Clinical Characteristics and Management Outcomes of Rhegmatogenous Retinal Detachments Due to Retinal Dialysis

Abstract

Purpose:

This work assesses the clinical characteristics and outcomes of various treatment methods in managing rhegmatogenous retinal detachments secondary to retinal dialysis.

Methods:

In this retrospective, consecutive case series, patients presenting to a tertiary referral vitreoretinal service from 2015 to 2020 were chosen. The primary outcome measure was the single-surgery anatomic success (SSAS) rate in the first 90 days after surgery and overall SSAS rate until the last visit.

Results:

Eighty-six eyes of 84 patients were included. Mean (SD) age was 30 (14.8) years (range, 7-71 years). Fifty-nine (70%) patients were men and 55 eyes (64%) had a history of trauma. Dialysis occurred in the inferotemporal quadrant in 50 eyes (58%) of the entire cohort and 25 of 55 eyes (45%) with prior trauma. Fifty-one eyes (59.3%) were managed by scleral buckle (SB), 25 eyes (29.2%) with laser retinopexy, and 10 eyes (11.6%) with pars plana vitrectomy (PPV). The SSAS rate in the first 90 days after surgery was 94.1% (81 of 86 eyes), including 90.1% of eyes with SB, 100% of eyes with laser retinopexy, and 100% of eyes with PPV (P = .07). The overall SSAS rate, however, was 89.5% (77 of 86) eyes, including 90.1% of eyes with SB, 84% of eyes with laser retinopexy, and 100% of eyes with PPV (P = .4). Final attachment rate was 100% in the entire cohort.

Conclusions:

With careful consideration of underlying pathology, all 3 treatment modalities can provide high rates of anatomic and functional success.

Introduction

Retinal dialysis is a circumferential retinal break located at the ora serrata, and it is a common cause of retinal detachments (RDs) in children and young adults. ¹ Unlike a retinal tear, in a retinal dialysis the retina remains attached to the vitreous base by its posterior edge. ² Ocular trauma is the most common cause of retinal dialysis, with a reported prevalence of 55% to 88% of all unilateral dialyses. ^3,4 Bilateral retinal dialyses are far less common and may be due to a developmental anomaly rather than trauma alone. ³

Retinal dialysis with minimal or no subretinal fluid (SRF) can be managed by in-office barrier laser photocoagulation. ⁵ Scleral buckling (SB) is the long-standing treatment of choice for RD due to retinal dialysis, and it has yielded good results. ⁶ This technique preserves the lens and accommodation in young patients while avoiding premature detachment of the posterior hyaloid face that would occur during vitrectomy. ⁷ SB may also avoid the need for positioning and reduces the risk of premature cataract formation, iatrogenic retinal breaks, and gas- or silicone oil–related complications. ⁸ However, postoperative SB-related complications including exposure, infection, strabismus, and subretinal hemorrhage are potential risks. ^9,10

In recent years, pars plana vitrectomy (PPV) has emerged as an increasingly popular alternative to SB in repairing RDs. ^11,12 A recent study showed that PPV is an effective technique for RD due to retinal dialysis in cases where a primary SB alone is insufficient or challenging because of media opacity. ¹³ With continued advancement in technique and instrumentation, PPV could serve as a reliable alternative in selected retinal dialysis cases. ¹³ Nevertheless, SB and PPV both are invasive procedures; alternative methods such as laser retinopexy may be a preferred option in eligible eyes.

In this study we describe the outcomes of different interventions, including SB, laser retinopexy, and PPV, for retinal dialysis. We report the baseline clinical features, single-surgery anatomic success (SSAS) rate, and visual outcomes of these 3 methods for repair of rhegmatogenous RD (RRD) secondary to retinal dialysis at a single vitreoretinal tertiary referral center. We also investigate potential correlations between patients’ baseline characteristics and the intervention chosen for retinal dialysis repair.

Methods

A retrospective analysis was performed of all patients regardless of age diagnosed with and treated for retinal dialysis at the retina service of Wills Eye Hospital and the offices of Mid Atlantic Retina, Philadelphia, PA from February 2015 to December 31, 2020. A computerized search was performed of both electronic health record systems using International Statistical Classification of Disease and Related Health Problems, 10th Revision diagnosis codes for retinal dialysis (H33.041-H33.049). A medical-record review was performed for each patient to confirm the diagnosis of retinal dialysis. Patients with a history of successfully treated retinal dialysis from an outside provider were excluded from the study.

Patient demographics were collected including any history of ocular trauma, defined as a patient-reported history of ocular trauma or examination findings consistent with ocular trauma. The following baseline characteristics at the time of diagnosis were collected: duration of symptoms, best available Snellen visual acuity (VA) with habitual correction or pinhole, intraocular pressure, lens status, previous RD, prior ophthalmic surgical procedures, prior laser treatments, and prior cryotherapy. The type of trauma was also recorded for each patient. Preoperative examination findings of choroidal detachment or hemorrhage, vitreous hemorrhage (VH), retinal hemorrhage, uveitis, posterior vitreous detachment, ocular hypertension, demarcation lines, and proliferative vitreoretinopathy (PVR) were recorded. Postoperative VA was documented at 3 months and at the final visit.

Types of interventions included SB placement, laser indirect ophthalmoscopy, and PPV. Intraoperative surgical data collected included the location and extent of dialysis, macular involvement, presence and extent of RD, presence of other retinal breaks, presence of lattice degeneration, presence of posterior vitreous detachment, and method of SRF drainage (if applicable). The chronicity of RD was noted by the presence of retinal demarcation lines and de novo PVR. Recorded data also included the settings for laser indirect ophthalmoscopy, SB or band type, PPV tamponade used, and method of SRF drainage. All subsequent medical and surgical interventions after the initial treatment were also recorded.

Spectral-domain optical coherence tomography images were individually reviewed by 2 observers (D.H. and M.S.) to confirm the macula-on or macula-off status, defined as the absence or presence, respectively, of SRF in the center of the macula. Digital retinal drawings were reviewed by 2 observers (D.H. and R.M.). The images of the retina were divided into 4 quadrants centered at the fovea: superotemporal, superonasal, inferotemporal, and inferonasal. The primary outcome measure was VA change from before surgery to final follow-up. The secondary outcome measures were the rate of final attachment and the rate of SSAS, defined as complete retinal attachment after SB and PPV and stable demarcation of SRF after laser treatment with no further intervention within 90 days.

All data were analyzed with SPSS (IBM SPSS Statistics, version 22.0; IBM Corp). Snellen VA was converted to logMAR for analysis. Descriptive statistics were performed using mean and SD for continuous measures and proportions for categorical variables. The McNemar test was used to compare the proportion of categorical variables before and after surgery. The Pearson χ² test was used to compare proportions between the study groups. The normal distribution of data was assessed using the Shapiro-Wilk test. For normally distributed data, continuous variables were analyzed with an independent 2-sample t test. One-way analysis of variance was used to compare the differences in continuous variables between 2 or more groups. For nonparametric data, continuous variables were analyzed with the Wilcoxon signed rank test. The Kruskal-Wallis test was used to compare the differences in continuous variables between 2 or more groups. A P value of less than .05 was considered to be statistically significant.

Results

Eighty-six eyes of 84 patients were included in this study. Mean (SD) age was 30 (14.8) years (range, 7-71 years). Fifty-nine (70%) patients were men. Forty-six (54%) cases involved the right eye. History of trauma was present in 55 eyes (64%). Of the 31 eyes (36%) that had no known cause, 2 patients had bilateral retinal dialysis. Fifty-five eyes (64%) were macula-on detachments, and 85 eyes (98.8%) were phakic at the time of presentation. Eleven patients (13%) had dialysis in more than 2 quadrants. The overall inferotemporal involvement rate was 58% (50 of 86 eyes) in all patients and 45% (25 of 55 eyes) in trauma cases. Table 1 shows the baseline characteristics of all cases and trauma groups separately.

Table 1.

Baseline Characteristics of All Eyes and Traumatic Eyes Diagnosed With Dialysis.

	All eyes (N = 86)	Traumatic eyes (n = 55)
Age, mean ± SD, y	30 ± 14.8	28.4 ± 14.5
Male sex, No. (%)	59 (70)	40 (72.7)
Right eye laterality, No. (%)	46 (54)	27 (49.1)
Macula status, No. (%)
On	55 (64)	37 (67.3)
Off	31 (36)	18 (32.7)
Phakic, No. (%)	85 (98.8)	55 (100)
IOP, mean ± SD, mm Hg	16.2 ± 5.2	16.6 ± 5.8
VH, No. (%)	14 (16.3)	13 (23.6)
PVD, No. (%)	10 (11.6)	7 (12.7)
Dialysis quadrant, No. (%)
Inferotemporal	50 (58)	25 (45)
Superotemporal	17 (20)	9 (16.4)
Superonasal	12 (14)	15 (27.3)
Inferonasal	3 (3.5)	–
Dialysis, mean ± SD, clock h	2.7 ± 1.7	2.4 ± 1.7
RD, mean ± SD, clock h	4.3 ± 2.8	4.6 ± 2.9
Dialysis, mean ± SD, quadrants	1.9 ± 0.6	1.9 ± 0.6
RD, mean ± SD, quadrants	2.4 ± 0.8	2.4 ± 0.9
PVR presence, No. (%)	18 (20.5)	11 (20)
PVR grade, No. (%)
A	9 (10)	6 (11)
B	6 (7)	3 (5.4)
C	3 (3.5)	2 (3.6)
Presence of other breaks, No. (%)	17 (19.8)	12 (21.8)
Lattice degeneration, No. (%)	13 (15.1)	7 (12.7)
LogMAR VA, mean ± SD (Snellen)	0.73 ± 0.82 (20/107)	0.74 ± 0.75 (20/109)
Duration of symptoms, mean ± SD, d	32.6 ± 48	27.9 ± 40.1
Follow-up duration after surgery, mean ± SD, d	514.7 ± 522.7	455.3 ± 511.2

Abbreviations: IOP, intraocular pressure; PVD, posterior vitreous detachment; PVR, proliferative vitreoretinopathy; RD, retinal detachment; VA, visual acuity; VH, vitreous hemorrhage.

Among eyes with trauma, a history of altercation was the most common cause (19 of 55, 34.5%) followed by sports (18 of 55, 32.7%), projectile trauma (7 of 55, 12/7%), and motor vehicle accidents (5 of 55, 9.1%). Among cases related to projectile trauma, Nerf darts (Hasbro Inc) accounted for 7.2% (4 of 55 eyes) of dialysis in traumatic cases. Table 2 summarizes the cause of dialysis in cases with a history of trauma.

Table 2.

Cause of Dialysis in Eyes With a History of Trauma.

	All traumatic eyes (n = 55)
Sports	18 (32.7%)
Soccer	5
Baseball	3
Bungee cord	2
Box	2
Football	1
Tennis	1
Hockey stick	1
Basketball	1
Swimming	1
Skateboard	1
Altercation	19 (34.5%)
Motor vehicle accident	5 (9.1%)
Projectile	7 (12.7%)
Nerf dart	4
Unknown	6 (10.9%)

Fifty-one of 86 eyes (59.3%) were managed by an SB, 25 of 86 (29.2%) with laser retinopexy, and 10 of 86 (11.6%) with PPV alone. (No eyes undergoing primary PPV had SB.) The baseline size of RD and dialysis extension was significantly larger in patients who underwent PPV (P = .02 and P = .04, respectively). Baseline VA was significantly worse in the PPV group (P < .001). VH was present in 100% of the PPV group, a significantly higher incidence compared with the other groups (P = .003). The duration of symptoms before undergoing surgery was significantly longer in the PPV group (P < .001). PVR was present in 15% of patients who underwent PPV, which was more than 2 other groups but not by a significant degree (P = .3). There was no macula-off eye in the laser-treated patients, which was significantly different from the other 2 groups (P < .001). The baseline characteristics of eyes managed by different surgical methods are shown in Table 3.

Table 3.

Baseline Characteristics of Eyes Based on Intervention Method.

	SB (n = 51, 59.3%)	Laser retinopexy (n = 25, 29.2%)	PPV (n = 10, 11.6%)	P value
	% or mean ± SD	% or mean ± SD	% or mean ± SD
Age, y	28.8 ± 14.3	30.1 ± 14.1	35.2 ± 19.2	.47a
Size of dialysis, in clock h	2.7 ± 1.8	1.6 ± 1.3	2.9 ± 1.3	.04a
Size of RD, in clock h	4.8 ± 2.8	2.7 ± 2.2	4.6 ± 3.3	.02a
Preoperative macula status
On	49	100	40	<.001b
Off	51	0	60
Mean logMAR VA ± SD (Snellen)	0.89 ± 0.8 (20/155)	0.22 ± 0.3 (20/33)	1.23 ± 0.9 (20/339)	<.001c
No. of other breaks	1.1 ± 1.1	0.33 ± 0.5	1.7 ± 0.9	.2a
Trauma history	29	36	50	.4b
VH presence	12	20	100	.03b
PVR presence	10	5	15	.3b
Phakic	100	100	90	.01b
Dialysis location
Superior	39	25	40
Inferior	47	70	40	.02b
Both	14	5	20
Duration of symptoms, d	33.4 ± 41.1	11.2 ± 14.7	75.2 ± 64.2	<.001a

Abbreviations: PPV, par plana vitrectomy; PVR, proliferative vitreoretinopathy; RD, retinal detachment; SB, scleral buckle; VA, visual acuity; VH, vitreous hemorrhage.

^a Analysis of variance test.

^bχ² test.

^cKruskal-Wallis test.

The majority of patients who underwent SB surgery had cryopexy (92.2%). For the eyes repaired with SB only, most cases (78.4%) were encircled with either a 41 or 42 band (Dutch Ophthalmic) (35.3% and 43.1%, respectively) and 21.6% had placement of a circumferentially oriented, segmental element. Most of these eyes (62.7%) did not have drainage of SRF, and 43.1% of eyes received gas or oil tamponade. The most common tamponade was perfluoropropane (C₃F₈) gas in 81% of cases. Unlike those with SB surgery, 80% of patients who underwent PPV had SRF drainage (internally) and 100% had retinal tamponade. Sulfur hexafluoride (SF₆) was used as the tamponade agent in 40% of these eyes, silicone oil in 30%, C₃F₈ in 20%, and air in 10%. The mean (SD) laser power used in laser retinopexy–only cases was 272.2 mW (66.4 mW). Mean (SD) laser spot duration was 60.1 ms (18.1 ms) and mean (SD) number of laser spots that were applied was 531.2 (332.6). The details of surgical interventions are summarized in Table 4.

Table 4.

Intervention Details.

	No. (%) or mean (± SD)
Buckle
Band type
41	18/51 (35.3)
42	22/51 (43.1)
Other	11/51 (21.6)
SRF drainage
Yes	19/51 (37.3)
No	32/51 (62.7)
Tamponade
Yes	22/51 (43.1)
No	29/51 (56.9)
Tamponade type
Air	1/22 (5)
C3F8	18/22 (81)
SF6	3/22 (14)
Retinal cryopexy
Yes	47/51 (92.2)
No	4/51 (7.8)
Laser retinopexy
Laser characteristics
Power, mW	272.2 (± 66.4)
Duration, ms	60.1 (± 18.1)
No. of spots	531.2 (± 332.6)
PPV
SRF drainage
Yes	8/10 (80)
No	2/10 (20)
Tamponade
Yes	10/10 (100)
Tamponade type
SF6	4/10 (40)
C3F8	2/10 (20)
SO	3/10 (30)
Air	1/10 (10)

Abbreviations: C₃F₈, perfluoropropane; PPV, pars plana vitrectomy; SF₆, sulfur hexafluoride; SO, silicone oil; SRF, subretinal fluid.

SSAS was defined as a successfully attached retina with no further intervention within 90 days. Patients who underwent SB had an SSAS rate of 46 of 51 (90.1%) eyes, which included 23 of 25 (92%) macula-on and 23 of 26 (88.5%) macula-off eyes. All patients (25 of 25) who received laser and all patients (10 of 10) who underwent PPV had a 100% SSAS rate (P = .07).

Of the entire cohort, 9 of 86 (10.5%) required a second surgery. Of these 9 patients, 5 (55.6%) had received SB as the first intervention and 4 (44.4%) had received laser retinopexy. The overall SSAS rate was 77 of 86 (89.5%) for the entire cohort, including 46 of 51 (90.1%) in SB, 21 of 25 (84%) in laser retinopexy, and 25 of 25 (100%) PPV patients, which was not statistically different between the 3 groups (P = .4). The mean (SD) number of days between the first and second surgery in SB patients was 40 (19.8) (range, 9-64 days) and in laser patients was 883 (799) (range, 138-1881 days). The laser-treated group had a significantly longer interval until redetachment (P = .002). The overall mean (SD) days between the first and second surgery was 415 (661) (range, 9-1881 days). Five of 9 patients (55.6%) had a history of trauma, whereas 4 of 9 had no history of trauma (P = .2). The overall final attachment rate was 100% in all cohorts and subgroups at a mean (SD) of 17.1 (17.4) months (range, 4.2-76.4 months). Table 5 shows the outcomes of the different intervention methods.

Table 5.

Surgical Outcomes Based on Intervention Method.

	SB (n = 51, 59.3%)	Laser retinopexy (n = 25, 29.2%)	PPV (n = 10, 11.6%)	All (n = 86)
	No. (%) or mean ± SD	No. (%) or mean ± SD	No. (%) or mean ± SD	No. (%) or mean ± SD
SSAS
Macula-on	23/25 (92)	25/25 (100)	7/7 (100)	55/57 (96.4)
Macula-off	23/26 (88.5)	–	3/3 (100)	26/29 (89.6)
Overall	46/51 (90.1)	25/25 (100)	10/10 (100)	81/86 (94.1)
Final attachment
Macula-on	25/25 (100)	25/25 (100)	7/7 (100)	57/57 (100)
Macula-off	26/26 (100)	–	3/3 (100)	29/29 (100)
Overall	51/51 (100)	25/25 (100)	10/10 (100)	86/86 (100)
Mean logMAR VA ± SD (Snellen)
Baseline	0.89 ± 0.8 (20/155)	0.22 ± 0.3 (20/33)	1.23 ± 0.9 (20/339)	0.73 ± 0.82 (20/107)
3 mo after surgery	0.85 ± 0.7 (20/141)	0.21 ± 0.3 (20/32)	1.05 ± 0.9 (20/224)	0.69 ± 0.69 (20/98)
P valuea	.68	.77	.75	.72
Final visit	0.72 ± 0.67 (20/104)	0.09 ± 0.09 (20/24)	1.1 ± 0.9 (20/252)	0.59 ± 0.69 (20/77)
P valuea	.09	.4	.07	.01
Final VA gainb	–0.17 ± 0.67	–0.12 ± 0.65	–0.15 ± 0.35	–0.16 ± 0.58c

Abbreviations: PPV, pars plana vitrectomy; SB, scleral buckle; SSAS, single-surgery anatomic success; VA, visual acuity.

^a Wilcoxon signed ranked test.

^bKruskal-Wallis test.

^c P equals .8.

The mean (SD) final VA (0.59 [0.69], Snellen equivalent, 20/77) was significantly improved compared with baseline VA (0.73 [0.82], Snellen 20/107) in all cohorts regardless of procedure type (P = .01). There was no significant difference in the change in VA from baseline between the different procedures, and patients had similar improvements in vision (P = .8). Mean (SD) final VA in patients with macula-off detachments (0.81 [0.59], Snellen 20/129) was worse than those with macula-on detachments (0.45 [0.72], Snellen 20/56, P = .02). Patients with SB had worse mean (SD) visual outcomes in macula-off eyes (0.85 [0.56], Snellen 20/141) than macula-on eyes (0.56 [0.76], Snellen 20/72, P = .01). The amount of VA gain was also significantly different between macula-on and macula-off patients in all cohorts and SB eyes (P < .001). Patients who underwent PPV had no significant difference in the mean (SD) final VA between macula-off (1.31 [0.99], Snellen 20/408) and macula-on eyes (1.05 [0.91], Snellen 20/224, P = .2). The VA gain was also not different in PPV eyes (P = .2).

Two of 4 macula-on cases had severe trauma and either secondary severe cataract or secondary choroidal rupture with final VA of hand motion and counting fingers, respectively, which adversely affected the mean final VA in the macula-on group of 4 eyes as a whole. There was no patient with macula-off RD who initially underwent laser retinopexy. Table 5 summarizes the visual outcomes based on the different interventions. Figure 1 shows the widefield (Optos) fundus photographs that were taken after laser retinopexy in a localized macula-on inferotemporal RRD secondary to retinal dialysis. There was no history of trauma in this eye and the presenting VA was 20/25, which was stable after 2 years of follow-up. Figure 2 shows an example of a widefield fundus photograph of a patient with a history of projectile trauma 14 days prior to presentation. Examination revealed near-total macula-off RD with nasal dialysis. He underwent SB and his VA improved from counting fingers to 20/100 over 9 months.

Figure 1.

A localized inferotemporal retinal dialysis that was treated by laser retinopexy and had stable vision (20/25) after 2 years of follow-up.

Figure 2.

Traumatic retinal dialysis and subsequent total macula-off retinal detachment before and after scleral buckle. Widefield fundus photograph of right eye (A) before surgical repair with an encircling scleral buckle and (B) two months after repair demonstrates complete retinal reattachment.

Conclusions

Our study analyzed the baseline clinical characteristics, SSAS rate, and visual outcomes of eyes with RRD secondary to retinal dialysis. In this series, we found a comparably high rate of SSAS between SB, laser retinopexy, and PPV. On average, the patients who underwent PPV had a larger size of dialysis and associated RD, worse baseline VA, a higher rate of VH, a higher rate of trauma, and a longer duration of symptoms prior to diagnosis. In general, patients who had laser retinopexy had the smallest size of dialysis and RD, better baseline VA, greater incidence of inferiorly located dialysis, sparing of the macula, and a shorter duration of symptoms before diagnosis. Despite these significant baseline differences, the eyes in each intervention group had an overall similar SSAS rate, visual gain, and final attachment rate. These findings may help to provide guidance for surgeons deciding on a treatment modality for dialysis repair using initial clinical presentation and patient baseline characteristics.

The overall SSAS rate in our study was 94.1%, which is comparable to the primary reattachment rate of 92% described by Kwong and colleagues and 95.8% by Jan et al. ^1,10 Recently Yap et al reported the outcomes of PPV with or without circumferential SB in 30 patients with RD secondary to retinal dialysis. Of the 30 eyes that underwent primary PPV, 24 also received encircling SB. Only 6 eyes had PPV alone as treatment. The authors documented primary anatomical success in 90% of all eyes. ¹³ Indications for PPV in their study included difficult retinal visualization and PVR (23% of eyes). In our study, the main reason for undergoing vitrectomy was hazy media in the setting of VH and PVR.

To the best of our knowledge, this is the first paper that describes the baseline features and outcomes of patients who underwent laser retinopexy to demarcate an RD associated with retinal dialysis. The rate of SSAS and visual outcome supports the use of laser retinopexy as an effective treatment option, with the primary failure rate being comparable to the SB and PPV groups. However, redetachment following laser retinopexy occurred later in the disease course. Of the 4 delayed redetachments in the laser group, 2 eyes were in a 14-year-old patient with bilateral idiopathic dialysis. Although the retina remained attached in both eyes up to 90 days post procedure, redetachment occurred by the end of the first year after initial presentation. The other 2 redetachments were in unilateral cases that happened approximately 3 and 5 years after the initial laser retinopexy.

Based on these findings, it is important for both patients and providers to be aware of the possibility of late progression of the detachment through the laser barricade, which occurred at a relatively high rate of 16% in our series. Caution should be exercised when considering laser alone for the treatment of dialysis-associated RD if a relatively large amount of SRF is present.

Chignell and Ross have each reported that the prevalence of idiopathic, bilateral inferotemporal dialysis was 7% and 4%, respectively, of all retinal dialysis cases. ^2,14 Our study found a prevalence of 2.4% (2 of 84) for bilateral idiopathic dialysis in all retinal dialysis cases, a lower percentage compared with the published literature. Brown and Tasman have described familial retinal dialysis in 2 brothers. One patient had presented with bilateral inferotemporal retinal dialyses and the other with unilateral inferotemporal dialysis. They recommended that when a patient with bilateral nontraumatic retinal dialyses is discovered, other family members should be examined. ¹⁵

Our analysis showed a history of trauma in about 64% of dialysis cases, which is similar to previous reports of 57% and 55%. ^4,13 In our study, the inferotemporal quadrant was the most involved section of retina overall and in eyes with a history of trauma (58% and 45%, respectively). Animal models previously hypothesized that direct trauma to the eye’s center could cause superonasal dialysis, whereas temporal trauma generally results in inferotemporal dialysis. ¹⁶ In other studies, inferotemporal involvement of the retina has been reported in 43.8% to 81% of all dialysis cases. ^1,6,9,10 Despite the relatively wide range, there is consensus in these studies that the inferotemporal quadrant is the most commonly involved area, similar to what was found in our study.

Limitations of this study are inherent in its retrospective nature, with selection bias in surgical repair choice by the treating physician. The use of Snellen VA with the patient’s usual correction or pinhole and without a standardized, best-refracted VA measurement may have underestimated VA results. We also included all patients treated for the first time in this center and excluded all other eyes with prior interventions from other centers or providers. This may underestimate the rate of failure in different treatment groups.

This study demonstrates that SB, laser retinopexy, and PPV are all potentially useful methods in the management of RDs secondary to retinal dialyses. Although primary SB is considered the initial treatment of choice given the generally young age at presentation, surgeons may consider PPV and laser retinopexy in select eyes with relatively good anatomical and functional success.