Purchase this article with an account.
Aleksandr Birg, Giovannella Carboni, Gina Forma, Grazyna Adamus, Alessandro Iannaccone; Spectral-Domain Optical Coherence Tomography (SD-OCT) Findings in Patients with Autoimmune Neuro-Retinopathy (AINR). Invest. Ophthalmol. Vis. Sci. 2011;52(14):3690.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To analyze SD-OCT scans from AINR patients and compare them to controls and cases of confirmed retinitis pigmentosa (RP) to characterize SD-OCT changes in AINR.
SD-OCT data were analyzed for 22 AINR patients, each positive for at least 1 anti-optic nerve and/or anti-retinal auto-antibody (auto-Ab). The most common auto-Ab recognized alpha-enolase (8/22). More than half of patients had 2 or more Auto-Ab bands, and as many as 5. Custom segmentation analyses on SD-OCT horizontal macular scans obtained at presentation (mostly with Spectralis, Heidelberg Eng.) were performed with ImageJ software (repeated x3 and averaged). We measured mean total retinal thickness (MTRT) and, separately, retinal nerve fiber layer (RNFL), outer nuclear layer (ONL), photoreceptor inner (IS) and outer (OS) segments. Percent interocular differences (IOΔ%) between layers were also calculated as an index of disease asymmetry. These analyses were done also on SD-OCTs from 5 controls and 6 RP patients representative of various phenotypes and inheritance modes. Conventional peripapillary (PP-) RNFL analyses (Cirrus, Zeiss Meditec) were also characterized in AINR patients.
Custom segmentation: MTRT in AINR (284±37µ) was similar to controls (294±8µ) but thicker than RP (256±41µ; p-value vs. AINR: 0.051; vs. controls: 0.009). However, in AINR, RNFL was thicker than normal (p=0.0186) [especially in anti-alpha-enolase-positive patients (p=0.0304)], the ONL was marginally thinner (p=0.0443), and IS and OS were markedly thinner (p=0.0001 & 0.0032, resp.) than normal. Unlike AINR, all layers were much thinner in RP vs. controls (including the ONL, which was 42% less than normal, p=0.01x10-7) but the RNFL, which was normal in thickness. No significant IOΔ% was seen for any layer in RP, but seen for all layers in AINR (p<0.01 in all cases but the RNFL). The RNFL was symmetric in 17/22 AINR cases but 14-31% asymmetric in 5/22 (normal: 3.4±2.4%). PP-RNFL: Both thickening (71%) and thinning (52%) of the RNFL in at least 1 quadrant were found in AINR, and RNFL symmetry between eyes was markedly diminished (on avg., only 67%).
AINR has distinctive imaging features that differentiate it from RP. MTRT is an inadequate measure of retinal health in AINR as RNFL thickening can mask thinning of other layers. RNFL changes are very common in AINR, which is consistent with the frequent clinical finding of changes at or around the disc in AINR (Rhadakrishnan et al. ARVO 2009).
This PDF is available to Subscribers Only