Abstract
Purpose: :
Optos® Panoramic200A utilizes scanning laser ophthalmoscope technology to capture ultra–wide field, high–resolution images (2000 x 2000 pixels). It has a potential advantage over standard digital photography by imaging a wider field of retina (approximately 200o vs 50o). In this collaborative trial, we compared fluorescein angiograms obtained by these two separate technologies to quantify possible differences in detection of diabetic retinopathy.
Methods: :
33 eyes of 33 diabetic patients from 3 centers (UPMC Eye Center, Jules Stein Eye Inst, and Miami VAMC) were recruited to this open, non–randomized, pilot trial. Each patient underwent fluorescein angiography on 2 separate occasions. On the first visit, the Panoramic200A was used and 1 week later, the same eye was imaged with a standard digital camera (Zeiss, Topcon TRC 501X and/or A or RC6000). For each eye, both an arteriovenous phase image and a late phase image were reviewed. An image review grid, centered on the macula, was digitally placed on each image and indexed to the specific optic nerve diameter to insure uniformity. The grid, comprised of 12 concentric rings (each 1 disc diameter (DD) apart) was subdivided into 8 equal 45 degree sectors. The combination of rings and sectors created as many as 96 image grid blocks on a single image, depending upon how far out each eye was imaged. Results from the Panoramic 200A and conventional camera images using the above grid were then graded and tabulated by a masked reviewer.
Results: :
The mean extent of the retina imaged for the Panoramic200A and standard digital camera was 8.7±1.60 DD and 3.4±0.76 DD respectively. The mean number of image sectors containing retinal ischemia was 16.9±15.04 and 3.4±4.26 in the arteriovenous phase and 17.6±16.77 and 3.1±3.36 in the late phase respectively. The mean number of sectors containing retinal neovascularization was 1.8±3.16 (arteriovenous phase) and 2.6±3.35 (late phase) for the Panoramic200A, versus 0.5±1.28 (arteriovenous phase) and 0.6±1.19 (late phase) for the standard digital camera. The differences between these sets of values were each statistically significant (p<0.01).
Conclusions: :
The Panoramic200A imaged a significantly larger area of the retina and allowed more lesions to be displayed within the mid–peripheral and peripheral retina. The ability of a clinician to readily detect clinically relevant lesions in a single image with the Panoramic200A in diabetic patients may result in better overall care of the retinopathy in these patients.
Keywords: imaging/image analysis: clinical • diabetic retinopathy