Eur. Radiol. 5, 160-164 (1995) © Springer-Verlag 1995

European Radiology

Head and Neck

Non-invasive aspergillosis of the paranasal sinuses: CT and MRI findings Y. Robert 1, O. Lamy 1, D. Chevalier 2, N. R o c o u r t 1, J. Darras 2, J.-J. Piquet 2, L. Lemaitre 1 1Service de Radiologie Ouest, H6pital Claude Huriez, CHU Lille, 1 place de Verdun, F-59037 Lille cedex France 2 Service d'orl, H6pital Claude Huriez, CHU Lille, i place de Verdun, F-59037 Lille cedex, France Received 16 January 1994; Revision received 19 May 1994; Accepted 17 June 1994 der to allow complete surgical removal of the fungal mucus. The purpose of the present study was to review the CT and M R I findings of 20 cases of proven fungal sinusitis, suggestive of the diagnosis.

Methods

Introduction Fungal infection of the paranasal sinuses is an uncomm o n disease, most often related to Aspergillus furnigatus. Aspergillus of the paranasal sinuses is caused by a spore-forming fungus of the class Ascomycetes that occurs world-wide [1, 2]. Fungal spores b e c o m e secondary saprophytic inhabitants of the nose and paranasal sinuses when inhaled. On occasions the fungi b e c o m e pathogenic, and most often aspergillosis appears as a chronic sinusitis that does not resolve with antibiotic therapy [3]. Therefore, its recognition is necessary in or-

Correspondence to: Y. Robert

Twenty patients (13 women, 7 men), aged from 30 to 74 years (mean 45 years), were referred for CT examination because of chronic sinusitis. This was unresolved by medical treatment in all 20 cases, with, in 1 case, headache spreading to the vertex, suggestive of a sphenoid sinus location. Dental root canal obturation in the months preceding the clinical symptoms was noted in 13 patients. In no case was immunocompromised status observed. All the patients underwent CT examination, perf o r m e d on a Elscint 2400, Elscint Elite Plus (Elscint, Israel) or Somatom D R H (Siemens, Germany). The slice thickness was 2 ram, and the table increment 4 mm. Transverse and frontal planes were both used to improve the demonstration of the disease and its extent. For the same reasons iodine contrast material was intravenously injected in 8 cases. Bony and soft tissue window settings were used. The CT scans were analysed for the presence of a soft tissue mass, with areas of increased attenuation, in the paranasal sinuses or in the nasal fossa; calcifications or foreign material of dental origin; and bony modifications (hyperosteosis, bony deformation related to a mass effect or osteolysis). In 2 cases, M R I was performed with a 0.5 T magnet (MR-Max, G E - C G R , Milwaukee, USA), using a head coil. The Tl-weighted images were obtained with a short repetition time (TR) of 550 ms, a short echo time (TE) of 25 ms and two or four excitations. The T2weighted images were obtained with a long T R of 2000 ms, a long T E of 50 ms and two excitations. The slice thickness was 5 m m with an intersection gap of 1 mm. Images were reconstructed with a 224 x 224 matrix. G a d o l i n i u m - D O T A (Guerbet, France) was used in

161

Y. Robert et al.: Non-invasive aspergillosis of the paranasal sinuses Table 1. CT findings Case no.

Age (years)

Sex

Location

Hyperdense retention

Calcification

Dental material

+ +

+

+

+ + +(A) +(A) +(A)

Bony modifications Sclerosis

1 67 F MSO-NC + 2 38 F MS + 3 58 M MSO-NC + 4 64 F MS + 5 40 F MS + 6 30 F MS + 7 33 M MS 8 32 M MS + 9 35 M MS 10 62 F MS-MSO-NC + 11 67 F MS + 12 74 F SS + 13R 34 F MS + 13L 34 F MS-MSO-NC + 14 43 M MS-MSO + 15 45 F MS-MSO-NC + 16 31 F MS + 17 32 F MS-MSO-NC + 18 44 F MS 19 39 M MS-MSO + 20 38 M MS-MSO + R, right; L, left; M, male; F, female; MS, maxillary sinus; MSO, maxillary a Bony erosion related to fungal mass 1 of the cases. The signal intensity was c o m p a r e d with that of the normal mucosa. Final diagnosis was obtained f r o m the surgical and histological data.

Results The results are s u m m a r i s e d in Table 1. The middle m e a tus and maxillary sinus were affected in 19 patients, bilaterally in only 1 of the patients (n -- 20). The fungal mass was located inside the maxillary sinus in 11 cases, extending to the maxillary sinus infundibulum and ostium (which were enlarged) in 7 cases, and extending into the middle meatus in 4 cases. In 2 cases the location was the middle meatus closely related to the maxillary sinus ostium. The sphenoid sinus was involved in 1 case. The sinus cavity was completely (n = 13) or partially (n = 8) filled by soft tissue, in which an area of high attenuation was observed in 18 cases. These hyperdense foci were associated with one or m o r e calcifications in 10 cases and dental material in 13 cases. Both calcification and dental material were observed in 7 cases. D e n tal a m a l g a m (n -- 4) was easily recognized inside the sinus cavity because of its very high attenuation, causing radiating artefacts (Fig. 1 a). Pulp canal sealer (n = 9) was also hyperdense, but no surrounding artefacts were observed, and its margins were sharp and regular (Fig.2). Calcifications were smaller, slightly irregular, and linear or punctiform (Figs. 1 b, 5 a). In 16 cases the maxillary or sphenoid sinus wall thickness was increased as a result of b o n y sclerosis related to chronic infection (Figs. 2, 3). In 2 cases the sinusonasal wall was eroded as a consequence of the fungal mass effect (Fig. 4).

+ + + +

+ +(A) +

+ +

+ + + + + +

+ + +

Erosion

+ + + + + + + +

~a +a

+ + + + sinus ostium; NC, nasal cavity; SS, sphenoid sinus; A, amalgam

On M R I the fungal mass displayed a low signal intensity on T l - w e i g h t e d images c o m p a r e d with the normal m u c o s a surrounding the turbinates. On T2-weighted images the fungal mass displayed a hypointense signal, particularly in the first case. The adjacent soft tissue displayed a hyperintense signal, similar to normal mucosa (Fig.5). A f t e r gadolinium injection, p e r f o r m e d in 1 c a s e (case 17), the fungal mass was not enhanced (Fig. 6).

Discussion

Aspergillus is a ubiquitous fungus which grows in soil, dust, plants and decaying matter, forming spores [1, 4]. These spores are inhaled and usually found as saprophytic inhabitants in the u p p e r airways and in the bronchop u l m o n a r y system. In N o r t h A m e r i c a and E u r o p e aspergillosis is most often related to Aspergillus fumigatus; A. flavus is m o r e frequently found in the Middle East. Others species, such as A. niger or A. oryzae, are u n c o m m o n . Aspergillus b e c o m e s pathogenic when the local or general conditions favour its growth. Non-invasive aspergillosis is related to propitious local anaerobic conditions: impaired mucosa, obstructed sinus cavity subsequent to septal deviation, nasal polyps, infectious and allergic rhinosinusitis [5-7]. Foreign bodies, such as dental material, favour chronic mucosal reaction and Aspergillus d e v e l o p m e n t [8]. Such conditions, noted in 13 of the 21 cases, were the main cause of maxillary sinus non-invasive aspergillosis. G e n e r a l i m m u n o c o m p r o m i s e d states such as diabetes mellitus, aplasia, cancer, or reduced resistance as a result of i m m u n o s u p pressive or oncological drugs, also favour the developm e n t of Aspergillus [1, 9]. Paranasal sinus aspergillosis

162

Y. Robert et al.: Non-invasive aspergillosis of the paranasal sinuses

Fig.2. Aspergillosis of the left maxillary sinus (unenhanced CT study) related to endodontal material (black arrow), which appears as a round hyperdense concretion with regular margins. The endodontal material was surrounded by a hyperdense retention (black arrowhead). Note the thickening of the maxillary sinus walls (small black arrows) in comparison with normal contralateral side Fig.3. Aspergillosis of the sphenoid sinus, revealed by chronic sinusitis and headache spreading to the vertex. Coronal CT scan (bony window setting) shows a markedly increased bony wall thickness (small black arrows) Fig.4. Aspergillosis of the left maxillary sinus (pseudotumoral form). The fungal mass (large black arrow) eroded the medial wall of the maxillary sinus and protruded into the nasal cavity (white arrowhead). Note the bony sclerosis of the posterior wall (black arrowhead) and the calcification (small black arrow). Such associated findings are suggestive of aspergillosis rather than tumour Fig.la, b. Bilateral aspergillosis of the maxillary sinus (unenhanced CT study). In the left maxillary sinus a typical amalgam (large black arrow in a), with its radiating artefacts, was observed and associated with hyperdense retention (white arrowhead in b) and calcifications (black arrowhead in b). Note that the fungal mass was bulging into the middle meatus of the nasal cavity (white arrow in b). In the right maxillary sinus a a hyperdense concretion (small black arrow) was noted and considered to be endodontal material. It was surrounded by a hyperdense retention

(white arrowhead)

has s e v e r a l clinical forms: n o n - i n v a s i v e , invasive, f u l m i n a n t a n d a l l e r g i c [10]. T h e n o n - i n v a s i v e f o r m is t h e m o s t c o m m o n a n d p r e s e n t s as c h r o n i c sinusitis (rhinorrhoea, nasal obstruction, headaches), unresponsive to m e d i c a l t r e a t m e n t , in n o n - i m m u n o c o m p r o m i s e d p a tients. A s d i d K o p p et al. [2], w e f o u n d t h a t p a r a n a s a l sinus a s p e r g i l l o s i s is m o r e f r e q u e n t in w o m e n t h a n in men. The invasive or fulminant forms are usually enc o u n t e r e d in i m m u n o c o m p r o m i s e d i n d i v i d u a l s [1, 10],

Y. Robert et al.: Non-invasive aspergillosis of the paranasal sinuses

163

Fig.Sa-c. Aspergillosis located in the middle meatus, a On the transverse CT scan calcifications (black arrowhead) surrounding endodontal material (small black arrow) are clearly seen. There is right maxillary sinus mucosal hyperplasia, related to the ostiomeatal location of the fungal mass (whitearrowheads), b On transverse Tl-weighted MRI image (TR/TE 550/25 ms) the fungal mass displays a low signal (whitearrowhead), less than that of the mucosal hyperplasia (white arrows), e On the T2-weighted MRI image (TR/TE 2000/50ms ) the signal of the fungal mass (white arrowhead) is very low, whereas the reactive mucosal hyperplasia is hyperintense (white arrows). Note that neither calcifications nor endodontal material are depicted by MRI Fig. 6. Aspergillosis of the maxillary sinus, extending into the middle meatus. On the coronal Tl-weighted MR scan (TR/TE 550/ 25 ms) after gadolinium injection, the fungal mass (white arrow) is not enhanced, whereas the inflammatory mucosa is enhanced (black arrow). Note the low signal defect corresponding to dental material (white arrowhead)

but may be seen in patients with no apparent predisposing cause [11, 12]. The main difference from the non-invasive form is the bony erosion and intracranial and/or orbital extension, which may simulate a tumour, mucormycosis or midline granuloma [5, 6, 13, 14]. Allergic Aspergillus sinusitis is characterised by the accumulation of mucoid material, in which CharcotL e y d e n crystals, scattered Aspergillus hyphae, eosinophilia and inflammatory infiltrates are found histologically [15].

Maxillary sinus is the most frequently involved paranasal cavity and usually unilaterally. Our results are similar to the findings of Kopp et al. and Patel et al. [2, 13], but the location was less frequently bilateral than reported by Patel et al. Sphenoid and ethmoid sinus locations are less usual and the frontal sinus is rarely affected [3, 13]. On plain radiography aspergillosis displays a homogeneous opacity, representing the mass of mycelia with mucosal reaction, which is observed in 50 % of the cases, without an air-fluid level. However, these signs are not sufficiently specific [2, 3]. When the medial maxillary wall is not seen, aspergillosis may be confused with a malignant neoplastic process. However, the visualisation of a very dense luminal concretion is a characteristic finding, found in more than half the cases in Kopp et al.'s series [2]. CT allows a better analysis of the sinus cavity opacity and more readily detects areas of increased attenuation, corresponding to the fungal mass [3, 13]. However, such a finding may be observed in desiccated secretions, intrasinus haemorrhage or chronic polyps [3, 16]. Thus, an area of increased attenuation is not specific and other signs, such as bony modifications and calcifications or endodontic material, are necessary to avoid misdiagnosis. Moreover, hyperdense retention may be absent. This was responsible for three false negatives in the study by Zinreich et al. [3], identified as having fungal sinusitis after histopathological findings proved posi-

164

Y. Robert et al.: Non-invasive aspergillosis of the paranasal sinuses

tive. In such cases other suggestive findings may be useful in the diagnosis. In our study osteosclerosis, with increased wall thickness, was more frequent than bony erosion, related to the chronic retention and the fungal mass effect respectively. Our results differ from Patel et al.'s findings, in which bone erosion is more frequent than bone sclerosis [13]. H y p e r d e n s e concretions are clearly seen on CT scans and are characteristic, but this sign may be absent, as in 6 of our cases [2, 3, 9]. Stammberger et al. [9] reported that these concretions, the attenuation values of which are high (about 2000 Hounsfield units), are mainly made of tertiary calcium phosphate and to a lesser degree of calcium sulphate and heavy-metal salts, deposited in necrotic areas. The differential diagnosis includes metallic foreign bodies or misplaced dental fillings, a possible cause of aspergillosis development [2, 8]. In Patel et al.'s study [13] calcifications, found in 7 of the 8 cases, had a much lower average attenuation value on CT. In our patients, 2 types of high-density opacity were observed: the first was slightly hyperdense, of irregular shape, and consistent with calcifications, as reported by Patel et al.; the second was more hyperdense, regular and considered to be endodontic material, after consulting clinical data. However, this hyperdense material was very similar to the hyperdense concretions as reported by Kopp et al. [2] and Stammberger et al. [9]. Thus, distinguishing between these two entities may at times be difficult and require comparison with clinical data. However, both are suggestive of paranasal sinus aspergillosis. Overall, only i of our patients had none of the CT criteria. Our M R I data are similar to the findings of Zinreich et al. [3]: the fungal mass displays a low or intermediate signal on T l - w e i g h t e d images, a low signal on T2weighted images and is not enhanced with gadolinium. This hypointense signal, which might be caused by calcium and ferromagnetic elements [3], is suggestive of the diagnosis, although it may be seen in chronic secretions with a high protein and poor water content, acute haemorrhage, calcium bone, enamel, endodontal material and air [16-18]. Nevertheless, such findings are not observed in tumours. Thus, M R I can be useful for differentiating neoplastic process from a pseudotumoral fungal mass. However, calcifications and bony modifications are more clearly seen on CT than on MRI, and CT may be useful for recognising the cause of hypointense signal, demonstrating the complementarity of the two methods. In conclusion, non-invasive paranasal sinus aspergillosis is characterised by the presence of a hyperdense retention, with calcifications or endodontal material, combined with increased wall thickness on CT. Such findings are suggestive of non-invasive aspergillosis, which re-

quires treatment by surgical removal of the whole fungal mass. In doubtful cases, such as pseudotumoral forms, M R I may have a complementary role in diagnosis, as it shows a suggestive low signal intensity on T1and T2-weighted images.

References 1. McGuirt WF, Harril JA (1979) Paranasal sinus aspergillosis. Laryngoscope 89:1563 2. Kopp W, Fotter R, Steiner H, Beaufort F, Stammberger H (1985) Aspergillosis of the paranasal sinuses. Radiology 156: 715 3. Zinreich SJ, Kennedy DW, Malat J, Curtin HD, Epstein JI, Huff LC, Kumar AJ, Johns ME, Rosenbaum AE (1988) Fungal sinusitis: diagnosis with CT and MR imaging. Radiology 169:439 4. Quiney RE, Rogers MJC, Davidson RN, Cheesman AD (1988) Craniofacial resection for extensive paranasal sinus aspergilloma. J Laryngol Otol 102:1172 5. Horton WD, Osguthorpe JD (1989) CT findings in sphenoid sinus aspergillosis. Otolaryngol Head Neck Surg 100:606 6. Gupta R, Singh AK, Bishnu R Malhotra V (1990) Intracranial aspergillus granuloma simulating meningioma on MR imaging. J Comput Assist Tomogr 14:467 7. Fuchs HA, Evans RM, Gregg CR (1985) Invasive aspergillosis of the sphenoid manifested as a pituitary tumor. South Med J 78:1365 8. Beck-Manegetta J, Necek K, Grasserbauer M (1983) Solitary aspergillosis of maxillary sinus: a complication of dental treatment. Lancet II: 1260 9. Stammberger H, Jakes R, Beaufort F (1984) Aspergillosis of the paranasal sinuses: X-ray diagnosis, histopathology, and clinical aspects. Ann Otol Rhinol Laryngol 93:251 10. Sarti EJ, Blaugrund SM, Tang Lin P, Camins MB (1988) Paranasal sinus disease with intracranial extension: aspergillosis versus malignancy. Laryngoscope 98:632 11. Coulthard A, Gholkar A, Sengupta RP (1991) Frontal aspergilloma: a complication of paranasal aspergillosis. Clin Radio144: 425 12. Centeno RS, Bentson JR, Mancuso AA (1981) CT scanning in rhinocerebral mucormycosis and aspergillosis. Radiology 140: 383 13. Patel PJ, Kolawole TM, Malabrey TM, Hulailah A, Hamid F, Chakaki M (1992) CT findings in paranasal aspergillosis. Clin Radio145:319 14. Kilpatrick C, Tress B, King J (1984) Computed tomography of rhinocerebral mucormycosis. Neuroradiology 26:71 15. Mt~nzel MA (1989) Die allergische aspergillus-sinusitis. Laryngo-Rhino-Otol 8:33 16. Sore PM, Dillon WR Curtin HD, Fullerton GD, Lidov M (1990) Hypointense paranasal sinus loci: differential diagnosis with MR imaging and relation to CT findings. Radiology 176:777 17. Som PM, Dillon WP, Curtin HD, Fullerton GD, Zimmerman RA, Rajagopalan B, Maron Z (1989) Chronically obstructed sinonasal secretions: observations on T1 and T2 shortening. Radiology 172:515 18. Zimmerman RA, Bilaniuk LT, Hackney DB, Goldberg HI, Grossman RI (1987) Paranasal sinus hemorrhage: evaluation with MR imaging. Radiology 162:499