OCT-A After Macular Hole Surgery Can Show Retinal Microvasculature Changes

Full thickness macular hole surgery improves best corrected visual acuity in many eyes, in addition to restoration of retinal vascular density in the deep vascular plexus, according to research results published in Retina. 

In the current study, researchers sought to evaluate changes in retinal microvasculature after vitreoretinal surgery, determining whether those changes might influence visual recovery. While the routine use of optical coherence tomography angiography (OCT-A) has created a better understanding of changes during the disease process, differing results of OCT-A have been described in the literature. 

Researchers evaluated a retrospective, observational case series of 33 eyes from 33 patients (26 women) with a full thickness macular hole who underwent epiretinal membrane removal. All patients underwent preoperative ophthalmic evaluation, measurement of best-corrected visual acuity (BCVA) and slit-lamp examination of anterior and posterior segments, among other measurements. Postoperatively, patients were evaluated at 1, 3, and 6 months. 

The study cohort included 14 right eyes and 19 left eyes. At postoperative month 1, all eyes had complete full thickness macular hole closure. BCVA improvement was seen in 21 eyes, while 12 eyes did not improve; BCVA at baseline, 1, 3, and 6 months was 58.94±9.7, 66.42±10, 70.8±9.8, and 72.18±9.9 ETDRS, respectively. 

At baseline, macular hole diameter was 470±166.28 µm. 

Over the whole area, the vascular density (VD) of the superficial vascular plexus (SVP) was 21.27±1.9, 22.06±1.6, 21.85±1.8, and 21.70±1.9 at baseline, 1 month, 3 months, and 6 months, respectively. VD of the deep vascular plexus (DVP) at the same timepoints was 14.82±2.7, 16.58±3.4, 17.73±3.3, and 18.39±3.2, respectively. 

Researchers considered the status of the foveal ellipsoid zone (EZ) and external limiting membrane (ELM) at baseline and the end of follow-up as a predictor of postoperative macular sensitivity restitution. Foveal EZ and ELM were evaluated in each eye to test integrity and determine if they were intact or disrupted. Intact eyes had a “regular and continuous hyperreflective line” corresponding with the EZ or ELM. Grading was achieved by the agreement by 2 authors. 

There was a statistically significant overall difference in inner SVP at all time points. Pairwise comparison showed that the only statistically significant difference was between the mean SVP (inner) at baseline and 1 month. There were no statistically significant differences in whole or outer SVP at varying timepoints, but post hoc comparisons highlighted significant differences in mean outer and whole SVP between baseline and 1 month. 

VD evaluation demonstrated a statistically significant overall difference between DVP (whole), DVP (inner), and DVP outer at all timepoints. 

DVP features were also further evaluated in 2 groups — those with and without postoperative BCVA improvement — to identify possible vascular parameters from OCT-A that were related to patients’ visual prognosis. 

BCVA improvement was seen in 21 eyes. Eyes without BCVA improvement generally experienced “very little” VD restoration postoperatively. There were also no statistically significant differences in VD of the DVP in any 3 areas tested between eyes with or without an improvement in BCVA. 

Repeated measures ANOVA showed a statistically significant change in mean BCVA between timepoints; post hoc Bonferroni correction tests showed that BCVA increased from baseline at all postoperative visits, with a statistically significant change in FAZ area at each visit postoperatively. 

A 2-way mixed ANOVA was performed to determine the possible effect of age on BCVA. Patients between 50 and 65 years of age showed an increase in mean BCVA of 15.17 ETDRS letters, while those >65 years of age had an increase of 10.93 ETDRS letters. Pairwise comparisons showed that for all possible visit combinations, differences in BCVA were statistically significant within each group. 

Study limitations include challenges in determining the correct segmentation, the retrospective study design, and the small sample size. 

“The results of our study suggest that the retinal vascular plexues are affected by vitrectomy and ILM peeling during [full thickness macular hole] surgery,” according to the researchers. “Further multicenter studies and…artificial intelligence could be used…to determine if preoperative VD may be a prognostic indicator of visual recovery after vitrectomy.”

Reference

Savastano A, Bacherini D, Savastano MC, et al. Optical coherence tomography angiography findings before and after vitrectomy for macular holes: Useful or useless? Retina. 2021;41(7):1379-1388. doi:10.1097/IAE0000000000003059