Comparing the diagnostic accuracy of telemedicine utilization versus in-person clinical examination for retinopathy of prematurity in premature infants: a systematic review.

Purpose: To synthesize the literature assessing the diagnostic accuracy of telemedicine evaluation compared with clinical examination for retinopathy of prematurity (ROP) in premature infants.

Methods: Covidence software was used to conduct a systematic literature search from September 14, 2020, through September 27, 2020, on MEDLINE (Ovid), EMBASE (Ovid), CINAHL, and the gray literature to identify studies relevant to telemedicine utilization for ROP detection. After duplicate removal and two-levels of screening, studies comparing telemedicine evaluation with binocular indirect ophthalmoscopic examination were included. Risk of bias assessment was conducted for the included studies following data extraction. A qualitative review was performed to summarize estimates of accuracy of ROP evaluation by telemedicine.

Results: A total of 507 studies were reviewed, of which 323 were found in EMBASE, 115 in MEDLINE, and 79 in CINAHL. Three possibly relevant conference abstracts were found. Following duplicate removal, 410 studies were reviewed based on titles and abstracts. Subsequently, 19 articles were thoroughly examined, and 14 studies (2,655 participants) were included. Most studies found that telemedicine performance for detecting ROP was comparable to ophthalmic examination, especially with regard to identifying treatment-requiring ROP.

Conclusions: Telemedicine evaluation can reliably detect ROP. Incorporation of telemedicine into conventional neonatal care has the potential to improve access to ROP care.