Analysis of Choroidal Thickness Changes in Recurrent Rhegmatogenous Retinal Detachment

Article information

Abstract

Materials and Methods

Study design

We included patients diagnosed and treated for recurrent RRD between November 2019 and March 2023. Recurrent RRD was defined as stable retinal attachment just after surgery and detached again thereafter. Patients with high myopia (axial length ≥28 mm or spherical equivalent ≤–6), macular hole RD, traumatic RD, age-related macular degeneration, pachychoroidal neovascularization, retinal vascular occlusion, or diabetic retinopathy were excluded as these conditions could affect choroidal thickness. Recurrence within 2 weeks and aged <18 years were also excluded.

Baseline choroidal thickness was defined as the choroidal thickness with macular adhesions immediately prior to the time of recurrence detection and collected at 2 and 6 months after reoperation. Choroidal thickness was defined as the distance from the lower part of the retinal pigment epithelial layer to the choroid-sclera interface on the horizontal section of the optical coherence tomography (OCT), and thickness measurement was performed using the method proposed by Manjunath et al. [7] and Ban et al. [8] (Fig. 1). In brief, measurement was done using built-in OCT software at 500-μm intervals up to 2,500 μm temporal and nasal to the fovea and mean value was used as choroidal thickness. Spectral-domain OCT with enhanced depth imaging (Spectralis, Heidelberg Engineering) was utilized. Choroidal thickness was measured regardless of tamponade status. Manual measurements made by two researchers using manual calipers were compared, and the average of the two measurements was used. If t he measurements of the two investigators differed by more than 10%, a third investigator arbitrated and the mean of the values differing by less than 10% was used. The choroidal thickness of the contralateral eye was also measured at the same time point and used as a control for analysis. Interobserver reliability was evaluated for choroidal thickness measurement.

Fig. 1

Optical coherence tomography (OCT) image shows choroidal thickness in the region of macular involvement due to retinal detachment. The OCT measurements were performed at 500 μm intervals across a 6 × 6-mm posterior pole cube scan. Perpendicular lines were drawn from the posterior boundary of the retinal pigment epithelium to the choroid/sclera interface using the built-in OCT software. Double arrows illustrate an example of measuring choroidal thickness.

In addition to choroidal thickness, factors such as age, sex, slit-lamp examination results, spherical equivalent, axial length, surgical methods used during the initial and subsequent surgeries, time until recurrence of RRD after the first surgery, extent of RD (superior/inferior), presence of macular detachment, and proliferative vitreoretinopathy (PVR) were evaluated. Best-corrected visual acuity, intraocular pressure, and central retinal thickness were also examined prior to recurrence, and 2 and 6 months after reoperation. The average retinal thickness within the central 1,000-μm area from the OCT macular thickness map was used to determine the central retinal thickness (CRT).

Results

A total of 33 patients were enrolled in the study. The mean age of the patients was 54.5 ± 17.3 years, with 20 men (60.6%) and 13 women (39.4%). At the time of recurrence, the state of the macula was attached in 24 eyes (72.7%) and detached in 9 eyes (27.3%). Extent of RRD showed superior detachment in 9 eyes (27.3%), inferior detachment in 22 eyes (66.7%), and both superior and inferior involvement in 2 eyes (6.1%). At the time of recurrence, 29 eyes were diagnosed with PVR (87.9%), with 6 eyes (18.2%) showing PVR grade A, 12 eyes (36.4%) showing PVR grade B, and 11 eyes (33.3%) showing PVR grade C. The median axial length was 25.5 mm (range, 22.6–27.7 mm). The mean time to recurrence after the primary surgery was 126.8 days (median, 60 days; range, 30–1,139 days) (Table 1).

Table 1

Characteristics of subject patients at the time of recurrence (n = 33)

At the primary surgery, 8 eyes (24.2%) underwent explant buckling, 22 eyes (66.7%) had pars plana vitrectomy (PPV) alone, and 3 eyes (9.1%) had both explant buckling and PPV. During reoperation, 26 eyes (78.8%) received PPV alone, and 7 eyes (21.2%) had both encircling and PPV. The surgical methods and types of tamponades used during reoperation are listed in Table 2.

Table 2

Summary of surgical techniques and tamponades for primary and secondary RRD (n = 33)

The baseline choroidal thickness for the recurrent and fellow eyes was 201.32 ± 80.54 and 200.12 ± 78.41 μm, respectively. At postoperative 2 months, the measurements were 195.18 ± 72.15 μm for the recurrent eye and 198.93 ± 78.85 μm for the fellow eye. At postoperative 6 months, measurements were 186.32 ± 72.96 μm for the recurrent eye and 197.46 ± 77.63 μm for the fellow eye (Fig. 2). When comparing the choroidal thickness of the recurrent and fellow eyes, a significant decrease in choroidal thickness over time was observed in the recurrent eye, but not in the fellow eye (repeated measures ANOVA; recurrent eye: Greenhouse-Geisser, p = 0.009 and Huynh-Feldt, p = 0.009; fellow eye: Greenhouse-Geisser, p = 0.214 and Huynh-Feldt p = 0.213). In the post hoc analysis, there was no significant difference from baseline to 2 months. However, significant reductions in choroidal thickness were observed in the recurrent eye from baseline to 6 months (p = 0.089) and from 2 to 6 months (p = 0.010; p = 0.012, after applying Bonferroni correction, with a significance level set at p < 0.0167).

Fig. 2

Choroidal thickness changes after secondary surgical intervention of recurrent rhegmatogenous retinal detachment. The recurrent eye showed a significant decrease in choroidal thickness over time (baseline, 201.32 ± 80.54 μm; 2 months, 195.18 ± 72.15 μm; 6 months, 186.32 ± 72.96 μm; Greenhouse-Geisser, p = 0.009), while the fellow eye remained relatively stable (baseline, 200.12 ± 78.41 μm; 2 months, 198.93 ± 78.85 μm; 6 months, 197.46 ± 77.63 μm; Greenhouse-Geisser, p = 0.214). In the post hoc analysis, choroidal thickness in the recurrent eye showed significant reductions from baseline to 6 months (p = 0.010) and from 2 to 6 months (p = 0.012; significance level, p < 0.0167 after Bonferroni correction).

The baseline CRT for the recurrent and fellow eyes was 313.44 ± 115.08 and 308.21 ± 95.76 μm, respectively. At postoperative 2 months, the measurements were 291.72 ± 80.37 μm for the recurrent eye and 310.58 ± 96.23 μm for the fellow eye. At postoperative 6 months, the measurements were 311.34 ± 97.16 μm for the recurrent eye and 311.51 ± 94.45 μm for the fellow eye.

In the univariate analysis, change in choroidal thickness showed correlation with change in CRT from baseline to 2 months (r = 0.347, p = 0.048). However, no significant correlation was observed from 2 to 6 months (p = 0.497) and from baseline to 6 months (p = 0.260). Worse grading of PVR showed a greater decrease of choroidal thickness from 2 to 6 months and from baseline to 6 months, with correlation coefficients of −0.448 (p = 0.009) and −0.404 (p = 0.020), respectively. No significant correlation was found from baseline to 2 months (p = 0.229) (Table 3). In the multivariate analysis, an association was identified between the changes in choroidal thickness from 2 to 6 months and the grade of PVR (p = 0.030) (Table 4).

Table 3

Univariate analysis of clinical variables associated with choroidal thickness changes

Table 4

Multivariate analysis of clinical variables associated with choroidal thickness changes

Our study demonstrated interobserver reliability for choroidal thickness measurements, with a single measures ICC of 0.984 (95% confidence interval, 0.367–0.996; p < 0.001) and an average measures ICC of 0.992 (95% confidence interval, 0.537–0.998; p < 0.001)

Discussion

The results of this study showed that the choroidal thickness in eyes with recurrent RRD was statistically significantly reduced compared to the fellow eye 6 months after surgical treatment. The change in choroidal thickness varied over time but was correlated with the grade of PVR and the change in central retinal thickness.

Several studies have reported on the changes in choroidal thickness after primary RRD surgery. Some studies have shown a significant increase in subfoveal choroidal thickness immediately after surgery, and from postoperative 1 month, the thickness returned to the level of a normal eye [4,9]. A report from Korea observed a significant decrease in choroidal thickness in the second month after surgery, and a significant decrease in choroidal thickness was also observed in the group that had more than 30 days elapse from symptom onset to the day of surgery [8]. This finding aligns with the results of our study. However, a limitation of all previous studies is that they set the control group as choroidal thickness in the fellow eye.

As the results of this study, changes in choroidal thickness vary over time. In this study, patients who showed recurrence within 2 weeks were excluded, as it was difficult to distinguish recurrence from primary failure. As a result, among the patients enrolled in the study, the shortest duration between the first and second surgeries was 30 days, with a median of 60 days. Therefore, it was assumed that for these patients, the choroidal thickness had stabilized after the initial adhesion between the neurosensory retina and the choroid considering normalization after 1 month [4], and therefore the authors speculated that baseline choroidal thickness was measured after this stabilization period and can be used as baseline value.

In this study, the change in CRT and the change in choroidal thickness showed a correlation, with a positive correlation observed between baseline and 2 months. After that period, no correlation could be found. Among the 33 eyes studied, cystoid macular edema was present in 6 eyes at the 2-month and 17 at the 6-month mark, which may have made it difficult to find a significant relationship between CRT and choroidal thickness after the secondary surgery.

In the multivariate analysis, change of choroidal thickness significantly associated with the grade of PVR. During RRD, many metabolic processes are disrupted, which may affect choroid circulation to change in choroidal thickness. PVR also occurs in RD especially in chronic cases. We could assume that chronicity might account for association of PVR and choroidal thickness, although there was no correlation between time to recurrence and change in choroidal thickness. Further larger study should be done.

After attachment, the thickness of the choroid decreased, and one possible mechanism for this could be the effect of silicone oil tamponade. According to previous reports, there are several studies suggesting that the use of silicone oil leads to a decrease in choroidal thickness thereafter [5,10,11]. The direct pathophysiological mechanism behind the decrease in central choroidal thickness due to silicone oil has not yet been clarified. However, there are study results suggesting that central retinal degeneration occurs due to the failure of Müller cells to absorb potassium due to silicone oil toxicity. This, in turn, could have a negative impact on the central choroidal blood flow, leading to a decrease in central choroidal thickness [12,13]. Additionally, there are reports suggesting that the infiltration of silicone oil into the retina can cause neural damage to ganglion and bipolar cell synapses, leading to a decrease in macular function [14]. There are reports suggesting that mechanical stress from silicone oil tamponade and inflammation caused by the subretinal migration of silicone oil can also be contributing factors [15–17]. As other possible mechanisms, there are reports that performing laser photocoagulation not only destroys the structure of the retina but also reduces the thickness of the choroid [18,19]. In the current study, among the tamponades used for retinal reattachment surgeries, silicone oil was used in 81.8% of cases. Based on previous research, the observed decrease in choroidal thickness in this study could potentially be attributed to the use of silicone oil [5,10,11]. However, no significant differences were found between the groups according to the type of tamponade used in this study.

We attempted to analyze the duration of re-detachment, defining it as the period from symptom onset to the date of reoperation, or from the date of detection to the date of reoperation. However, we were unable to derive meaningful results. In all cases, reoperation was performed within 2 weeks of detection, with many cases proceeding as emergency operations on the day of detection.

The limitations of this study include the small number of subjects, which may limit statistical interpretation, and the lack of analysis on the area of PVR and laser treatment extent, preventing the analysis of choroidal thickness changes according to PVR area and laser treatment area. The inability to clarify the correlation between choroidal thickness and visual function is thought to be due to the retrospective nature and the small number of subjects. Chronicity should be evaluated as our group reported it is related with change in choroidal thickness in primary RRD [8]. Due to the retrospective nature of the study, we could not be able to get precise duration of re-detachment. Additionally, we could not be able to address how PVR influences choroidal thickness through its pathophysiology. Despite these limitations, this study is significant as it is a report to measure the changes in choroidal thickness in the same eye, rather than the contralateral eye, before and after surgery in patients with RRD.

In conclusion, in patients with recurrent RRD, the thickness of the choroid significantly decreased after reattachment of the retina, and the greater the severity of PVR, the more the choroidal thickness decreased.

Article information Continued