Prognostic Factors in Adults with Bacterial Meningitis
http://www.100md.com
《新英格兰医药杂志》
To the Editor: In a study of adults with bacterial meningitis, van de Beek et al. (Oct. 28 issue)1 noted the following: 14 percent were comatose on admission, those with opening pressures above 400 mm of water were more likely to be admitted in a coma (24 percent, vs. 11 percent of those with lower opening pressures), and a low score on the Glasgow Coma Scale was a strong predictor of an unfavorable outcome in the multivariate analysis. The authors state, "In those with moderate or severe impairment of consciousness . . ., lumbar puncture should be preceded by cranial CT [computed tomography]," presumably "to predict the likelihood and avoid the possibility of brain herniation" after lumbar puncture.
Cranial CT is useful for detecting contraindications to lumbar puncture. However, a normal cranial CT scan does not mean that lumbar puncture is safe. Clinical signs of impending herniation, such as coma, may be the best predictors of when to delay lumbar puncture in bacterial meningitis. Brain herniation shortly after lumbar puncture in patients with bacterial meningitis who have normal results on cranial CT has been reported,2,3,4 and a poor correlation of intracranial pressure with the results of cranial CT in bacterial meningitis has also been reported.5,6 Could some unfavorable outcomes have been prevented by delaying lumbar puncture in those patients who were comatose on admission?
Ari R. Joffe, M.D.
Stollery Children's Hospital
Edmonton, AB T6G 2B7, Canada
ajoffe@cha.ab.ca
References
van de Beek D, de Gans J, Spanjaard L, Weisfeldt M, Reitsma JB, Vermeulen M. Clinical features and prognostic factors in adults with bacterial meningitis. N Engl J Med 2004;351:1849-1859.
Shetty AK, Desselle BC, Craver RD, Steele RW. Fatal cerebral herniation after lumbar puncture in a patient with a normal computed tomography scan. Pediatrics 1999;103:1284-1287.
Kastenbauer S, Winkler F, Pfister HW. Cranial CT before lumbar puncture in suspected meningitis. N Engl J Med 2002;346:1248-1251.
Rennick G, Shann F, de Campo J. Cerebral herniation during bacterial meningitis in children. BMJ 1993;306:953-955.
Winkler F, Kastenbauer S, Yousry TA, Maerz U, Pfister HW. Discrepancies between brain CT imaging and severely raised intracranial pressure proven by ventriculostomy in adults with pneumococcal meningitis. J Neurol 2002;249:1292-1297.
Grande PO, Myhre EB, Nordstrom CH, Schliamser S. Treatment of intracranial hypertension and aspects on lumbar dural puncture in severe bacterial meningitis. Acta Anaesthesiol Scand 2002;46:264-270.
To the Editor: Van de Beek et al. identified several independent risk factors for an unfavorable outcome of bacterial meningitis in adults. However, in the multivariate logistic-regression analysis, they did not include the strongest predictor of a poor outcome — namely, having meningitis due to Streptococcus pneumoniae (odds ratio as compared with meningitis due to Neisseria meningitides, 6; 95 percent confidence interval, 2.6 to 13.9; P<0.001) — a fact that limits the interpretation of their results. Probably such prognostic factors as the presence of otitis media or sinusitis, the absence of rash, or a positive blood culture identify patients with pneumococcal meningitis, rather than those likely to have a poor outcome. By contrast, the presence of otitis media was indeed an independent prognostic factor for survival among 189 patients with pneumococcal meningitis in Denmark during a two-year national survey (odds ratio in the multivariate logistic-regression analysis, 6.4; 95 percent confidence interval, 2.1 to 20.1; P=0.001) (unpublished data). Which independent risk factors could be identified with the inclusion of the causative pathogen in the multivariate analysis?
Christian ?stergaard, M.D., Ph.D.
Hvidovre University Hospital
DK-2650 Hvidovre, Denmark
coa@ssi.dk
To the Editor: Van de Beek et al. found that otitis and sinusitis are risk factors for an unfavorable outcome in acute bacterial meningitis. However, the authors provide no details about their definition of otitis and sinusitis. Signs of otitis or sinusitis were found in 48 patients on cranial computed tomography (CT), but the radiologic definition is not given. It is unclear how the diagnosis was made in the other 128 patients. The evidence of cranial CT findings is doubtful, since the paranasal sinuses and temporal bones are often not shown completely. Patients in intensive care units often have opacifications in the paranasal sinuses or in mastoid cells without having clinical disease. For patients with such findings, otolaryngologists are often asked to perform paranasal-sinus surgery or a mastoidectomy. Frequently, the intraoperative findings do not confirm the suspected focus of infection. To shed light on this problem, the authors should give information on how often a swab from the middle ear or the sinuses, or a sample obtained during surgery, revealed the same microorganisms as were isolated from cerebrospinal fluid when otitis or sinusitis was suspected.
Jens Peter Klussmann, M.D.
Orlando Guntinas-Lichius, M.D.
University of Cologne
D-50924 Cologne, Germany
orlando.guntinas@uni-koeln.de
To the Editor: In reading the recent Perspective article by Swartz (Oct. 28 issue),1 I was struck by the fact that the first and perhaps most dramatic break in the mortality curve for bacterial meningitis came not from the introduction of sulfonamides or even penicillin, but from the use of humble, dirty antiserums. In an era of increasing bacterial resistance and flattening of mortality curves for bacterial meningitis — and with our now remarkable ability to produce antibodies — the use of antiserums needs to again be strongly considered and tested as adjuvant therapy for bacterial meningitis.
Eric L. Altschuler, M.D., Ph.D.
Mount Sinai School of Medicine
New York, NY 10029
eric.altschuler@mssm.edu
References
Swartz MN. Bacterial meningitis -- a view of the past 90 years. N Engl J Med 2004;351:1826-1828.
Drs. van de Beek and de Gans reply: The question asked by Joffe reflects an interesting dilemma. Case reports of fatal herniation after lumbar puncture with normal CT results are rare and should be interpreted cautiously. The argument that a normal cranial CT scan does not exclude the possibility of herniation after lumbar puncture should be qualified by the problems in diagnosing brain herniation, the limitations of CT of the posterior fossa, and a caveat about the causal role of lumbar puncture.1 In the first case report cited by Joffe, the CT study was not shown and the time between CT and lumbar puncture was not given.2 In our study, patients with opening pressures of more than 400 mm of water were more likely to be admitted in a coma than were those with lower opening pressures; the percentage of patients with an unfavorable outcome was similar in the two groups. In acute bacterial meningitis, CT should be performed to detect brain shift, not raised cerebrospinal fluid pressure, which is present anyway in most cases.1,3 The relation between high opening pressures and low Glasgow Coma Scale scores in our study is neither an argument nor motivation for delaying lumbar puncture in patients with no signs of brain shift on CT.
?stergaard asks which independent risk factors could be identified with the inclusion of the causative pathogen in the multivariate analysis. In a model including this variable, the predictive effect of advanced age, a heart rate of more than 120 beats per minute, a low score on the Glasgow Coma Scale, a cerebrospinal fluid white-cell count below 1000 per cubic millimeter, a positive blood culture, an elevated erythrocyte sedimentation rate, and a reduced platelet count remained significant.
Klussmann and Guntinas-Lichius ask for information on swabs from the middle ear or the sinuses in patients with meningitis who had otitis, sinusitis, or both. In general, the diagnosis of otitis or sinusitis is made with a degree of uncertainty. In addition, patients with otitis do not routinely undergo middle-ear swabbing.4 In our study, the diagnosis of otitis or sinusitis was made by the treating physician on clinical grounds, not on the basis of CT studies or swabs.
Diederik van de Beek, M.D., Ph.D.
Jan de Gans, M.D., Ph.D.
University of Amsterdam
1100 DD Amsterdam, the Netherlands
d.vandebeek@amc.uva.nl
References
van Crevel H, Hijdra A, de Gans J. Lumbar puncture and the risk of herniation: when should we first perform CT? J Neurol 2002;249:129-137.
Shetty AK, Desselle BC, Craver RD, Steele RW. Fatal cerebral herniation after lumbar puncture in a patient with a normal computed tomography scan. Pediatrics 1999;103:1284-1287.
van de Beek D, de Gans J, Spanjaard L, Weisfeldt M, Reitsma JB, Vermeulen M. Clinical features and prognostic factors in adults with bacterial meningitis. N Engl J Med 2004;351:1849-1859.
Mandell GL, Bennett JE, Dolin R, eds. Mandell, Douglas, and Bennett's principles and practice of infectious diseases. New York: Elsevier/Churchill Livingstone, 2005.
Dr. Swartz replies: Dr. Altschuler raises the interesting possibility of employing specific antiserums as adjunctive therapy for community-acquired bacterial meningitis. He notes that their use early in the 20th century had a salutory effect in this formerly almost uniformly fatal disease. However, the successes involved intrathecal treatment with antiserum to either Haemophilus influenzae1 or N. meningitidis.2 With the former, essentially all cases involved a single serotype, type B, greatly simplifying antiserum selection; with the latter, most cases involved epidemic strains, mainly serogroup A, again simplifying treatment. In contrast, the results of treatment of pneumococcal meningitis were dismal, probably reflecting the multiplicity of serotypes involved and the greater severity of disease.
Immunization against H. influenzae type B has largely eliminated this microorganism as a cause of childhood meningitis in the developed world. Almost all strains of N. meningitidis remain susceptible to penicillin and third-generation cephalosporins. Thus, only S. pneumoniae is a continuing concern, because of the resistance of 35 percent of isolates to penicillin. Unfortunately, the multiplicity of serotypes and the need for rapid serotyping of isolates and prompt intrathecal administration of antiserum would make such adjuvant therapy problematic.
Morton N. Swartz, M.D.
Massachusetts General Hospital
Boston, MA 02114
Cranial CT is useful for detecting contraindications to lumbar puncture. However, a normal cranial CT scan does not mean that lumbar puncture is safe. Clinical signs of impending herniation, such as coma, may be the best predictors of when to delay lumbar puncture in bacterial meningitis. Brain herniation shortly after lumbar puncture in patients with bacterial meningitis who have normal results on cranial CT has been reported,2,3,4 and a poor correlation of intracranial pressure with the results of cranial CT in bacterial meningitis has also been reported.5,6 Could some unfavorable outcomes have been prevented by delaying lumbar puncture in those patients who were comatose on admission?
Ari R. Joffe, M.D.
Stollery Children's Hospital
Edmonton, AB T6G 2B7, Canada
ajoffe@cha.ab.ca
References
van de Beek D, de Gans J, Spanjaard L, Weisfeldt M, Reitsma JB, Vermeulen M. Clinical features and prognostic factors in adults with bacterial meningitis. N Engl J Med 2004;351:1849-1859.
Shetty AK, Desselle BC, Craver RD, Steele RW. Fatal cerebral herniation after lumbar puncture in a patient with a normal computed tomography scan. Pediatrics 1999;103:1284-1287.
Kastenbauer S, Winkler F, Pfister HW. Cranial CT before lumbar puncture in suspected meningitis. N Engl J Med 2002;346:1248-1251.
Rennick G, Shann F, de Campo J. Cerebral herniation during bacterial meningitis in children. BMJ 1993;306:953-955.
Winkler F, Kastenbauer S, Yousry TA, Maerz U, Pfister HW. Discrepancies between brain CT imaging and severely raised intracranial pressure proven by ventriculostomy in adults with pneumococcal meningitis. J Neurol 2002;249:1292-1297.
Grande PO, Myhre EB, Nordstrom CH, Schliamser S. Treatment of intracranial hypertension and aspects on lumbar dural puncture in severe bacterial meningitis. Acta Anaesthesiol Scand 2002;46:264-270.
To the Editor: Van de Beek et al. identified several independent risk factors for an unfavorable outcome of bacterial meningitis in adults. However, in the multivariate logistic-regression analysis, they did not include the strongest predictor of a poor outcome — namely, having meningitis due to Streptococcus pneumoniae (odds ratio as compared with meningitis due to Neisseria meningitides, 6; 95 percent confidence interval, 2.6 to 13.9; P<0.001) — a fact that limits the interpretation of their results. Probably such prognostic factors as the presence of otitis media or sinusitis, the absence of rash, or a positive blood culture identify patients with pneumococcal meningitis, rather than those likely to have a poor outcome. By contrast, the presence of otitis media was indeed an independent prognostic factor for survival among 189 patients with pneumococcal meningitis in Denmark during a two-year national survey (odds ratio in the multivariate logistic-regression analysis, 6.4; 95 percent confidence interval, 2.1 to 20.1; P=0.001) (unpublished data). Which independent risk factors could be identified with the inclusion of the causative pathogen in the multivariate analysis?
Christian ?stergaard, M.D., Ph.D.
Hvidovre University Hospital
DK-2650 Hvidovre, Denmark
coa@ssi.dk
To the Editor: Van de Beek et al. found that otitis and sinusitis are risk factors for an unfavorable outcome in acute bacterial meningitis. However, the authors provide no details about their definition of otitis and sinusitis. Signs of otitis or sinusitis were found in 48 patients on cranial computed tomography (CT), but the radiologic definition is not given. It is unclear how the diagnosis was made in the other 128 patients. The evidence of cranial CT findings is doubtful, since the paranasal sinuses and temporal bones are often not shown completely. Patients in intensive care units often have opacifications in the paranasal sinuses or in mastoid cells without having clinical disease. For patients with such findings, otolaryngologists are often asked to perform paranasal-sinus surgery or a mastoidectomy. Frequently, the intraoperative findings do not confirm the suspected focus of infection. To shed light on this problem, the authors should give information on how often a swab from the middle ear or the sinuses, or a sample obtained during surgery, revealed the same microorganisms as were isolated from cerebrospinal fluid when otitis or sinusitis was suspected.
Jens Peter Klussmann, M.D.
Orlando Guntinas-Lichius, M.D.
University of Cologne
D-50924 Cologne, Germany
orlando.guntinas@uni-koeln.de
To the Editor: In reading the recent Perspective article by Swartz (Oct. 28 issue),1 I was struck by the fact that the first and perhaps most dramatic break in the mortality curve for bacterial meningitis came not from the introduction of sulfonamides or even penicillin, but from the use of humble, dirty antiserums. In an era of increasing bacterial resistance and flattening of mortality curves for bacterial meningitis — and with our now remarkable ability to produce antibodies — the use of antiserums needs to again be strongly considered and tested as adjuvant therapy for bacterial meningitis.
Eric L. Altschuler, M.D., Ph.D.
Mount Sinai School of Medicine
New York, NY 10029
eric.altschuler@mssm.edu
References
Swartz MN. Bacterial meningitis -- a view of the past 90 years. N Engl J Med 2004;351:1826-1828.
Drs. van de Beek and de Gans reply: The question asked by Joffe reflects an interesting dilemma. Case reports of fatal herniation after lumbar puncture with normal CT results are rare and should be interpreted cautiously. The argument that a normal cranial CT scan does not exclude the possibility of herniation after lumbar puncture should be qualified by the problems in diagnosing brain herniation, the limitations of CT of the posterior fossa, and a caveat about the causal role of lumbar puncture.1 In the first case report cited by Joffe, the CT study was not shown and the time between CT and lumbar puncture was not given.2 In our study, patients with opening pressures of more than 400 mm of water were more likely to be admitted in a coma than were those with lower opening pressures; the percentage of patients with an unfavorable outcome was similar in the two groups. In acute bacterial meningitis, CT should be performed to detect brain shift, not raised cerebrospinal fluid pressure, which is present anyway in most cases.1,3 The relation between high opening pressures and low Glasgow Coma Scale scores in our study is neither an argument nor motivation for delaying lumbar puncture in patients with no signs of brain shift on CT.
?stergaard asks which independent risk factors could be identified with the inclusion of the causative pathogen in the multivariate analysis. In a model including this variable, the predictive effect of advanced age, a heart rate of more than 120 beats per minute, a low score on the Glasgow Coma Scale, a cerebrospinal fluid white-cell count below 1000 per cubic millimeter, a positive blood culture, an elevated erythrocyte sedimentation rate, and a reduced platelet count remained significant.
Klussmann and Guntinas-Lichius ask for information on swabs from the middle ear or the sinuses in patients with meningitis who had otitis, sinusitis, or both. In general, the diagnosis of otitis or sinusitis is made with a degree of uncertainty. In addition, patients with otitis do not routinely undergo middle-ear swabbing.4 In our study, the diagnosis of otitis or sinusitis was made by the treating physician on clinical grounds, not on the basis of CT studies or swabs.
Diederik van de Beek, M.D., Ph.D.
Jan de Gans, M.D., Ph.D.
University of Amsterdam
1100 DD Amsterdam, the Netherlands
d.vandebeek@amc.uva.nl
References
van Crevel H, Hijdra A, de Gans J. Lumbar puncture and the risk of herniation: when should we first perform CT? J Neurol 2002;249:129-137.
Shetty AK, Desselle BC, Craver RD, Steele RW. Fatal cerebral herniation after lumbar puncture in a patient with a normal computed tomography scan. Pediatrics 1999;103:1284-1287.
van de Beek D, de Gans J, Spanjaard L, Weisfeldt M, Reitsma JB, Vermeulen M. Clinical features and prognostic factors in adults with bacterial meningitis. N Engl J Med 2004;351:1849-1859.
Mandell GL, Bennett JE, Dolin R, eds. Mandell, Douglas, and Bennett's principles and practice of infectious diseases. New York: Elsevier/Churchill Livingstone, 2005.
Dr. Swartz replies: Dr. Altschuler raises the interesting possibility of employing specific antiserums as adjunctive therapy for community-acquired bacterial meningitis. He notes that their use early in the 20th century had a salutory effect in this formerly almost uniformly fatal disease. However, the successes involved intrathecal treatment with antiserum to either Haemophilus influenzae1 or N. meningitidis.2 With the former, essentially all cases involved a single serotype, type B, greatly simplifying antiserum selection; with the latter, most cases involved epidemic strains, mainly serogroup A, again simplifying treatment. In contrast, the results of treatment of pneumococcal meningitis were dismal, probably reflecting the multiplicity of serotypes involved and the greater severity of disease.
Immunization against H. influenzae type B has largely eliminated this microorganism as a cause of childhood meningitis in the developed world. Almost all strains of N. meningitidis remain susceptible to penicillin and third-generation cephalosporins. Thus, only S. pneumoniae is a continuing concern, because of the resistance of 35 percent of isolates to penicillin. Unfortunately, the multiplicity of serotypes and the need for rapid serotyping of isolates and prompt intrathecal administration of antiserum would make such adjuvant therapy problematic.
Morton N. Swartz, M.D.
Massachusetts General Hospital
Boston, MA 02114