|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pediatric Imaging |
1 From the Departments of Radiology (J.P., S.R.) and Oto-Rhino-Laryngology (A.P.), Helsinki University Hospital, Haartmaninkatu 4, POB 340, FIN-00029 HUS, Finland; Departments of Oto-Rhino-Laryngology (A.J., W.D.B.) and Radiology (C.C., S.R.), Vienna University Hospital, Vienna, Austria; and Department of Health and Disability, National Public Health Institute of Finland, Helsinki (M.H.). Received September 1, 2002; revision requested November 9; final revision received May 23, 2003; accepted June 9. Address correspondence to J.P. (e-mail: johanna.pekkola@hus.fi).
| ABSTRACT |
|---|
|
|
|---|
MATERIALS AND METHODS: Temporal-bone thin-section CT images obtained in 73 patients aged 09 years (20 Austrian, 53 Finnish; 36 female and 37 male patients) were evaluated for the presence of localized hypoattenuating foci in the region of the fissula ante fenestram of the otic capsule. Clinical information collected for all patients was also evaluated. The data were analyzed with a logistic regression model.
RESULTS: Hypoattenuating areas in the region of the fissula ante fenestram were observed in 23 of 73 patients (32%). Hypoattenuating foci were substantially more prevalent in patients younger than 3 years than in those 3 years or older (odds ratio, 0.14; 95% CI: 0.04, 0.52; P = .001). The prevalence did not differ between sexes or according to clinical diagnosis. Only three of the 23 patients with hypoattenuating foci had clinical findings suggestive of otosclerosis, and none had osteogenesis imperfecta. After adjustment for age and sex, the finding was more prevalent among the Finnish patients (odds ratio, 5.4; 95% CI: 1.19, 24.52; P = .02) than among the Austrian patients.
CONCLUSION: Hypoattenuating areas in the region of the fissula ante fenestram in the otic capsule at thin-section CT are prevalent among children younger than 3 years in the absence of clinical evidence of otosclerosis or osteogenesis imperfecta and appear in children up to 9 years old.
© RSNA, 2003
Index terms: Computed tomography (CT), in infants and children, 21.12115, 21.12118 Ear, anatomy, 21.92 Ear, CT, 21.12115, 21.12118
| INTRODUCTION |
|---|
|
|
|---|
The normal appearance of the cochlear otic capsule at thin-section CT is described as a sharply defined, homogeneous area of bone attenuation that outlines the lumen of the cochlea (35).
Localized lucent areas in the bony cochlear capsule occur in various pathologic conditions, most typically otosclerosis and osteogenesis imperfecta (35). At thin-section CT, hypoattenuating foci usually seen in the area of the fissula ante fenestram are observed in active otospongiosis and otosclerosis (35). Purely sclerotic changes, which may represent a later development in the disease process, may cause irregularity in the outline of the cochlear capsule and are more difficult to depict (3,4). Osteogenesis imperfecta may cause similar but usually more extensive hypoattenuating changes in the otic capsule (5,6).
Pericochlear hypoattenuating foci at temporal-bone thin-section CT have previously been recognized in adult subjects with normal hearing and no clinical evidence of otosclerosis or osteogenesis imperfecta (3,7).
To our knowledge, no previous studies on pericochlear hypoattenuating foci at thin-section CT in children have been published. Thus, the purpose of this retrospective study was to evaluate the prevalence of localized lucent areas in the cochlear otic capsule at temporal-bone thin-section CT in pediatric patients and correlate the findings with clinical information.
| MATERIALS AND METHODS |
|---|
|
|
|---|
For comparison, 21 consecutive temporal-bone thin-section CT studies obtained in 20 children aged 07 years (eight male and 12 female patients; mean age, 3.2 years ± 2.1) from Vienna University Hospital (Austria) were collected from May 1, 2000, to April 30, 2002.
Institutional review board approval and patient informed consent are not required in Finland or Austria for studies of preexisting image material and clinical records.
Four of the authors, three otorhinolaryngologists (A.P., A.J., W.D.B.) and one radiologist (J.P.), jointly collected data from the medical records of the patients concerning clinical otologic diagnosis at the time of the examination and the reason for referral of the patient for temporal-bone thin-section CT.
Clinical diagnoses included complicated middle-ear infectious disease and/or suspicion of cholesteatoma (27 patients), sensorineural hearing loss or deafness of an infant (18 patients), congenital external canal or auricular malformation (12 patients), conductive hearing loss without explanatory clinical findings (eight patients), and other diagnoses (eight patients). The other diagnoses included trauma in two patients and external canal scarring after a dog bite, facial paresis, craniofacial dysplasia, ossified tympanic membrane, recurrent tympanic membrane polyp, and tinnitus (one patient each).
Imaging and Interpretation
The studies were performed in the transverse plane with a model MX8000 (Marconi, Cleveland, Ohio) fourdetector row helical CT scanner with 4.0 x 0.5-mm collimation (36 studies), a Somatom plus 4 (Siemens, Erlangen, Germany) singledetector row helical CT scanner with 1.5-mm section thickness (19 studies), and a Sekura (Philips, Best, the Netherlands) singledetector row helical CT scanner with 1-mm section thickness (21 studies) (three patients underwent imaging twice). Thin-section images in coronal and other desired planes were created from the transverse imaging data with workstations (FD Trinitron, Sony, Tokyo, Japan or model 445 Pro, Nokia, Espoo, Finland).
With the Marconi and Siemens scanners, window width was 4,000 HU (mean, 900 HU), and with the Philips scanner, window width was 2,000 HU (mean, 400 HU). Images were produced with a high-resolution bone algorithm.
All images were evaluated retrospectively. Two readers (S.R., J.P.) in consensus assessed the presence and extent of hypoattenuating foci in the cochlear otic capsule.
Statistical Analysis
Data were analyzed with a logistic regression model to compute odds ratios with 95% CIs. Age, sex, nationality, and clinical diagnosis were considered to be determinants for pericochlear hypoattenuating foci. Each factor was tested separately in the multivariate analyses. Statistical significance of the covariate terms was tested with the likelihood ratio test and expressed as exact P values. The temporally first studies were used in the statistical analysis for the three patients who underwent imaging twice during the observation period.
| RESULTS |
|---|
|
|
|---|
|
|
|
|
|
|
|
|
| DISCUSSION |
|---|
|
|
|---|
In our series of 73 pediatric patients, one-half of the children younger than 3 years and one-fifth of those aged 39 years were found to have localized pericochlear hypoattenuating foci at temporal-bone thin-section CT. Of the 23 pediatric patients with hypoattenuating foci in our study, only three (aged 6 years 0 months, 7 years 11 months, and 8 years) presented with unexplained conductive hearing loss, and otosclerosis was suspected. None of the 23 patients had osteogenesis imperfecta.
Earlier observations of pericochlear hypoattenuating foci at thin-section CT in adults without clinical evidence of otosclerosis or osteogenesis imperfecta have been published. The hypoattenuating foci were interpreted by the authors as probable normal variants (7) or possible incipient otosclerosis (3). Association with osteoporotic changes in the endosteal layer of the cochlear capsule was also suggested (7).
The region of the fissula ante fenestram is the classic site of otosclerotic plaques. Histologically confirmed otosclerosis, however, is very rare in children younger than 5 years (8). Compared with the most typical site of otosclerotic lesions, the sites of the hypoattenuating foci observed in our patients were more caudal in the transverse plane.
Clinical situations among patients with hypoattenuating foci were variable and included conditions with no known connection to otosclerosis, such as external auditory meatus scarring after a dog bite, congenital external auditory canal atresia, and ossified tympanic membrane for unknown reasons. The greater prevalence of the finding among younger children might suggest a nonpathologic, possibly developmental nature.
Development of the bony otic capsule is complex. The capsule consists of three layers: the thin inner endosteal layer, the middle layer of combined endochondral and intrachondral bone, and the outer periosteal layer, which resembles the periosteal layer of long bones (9,10). The structure of the middle layer is exceptional: It contains calcified hyaline cartilage with islands of true bone in the original cartilaginous lacunae surrounded by primary endochondral bone. The middle layer of the otic capsule is the only location in the human skeleton where this peculiar type of bone appears, and it is the location for the pathologic process of otosclerosis (9,10).
The cartilaginous model of the otic capsule reaches its adult size in the 16th week of fetal life. At that time, the ossification process starts in 14 separate ossification centers, which ultimately fuse into a complex structure that hosts the organs of hearing and balance (9,10). Ossification progresses differently in different parts of the capsule, with the areas adjacent to the fissula ante fenestram and oval window being the last to ossify (911). The process is completed first in the inner and outer layers. The middle layer stays partly cartilaginous in a near-term fetus, when a rapid increase in bone formation occurs and most marrow spaces are replaced by dense bone (911). In a term infant, the whole capsule is almost completely ossified (9,10). In a histologic study, some small bone marrow spaces persisted in the middle layer in a child up to age 15 months (12).
The fissula ante fenestram is a cleft of fibrocartilaginous tissue between the inner and middle ears just anterior to the anterior margin of the oval window. The size and appearance have great individual variance. Usually, a thin quiescent cartilaginous rim surrounds the fissula (13,14) and persists to adulthood. In various histologic studies, aberrant, sometimes bulky masses of histologically active cartilage were observed in the region of the fissula ante fenestram in up to 80% of fetuses and young infants (13,15,16). They have been observed in conjunction with wide fissulae and could represent an attempt to narrow a wide fissula (13,15). These active, ossifying cartilage masses are postulated to be histologically unstable and a possible connection to later development of the otosclerotic process has been speculated (15).
The "incidental" hypoattenuating foci visible at temporal-bone thin-section CT are observed in one of the last areas of the otic capsule to be completely ossified; therefore, they might represent a variation or delay in the capsule ossification process. Another developmental process that could produce a hypoattenuating area in the antefissular region at temporal-bone thin-section CT might be the development of cartilaginous masses.
A reason for the difference in the prevalence of hypoattenuating foci between the Finnish and Austrian patient populations can only be speculated. Vitamin D is required for the normal ossification process; therefore, a vitamin D supplement is uniformly used in young infants in Finland to prevent deficiency states caused by the lack of sunlight during the dark winter months. It might be possible that vitamin D requirements for ossification in endosteal bone differ from those for other types of bone, and a relative lack of vitamin D would delay the bone formation in the endosteal layer among the Finnish patients. Also, because Finland has been relatively isolated for long historic periods compared with middle European countries, genetic differences that affect the ossification process might play a part. Issues related to the imaging technique and scanning parameters can affect visualization of small structures. In our study, three CT scanners were used, with section thicknesses ranging from 0.5 to 1.5 mm. Both the thinnest (0.5-mm) and thickest (1.5-mm) sections were obtained in Finnish patients. Thus, we think it is safe to assume that the difference between Austrian and Finnish patients is not greatly affected by the different imaging techniques.
Hypoattenuating areas visible at thin-section CT in the region of the fissula ante fenestram of the cochlear capsule are highly prevalent among infants and pediatric patients without clinical evidence of the diseases that typically lead to demineralization in the otic capsule. We believe that these findings should be considered normal variants unless there is clinical evidence that clearly points to a pathologic process. The development of imaging methods will show radiologists new details of normal anatomy, and the interpretation of these appearances should be assessed carefully.
In conclusion, hypoattenuating areas in the region of the fissula ante fenestram are seen at thin-section CT in a substantial portion of pediatric patients, but they should not have any pathologic implication without clinical evidence of otosclerosis or osteogenesis imperfecta.
| FOOTNOTES |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. Moser, F. Veillon, H. Sick, and S. Riehm The Hypodense Focus in the Petrous Apex: A Potential Pitfall on Multidetector CT Imaging of the Temporal Bone AJNR Am. J. Neuroradiol., January 1, 2008; 29(1): 35 - 39. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |