Clinical predictors of invasive fungal rhinosinusitis in a tertiary pediatric hospital.

Objectives: Invasive fungal rhinosinusitis (IFRS) is a potentially fatal disease that affects the severely immunocompromised and requires aggressive treatment. The objective of this study is to better describe predictors of biopsy positivity in patients at high risk of IFRS at a pediatric hospital.

Methods: This was a single-center case-control study of 36 patients (37 total biopsies) ≤ 21 years old with one of five high-risk oncologic/hematologic diagnoses who underwent operative endoscopy for clinical suspicion for IFRS. IFRS positivity was defined histologically. Collected information included patient demographics, primary diagnosis, oncologic relapses, time from diagnosis to biopsy, clinical characteristics, and endoscopic findings. These data were used to create a simple predictive scoring system.

Results: 17 patients had biopsy-proven IFRS (IFRS(+)) for an overall incidence of 2.1% in the designated high-risk population. Average time from most recent oncologic development (diagnosis, relapse, or hematopoietic stem-cell transplant) to biopsy in the IFRS(+) group was 2.09 months (SD = 2.26), and 7.28 months in the IFRS(-) group (SD = 9.17) (p = 0.009). Clinical characteristics did not differentiate between IFRS(+) and IFRS(-). Bedside endoscopy performed poorly, as it was interpreted as normal in 42.8% of IFRS(+) and 53.8% of IFRS(-). In contrast, the presence of any positive endoscopic finding intra-operatively was highly specific for IFRS(+) (94%) with moderate sensitivity (70%), and the difference in rate of positivity between groups was statistically significant (p < 0.001).

Conclusion: Operative endoscopy with biopsy remains the gold-standard to rule-out IFRS in the setting of high clinical suspicion. Time elapsed from most recent oncologic development to clinical concern for IFRS may influence the likelihood of disease, though this requires further study. Clinical symptoms and bedside endoscopy were not predictive and should be used with caution in decision-making.