Combined OCT and colour fundus photography in virtual clinic assessments of wet AMD patients

We would like to commend Hibbs et al¹ on their study emphasising the need for colour photography in screening patients for wet age-related macular degeneration (AMD). We agree with the authors that colour photography is essential, but would like to describe a service evaluation performed at Hull and East Yorkshire Eye Hospital. Data from this evaluation suggested that even with colour photography combined with optical coherence tomography (OCT), wet AMD patients with retinal haemorrhage and no other features of disease activity can be under-treated without slit lamp assessment.

In our evaluation, we used spectral domain OCT with associated colour fundus photography (Topcon 3D OCT-1000) to examine 242 patients with wet AMD undergoing follow-up, to assess the need for Ranibizumab therapy. Clinicians were asked to record re-treatment decisions based on imaging and ETDRS visual acuity only as a ‘virtual' assessment. Each patient was then immediately reviewed with slit lamp examination (SLE) and any change in management decision documented.

A total of 242 patients were evaluated and the results were as follows.

For 200 patient episodes, SLE did not provide any other useful clinical information. For 42 patients, the SLE provided additional information when compared with virtual assessment. We further examined these questionnaires.

In 21 patients ocular pathology unrelated to Ranibizumab therapy was detected, whereas in 6 patients OCT quality was too poor to allow virtual assessment. Of these 27 patients, SLE made no change in management of any ophthalmic disease for 11 patients. In our cohort, only 15 of the 242 (6.2%) patients had haemorrhage on SLE that was otherwise not detected on OCT/colour photography alone, but in only 6/15 (2.5% of the total) were there no other associated OCT features indicating disease activity.

We concluded that in our cohort 2.5% of patients would require SLE to detect isolated retinal haemorrhage undetected with colour imaging where there were no other OCT features of disease activity. However, in the context of the capacity issues associated with reviewing all patients on a monthly basis, it was felt that this was an acceptable percentage when set against the benefits of higher throughput in virtual clinics.

The authors declare no conflict of interest.