Factors associated with the response to fluocinolone acetonide 0.19 mg in diabetic macular oedema evaluated as the area-under-the-curve

Abstract

Objectives: The area-under-the-curve (AUC) measures the average drug effect over time. We investigated the impact of baseline clinical and optical coherence tomography (OCT) factors on the response to fluocinolone acetonide (FAc) 0.19 mg implant in patients with diabetic macular oedema (DMO) as the AUC over 36 months.

Methods: Retrospective study of DMO eyes undergoing FAc with follow-up from 12 to 36 months. The AUC of the best-corrected visual acuity (BCVA) and the central macular thickness (CMT) were calculated with the trapezoidal rule. Demographic and clinical data at the time of FAc administration were collected, and associations with BCVA and CMT changes were investigated with linear mixed models.

Results: Eighty-nine eyes of 63 patients were enroled; median follow-up was 26 months. Mean±standard deviation (SD) AUC_BCVA and AUC_CMT after FAc injection were 0.24 ± 0.17 LogMAR/month and 179.6 ± 54.3 μm/month, respectively. Worse baseline BCVA (β = 0.30 LogMAR/month, p < 0.001), higher AUC_CMT after FAc administration (β = 0.08 LogMAR/month, p < 0.001), diagnosis of type 1 diabetes (β = -0.04 LogMAR/month, p = 0.04), and absent ELM/EZ layers (β = 0.06 LogMAR/month, p = 0.01) were associated with worse vision over time (higher AUC_BCVA). Eyes with higher CMT at baseline (β = 9.61 μm/month, p < 0.001) and those with tractional DMO (β = 24.7 μm/month, p = 0.01) had worse anatomic outcomes (higher AUC_CMT). The need for additional treatments after FAc was also associated with higher AUC_CMT (β = 33.9 μm/month, p = 0.001).

Conclusion: Baseline better visual acuity, lower macular thickness, and photoreceptors' layers integrity are associated with better functional response to FAc in DMO. Eyes with severe DMO at the time of implant or tractional oedema have worse anatomic response. These findings might guide clinicians in a more informed decisional algorithm in treating DMO.

Fig. 1. Method of calculation of the area-under-the-curve for the best-corrected visual acuity (BCVA, A) and central macular thickness (CMT, B) over the follow-up.

For each eye, the BCVA and CMT values were plotted as a function of follow-up time. Each visit corresponded to a different observation (x axis). Adjacent time points (showed in magenta) were joined by a straight line, forming multiple sub-intervals with a trapezoidal shape. The areas of each trapezium were summed together, giving the total AUC for each eye. Panels A and B show the curve of the same eye.

Fig. 2. Changes in the best-corrected visual acuity (BCVA) after fluocinolone acetonide implant injection.

A Boxplots illustrating the overall changes in BCVA. Straight bold lines within the boxplot indicate the median values, diamonds the mean values, and single points the outliers. B BCVA changes with eyes stratified as a function baseline visual acuity. Eyes with worse baseline BCVA had the greatest improvement after FAc implant, but the worst BCVA at the end of follow-up. C Boxplots of the cohort stratifying eyes based on the history of trabeculectomy. Eyes undergoing trabeculectomy had a worse vision over time. D Linear correlation between the mean macular thickness over time (AUC_CMT) and the mean visual acuity (AUC_BCVA) over time. The regression coefficient was calculated with a Pearson’s correlation.

Fig. 3. Risk of trabeculectomy after fluocinolone acetonide implant injection.

The figure displays a Kaplan–Meier curve of the cumulative probability of trabeculectomy in all the patients who underwent fluocinolone acetonide implant injection (n = 112 eyes).