Aqueous cell differentiation in anterior uveitis using Fourier-domain optical coherence tomography

Abstract

Purpose: The differential diagnosis of a patient presenting with anterior uveitis is broad and can present a diagnostic challenge. In this study, we evaluate the characteristic findings of inflammatory cells on optical coherence tomography (OCT) both in vitro and in vivo.

Methods: Blood from two healthy volunteers was prepared using standardized methods for cell sorting with a flow cytometer (FASCAria). Neutrophils, lymphocytes, monocytes, and red blood cells were placed in suspension and scanned with a 26-kHz Fourier-domain OCT system (RTVue) with 5-μm axial resolution. Custom software algorithms were used to identify cells based on their reflectance distribution. These algorithms were then applied to OCT images obtained from uveitis patients with active anterior chamber inflammation.

Results: On OCT images the cells appeared as hyperreflective spots. In vitro, cell reflectance was statistically significantly different between all of the cell types (neutrophils, monocytes, lymphocytes, and red blood cells, P < 0.001, Mann-Whitney test). In vivo, the relationship between underlying disease and cell type imaged on OCT was highly statistically significant, with human leukocyte antigen (HLA)-B27-associated uveitis patients having a predominantly polymorphonuclear pattern on OCT and sarcoidosis and inflammatory bowel disease patients having a predominantly mononuclear pattern on OCT (P < 0.001, Fisher's exact test).

Conclusions: These in vitro and in vivo data demonstrate the potential of OCT to evaluate cells in the anterior chamber of patients noninvasively. Optical coherence tomography may be a useful adjunct to guide the diagnosis and treatment of ocular inflammatory conditions.

Keywords: Fourier-domain optical coherence tomography; anterior uveitis; image analysis; mononuclear leukocytes; polymorphonuclear leukocytes.

Figure 1

Optical coherence tomography (OCT) images of (A) neutrophils, (B) lymphocytes, (C) monocytes, and (D) red blood cells in suspension. The cells, identified by customized software, are marked in OCT images (circles) and magnified images (arrows). The locations of the right-side magnified images are marked in left-side OCT images with squares. The bright bands on top of the OCT image are artifact from the tube wall.

Figure 2

Histogram plots of cell reflectance. The x-axes are plotted on a logarithmic scale.

Figure 3

Optical coherence tomography (OCT) cross-sectional image (left) and cell reflectance histogram plots (right) of two eyes with anterior uveitis. (A) An HLA-B27-positive scleritis case with sudden onset associated with predominantly polymorphonuclear (PMN) leukocyte response. The OCT cell differentiation algorithm determined that there were 85% bright cells (PMN) and 15% dim cells (mononuclear) in the anterior chamber. (B) A sarcoidosis case characterized with a mononuclear infiltrate. The OCT cell differentiation algorithm determined that there were 100% dim cells (mononuclear) in the anterior chamber. The OCT cell differentiation results were consistent with the clinical diagnoses.