Feasibility of the Digital Retinography System Camera in the Pediatric Emergency Department

Direct ophthalmoscopy may be difficult without pupillary dilation and patient cooperation. Nonmydriatic ocular fundus photography (NMOFP) has been shown to be easily and efficiently accomplished by medical providers and improve the detection of abnormalities in adult emergency department (ED) patients. Nonmydriatic ocular fundus photography for pediatric ED patients has not been studied. The purpose of this study was to assess the ease of use of the Digital Retinography System (DRS) camera for NMOFP in ED patients aged 5 to 12 years and the quality of retinal images obtained with the DRS.

Retinal images were obtained with the DRS by a pediatric emergency medicine physician using a convenience sample of ED patients aged 5 to 12 years. Time to procedure completion, patient cooperation (Likert scale 1-5, with 5 being most cooperative), and satisfaction with the images (Likert scale 1-5, with 5 being completely satisfied) were recorded. Any satisfaction score less than 5 required the physician to describe a reason for dissatisfaction (brightness, field of view, focus). An ophthalmologist was consulted regarding any abnormal image. The accompanying parent completed a survey following the procedure. Estimated time to completion of the procedure and a rating of the overall comfort and cooperation of the child during the procedure (Likert scale 1-5) were recorded. A second pediatric emergency medicine physician reviewed all images and rated the level of satisfaction, reasons for dissatisfaction, and whether the images were normal. Descriptive statistics were used to analyze survey responses. A Mann-Whitney U test was used to compare continuous data for age groups 5 to 8 and 9 to 12 years. A Krippendorff α or κ coefficient was used to measure agreement between the physician obtaining the images and the secondary reviewer for image satisfaction and image abnormalities.

One hundred three patients were enrolled: 50 aged 5 to 8 years and 53 aged 9-12 years (mean, 9.1 [SD, 2.1] years). Five patients failed to cooperate, and no images were obtained. The mean length of time (LOT) to procedure completion was 1.8 (SD, 0.86) minutes. Overall, mean cooperation score was 4.4, and mean image satisfaction score was 4.6. One or more reasons for image dissatisfaction were given in 27 patients (imperfect focus most commonly). There was moderate agreement between the 2 physicians for image satisfaction (Krippendorff α coefficient = 0.48) and image abnormalities (κ coefficient = 0.38). Mean LOT did not differ between 5- to 8-year-olds and 9- to 12-year-olds (P = 0.23). Older patients had higher mean cooperation scores and image satisfaction scores (P < 0.001 and P = 0.04 respectively). Parental mean score for perceived LOT was 4.6 (5 = very short), 4.8 for patient comfort (5 = very comfortable), and 4.8 for patient cooperation (5 = very cooperative).

Our data suggest that NMOFP using the DRS camera is a rapid and easy method of obtaining high-quality images of the retina in pediatric ED patients.