Correspondence
Correction
Case 1-1999: Acute Hemorrhagic Leukoencephalitis
N Engl J Med 1999; 341:1314-1316October 21, 1999
- Article
To the Editor:
The evaluation of a 53-year-old man with fever and rapid neurologic deterioration in the January 14 Case Record1 is a good example of the use of laboratory tests for the diagnosis of bacterial meningitis that have little or no clinical utility. A lumbar puncture was performed, tests for bacterial antigens (Haemophilus influenzae, Streptococcus pneumoniae, and Neisseria meningitidis) and syphilis were performed on cerebrospinal fluid, and cold agglutinins were tested in serum.
Although their use was in favor during the 1980s, bacterial-antigen tests have been shown to be insensitive and nonspecific and to have little, if any, effect on the care of patients with suspected bacterial meningitis.2,3 Gram's staining in such patients is highly sensitive and much more specific.4 The Venereal Disease Research Laboratory test performed on cerebrospinal fluid, although highly specific, is insensitive and should only be performed in patients with serologic (i.e., serum) evidence of syphilis.5 Finally, a cold-agglutinin test, although simple to perform, is insensitive and nonspecific for mycoplasma infection.6
6 ReferencesIrving Nachamkin, Dr.P.H.
University of Pennsylvania, Philadelphia, PA 19104-42831
Case Records of the Massachusetts General Hospital (Case 1-1999)
Full Text | Web of Science | Medline2
Kiska DL ,Jones MC ,Mangum ME ,Orkiszewski D ,Gilligan PH . Quality assurance study of bacterial antigen testing of cerebrospinal fluid. J Clin Microbiol 1995;33:1141-1144
Web of Science | Medline3
Perkins MD ,Mirrett S ,Reller LB . Rapid bacterial antigen detection is not clinically useful. J Clin Microbiol 1995;33:1486-1491
Web of Science | Medline4
Dunbar SA ,Eason RA ,Musher DM ,Clarridge JE III . Microscopic examination and broth culture of cerebrospinal fluid in diagnosis of meningitis. J Clin Microbiol 1998;36:1617-1620
Web of Science | Medline5
Dans PE ,Cafferty L ,Otter SE ,Johnson RJ . Inappropriate use of the cerebrospinal fluid Venereal Disease Research Laboratory (VDRL) test to exclude neurosyphilis. Ann Intern Med 1986;104:86-89
Web of Science | Medline6
Mycoplasmas and ureaplasmas. In: Koneman EW, Allen SD, Janda WM, Schreckenberger PC, Winn WC Jr, eds. Color atlas and textbook of diagnostic microbiology. 5th ed. Philadelphia: Lippincott, 1997:857-92.
To the Editor:
In the January 14 Case Record, I was surprised by the repeated use of lumbar puncture in a comatose, febrile patient with pharmacologically induced paralysis in whom a cranial computed tomographic (CT) scan obtained before the initial lumbar puncture showed massive, diffuse edema of the brain. After the second lumbar puncture, the patient was reported to have bilateral decerebrate posturing, and a repeated cranial CT scan showed transtentorial herniation, the eventual cause of death. Even though the eventual outcome of death from postinfectious encephalitis may have been inevitable, it may also have been unnecessarily hastened by the lumbar punctures.1-4 A lumbar puncture should not be performed in this clinical setting because of the risk of brain herniation caused by a rapid lowering of spinal-canal pressure. When the ventricles are easily visible (as in this patient), a neurosurgeon can obtain cerebrospinal fluid safely and rapidly by means of ventriculostomy. A ventriculostomy may provide continuous ventricular drainage of cerebrospinal fluid and is useful in treating elevated intracranial pressure because it rapidly decreases intracranial volume and provides a reliable means of pressure monitoring. It may also provide an efficacious route for the intrathecal delivery of medication.
4 ReferencesBernhard Zünkeler, M.D.
St. Joseph's Medical Center, Baltimore, MD 212041
Evans RW . Complications of lumbar puncture. Neurol Clin 1998;16:83-105
CrossRef | Web of Science | Medline2
Addy DP . When not to do a lumbar puncture. Arch Dis Child 1987;62:873-875[Erratum, Arch Dis Child 1987;62:1293.]
CrossRef | Web of Science | Medline3
Richards PG ,Towu-Aghantse E . Dangers of lumbar puncture. BMJ 1986;292:605-606
CrossRef | Web of Science | Medline4
Gower DJ ,Baker AL ,Bell WO ,Ball MR . Contraindications to lumbar puncture as defined by computed cranial tomography. J Neurol Neurosurg Psychiatry 1987;50:1071-1074
CrossRef | Web of Science | Medline
To the Editor:
In his discussion of Case 1-1999, Dr. Vartanian stated that the pathogenesis of acute hemorrhagic leukoencephalitis is an immune-mediated attack on the oligodendrocyte–myelin unit. We believe that current evidence implicates endothelial cells in this disorder and that perivenous demyelination is a secondary event.1 Indeed, the histopathological findings in the case under discussion beautifully illustrate the cardinal feature of acute hemorrhagic leukoencephalitis: necrotized endothelial cells.
In our view, cerebral endothelial cells are the main target in primary demyelinating diseases1 and highlight the role of the blood–brain barrier in the pathogenesis of these disorders.
1 ReferencesAbhijit Chaudhuri, D.M., M.D.
Peter O. Behan, D.Sc., M.D.
Southern General Hospital, Glasgow G51 4TF, United Kingdom1
Behan PO ,Geschwind N ,Lamarche JB ,Lisak RP ,Kies MW . Delayed hypersensitivity to encephalitogenic protein in disseminated encephalomyelitis. Lancet 1968;2:1009-1012
CrossRef | Web of Science | Medline
Dr. Zünkeler's letter was referred to Dr. Anne Young, the patient's attending physician, who offers the following response:
To the Editor: Zünkeler's comments on lumbar puncture are certainly well taken. Unfortunately, the clinical synopsis was inaccurate. Figure 1, which was cited on page 127, should have been cited on page 128 in the discussion of the third hospital day. The initial CT scan (which was not shown) showed bifrontoparietal edema, with obliterated sulci and diffuse hypodensity of white matter underlying the same regions, intact sulci in the occipital lobe, mild effacement of the third and lateral ventricles, a normal fourth ventricle, and no obliteration of the basilar cisterns. There was also no midline shift. Under these conditions and considering the critical importance of making a prompt and proper diagnosis of a presumed meningitis or encephalitis (or both), we believed that the lumbar puncture was necessary. Although it was not mentioned, an intracranial-pressure monitor was placed the day after admission. The lateral ventricles were not amenable to a ventriculostomy. The patient did have intermittent decerebrate posturing 10 hours after the lumbar puncture, but at that time brain-stem function was clinically intact, with small, reactive pupils and intact doll's-eye maneuvers. The results of the repeated CT performed at that time (also not shown) were unchanged from the previous ones, suggesting that the neurologic signs were secondary to the bilateral hemispheric edema and not to herniation or brain-stem compromise. A CT scan obtained on the third hospital day (which was shown as Figure 1 of the article), 72 hours after the lumbar puncture, did show transtentorial herniation and distortion of the brain stem. This massive swelling was considered to be due to bihemispheric parenchymal disease and not to herniation. Although lumbar puncture in the presence of known cerebral edema is risky, studies have shown that in the absence of obstruction of cerebrospinal fluid flow or a clinically significant midline shift (or both), it rarely precipitates herniation.1-3 In this case, the risk of delayed diagnosis or misdiagnosis outweighed the risk of herniation.
3 ReferencesAnne B. Young, M.D., Ph.D.
Massachusetts General Hospital, Boston, MA 021141
Gower DJ ,Baker AL ,Bell WO ,Ball MR . Contraindications to lumbar puncture as defined by computed cranial tomography. J Neurol Neurosurg Psychiatry 1987;50:1071-1074
CrossRef | Web of Science | Medline2
Evans RW . Complications of lumbar puncture. Neurol Clin 1998;16:83-105
CrossRef | Web of Science | Medline3
Practice parameters: lumbar puncture (summary statement)Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 1993;43:625-627
Web of Science | Medline
Author/Editor Response
Dr. Vartanian replies:
To the Editor: The pathophysiology of necrotizing angiitis in patients with acute hemorrhagic leukoencephalitis is poorly understood. Although necrotizing angiitis is a distinguishing feature of this disorder, an individual patient may have some lesions at autopsy that resemble acute hemorrhagic leukoencephalitis and others that resemble acute disseminated encephalomyelitis (i.e., inflammatory demyelinating lesions with and without necrotizing angiitis). This fact, along with data from animals showing that acute and hyperacute experimental allergic encephalitis can be elicited by the same antigen, favors the argument that these diseases are part of a spectrum.
Does the necrotizing angiitis of acute hemorrhagic leukoencephalitis result from a specific insult to the cerebral microvasculature, or is it the result of an intense inflammatory response in the brain parenchyma, with secondary destruction of endothelial cells? Destruction of the cerebral vasculature by the inflammatory response, perhaps due to the generation of reactive oxygen intermediates, would help explain the lack of clinically significant involvement of blood vessels in other organs. However, endothelial cells within the central nervous system, in contrast to most other organs, have unique molecular components that contribute to the function of the blood–brain barrier. Thus, it is possible that they are uniquely susceptible or specifically targeted in acute hemorrhagic leukoencephalitis. This important question of whether there is a primary insult to the blood–brain barrier in demyelinating diseases, which has been considered by many investigators throughout the century, has yet to be answered.
Timothy Vartanian, M.D., Ph.D.
Harvard Institutes of Medicine, Boston, MA 02115
