Abstract
Purpose: :
To report the 4-year experience of the Stanford University Network for Diagnosis of Retinopathy of Prematurity (SUNDROP) telemedicine initiative to reduce the risk of blindness from retinopathy of prematurity (ROP).
Methods: :
Retrospective analysis of the SUNDROP archival data gathered between 12/1/05 and 11/30/09 to evaluate the efficacy of this diagnostic technology for ROP screening. Wide-angle digital fundus RetCam II cameras were installed in the four neonatal intensive care units (NICUs) comprising the SUNDROP network. A total of 337 infants meeting ROP examination criteria were consecutively enrolled. Nurses obtained five or six images in each eye. All patients were screened at least once using the RetCam II and evaluated remotely at the SUNDROP reading center at Stanford University without a simultaneous indirect ophthalmological examination unless this was eventually indicated by photographic findings. Outcomes included treatment-warranted ROP (TW-ROP) and anatomical outcomes. TW-ROP was defined as any Type 2 disease or greater, threshold disease, any plus disease, and any stage 4 or higher disease, in following with Early Treatment Retinopathy of Prematurity study recommendations. All patients were examined via indirect ophthalmoscopy within one week of discharge.
Results: :
The SUNDROP telemedicine screening initiative has not missed any TW-ROP in its 4-year evaluation period. A total of 337 infants (674 eyes) were imaged, resulting in 1410 unique examinations and 16,513 unique images. An average of 12.0 images was obtained per patient (range of 2-40 images). Fourteen infants were identified with TW-ROP; thirteen underwent laser photocoagulation and completely regressed. The one remaining patient regressed spontaneously, and did not require intervention. The calculated sensitivity was 100%, with a specificity of 99.7% for detection of TW-ROP. No patient progressed to retinal detachment, vision loss, or other adverse anatomical outcomes.
Conclusions: :
The SUNDROP telemedicine initiative for ROP demonstrates a high degree of sensitivity (100%) and specificity (99.7%) for identification of treatment-warranted disease. All cases of treatment-warranted disease were captured, and staff training was easily implemented. There were no adverse anatomical outcomes.
Keywords: retinopathy of prematurity • clinical (human) or epidemiologic studies: health care delivery/economics/manpower