Ann Otol 88 :1979
NECROTIZING ("MALIGNANT") EXTERNAL OTITIS HISTOPATHOLOGIC PROCESSES ROBERT
I.
KOHUT,
MD
WINSTON-SALEM, NORTH CAROLINA
JOHN
R.
LINDSAY,
MD
CmCAGO, ILLINOIS
The histologic findings in a serially sectioned temporal bone, from a patient who succumbed to brain abscess secondary to necrotizing ("malignant") external otitis, are described. The mechanism of invasion of the ear canal appears to be due to local bone necrosis. This in turn extends to the submucosal vasculature of the pneumatic spaces. The infective process extends submucosally, establishing one or several sites of bone destruction. The lumen of the pneumatic space is not involved. In this process, the periphery of the fibrous inflammatory tissue formation is the site of active bone destruction. In pneumococcal petrositis, the peripheral fibrous elements are protective. The process in malignant external otitis may extend directly to adjacent central nervous system structures inoculating the structure with Pseudomonas. Development of Pseudomonas brain abscesses can be slow, allowing for new bone closure of the site from which the infection spreads as demonstrated in this specimen. Therefore, apparent local control of the disease can be established while a central infective process progresses.
acteristics of osteomyelitis. Schuknecht has agreed with this conclusion.'
The attempts to understand the pathogenic processes of lethal inflammatory ear disease during the 1880s followed centuries of such efforts. At that time Bricheteau (1834) 1 may have been discussing the fulminating form of external otitis which was identified by Meltzer and Keleman in 1959.2 Chandler" later described the entity in clinical detail and coined the name, malignant external otitis. Some of the early French authors interpreted the presence of purulent material from the ear as a spontaneous rupture of a brain abscess; the ear, they thought, provided the route for external drainage. Though now recognized as fallacious, the association of brain abscess and fulminating ear disease, was becoming firmly established and has relevance to our current knowledge and investigations concerning malignant external otitis.
Pseudomonas has been regularly cultured from patients having this disorder. However, it was recently stated" that appropriately prepared anaerobic cultures in some patients revealed anaerobes which responded to specific medications chosen on the basis of sensitivity studies.
Meltzer and Keleman" indicated that the histologic specimen showed scattered areas of involvement of an inflammatory process. From this they concluded that the process was blood-borne and involved diploic bone, the essential char-
It is curious that some papers report death from brain abscesses or unidentified causes after the otitic process has cleared, and apparent resolution of the disease had occurred." The purpose of this paper is to de-
Chandler" identified the potential rapid clinical course of the disease and the characteristics that allowed it to be distinguished from other processes. He indicated that diabetes was a common disease in patients having the disorder. The surgical management proposed for this entity was appropriately extensive based on its rapid course. Currently, medical management satisfactorily controls promptly treated cases.
·Unldentlfled verbal dIscussIon from audIence to Dr. Ryan Chandler's presentation on Malignant External Otitis. American College of Surgeons Meeting, 1976. From the DIvIsIon of Otolaryngology, UnIversIty of CalifornIa. Irvine, CalifornIa, and the DIvIsIon of Otolaryngology. University of Chicago, Ohlcago, Illinois. Presented at the meeting of the American OtologIcal Society, Inc., Los Angeles, California, March 31-April 1, 1979.
714
Downloaded from aor.sagepub.com at Kungl Tekniska Hogskolan / Royal Institute of Technology on September 2, 2015
EXTERNAL OTITIS
scribe the histologic findings in a serially sectioned temporal bone, from a patient who succumbed to brain abscess secondary to malignant external otitis. This material gives us information regarding: 1) the mechanisms of invasion; 2) the details of the routes of spread; 3) the process by which normally resistant bone is destroyed; and 4) the sequential events which may account for the phenomenon of late complications in apparently cured patients. The pathologic process is contrasted to that of pneumococcal petrositis. CASE REPORT
The patient (AD), a 76-year-old Caucasian male. was well until August 1 of the year of his death, when while at the beach a wave hit him in the left ear and he experienced some pain. He continued to have pain and saw his doctor who placed him on ear drops and antibiotics. The disease continued and an otologist diagnosed severe external otitis. He was treated medically with drops and antibiotics. Abscesses of the canal skin were biopsied and curetted twice early in October. On August 17, a modified radical mastoidectomy was performed. This was revised November 1. A facial paralysis had developed between the two procedures. At the time of the second procedure, the nerve was found to be intact, but surrounded by an inflammatory process. The mandibular fossa was bathed in pus. Cultures grew Pseudomonas. The patient was transferred to the University of Florida hospital with residual severe pain in his left ear and jaw. On examination, a left posterior auricular sinus was present (previous site of Penrose drain). The mastoid cavity and external canal contained granular tissue and purulent exudate. The external ear was normal. There was a complete left facial naralvsis. He was identified as being midlv diabetic, a condition which was controllable with diet and oral medications. On November 22. surgery was performed. The ear was debrided of all granulation tissue. Culture taken at that time grew Pseudomonas. He gradually improved. The ear was drying and gran-
715
ulations resolving. On approximately December 15,24 days after his last surgery and 3~ months since the onset of his disease, an insidious beginning of a changing mental status occurred. This consisted of occasional disorientation occurring at night. Neurologic consultants felt this to be normal for the patient's age. There were no other new neurological abnormalities. The mental status defect became more prominent and a brain scan was ordered. This showed some increased activity in the left posterotemporo-occipital area, consistent with recent aural surgery. Neurosurgical consultants recommended skull xrays and an electroencephalogram. On December 27 at 4:00 AM, the patient was found to be unresponsive and flaccid. A lumbar puncture produced grossly purulent spinal fluid. A cerebral angiogram was performed and revealed a left temporal lobe mass. Evacuation of a left temporal lobe brain abscess was performed. The patient expired on December 30. An autopsy was performed four hours after death. The temporal bones were removed and prepared for processing. 0 On removal of the calvarium there was some serous drainage, but no epidural hemorrhage. The brain had a normal shape. There was clouding of the meninges. In the left temporal area there was softening and several hemorrhagic sites were noted corresponding to the sites of surgical drainage. Examination of the inside of the skull revealed no drainage or opening from the temporal and mastoid bones into the cranial cavity. The transverse and sigmoid sinuses were examined and there appeared to be clotted blood of antemortem nature in the left sigmoid sinus. Microscopic evaluation of the brain showed a small left temporal lobe abscess, with focal meningitis and ventriculitis, secondary to a rupture of the small brain abscess into the left lateral ventrical. TEMPORAL BONE FINDINGS
Left Ear. The middle ear cavity had minimal submucosal fibrosis, but no evidence of active bone destruction (Fig. 1 ). In the inner ear there was no inflam-
-The temporal bones were processed In the Otolaryngology Temporal Bone Laboratory the University ot Florida School ot Medicine, Gainesville, Florida. '
Downloaded from aor.sagepub.com at Kungl Tekniska Hogskolan / Royal Institute of Technology on September 2, 2015
716
KOHUT-LINDSAY
Fig. 1. A) Stapes footplate. There is minimal submucosal thickening over the stapes (a). B) Round window membrane. There is no evidence of an inflammatory process in the perilymphatic space (b). On the external surface of the round window membrane, thickening and submucosal fibrosis has occurred.
Fig. 2. Petrous apex (low power) adjacent to the superior semicircular canal (a). Centrally (b) there is an area of a fibrotic inflammatory process. Destruction of trabecular and compact bone has occurred at the periphery of this area. There is thickening of the mucosa within pneumatic spaces (c). The lumen of all pneumatic spaces are free of pus; some contain protein debris. The semicircular canal contains no inflammatory debris. (X2.5)
Downloaded from aor.sagepub.com at Kungl Tekniska Hogskolan / Royal Institute of Technology on September 2, 2015
EXTERNAL OTITIS
717
Fig. 3. A) Perilymphatic duct showing minimal subepithelial thickening in its midportion. B) Direct extension of the fibroinflammatory process to the duct's widened portion (a). There is in this area marrow space involvement (b). A pneumatic space ( c) has subepithelial fibrosis, but no lumenal purulent material. An area of compact bone destruction is visible (d).
matory involvement of the membranous cochlea, vestibule, or semicircular canals (Figs. 1 and 2). At the midportion of the cochlear duct, a subepithelial inflammatory process was observed. This process extended from the duct's midportion towards its arachnoid orifice. Near the external dural aperture of the cochlear duct at its widened portion, direct communication occurred with a fibrous nonpurulent type of inflammatory process. There was active bone destruction at the periphery of this process (Fig. 3). The dura adjacent to this area was slightly thickened, but there was no evidence of direct spread to the central nervous system structures. OTIC CAPSULE AND PNEUMATIC BONE
The residual defect of the mastoid cavity was filled with connective tissue, in some areas covered with a thin layer
of squamous epithelium. This fibrous process extended superiorly to the apex adjacent to the arcuate eminence (Fig. 2), inferior-medially to the perilymphatic duct (Fig. 3), and anteriorly towards the middle ear space. In the apical region new bone formation had occurred, presumably forming the new barrier between the fibrous process and the middle fossa. The mode of bone destruction in all areas appeared to be by osteoblastic and osteoclastic activity and occurred at the periphery of the fibrous tissue. The compact bone was widely involved in this destructive process. Pneumatic trabeculae were not widely destroyed in any coalescent process. The lumen of the pneumatic spaces contained proteinaceous debris, but did not contain elements of an inflammatory process. The mucosa lining the pneumatic spaces was commonly thickened
Downloaded from aor.sagepub.com at Kungl Tekniska Hogskolan / Royal Institute of Technology on September 2, 2015
718
KOHUT-LINDSAY
Fig. 4. Destruction of compact bone (a). Adjacent pneumatic spaces (b). There is submucosal thickening of the pneumatic spaces and subepithelial new bone formation ( c ). The bony walls of some pneumatic spaces have been destroyed, but the mucosa remains intact (d).
and involved in a fibrous process similar to that seen in the larger destructive areas. The pneumatic spaces were rarely obliterated by this process. The bony walls of the pneumatic spaces often contained submucosal new bone or osteoid. In some areas the pneumatic spaces were separated by new bone from the areas of connective tissue. Bone destruction had occurred at the periphery of these areas of fibrous tissue, involving both endochrondral and periosteal bone (Fig. 4). The bone destruction extended directly to the facial nerve. The nerve was partially destroyed and replaced with fibrous tissue. DISCUSSION
This process should be distinguished from the process seen in pneumococcal petrositis in which the lumen of the pneumatic space contains exudate and the bone destruction involves mainly trabecular walls, with a protective fibrous process at the periphery of the abscess formation.' In this instance (pneumococcal petrositis), no bone destruction occurs at the periphery of the protective fibrous walls.
Ordinarily in pneumococcal petrositis, as fibrosis occurs the bone destruction stops, but in malignant external otitis the bone destruction continues in the areas of mature fibrosis. In pneumococcal abscess formation, the destruction process is central about the abscess, with a protective reactive fibrosis forming peripherally in which there is little or no bone destruction. Whereas, in malignant external otitis, the striking bone destruction is in the peripheral fibrous tissue which extends on a broad front without respect to pneumatic or marrow spaces, or compact bone. The spread of the process is facilitated by the pneumatic space submucosa. This is evidenced by the observation that in any major area of development, there is submucosal pneumatic space involvement. The opposite is not true, that is, there is never isolated marrow space involvement. The process is probably dependent on vascular spread and although the marrow spaces are vascularly connected, the histologic evidence suggests that the submucosal blood sup-
Downloaded from aor.sagepub.com at Kungl Tekniska Hogskolan / Royal Institute of Technology on September 2, 2015
EXTERNAL OTITIS
ply of the pneumatic space provides the major pathway for extension. It is interesting to note some differences between the findings in this temporal bone and those identified by Meltzer and Keleman." In their specimen there was extensive involvement of diploic bone, the petrous tip was transformed into a large abscess, trabecular destruction was prominent and the pneumatic spaces were filled with pus or in other areas obliterated.
The striking thing in this temporal bone specimen is the lack of destruction of the pneumatic spaces with marked widespread bone distribution outside of the pneumatic spaces without particular involvement of diploic bone. It seems that the compact bone in malignant external otitis is more susceptible than the bone of the pneumatic spaces. This is quite the opposite to the process seen in pneumococcal petrositis where the abscess forms by breaking down the pneumatic spaces with the formation of a protective fibrosis and new bone formation in the adjacent marrow spaces/ The presence of a small brain abscess in the temporal lobe suggests that the otitic process had spread to the brain from the middle fossa surface of the temporal bone. At autopsy, no defect was detected on the temporal bone surface. On histologic evaluation of the apex, new bone formation had formed, presumably developing a new barrier to the middle fossa. This would suggest that the central spread of the infection had occurred early in the disease process, allowing for new bone formation to heal the site from which the infection had extended to the brain. The slow rate of growth in some brain abscesses (Pseudomonas) is consistent with this proposed mechanism.
NECROTIZING ("MALIGNANT") EXTERNAL OTITIS HISTOPATHOLOGIC PROCESSES ROBERT
I.
KOHUT,
MD
WINSTON-SALEM, NORTH CAROLINA
JOHN
R.
LINDSAY,
MD
CmCAGO, ILLINOIS
The histologic findings in a serially sectioned temporal bone, from a patient who succumbed to brain abscess secondary to necrotizing ("malignant") external otitis, are described. The mechanism of invasion of the ear canal appears to be due to local bone necrosis. This in turn extends to the submucosal vasculature of the pneumatic spaces. The infective process extends submucosally, establishing one or several sites of bone destruction. The lumen of the pneumatic space is not involved. In this process, the periphery of the fibrous inflammatory tissue formation is the site of active bone destruction. In pneumococcal petrositis, the peripheral fibrous elements are protective. The process in malignant external otitis may extend directly to adjacent central nervous system structures inoculating the structure with Pseudomonas. Development of Pseudomonas brain abscesses can be slow, allowing for new bone closure of the site from which the infection spreads as demonstrated in this specimen. Therefore, apparent local control of the disease can be established while a central infective process progresses.
acteristics of osteomyelitis. Schuknecht has agreed with this conclusion.'
The attempts to understand the pathogenic processes of lethal inflammatory ear disease during the 1880s followed centuries of such efforts. At that time Bricheteau (1834) 1 may have been discussing the fulminating form of external otitis which was identified by Meltzer and Keleman in 1959.2 Chandler" later described the entity in clinical detail and coined the name, malignant external otitis. Some of the early French authors interpreted the presence of purulent material from the ear as a spontaneous rupture of a brain abscess; the ear, they thought, provided the route for external drainage. Though now recognized as fallacious, the association of brain abscess and fulminating ear disease, was becoming firmly established and has relevance to our current knowledge and investigations concerning malignant external otitis.
Pseudomonas has been regularly cultured from patients having this disorder. However, it was recently stated" that appropriately prepared anaerobic cultures in some patients revealed anaerobes which responded to specific medications chosen on the basis of sensitivity studies.
Meltzer and Keleman" indicated that the histologic specimen showed scattered areas of involvement of an inflammatory process. From this they concluded that the process was blood-borne and involved diploic bone, the essential char-
It is curious that some papers report death from brain abscesses or unidentified causes after the otitic process has cleared, and apparent resolution of the disease had occurred." The purpose of this paper is to de-
Chandler" identified the potential rapid clinical course of the disease and the characteristics that allowed it to be distinguished from other processes. He indicated that diabetes was a common disease in patients having the disorder. The surgical management proposed for this entity was appropriately extensive based on its rapid course. Currently, medical management satisfactorily controls promptly treated cases.
·Unldentlfled verbal dIscussIon from audIence to Dr. Ryan Chandler's presentation on Malignant External Otitis. American College of Surgeons Meeting, 1976. From the DIvIsIon of Otolaryngology, UnIversIty of CalifornIa. Irvine, CalifornIa, and the DIvIsIon of Otolaryngology. University of Chicago, Ohlcago, Illinois. Presented at the meeting of the American OtologIcal Society, Inc., Los Angeles, California, March 31-April 1, 1979.
714
Downloaded from aor.sagepub.com at Kungl Tekniska Hogskolan / Royal Institute of Technology on September 2, 2015
EXTERNAL OTITIS
scribe the histologic findings in a serially sectioned temporal bone, from a patient who succumbed to brain abscess secondary to malignant external otitis. This material gives us information regarding: 1) the mechanisms of invasion; 2) the details of the routes of spread; 3) the process by which normally resistant bone is destroyed; and 4) the sequential events which may account for the phenomenon of late complications in apparently cured patients. The pathologic process is contrasted to that of pneumococcal petrositis. CASE REPORT
The patient (AD), a 76-year-old Caucasian male. was well until August 1 of the year of his death, when while at the beach a wave hit him in the left ear and he experienced some pain. He continued to have pain and saw his doctor who placed him on ear drops and antibiotics. The disease continued and an otologist diagnosed severe external otitis. He was treated medically with drops and antibiotics. Abscesses of the canal skin were biopsied and curetted twice early in October. On August 17, a modified radical mastoidectomy was performed. This was revised November 1. A facial paralysis had developed between the two procedures. At the time of the second procedure, the nerve was found to be intact, but surrounded by an inflammatory process. The mandibular fossa was bathed in pus. Cultures grew Pseudomonas. The patient was transferred to the University of Florida hospital with residual severe pain in his left ear and jaw. On examination, a left posterior auricular sinus was present (previous site of Penrose drain). The mastoid cavity and external canal contained granular tissue and purulent exudate. The external ear was normal. There was a complete left facial naralvsis. He was identified as being midlv diabetic, a condition which was controllable with diet and oral medications. On November 22. surgery was performed. The ear was debrided of all granulation tissue. Culture taken at that time grew Pseudomonas. He gradually improved. The ear was drying and gran-
715
ulations resolving. On approximately December 15,24 days after his last surgery and 3~ months since the onset of his disease, an insidious beginning of a changing mental status occurred. This consisted of occasional disorientation occurring at night. Neurologic consultants felt this to be normal for the patient's age. There were no other new neurological abnormalities. The mental status defect became more prominent and a brain scan was ordered. This showed some increased activity in the left posterotemporo-occipital area, consistent with recent aural surgery. Neurosurgical consultants recommended skull xrays and an electroencephalogram. On December 27 at 4:00 AM, the patient was found to be unresponsive and flaccid. A lumbar puncture produced grossly purulent spinal fluid. A cerebral angiogram was performed and revealed a left temporal lobe mass. Evacuation of a left temporal lobe brain abscess was performed. The patient expired on December 30. An autopsy was performed four hours after death. The temporal bones were removed and prepared for processing. 0 On removal of the calvarium there was some serous drainage, but no epidural hemorrhage. The brain had a normal shape. There was clouding of the meninges. In the left temporal area there was softening and several hemorrhagic sites were noted corresponding to the sites of surgical drainage. Examination of the inside of the skull revealed no drainage or opening from the temporal and mastoid bones into the cranial cavity. The transverse and sigmoid sinuses were examined and there appeared to be clotted blood of antemortem nature in the left sigmoid sinus. Microscopic evaluation of the brain showed a small left temporal lobe abscess, with focal meningitis and ventriculitis, secondary to a rupture of the small brain abscess into the left lateral ventrical. TEMPORAL BONE FINDINGS
Left Ear. The middle ear cavity had minimal submucosal fibrosis, but no evidence of active bone destruction (Fig. 1 ). In the inner ear there was no inflam-
-The temporal bones were processed In the Otolaryngology Temporal Bone Laboratory the University ot Florida School ot Medicine, Gainesville, Florida. '
Downloaded from aor.sagepub.com at Kungl Tekniska Hogskolan / Royal Institute of Technology on September 2, 2015
716
KOHUT-LINDSAY
Fig. 1. A) Stapes footplate. There is minimal submucosal thickening over the stapes (a). B) Round window membrane. There is no evidence of an inflammatory process in the perilymphatic space (b). On the external surface of the round window membrane, thickening and submucosal fibrosis has occurred.
Fig. 2. Petrous apex (low power) adjacent to the superior semicircular canal (a). Centrally (b) there is an area of a fibrotic inflammatory process. Destruction of trabecular and compact bone has occurred at the periphery of this area. There is thickening of the mucosa within pneumatic spaces (c). The lumen of all pneumatic spaces are free of pus; some contain protein debris. The semicircular canal contains no inflammatory debris. (X2.5)
Downloaded from aor.sagepub.com at Kungl Tekniska Hogskolan / Royal Institute of Technology on September 2, 2015
EXTERNAL OTITIS
717
Fig. 3. A) Perilymphatic duct showing minimal subepithelial thickening in its midportion. B) Direct extension of the fibroinflammatory process to the duct's widened portion (a). There is in this area marrow space involvement (b). A pneumatic space ( c) has subepithelial fibrosis, but no lumenal purulent material. An area of compact bone destruction is visible (d).
matory involvement of the membranous cochlea, vestibule, or semicircular canals (Figs. 1 and 2). At the midportion of the cochlear duct, a subepithelial inflammatory process was observed. This process extended from the duct's midportion towards its arachnoid orifice. Near the external dural aperture of the cochlear duct at its widened portion, direct communication occurred with a fibrous nonpurulent type of inflammatory process. There was active bone destruction at the periphery of this process (Fig. 3). The dura adjacent to this area was slightly thickened, but there was no evidence of direct spread to the central nervous system structures. OTIC CAPSULE AND PNEUMATIC BONE
The residual defect of the mastoid cavity was filled with connective tissue, in some areas covered with a thin layer
of squamous epithelium. This fibrous process extended superiorly to the apex adjacent to the arcuate eminence (Fig. 2), inferior-medially to the perilymphatic duct (Fig. 3), and anteriorly towards the middle ear space. In the apical region new bone formation had occurred, presumably forming the new barrier between the fibrous process and the middle fossa. The mode of bone destruction in all areas appeared to be by osteoblastic and osteoclastic activity and occurred at the periphery of the fibrous tissue. The compact bone was widely involved in this destructive process. Pneumatic trabeculae were not widely destroyed in any coalescent process. The lumen of the pneumatic spaces contained proteinaceous debris, but did not contain elements of an inflammatory process. The mucosa lining the pneumatic spaces was commonly thickened
Downloaded from aor.sagepub.com at Kungl Tekniska Hogskolan / Royal Institute of Technology on September 2, 2015
718
KOHUT-LINDSAY
Fig. 4. Destruction of compact bone (a). Adjacent pneumatic spaces (b). There is submucosal thickening of the pneumatic spaces and subepithelial new bone formation ( c ). The bony walls of some pneumatic spaces have been destroyed, but the mucosa remains intact (d).
and involved in a fibrous process similar to that seen in the larger destructive areas. The pneumatic spaces were rarely obliterated by this process. The bony walls of the pneumatic spaces often contained submucosal new bone or osteoid. In some areas the pneumatic spaces were separated by new bone from the areas of connective tissue. Bone destruction had occurred at the periphery of these areas of fibrous tissue, involving both endochrondral and periosteal bone (Fig. 4). The bone destruction extended directly to the facial nerve. The nerve was partially destroyed and replaced with fibrous tissue. DISCUSSION
This process should be distinguished from the process seen in pneumococcal petrositis in which the lumen of the pneumatic space contains exudate and the bone destruction involves mainly trabecular walls, with a protective fibrous process at the periphery of the abscess formation.' In this instance (pneumococcal petrositis), no bone destruction occurs at the periphery of the protective fibrous walls.
Ordinarily in pneumococcal petrositis, as fibrosis occurs the bone destruction stops, but in malignant external otitis the bone destruction continues in the areas of mature fibrosis. In pneumococcal abscess formation, the destruction process is central about the abscess, with a protective reactive fibrosis forming peripherally in which there is little or no bone destruction. Whereas, in malignant external otitis, the striking bone destruction is in the peripheral fibrous tissue which extends on a broad front without respect to pneumatic or marrow spaces, or compact bone. The spread of the process is facilitated by the pneumatic space submucosa. This is evidenced by the observation that in any major area of development, there is submucosal pneumatic space involvement. The opposite is not true, that is, there is never isolated marrow space involvement. The process is probably dependent on vascular spread and although the marrow spaces are vascularly connected, the histologic evidence suggests that the submucosal blood sup-
Downloaded from aor.sagepub.com at Kungl Tekniska Hogskolan / Royal Institute of Technology on September 2, 2015
EXTERNAL OTITIS
ply of the pneumatic space provides the major pathway for extension. It is interesting to note some differences between the findings in this temporal bone and those identified by Meltzer and Keleman." In their specimen there was extensive involvement of diploic bone, the petrous tip was transformed into a large abscess, trabecular destruction was prominent and the pneumatic spaces were filled with pus or in other areas obliterated.
The striking thing in this temporal bone specimen is the lack of destruction of the pneumatic spaces with marked widespread bone distribution outside of the pneumatic spaces without particular involvement of diploic bone. It seems that the compact bone in malignant external otitis is more susceptible than the bone of the pneumatic spaces. This is quite the opposite to the process seen in pneumococcal petrositis where the abscess forms by breaking down the pneumatic spaces with the formation of a protective fibrosis and new bone formation in the adjacent marrow spaces/ The presence of a small brain abscess in the temporal lobe suggests that the otitic process had spread to the brain from the middle fossa surface of the temporal bone. At autopsy, no defect was detected on the temporal bone surface. On histologic evaluation of the apex, new bone formation had formed, presumably developing a new barrier to the middle fossa. This would suggest that the central spread of the infection had occurred early in the disease process, allowing for new bone formation to heal the site from which the infection had extended to the brain. The slow rate of growth in some brain abscesses (Pseudomonas) is consistent with this proposed mechanism.
