MRI Versus CT in Acute Invasive Fungal Sinusitis: Mapping Diagnostic Accuracy Across Sinonasal Subsites

Acute invasive fungal sinusitis (AIFS) is a rare, aggressive infection characterized by fungal colonization of the nasal cavity and paranasal sinuses with subsequent tissue invasion affecting immunocompromised patients [1]. Prompt diagnosis is crucial as treatment delays are linked to significantly worse patient outcomes [2]. Imaging modalities such as computed tomography (CT) and magnetic resonance imaging (MRI) play a vital role in the early diagnosis of AIFS and facilitate immediate intervention. While typical CT and MRI findings have been reported [3], few studies have directly compared their diagnostic performances. Hence, we aim to evaluate and compare the diagnostic utility of CT and MRI across various sinonasal subsites in AIFS. A retrospective chart review (July 2016–June 2023) was conducted at an academic tertiary care center in Tucson, Arizona. All patients with a biopsy-confirmed diagnosis of AIFS in any sinonasal subsite, with available CT and MRI scans, were included. Before data collection, all studies were read by a neuroradiologist and independently reviewed by a rhinologist and skull base surgeon. The presence of an abnormality on imaging within a specific sinonasal subsite was considered a positive finding for that subsite. The imaging findings for both CT and MRI were independently compared to the intraoperative pathology results for each subsite. Descriptive statistics were used to summarize the data. Continuous variables are presented as means and standard deviations, and categorical variables as frequencies and proportions. Chi-square tests were performed to evaluate the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for each sinonasal subsite and for both CT and MRI imaging modalities. Fifty-one patients were included in this study, with a mean age of 53.5±16.8 years, 66.7% of which were male. Diabetes mellitus was the most prevalent immunocompromising condition (64.7%). The time from symptom onset to presentation varied, with the largest proportion (41.1%), presenting within 7 days. The most common presenting symptoms and baseline clinical characteristics are presented in Tables S1 and S2. Bedside endoscopy was suspicious for AIFS in 44% of cases. Common imaging findings on CT and MRI are described in Table 1. On subsite analysis, MRI significantly outperformed CT in sensitivity within the turbinates (88.4% vs. 0%, p < 0.001), nasal septum (80% vs. 13.6%, p < 0.001), and hard palate (92.8% vs. 64.2%, p < 0.001). There were no significant differences in detecting abnormalities within the paranasal sinuses, orbit, and skull base between CT and MRI (Figure 1). Septum Comparison of CT and MRI performance in terms of sensitivity, specificity, positive Predictive value (PPV), and negative predictive value (NPV) across (A) paranasal sinuses, (B) Turbinates, (C) septum, (D) hard palate, (E) orbit, and (F) skull base. Black bars represent CT, while Gray bars represent MRI. Statistically significant differences between CT and MRI are indicated by Asterisks (*p < 0.05, **p < 0.01, ***p < 0.001). Early diagnosis of AIFS, particularly when localized, improves prognosis, with imaging playing a critical role in assessment. Some studies advocate for CT as the first-line modality [4], while the American College of Radiology's expert panel on neurologic imaging supports MRI for initial screening and CT for intraoperative management [5]. Studies comparing CT and MRI characteristics in AIFS are limited. Herein, we address this by evaluating imaging utility based on specific sinonasal subsites. Our study demonstrates MRI's superior sensitivity over CT in detecting AIFS within the turbinates, septum, and hard palate. All patients with turbinate involvement demonstrated loss of contrast enhancement (LoCE) on MRI, a key marker of vascular invasion and tissue necrosis (Figure S1). The “black turbinate sign” on MRI has been thought to indicate devitalized mucosa, though previous studies caution that similar findings may appear in immunocompetent patients without AIFS [1, 6]. Similarly, MRI demonstrated significantly improved septum and hard palate sensitivity. Nasal septal ulceration on CT (Figure S1), a well-described diagnostic finding, was noted in three patients confirmed to have AIFS [7]. Interestingly, most patients had no suspicious CT findings within the septum, resulting in 19 false-negative scans. There was no significant difference between CT and MRI within the orbit. However, MRI detected several well-known key features of AIFS in the orbit [7], including LoCE, likely due to its superior soft tissue contrast resolution. LoCE has been previously described in AIFS, noting that healthy extraocular muscles typically enhance on T1-weighted imaging [8]. Within the paranasal sinuses, most patients presented with paranasal sinus edema or opacification, common findings in sinusitis, and non-specific for AIFS [9]. However, hyperdense content on CT was observed in all cases (Figure S1), suggesting its potential role in identifying fungal hyphae, seen in both invasive and non-invasive types of fungal infection [7]. All patients with skull base involvement demonstrated fat stranding in the infratemporal or pterygopalatine fossa on CT. The pterygopalatine fossa is considered a potential reservoir for AIFS, and its debridement may help contain disease spread [10]. Notably, all patients with cavernous sinus thrombosis or involvement on MRI (Figure S1) had biopsy-confirmed AIFS in this location, underscoring the diagnostic value of these features in assessing disease extent. Our cohort largely reflected the most common presentation of AIFS, with most of our patient population having presented with diabetes and mucormycosis. However, the retrospective, single-center design of this study inherently carries potential biases. Locations chosen to biopsy were driven by imaging findings affecting the overall accuracy of our findings. Furthermore, since this study exclusively included patients with biopsy-confirmed AIFS, it may introduce a selection bias toward more advanced disease, potentially overestimating the PPV, NPV, sensitivity, and specificity of imaging modalities and our findings. In conclusion, MRI with contrast should be part of the first-line investigations for AIFS, along with CT. It is essential for the detection of early AIFS and for surgical planning. Given the aggressive nature of AIFS and high mortality rates, optimizing imaging strategies may facilitate earlier interventions and potentially improve patient outcomes. Future research should focus on refining standardized imaging protocols to optimize image timing and conducting multicenter prospective studies to validate these findings and assess their impact on reducing delays. The authors have nothing to report. The authors declare no conflicts of interest. Supplementary Figure 1: Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) findings of patients with acute invasive fungal sinusitis. Coronal CT (A) and post-contrast MRI (B) demonstrating lack of contrast enhancement left inferior turbinate. Axial CT (C) and post-contrast MRI (D) images reveal a 1.5 cm anterior nasal septal defect. Axial CT (E) and post-contrast MRI (F) images demonstrating hyperdense foci within right maxillary sinus. Axial CTA (G) and post-contrast MRI (H) demonstrating left cavernous sinus thrombosis. Supplementary Table 1: Baseline Demographic and Clinical Characteristics of AIFS Patients Supplementary Table 2: Clinical presentations of AIFS Supplementary Table 3 Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.