Cerebrospinal fluid lactate level as a diagnostic biomarker for bacterial meningitis in children
© Mekitarian Filho et al.; licensee Springer. 2014
Received: 15 October 2013
Accepted: 3 February 2014
Published: 27 February 2014
Cerebrospinal fluid (CSF) lactate is a potential biomarker for bacterial meningitis in children. To this end, we performed a single-center retrospective cohort study of children from Sao Paulo, Brazil, with CSF pleocytosis to evaluate the ability of CSF lactate to distinguish between children with bacterial and aseptic meningitis. We determined the optimum cutoff point for CSF lactate using receiver-operator curve (ROC) analysis.
We identified 451 children of whom 40 (9%) had bacterial meningitis. Children with bacterial meningitis had a higher median CSF lactate level [9.6 mmol/l, interquartile range (IQR) 3.2-38.5 mmol/l bacterial meningitis vs. 2.0 mmol/l, IQR 1.2-2.8 mmol/l aseptic meningitis]. A CSF lactate cutoff point of 3.0 mmol/l had a sensitivity of 95% [95% confidence interval (CI) 83-99%), specificity of 94% (95% CI 90-96%) and negative predictive value of 99.3% (95% CI 97.7-99.9%) for bacterial meningitis.
In combination with a validated meningitis clinical prediction rule, the CSF lactate level can be used to distinguish between bacterial and aseptic meningitis in children with CSF pleocytosis.
KeywordsBacterial meningitis Aseptic meningitis Cerebrospinal fluid Lactate
While bacterial meningitis causes significant morbidity and mortality despite advances in antibiotic therapy, aseptic meningitis is typically a benign condition requiring only supportive care . Rapid differentiation between bacterial and aseptic meningitis allows early initiation of appropriate therapy for children at risk for having bacterial meningitis without overtreating low-risk children. Although available conjugate vaccinations against Streptococcus pneumoniae and Neisseria meningitidis have decreased the bacterial meningitis incidence, incomplete vaccine uptake as well as infections caused by bacterial serogroups not included in the vaccine makes bacterial meningitis a clinical concern in children, especially in resource-poor settings .
The gold standard for the diagnosis for bacterial meningitis is culture, which requires several days to return results [3, 4]. Rapidly available biomarkers such as cerebrospinal (CSF) lactate have been studied to distinguish bacterial from aseptic meningitis before the results of bacterial cultures become available. CSF lactate levels are high in children with bacterial meningitis as lactate is produced by both bacterial anaerobic metabolism as well as ischemic brain tissue . The predictive ability of CSF lactate for bacterial meningitis has been examined in two recently published meta-analyses [6, 7]. In our large retrospective cohort of children with meningitis, we sought to establish the optimal CSF lactate cutoff point to accurately distinguish between bacterial and aseptic meningitis .
Materials and methods
To this end, we performed a retrospective cohort study of children with meningitis between 1 month and 15 years of age who presented to a single emergency department (ED) of the University Hospital over the 12-year period (2001–2012). The study was approved by the institutional review board. Details of the study protocol have been described previously . We included children who had clinical findings consistent with meningitis (e.g., fever, headache, vomiting or nausea, and neck stiffness) along with CSF white blood cells (WBC) ≥10 cells/μl [corrected for the presence of CSF red blood cells (RBC) using a standard 1:500 ratio of leukocytes to erythrocytes] and who had both a CSF lactate and CSF culture obtained. We further limited our study population to those children who had CSF lactate obtained by the treating clinician. We excluded children with any of the following: critical illnesses, any purpura, presence of a ventricular shunt or recent neurosurgery, immunosuppression and other bacterial infections requiring parenteral antibiotics. We also excluded children who had received any antibiotic pretreatment within 72 h of diagnostic lumbar puncture as pretreatment can render cultures falsely negative.
We reviewed the medical records for all eligible children and abstracted relevant clinical and laboratory factors for all study patients. The CSF lactate level was measured by the hospital clinical laboratory using standard enzymatic methods (ADIVA Chemistry, Siemens, Bayswater Victoria, Australia).
We defined a case of bacterial meningitis as a child with either a positive CSF culture or CSF pleocytosis with a positive blood culture for a bacterial pathogen . We defined a case of aseptic meningitis as a child with meningitis who had not received antibiotics prior to diagnostic lumbar puncture who had a negative CSF bacterial culture .
For statistical analysis, we first selected a range of cutoff points for the CSF lactate biomarker to distinguish bacterial from aseptic meningitis. We reported the sensitivity, specificity and negative predictive value (NPV) for bacterial meningitis of each CSF lactate cutoff point. Next, we generated a receiver-operator curve (ROC) to visually represent the trade-off between sensitivity and specificity. We utilized ROC curve analysis to select the optimal CSF lactate cutoff point to minimize both the number of false positives (children with aseptic meningitis with CSF lactate above the chosen cutoff point) as well as false negatives (children with bacterial meningitis with CSF lactate below the cutoff point). A Mann–Whitney U-test was performed to compare lactate levels between children with bacterial and aseptic meningitis. We used PASW Statistics (version 21.0, Chicago, IL, 2012) for all statistical analyses.
We identified 494 children with meningitis who met the study criteria, of which 43 (9%) did not have a CSF lactate analysis performed. Of the 451 study patients, the mean age was 4.9 years [median 4.0 years;interquartile range (IQR) 0.8-7.3 years], and 306 (68%) were male. Of the 40 children with bacterial meningitis (9% of study patients), the two most common bacterial pathogens were Neisseria meningitidis (55% of bacterial meningitis cases) and Streptococcus pneumoniae (33% of cases). Of the 22 children with Neisseria meningitidis meningitis, 8 had a positive CSF culture alone, 9 had positive CSF and blood cultures, and 5 had a positive blood culture with CSF pleocytosis. Of the 13 children with Streptococcus pneumoniae meningitis, 2 had a positive CSF culture alone, 3 had positive CSF and blood cultures, and 8 had a positive blood culture with CSF pleocytosis. The remaining five children with bacterial meningitis had a positive CSF culture.
CSF lactate concentration in children with bacterial and aseptic meningitis
(n = 40)
(n = 411)
Mean CSF lactate, mmol/l
Median CSF lactate, mmol/l
The diagnostic accuracy of CSF lactate for bacterial meningitis by cutoff point
CSF lactate (mmol/l)
n = 451
Negative predictive value (NPV)
n (%) (95% CI)
n(%) (95% CI)
n% (95% CI)
Two children with bacterial meningitis had a CSF lactate value < 3.0 mmol, but both had a positive CSF gram stain as well as peripheral and CSF leukocytosis. One of these children had Staphylococcus aureus meningitis and the other Streptococcus pyogenes meningitis. Both children were hospitalized and received broad-spectrum antibiotics, with good clinical outcomes.
In our large retrospective cohort of children with meningitis, CSF lactate ≥3.0 mmol/l accurately distinguished between bacterial and aseptic meningitis. However, two children (5% of children with bacterial meningitis) had CSF lactate below the cutoff point. Therefore, CSF lactate should not be used alone but rather combined with a validated clinical prediction model such as the Bacterial Meningitis Score [8–10]. Importantly, this prediction model appropriately identified both children with bacterial meningitis and CSF lactate <3.0 mmol/l.
Two recent meta-analyses evaluated the ability of CSF lactate to distinguish aseptic from bacterial meningitis [6, 7]. These two studies included 38 unique studies with between 15 and 201 patients per study. Using a CSF lactate cutoff point between 2.1 and 4.4 mmol/l, the aggregated sensitivity of CSF lactate for bacterial meningitis was 93% or 96% and specificity 94% or 96%, respectively. Our findings were similar, but our study was larger than any of the included studies. Additionally, our study was the only pediatric study conducted in an ED setting where initial management decisions for children with meningitis must occur.
The management of children with meningitis may differ in resource-poor settings asthe incidence of bacterial meningitis may be considerably higher and clinicians may have limited access to diagnostic laboratories . Available hand-held testing platforms can provide point-of-care testing for CSF lactate. In a small study conducted in Uganda, CSF lactate analysis using a hand-held device had good sensitivity and specificity for bacterial meningitis . Larger, likely multi-centered, studies are needed to further evaluate the diagnostic accuracy of CSF lactate in resource-poor settings.
Our study has several limitations. First our study was retrospective, and we could not assess general appearance of the children. Children who appear unwell at presentation require inpatient therapy regardless of the risk of bacterial meningitis. Second, not all children with aseptic meningitis had a specific infectious etiology identified. However, we reliably excluded bacterial meningitis in these children as all had a CSF bacterial culture with no antibiotic pretreatment within 72 h of diagnostic LP. Third, not all children had CSF lactate obtained and were not included in our study. In fact, the proportion of children who had CSF lactate increased over the study time period. However, the clinical and laboratory presentations of children who did and did not have CSF lactate obtained were similar (data not shown). Last, we had a relatively small number of children with bacterial meningitis, which limited our ability to measure the diagnostic accuracy of CSF lactate. We were also unable to compare CSF lactate by bacterial meningitis pathogen.
The findings of this study suggest that CSF lactate is a reliable biomarker to distinguish between bacterial meningitis from aseptic meningitis, albeit with imperfect sensitivity. In combination with other validated clinical prediction models, CSF lactate can assist with clinical decision-making for children with meningitis.
Written informed consent was obtained from the patient’s guardian/parent/next of kin for the publication of this report and any accompanying images.
- Thigpen MC, Whitney CG, Messonnier NE, Zell ER, Lynfield R, Hadler JL, Harrison LH, Farley MM, Reingold A, Bennett NM, Craig AS, Schaffner W, Thomas A, Lewis MM, Scallan E, Schuchat A, Emerging Infections Programs Network: Bacterial meningitis in the United States, 1998–2007. N Engl J Med 2011,364(21):2016–2025. 10.1056/NEJMoa1005384View ArticlePubMedGoogle Scholar
- Souza SF, Costa Mda C, Paim JS, Natividade MS, Pereira SM, Andrade AM, Teixeira MG: Bacterial meningitis and living conditions. Rev Soc Bras Med Trop Jun 2012,45(3):323–328. 10.1590/S0037-86822012000300009View ArticleGoogle Scholar
- Jansen GJ, Mooibroek M, Idema J, Harmsen HJ, Welling GW, Degener JE: Rapid identification of bacteria in blood cultures by using fluorescently labeled oligonucleotide probes. J Clin Microbiol 2000,38(2):814–817.PubMed CentralPubMedGoogle Scholar
- Poppert S, Essig A, Stoehr B, Steingruber A, Wirths B, Juretschko S, Reischl U, Wellinghausen N: Rapid diagnosis of bacterial meningitis by real-time PCR and fluorescence in situ hybridization. J Clin Microbiol 2005,43(7):3390–3397. 10.1128/JCM.43.7.3390-3397.2005PubMed CentralView ArticlePubMedGoogle Scholar
- Guerra-Romero L, Tauber MG, Fournier MA, Tureen JH: Lactate and glucose concentrations in brain interstitial fluid, cerebrospinal fluid, and serum during experimental pneumococcal meningitis. J infect dis Sep 1992,166(3):546–550. 10.1093/infdis/166.3.546View ArticleGoogle Scholar
- Huy NT, Thao NT, Diep DT, Kikuchi M, Zamora J, Hirayama K: Cerebrospinal fluid lactate concentration to distinguish bacterial from aseptic meningitis: a systemic review and meta-analysis. Crit Care 2010,14(6):R240. 10.1186/cc9395PubMed CentralView ArticlePubMedGoogle Scholar
- Sakushima K, Hayashino Y, Kawaguchi T, Jackson JL, Fukuhara S: Diagnostic accuracy of cerebrospinal fluid lactate for differentiating bacterial meningitis from aseptic meningitis: a meta-analysis. J Infect Apr 2011,62(4):255–262. 10.1016/j.jinf.2011.02.010View ArticleGoogle Scholar
- Mekitarian Filho E, Horita SM, Gilio AE, Alves AC, Nigrovic LE: The bacterial meningitis score to distinguish bacterial from aseptic meningitis in children from Sao Paulo, Brazil. Pediatr Infect Dis J 2013,32(9):1026–1029. 10.1097/INF.0b013e3182913e84View ArticlePubMedGoogle Scholar
- Nigrovic LE, Kuppermann N, Macias CG, Cannavino CR, Moro-Sutherland DM, Schremmer RD, Schwab SH, Agrawal D, Mansour KM, Bennett JE, Katsogridakis YL, Mohseni MM, Bulloch B, Steele DW, Kaplan RL, Herman MI, Bandyopadhyay S, Dayan P, Truong UT, Wang VJ, Bonsu BK, Chapman JL, Kanegaye JT, Malley R, Pediatric Emergency Medicine Collaborative Research Committee of the American Academy of Pediatrics: Clinical prediction rule for identifying children with cerebrospinal fluid pleocytosis at very low risk of bacterial meningitis. JAMA 2007,297(1):52–60. 10.1001/jama.297.1.52View ArticlePubMedGoogle Scholar
- Nigrovic LE, Malley R, Kuppermann N: Meta-analysis of bacterial meningitis score validation studies. Arch dis child 2012,97(9):799–805. 10.1136/archdischild-2012-301798View ArticlePubMedGoogle Scholar
- Scarborough M, Thwaites GE: The diagnosis and management of acute bacterial meningitis in resource-poor settings. Lancet Neurol 2008,7(7):637–648. 10.1016/S1474-4422(08)70139-XView ArticlePubMedGoogle Scholar
- Majwala A, Burke R, Patterson W, Pinkerton R, Muzoora C, Wilson LA, Moore CC: Handheld point-of-care cerebrospinal fluid lactate testing predicts bacterial meningitis in Uganda. J Trop Med Hyg 2013,88(1):127–131. 10.4269/ajtmh.2012.12-0447View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.