A 77-year-old right-handed man was admitted to the hospital because of the recent onset of pain in the ear, difficulty speaking, and altered mental status. The patient had been well until the day before admission, when he awoke in the morning with nasal congestion and pain on the right side of his face. That evening, pain, accompanied by drainage, developed in the right ear; his wife administered ciprofloxacin eardrops. The next morning, the temperature was 37.8°C. At about 9 a.m., the patient's wife noted that his speech was slurred. The patient said he was tired and retired to nap. Ninety minutes later, his wife found him on the floor, unresponsive, and she called emergency medical services.
When the emergency medical technicians arrived, the patient was conscious, had incomprehensible speech, and could not walk. The blood pressure was 140/94 mm Hg, the pulse 160 beats per minute, and the oxygen saturation 100% while the patient was receiving supplemental oxygen; the respiratory rate was 20 breaths per minute. Dried blood was present in the nares and mouth. The pupils were equal and reactive to light, the facial expression was symmetrical, the right-hand grip was weak, and the torso leaned to the right. The patient did not move his feet when requested to do so. He was transported by ambulance to the emergency department of this hospital, arriving at 12:30 p.m.
Several days before admission, the patient had slipped on an icy sidewalk and struck his head. He did not lose consciousness and did not seek medical attention. He had hypertension, adenomatous colonic polyps, and a torn left medial meniscus; the baseline creatinine level was 1.3 mg per deciliter (115 µmol per liter). Lip swelling suggestive of angioedema had occurred after he consumed shellfish; there were no allergies to medications. Medications included omeprazole, diphenoxylate, lisinopril, chlorpheniramine, hydrochlorothiazide, and triamterene. The patient had been born in China and had immigrated to the United States in his third decade. He was a retired university professor who lived with his wife and traveled to Martha's Vineyard, Massachusetts, and China frequently. He drank wine daily, did not smoke, and had no recent exposure to animals. His father had died from a cerebral hemorrhage, and his mother had died from complications of diabetes.
In the emergency department, the blood pressure was 95/77 mm Hg, rising to 163/94 mm Hg within 5 minutes after arrival; the pulse was 147 beats per minute, the respiratory rate was 20 breaths per minute, and the oxygen saturation was 96% while the patient was breathing 5 liters of oxygen per minute by means of a nasal cannula. The temperature was 37.4°C. On examination by a neurologist, the patient could be roused with mild stimulation; he had aphasia and did not follow commands. The gaze was midline without deviation; the pupils were equal and each 4 mm in diameter, decreasing to 2 mm on direct illumination. There was no ptosis or facial droop. Dried blood was present in the right external ear canal, both nares, and the oropharynx. The neck was supple, and rhonchi were heard in both lungs. The heart sounds were normal, and the abdomen was distended and tender in the right upper quadrant, without bowel sounds. The patient moved both arms and legs purposefully in response to noxious stimuli. The reflexes were 1+ in the arms, trace at the patellar tendons, and absent at the Achilles' tendons; the plantar responses were flexor.
An electrocardiogram showed sinus tachycardia with first-degree atrioventricular block but was otherwise normal. Laboratory-test results are shown in Table 1. Urinalysis revealed few bacteria and three to five granular casts per high-power field. Thiamine and glucose were administered intravenously, followed by lorazepam (4 mg).
|
Computed tomography (CT) of the head performed 1 hour after the patient's arrival in the emergency department, without the administration of contrast material, revealed pneumocephalus, opacified right mastoid air cells, and age-related parenchymal changes. The initial interpretation suggested the presence of a longitudinal fracture of the right temporal bone. A chest radiograph was normal.
Ninety minutes after arrival, the temperature rose to 38.6°C, and blood specimens were sent for culture. Acetaminophen, ceftriaxone (2 g), metronidazole (500 mg), and vancomycin (1 g) were administered. Three hours after arrival, a lumbar puncture was performed; the opening pressure was 270 mm of water. The cerebrospinal fluid was xanthochromic; the results of laboratory tests are shown in Table 2. Specimens were sent for fungal, bacterial, and viral cultures; Gram's staining; and testing for cryptococcal antigen and for herpes simplex virus DNA. The closing pressure was 170 mm of water.
|
Four hours after arrival, CT of the temporal bones, performed without the administration of contrast material, revealed pneumocephalus in the right middle and anterior cranial fossae, fluid within the middle ear and mastoid air cells, and loss of the bony integrity of the tegmen tympani. There was no evidence of a fracture of the temporal bone.
The results of a diagnostic test were received.
Differential Diagnosis
Dr. Martin A. Samuels: Dr. Gonzalez, may we see the images?
Dr. R. Gilberto Gonzalez: CT of the head showed pneumocephalus (Figure 1A) and opacification of the right middle ear and mastoid air cells. There was a question of a longitudinal temporal-bone fracture, and a dedicated temporal-bone CT scan was obtained, which showed opacification of the right middle ear (Figure 1B). Sagittal images demonstrated loss of integrity of the tegmen tympani (Figure 1C). There was no evidence of a fracture.
|
Dr. Samuels: Was the air in the epidural, subdural, or subarachnoid space?
Dr. Gonzalez: It appears to be subarachnoid. The air pocket in the middle cranial fossa is round, with an air–liquid level. If it were in the subdural or the epidural space, it would conform to the shape of the skull.
Dr. Samuels: Although this is clearly a case of otitic meningitis, it raises several important questions: the cause of the otorrhea, the meaning of the pneumocephalus, the risks of lumbar puncture, the cause of the aphasia, the likely causal organism, and the cause of the defect in the tegmen tympani.
Otorrhea
This patient had drainage from his ear that was variously described as a discharge and as dried blood. Was this an infectious discharge from the external ear, or was it cerebrospinal fluid otorrhea? Cerebrospinal fluid otorrhea refers to the presence of cerebrospinal fluid in the external auditory canal due to an abnormal connection between the sub-arachnoid space and the tympanomastoid space. A rent in the tympanic membrane is required for the cerebrospinal fluid to enter the external ear. Otherwise, fluid accumulating in the middle ear is drained via the eustachian tube into the nasopharynx and often goes unrecognized.
An important clinical challenge is to recognize when fluid emanating from the ear is, in fact, cerebrospinal fluid. Cerebrospinal fluid otorrhea may be serosanguineous and mistaken for blood, as was probably the case in this patient. Otorrhea and rhinorrhea may be recognized as containing cerebrospinal fluid when a handkerchief soaked in the fluid does not stiffen when dry, because of the fluid's relatively low protein level (i.e., <2> A fluid glucose level of greater than 40 mg per deciliter (2.2 mmol per liter) also suggests that the fluid is cerebrospinal fluid. The most accurate test is a measurement of β2-transferrin, a protein that is found only in cerebrospinal fluid.1 In this case, the otic discharge may well have been cerebrospinal fluid, but no test was obtained to document that fact.
One of the causes of cerebrospinal fluid otorrhea is fracture of the temporal bone. Temporal-bone fractures are classified as either transverse or longitudinal on the basis of their relationship to the otic capsule and the axis of the petrous ridge. In reality, many are oblique, with some features of both types. A longitudinal fracture, which was suspected in this case, may cause puncture of the tympanic membrane, producing otorrhea, whereas transverse fractures usually do not cause puncture of the tympanic membrane. In one study of 820 temporal-bone fractures,2 15% had cerebrospinal fluid fistulas, 85% of which involved the ear; 95% of the cerebrospinal fluid fistulas closed spontaneously within 14 days, and 7% of the patients contracted meningitis. Antibiotic prophylactic therapy administered until the fistula is closed may be helpful in preventing meningitis.3
Pneumocephalus
Pneumocephalus, the accumulation of intracranial air, was seen on CT shortly after the patient's admission to the emergency department. It can be due to trauma, neoplasm, infection, or surgery. Otogenic pneumocephalus is not rare: among 59 consecutive patients with pneumocephalus, the cause was trauma in 36% of patients, otitis media in 31%, otic surgery in 31%, and congenital defects in 2%.4 Rarely, pneumocephalus may be caused by brain abscess involving air-forming organisms,5 but there is no evidence for such an infection in this case. In this patient, the presence of pneumocephalus is evidence of a cerebrospinal fluid fistula, which allowed cerebrospinal fluid to exit through the ear and air to enter from the mastoid air cells of pneumatized bone or from the external auditory canal.
Pros and Cons of Lumbar Puncture
The chance of increased intracranial pressure is substantial in the presence of otic infection. In the era before antibiotics, hydrocephalus in the presence of otic infection was termed "otitic hydrocephalus"; this was a misnomer in most cases, since it was usually increased levels of interstitial brain water (pseudotumor) related to poor venous drainage from the failure of one or more of the dural sinuses. However, in the presence of leptomeningitis, an authentic hydrocephalus may develop because of the slowing of cerebrospinal fluid flow and blockage of the resorption of cerebrospinal fluid into the venous system at the arachnoid granulations. The two processes (hydrocephalus and pseudotumor) could therefore coexist in this patient, producing increased intracranial pressure while leaving ventricular size relatively normal.
Fortunately, the neurologists caring for this patient were not afraid to perform a lumbar puncture, despite the risk of increased intracranial pressure. The risks of the procedure, though real, are generally overrated, and I believe it is likely that many more patients have had a deteriorating condition from the failure to perform the procedure than have had a complication or death resulting from the procedure. The presence of a mass, particularly near the midline or in the posterior fossa, probably increases the risks. If the clinician is concerned about performing the procedure in a patient in whom it is indicated because of the real threat of infection, rapid imaging of the brain should be performed before the procedure is undertaken.6
The cerebrospinal fluid in this patient was, indeed, under increased pressure (270 mm of water) and showed a neutrophilic pleocytosis, an elevated protein level, and a glucose level that was about 45% of the blood glucose level. All these findings strongly suggests a bacterial leptomeningitis. It is likely that the lumbar subarachnoid space, accessible with the use of the spinal needle, did not fully reflect the vigorousness of the inflammatory response around the brain, because the fluid was not flowing freely. The elevated cerebrospinal fluid protein level of 322 mg per deciliter may, in part, reflect stagnation of cerebrospinal fluid, known as Froin's syndrome.
Aphasia
One of the more interesting and cryptic aspects of this case is the aphasia. Disorders of speech and language are complex and often difficult to characterize accurately in the presence of generalized encephalopathy. I shall assume that the neurologist was correct in characterizing this patient's problem as aphasia, a disorder of language. For virtually all right-handed people and most left-handed people, the major systems for language are strongly lateralized to the left hemisphere. In general, aphasias are produced by disorders that affect the cerebrum, its subcortical connections, and the thalamus. It is uncommon, for example, to see aphasia with a subdural hematoma, a lesion on the surface of the brain, until the patient is so drowsy that the language problem is difficult to distinguish from a disorder of consciousness.
What disorder of the left hemisphere could have occurred in this patient? Septic dural sinus thrombosis is a common complication of infection of the tympanomastoid space,7,8 but that should have affected the right transverse sinus, which should not produce aphasia. Dural sinus septic thrombophlebitis only rarely spreads to the opposite transverse sinus. Cerebritis or brain abscess in the left hemisphere could have occurred by hematogenous spread of infection in the right ear, but the imaging studies showed no such lesion.
The most likely explanation is cerebral vasospasm in reaction to the pus in the subarachnoid space. To produce aphasia, the middle cerebral arteries should be affected. Prolonged, severe vasospasm can lead to cerebral infarction, which may be the major sequela in survivors of bacterial, fungal, and tuberculous meningitis.
The Organism
What organism or organisms are responsible for this otitic meningitis? The most common bacteria are Streptococcus pneumoniae (about half of cases), Neisseria meningitidis (about one fifth of cases), Listeria monocytogenes (about one tenth of cases), and Haemophilus influenzae (about one tenth of cases).9,10,11 If one hypothesizes that the infection in the ear actually caused the bony erosion that led to the cerebrospinal fluid fistula, one might favor more aggressive organisms such as Staphylococcus aureus, gram-negative organisms, and group A streptococcus. However, the relationship of the tegmen defects to the course of the ear infection is not known, so there is no way to predict the identity of the organism. The decision to treat the patient with broad-spectrum antibiotics, including coverage for anaerobic organisms, was well founded. The only additional intervention might have been intravenous high-dose corticosteroids, but the data supporting this approach are equivocal.6
Defects of the Tegmen Tympani
The final issue revolves around the defects of the tegmen tympani found on the temporal-bone CT. The tegmen tympani is the portion of the temporal bone that overlies the tympanic and mastoid cavities, forming the floor of the middle fossa. A few (less than five) defects in the tegmen are common and have been found in 15 to 34% of carefully examined temporal bones; however, five or more defects are rare (found in <1%> More than five defects are associated with an increased risk of neuro-otologic problems, including otorrhea and meningitis.12 Tegmen defects may be congenital, leading to early and recurrent episodes of meningitis. Tegmen defects that arise late in life could be due to infection, trauma, or the long-term effect of a congenital defect. Chronic infection, with or without cholesteatoma, in the tympanomastoid space can lead to erosion of the tegmen. In the era before antibiotic treatment, this was probably a fairly common mechanism of otitic meningitis.
The arachnoid granulations (pacchionian granulations) are pockets of the subarachnoid space that project through the dura into the dural venous sinuses. Most of these are along the superior sagittal sinus, allowing for the return of cerebrospinal fluid to the venous system; however, aberrant arachnoid granulations can impinge on pneumatized bone in the posterior and middle fossa, erode the tegmen tympani, and lead to cerebrospinal fluid fistulas.13 It is possible that minor head trauma could have exacerbated preexisting defects in the tegmen tympani.
Summary
To synthesize the sequence of events in this case, there was chronic silent otitis media, and aberrant arachnoid granulations produced discontinuities in the tegmen tympani. Minor head trauma disrupted the fragile tegmen and, together with the chronic otitis media, produced a new rent in the tympanic membrane. Cerebrospinal fluid otorrhea, pneumocephalus, and bacterial meningitis ensued. The organism was aggressive, further eroding the tegmen tympani. Pus in the subarachnoid space caused cerebral vasospasm, producing aphasia. The dural sinuses, particularly the right transverse sinus, became incompetent, leading to increased intracranial pressure. The diagnostic tests were probably the blood culture and cerebrospinal fluid culture, as well as magnetic resonance imaging (MRI), angiography, and venography.
Dr. Nancy Lee Harris (Pathology): Dr. Caviness, you cared for this patient on the Neurology Service; can you comment on your thinking?
Dr. Verne S. Caviness, Jr. (Neurology): Dr. Alireza Atri was the neurologist called to see the patient in the emergency department. Aphasia was present before the patient's mental status deteriorated; we considered in retrospect that the patient might have had a seizure and that the aphasia and transient right hemiparesis might have been postictal. Dr. Atri immediately recognized that this was meningitis, instituted three-drug therapy, and performed the lumbar puncture. At that time, a temporal-bone fracture was considered and searched for but could not be identified. We considered a diastatic suture in the medial temporal bone. A defect in the tegmen tympani was not reported at the time.
Dr. Harris: Dr. Kim, what was the opinion of the consultants from the Department of Infectious Disease?
Dr. Arthur Y. Kim: We suspected bacterial meningitis from an otitic or a mastoid source or both.
Clinical Diagnosis
Bacterial meningitis due to otitis or mastoiditis or both.
Dr. Martin A. Samuels's Diagnosis
Otitic bacterial leptomeningitis, secondary to tegmen tympani defects caused by arachnoid granulations, possibly aggravated by minor head trauma.
Pathological Discussion
Dr. Kim: The diagnostic test was Gram's staining of the cerebrospinal fluid specimen, which revealed moderate numbers of gram-positive cocci in pairs and chains, suggesting streptococcal species. Although S. pneumoniae remained a primary consideration, we also considered other organisms, including group A streptococci and enterococci. We recommended broad-spectrum antibiotics to cover these possibilities. Within hours, group A streptococci were grown on blood culture and cerebrospinal fluid culture.
Group A streptococci are a rare cause of bacterial meningitis, accounting for 0.5 to 1.5% of community-acquired cases10,11,14; the mortality rate (27%) is similar to that for pneumococcal meningitis (30%).6 Otitis media due to group A streptococci is associated with high rates of local invasion, including tympanic perforation and mastoiditis, as was seen in this case,15 and it is the most important risk factor for group A streptococcal meningitis among adult patients.16 Most patients with group A streptococcal meningitis do not have the clinical features of septic shock associated with invasive group A streptococcal disease, and the incidence of the meningitis has not increased despite an increasing incidence of other forms of invasive disease.16,17 After the culture results were received, additional imaging studies were performed.
Dr. Gonzalez: Brain MRI was performed before and after the administration of contrast material, as was magnetic resonance angiography and venography. In addition to the pneumocephalus, subtle areas of increased signal involved the cortex of the right temporal, parietal, and frontal lobes. Diffusion MRI did not show evidence of an acute ischemic event. The magnetic resonance venogram showed flow within all the major veins and sinuses. The angiogram was limited by the artifact of the technique, but it did not show good flow distal to the internal carotid arteries or distal to the basilar arteries. This can sometimes be seen in normal persons as an artifact, but we cannot rule out a process such as vasospasm.
Dr. Kim: Persistent hypotension and multiorgan failure ensued; vasopressor support was withdrawn, and death occurred 19 hours after admission to the emergency department. Permission for an autopsy was obtained.
Dr. Anat Stemmer-Rachamimov: At autopsy, multiple petechial hemorrhages were noted in the dura overlying the right middle fossa, and the underlying right mastoid bone was swollen, soft, and hemorrhagic. The venous sinuses were free of thrombi. The leptomeninges appeared cloudy, with small collections of white, thick fluid surrounding small meningeal vessels that were most prominent in the leptomeninges overlying the cerebral hemispheres (Figure 2A).
|
Microscopical examination showed a dense inflammatory infiltrate in the subarachnoid space, which was composed of neutrophils and immature mononuclear cells (Figure 2B). Fibrinoid necrosis and thrombosis were noted in some of the meningeal arterioles. In the brain parenchyma, inflammatory cells were clustered around cortical arterioles and scattered in the neuropil. The inflammation involved the left temporal lobe as well as other areas of the brain.
Sections of the mastoid bone were reviewed with Dr. Saumil Merchant (Massachusetts Eye and Ear Infirmary). There was no evidence of fracture, a preexisting chronic inflammatory condition (such as a cholesteatoma) in the middle ear, or labyrinthitis in the semicircular canals. In contrast, the mastoid mucosal lining was thickened by means of vascular engorgement, edema, and inflammatory-cell infiltrates (Figure 2C). Large inflamed and necrotic arachnoid granulations (>3 mm in diameter) were noted in sections of the mastoid bone (Figure 2D), and aggregates of gram-positive cocci were present in the subarachnoid space and in the mastoid air cells.
Arachnoid granulations may increase in size with age, and intermediate and large granulations (3.0 to 9.5 mm in diameter) may herniate through dural defects and erode bone (aberrant granulations).12,13,18,19 The presence of large arachnoid granulations and the absence of fracture, preexisting middle-ear disease, and labyrinthitis suggest that a dural defect and bone erosion caused by aberrant arachnoid granulations are the most probable route for intracranial extension of the suppurative mastoiditis in this case, which then led to acute meningoencephalitis.
Dr. Harris: Would either of the patient's sons, Dr. Thomas Lee or Dr. Richard Lee, like to comment?
Dr. Thomas H. Lee, Jr. (Internal Medicine, Brigham and Women's Hospital): We learned from this case that the care of patients does continue after they die. The performance of the autopsy and the careful review of all aspects of this case for this conference provided us with two opportunities that I believe every family who loses a loved one would appreciate. The first was the opportunity to know what caused his death. During hospitalization, we knew that there was a suspicion of a fracture on the CT scan; after my father's death, we left the hospital believing that his fall on the ice had caused the fracture and therefore his death. After the autopsy, which showed no fracture, we realized that perhaps the fall had not caused his death. After the discussion at this conference, we now have a final answer.
The second opportunity is to have something good arise from tragedy. My brother and I know that our father, as a professor at the Massachusetts Institute of Technology, would have been completely in favor of having his death contribute to the education of physicians who will care for patients in the future. He would have joked that this was his final publication, and I would have said to him, "It is your first in a journal worth reading."
Anatomical Diagnosis
Acute bacterial meningoencephalitis, acute right otitis media and mastoiditis, and aberrant arachnoid granulations with erosion of the temporal bone.
Dr. Samuels reports receiving consulting fees from M/C Communications. No other potential conflict of interest relevant to this article was reported.
Source Information
From the Department of Neurology, Brigham and Women's Hospital (M.A.S.); the Departments of Radiology (R.G.G.), Infectious Diseases (A.Y.K.), and Pathology (A.S.-R.), Massachusetts General Hospital; and the Departments of Neurology (M.A.S.), Radiology (R.G.G.), Medicine (A.Y.K.), and Pathology (A.S-R.), Harvard Medical School.
References
- Oberascher G. Cerebrospinal fluid otorrhea -- new trends in diagnosis. Am J Otol 1988;9:102-108. [ISI][Medline]
- Brodie HA, Thompson TC. Management of complications from 820 temporal bone fractures. Am J Otol 1997;18:188-197. [ISI][Medline]
- Brodie HA. Prophylactic antibiotics for posttraumatic cerebrospinal fluid fistulae: a meta-analysis. Arch Otolaryngol Head Neck Surg 1997;123:749-752.
[Free Full Text] - Andrews JC, Canalis RF. Otogenic pneumocephalus. Laryngoscope 1986;96:521-528. [CrossRef][ISI][Medline]
- Parmar MS. Pneumocephalus associated with Bacteroides fragilis meningitis. J Postgrad Med 2004;50:272-273. [Medline]
- van de Beek D, de Gans J, Tunkel AR, Wijdicks EFM. Community-acquired bacterial meningitis in adults. N Engl J Med 2006;354:44-53.
[Free Full Text] - Southwick FS, Richardson EP Jr, Swartz MN. Septic thrombosis of the dural venous sinuses. Medicine (Baltimore) 1986;65:82-106. [Medline]
- Stam J. Thrombosis of the cerebral veins and sinuses. N Engl J Med 2005;352:1791-1798.
[Free Full Text] - Schuchat A, Robinson K, Wenger JD, et al. Bacterial meningitis in the United States in 1995. N Engl J Med 1997;337:970-976.
[Free Full Text] - Durand ML, Calderwood SB, Weber DJ, et al. Acute bacterial meningitis in adults: a review of 493 episodes. N Engl J Med 1993;328:21-28.
[Free Full Text] - van de Beek D, de Gans J, Spanjaard L, Weisfelt M, Reitsma JB, Vermeulen M. Clinical features and prognostic factors in adults with bacterial meningitis. N Engl J Med 2004;351:1849-1859. [Erratum, N Engl J Med 2005;352:950.]
[Free Full Text] - Merchant SN, McKenna MJ. Neurotologic manifestations and treatment of multiple spontaneous tegmental defects. Am J Otol 2000;21:234-239. [CrossRef][ISI][Medline]
- Gacek RR. Arachnoid granulation cerebrospinal fluid otorrhea. Ann Otol Rhinol Laryngol 1990;99:854-862. [ISI][Medline]
- Swartz MN, Dodge PR. Bacterial meningitis -- a review of selected aspects: general clinical features, special problems, and unusual meningeal reactions mimicking bacterial meningitis. N Engl J Med 1965;272:725-731. [ISI][Medline]
- Segal N, Givon-Lavi N, Leibovitz E, Yagupsky P, Lieberman A, Dagan R. Acute otitis media caused by Streptococcus pyogenes in children. Clin Infect Dis 2005;41:35-41. [CrossRef][ISI][Medline]
- van de Beek D, de Gans J, Spanjaard L, Sela S, Vermeulen M, Dankert J. Group A streptococcal meningitis in adults: report of 41 cases and a review of the literature. Clin Infect Dis 2002;34:e32-e36. [CrossRef][ISI][Medline]
- Sommer R, Rohner P, Garbino J, et al. Group A beta-hemolytic streptococcus meningitis: clinical and microbiological features of nine cases. Clin Infect Dis 1999;29:929-931. [ISI][Medline]
- Gacek RR. Evaluation and management of temporal bone arachnoid granulations. Arch Otolaryngol Head Neck Surg 1992;118:327-332.
[Free Full Text] - Rosenberg AE, O'Connell JX, Ojemann RG, Plata MJ, Palmer WE. Giant cystic arachnoid granulations: a rare cause of lytic skull lesions. Hum Pathol 1993;24:438-441. [CrossRef][ISI][Medline]
No comments:
Post a Comment