Telemedicine a viable solution for ROP evaluation

November 15, 2006

Calgary, Alberta-Serial digital fundus imaging with remote interpretation by experienced physician graders is accurate for detecting clinically significant retinopathy of prematurity (CSROP), according to the results of the Photographic Screening Trial for Retinopathy of Prematurity (PhotoROP).

Calgary, Alberta-Serial digital fundus imaging with remote interpretation by experienced physician graders is accurate for detecting clinically significant retinopathy of prematurity (CSROP), according to the results of the Photographic Screening Trial for Retinopathy of Prematurity (PhotoROP).

The study also found that detection of CSROP through telemedicine occurred on average 1.5 weeks earlier compared with the gold standard of bedside evaluation with indirect ophthalmoscopy, reported Anna L. Ells, MD, a retinal specialist and pediatric ophthalmologist at Alberta Children's Hospital, Calgary, Canada, and clinical associate professor of ophthalmology, University of Calgary, Alberta.

The PhotoROP was an international, prospective, longitudinal, multicenter cohort study. At weekly intervals, the enrolled premature infants were evaluated with indirect ophthalmoscopy and had digital fundus photographic images obtained (RetCAM-120 Digital Retinal Camera, Clarity Inc.) using a standard image captivation protocol. The images were transferred via the Internet to a reading center where they were interpreted by two expert physician graders.

"A telemedicine approach has the potential to address accurately and efficiently the tremendous burden and need for a better evaluation program for ROP, and without missing severe disease, which left untreated will progress to blindness," Dr. Ells said. "Further, because it may result in earlier detection of clinically significant ROP and enable earlier treatment, it may also result in a better visual outcome. We know from the Early Treatment for ROP (ET-ROP) Study that early identification and treatment of severe disease is important for best visual prognosis."

Dr. Ells was one of the investigators in PhotoROP. The principal investigators for the study were Antonio Capone Jr., MD, and Michael T. Trese, MD, both of Royal Oak, MI, who performed the remote interpretation of the digital images.

Challenges of ROP evaluation

Evaluation of premature infants for ROP currently poses numerous challenges relating to staffing limitations, geographic diversity, and medicolegal liability. Children at risk need to be evaluated regularly during the 8 to 12 weeks after birth when ROP can develop. Marked regional differences exist with respect to features defining infants at greatest risk and access to care. In addition, ROP diagnosis presents a tenuous medicolegal scenario, because it results in one of the highest average settlements paid out in ophthalmology.

"The potential solution is to design an ROP program that evaluates all infants eligible for an examination but refers for on-site expert evaluation only the approximately 10% most at risk for severe disease. Telemedicine allows high-volume screening and also enables appropriate access to subspecialty care," Dr. Ells explained.

Six clinical centers participated in the PhotoROP. Eligible infants had to have a gestational age under 31 weeks and a birth weight under 1,000 g. Infants were included in the data analysis if they had at least two digital exams. Referral was warranted for eyes diagnosed by the reading center as having CSROP, which was defined in this study as any ROP in zone 1, stage 2 or 3 ROP in zone 2, or one quadrant of plus disease. This definition was established to identify eyes requiring close scrutiny with ample lead time prior to the need for laser treatment.

Nearly half of the enrolled infants were Caucasian (45% of total), and almost all of the rest were African-American (39% of total). Ninety-two percent of all digital images captured were interpretable by the reading center.

The children had between two and 10 serial examinations. The high rate of CSROP diagnosed in this study (81%) is reflective of the inclusion criteria, Dr. Ells explained.

Related Content:

News