In reply

Abstract

We appreciate the interest of Bijlsma and colleagues in our report on risk assessment in patients who have meningitis and negative Gram stains, and we applaud their efforts to test the validity of our model in a large prospectively collected database of Danish patients who had bacterial meningitis. Specifically, Bijlsma and associates evaluated our risk score in a cohort of 218 patients who had bacterial meningitis and a negative Gram stain. They discovered that our model had a high sensitivity (98.5%) but a very low specificity (9.2%) for identifying patients with meningitis and a negative Gram stain at risk for an adverse clinical outcome. We derived and validated our prognostic model in a very large and heterogeneous patient population from the Houston, Texas, area that included meningitis caused by viruses, bacteria, fungi, and mycobacteria and in which only 22 of 567 patients (3.7%) had culture-proven bacterial meningitis. In contrast, the analysis by Bijlsma and colleagues included only bacterial meningitis cases, which are associated with higher adverse clinical outcomes. In the United States, the use of conjugate vaccines has reduced the incidence of bacterial meningitis from 5.5 cases per 100,000 persons in 1986 to 1.38 per 100,000 persons in 2007.1McIntyre P.B. O’Brien K.L. Greenwood B. van de Beek D. Effect of vaccines on bacterial meningitis worldwide.Lancet. 2012; 380: 1703-1711Abstract Full Text Full Text PDF PubMed Scopus (216) Google Scholar Thus, although the fraction of bacterial meningitis cases we reported differs considerably from that of Bijlsma and associates, our data accurately reflect the current epidemiology of community-acquired meningitis with a negative Gram stain in US hospitals.2Khoury N.T. Hossain M.M. Wootton S.H. Salazar L. Hasbun R. Meningitis with a negative cerebrospinal fluid gram stain in adults: risk classification for an adverse clinical outcome.Mayo Clinic Proc. 2012; 87: 1181-1188Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar It is reassuring to see that 203 of 218 patients (93%) with bacterial meningitis and a negative Gram stain in the Dutch Meningitis Cohort Study were identified by our high-risk score and that their adverse clinical outcome rate was 30%. The low-risk score in that study was seen in only 15 of the 218 patients (7%), and only 1 of the 15 (7% of the patients with a low-risk score; 0.5% of the 218 patients with bacterial meningitis with a negative Gram stain) had an adverse clinical outcome. The single low-risk patient who had an adverse outcome had nosocomial meningitis and would have been excluded from our community-acquired meningitis study; hence, applying our calculations to this adverse clinical outcome in the analysis by Bijlsma and colleagues yielded a 0% adverse outcome rate. Despite these reassuring findings, we agree with Bijlsma and associates that a risk score specifically derived and validated in patients with bacterial meningitis is preferred in predicting outcomes in this patient population.3Aronin S.I. Peduzzi P. Quagliarello V.J. Community-acquired bacterial meningitis: risk stratification for adverse clinical outcome and effect of antibiotic timing.Ann Intern Med. 1998; 129: 862-869Crossref PubMed Google Scholar, 4Weisfelt M. van de Beek D. Spanjaard L. Reitsma J.B. de Gans J. A risk score for unfavorable outcome in adults with bacterial meningitis.Ann Neurol. 2008; 63: 90-97Crossref PubMed Scopus (57) Google Scholar Prognostic Risk Score for Pleocytosis With a Negative Gram Stain: Valid but of Limited Utility in Bacterial Meningitis PatientsMayo Clinic ProceedingsVol. 88Issue 4PreviewMeningitis with a negative cerebrospinal fluid (CSF) Gram stain has an extensive differential diagnosis including both benign and life-threatening etiologies. Reliable prognostic risk stratification at presentation could aid physicians in management decisions and patient counseling in the emergency department. In the December 2012 issue of Mayo Clinic Proceedings, Khoury et al1 reported on the derivation and validation of a risk score for an adverse clinical outcome in this patient group. Patients were classified as low risk if they had normal neurologic examination findings, had a CSF glucose level of 45 mg/dL or higher, and were 60 years old or younger. Full-Text PDF