Transcutaneous retrobulbar amphotericin B for rhino-orbital-cerebral mucormycosis: a multi-center retrospective comparative study

Purpose: To assess whether transcutaneous retrobulbar amphotericin B injections (TRAMB) reduce exenteration rate without increasing mortality in rhino-orbital-cerebral mucormycosis (ROCM). Methods: In this retrospective case–control study, 46 patients (51 eyes) with biopsy-proven ROCM were evaluated at 9 tertiary care institutions from 1998 to 2021. Patients were stratified by radiographic evidence of local orbital versus extensive involvement at presentation. Extensive involvement was defined by MRI or CT evidence of abnormal or loss of contrast enhancement of the orbital apex with or without cavernous sinus, bilateral orbital, or intracranial extension. Cases (+TRAMB) received TRAMB as adjunctive therapy while controls (−TRAMB) did not. Patient survival, globe survival, and vision/motility loss were compared between +TRAMB and -TRAMB groups. A generalized linear mixed effects model including demographic and clinical covariates was used to evaluate the impact of TRAMB on orbital exenteration and disease-specific mortality. Results: Among eyes with local orbital involvement, exenteration was significantly lower in the +TRAMB group (1/8) versus -TRAMB (8/14) (p = 0.04). No significant difference in mortality was observed between the ±TRAMB groups. Among eyes with extensive involvement, there was no significant difference in exenteration or mortality rates between the ±TRAMB groups. Across all eyes, the number of TRAMB injections correlated with a statistically significant decreased rate of exenteration (p = 0.048); there was no correlation with mortality. Conclusions: Patients with ROCM with local orbital involvement treated with adjunctive TRAMB demonstrated a lower exenteration rate and no increased risk of mortality. For extensive involvement, adjunctive TRAMB does not improve or worsen these outcomes.