Sir,

Retinopathy of prematurity (ROP) affects 50% of babies screened and, of these, 4% will develop sight threatening (Type 1) ROP and require treatment.1 Recognition of progression to Type 1 ROP and timely laser treatment is essential to prevent permanent visual loss, which accounts for 3% of childhood visual loss in the UK.2

Current issues with ophthalmologist delivered screening in the UK include recruitment difficulties, inexperience of screeners and the lack of a certified programme of continued medical education. Wide-field digital retinal imaging is an effective and accurate method for ROP telemedicine but the two commercially available cameras are expensive, making largescale NHS purchase unfeasible.3, 4 An alternative is digital indirect ophthalmoscopy (DIO) which has been found to have between 94–100% sensitivity and 89–94% specificity for grading pre-plus or plus disease.5

The East of England neonatal network comprises 11 units referring to one treating centre, around one half of babies treated have been transferred from regional units specifically for ROP management (Figure 1). In 2016, nine of these units agreed to participate in an ROP telemedicine service. Implementation of the service reduced the number of unnecessary transfers from 56 to 21% and the proportion of treatments performed on babies with Type 2 disease from 21 to 7% (Figure 2). Three babies were transferred without images (two of these did not require treatment) and of those sent with images 17 out of 18 were gradable. Previously undiagnosed bilateral Stage 4 ROP was identified in one baby, enabling urgent transfer for vitreo-retinal surgery.

Figure 1
figure 1

Proportion of babies treated for ROP who were transferred from regional units.

Full size image
Figure 2
figure 2

Pie charts showing outcome of transfer (a) pre and (b) post telemedicine service implementation.

Full size image

Problems have included inadequate, and sometimes obstructive, information technology (IT) support from participating hospitals which has delayed implementation in several units, and issues relating to image transfer, which is currently onerous for the referring ophthalmologist.

The DIO telemedicine service has saved £21,000 (€23,300, $27,500) in transport costs in the first year alone. The recent British Ophthalmic Surveillance Unit study indicated that nearly 30% of babies treated have Type 2 disease.1 Potentially unnecessary laser treatment for Type 2 ROP, with all the risks it entails, was prevented in most cases by telemedicine in this study.

Telemedicine for ROP using wide-angle digital retinal imaging is likely to remain the gold standard and enables screening by technicians and nurses. The advantages of the DIO based telemedicine service is its affordability: for the cost of one wide-angle retinal camera, seven systems were purchased. The disadvantage of DIO based telemedicine is the continued need for an ophthalmologist to screen.

This service describes one method for ROP telemedicine, but all methods require local and national IT support and the use of a digital network such as Picture Archiving and Communications System (PACS). This would make telemedicine easier to use and facilitate image sharing and cross-cover between units and regions. Current difficulty with the recruitment and retention of experienced ophthalmologists underlines the importance of developing a telemedicine strategy to ensure optimal management and prevention of childhood blindness due to ROP.