PURPOSE: To report the 24-month experience of the Stanford University Network for Diagnosis of Retinopathy of Prematurity (SUNDROP) telemedicine initiative. METHODS: Retrospective analysis of the SUNDROP archival data gathered between 1 December 2005 and 30 November 2007 to evaluate this diagnostic technology for ROP screening. One hundred and sixty consecutively enrolled infants meeting ROP examination criteria were screened with the RetCam II and evaluated by the SUNDROP reading centre at Stanford University. Nurses obtained five or six images in each eye. All patients also received a dilated examination within 1 week of discharge. Outcomes included treatment-warranted retinopathy of prematurity (TW-ROP) and anatomical outcomes. RESULTS: In the initial 24-month period, the SUNDROP telemedicine initiative has not missed any TW-ROP. A total of 160 infants (320 eyes) were imaged, resulting in 669 exams and 7556 images. Seven infants were identified with TW-ROP; six underwent laser photocoagulation and one regressed spontaneously. The sensitivity was 100%, with specificity of 99.4%. No patient progressed to retinal detachment or other adverse outcomes. CONCLUSION: The SUNDROP telemedicine screening initiative for ROP has been proven to have a high degree of sensitivity and specificity for the identification of treatment-warranted disease. All cases of treatment-warranted disease were captured. There were no adverse outcomes.