Abstract
Introduction
Based on published studies, untreated symptomatic retinal breaks lead to rhegmatogenous retinal detachments (RRD) in 35% to 47% of patients,1,2 carrying the incumbent risk of long-term permanent visual impairment. 3 Laser retinopexy, the current standard treatment, applies thermal energy to induce adhesions between the retina and retinal pigment epithelium (RPE) to block the ingress of subretinal fluid (Figure 1, A–D), thereby reducing the rate of RRD. 4 The American Academy of Ophthalmology Preferred Practice Pattern recommends prompt treatment of acute symptomatic horseshoe tears; however, the management recommendations are left to the physician’s discretion for operculated holes associated with acute posterior vitreous detachment and asymptomatic horseshoe tears. 5
(A) Initial fundus photograph of the left eye demonstrating a horseshoe tear in the superotemporal periphery. Indirect laser photocoagulation was performed, with 2 to 3 rows surrounding the horseshoe tear without direct treatment of the tear. (B) Fundus photograph of the left eye 4 weeks after laser retinopexy. (C) Fundus photograph of the left eye 5 weeks after laser retinopexy. (D) Fundus photograph of the left eye 23 weeks after laser retinopexy.
Contemporary data with extended follow-up from large tertiary referral practices in the United States are limited. Prior studies have often focused on short-term RRD rates, with less emphasis on differentiating retreatment of the original break from newly identified breaks, fellow-eye involvement, and clinical factors associated with escalation to additional laser or surgical intervention. In the present study, we report the rates of subsequent RRD, retreatment of the original break vs formation of a new break, and fellow-eye laser retinopexy after the initial treatment of symptomatic retinal break(s). In addition, the clinical factors associated with additional laser retinopexy or vitreoretinal surgery are identified.
Methods
This retrospective interventional case series was conducted at the Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, with approval by the Institutional Review Board of the University of Miami and in accordance with the Declaration of Helsinki. All patients diagnosed with a retinal break and treated with laser retinopexy prophylaxis (Current Procedural Terminology code 61745) at any Bascom Palmer Eye Institute clinic site between January 1, 2020, and December 31, 2024, were assessed for inclusion.
Laser retinopexy was performed with a pattern scan laser system (Pascal), multifunctional laser system (Lumenis), or Vision One (Lumenis) laser system. The delivery choice of slitlamp laser with a contact lens or laser indirect ophthalmoscopy was at the treating physician’s discretion. Although there was no standard procedure protocol, in general, 2 to 3 concentric, confluent rows of laser were applied to surround the retinal defect and any subretinal fluid, if present. Direct treatment of the bare RPE beneath the retinal break was avoided. Laser spots are generally extended to the ora serrata for peripheral tears anterior to the equator. Retinal breaks located at the 12:00, 3:00, 6:00, and 9:00 meridians were categorized based on the adjacent quadrant with greater involvement.
The primary outcome measures were the occurrence of RRD and performance of additional laser retinopexy in the same or fellow eye after the initial procedure. The decision to retreat was at the discretion of the treating physician; there was no protocol to define retreatment. Additional laser retinopexy was further classified as treatment of the original break or a new break. The time interval until subsequent treatment was recorded. The RRD status at the time of surgery (macula-on or macula-off) was documented. Outcomes were compared between eyes initially treated by attending physicians and resident physicians.
Covariates
Demographics, baseline ocular characteristics, and clinical findings were included as covariates, including laterality (left or right), sex, age at time of initial laser retinopexy, race, ethnicity, lens status (phakic vs pseudophakic), logMAR, best-corrected visual acuity (BCVA), vitreous hemorrhage (VH), number of retinal breaks (single vs multiple, defined as 2 or more), and location of the retinal break by quadrant (superonasal, superotemporal, inferotemporal, inferonasal).
Statistical Analysis
Univariable logistic regression was used to estimate the odds ratio (OR) and 95% CI between the primary outcomes and independent variables. For analyses evaluating whether laser retinopexy in 1 eye was associated with outcomes in the fellow eye, generalized estimation equations with robust sandwich variation estimators were used to account for inter-eye correlation. Multivariable logistic regression models were constructed to estimate adjusted OR and 95% CI, using a backward stepwise elimination approach. Models initially included demographic variables (eye laterality, sex, race, ethnicity) and all statistically significant variables in univariate analysis. All 2-way interactions of significant independent variables were tested. All analyses were conducted with Stata 18.5 (StataCorp LP). Statistical significance was defined as P < .05.
Results
Study Population
In total, 950 eyes of 922 patients treated with initial laser retinopexy were identified in the medical records; 336 eyes were excluded for having less than 6 months of follow-up from initial laser retinopexy, and 14 eyes were excluded due to prior history of vitreoretinal surgery. The final study cohort included 600 eyes of 572 patients treated with laser retinopexy for retinal breaks. The median patient age at initial laser retinopexy was 62 years (interquartile range, 56-67). The study cohort included 316 (52.7%) men and 299 (49.8%) right eyes. Four hundred ninety-three patients were White (82.2%, including self-identified Hispanic), and 226 (37.7%) were Hispanic. The mean duration of follow-up was 31.7 ± 15.6 months. The overall mean number of retinal breaks per eye was 1.04 ± 0.26. Among eyes with multiple breaks, the mean number of breaks was 2.41 ± 0.62. Retinal break morphology included 503 horseshoe tears (83.8%), 73 operculated holes (12.2%), and 24 unspecified retinal breaks (4.0%). Overall, 498 eyes (83.0%) presented with symptoms, defined as floaters, flashes (photopsia), or blurred vision, while 102 eyes (17.0%) were asymptomatic. The average refractive error in spherical equivalent was −1.25 ± 3.19 diopters before procedure, −1.46 ± 3.42 diopters on the day or immediately after the procedure, and −1.16 ± 2.95 diopters at last follow-up. The mean BCVA was 0.16 ± 0.32 logMAR (20/29) before the procedure, 0.18 ± 0.30 logMAR (20/30) on/after the procedure, and 0.15 ± 0.27 logMAR (20/28) at last follow-up. No significant differences were found with pairwise t tests (all P > .12) (Table 1).
Demographic Characteristics of Patients With Treated Retinal Breaks.
Abbreviations: BCVA, best-corrected visual acuity; VA, visual acuity.
Vitreoretinal surgery was performed in 43 eyes (7.2%), most commonly for macula-on RRD (37 eyes [86.0%]), while other indications were for macula-off RRD and nonclearing VH. In eyes that underwent surgery, the mean BCVA was 0.28 ± 0.49 logMAR (20/38) before the procedure, 0.35 ± 0.67 logMAR (20/45) on/after the procedure, and 0.32 ± 0.32 logMAR (20/42) at the most recent follow-up. No significant differences were found with pairwise t tests (all P > .15) (Table 2). Among the 66 eyes (11%) receiving additional laser, an average of 1.13 additional sessions were performed, 53 (80.3%) for the original break and 13 (19.7%) for a newly identified break identified during follow-up. VH at baseline was present in 19 eyes (28.8%) and pseudophakia in 15 (22.7%). The additional laser retinopexy was performed within 1 week of the original procedure in 16 eyes (24.2%), between 1 week and 1 month in 14 (21.2%), and after 1 month in 36 (54.6%) (Table 2). In eyes that received additional laser, the mean BCVA was 0.07 ± 0.11 logMAR (20/24) before the procedure, 0.18 ± 0.25 logMAR (20/30) on/after the procedure, and 0.14 ± 0.19 logMAR (20/28) at the most recent follow-up. No significant differences were found with pairwise t tests (all P > .29) (Table 2).
Outcomes of Patients Treated With Laser Retinopexy for Retinal Breaks (N = 600).
Abbreviations: BCVA, best-corrected visual acuity; VA, visual acuity.
Initial laser retinopexy was performed by attending physicians (including fellows) in 572 eyes (95.3%) and by residents under attending supervision in 28 eyes (4.7%). The rate of additional laser retinopexy treatments performed by residents was similar to the rate performed by attendings (14.3% vs 10.8%; P = .56). Similar rates were also found between residents and attendings performing vitreoretinal surgery (10.7% vs 7.0%; P = .47). The mean BCVA at last follow-up did not differ significantly between resident-treated eyes (0.12 ± 0.12 logMAR, 20/26) and attending-treated eyes (0.15 ± 0.28 logMAR, 20/28; P = .62) (Table 3).
Comparison of Residents and Attendings in Laser Treatment for Retinal Breaks.
Univariate Analysis
Characteristics associated with higher odds of additional laser retinopexy included multiple retinal breaks (39.1% vs 9.9%; OR, 5.9; 95% CI, 2.4-14.2; P = .001), laser retinopexy in the fellow eye (46.4% vs 9.3%; OR, 8.49; 95% CI, 3.84-20.6; P = .001), and inferotemporal location (55.7% vs 34.8%; OR, 2.34; 95% CI 1-5.4; P = .047). Factors associated with lower odds included pseudophakia (36.6% vs 57.9%; OR, 0.42; 95% CI, 0.20-0.90; P = .025), VH (38.0% vs 57.3%; OR, 0.46; 95% CI, 0.22-0.94; P = .033), and poorer baseline BCVA per 0.1 logMAR (OR, 0.70; 95% CI, 0.49-1.0; P = .049). Surgical intervention (RRD or nonclearing VH) was associated with multiple retinal breaks (21.7% vs 6.6%; OR, 3.94; 95% CI, 1.4-11.2; P = .001), pseudophakia (51.2% vs 25.0%; OR, 3.15; 95% CI, 1.4-6.9; P = .004), and VH (44.0% vs 25.6%; OR, 2.28; 95% CI, 1.1-4.8; P = .030). Female sex was associated with lower odds of subsequent surgical intervention (2.8% vs 11.1%; OR, 0.23; 95% CI, 0.1-0.5; P = .001).
Multivariable Analysis
Pseudophakia was associated with lower odds of performing additional laser retinopexy (adjusted OR, 0.39; 95% CI, 0.17-0.90; P = .028), while laser retinopexy in the fellow eye (adjusted OR, 7.78; 95% CI, 1.63-37.1; P = .01) was associated with higher odds. Female sex was associated with lower odds of performing additional laser retinopexy (adjusted OR, 0.35; 95% CI, 0.14-0.90; P = .029), while VH was associated with higher odds for surgical intervention (adjusted OR, 2.31; 95% CI, 1.04-5.14; P = .038).
Conclusions
The present study found a rate of 7.2% for subsequent RRD and 11% for additional laser retinopexy after initial treatment of a symptomatic retinal break. Historically, the 6-month rates of RRD after laser retinopexy are reported to be between 2.1% and 8.8% depending on risk factors and break morphology.6–9
The present study expands upon prior reports by providing extended follow-up from a large contemporary US tertiary referral cohort and offering more specific characterization of retreatment patterns after initial laser retinopexy. In addition to reporting rates of subsequent RRD, this analysis distinguishes retreatment of the original break from newly identified break(s) and evaluates fellow-eye involvement and clinical predictors of treatment escalation. These findings provide a more detailed understanding of the clinical course after laser retinopexy in contemporary practice and may assist clinicians in risk stratification and counseling regarding the likelihood and timing of additional intervention.
Univariate analysis identified a 4-fold increased risk associated with multiple retinal breaks for both vitreoretinal surgical intervention and additional laser retinopexy, consistent with previous studies.9,10 VH and pseudophakia in the present study were associated with lower odds of additional laser but higher odds (3- to 4-fold) of surgical intervention, which differs from other studies. VH and pseudophakia have been hypothesized to compromise the view and ability to deliver laser,10–13 which may factor into the decision-making process to perform surgery. Pseudophakia has also been associated with an increased risk of RRD, potentially due to accelerated posterior vitreous detachment after cataract surgery.14,15 In this context, pseudophakic eyes may be more likely to progress directly to clinically significant detachment rather than undergoing localized progression amenable to repeat laser treatment. However, this relationship is unable to be corroborated in the present study because it was not designed to evaluate causality.
Inferotemporal location was also associated with higher odds of additional laser retinopexy. Inferotemporal retinal breaks may be more susceptible to persistent subretinal fluid accumulation due to gravity, and the temporal retina is relatively thinner. 16 Moreover, inferior retinal breaks may be more technically challenging to visualize and fully barricade, particularly in the setting of media opacity or patient positioning (eg, Bell’s reflex). These factors may contribute to a higher likelihood of incomplete initial adhesion or subsequent extension requiring retreatment; again, it is important to emphasize this relationship cannot be supported by the present study because it did not evaluate causality.
The 11% rate of additional laser retinopexy for the same or new break, while considerable, is in line with retrospective studies and systematic reviews reporting rates of up to 22% to 42% for retreatment with laser.6,16 Moussa et al 7 reported that within 6 months, the rate of retreatment with laser was 22%. Garoon et al 9 reported additional laser retinopexy in 18.7% of eyes in a previous study from our institution; 48.7% of retreated eyes involved new breaks, and 60.0% of retreated eyes received retreatment within the first month. The present study reports new breaks accounting for 19.7% of retreated eyes and 45.5% of retreatments occurring within the first month. 9 For retreatments, 80.3% were performed on the original break. In comparison, Garoon et al reported that approximately 51.3% of retreatments were performed for the same break. The reason for this difference is not fully known but may reflect incomplete initial retinopexy, persistent vitreoretinal traction, large break morphology, or limited patient tolerance during the initial procedure.
Moreover, this study reports that patients are at increased odds (8-fold) of undergoing additional laser retinopexy (but not surgery) if they received laser retinopexy in the fellow eye. The likelihood of undergoing laser retinopexy specifically for a new retinal break in the fellow eye (as opposed to developing an RRD) remains less quantified in the literature.17,18 In a retrospective study on fellow eye RRD, Al-Dwairi et al 17 demonstrated prophylactic laser retinopexy for high-risk retinal breaks in the fellow eye decreased the likelihood of fellow eye RRD. Thus, the rate of laser retinopexy for a retinal break in the fellow eye is considerable. Bilateral laser retinopexy may warrant heightened surveillance and frequent follow-up evaluations.
In the present study, there were no significant differences in outcome for eyes treated by residents and those treated by attendings for additional laser retinopexy, surgical intervention, or visual outcomes at last follow-up, as previously reported. 9
The findings of the present study are subject to inherent limitations applicable to all retrospective studies. Furthermore, this case series is subject to biases of nonstandardized management, documentation, and follow-up regimen. There was no standardized protocol to follow for retreatments in the same eye or the fellow eye, and it is unknown whether new retinal breaks that received additional laser retinopexy could have been avoided. There was also a substantial proportion of patients (>300) who were excluded for having less than 6 months of follow-up, which may have represented selection bias.
Progression from retinal break to RRD occurs infrequently. Retreatment of retinal breaks after initial laser retinopexy is not uncommon, underscoring the need for educating the patient to return promptly if there are new symptoms, as well as continued surveillance and risk stratification, particularly in the early post treatment period when incomplete treatment or new breaks may occur.
Footnotes
Ethical Considerations
This report was conducted in accordance with the principles of the Declaration of Helsinki. Institutional review board approval was required and obtained.
Consent to Participate
Written informed consent was obtained from the patient(s) and/or guardian(s) for publication of this report and any accompanying images.
Consent for Publication
Written informed consent was obtained from the patient(s) and/or guardian(s) for publication of this report and any accompanying images.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by an unrestricted grant from Research to Prevent Blindness (GR004596-1).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
Data are available upon request to the corresponding author.