Improving the diagnosis of retinopathy of prematurity: a leading cause of childhood blindness

Why is this research needed?

Retinopathy of prematurity (ROP) is a leading cause of childhood blindness, which occurs when the blood vessels that supply the retina develop in an abnormal way. When a baby is born pre-term, the retinal blood vessels have not had a chance to develop fully. In most cases, these blood vessels will continue to form normally after the baby is born. In some cases, however, abnormal blood vessels grow out of the retina, creating scar tissue that can cause the retina to detach, resulting in blindness.

ROP is classed as mild or severe according to how far it has advanced and how much of the retina is involved. Mild ROP will usually resolve by itself as the retina develops, but requires continuous monitoring, whereas severe ROP requires treatment.

If detected early, ROP can be treated - but the existing screening process is lengthy and distressing for babies, and is also inefficient. UK guidelines aim for 100% detection rate for severe ROP, so that no child misses treatment they might potentially need. In fact, fewer than 10% of babies born with ROP will go on to need treatment, but as current screening methods cannot predict which children are most at risk of developing the severe form of the disease, thousands of babies in the UK undergo invasive, and unnecessary, assessments every year.

Over the past 25 years, the development of optical coherence tomography (OCT), which enables the eye to be examined at near microscopic levels, has revolutionised the diagnosis and treatment of eye diseases in adults. Until recently, very young children have been deprived of this technology because the standard table-mounted devices used in OCT screening require the patient to remain still with their head in an upright position, and are therefore impossible to use with babies and infants.

Dr Proudlock’s team is exploring a new approach to screening using a hand-held optical coherence tomography device (HH-OCT) that completes the scan in less than 2 seconds without touching the baby’s eye. Results from a preliminary study suggest that this equipment could be helpful in predicting the onset of severe ROP, but a full scale study is needed to confirm its diagnostic potential.

What is the aim of the project

Dr Proudlock is leading a 3-year research project using the HH-OCT, alongside current screening methods, to gather data from premature babies admitted to the neonatal units at University Hospital of Leicester NHS Trust and Birmingham Women’s and Children’s Hospital NHS Foundation Trust.

All babies in the study will undergo routine screening sessions according to current guidelines. Demographic and clinical parameters such as birth weight, gestation age, sex, weight, respiratory status, and any other concurrent diseases will also be documented.

In addition, following promising findings from a previous study, babies born at a postmenstrual age of between 32 and 34 weeks will also undergo an examination with the HH-OCT, to capture images of the fovea and optic nerves of both eyes. Data relating to the retinal and optic nerve parameters will be analysed in order to determine whether this information can help to predict the onset of severe ROP and improve the accuracy of the current diagnostic model.

How will this research help to benefit patients?

If the HH-OCT can help to identify those babies for whom screening can be reduced or stopped earlier than is currently recommended, this study could help to transform clinical practice, sparing thousands of premature babies from distressing, lengthy and unnecessary examinations.

Thank you

This project is being made possible thanks to the generosity of several charitable trusts. In particular we would like to thank the H B Allen Charitable Trust, Sir Samuel Scott of Yews Trust, The C M Lowe Charitable Trust, Clara E Burgess Charity, The Mackintosh Foundation, The Sir Robert Gooch Trust, The Cowslip Green Charity 1994, and four further charitable trusts who wish to remain anonymous.