Effect of Antibiotic Pretreatment on Cerebrospinal Fluid Profiles of Children With Bacterial Meningitis Lise E.

Nigrovic, Richard Malley, Charles G. Macias, John T. Kanegaye, Donna M. Moro-Sutherland, Robert D. Schremmer, Sandra H. Schwab, Dewesh Agrawal, Karim M. Mansour, Jonathan E. Bennett, Yiannis L. Katsogridakis, Michael M. Mohseni, Blake Bulloch, Dale W. Steele, Ron L. Kaplan, Martin I. Herman, Subhankar Bandyopadhyay, Peter Dayan, Uyen T. Truong, Vince J. Wang, Bema K. Bonsu, Jennifer L. Chapman and Nathan Kuppermann Pediatrics 2008;122;726 DOI: 10.1542/peds.2007-3275

The online version of this article, along with updated information and services, is located on the World Wide Web at:
http://pediatrics.aappublications.org/content/122/4/726.full.html

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly publication, it has been published continuously since 1948. PEDIATRICS is owned, published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois, 60007. Copyright 2008 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on January 4, 2012

ARTICLE

Effect of Antibiotic Pretreatment on Cerebrospinal Fluid Proles of Children With Bacterial Meningitis
Lise E. Nigrovic, MD, MPHa, Richard Malley, MDa, Charles G. Macias, MD, MPHb, John T. Kanegaye, MDc, Donna M. Moro-Sutherland, MDd,e, Robert D. Schremmer, MDf, Sandra H. Schwab, MDg, Dewesh Agrawal, MDh, Karim M. Mansour, MDi, Jonathan E. Bennett, MDj,k, Yiannis L. Katsogridakis, MD, MPHl, Michael M. Mohseni, MDm,n, Blake Bulloch, MDo, Dale W. Steele, MDp, Ron L. Kaplan, MDq, Martin I. Herman, MDr, Subhankar Bandyopadhyay, MDs,t, Peter Dayan, MD, MScu, Uyen T. Truong, MDv, Vince J. Wang, MDw, Bema K. Bonsu, MDx, Jennifer L. Chapman, MDx, Nathan Kuppermann, MD, MPHv, for the American Academy of Pediatrics, Pediatric Emergency Medicine Collaborative Research Committee Childrens Hospital Boston and Harvard Medical School, Boston, Massachusetts; bTexas Childrens Hospital and Baylor College School of Medicine, Houston, Texas; cRady Childrens Hospital San Diego and University of California, San Diego, School of Medicine, San Diego, California; dWakeMed Health and Hospitals, Raleigh, North Carolina; eUniversity of North Carolina at Chapel Hill, Chapel Hill, North Carolina; fChildrens Mercy Hospital and University of Missouri-Kansas City, Kansas City, Missouri; gChildrens Hospital of Philadelphia and University of Pennsylvania, Philadelphia, Pennsylvania; hChildrens National Medical Center and George Washington University School of Medicine, Washington, DC; iOakland Childrens Hospital and Research Center and University of California, Oakland, School of Medicine, Oakland, California; jA. I. duPont Hospital for Children, Wilmington, Delaware; kJefferson Medical College, Philadelphia, Pennsylvania; lChildrens Memorial Hospital and Northwestern Feinberg School of Medicine, Chicago, Illinois; mChildrens Medical Center and Medical College of Georgia, Augusta, Georgia; nMayo Clinic Jacksonville and Mayo School of Graduate Medical Studies, Jacksonville, Florida; oPhoenix Childrens Hospital and University of Arizona College of Medicine, Phoenix, Arizona; pHasbro Childrens Hospital and Warren Alpert Medical School of Brown University, Providence, Rhode Island; qSeattle Childrens and Regional Medical Center and University of Washington School of Medicine, Seattle, Washington; rChildrens Hospital Memphis and University of Tennessee Graduate School of Medicine, Memphis, Tennessee; sChildrens Hospital of Wisconsin and Medical College of Wisconsin, Milwaukee, Wisconsin; tChildrens Healthcare of Atlanta at Scottish Rite, Atlanta, Georgia; uMorgan Stanley Childrens Hospital of New York-Presbyterian and Columbia University College of Physicians and Surgeons, New York, New York; vUniversity of California, Davis, Medical Center and University of California, Davis, School of Medicine, Davis, California; wChildrens Hospital Los Angeles and Keck School of Medicine, University of Southern California, Los Angeles, California; xColumbus Childrens Hospital and Ohio State University, Columbus, Ohio
The authors have indicated they have no nancial relationships relevant to this article to disclose.
a

Whats Known on This Subject

Administration of antibiotics to children before diagnostic lumbar punctures increases the number of false-negative CSF bacterial cultures. The effect of antibiotic pretreatment on CSF cellular and chemical proles has been studied but remains a matter of controversy.

What This Study Adds

We demonstrate that antibiotic pretreatment of children with bacterial meningitis affects CSF proles, specically resulting in increases in CSF glucose levels and decreases in CSF protein levels, which become apparent within several hours of pretreatment.

ABSTRACT OBJECTIVE. The goal of this study was to evaluate the effect of antibiotic administration before lumbar puncture on cerebrospinal uid proles in children with bacterial meningitis.
METHODS. We reviewed the medical records of all children (1 month to 18 years of age)

www.pediatrics.org/cgi/doi/10.1542/ peds.2007-3275 doi:10.1542/peds.2007-3275

Key Words bacterial meningitis, antibiotic pretreatment, cerebrospinal uid, children Abbreviations CSF cerebrospinal uid ANCabsolute neutrophil count IQRinterquartile range WBCwhite blood cell
Accepted for publication Jan 7, 2008

with bacterial meningitis who presented to 20 pediatric emergency departments between 2001 and 2004. Bacterial meningitis was dened by positive cerebrospinal uid culture results for a bacterial pathogen or cerebrospinal uid pleocytosis with positive blood culture and/or cerebrospinal uid latex agglutination results. Probable bacterial meningitis was dened as positive cerebrospinal uid Gram stain results with negative results of bacterial cultures of blood and cerebrospinal uid. Antibiotic pretreatment was dened as any antibiotic administered within 72 hours before the lumbar puncture.
RESULTS. We identied 231 patients with bacterial meningitis and another 14 with prob-

Address correspondence to Lise E. Nigrovic, able bacterial meningitis. Of those 245 patients, 85 (35%) had received antibiotic MD, MPH, Division of Emergency Medicine, pretreatment. After adjustment for patient age, duration and severity of illness at preChildrens Hospital Boston, 300 Longwood sentation, and bacterial pathogen, longer duration of antibiotic pretreatment was not Ave, Boston, MA 02115. E-mail: lise.nigrovic@ childrens.harvard.edu signicantly associated with cerebrospinal uid white blood cell count, cerebrospinal PEDIATRICS (ISSN Numbers: Print, 0031-4005; uid absolute neutrophil count. However, antibiotic pretreatment was signicantly Online, 1098-4275). Copyright 2008 by the associated with higher cerebrospinal uid glucose and lower cerebrospinal uid protein American Academy of Pediatrics levels. Although these effects became apparent earlier, patients with 12 hours of pretreatment, compared with patients who either were not pretreated or were pretreated for 12 hours, had signicantly higher median cerebrospinal uid glucose levels (48 mg/dL vs 29 mg/dL) and lower median cerebrospinal uid protein levels (121 vs 178 mg/dL).

CONCLUSIONS. In patients with bacterial meningitis, antibiotic pretreatment is associated with higher cerebrospinal uid

glucose levels and lower cerebrospinal uid protein levels, although pretreatment does not modify cerebrospinal uid white blood cell count or absolute neutrophil count results. Pediatrics 2008;122:726730

726

NIGROVIC et al

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on January 4, 2012

DMINISTRATION OF ANTIBIOTICS to children before the performance of diagnostic lumbar punctures increases the number of false-negative cerebrospinal uid (CSF) bacterial culture results.1,2 The effects of antibiotic pretreatment on CSF cellular and chemical proles were studied previously but remain a matter of controversy. Some previous work suggested that antibiotic pretreatment does not affect1,3,4 or has only a small inuence on5,6 CSF white blood cell (WBC) proles in patients with bacterial meningitis. One previous study demonstrated a decrease in CSF protein levels after pretreatment with antibiotics.5 Those previous studies, however, were performed before the introduction of conjugate bacterial vaccines against Haemophilus inuenzae type b and Streptococcus pneumoniae, which have transformed the epidemiological features of bacterial meningitis.7,8 Clinical prediction rules for the identication of bacterial meningitis combine clinical and laboratory parameters to estimate the risk of bacterial meningitis. Our group developed9 and then validated10 in a multicenter study the Bacterial Meningitis Score, a clinical prediction rule that accurately identies children with CSF pleocytosis who are at very low risk of bacterial meningitis. In those studies, we excluded all patients who had received any antibiotic pretreatment, because accurately assigning a diagnosis to pretreated patients with negative bacterial culture results is often difcult. Therefore, the accuracy of the bacterial meningitis score in children with pretreated bacterial meningitis is untested. To explore further the relationship between antibiotic pretreatment and CSF parameters in the era of widespread conjugate vaccination against both H inuenzae type b and S pneumoniae, we studied a large cohort of children with bacterial meningitis evaluated in the emergency departments of 20 pediatric centers in the United States.

METHODS We conducted a retrospective cohort study endorsed by the Pediatric Emergency Medicine Collaborative Research Committee of the American Academy of Pediatrics. Twenty emergency departments in pediatric centers located in diverse geographic areas of the United States participated in the study. Details of the study design were described previously.10 In brief, we identied cases of bacterial meningitis in 2 ways, that is, (1) International Classication of Diseases discharge diagnosis codes revision 9.0 for bacterial meningitis (bacterial meningitis, codes 320.0 320.9, or unspecied meningitis, codes 321.0 322.9) or Neisseria infection (code 036.0) and (2) review of microbiology logs for CSF cultures testing positive for bacterial pathogens. We included all children 29 days and 18 years of age who presented to the emergency departments of the participating centers between January 2001 and July 2004. We reviewed the medical records of all potential study patients to determine clinical features, including the occurrence and timing of antibiotic pretreatment, and laboratory results. We dened a case of bacterial meningitis as a patient presenting with a CSF culture testing positive for a bacterial pathogen known to cause bacterial meningitis or

CSF pleocytosis (CSF WBC count of 10 cells per L) with positive blood culture results and/or positive CSF latex agglutination study results. Consistent with clinical practice, we categorized patients with positive CSF Gram-stain results but negative results of bacterial cultures of blood and CSF as having probable bacterial meningitis. We included the patients with probable bacterial meningitis in the bacterial meningitis group in the analysis. We excluded patients with CSF shunts, recent neurosurgery (within 1 month), or suppressed immune systems (either primary or acquired immunodeciency or immunosuppressive therapy), because the etiologic agents and the inammatory response may be different from those in a previously healthy population. We categorized a patient as having pretreated bacterial meningitis if a patient with bacterial meningitis or probable bacterial meningitis was given an antibiotic within 72 hours before the performance of the lumbar puncture. We dened the duration of antibiotic pretreatment as the time between administration of the antibiotic and performance of the lumbar puncture. We compared proportions by using 2 analysis. We compared group medians by using the Mann-Whitney U test (2 groups) or the Kruskal-Wallis test (3 groups). We used multivariate linear regression to determine the relationship between duration of antibiotic pretreatment and each of the following CSF parameters: CSF WBC count, CSF absolute neutrophil count (ANC), CSF protein level, and CSF glucose level. In each case, we adjusted for the following 4 clinically important covariates in the multivariate analyses: patient age, duration of fever, hypotension requiring vasoactive medication, and respiratory failure requiring intubation. We selected the covariates based on our a priori belief that these parameters have associations with patients inammatory responses to bacterial meningitis, as reected in CSF ndings. First, bacterial pathogens are known to differ according to the age of the patient.7 Second, the duration of fever or illness may affect both the CSF inammatory response and the likelihood of being pretreated with antibiotics.11 Finally, previous work showed that patients with more severe infections such as meningococcal disease have reduced CSF neutrophil responses.12 Patients with missing information regarding either the main predictor (duration of antibiotic pretreatment) or any of the covariates were not included in the multivariate analyses. We then performed a threshold analysis for antibiotic effect by using the median for each of the 4 CSF parameters at 3 clinically sensible cutoff points (4, 12, and 24 hours of antibiotic pretreatment). We used SPSS 15.0 (SPSS, Chicago, IL) for all analyses. RESULTS We identied a total of 255 patients with bacterial meningitis. We excluded 10 patients because of the presence of a CSF shunt or recent neurosurgery. Of the remaining 245 patients with bacterial meningitis, 14 (5.7%) were categorized as having had probable bacterial meningitis on the basis of positive Gram-stain results alone (and are included in the bacterial meningitis group in subsequent references and analyses). Of the 245 study patients, 159
PEDIATRICS Volume 122, Number 4, October 2008 727

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on January 4, 2012

200

Duration of pretreatment, h

150

FIGURE 1 Duration of antibiotic pretreatment according to route of antibiotic administration (oral, parenteral, or both) for children with pretreated bacterial meningitis. The horizontal lines represent the median duration of antibiotic pretreatment for each route of administration.

100

50

0 Oral antibiotics Parenteral antibiotics Oral and parenteral antibiotics

Pretreatment type

(65%) had nonpretreated bacterial meningitis, 85 (35%) had pretreated bacterial meningitis, and 1 (0.4%) lacked data about antibiotic pretreatment. Of the 14 patients with probable bacterial meningitis, 9 had been pretreated with antibiotics before lumbar puncture. The children with pretreated bacterial meningitis were slightly older than those with nonpretreated bacterial meningitis, although this difference did not reach statistical signicance (age of patients with pretreated bacterial meningitis: median: 0.8 years; interquartile range [IQR]: 0.3 4.1 years; age of patients with nonpretreated bacterial meningitis: median: 0.4 years; IQR: 0.23.8 years; P .09). Of the 85 patients who received antibiotic pretreatment, 20 (24%) received orally administered antibiotics alone, 59 (69%) parenterally administered antibiotics alone, and 6 (7%) parenterally and orally administered antibiotics. The duration of antibiotic pretreatment varied according to the route of administration (orally administered alone: median: 24 hours; IQR: 24 54 hours; parenterally administered alone: median: 4 hours; IQR: 124 hours; both orally and parenterally administered: median: 24 hours; IQR: 10.5 60 hours; P .001) (Fig 1). Of the 25 patients who were pretreated for 4 hours before lumbar puncture, 24 (96%) were given parenterally administered ceftriaxone. Fourteen patients had missing information about the duration of antibiotic pretreatment. We rst examined the effect of antibiotic pretreatment on CSF Gram-stain results and bacterial culture positivity among the 231 patients with denite bacterial meningitis (Table 1). Although the rates of positive CSF Gram-stain results did not differ according to pretreatment status, those who had received any antibiotic pretreatment before blood or CSF specimens were obtained for culture had lower rates of corresponding positive culture results than did patients who had not received any antibiotic treatment (Table 1). Because blood and CSF culture specimens were not always obtained simultaneously, the pretreatment status sometimes differed according to culture source. Of the 85 patients who received antibiotics before lumbar puncture, 46 (54%) had blood culture specimens obtained before the admin728 NIGROVIC et al

istration of antibiotics. For 16 (35%) of the 46 patients, the bacterial meningitis pathogen was identied on the basis of blood culture results alone. We then examined the effect of antibiotic pretreatment on CSF proles with adjustment for clinically important covariates (patient age, duration of fever, hypotension requiring vasoactive medications, and respiratory failure requiring intubation). The duration of antibiotic pretreatment was not associated with either CSF WBC count ( 0.15; P .20) or CSF ANC ( 0.16; P .21). However, we identied an association between duration of antibiotic pretreatment and both CSF glucose level ( 0.19; P .04) and CSF protein level ( 0.31; P .01). The results of the analyses were similar when we excluded the 14 patients with probable bacterial meningitis dened on the basis of positive Gram-stain results alone (data not shown). We further explored the relationship between the duration of antibiotic pretreatment and the observed effects on CSF proles by using clinically sensible time cutoff points (Table 2). At 4 hours of pretreatment, the glucose level was higher (P .045) and the protein level was lower (P .24), compared with patients who received no pretreatment, although the differences in CSF protein levels did not achieve statistical signicance.

TABLE 1 Frequency of Positive Gram-Stain and Blood and CSF Bacterial Culture Results for Patients With Denite Bacterial Meningitis, According to Antibiotic Pretreatment Status
n/N (%) No Antibiotic Positive CSF Gram-stain resultsa Positive blood culture resultsb Positive CSF culture resultsc 95/150 (63) 123/187 (66) 136/154 (88) Any Antibiotic 46/74 (62) 16/33 (48) 53/76 (70) .86 .05 .001 P

a Excluding 7 patients (6 with no Gram-stain results and 1 with missing data on lumbar puncture pretreatment status). b Excluding 11 patients (5 with no blood culture results and 6 with missing data on blood culture pretreatment status). c Excluding 1 patient (with missing data on lumbar puncture pretreatment status).

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on January 4, 2012

TABLE 2 Threshold Analysis of Differences in CSF Parameters According to Duration of Antibiotic Pretreatment
CSF Parameter CSF WBC count, median (IQR), cells per L CSF ANC, median (IQR), cells per L CSF glucose level, median (IQR), mg/dL CSF protein level, median (IQR), mg/dL Positive CSF bacterial culture results, n (%) No Pretreatment (n 159) 1020 (1903480) 787 (1272780) 29 (1053) 174 (84278) 146 (84) Pretreatment for 4 h (n 25) 547 (1461695) 458 (621606) 31 (1153) 175 (107302) 18 (72) Pretreatment for 4 h (n 47) 1275 (2165594) 968 (964613) 45 (1759) 145 (73247) 26 (55) Pretreatment for 12 h (n 33) 1081 (2154029) 770 (763625) 49 (4263) 121 (73206) 19 (58) Pretreatment for 24 h (n 29) 1012 (2152666) 627 (761744) 50 (3268) 115 (70195) 17 (59)

Fourteen patients had missing information on the duration of antibiotic pretreatment.

However, when we compared patients who had received no or 12 hours of pretreatment with patients who had been pretreated for 12 hours, the median CSF glucose level was signicantly higher (P .005) in pretreated patients and the median CSF protein level was signicantly lower (P .008). No signicant differences in CSF WBC count or CSF ANC results were noted at any of the time points. The effects of antibiotic pretreatment on the CSF parameters tested were very similar for a threshold of 24 hours of pretreatment and for a threshold of 12 hours of pretreatment (differences in CSF glucose levels: P .005; differences in CSF protein levels: P .009). When we adjusted for patient age, duration of fever, hypotension, and respiratory failure in a multivariate analysis, CSF glucose and protein levels both became signicantly different after 12 hours of antibiotic pretreatment. DISCUSSION This study examined the effect of antibiotic pretreatment on CSF proles in the era of widespread use of bacterial conjugate vaccines against H inuenzae type b and S pneumoniae. In our cohort of 245 children with bacterial meningitis, more than one third received antibiotics before the initial lumbar puncture was performed. We found that antibiotic pretreatment did not signicantly affect CSF WBC count or CSF ANC. However, antibiotic pretreatment was associated with higher CSF glucose levels and lower CSF protein levels and a lower rate of CSF culture positivity. Although these effects on CSF parameters became apparent after 4 hours of pretreatment, glucose and protein levels were signicantly higher and lower, respectively, after 12 hours of pretreatment, after adjustment for other factors that might affect CSF proles. Numerous previous investigations investigated the relationship between antibiotic pretreatment and CSF proles.16 No consistent effect of antibiotic therapy was suggested by all of those studies, which might be attributable in part to the differences in denitions and patient populations among the studies. Our study differed from those studies in several important ways. First, all previously published studies were performed before the introduction of universal H inuenzae type b immunization, whereas our study reects the epidemiological features of bacterial meningitis in the setting of nearly universal H inuenzae type b and conjugate pneumococcal vaccination (in the United States). One previous group dened bacterial meningitis by using CSF param-

eters (elevated CSF WBC count, low CSF glucose level, and elevated CSF protein level) without requiring positive blood or CSF culture results,1 potentially obscuring any impact of pretreatment on those parameters. Because of this concern, we conducted our analyses with only patients who had microbiologic conrmation of bacterial meningitis or positive CSF Gram-stain results. Furthermore, by seeking a threshold duration of antibiotic pretreatment at which changes became apparent, we were able to detect differences in CSF protein levels (as shown by others5) and glucose proles that could not be appreciated by simply comparing patients who received pretreatment and those who did not receive pretreatment. Finally, we applied multivariate statistical methods to determine the independent contribution of antibiotic pretreatment after adjustment for other clinically important variables. Our study has some important limitations. First, our study was retrospective and is subject to the limitations of retrospective data gathering. However, we had minimal missing data and included only objective parameters that were likely to be recorded accurately in the medical records. Second, the route of antibiotic pretreatment was strongly related to the duration of pretreatment, because all except 1 patient who had been pretreated for 4 hours had received parenterally administered ceftriaxone, rather than orally administered antibiotics. The study had insufcient power, however, to separate statistically the effects of oral versus parenteral antibiotic pretreatment, beyond the duration of pretreatment. We did not detect any signicant changes in WBC counts or ANCs with increasing duration of antibiotic treatment. Although the study might have been underpowered to detect true differences in CSF WBC counts or CSF ANCs according to pretreatment duration, demonstration of signicant differences in these CSF parameters would only reinforce our conclusion not to apply bacterial meningitis prediction rules to pretreated patients. The most important potential limitation, however, is that our study could not evaluate the effect of antibiotic pretreatment on the CSF proles of children with negative CSF Gram-stain results and negative blood and CSF culture results, who might have had bacterial meningitis but whose cultures yielded negative results because of antibiotic pretreatment. Because accurate assignment of pretreated patients with negative culture and Gram-stain results into bacterial or aseptic meningitis groups was not feasible, such patients were excluded from this analysis. Some of those patients might in fact have had
PEDIATRICS Volume 122, Number 4, October 2008 729

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on January 4, 2012

bacterial meningitis. It is conceivable, but undeterminable, that the effect of antibiotic pretreatment on the CSF proles of those patients is different from the effect on the proles of patients who retain microbiologic evidence of bacterial meningitis. There are important implications of this study. We have demonstrated that antibiotic pretreatment of children with bacterial meningitis affects CSF proles, specically resulting in increases in CSF glucose levels and decreases in CSF protein levels that become apparent within several hours of pretreatment, in addition to the decreased rate of CSF culture positivity. Therefore, the CSF proles of children who have received antibiotic pretreatment before lumbar puncture should be interpreted with caution. In particular, predictive models that discriminate bacterial from aseptic meningitis often include CSF parameters such as glucose and/or protein levels,9,10,1316 including the Bacterial Meningitis Score developed by our group.9,10 Our results suggest that predictive models that rely on these variables should not be applied to patients who have received antibiotics before lumbar puncture. ACKNOWLEDGMENTS This work was supported by an Ambulatory Pediatric Association Young Investigator Grant and a National Research Service Award (grant T32 HD40128-01, Research Training in Pediatric Emergency Medicine, to Dr Nigrovic). We thank other members of the study group, including those who helped with chart abstraction, data entry, and database management, as follows: Elizabeth R. Alpern, MD MSCE (Childrens Hospital of Philadelphia and University of Pennsylvania), Troy Bush (Texas Childrens Hospital and Baylor College of Medicine), Joseph M. Campos, PhD (Childrens National Medical Center), Christopher R. Cannavino, MD (Rady Childrens Hospital San Diego and University of California, San Diego, School of Medicine), Murray Edelberg, PhD (Carlisle, MA), Kim Fisher, PhD (Center for Pediatric Research, University of Tennessee), Marissa Hauptman, MPH (New York University School of Medicine), Paul Ishimine, MD (Rady Childrens Hospital San Diego and University of California, San Diego, School of Medicine), Daniel M. Kaplan (Childrens National Medical Center), John Leake, MD, and R. Ian McCaslin, MD, MPH (Rady Childrens Hospital San Diego and University of California, San Diego, School of Medicine), Umair Salim (Center for Pediatric Research, University of Tennessee), James Wilde, MD (Childrens Medical Center Augusta and Medical College of Georgia), and Xiaohi Zhao (Brookline, MA).

We thank the members of the Pediatric Emergency Medicine Collaborative Research Committee for their critical review and endorsement of the study protocol. REFERENCES
1. Dalton HP, Allison MJ. Modication of laboratory results by partial treatment of bacterial meningitis. Am J Clin Pathol. 1968; 49(3):410 413 2. Kanegaye JT, Soliemanzadeh P, Bradley JS. Lumbar puncture in pediatric bacterial meningitis: dening the time interval for recovery of cerebrospinal uid pathogens after parenteral antibiotic pretreatment. Pediatrics. 2001;108(5):1169 1174 3. Winkelstein JA. The inuence of partial treatment with penicillin on the diagnosis of bacterial meningitis. J Pediatr. 1970; 77(4):619 624 4. Davis SD, Hill HR, Feigl P, Arnstein EJ. Partial antibiotic therapy in Haemophilus inuenzae meningitis: its effect on cerebrospinal uid abnormalities. Am J Dis Child. 1975;129(7): 802 807 5. Kaplan SL, Smith EO, Wills C, Feigin RD. Association between preadmission oral antibiotic therapy and cerebrospinal uid ndings and sequelae caused by Haemophilus inuenzae type b meningitis. Pediatr Infect Dis. 1986;5(6):626 632 6. Converse GM, Gwaltney JM Jr, Strassburg DA, Hendley JO. Alteration of cerebrospinal uid ndings by partial treatment of bacterial meningitis. J Pediatr. 1973;83(2):220 225 7. Schuchat A, Robinson K, Wenger JD, et al. Bacterial meningitis in the United States in 1995: Active Surveillance Team. N Engl J Med. 1997;337(14):970 976 8. Whitney CG, Farley MM, Hadler J, et al. Decline in invasive pneumococcal disease after the introduction of proteinpolysaccharide conjugate vaccine. N Engl J Med. 2003;348(18): 17371746 9. Nigrovic LE, Kuppermann N, Malley R. Development and validation of a multivariable predictive model to distinguish bacterial from aseptic meningitis in children in the postHaemophilus inuenzae era. Pediatrics. 2002;110(4):712719 10. Nigrovic LE, Kuppermann N, Macias C, et al. A clinical prediction rule for identifying children with cerebrospinal uid pleocytosis at very low risk of bacterial meningitis. JAMA. 2007; 297(1):52 60 11. Bonsu BK, Harper MB. Fever interval before diagnosis, prior antibiotic treatment, and clinical outcome for young children with bacterial meningitis. Clin Infect Dis. 2001;32(4):566 572 12. Malley R, Inkelis SH, Coelho P, Huskins WC, Kuppermann N. Cerebrospinal uid pleocytosis and prognosis in invasive meningococcal disease in children. Pediatr Infect Dis J. 1998;17(10): 855 859 13. Spanos A, Harrell FE Jr, Durack DT. Differential diagnosis of acute meningitis: an analysis of the predictive value of initial observations. JAMA. 1989;262(19):2700 2707 14. Bonsu BK, Harper MB. Differentiating acute bacterial meningitis from acute viral meningitis among children with cerebrospinal uid pleocytosis: a multivariable regression model. Pediatr Infect Dis J. 2004;23(6):511517 15. Bonsu BK, Harper MB. Explanation of mathematical model. Pediatr Infect Dis J. 2004;23(9):893 16. Oostenbrink R, Moll HA, Moons KG, Grobbee DE. Predictive model for childhood meningitis. Pediatr Infect Dis J. 2004; 23(11):1070 1071

730

NIGROVIC et al

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on January 4, 2012

Effect of Antibiotic Pretreatment on Cerebrospinal Fluid Profiles of Children With Bacterial Meningitis Lise E. Nigrovic, Richard Malley, Charles G. Macias, John T. Kanegaye, Donna M. Moro-Sutherland, Robert D. Schremmer, Sandra H. Schwab, Dewesh Agrawal, Karim M. Mansour, Jonathan E. Bennett, Yiannis L. Katsogridakis, Michael M. Mohseni, Blake Bulloch, Dale W. Steele, Ron L. Kaplan, Martin I. Herman, Subhankar Bandyopadhyay, Peter Dayan, Uyen T. Truong, Vince J. Wang, Bema K. Bonsu, Jennifer L. Chapman and Nathan Kuppermann Pediatrics 2008;122;726 DOI: 10.1542/peds.2007-3275
Updated Information & Services References including high resolution figures, can be found at: http://pediatrics.aappublications.org/content/122/4/726.full.ht ml This article cites 16 articles, 6 of which can be accessed free at: http://pediatrics.aappublications.org/content/122/4/726.full.ht ml#ref-list-1 This article has been cited by 8 HighWire-hosted articles: http://pediatrics.aappublications.org/content/122/4/726.full.ht ml#related-urls This article, along with others on similar topics, appears in the following collection(s): Infectious Disease & Immunity http://pediatrics.aappublications.org/cgi/collection/infectious_ disease Information about reproducing this article in parts (figures, tables) or in its entirety can be found online at: http://pediatrics.aappublications.org/site/misc/Permissions.xht ml Information about ordering reprints can be found online: http://pediatrics.aappublications.org/site/misc/reprints.xhtml

Citations

Subspecialty Collections

Permissions & Licensing

Reprints

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly publication, it has been published continuously since 1948. PEDIATRICS is owned, published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois, 60007. Copyright 2008 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from pediatrics.aappublications.org at Indonesia:AAP Sponsored on January 4, 2012