Abstract
Objective
Describe cerebrospinal fluid parameters in infants with culture-proven Group B streptococcal meningitis in the era of intrapartum antibiotic prophylaxis.
Study Design
Cohort study of the first lumbar puncture from 13,495 infants cared for at 150 neonatal intensive care units. We compared cerebrospinal fluid parameters [white blood cell count, red blood cell count, glucose, and protein], demographics, and outcomes between infants with and without Group B streptococcal meningitis.
Results
We identified 46 infants with Group B streptococcal meningitis. The median cerebrospinal fluid white blood cell count was 271 cells/mm3 for infants with Group B streptococcal meningitis and 6 cells/mm3 for infants without meningitis (p=0.0001). Of the infants with Group B streptococcal meningitis, 9/46 (20%) had negative blood cultures. Meningitis complicated 22/145 (15%) of episodes of early onset Group B streptococcal sepsis and 13/23 (57%) of episodes of late onset Group B streptococcal sepsis.
Conclusions
Group B streptococcal meningitis occurs in the presence of negative blood cultures. In hospitalized infants who undergo a lumbar puncture, Group B streptococcal sepsis is frequently complicated by GBS meningitis.
Keywords: Group B streptococcus, intrapartum antibiotic prophylaxis, meningitis
Introduction
Group B streptococcus (GBS) is the most common cause of neonatal meningitis1,2. Mortality of GBS meningitis approaches 30%, and 50% of survivors have significant neurological sequelae including: hydrocephalus, deafness, blindness, and developmental disabilities.3 Although the institution of intrapartum antibiotic prophylaxis (IAP) in the late 1990s has decreased the burden of GBS disease, neonatal exposure to IAP and empirical antibiotics prior to lumbar puncture (LP) often compromises the diagnosis of meningitis based on culture of the cerebrospinal fluid (CSF).4,7 In these situations, clinicians must often rely on the analysis of CSF white blood cell (WBC) count, red blood cell (RBC) count, glucose, and protein for the diagnosis of meningitis.
Because descriptions of CSF parameters in infants with GBS meningitis occurred prior to IAP, the purpose of this study is to compare the CSF parameters, demographics, and outcome of infants with culture proven GBS meningitis to infants with negative CSF cultures in the era of IAP.8
Methods
We examined the results of the first LP from infants discharged from 150 neonatal intensive care units (NICUs) managed by the Pediatrix Medical Group, Inc. from 1997 to 2004. Samples were processed in the local clinical microbiology laboratories according to the local hospital standards. GBS meningitis was defined by a positive CSF culture. We excluded neonates with CSF reservoirs/shunts, and CSF results from infants positive for viral pathogens and bacterial species other than GBS. The Duke University Institutional Review Board provided permission to conduct this investigation. The study used an administrative database for which data were collected in a manner described previously.2
Normal CSF parameters for premature infants (< 37 weeks) were defined as white blood cell (WBC) count < 26 cells/mm3, glucose level > 23 mg/dL and protein < 151 mg/dL.9 Normal CSF parameters for term infants (≥ 37 weeks) were defined as WBC count < 23 cells/mm3, glucose level > 33 mg/dL and protein < 171 mg/dL.9 We also examined the blood culture results in patients with GBS meningitis and the rate of meningitis complicating early and late-onset GBS sepsis. Early-onset infections were defined as positive cultures in the first week of life.
Demographics were compared using Kruskall-Wallis for continuous variables or Fisher's exact tests for categorical variables. CSF parameters were compared using median linear regression.10 Sensitivity, specificity, positive predictive values, negative predictive values, positive likelihood ratios and negative likelihood ratios were calculated for each CSF parameters as defined above for predicting culture proven meningitis. STATA 10 (College Station, TX) and SAS (Cary, NC) were used for statistical analysis. All P values were two-tailed and statistical significance was defined as a P ≤ 0.05.
Results
Of 13,495 infants that underwent at least one LP, 46 (0.3%) infants had positive CSF cultures for GBS. Of the 46 infants with GBS meningitis, 33 (72%) were born at term, and 7 (16%) were < 1000 g birth weight. The proportion of infants with GBS meningitis, GBS bacteremia and culture-negative CSF, and negative blood and CSF cultures exposed to IAP were 7/46 (15%), 34/133 (26%), and 4186 (38%), respectively, p=0.001 (Table 1). In the group of infants with GBS meningitis, 3/26 (12%) of the mothers were GBS positive. In the group of infants with GBS bacteremia but negative CSF, 16/90 (18%) of the mothers were GBS positive. In the group negative for GBS meningitis and bacteremia, 1693/6419 (26%) of the mothers were GBS positive. Calculated p-values found that significant demographic variables were gestational age (p=0.0001), birth weight (p=0.0001), delivery (p=0.02), and IAP use (p=0.001).
Table 1.
Demographics
| GBS meningitis (%) | GBS bacteremia, culture negative CSF (%) | Negative Blood and CSF Culture (%) | |
|---|---|---|---|
| N | 46 | 133 | 13,316 |
| Gestational age (weeks) - median [IQR] | 38 [36, 39] | 38 [35, 40] | 37 [31, 39] |
| Birth weight (g) - median [IQR] | 3122 [2735, 3570] | 3301 [2696, 3625] | 2740 [1825,3425] |
| Delivery - N (%) | |||
| Caesarian-section | 12 (28) | 47 (35) | 5746 (43) |
| Maternal GBS status | |||
| Positive | 3 (12) | 16 (18) | 1938 (25) |
| Negative | 17 (65) | 50 (56) | 3438 (44) |
| Unknown | 6 (23) | 24 (27) | 2381 (31) |
| Intrapartum antibiotic prophylaxis - N (%) | |||
| Yes | 7 (15) | 34 (26) | 5203 (39) |
| Race/ethnicity - N (%) | |||
| Black | 6 (14) | 24 (18) | 2462 (19) |
| Hispanic | 13 (31) | 28 (22) | 3890 (30) |
| White | 23 (55) | 75 (58) | 5923 (46) |
| Other | 0 (0) | 3 (2) | 595 (5) |
| Sex - N (%) | |||
| Male | 20 (43) | 68 (51) | 7498 (56) |
| Death-N(%) | |||
| Yes | 2 (5) | 1 (1) | 218 (2) |
The median WBC count for the infants with GBS meningitis, 271/mm3 [69, 1624], was higher than the infants with GBS bacteremia and culture-negative CSF, 7/mm3 [2, 14], and infants with negative CSF and blood cultures, 6 /mm3 [2, 16], p<0.0001. The median CSF glucose for the infants with GBS meningitis, 13 mg/dL [2, 44], was lower than infants who had GBS bacteremia and culture-negative CSF, 52 mg/dL [43, 58], and the infants who had both negative blood and CSF cultures, 49 mg/dL [42, 58], p<0.0001. The median CSF protein for the infants with GBS meningitis, 322 mg/dL [268, 852], was significantly higher than the infants with GBS bacteremia but negative CSF, 98 mg/dL [78, 127] and the infants with both negative CSF and blood cultures, 114 mg/dL [84, 155], p<0.0001.
Out of the 46 infants with culture proven GBS meningitis, 29 (63%) had early-onset disease and 17 (37%) had late-onset infection. The median CSF WBC count in the early-onset group was 1529 /mm3 [163, 6120] and 102 /mm3 [57, 276] in the late-onset group, p=0.01. The median CSF glucose in the early-onset group was 6 mg/dl [2, 30] and 36 mg/dL [6, 53] in the late-onset group, p=0.12. The median CSF protein in the early-onset group was 404 mg/dl [273, 852] and 291 mg/dL [178, 822] in the late-onset group, p=0.38.
Nine of the 46 infants (20%) with GBS meningitis had negative blood cultures. Meningitis complicated 22/155 (14%) of episodes of early-onset GBS sepsis and 13/24 (54%) of episodes of late-onset GBS sepsis, p<0.01.
An elevated CSF protein was the most sensitive CSF parameter for detecting culture proven GBS meningitis (92.6%) and a low CSF glucose was the most specific (95.7%), Table 2. An infant with all 3 CSF parameters in the abnormal range had a 32 fold increase in the odds of having a CSF culture positive for GBS.
Table 2.
CSF parameter comparison
| Sensitivity | Specificity | PPV | NPV | + LR | − LR | |
|---|---|---|---|---|---|---|
| Elevated WBC | 89.3 | 81.7 | 1.8 | 99.9 | 4.9 | 0.1 |
| Elevated protein | 92.6 | 75.7 | 1.5 | 100.0 | 3.8 | 0.1 |
| Low glucose | 60.7 | 95.7 | 5.4 | 99.8 | 14.0 | 0.4 |
| Any abnormal value | 96.7 | 67.2 | 1.1 | 100.0 | 3.0 | 0.1 |
| All values abnormal | 58.3 | 98.2 | 11.1 | 99.8 | 32.0 | 0.4 |
PPV (positive predictive value), NPV (negative predictive value), +LR (positive likelihood ratio), −LR (negative likelihood ratio)
Discussion
Prior to the introduction of IAP for maternal GBS colonization, the incidence of early-onset sepsis from GBS was approximately 1.5 cases per 1000 live births.5 In 1996 the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists recommended a screening-based or a risk-based approach for identifying women that should receive IAP as a means of preventing neonatal GBS infection.5 These efforts resulted in a 65% decrease in early-onset GBS infection between 1993 and 1998.4 In 2002 a population-based study comparing the screening-based approach to the risk-based approach found that the screening-based approach was 50% more effective at preventing perinatal GBS disease. The findings of this study led to revised CDC recommendations in 2002 that called for universal GBS screening of all pregnant women at 35–37 weeks gestation. The incidence of GBS disease in infants dropped to 0.32 cases per 1000 live births in 2003, the lowest ever recorded in the United States.5,6
Because diagnosis of GBS meningitis may be complicated by prior exposure of the infant to antibiotics, we decided to evaluate the CSF parameters under current care practices including IAP. We observed an elevated CSF WBC count and protein and a decrease in glucose in the CSF of our infants with GBS meningitis compared to infants with negative CSF cultures.9 Infants with GBS bacteremia and meningitis were also less likely to have been exposed to IAP (Table 1). IAP may have decreased the sensitivity of the blood and CSF cultures and increased the likelihood of false negative results in those infants classified as being free of GBS disease.
Although GBS disease occurred among infants exposed to IAP, this should not be construed as failed therapy. The doses and frequency of IAP was not known in this study and may be a limiting factor in our results. Nevertheless, physicians should not automatically assume that the risk of GBS is eliminated when a mother is given IAP or is culture negative for GBS.11
Historically, early-onset GBS disease typically presents as fulminant sepsis in the first week of life, while late onset disease presents as a focal infection (e.g., meningitis) after the first week of life.5 Our study was consistent with previous reports of higher rates of meningitis complicating late-onset GBS sepsis.12
Although a low CSF glucose is associated with the highest positive predictive value (5.4%) of any of the 3 CSF parameters, it has the lowest sensitivity (60.7%). Fifty-eight percent of infants with GBS meningitis had an abnormal CSF WBC count, glucose and protein in contrast to only 18% of premature infants from the same cohort with meningitis caused by any pathogen (Gram-negative, Gram-positive, or Candida).13 Nearly all (96.7%) infants with GBS meningitis have at least 1 abnormal CSF value. Among infants from the same cohort with Candida meningitis only 4/7 (57%) had an abnormal CSF glucose, protein or WBC count.14
Clinicians often defer the LP until after the blood culture becomes positive. This practice has been questioned by several investigators.1,2,16 In this cohort, 20% of infants with GBS meningitis had negative blood cultures. Future work with GBS meningitis should also include prospective studies of survival and neurodevelopmental follow-up to assess the utility of CSF parameters in diagnosing meningitis. This data supports the conclusion that the LP is important in the initial sepsis evaluations of infants provided that the patient has sufficient clinical stability to undergo the procedure.
Glossary
Abbreviations
- GBS
Group B Streptococcus
- CSF
cerebrospinal fluid
- IAP
intrapartum
- LP
lumbar puncture
- WBC
white blood cell
- RBC
red blood cell
- NICU
neonatal intensive care units
- IQR
interquartile range
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Wiswell TE, Baumgart S, Gannon CM, Spitzer AR. No lumbar puncture in the evaluation for early neonatal sepsis: will meningitis be missed? Pediatrics. 1995;95(6):803–6. [PubMed] [Google Scholar]
- 2.Garges HP, Moody MA, Cotten CM, et al. Neonatal meningitis: what is the correlation among cerebrospinal fluid cultures, blood cultures, and cerebrospinal fluid parameters? Pediatrics. 2006;117(4):1094–1100. doi: 10.1542/peds.2005-1132. [DOI] [PubMed] [Google Scholar]
- 3.Leib SL, Kim YS, Chow LL, et al. Reactive oxygen intermediates contribute to necrotic and apoptotic neuronal injury in an infant rat model of bacterial meningitis due to Group B streptococci. J Clin Invest. 1996;98(11):2632–9. doi: 10.1172/JCI119084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Schrag SJ, Zywicki S, Farley MM, et al. Group B streptococcal disease in the era of intrapartum antibiotic prophylaxis. N Engl J Med. 2000;342(1):15–20. doi: 10.1056/NEJM200001063420103. [DOI] [PubMed] [Google Scholar]
- 5.Schrag SJ, Gorwitz R, Fultz-Butts K, Schuchat A. Prevention of perinatal group B streptococcal disease (Revised guidelines from CDC) MMWR Recomm Rep. 2002;51(RR11):1–22. [PubMed] [Google Scholar]
- 6.Brooks S, Apostol M, Nadle J, et al. Diminishing racial disparities in early-onset neonatal group B streptococcal disease-United States, 2000–2003. MMWR Weekly. 2004;53(23):502–505. [PubMed] [Google Scholar]
- 7.Wendel GD, Jr., Leveno KJ, Sanchez PJ, et al. Prevention of neonatal group B streptococcal disease: a combined intrapartum and neonatal protocol. Am J Obstet Gynecol. 2002;186(4):618–26. doi: 10.1067/mob.2002.122970. [DOI] [PubMed] [Google Scholar]
- 8.Sarff LD, Platt LH, McCracken GH. Cerebrospinal fluid evaluation in neonates: comparison of high risk infants with and without meningitis. J Pediatr. 1976;88(3):473–7. doi: 10.1016/s0022-3476(76)80271-5. [DOI] [PubMed] [Google Scholar]
- 9.Custer JW. Blood chemistries and body fluids. In: Custer JW, Rau RE, editors. The Harriet Lane Handbook. Elsevier Mosby; Philadelphia: 2008. pp. 686–688. [Google Scholar]
- 10.Kottas A, Gelfand AE. Bayesian semiparametric median regression modeling. J American Statistical Assn. 2001;96(456):1458–1468. [Google Scholar]
- 11.Puopolo KM, Madoff LM, Eichenwald EC. Early-onset group B streptococcal disease in the era of maternal screening. Pediatrics. 2005;115(5):1240–6. doi: 10.1542/peds.2004-2275. [DOI] [PubMed] [Google Scholar]
- 12.Edwards MS, Nizet V, Baker CJ. Group B streptococcal infections. In: Remington JS, Klein JO, Wilson CB, Baker CJ, editors. Infectious diseases of the fetus and newborn infant. Elsevier Saunders; Philadelphia: 2006. pp. 403–464. [Google Scholar]
- 13.Smith PB, Garges HP, Cotton CM, et al. Meningitis in preterm neonates: importance of cerebrospinal fluid parameters. Am J Perin. 2008;25(7):421–426. doi: 10.1055/s-0028-1083839. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Cohen-Wolkowiez M, Smith PB, Mangum B, et al. Neonatal candida meningitis: significance of cerebrospinal fluid parameters and blood cultures. J of Perin. 2007;27:97–100. doi: 10.1038/sj.jp.7211628. [DOI] [PubMed] [Google Scholar]
- 15.Stoll BJ, Hansen N, Fanaroff AA, et al. To tap or not to tap: high likelihood of meningitis without sepsis among very low birth weight infants. Pediatrics. 2004;113(5):1181–6. doi: 10.1542/peds.113.5.1181. [DOI] [PubMed] [Google Scholar]