Acta Oto-Laryngologica. 2015; Early Online, 1–5
ORIGINAL ARTICLE
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 06/09/15 For personal use only.
Correlation of computed tomography, echo-planar diffusion-weighted magnetic resonance imaging and surgical outcomes in middle ear cholesteatoma
MURAT SONGU1, CANAN ALTAY2, KAZIM ONAL1, SECIL ARSLANOGLU1, MUSTAFA KORAY BALCI1, MEHMET UCAR1, EJDER CIGER1 & AYLIN KOPAR1 1
Department of Otorhinolaryngology, Izmir Katip Celebi University Ataturk Training and Research Hospital, Izmir and Department of Radiology, Izmir Dokuz Eylul University Faculty of Medicine, Izmir, Turkey
2
Abstract Conclusion: Echo-planar diffusion-weighted magnetic resonance imaging (DW MRI) is more reliable than high-resolution computed tomography (HRCT) in predicting the presence and localization of cholesteatoma before tympanomastoid surgery. Objectives: To evaluate the diagnostic accuracy of HRCT and echo-planar DW MRI in the detection and localization of cholesteatoma. Methods: Fifty-nine patients were prospectively included in this study. Patients with suspected primary cholesteatoma were evaluated by HRCT and echo-planar DW MRI before tympanomastoid surgery. Radiological findings were correlated with intraoperative findings. Results: HRCT and echo-planar DW MRI accurately predicted the presence or absence of cholesteatoma in 40/59 (67.8%) and 52/59 (88.1%) patients, respectively. The sensitivity, specificity, and positive and negative predictive values of HRCT were 68.97%, 66.67%, 66.67%, and 68.97%, respectively. However, sensitivity, specificity, and positive and negative predictive values of echo-planar DW MRI were 85.71%, 90.32%, 88.89%, and 87.50%, respectively.
Keywords: Mastoidectomy, tympanomastoid surgery, HRCT, high-resolution computed tomography
Introduction In the detection and description of the extension of a suspected middle ear cholesteatoma, thin-section, high-resolution computed tomography (HRCT) is still considered the imaging method of choice [1]. It can demonstrate ossicular erosion and possible complications such as tegmental disruption and fistulization through the lateral semicircular canal. However, HRCT has significant limitations in distinguishing cholesteatoma from granulation tissue, cholesterol granuloma, and other soft tissue changes [2,3]. Magnetic resonance imaging (MRI) may provide additional information for the evaluation of cholesteatoma extension and may lead to a more accurate
diagnosis [4]. Apart from an excellent soft tissue contrast, MRI offers the possibility of using various pulse sequences, including echo-planar imaging diffusion-weighted imaging (DWI), and temporal bone MRI with administration of intravenous contrast material. DWI is much faster than delayed post contrast T1-weighted imaging and does not require contrast administration [5]. In the past few years, MRI including DWI has gained increasing importance in the detection of acquired middle ear cholesteatoma [6]. The aim of this prospective study was to evaluate the accuracy of HRCT and DWI in the detection of primary cholesteatoma before tympanomastoid surgery.
Correspondence: Murat Songu MD, Associate Professor, Department of Otorhinolaryngology, Izmir Katip Celebi University Ataturk Training and Research Hospital, Izmir, Turkey. Tel: +90 5326858484. Fax: +90 2322431530. E-mail: [email protected]
(Received 22 September 2014; accepted 16 February 2015) ISSN 0001-6489 print/ISSN 1651-2251 online 2015 Informa Healthcare DOI: 10.3109/00016489.2015.1021931
2
M. Songu et al.
Material and methods
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 06/09/15 For personal use only.
Study design This prospective study was approved by the local Institutional Review Board (112/28122012). Patients admitted to our tertiary center with clinically suspected primary cholesteatoma between December 2008 and February 2013 constituted the study group. The information on clinical, radiologic, and surgical findings of the ear disease was collected from the database for all patients. Cases who had undergone tympanomastoidectomy for chronic otitis media with or without cholesteatoma were excluded from the study. All patients underwent HRCT and DWI before their operation. Data were recorded for the presence or absence of cholesteatoma.
Outcome parameters A radiologist who was blinded to the clinical condition of the patients reported all HRCT and DWI findings. The interpretation of HRCT was based on the presence of intratympanic non-dependent soft tissue density, bone expansion and thinning, and bone
erosions involving the ossicles and adjacent structures (Figure 1a). On DWI, a consensus agreement was made for the evaluation of primary or residual cholesteatomas, which were marked either as positive (cholesteatoma present) or as negative (cholesteatoma absent) (Figure 1b, c). Interpretation relied on the signal intensity of the soft tissue extension on nonenhanced T1-weighted images and T2-weighted images. All CT scans were performed on a 64-row multislice CT scanner (Toshiba Aquilion, Tochigi, Japan) using axial, coronal, and sagittal images. MRI was performed on a 1.5 T superconductive unit (GE, Milwaukee, WI, USA) with use of the eight-channel head and neck coil. Routine temporal bone MRI was performed with axial 5 mm thick turbo-spin-echo T2-weighted images (TR 4.300 ms, TE 85 ms, FOV 250 250, matrix 128 128) and axial and coronal 5 mm thick turbo-spin-echo T1-weighted images (TR 6000 ms, TE 85 ms, FOV 250 250, matrix 128 128) with and without contrast material administration. After the injection of gadolinium (0.1 mmol/kg), axial and coronal T1-weighted spinecho images were acquired with the same parameters as those used before administration of contrast
a
15.6 mm
b
c
13.3 mm 16.7 mm
Figure 1. A 59-year-old female patient with cholesteatoma in the left attic cavity and mastoid antrum. On high-resolution CT image (a), a hypodense soft tissue mass in left middle ear cavity is seen. On diffusion-weighted MR image (b), the lesion is hyperintense compared with the adjacent tissues. Hypointensity on the ADC map (c) regarding apparent diffusion decreased.
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 06/09/15 For personal use only.
CT, EP-DW MRI, and surgical outcomes in cholesteatoma material. DWI was performed by the echo planar imaging (EPI) technique (TR 2745 ms, TE 68 ms, FOV 250 250 mm, matrix 140 108, slice thickness 5 mm). The b-values were accepted at 0 and 1000 s/mm2. After the tympanomastoid surgery, histological confirmation was obtained in all patients. Truepositive (TP) patients were cases with cholesteatoma found during surgery and previously diagnosed by the imaging technique. False-positive (FP) patients were cases with no cholesteatoma found during surgery, but previously diagnosed by the imaging technique. True-negative (TN) patients were cases with no cholesteatoma found during surgery and previously diagnosed as having no cholesteatoma by the imaging technique. False-negative (FN) patients were cases with cholesteatoma found during surgery, but not previously diagnosed by the imaging technique. Statistical analyses Data were analyzed using the Statistical Package for Social Sciences 19.0 for Windows (SPSS Inc., Chicago, IL, USA). Independent-samples t test and Mann–Whiney U test were used to compare independent groups. One-way analysis of variance (ANOVA) was used to compare groups of independent continuous variables and Bonferroni post hoc analysis was used for multiple comparison tests. The partial correlation test was used to calculate correlation coefficients. Data are expressed as mean ± SD. All differences associated with a chance probability of 0.05 or less were considered statistically significant.
3
Table I. Correlation between HRCT and surgical findings. Status
Positive on HRCT (n = 30)
Negative on HRCT (n = 29)
Positive on True-positive (n = 20) False-negative (n = 9) surgery (n = 37) Size ‡4 mm (n = 17)
Size ‡4 mm (n = 3)
Size
ORIGINAL ARTICLE
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 06/09/15 For personal use only.
Correlation of computed tomography, echo-planar diffusion-weighted magnetic resonance imaging and surgical outcomes in middle ear cholesteatoma
MURAT SONGU1, CANAN ALTAY2, KAZIM ONAL1, SECIL ARSLANOGLU1, MUSTAFA KORAY BALCI1, MEHMET UCAR1, EJDER CIGER1 & AYLIN KOPAR1 1
Department of Otorhinolaryngology, Izmir Katip Celebi University Ataturk Training and Research Hospital, Izmir and Department of Radiology, Izmir Dokuz Eylul University Faculty of Medicine, Izmir, Turkey
2
Abstract Conclusion: Echo-planar diffusion-weighted magnetic resonance imaging (DW MRI) is more reliable than high-resolution computed tomography (HRCT) in predicting the presence and localization of cholesteatoma before tympanomastoid surgery. Objectives: To evaluate the diagnostic accuracy of HRCT and echo-planar DW MRI in the detection and localization of cholesteatoma. Methods: Fifty-nine patients were prospectively included in this study. Patients with suspected primary cholesteatoma were evaluated by HRCT and echo-planar DW MRI before tympanomastoid surgery. Radiological findings were correlated with intraoperative findings. Results: HRCT and echo-planar DW MRI accurately predicted the presence or absence of cholesteatoma in 40/59 (67.8%) and 52/59 (88.1%) patients, respectively. The sensitivity, specificity, and positive and negative predictive values of HRCT were 68.97%, 66.67%, 66.67%, and 68.97%, respectively. However, sensitivity, specificity, and positive and negative predictive values of echo-planar DW MRI were 85.71%, 90.32%, 88.89%, and 87.50%, respectively.
Keywords: Mastoidectomy, tympanomastoid surgery, HRCT, high-resolution computed tomography
Introduction In the detection and description of the extension of a suspected middle ear cholesteatoma, thin-section, high-resolution computed tomography (HRCT) is still considered the imaging method of choice [1]. It can demonstrate ossicular erosion and possible complications such as tegmental disruption and fistulization through the lateral semicircular canal. However, HRCT has significant limitations in distinguishing cholesteatoma from granulation tissue, cholesterol granuloma, and other soft tissue changes [2,3]. Magnetic resonance imaging (MRI) may provide additional information for the evaluation of cholesteatoma extension and may lead to a more accurate
diagnosis [4]. Apart from an excellent soft tissue contrast, MRI offers the possibility of using various pulse sequences, including echo-planar imaging diffusion-weighted imaging (DWI), and temporal bone MRI with administration of intravenous contrast material. DWI is much faster than delayed post contrast T1-weighted imaging and does not require contrast administration [5]. In the past few years, MRI including DWI has gained increasing importance in the detection of acquired middle ear cholesteatoma [6]. The aim of this prospective study was to evaluate the accuracy of HRCT and DWI in the detection of primary cholesteatoma before tympanomastoid surgery.
Correspondence: Murat Songu MD, Associate Professor, Department of Otorhinolaryngology, Izmir Katip Celebi University Ataturk Training and Research Hospital, Izmir, Turkey. Tel: +90 5326858484. Fax: +90 2322431530. E-mail: [email protected]
(Received 22 September 2014; accepted 16 February 2015) ISSN 0001-6489 print/ISSN 1651-2251 online 2015 Informa Healthcare DOI: 10.3109/00016489.2015.1021931
2
M. Songu et al.
Material and methods
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 06/09/15 For personal use only.
Study design This prospective study was approved by the local Institutional Review Board (112/28122012). Patients admitted to our tertiary center with clinically suspected primary cholesteatoma between December 2008 and February 2013 constituted the study group. The information on clinical, radiologic, and surgical findings of the ear disease was collected from the database for all patients. Cases who had undergone tympanomastoidectomy for chronic otitis media with or without cholesteatoma were excluded from the study. All patients underwent HRCT and DWI before their operation. Data were recorded for the presence or absence of cholesteatoma.
Outcome parameters A radiologist who was blinded to the clinical condition of the patients reported all HRCT and DWI findings. The interpretation of HRCT was based on the presence of intratympanic non-dependent soft tissue density, bone expansion and thinning, and bone
erosions involving the ossicles and adjacent structures (Figure 1a). On DWI, a consensus agreement was made for the evaluation of primary or residual cholesteatomas, which were marked either as positive (cholesteatoma present) or as negative (cholesteatoma absent) (Figure 1b, c). Interpretation relied on the signal intensity of the soft tissue extension on nonenhanced T1-weighted images and T2-weighted images. All CT scans were performed on a 64-row multislice CT scanner (Toshiba Aquilion, Tochigi, Japan) using axial, coronal, and sagittal images. MRI was performed on a 1.5 T superconductive unit (GE, Milwaukee, WI, USA) with use of the eight-channel head and neck coil. Routine temporal bone MRI was performed with axial 5 mm thick turbo-spin-echo T2-weighted images (TR 4.300 ms, TE 85 ms, FOV 250 250, matrix 128 128) and axial and coronal 5 mm thick turbo-spin-echo T1-weighted images (TR 6000 ms, TE 85 ms, FOV 250 250, matrix 128 128) with and without contrast material administration. After the injection of gadolinium (0.1 mmol/kg), axial and coronal T1-weighted spinecho images were acquired with the same parameters as those used before administration of contrast
a
15.6 mm
b
c
13.3 mm 16.7 mm
Figure 1. A 59-year-old female patient with cholesteatoma in the left attic cavity and mastoid antrum. On high-resolution CT image (a), a hypodense soft tissue mass in left middle ear cavity is seen. On diffusion-weighted MR image (b), the lesion is hyperintense compared with the adjacent tissues. Hypointensity on the ADC map (c) regarding apparent diffusion decreased.
Acta Otolaryngol Downloaded from informahealthcare.com by Nyu Medical Center on 06/09/15 For personal use only.
CT, EP-DW MRI, and surgical outcomes in cholesteatoma material. DWI was performed by the echo planar imaging (EPI) technique (TR 2745 ms, TE 68 ms, FOV 250 250 mm, matrix 140 108, slice thickness 5 mm). The b-values were accepted at 0 and 1000 s/mm2. After the tympanomastoid surgery, histological confirmation was obtained in all patients. Truepositive (TP) patients were cases with cholesteatoma found during surgery and previously diagnosed by the imaging technique. False-positive (FP) patients were cases with no cholesteatoma found during surgery, but previously diagnosed by the imaging technique. True-negative (TN) patients were cases with no cholesteatoma found during surgery and previously diagnosed as having no cholesteatoma by the imaging technique. False-negative (FN) patients were cases with cholesteatoma found during surgery, but not previously diagnosed by the imaging technique. Statistical analyses Data were analyzed using the Statistical Package for Social Sciences 19.0 for Windows (SPSS Inc., Chicago, IL, USA). Independent-samples t test and Mann–Whiney U test were used to compare independent groups. One-way analysis of variance (ANOVA) was used to compare groups of independent continuous variables and Bonferroni post hoc analysis was used for multiple comparison tests. The partial correlation test was used to calculate correlation coefficients. Data are expressed as mean ± SD. All differences associated with a chance probability of 0.05 or less were considered statistically significant.
3
Table I. Correlation between HRCT and surgical findings. Status
Positive on HRCT (n = 30)
Negative on HRCT (n = 29)
Positive on True-positive (n = 20) False-negative (n = 9) surgery (n = 37) Size ‡4 mm (n = 17)
Size ‡4 mm (n = 3)
Size
