Anti-vascular endothelial growth factor dosing frequency and visual outcomes in macular oedema following branch retinal vein occlusion

Disposition and patient characteristics

Overall, 53,683 eyes with MO secondary to BRVO were assessed for eligibility (Fig. 1). Of these, 9439 received their initial anti-VEGF injection between 1 January 2012 and 31 May 2016 and had a VA reading on the same day as the index injection. After excluding eyes that did not have all required quarterly VA readings, those without sex identification, and those who had treatment breaks longer than 11 months during follow-up, 3099 and 1469 eyes were included in the year 1 and year 2 cohorts, respectively. In total, 47% (n = 1469) of eyes in the year 1 cohort qualified for inclusion in the year 2 cohort. Seven percent (n = 222) of eyes in the year 1 cohort did not have any visits in year 2, and the remainder were excluded due to incomplete VA readings at baseline or every quarter, or treatment breaks longer than 11 months over the 2 years of follow-up.

Fig. 1: Disposition of eyes with MO secondary to BRVO.figure 1

aEyes from 3070 patients. bEyes from 1457 patients. BRVO branch retinal vein occlusion, MO macular oedema, VA visual acuity, VEGF vascular endothelial growth factor.

In the year 1 cohort, 12% (n = 387) of eyes received aflibercept, 30% (n = 937) received bevacizumab, 31% (n = 972) received ranibizumab, and 26% (n = 803) received more than one anti-VEGF therapy. In the year 2 cohort, 15% (n = 217) of eyes received aflibercept, 30% (n = 436) received bevacizumab, 28% (n = 413) received ranibizumab, and 27% (n = 403) received more than one anti-VEGF therapy. Injection frequencies were similar between anti-VEGF treatment types used for both year 1 and year 2 (Supplementary Table 1).

Regardless of anti-VEGF treatment type, eyes in the year 1 cohort (n = 3099) received a mean of 7.2 (median, 7.0) injections during year 1, and were divided into 2 subcohorts according to receipt of ≤6 injections (1197/3099; 38.6%) or ≥7 injections (1902/3099; 61.4%) through year 1. Baseline characteristics, including baseline vision, were similar between the ≤6-injections and ≥7-injections subcohorts (Table 1). In year 1, 24.6% (295/1197) of eyes in the ≤6-injections subcohort and 21.1% (402/1902) of eyes in the ≥7-injections subcohort had received concomitant steroid and/or laser therapy, with a mean (range) of 4.3 (2–6) treatments and 9.5 (7–13) treatments, respectively (Table 1).

Table 1 Demographic and baseline characteristics of patients with MO secondary to BRVO in the year 1 cohort by injection frequency during year 1.Year 1 outcomes

For eyes that were treated for ≥1 year, those in the ≤6-injections subcohort received a mean of 4.6 (range, 2–6) injections and eyes in the ≥7-injections subcohort received a mean of 8.8 (range, 7–14) injections in year 1. In the ≤6-injections subcohort, mean VA increased from 53 letters at baseline to 63 letters after 1 year of treatment (Fig. 2), and in the ≥7-injections subcohort, from 52 letters at baseline to 66 letters at year 1. Mean VA gain from baseline was significantly lower in the ≤6-injections subcohort than the ≥7-injections subcohort at year 1 (10.4 vs 13.9; p < 0.001). Consistent with this finding, a sensitivity analysis indicated similarly lower mean VA gains in eyes that received ≤6-injections compared with those that received ≥7-injections among a subgroup of eyes that did not receive steroids or laser (n = 2402; 11.1 vs 14.3, p < 0.001), as well as those that received steroids and/or laser (n = 697; 8.5 vs 12.3, p = 0.041) (Supplementary Fig. 1). Eyes that received laser treatment received a similar mean number of anti-VEGF injections during year 1 as eyes that did not receive laser (7.0 vs 7.3 injections).

Fig. 2: Mean VA through year 1 in eyes with MO secondary to BRVO in the ≤6-injections and ≥7-injections subcohorts.figure 2

BRVO branch retinal vein occlusion, BSL baseline, ETDRS Early Treatment Diabetic Retinopathy Study, MO macular oedema, Q quarter, SE standard error, VA visual acuity.

To assess the relationship between the frequency of treatment and visual and anatomic outcomes, eyes were further stratified into 4 subcohorts based on injection frequency: 1–3, 4–6, 7–9, and ≥10 injections over year 1. Across the 4 subcohorts of eyes for which VA was available, mean VA ranged from 52–55 letters at baseline and improved to 63–67 letters at year 1 (Fig. 3). Overall and on average, VA gain increased with increasing injection frequency. In a subset of these eyes with both VA and foveal thickness measurements available, a trend showing improved (thinner) foveal thickness with increasing injection frequency was also observed, with eyes in the 1–3-injections subcohort being outliers in this trend as their visual and anatomic outcomes were relatively better than those in the 4–6 injection subcohort (Supplementary Fig. 2). In Year 1, 3.9% (121/3099) of eyes received a mean (range) of 2.7 (2–4) injections in Q1 with no additional injections in Q2, Q3, or Q4; these eyes were all in the ≤6-injections subcohort.

Fig. 3: Mean VA change through year 1 by injection frequency during year 1 in eyes with MO secondary to BRVO.figure 3

Analysis included eyes with available VA measurements through Year 1. BRVO branch retinal vein occlusion, BSL baseline, ETDRS Early Treatment Diabetic Retinopathy Study, MO macular oedema, Q quarter, VA visual acuity.

Year 2 outcomes

Eyes that received ≤6 injections in year 1 and also subsequently received ≤6 injections in year 2 (n = 250) received a mean of 5.2 and 4.2 injections in years 1 and 2, respectively. At the start of year 2, mean VA of these eyes was 64 letters, and this score was maintained through the end of year 2 (Fig. 4). Eyes that received ≤6 injections in year 1 and ≥7 injections in year 2 (n = 41) were administered a mean of 5.2 and 7.9 injections in years 1 and 2, respectively. These eyes started year 2 with a mean VA of 64 letters and ended year 2 with a mean VA of 66 letters (p = 0.69).

Fig. 4: Mean VA at the start and end of year 2 by injection frequency in years 1 and 2 in eyes with MO secondary to BRVO.figure 4

*P < 0.001 compared with the change in the ≥7/≥7-injections subcohort. BRVO branch retinal vein occlusion, ETDRS Early Treatment Diabetic Retinopathy Study, MO macular oedema, VA visual acuity.

Eyes that received ≥7 injections in year 1 and subsequently ≤6 injections in year 2 (n = 605) received a mean of 7.5 and 4.6 injections in years 1 and 2, respectively. These eyes started year 2 with a mean VA of 67 letters and ended year 2 with a mean of 64 letters, showing an average loss of 3 letters (Fig. 4). Eyes that received ≥7 injections in year 1 and continued to receive ≥7 injections in year 2 (n = 573) received a mean of 9.7 and 8.5 injections in years 1 and 2, respectively. These eyes started year 2 with a mean of 67 letters and ended year 2 with a mean of 68 letters, gaining an average of 0.7 letters. The change in mean VA between the start and end of year 2 in the subcohort of eyes that received ≤6 injections in year 2 and ≥7 injections in year 1 differed significantly from the subcohort of eyes that received ≥7 injections in both years 1 and 2 (–3 vs +1 letter; p < 0.001). It is important to note that patients who received ≥7 injections in years 1 and 2 avoided loss of approximately one line of VA compared with those that received ≥7 injections in year 1 and ≤6 injections in year 2 with an additional three anti-VEGF injections per year.

In a subset of eyes in the ≤6-injections or ≥7-injections subcohorts for both years 1 and 2, and for which measurements for both VA and foveal thickness at baseline and all 8 quarters through year 2 were available, there was a trend toward increased VA and decreased foveal thickness with increasing injection frequency (Supplementary Fig. 3).

Eyes that received laser treatment received a similar mean number of anti-VEGF injections during year 2 as eyes that that did not receive laser (14.3 vs 14.4 injections).

Annual trend in treatment frequency during year 1

The mean number of anti-VEGF injections given during year 1 was similar between 2012 and 2016 for both the ≤6-injections subcohort (4.4–4.7 injections) and the ≥7-injections subcohort (8.6–8.9 injections) (Supplementary Fig. 4). However, numerically greater proportions of eyes received ≥7 injections during year 1 between 2014 and 2016 (63–65%) than in either 2012 (41%) or 2013 (51%) (Supplementary Fig. 4).

留言 (0)

沒有登入
gif