# Evaluating the long-term predictive value of macular thickness fluctuations on diabetic macular oedema response to anti-VEGF treatment

## TL;DR

This study found that changes in retinal thickness over time do not reliably predict long-term vision outcomes in diabetic macular oedema patients treated with anti-VEGF therapy.

## Contribution

The study provides new evidence that macular thickness fluctuations are not predictive of long-term visual acuity in diabetic macular oedema patients.

## Key findings

- Macular thickness variability was not significantly associated with changes in visual acuity at 3 and 5 years.
- Baseline visual acuity and number of anti-VEGF injections were predictive of long-term outcomes.
- Patients with the highest retinal thickness variability had lower final vision scores but not due to changes over time.

## Abstract

This study assesses the predictiveness of retinal thickness variability on long-term visual acuity (VA) outcomes in patients with diabetic macular oedema (DMO) undergoing anti-VEGF therapy.

Retrospective chart review at a single institution.

184 and 138 patients underwent initial treatment for DMO and continued to follow-up at 3 and 5 years, respectively.

Baseline demographics, past medical, and clinical data were collected through electronic medical record query. Central subfield thickness (CST), choroidal volume (CV), and cube average thickness (CAT) variability over the first year of treatment were calculated using standard deviation, amplitude, and area under the curve. CST, CV, CAT, and best corrected visual acuity (BCVA) were noted at the initial treatment date and 3 and 5 years. CST variability quartiles were compared on baseline, final, and change in BCVA alongside final and change in CST. Multiple linear regression was used to evaluate factors associated with final BCVA at 3 and 5 years while adjusting for confounders.

There is no significant association between larger fluctuations in macular thickness and change in BCVA at 3 and 5 years. Individuals in the highest CST variability quartile had the lowest final BCVA but also the lowest baseline BCVA and no difference in BCVA change. Linear regression revealed that CST variability was not predictive of final BCVA. Baseline BCVA and total number of anti-VEGF injections were predictive of BCVA at 3 years and 5 years respectively.

Macular thickness variability did not predict long-term visual outcomes at 3 and 5 years.