Fungal rhinosinusitis (FRS) is a spectrum of sinus diseases that is caused by the fungal invasion of the mucosa of the paranasal sinus. FRS has been remarkably expanded in the last 3 decades, mainly due to an increasing number of immunocompromised hosts and the improvement in the diagnostic imaging tools and histopathology techniques, first described by Plaignaud in 1791 as a complication of the maxillary fungal sinusitis.

There are two main types of FRS – invasive and non-invasive. There are three entities of non-invasive FRS: allergic fungal rhinosinusitis (AFRS), fungal ball (FB), and saprophytic fungal infestation (SFI). Invasive FRS can be acute invasive fungal rhinosinusitis (AIFRS), chronic invasive fungal rhinosinusitis (CIFRS) or granulomatous invasive fungal rhinosinusitis (GIFRS). This classification is directly relevant to treatment, because each of the subtypes will need a different treatment approach.

In FRS, immune status of the host plays a large role in the pathogenesis of the disease. Most cases of non-invasive forms occur in immunocompetent individuals, and invasive FRS is more likely to occur in immunocompromised patients such as patients with diabetes mellitus, patients with hematologic malignancies, patients with HIV/AIDS or those who are receiving prolonged courses of chemotherapeutics or corticosteroids. The latter is associated with a 50% mortality rate and has the potential of extending intracranially where it is liable to cause life threatening neurological complications.

Aspergillus species predominate as causative organisms, the most common in allergic fungal sinusitis as well as the invasive varieties. A unique and aggressive group is mucorales (Mucor and Rhizopus) which cause mucormycosis and are the only fungi that invade the blood vessels and cause tissue necrosis.

Clinically, FRS has a heterogenous clinical spectrum of symptoms and signs, such as nasal obstruction, headache, rhinorrhea, epistaxis, hyposmia, and proptosis, cranial nerve palsies and periorbital cellulitis in later stages. These symptoms can be non-specific, causing a delay in diagnosis. Although the definitive diagnosis is made with histopathological confirmation of fungal elements in tissue samples, a significant role is reserved for imaging before surgery, not only for the characterization of the extent of disease, but also for the planning of surgical procedure, and for screening for serious complications like erosion of the skull base and intracranial extension.

Currently, the standard-of-care modality of imaging the paranasal sinuses before surgery is computed tomography (CT) with the FESS protocol. FESS CT also allows high-resolution multiplanar imaging for better identification of anatomic variants, evaluation of involvement of the ostiomeatal complex and also to show characteristic radiological features of FRS including hyperdense sinus opacification, calcification and osseous erosion. The sensitivity and specificity and overall diagnostic accuracy of FESS CT versus histopathology (the gold standard) has been inconsistently reported for different clinical settings, especially in Pakistan.

This study was then planned to find the diagnostic accuracy of MDCT images obtained using the FESS protocol for confirmation of fungal sinusitis according to age and gender in comparison with histopathology as the gold standard.

LITERATURE REVIEW

The role of computed tomography (CT) imaging in the diagnosis of fungal sinusitis has been reviewed by a growing number of studies, which have yielded a spectrum of sensitivities and specificities depending on the type of fungal sinusitis, the patient population, and the method used as the reference standard.

Junaid et al. (2021) performed a cross-sectional study on 120 patients with clinical suspicion of FRS and found a sensitivity of 96.19%, specificity of 93.33%, PPV of 99.01%, NPV of 77.77% and diagnostic accuracy of 95.83% in CT. The kappa statistics showed 82% agreement between CT and smear analysis which was near to substantial agreement.

In a study conducted at Civil Hospital and Dow Medical College Karachi, Nasreen et al. 2016, screened 98 patients with sinonasal polyposis with CT and compared the results with the gold standard of histopathology and found that the sensitivity was 89.3%, specificity was 86.9%, PPV 95.7% and NPV 71.4% and diagnostic accuracy of 88.7%. Interestingly, 47.9% of patients had intraorbital extension and 41.8% had intracranial extension on CT.

In a descriptive cross-sectional study, Iqbal et al (2017) found that sensitivity was 96.19%, specificity 93.33%, PPV 99.01%, NPV 77.77% and overall diagnostic accuracy was 95.83% which is similar to the results found by Junaid et al.

Awan et al. (2021) assessed 216 patients with chronic rhinosinusitis (CRS), of which, 45 (20.8%) were diagnosed with AFRS. In all AFRS cases, CT showed heterogeneous opacity (100%), and sinus expansion (62.2%), and in 28.9% showed bone destruction. The mean age of patient was 29.49 ± 9.16 years and there was male preponderance.

John et al. (2013) evaluated 23 immunocompromised patients with proven invasive fungal sinusitis (IFS) and found that 21 patients had severe mucosal edema of the nose, 8 had bone erosion, 6 had orbital invasion, and no patient had intracranial involvement. Unilateral disease was typical (21 of 23 patients) and substantial differences existed in CT findings between IFS patients and controls with non-fungal sinusitis (p<0.001).

Hafeez et al., (2022) performed a cross-sectional study in which 63 patients with a mean age of 41.82 ± 11.33 years were included, with a sensitivity of 95%, specificity of 68%, PPV of 87% and NPV of 86% for non-contrast CT as compared with culture. There was also a male preponderance (54%).

Together the studies suggest that CT has diagnostic utility in FRS and there is a variation in performance measures due to the differences in imaging protocol, the gold standard diagnostic method (histopathologic vs culture vs smear analysis), and the study population heterogeneity. The present study adds to the evidence base by specifically using the FESS CT protocol, and using histopathology as the reference standard in a Pakistani tertiary care environment

Study Design and Setting This was a cross sectional study done in Department of Radiology, Govt. Over a period of three months after the approval of the synopsis in the institution, Lahore, Pakistan in the Kot Khawaja Saeed Teaching Hospital. The study has been approved by the Institutional Review Board of King Edward Medical University, Lahore under the ethical clearance certificate number: 578/03/2021. All the participants have given written informed consent before entering the study. Samples and sampling technique Based on the prevalence of fungal sinusitis (18.9%), sensitivity (100%), and specificity (15.8%) and using an absolute precision margin of 10% and a 95% confidence level, the estimated sample size was 65 patients. A convenient sampling method was used but it was not a probability sampling. Inclusion / Exclusion Criteria Patients ranging from 14 to 80 years of age, of both sexes, who presented with the clinical features of paranasal sinus disease, including nasal obstruction, headache and rhinorrhea, and CT features of sinonasal disease, such as thickening of the paranasal sinus mucosa, polyp formation, homogeneous opacification, erosion of the bone, air fluid levels, involvement of intraorbital sinuses, intracranial sinuses, retroantral sinuses, and overlying soft tissues, were included. Those who were pregnant and patients who had undergone previous surgery for the same disease in the sinonasal cavity were excluded. Imaging Protocol Non-contrast MDCT of the paranasal sinuses was performed in all patients on Toshiba Aquilion 16-slice Multidetector Computed Tomography (MDCT) scanner. The FESS protocol was used that involves multiplanar reconstruction (MPR) in the axial, coronal and sagittal planes. The soft-tissue and bone window settings were used to assess the mucosal disease and osseous involvement, respectively. All FESS CT findings were carefully recorded, which comprised mucosal thickening, internal hyperdensity (double density sign), sinus expansion, bony erosion and thinning, intracranial extension, intraorbital extension, and anatomical variants. Reference Standard All patients with computed tomography (CT) positive and/or clinically suspected were forced to undergo the endoscopic sinus surgery (ESS) by an ENT surgeon. Biopsy samples were sent to histopathology department for formal analysis. The gold standard for diagnosis of fungal sinusitis was histopathological diagnosis and the findings of FESS CT were compared to it. Statistical Analysis SPSS version 25.0 was used to input the data and perform data analysis. Categorical variables were reported as frequencies (percentages) and continuous variables as mean ± standard deviation. The sensitivity, specificity, PPV, NPV, and overall diagnostic accuracy were calculated using a 2×2 contingency table. The results were also analyzed by age group (20-40, 41-60 and >60 years) and gender. Ninety-five percent confidence intervals were obtained for all the measures of diagnostic accuracy.

Patient Demographics

A total of 65 patients were enrolled. The mean age was 50.29 ± 13.38 years (range: 20–70 years). The cohort comprised 40 male (61.54%) and 25 female (38.46%) patients. The average time symptoms were present before the presentation was 8.2 ± 2.53 weeks (range 4–12 weeks).

Table 1. Patient Demographics and Clinical Characteristics

Clinical Presentation

The most common symptoms (92% each) were headache and hyposmia, followed by nasal blockage (89% of patients), allergic rhinitis (69%), rhinorrhea (62%), epistaxis (62%), and eye symptoms (62%).

CT Findings

100% (65) of patients showed mucosal thickening on FESS CT with hyperdensity in the mucosa. Sinus expansion was present in 58 patients (89.23%), bony erosion and thinning in 53 (81.54%), intracranial extension in 31 (47.69%), intraorbital extension in 29 (44.62%), and anatomical variants in 25 (38.46%). The CT criteria defined 38 patients (58%) as having fungal sinusitis.

Table 2. Paranasal Sinus CT (FESS Protocol) Findings

Histopathological Findings

Histopathological examination was positive in 39 patients (60%) and negative in 26 patients (40%) indicating fungal sinusitis.

Diagnostic Accuracy of MDCT

The CT results were compared with the histopathology results and 33 were TP (84.6%) and 6 FN (15.4%), 21 TN (80.8%) and 5 FP (19.2%). The resulting 2×2 contingency table and diagnostic accuracy parameters are shown below.

Table 3. Contingency Table: MDCT vs. Histopathology

Table 4. Diagnostic Accuracy Parameters of MDCT Using FESS Protocol

Age-Stratified Diagnostic Accuracy

Sensitivity increased with advancing age: 77.78% in the 20–40-year group, 84.21% in the 41–60-year group, and 90.91% in those over 60. Specificity was highest in the 41–60-year group (85.71%) and lowest in patients over 60 (66.67%). Diagnostic accuracy was highest in the 41–60-year group (84.85%).

Table 5. Age-Stratified Diagnostic Accuracy of MDCT (FESS Protocol)

Gender-Stratified Diagnostic Accuracy

Male patients demonstrated superior diagnostic performance across all parameters compared to female patients: sensitivity 88.88% vs. 75.00%, specificity 92.30% vs. 69.23%, PPV 96.00% vs. 69.23%, NPV 80.00% vs. 75.00%, and diagnostic accuracy 90.00% vs. 72.00%.

Table 6. Gender-Stratified Diagnostic Accuracy of MDCT (FESS Protocol)

The diagnostic value of MDCT with FESS protocol in the diagnosis of fungal sinusitis was assessed and compared with histopathology as the reference standard. The sensitivity was 84.62%, specificity was 80.77%, and the diagnostic accuracy was 83.08%, all of which were significant but fell short of 100%. These results place MDCT as a reliable screening and planning tool preoperatively, and emphasize the importance of histopathological confirmation.

Our findings are largely consistent with, but still below, the results of Junaid et al. (2021) and Iqbal et al. (2017) which both showed a diagnostic accuracy of >95% using smear analysis as the standard of diagnosis. This difference is clinically relevant: smear analysis and culture based methods are more sensitive for detecting the viable organisms while histopathology allows the identification of the pattern of tissue invasion useful for sub-classification. In the present study, the lower accuracy figures may have been a result of using histopathology as the reference standard as some cases found to be CT-positive without evidence of tissue invasion would not have been classified as FRS histopathologically.

Other study by Nasreen et al. (2016) also showed 88.7% diagnostic accuracy with histopathology being the gold standard, which is closer to our study. Their cohort also had high rates of intraorbital (47.9%) and intracranial (41.8%) extension on CT, which were similar to our findings of 44.62% and 47.69% respectively. The results demonstrate the importance of CT in identification of serious complications directly influencing surgical planning and prognosis.

A high prevalence of mucosal thickening with internal hyperdensity (100%, in our study) is consistent with the CT semiotics of fungal sinusitis. A well-known sign is the "double density sign" which is formed by the hyperdense fungal mucin and hypodense mucosal thickening in the periphery. Calcification is not always present but if it occurs, it is pathognomonic and is dystrophic calcification in inspissated fungal secretions. Bony erosion was observed in a high proportion of our cohort (81.54%), suggesting a higher proportion of invasive or aggressive non-invasive subtypes.

Age stratified analysis showed that sensitivity was greater with age (77.78% for 20–40 year group, 84.21% for 41–60 year group and 90.91% for >60 year group), possibly because older age groups were more likely to have co-morbid factors such as diabetes mellitus that would support invasive fungal growth and hence more likely to show the CT findings. Interestingly, specificity was lower in the oldest age group (66.67%), likely because of CT results of age-related sinonasal changes that mimicked fungal disease.

The results stratified by gender showed significantly better performance in male patients (90.00% diagnostic accuracy) than female patients (72.00%). Male predominance in FRS has been reported in several studies from the region, and the gender gap in CT accuracy could be due to differences in severity of disease of FRS or differences in the distribution of the different subtypes between the sexes. There may be a higher frequency of patients with non-invasive subtypes with more subtle CT findings, with lower sensitivity and specificity in female patients.

Hafeez et al. (2022) reported high sensitivity (95%) but low specificity (68%) as compared to our study (using culture as a gold standard). This inversion in comparison to the histopathology-based studies is to be expected as sensitivity is exaggerated by the detection of colonizing fungi which may not be causing invasion of tissue, and specificity is decreased when the true reference standard is histopathology.

This study has a number of limitations. Subgroup analyses were conducted with a relatively small sample size (n = 65) and thus may not have the precision necessary to draw valid conclusions from the analyses. This is a cross-sectional design, which is not suitable for follow-up and evaluation of treatment outcomes. The epidemiology of the data may be different in a single center (tertiary referral hospital). Additionally, because of sample size limitations, the study was not able to distinguish between CT performances at each FRS subtype.

In spite of the above, the results advocate that MDCT combined with FESS protocol is a mandatory pre-operative work-up in all cases of suspected fungal sinusitis. Beyond diagnosis, it is helpful in delineating anatomical variants and determining the presence of ostiomeatal complex obstruction and extension of extrasinus disease, which allows direct and useful guidance for surgical planning in the endoscopy era and helps to predict the presence of life-threatening complications.

The overall sensitivity, specificity, and diagnostic accuracy of MDCT with FESS protocol for diagnosis of fungal sinusitis are 84.62%, 80.77%, and 83.08%, respectively, when compared with the gold standard histopathology.

There was better performance in males and in patients older than 60 years. The imaging protocol is very good in determining disease extent, intracranial involvement, intraorbital involvement, and that is very important in determining surgical approach and prognosis.

The results support the use of FESS CT as a vital part of the pre-operative assessment of fungal sinusitis. But sensitivity and specificity are not absolute, and the histopathological confirmation is required for definitive diagnosis and subtyping of the lesion. Further multi-center prospective studies with larger cohorts and subtype stratification is warranted to further delineate the diagnostic performance of MDCT at the extremes of fungal rhinosinusitis.

1. Pilai S, Bhandarkar AM, Nair SS, Jha AT, Ramaswamy B, Al-Abri R. A radiological profile of fungal sinusitis. Asian J Pharm Clin Res. 2017;10(9):331–4.
2. Naz N, Ahmad Z, Malik S, Zahid T. Diagnostic accuracy of CT scan in fungal sinusitis, diagnosis and extent. Ann Pak Inst Med Sci. 2016;12(2):61–5.
3. Chatterjee SS, Chakrabarti A. Epidemiology and medical mycology of fungal rhinosinusitis. Otorhinolaryngol Clin. 2009;1:1–13.
4. Ferguson BJ. Mucormycosis of the nose and paranasal sinuses. Otolaryngol Clin North Am. 2000;33(2):349–65.
5. Waqar S, Ameer S, Naeem M, Bajwa MA, Bajwa SM, Bajwa GR. Diagnostic accuracy of MRI for acute invasive fungal sinusitis with histopathology as gold standard. Pak Armed Forces Med J. 2020;70(6):1810.
6. Singh V. Fungal rhinosinusitis: Unravelling the disease spectrum. J Maxillofac Oral Surg. 2019;18(2):164–79.
7. Gorovoy IR, Kazanjian M, Kersten RC, Kim HJ, Vagefi MR. Fungal rhinosinusitis and imaging modalities. Saudi J Ophthalmol. 2012;26(4):419–26.
8. Nicolai P, Lombardi D, Tomenzoli D. Fungus ball of the paranasal sinuses: experience in 160 patients treated with endoscopic surgery. Laryngoscope. 2009;119(11):2275–9.
9. Lutfi IA, Maheswari N, Devi D, Afridi FI. Accuracy of computed tomography in characterizing paranasal fungal infection. Pak J Rehabil. 2013;2(1):13–6.
10. Meng Y, Zhang L, Piao Y, Lou H, Wang K, Wang C. MRI in differential diagnosis of allergic fungal sinusitis and eosinophilic mucin rhinosinusitis. J Thorac Dis. 2019;11(8):3569.
11. Pant H, Kette FE, Smith WB. Eosinophilic mucus chronic rhinosinusitis: clinical subgroups or a homogeneous entity? Laryngoscope. 2006;116:1241–7.
12. Kanwar SS, Mital M, Gupta PK, Saran S. Evaluation of paranasal sinus diseases by CT and histopathological correlation. J Oral Maxillofac Radiol. 2017;5:46–52.
13. Krishnan KU, Agatha D, Selvi R. Fungal rhinosinusitis: a clinicomycological perspective. Indian J Med Microbiol. 2015;33:120–4.
14. Kandukuri R, Phatak S. Evaluation of sinonasal diseases by CT. J Clin Diagn Res. 2016;10(11):TC09.
15. Gupta K, Saggar K. Analysis of CT features of fungal sinusitis and correlation with endoscopy and histopathology. Ann Afr Med. 2014;13:119–23.
16. Sharifian H, Kashani SS, Mazaher H, Abhari M. Accuracy of paranasal sinuses limited CT scans in sinusitis diagnosis. Iran J Radiol. 2006;3(4):225–8.
17. Rao K. CT of paranasal sinus pathologies with FESS/nasal endoscopy correlation. Clin Rhinol. 2015;8(1):15–9.
18. Adelson RT, Marple BF. Fungal rhinosinusitis: state-of-the-art diagnosis and treatment. J Otolaryngol. 2018;34:18–23.
19. Lloyd G. The sinuses. In: Sutton D, ed. Textbook of Radiology and Imaging. 6th ed. New York: Churchill Livingstone; 2012:1297–1300.
20. Katzenstein AL, Sale SR, Greenberger PA. Allergic Aspergillus sinusitis: a newly recognized form of sinusitis. J Allergy Clin Immunol. 2019;72:89–93.
21. Lanza DC, Dhong HJ, Tantilipikorn P, et al. Fungus and chronic rhinosinusitis: from bench to clinical understanding. Ann Otol Rhinol Laryngol Suppl. 2016;196:27–34.
22. Junaid I, et al. Diagnostic accuracy of CT scan in fungal sinusitis using smear analysis as gold standard. [Cross-sectional study]. 2021.
23. Nasreen N, et al. Diagnostic accuracy of CT scan in fungal sinusitis in Karachi. Ann Pak Inst Med Sci. 2016.
24. Iqbal J, et al. Descriptive cross-sectional study on CT diagnosis of fungal sinusitis. 2017.
25. Awan N, et al. Diagnostic accuracy of fungal culture in allergic fungal rhinosinusitis patients. Mayo Hospital Lahore study. 2021.
26. John M, et al. CT imaging findings predictive of invasive fungal sinusitis in immunocompromised patients. 2013.
27. Hafeez R, et al. Non-contrast CT scan in fungal sinusitis: diagnostic accuracy against culture. 2022.
28. Soler ZM, Schlosser RJ. The role of fungi in diseases of the nose and sinuses. Am J Rhinol Allergy. 2012;26:351–8.
29. Marfani M, Jawaid M, Shaikh S, Thaheem K. Allergic fungal rhinosinusitis with skull base and orbital erosion. J Laryngol Otol. 2010;124:161–5.
30. Aribandi M, McCoy VA, Bazan C. Imaging features of invasive and noninvasive fungal sinusitis. Radiographics. 2017;27(5):1283–96.
31. Broglie MA, Tiguely M, Holzmann D. Diagnosing sinus fungal balls in paranasal sinus. Rhinology. 2019;47:37–9.
32. Bent JP, Kuhn FA. Diagnosis of allergic fungal sinusitis. Otolaryngol Head Neck Surg. 2014;111:580–8.
33. Hopkins C, Browne JP, Slack R. The Lund-Mackay staging system for chronic rhinosinusitis. Otolaryngol Head Neck Surg. 2017;137:555–561.
34. Grosjean P, Weber R. Fungus balls of the paranasal sinuses: a review. Eur Arch Otorhinolaryngol. 2017;264:461–470.