Bone Abstracts

Objectives: Optic neuropathy (ON) is a critical complication of fibrous dysplasia (FD). Early surgical decompression can prevent blindness; however, prophylactic intervention may cause blindness from intra-operative nerve damage. There is therefore a critical need to develop diagnostic tests for accurate and early detection of ON in patients with FD. Currently used methodologies, including radiographic evaluation of optic canal size and optic nerve length, are suboptimal and correlate poorly with objective vision loss. Optical coherence tomography (OCT) has emerged as a potential tool due to its ability to detect retinal nerve fiber layer (RNFL) thinning, an early sign of optic nerve atrophy. The purpose of this study was to assess the diagnostic utility of OCT for evaluation of ON in patients with FD.

Methods: Subjects with craniofacial FD underwent neuro-ophthalmologic evaluation and OCT imaging (N=70, mean age=23.5, range=5–66). ON was diagnosed clinically, based on visual acuity, visual fields, color vision, contrast vision, and fundoscopic examination. The diagnostic utility of RNFL thickness was determined using ROC curve analysis, and accuracy was compared to radiographic optic canal measurements. Longitudinal changes in RNFL thickness were assessed.

Results: 11 subjects had ON (8 unilateral/3 bilateral) for 14 total affected eyes. RNFL thickness accurately identified ON (AUC=0.997, P<0.0001) with 100% sensitivity and 97% specificity when using a threshold of ≤71 μm. RNFL thickness outperformed radiographic measurements of optic canal area and optic nerve length (P<0.05). In subjects with normal vision, RNFL thickness decreased at an average rate of 0.21 μm/year (P<0.01). When analyzed longitudinally, RNFL increased during adolescence and declined after puberty.

Conclusions: OCT measurement of RNFL thickness is a superior test for identifying ON than radiographic optic canal narrowing or optic nerve elongation. Using a threshold RNFL thickness ≤71 μm, clinicians can identify ON in FD patients with high sensitivity and specificity. By establishing the normal rate of age-related RNFL changes, clinicians can monitor for abnormal RNFL loss over time and identify at-risk patients prior to irreversible damage. These findings have immediate clinical application that may help identify appropriate surgical candidates, reduce unnecessary interventions, and decrease the incidence of blindness in FD.

Disclosure: The authors declared no competing interests.