success, efficiency, and complication rates. Ann Emerg Med. 1994;23 :225 –230[Web of Science][Medline]
- Muenning P. Designing and Conducting Cost-Effectiveness Analyses in Medicine and Health Care. San Francisco, CA: Jossey-Bass; 2002
- Roberts GJ, Gardner P, Simmons NA. Optimum sampling time for detection of dental bacteraemia in children. Int J Cardiol. 1992;35 :311 –315[CrossRef][Web of Science][Medline]
- Durack DT, Beeson PB. Experimental bacterial endocarditis. II. Survival of bacteria in endocardial vegetations. Br J Exp Pathol. 1975;53 :50 –53
- Durack DT. Experimental bacterial endocarditis. IV. Structure and evolution of very early lesions. J Pathol. 1975;115 :81 –89[CrossRef][Web of Science][Medline]
- Hoberman A, Chao HP, Keller DM, et al. Prevalence of urinary tract infection in febrile infants. J Pediatr. 1993;123 :17 –23[CrossRef][Web of Science][Medline]
- Shaw KN, Gorelick M, McGowan KL, et al. Prevalence of UTI in febrile young children in the emergency department.
Pediatrics. 1998;102
:1
–5
[Abstract/Free Full Text] - Roberts KB, Charney E, Sweren RJ, et al. Urinary tract infection in infants with unexplained fever: a collaborative study. J Pediatr. 1983;103 :864 –867[CrossRef][Web of Science][Medline]
- Fukushige J, Igarashi H, Ueda K. Spectrum of infective endocarditis during infancy and childhood: 20-year review. Pediatr Cardiol. 1994;15 :127 –131[Web of Science][Medline]
- Van Hare GF, Ben-Shachar G, Liebman J, et al. Infective endocarditis in infants and children during the past 10 years: a decade of change. Am Heart J. 1984;107 :1235 –1240[CrossRef][Web of Science][Medline]
- Honkinen O, Jahnukainen T, Mertsola J, et al. Bacteremic urinary tract infection in children. Pediatr Infect Dis J. 2000;19 :630 –634[CrossRef][Web of Science][Medline]
- Hoberman A, Wald ER, Hickey RW, et al. Oral versus initial intravenous therapy for urinary tract infections in young febrile children.
Pediatrics. 1999;104
:79
–86
[Abstract/Free Full Text] - Allan WR, Kumar A. Prophylactic mezlocillin for transurethral prostatectomy. Br J Urol. 1985;57 :46 –49[CrossRef][Web of Science][Medline]
- Bor DH, Himmelstein DU. Endocarditis prophylaxis for patients with mitral valve prolapse: a quantitative analysis. Am J Med. 1984;76 :711 –717[CrossRef][Web of Science][Medline]
- Clemens JD, Ransohoff DF. A quantitative assessment of pre-dental antibiotic prophylaxis for patients with mitral-valve prolapse. J Chron Dis. 1984;37 :531 –544[CrossRef][Web of Science][Medline]
- Slade N. Bacteraemia and septicaemia after urological operations. Proc R Soc Med. 1958;51 :331 –334[Web of Science][Medline]
- Sullivan NM, Sutter VL, Mims MM, et al. Clinical aspects of bacteremia after manipulation of the genitourinary tract. J Infect Dis. 1973;127 :49 –55[Web of Science][Medline]
- Everett ED, Hirschmann JV. Transient bacteremia and endocarditis prophylaxis. A review. Medicine. 1977;56 :61 –77[Medline]
- Idsoe O, Guthe T, Willcox RR, DeWeck A. Nature and extent of penicillin side-reactions, with particular reference to fatalities from anaphylactic shock. Bull World Health Organ. 1968;38 :159 –188[Web of Science][Medline]
- Cunha BA. Vancomycin. Med Clin North Am. 1995;79 :817 –831[Web of Science][Medline]
- Lortholary O, Tod M, Cohen Y, Petitjean O. Aminoglycosides. Med Clin North Am. 1995;79 :761 –787[Web of Science][Medline]
- Gold MR, Franks P, McCoy KI, Fryback DG. Toward consistency in cost-utility analyses: using national measures to create condition-specific values. Med Care. 1998;36 :778 –792[CrossRef][Web of Science][Medline]
- CDC. National Center for Health Statistics. National Vital Statistics Report. Available at: www.cdc.gov
- Medical Economics Staff. Drug Topics Red Book. Montvale, NJ: Medical Economics; 2001
- US Department of Labor. Bureau of Labor Statistics. Bureau of Labor Statistics Data. Available at: www.bls.gov
- US Department of Labor. Bureau of Labor Statistics. National Compensation Survey: Occupational Wages in the United States, 2000. Available at: www.bls.gov
- Williams RM. The costs of visits to emergency departments.
N Engl J Med. 1996;334
:642
–646
[Abstract/Free Full Text] - Agency for Healthcare Research and Quality. Healthcare Cost and Utilization Project. HCUPnet Database Selection. Available at: http://hcup.ahrq.gov/HCUPnet.asp
- Starke J. Infective endocarditis. In: Feigin RD, Cherry JD, eds. Pediatric Infectious Diseases. 4th ed. Philadelphia, PA: WB Saunders Company; 1998:315–338
PEDIATRICS (ISSN 1098-4275). ©2004 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  N. Brooks Prophylactic antibiotic treatment to prevent infective endocarditis: new guidance from the National Institute for Health and Clinical Excellence Heart, May 1, 2009; 95(9): 774 - 780. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
