ttp://www.bsava.com
PAPER
Diagnostic value of MRI in dogs with inflammatory nasal disease A. R. R. Furtado, A. Caine* and M. E. Herrtage† Centre for Small Animal Studies, Animal Health Trust, Kentford, Suffolk CB8 7UU *Department of Diagnostic Imaging, Dick White Referrals, Six Mile Bottom, Suffolk CB8 0UH †Department of Veterinary Medicine, University of Cambridge, Cambridge CB3 0ES
OBJECTIVES: To determine the value of low-field magnetic resonance imaging in differentiating sino-nasal aspergillosis from lymphoplasmacytic rhinitis in dogs. METHODS: A retrospective study of 41 dogs (25 with lymphoplasmacytic rhinitis and 16 with sino-nasal aspergillosis) that underwent magnetic resonance imaging scan of the nasal cavity was conducted. On magnetic resonance imaging, turbinate destruction was classified as mild, moderate or severe. The cribriform plate and vomer destruction were classified as present or absent. The intensity of fluid accumulation and turbinates was classified on T1-weighted and T2-weighted images as hypointense, hyperintense and isointense based on the brightest area on the same slice. RESULTS: Turbinate destruction was significantly (P=0·005) associated with sino-nasal aspergillosis. On T1-weighted images, sino-nasal aspergillosis was associated with turbinate hyperintensity, while lymphoplasmacytic rhinitis was significantly (P=0·007) associated with hypointensity. On T2-weighted images, this feature was shown not to be relevant. CLINICAL SIGNIFICANCE: This study has demonstrated that turbinate destruction is the most reliable feature to differentiate sino-nasal aspergillosis from lymphoplasmacytic rhinitis and that T1-weighted image was the most useful sequence.
Journal of Small Animal Practice (2014) 55, 359–363 DOI: 10.1111/jsap.12223 Accepted: 11 March 2014; Published online: 23 April 2014
INTRODUCTION Respiratory diseases are considered to be common and important causes of morbidity and mortality in animals and humans mainly because the respiratory tract is in direct contact with the physical environment and is exposed to airborne microorganisms such as viruses, bacteria, fungi and parasites. Sino-nasal aspergillosis (SNA) and lymphoplasmacytic rhinitis (LPR) are the most common inflammatory nasal diseases in the dog and a correct diagnosis is essential for accurate prognosis and treatment planning. First described in 1987 (Burgener et al. 1987), LPR is characterised by the histopathological appearance of inflammation of the nasal mucosa dominated by lymphocytes and plasma cells. It is typically not a destructive disease but occasionally turbinate destruction is present mimicking SNA (Kuehn 2009). Journal of Small Animal Practice
•
Vol 55
•
July 2014
•
The accurate diagnosis of both diseases remains challenging and no single test is definitive. The diagnosis of SNA is based on two or three cumulative positive test findings with appropriate clinical signs and in contrast LPR diagnosis is made by exclusion (Sullivan 1987, Harcourt-Brown 2006). Non-invasive diagnostic techniques such as magnetic resonance imaging (MRI) are now considered to have good resolution to identify early nasal disease, as they provide a thorough assessment of the nasal cavity and sinuses and provide superior insight to the nature and extent of the disease. MRI in canine nasal disease is considered more sensitive than computed tomography (CT) for soft tissue changes; however, there is no demonstrable difference between CT and MRI for diagnosing nasal cavity mycosis (Saunders et al. 2004). The aim of this study was to determine the most common MRI features of inflammatory nasal diseases and to determine if
© 2014 British Small Animal Veterinary Association
359
A. R. R. Furtado et al.
there are statistical associations in MRI abnormalities that could aid in the differentiation of SNA and LPR in dogs.
MATERIALS AND METHODS Medical records of dogs that underwent MRI because of nasal disease between February 2002 and December 2009 were reviewed. Images were qualitatively characterised with regard to the level of destruction and comparative intensity. Dogs diagnosed with inflammatory or fungal nasal disease were included in the study if results of nasal MRI were available. Dogs were excluded from the study if MRI results were not available, or if the final diagnosis was presence of foreign body, parasitic infection, trauma, neoplasia or if nasal disease was not present. LPR was diagnosed if inflammatory changes were confirmed by a board-certified histopathologist and there was no other associated cause for the nasal disease. SNA was confirmed by at least three positive ancillary diagnostic tests (radiography, visualization of fungal colonies via rhinoscopy, serology, culture and/or histopathology). MRI was performed under general anaesthesia, using a veterinary MRI unit (Vet-MR Grande; Esaote) incorporating an open 0·25-T permanent magnet. The dog’s head was placed in a dual-phase-array coil (143 to 158 mm). T1-weighted (T1W) and T2-weighted (T2W) images of the nasal cavities and paranasal sinuses were acquired in the dorsal, transverse and sagittal planes. The scans were blind-reviewed by the authors over a period of 1 week in several batches with patients in roughly chronological order. If a patient had more than one scan, the one used for the study was always the earliest examination. Vomer, cribriform plate and turbinate destruction were classified as present or absent. Turbinate destruction, when present, was classified as mild (50% destruction). This classification was done according to the transverse slice that appeared to show the greatest degree of destruction. The localisation of the lesions (right side, left side and bilateral) was also documented. The involvement of the frontal sinus (fluid and granulomas) was also classified as present or absent. The intensity of turbinates and fluid accumulation was examined on dorsal T1W (comparing with muscle) and on T2W (comparing with periorbital fat and brain) scans. These results were classified as unknown (when the brain, fat or muscle was not visible), hypointense, hyperintense and isointense based on the brightest area on the same slice. The intensity of the turbinates was considered on the most affected transverse slice where the turbinates appeared symmetrical on both sides. Fluid accumulation was defined when there was a localised obstruction of more than 5% of one of the sides of the nasal passages. Differentiating mucus from fungal colonies is quite difficult; so all the secretions seen were classified under the variable fluid. Statistical analysis of the data was performed using SPSS® 17·0 (SPSS Inc.). The variables were all cross-tabulated with 360
the variables LPR and SNA, and for each cross-tabulation nonparametric tests were used. The level of significance accepted was P
PAPER
Diagnostic value of MRI in dogs with inflammatory nasal disease A. R. R. Furtado, A. Caine* and M. E. Herrtage† Centre for Small Animal Studies, Animal Health Trust, Kentford, Suffolk CB8 7UU *Department of Diagnostic Imaging, Dick White Referrals, Six Mile Bottom, Suffolk CB8 0UH †Department of Veterinary Medicine, University of Cambridge, Cambridge CB3 0ES
OBJECTIVES: To determine the value of low-field magnetic resonance imaging in differentiating sino-nasal aspergillosis from lymphoplasmacytic rhinitis in dogs. METHODS: A retrospective study of 41 dogs (25 with lymphoplasmacytic rhinitis and 16 with sino-nasal aspergillosis) that underwent magnetic resonance imaging scan of the nasal cavity was conducted. On magnetic resonance imaging, turbinate destruction was classified as mild, moderate or severe. The cribriform plate and vomer destruction were classified as present or absent. The intensity of fluid accumulation and turbinates was classified on T1-weighted and T2-weighted images as hypointense, hyperintense and isointense based on the brightest area on the same slice. RESULTS: Turbinate destruction was significantly (P=0·005) associated with sino-nasal aspergillosis. On T1-weighted images, sino-nasal aspergillosis was associated with turbinate hyperintensity, while lymphoplasmacytic rhinitis was significantly (P=0·007) associated with hypointensity. On T2-weighted images, this feature was shown not to be relevant. CLINICAL SIGNIFICANCE: This study has demonstrated that turbinate destruction is the most reliable feature to differentiate sino-nasal aspergillosis from lymphoplasmacytic rhinitis and that T1-weighted image was the most useful sequence.
Journal of Small Animal Practice (2014) 55, 359–363 DOI: 10.1111/jsap.12223 Accepted: 11 March 2014; Published online: 23 April 2014
INTRODUCTION Respiratory diseases are considered to be common and important causes of morbidity and mortality in animals and humans mainly because the respiratory tract is in direct contact with the physical environment and is exposed to airborne microorganisms such as viruses, bacteria, fungi and parasites. Sino-nasal aspergillosis (SNA) and lymphoplasmacytic rhinitis (LPR) are the most common inflammatory nasal diseases in the dog and a correct diagnosis is essential for accurate prognosis and treatment planning. First described in 1987 (Burgener et al. 1987), LPR is characterised by the histopathological appearance of inflammation of the nasal mucosa dominated by lymphocytes and plasma cells. It is typically not a destructive disease but occasionally turbinate destruction is present mimicking SNA (Kuehn 2009). Journal of Small Animal Practice
•
Vol 55
•
July 2014
•
The accurate diagnosis of both diseases remains challenging and no single test is definitive. The diagnosis of SNA is based on two or three cumulative positive test findings with appropriate clinical signs and in contrast LPR diagnosis is made by exclusion (Sullivan 1987, Harcourt-Brown 2006). Non-invasive diagnostic techniques such as magnetic resonance imaging (MRI) are now considered to have good resolution to identify early nasal disease, as they provide a thorough assessment of the nasal cavity and sinuses and provide superior insight to the nature and extent of the disease. MRI in canine nasal disease is considered more sensitive than computed tomography (CT) for soft tissue changes; however, there is no demonstrable difference between CT and MRI for diagnosing nasal cavity mycosis (Saunders et al. 2004). The aim of this study was to determine the most common MRI features of inflammatory nasal diseases and to determine if
© 2014 British Small Animal Veterinary Association
359
A. R. R. Furtado et al.
there are statistical associations in MRI abnormalities that could aid in the differentiation of SNA and LPR in dogs.
MATERIALS AND METHODS Medical records of dogs that underwent MRI because of nasal disease between February 2002 and December 2009 were reviewed. Images were qualitatively characterised with regard to the level of destruction and comparative intensity. Dogs diagnosed with inflammatory or fungal nasal disease were included in the study if results of nasal MRI were available. Dogs were excluded from the study if MRI results were not available, or if the final diagnosis was presence of foreign body, parasitic infection, trauma, neoplasia or if nasal disease was not present. LPR was diagnosed if inflammatory changes were confirmed by a board-certified histopathologist and there was no other associated cause for the nasal disease. SNA was confirmed by at least three positive ancillary diagnostic tests (radiography, visualization of fungal colonies via rhinoscopy, serology, culture and/or histopathology). MRI was performed under general anaesthesia, using a veterinary MRI unit (Vet-MR Grande; Esaote) incorporating an open 0·25-T permanent magnet. The dog’s head was placed in a dual-phase-array coil (143 to 158 mm). T1-weighted (T1W) and T2-weighted (T2W) images of the nasal cavities and paranasal sinuses were acquired in the dorsal, transverse and sagittal planes. The scans were blind-reviewed by the authors over a period of 1 week in several batches with patients in roughly chronological order. If a patient had more than one scan, the one used for the study was always the earliest examination. Vomer, cribriform plate and turbinate destruction were classified as present or absent. Turbinate destruction, when present, was classified as mild (50% destruction). This classification was done according to the transverse slice that appeared to show the greatest degree of destruction. The localisation of the lesions (right side, left side and bilateral) was also documented. The involvement of the frontal sinus (fluid and granulomas) was also classified as present or absent. The intensity of turbinates and fluid accumulation was examined on dorsal T1W (comparing with muscle) and on T2W (comparing with periorbital fat and brain) scans. These results were classified as unknown (when the brain, fat or muscle was not visible), hypointense, hyperintense and isointense based on the brightest area on the same slice. The intensity of the turbinates was considered on the most affected transverse slice where the turbinates appeared symmetrical on both sides. Fluid accumulation was defined when there was a localised obstruction of more than 5% of one of the sides of the nasal passages. Differentiating mucus from fungal colonies is quite difficult; so all the secretions seen were classified under the variable fluid. Statistical analysis of the data was performed using SPSS® 17·0 (SPSS Inc.). The variables were all cross-tabulated with 360
the variables LPR and SNA, and for each cross-tabulation nonparametric tests were used. The level of significance accepted was P
