|
Correction
to
Durand et al.,
|
| |||||||||||||||||||||||||||||||||
Second, some of the "culture-negative" episodes might represent unrecognized anaerobic infection, especially in the first years of the study, when cerebrospinal fluid cultures were not routinely performed under anaerobic conditions. The reports of Propionibacterium acnes as a pathogen in neurosurgical infections4 contrast with the very low incidence reported by Durand et al. and emphasize the need for prolonged anaerobic culture to establish the diagnosis and for caution in dismissing "diphtheroid" organisms as contaminants. Moreover, the clinical presentation of neurosurgical meningitis due to P. acnes may be insidious, and the level of cerebrospinal fluid pleocytosis may be slight, so that some cases may not fit the criteria of Durand et al. for "culture-negative" bacterial meningitis.
Although the conclusions of their study would not change substantially if the aforementioned points were not considered, we think these points have clinical relevance for many patients.
Esteban Martinez, M.D.
Hospital de la Santa Creu i Sant Pau
08025 Barcelona, Spain
Angeles Marcos, M.D.
Hospital Clinic i Provincial
08036 Barcelona, Spain
References
During an 11-year period (1981 through 1992), there were 80 episodes of community-acquired meningitis in adults at our institution (a 480-bed university hospital). Seventy patients were immediately admitted to our intensive care unit because of organ failure (shock, 63 percent; respiratory failure, 44 percent; Glasgow score <6, 9 percent). The severity of disease on admission, expressed as a simplified acute physiological score, was 16.5 ±71. Multiorgan failure was present in 62 percent of the patients.
The hospital case fatality rate was 39 percent (31 of 80). The main risk factors for death were altered consciousness on admission (75 percent mortality; P<0.01), respiratory failure (P<0.001), an increased simplified acute physiological score (21.1 ±5.5 vs. 12.8 ±5.8; P<0.001), age of 60 years or more (mortality, 72 percent vs. 24 percent; P<0.001), and the extent of organ failure. The pathogens were Streptococcus pneumoniae (27 patients), Neisseria meningitidis (6), staphylococci (6), Listeria monocytogenes (6), group B streptococcus (3), and gram-negative bacilli (7). In 16 episodes, cerebrospinal fluid culture was negative. The prognosis was independent of seizures (34 percent), neurologic signs (30 percent), antibiotic therapy before initial lumbar puncture (29 percent), pathogen, bacteremia (43 episodes), and the ratio of cerebrospinal fluid glucose to blood glucose (mean, 0.43 ±0.3).
The prognosis was very poor despite early admission to the intensive care unit, identification of the bacteria in most cases (positive blood cultures in 61 percent), adequate antibiotic treatment, and aggressive therapy. The high mortality rate may have been due to several other factors, including severe progressive illness before admission2 and previously impaired health (impairment was present in 54 percent three months before admission).
Francois G. Brivet, M.D.
Michele Guibert, M.D.
Jean Dormont, M.D.
Hopital Antoine Beclere
92140 Clamart, France
References
To the Editor: We agree with Martinez and Marcos that it may be difficult to distinguish culture-negative bacterial meningitis from the posterior cranial fossa syndrome in patients who have recently had surgery of the posterior fossa and rapid tapering of postoperative steroids. However, in our study, only 2 of the 72 episodes of culture-negative meningitis occurred in patients who had recently had such surgery. In one of these patients, cerebrospinal fluid otorrhea developed postoperatively and then meningitis (cerebrospinal fluid white-cell count, 4550 per cubic millimeter, with 100 percent neutrophils). In the other patient, meningitis developed two weeks after surgery despite a relatively slow tapering of steroids (cerebrospinal fluid white-cell count, 1900 per cubic millimeter, with 98 percent neutrophils). Both patients were treated with parenteral antibiotic therapy for 10 days. Studies1 of postneurosurgical meningitis that compare only culture-positive cases of bacterial meningitis with presumably "aseptic" cases may inadvertently include episodes of culture-negative bacterial meningitis among aseptic episodes, unless the latter category specifically excludes patients treated with a full course of antibiotic therapy.
We also agree that P. acnes may be underrepresented as a cause of meningitis in patients with a break in the dural barrier, and that this organism usually causes subacute meningitis with only slight pleocytosis2. The 23 patients in our series who had culture-negative nosocomial meningitis and a break in the dural barrier (due to recent head trauma or a neurosurgical or otolaryngologic procedure) had acute presentations, and most had marked pleocytosis. The 19 who did not have positive cerebrospinal fluid on Gram's staining or bacterial meningitis at autopsy had a mean cerebrospinal fluid white-cell count of 3007 per cubic millimeter, with 89 percent neutrophils.
Brivet et al. found, as we did, that an obtunded mental state and an age of 60 years or more at admission were risk factors for mortality among patients with single episodes of community-acquired bacterial meningitis. They did not find seizures to be a risk factor, although we would like to emphasize that seizures with an early onset ( 
24 hours), but not those with a late onset, were a risk factor in our study. It would be interesting to know whether the higher in-hospital mortality rate in their study (39 percent vs. 25 percent in our study; P<0.05) reflected an older patient population than ours or perhaps a higher percentage of patients who presented with obtunded mental status.
We also note that in Table 5 of our article, the sixth variable should have been "Positive Gram's stain," not "Gram-positive," as printed.
Marlene L. Durand, M.D.
Stephen B. Calderwood, M.D.
Morton N. Swartz, M.D.
Massachusetts General Hospital
Boston, MA 02114
References
| |||||||||||||||||||||||||||||||||
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
