1 Introduction
Chronic invasive fungal rhinosinusitis (CIFRS) is a rare but serious opportunistic mycotic infection, accounting for an estimated 10%–15% of invasive fungal rhinosinusitis. It predominantly affects patients with mild immunocompromise
[1], such as diabetes mellitus, renal insufficiency, or low-dose glucocorticoid exposure, and, unlike acute invasive disease, pursues an indolent course over months to years, silently extending to the skull base, orbit, and cranial nerves. Skull base osteomyelitis secondary to CIFRS is a devastating complication that can erode the internal carotid artery (ICA), cranial nerves, and dura mater; ICA involvement raises the risk of life-threatening hemorrhage, stroke, or pseudoaneurysm.
We report a case in which the dominant clinical signature was not localized sinonasal symptoms but a progressive, multilevel cranial neuropathy that accrued over months, compounded by a non-diagnostic initial biopsy. The purpose of this report is to emphasize that signature, the differential between sinonasal CIFRS and central skull base osteomyelitis, and the rationale for maximal safe debridement when the disease encases both carotid arteries.
2 Case Presentation
2.1 History
A 77-year-old man was admitted on 4 March, 2026 (designated day −1, one day before treatment initiation), with a six-month history of left facial numbness and a 12-day history of hoarseness and dysphagia. Six months before admission he developed, without obvious precipitant, numbness of the left face with left nasal obstruction and mucopurulent rhinorrhea, left aural discharge, left tinnitus and hearing loss, and progressive left visual decline; occasional frontal or temporal headache was reported. He denied epistaxis, olfactory loss, facial pain, exophthalmos, diplopia, or weight loss. A computed tomography study at a referring hospital (20 February 2026) showed nasopharyngeal soft-tissue thickening with skull-base bone destruction, used as the comparison baseline. In December 2025 a biopsy was performed at that hospital; the pathology report described a benign lesion, and conservative management followed without symptomatic improvement. Twelve days before admission, hoarseness and progressive oropharyngeal dysphagia emerged, and a nasogastric tube was placed at the referring hospital for enteral feeding. He was referred to our institution with a presumptive diagnosis of left skull base lesion.
His prior history included type 2 diabetes mellitus of eight years’ duration managed with subcutaneous insulin (reportedly satisfactorily controlled), renal insufficiency of three months’ duration managed with oral medication, and a right clavicular fracture treated with internal fixation in 2023. He denied recent high-dose antibiotic exposure, sepsis, or glucocorticoid therapy
[2].
2.2 Physical examination
On admission the patient appeared chronically ill but was alert and cooperative. Vital signs showed a heart rate of 120 beats per minute with regular rhythm; blood pressure 130/88 mmHg, temperature 36.5 °C, respiratory rate 20 breaths/min. Cranial nerve examination revealed left visual impairment, left facial numbness (reported; formal dermatomal mapping of the trigeminal divisions was not documented), and left vocal fold fixation with impaired glottic closure on laryngoscopy; the right vocal fold moved normally. Nasal endoscopy showed congested nasal mucosa, no mass or discharge in either nasal cavity or choana, and postoperative changes in the nasopharynx. The septum was recorded as midline on endoscopy, whereas CT described leftward septal deviation, a within-source discrepancy noted rather than resolved. Otoscopy showed normal auricles, patent external auditory canals, normal tympanic membranes bilaterally, no middle-ear effusion, and no mastoid tenderness; imaging showed right mastoid effusion without external auditory canal bony erosion. No cervical lymphadenopathy was palpable. The remainder of the head and neck examination was unremarkable. Formal ophthalmologic visual acuity testing was performed on admission, revealing left eye visual acuity of 0.1 (right eye 0.8). A tracheostomy was planned for severe bulbar dysfunction; whether it was performed awaits confirmation from the operative note. Postoperative bedside swallow assessment confirmed severe oropharyngeal dysphagia.
2.3 Imaging
Preoperative imaging was obtained on admission (4 March 2026; radiology exam No. 202603040251) at our institution and comprised two studies: skull base computed tomography (CT) and contrast-enhanced nasopharyngeal magnetic resonance imaging (MRI) (Table 1).
Skull base CT (bone window) showed right nasal-cavity post-surgical changes with localized bone defects across multiple sinuses; bilateral sinus mucosal thickening with low-density lesions; a cyst about 1.5 cm in the left sphenoid sinus; irregular bone destruction of the right sphenoid sinus, pterygoid process, and greater wing of the sphenoid bone; diffuse bone destruction of the bilateral maxillary alveolar process and hard palate; left deviation of the nasal septum; soft-tissue thickening of the nasopharyngeal roof; and mild depression of the right medial orbital wall (Fig. 1).
Contrast-enhanced nasopharyngeal MRI showed irregular soft-tissue thickening of the right sphenoid sinus, greater wing of the sphenoid bone, and deep nasopharynx, with necrosis and restricted diffusion (diffusion-weighted imaging [DWI]-positive), invading adjacent soft-tissue spaces. The right middle cranial fossa meninges showed mild thickening and abnormal enhancement. The process lay adjacent to the right internal carotid artery (C3–C6 segments) with mild luminal narrowing of the foramen lacerum segment. Bilateral maxillary alveolar-process and hard-palate bone destruction was noted, along with right mastoid effusion and a nodule about 15 mm in the left maxillary sinus
[3] (Fig. 2).
Postoperative imaging comprised two contrast-enhanced skull-base MRI studies. The postoperative day 1 study (6 March 2026) showed right-sided multiple sinus bone defects, extensive surgical-cavity packing material and hemorrhage, residual enhancing soft tissue in the deep nasopharynx involving the sphenoid body, and a right greater-sphenoid-wing lesion with enhancement and necrosis; the right middle cranial fossa meninges were mildly thickened. Comparison with the preoperative study showed resection of the right sphenoidal soft-tissue disease, with the residual right greater-sphenoid-wing and deep-nasopharyngeal disease stable in extent. The postoperative day 10 study (14 March 2026, at a referring hospital in Fujian) showed no new abnormal intracranial signal or enhancement; postoperative sinus changes; partial absence of the right maxillary sinus wall and middle/inferior turbinates; a right nasal-vestibule about 5 mm nodule decreased in size with marked enhancement; resolution of a right maxillary sinus cystic signal; stable bilateral sinus mucosal thickening; diffuse T2-hyperintense signal in the bilateral maxilla and middle-cranial-fossa skull base with patchy/ring enhancement but no disease progression; parapharyngeal and right facial soft-tissue swelling; and stable mild enhancement of the right middle cranial fossa meninges with right mastoid T2-hyperintense effusion (Figs. 3 and 4).
2.4 Surgery
On 5 March, 2026 (day 0), the patient underwent surgical treatment under general anesthesia.
The operative intent was maximal safe debridement of necrotic and infected tissue while protecting the encased carotid arteries, cranial nerves, and dura, deliberately retaining right-sided disease in the deep nasopharynx and right greater sphenoid wing, abutting the cavernous sinus, orbital apex, and internal carotid artery. Whether dedicated neurosurgical and interventional-radiology standby was available during the procedure could not be confirmed from the records accessible to the authors, as the operative note was not retrievable at the time of submission; readers should not infer either its presence or its absence. A tracheostomy was planned preoperatively given the nasogastric dependence and bulbar (IX/X) dysfunction.
2.5 Pathology
Histopathological examination (pathology report No. 26-03205, issued 11 March 2026) was performed on specimens obtained at surgery. Three specimens were received: (1) right sphenoid sinus, (2) nasopharynx, and (3) left parapharyngeal space. The right sphenoid sinus specimen showed chronic granulomatous inflammation with fibroblastic proliferation and necrosis at the tissue periphery containing fungal hyphae and spores; special stains were positive for Grocott’s methenamine silver (GMS+) and periodic acid–Schiff (PAS+). The nasopharyngeal and left parapharyngeal specimens showed fibrogranulomatous tissue with confluent necrosis and scattered fungal hyphae; special stains were scant-positive for GMS and PAS. In aggregate, in conjunction with the clinical and radiological findings, the picture was considered invasive mould disease. No fungal culture or molecular (NGS) testing was performed on these specimens. Supplementary special stains for vascular-wall invasion (EVG, CD31, CD34) were not performed; assessment of vascular involvement therefore rests on imaging findings (Table 2).
2.6 Postoperative course
After surgery the patient was transferred to the intensive care unit and remained under intensive care unit (ICU) monitoring for 72 hours before ward transfer. He was discharged with a plan for outpatient review by the senior author (Q. Liu) two weeks after discharge. The patient was discharged on 2 Apri, 2026. Oral voriconazole 200 mg twice daily was initiated with therapeutic drug monitoring; the intended total duration of antifungal therapy is not separately documented in the source records available to the authors. During the available observation window the early postoperative course was favorable: the surgical bed showed expected packing and resolving hemorrhage without active bleeding, and objectively the systemic inflammatory markers declined (C-reactive protein 93.0 → 43.0 mg/L between postoperative days 3 and 4; hemoglobin 92 → 105 g/L; neutrophil percentage 85.0% → 83.1%). Bacterial culture and a six-pathogen respiratory nucleic-acid panel were negative. No long-term follow-up, recurrence, reoperation, treatment interruption, or drug-related adverse event could be documented within the available records; the observation window at the time of manuscript submission was 29 days (the inpatient admission, 4 March–2 April 2026), and no post-discharge follow-up, imaging, or therapeutic-drug-monitoring data were available (Figs. 3 and 4).
3 Discussion
CIFRS remains an uncommon but formidable complication of mild immunocompromise, most often diabetes mellitus, renal insufficiency, or low-dose glucocorticoid exposure. Its indolent tempo belies a capacity for relentless local advance, and the present case illustrates a clinical signature that, to our reading, deserves greater emphasis than the odontogenic mimicry that has dominated prior reports.
Progressive, multilevel cranial neuropathy was the organizing presentation. The patient’s symptoms accrued in a cranial-nerve-specific sequence over roughly six months: left visual decline (II), left facial hypoesthesia (V), left otologic symptoms with tinnitus and hearing loss (VII/VIII), and finally hoarseness and dysphagia with left vocal fold fixation (IX/X) necessitating nasogastric feeding
[4]. This stepwise, multimodal cranial-nerve involvement, rather than a single localized deficit, is the feature we would highlight as the case’s principal teaching point. In an elderly diabetic host it can be mistakenly attributed to cerebrovascular or neurodegenerative disease until mass effect on the skull base becomes radiologically apparent. The non-diagnostic initial biopsy, reported as a benign lesion at the referring hospital, added roughly three months of diagnostic delay and allowed the process to reach the carotid sheath and pharyngeal constrictors. We would caution that a single superficial biopsy from a necrotic field carries a real false-negative risk in CIFRS, and that a negative report should not terminate the diagnostic search when imaging suggests invasive disease.
CIFRS versus central skull base osteomyelitis. The imaging and pathological findings localize a fungal process centered on the nasopharynx, sphenoid sinus, and central skull base, with marrow signal loss, a clival ring-enhancing necrotic focus, and osseous destruction. Yet the diffuse destruction of the bilateral maxillary alveolar process and hard palate and the right medial orbital-wall depression raise a genuine differential: is this primarily
[5] sinonasal CIFRS extending inferolaterally, or central skull base osteomyelitis (CSBO) of odontogenic origin extending superomedially? In practice the two entities converge at the petrous apex and clivus, and in an elderly diabetic patient either pathogenesis is plausible. We favor a sinonasal CIFRS origin given the nasopharyngeal and sphenoid predominance, but the otologic component warrants explicit consideration rather than omission; the distinction matters because it shapes surveillance of the temporal bone and the external auditory canal.
Treatment and follow-up framework for skull base osteomyelitis. The decision to pursue maximal safe debridement rather than complete resection reflected the right internal carotid artery involvement (C3–C6, foramen lacerum mild stenosis) and the right-sided disease abutting the cavernous sinus and orbital apex
[6], anatomy that rendered aggressive clearance unsafe. The surgical approach should be individualized according to disease extent and the involvement of critical neurovascular structures; endoscopic skull base surgery is one option that affords visualization and precise debridement when anatomy permits. Judging the success of such debridement, however, requires a defined framework. Skull base osteomyelitis is assessed not by a single early scan but by the convergence of clinical response, serial acute-phase reactants (C-reactive protein, erythrocyte sedimentation rate), repeated MRI documenting evolution of marrow signal and enhancement, and, when uncertainty persists, nuclear medicine (gallium-67 scintigraphy or
18F-FDG PET-CT). Courses typically extend over months. Our available data are limited to an early postoperative window; we therefore make no claim of cure or of completed therapy, and we present the postoperative imaging as documentation of surgical extent and early course rather than as proof of disease control.
Antifungal stewardship in a frail host. Voriconazole is a rational agent for Aspergillus-like mould disease, but its use in a 77-year-old patient with renal insufficiency and hypoalbuminemia (albumin 29.2 g/L) demands vigilance. Renal impairment and low albumin both alter voriconazole exposure and augment toxicity risk, and therapeutic drug monitoring (TDM) is accordingly advisable to target therapeutic levels while surveilling hepatic, visual, and neurological adverse effects. We were unable to verify the administered regimen, dose, or duration from the available records and have not asserted them; when supplied, they should be coupled with TDM data and individualized to clinical and radiological response, which may span months to beyond a year.
In summary, this case reminds us that CIFRS in the mildly immunocompromised elderly may present as sequentially accruing multiple cranial neuropathies, that a non-diagnostic initial biopsy must not close the diagnostic effort, and that maximal safe debridement, with honest acknowledgment of intentionally retained disease, is sometimes the only defensible surgical goal.
4 Conclusion
Chronic invasive fungal rhinosinusitis is a slowly progressive infection of mildly immunocompromised hosts with a high propensity for skull base and vascular involvement. Its presentations are protean; in the elderly diabetic patient it may declare itself through sequentially accruing multiple cranial neuropathies rather than localized sinonasal symptoms, and a non-diagnostic initial biopsy can compound diagnostic delay. A high index of suspicion, histopathological confirmation with fungal stains, maximal safe debridement that honestly acknowledges intentionally retained disease, and individualized antifungal therapy guided by therapeutic drug monitoring constitute the cornerstones of management. Early postoperative imaging documents surgical extent and short-term course but cannot, in isolation, establish durable disease control.
5 Limitations
This report has several limitations. First, the fungal aetiology is established histopathologically as invasive mould disease with morphology compatible with an Aspergillus-like hyaline septate mould, but no definitive species identification was achieved; fungal culture and molecular testing, including next-generation sequencing (NGS) and panfungal PCR, were not performed on the available specimens, and we therefore cannot state the organism beyond morphological inference. Second, the observation window at submission was only 29 days; no long-term follow-up, recurrence, reoperation, or treatment-completion data are available, and no efficacy claim is made. Third, no dedicated vascular imaging (CTA or high-resolution vessel-wall MRI) was obtained; the internal carotid artery was assessed only on the contrast-enhanced nasopharyngeal MRI, which showed right ICA (C3–C6) involvement with mild foramen lacerum stenosis. We cannot quantify the degree of stenosis, characterize the vessel wall, or document stability or progression postoperatively, and no vascular-surgery, neurology, or neurosurgery consultation is documented in the available records. Fourth, the detailed operative description, including precise procedures, device models, blood loss, duration, frozen-section findings, and the status of the planned tracheostomy, were not available for review, as the primary operative note could not be retrieved at the time of manuscript preparation. Fifth, the initial referring-hospital biopsy and its wax block were not available for our re-review. Sixth, the two preoperative studies are dated 4 March 2026 (radiology exam No. 202603040251); the 4-versus-5 March discrepancy in the discharge summary has been resolved to 4 March 2026. The postoperative studies are dated 6 March (POD1) and 14 March (POD10) 2026. Finally, the availability of intraoperative neurosurgical and interventional-radiology standby could not be verified because the operative note was not retrievable at submission; the perioperative vascular-contingency arrangements therefore cannot be described.
The Author(s). This article is published by Higher Education Press at journal.hep.com.cn.
This is an open access article distributed under the Creative Commons Attribution License 4.0 (CC BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
PDF
(7296KB)
