Clinical Outcomes of Rhegmatogenous Retinal Detachment Treated With Pneumatic Retinopexy

Key Points

Question  What are the clinical practice setting outcomes associated with retinal detachment treated by pneumatic retinopexy?

Findings  In this cohort study including 9659 eyes from 9553 patients, single-operation success was achieved in 6613 eyes (68.5%). Success was associated with self-reporting as female, while current smoking status was associated with failure.

Meaning  The clinical practice settings outcomes associated with retinal detachment treated by pneumatic retinopexy in this study appear to be less favorable than those reported in large clinical trials.

Importance  Pneumatic retinopexy (PR) is the only clinic-based method of rhegmatogenous retinal detachment (RRD) repair. Registry-acquired clinical practice setting outcomes data with this procedure have not yet been reported.

Objective  To describe the clinical outcomes associated with RRD treated primarily with PR.

Design, Setting, and Participants  In this retrospective cohort study, data from patients 19 years and older with noncomplex RRD treated at academic and private ophthalmology practices participating in the American Academy of Ophthalmology IRIS Registry (Intelligent Research in Sight) were analyzed. Data were collected from January 1, 2013, to December 31, 2019, and data were analyzed from January to December 2020.

Exposures  Data from the IRIS Registry were queried for eyes that underwent PR for noncomplex RRD and had at least 3 months of follow-up. Cases were identified by a combination of diagnosis code for RRD and a Current Procedural Terminology code for PR.

Main Outcomes and Measures  The number of eyes that achieved single-operation success (SOS), defined as retinal reattachment without a subsequent retinal detachment surgery or repeated PR.

Results  Of 9553 included patients, 5827 (61.0%) were male, and the mean (SD) age was 62 (10) years. A total of 9659 eyes were identified. SOS was achieved in 6613 eyes (68.5%). Best-corrected visual acuity significantly differed 9 to 12 months after treatment between the SOS group, with a mean of 0.24 logMAR (95% CI, 0.23-0.25; approximate Snellen equivalent, 20/35), and the single-operation failure group, with a mean of 0.43 logMAR (95% CI, 0.40-0.46; approximate Snellen equivalent, 20/54). Among all patients, the mean time to maximal visual recovery was 268 days (95% CI, 260-276). Endophthalmitis was observed in 3 eyes (0.03%). SOS was associated with female sex (odds ratio, 1.51; 95% CI, 1.38-1.65), while current smoking status was associated single-operation failure (odds ratio, 0.78; 95% CI, 0.68-0.91).

Conclusions and Relevance  In this registry-based study, which encompasses a large number of eyes drawn from multiple, heterogenous electronic health record systems, SOS was achieved in 68.5% of eyes with noncomplex RRD treated by primary PR. It is unknown how these outcomes would have compared with other methods of RRD repair in this cohort.

Rhegmatogenous retinal detachment (RRD) is the most common type of retinal detachment (RD) and may result in significant morbidity and visual disability. Of the 3 procedures used to repair RRDs, only one method, pneumatic retinopexy (PR), is performed in the office rather than in the operating room. Introduced by Hilton and Grizzard in 1986,¹ PR has several potential advantages, including the faster restoration of visual acuity,² the avoidance of systemic anesthesia, a lower rate of cataract progression,³ and the potential for improved cost and utility.⁴ Despite its potential benefits, PR has most often been reserved for eyes with detachments with single or closely clustered retinal breaks (often in the superior clock hours), phakic lens status, and without additional other high-risk pathology, such as lattice degeneration, proliferative vitreoretinopathy (PVR), giant retinal tear, or dialysis.

A 2019 randomized clinical trial, the Pneumatic Retinopexy Versus Vitrectomy for the Management of Primary Rhegmatogenous Retinal Detachment Outcomes Randomized Trial (PIVOT),² compared the clinical outcomes associated with PR vs pars plana vitrectomy (PPV) for management of RRD. This trial included eyes with a single retinal break or a group of breaks in detached retina within 1 clock hour above the 8-o’clock to 4-o’clock meridian, with any number, location, and size of retinal breaks or lattice degeneration in attached retina. The study enrolled 176 patients and included both phakic and pseudophakic eyes. The trial disclosed that primary anatomic success was achieved in 81% of patients undergoing PR vs 93% of patients undergoing PPV, with 99% of eyes in each arm achieving overall anatomic success. However, eyes treated with PR had superior visual acuity, less vertical metamorphopsia, and reduced morbidity compared with eyes treated with PPV. While randomized clinical trials⁵ and results from moderately sized observation studies^6-9 support the use of PR, large meta-analyses¹⁰ and big data exploring clinical outcomes are currently lacking.

It has been estimated that 40% of all RRD could potentially be repaired via PR.¹¹ Although PR is a favorable alternative for many RRDs, it remains uncommonly used in the US. In 2014, Medicare beneficiaries received 19 288 PPVs for RRDs but only 2791 PRs, and the number of PRs has been steadily decreasing since its peak utilization in 2004.^12,13 Expansion of the use of PR could potentially allow for less patient discomfort, faster recovery, avoidance of anesthesia, preservation of refractive status, and significant cost savings for the health care system. Of note, a modeled cost analysis of RRD repair showed that at an assumed success rate of 75%, 5% less than the reported rate from PIVOT,² PR is approximately $2000 less expensive than PPV or scleral buckling.⁴ While trial data have shown that PR is an effective treatment when performed by physicians with dedicated expertise with PR and on a subset of favorable candidates, data are needed to evaluate the clinical practice setting outcomes associated with use of PR at a population-based level.

Quiz Ref IDThe purpose of this study is to evaluate the clinical efficacy and safety of PR in a population-based study using patients cared for by physicians participating in the American Academy of Ophthalmology IRIS Registry (Intelligent Research in Sight).

The methods of the IRIS Registry database have been described in the literature.¹⁴ Data are collected from electronic health record systems of ophthalmic practices that participate in the registry, which contain both clinical and billing information but not images nor operative notes. For this study, we only had access to deidentified patient-level information. Given the usage of deidentified patient data, this study was exempt from institutional review board approval per institutional policy. All cases were identified in the IRIS Registry via a combination of International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and ICD-10-CM codes and Current Procedural Terminology (CPT) codes (eAppendix in the Supplement). The ICD-9-CM and ICD-10-CM codes that were queried to identify patients were those specifying RRD. Exclusion criteria were prior history of RD, history of vitrectomy or scleral buckling procedure, tractional RD, traumatic RD and retinal dialysis, giant retinal tear–associated RD, PVR, and exudative or serous RD. All patients were required to be treated initially with PR, defined as CPT code 67110. The minimum required follow-up period after PR was defined as 3 months after the index date.

The main outcome measure was single-operation success (SOS), defined as anatomical retinal reattachment in a single procedure without a subsequent vitreoretinal surgery aimed at retinal reattachment (ie, CPT code 67110 for repeated PR, CPT code 67107 for scleral buckling, CPT code 67108 for vitrectomy with or without scleral buckling, or CPT code 67113 for repair of complex detachment). The secondary outcome measures were logMAR best-corrected visual acuity (BCVA), time to visual recovery of maximal BCVA, development of metamorphopsia, rate of postoperative vitreous hemorrhage, development of epiretinal membrane, rate of postoperative PVR, and the rate of postoperative endophthalmitis.

All patient data were included or excluded based on the diagnosis and CPT codes listed in the eAppendix in the Supplement, and neither the individual treating physicians nor practices had any impact in the process of case selection or analysis. All patients who were documented in the electronic health records of practices participating in the IRIS Registry were accessible to the investigation. Statistical analyses were performed with R version 4.0.2 (The R Foundation). Visual acuity was recorded in Snellen lines then translated to logMAR units for quantitative assessments. Two-tailed P values were calculated using the Wald test, and significance was set at P < .05.

In the IRIS Registry, between January 1, 2013, and December 31, 2019, a total of 9659 eyes from 9553 patients were identified to have met the inclusion criteria (Figure 1). A total of 5827 patients (61.0%) were male, and the mean (SD) patient age was 62 (10) years. A total of 5977 patients (62.6%) had never smoked, while 2263 (23.7%) were former smokers, 952 (10.0%) were active smokers, and 361 (3.8%) did not report their smoking status. The regional representation of patients included 3780 (39.6%) from the South, 1926 (20.2%) from the Midwest, 1921 (20.1%) from the Northeast, and 1810 (18.9%) from the West.

Quiz Ref IDThe primary outcome measure, SOS, was achieved in 6613 eyes (68.5%). Of the 9659 eyes requiring a subsequent vitreoretinal surgical intervention, 8596 (89.0%) received a procedure in the first 3 postoperative months and 8983 (93.0%) received a procedure in the first 6 months following PR. The date of the requirement for additional RD surgery is listed in Table 1. Time to experiencing a failure is depicted in Figure 2. In a generalized estimating equation analysis, SOS was associated with reporting as female (odds ratio, 1.51; 95% CI, 1.38-1.65), while current smoking status was associated with failure (odds ratio, 0.78; 95% CI, 0.68-0.91). Additional factors evaluated for association with SOS are listed in Table 2.

Quiz Ref IDRegarding the secondary clinical outcome measures, the mean (SD) BCVA was 0.31 (0.39) logMAR (approximate Snellen equivalent, 20/40) in the entire cohort at postoperative month 9 to 12. Mean BCVAs at other specified postoperative time intervals are available in Table 3. Of note, mean BCVA significantly differed at 9 to 12 months postoperatively between the SOS group, which had a mean BCVA of 0.24 logMAR (95% CI, 0.23-0.25; approximate Snellen equivalent, 20/35), and the single-operation failure (SOF) group, with a mean BCVA of 0.43 logMAR (95% CI, 0.40-0.46; approximate Snellen equivalent, 20/54). Among all patients, the mean time to maximal visual recovery was 268 days (95% CI, 260-276). Of the 3046 eyes that experienced SOF, 2193 (72.0%) received either a PPV or scleral buckling and PPV, 701 (23.0%) received a second PR, 365 (12.0%) received a complex RD repair, and 152 (5.0%) received a scleral buckling alone.

Adverse outcomes were uncommon. PVR was not diagnosed postoperatively in the medical record 36 months after the initial PR. Endophthalmitis occurred in 3 of 9659 eyes (0.03%) up to 14 days after the initial PR. Postoperative vitreous hemorrhage was noted in 307 eyes (3.2%). Metamorphopsia was observed in 14 eyes (0.1%) postoperatively. The rate of postoperative epiretinal membrane was 24.5% (2367 of 9659).

This big data retrospective cohort study from the IRIS Registry of 9659 eyes treated with PR for RRD disclosed a SOS rate of 68.5%. Female sex was positively associated with SOS, while current smoking was negatively associated with success. Of note, both groups had favorable long-term visual outcomes on average by postoperative month 9 to 12, with slightly better results in eyes that had achieved SOS. The mean (SD) time to maximal visual recovery was 268 days (95% CI, 260-276).

Previous randomized studies have reported higher rates of SOS, including 81% in PIVOT² and 82% in the study by Tornambe and Hilton.⁵ Of note, PIVOT evaluated primary anatomic success at 12 months and did not recognize eyes that received PR and required an additional gas injection after initial treatment as having experienced a primary failure. If those eyes had been considered failures (as they would have been using the current study’s methodology), the SOS would have been 72%.

A meta-analysis of 81 studies and 4138 eyes disclosed an SOS of 74%.¹⁵ Additionally, a multicenter study of outcomes of PR performed by vitreoretinal fellows in the US reported an SOS of 67%.¹⁶ Of note, fellows in that study reported a median of 7 cases during their vitreoretinal training. The use of PR greatly varies based on region of the US, operating room availability, and country (for instance, in Canada, it is more commonly used as a primary repair method and a more substantial part of resident and fellow training).

The factors associated with anatomic success and failure found in this study are consistent with prior literature. Smoking has been associated with a higher risk of PVR.^17,18 Although female sex has not been reliably associated with more favorable anatomic outcomes in RRD, recent evidence has pointed to possible differences in time to repair between self-reported male and self-reported female individuals.¹⁹

The patients in this cohort had favorable visual outcomes across both the SOS and SOF groups. It has been shown in retrospective studies²⁰ as well as PIVOT² that even eyes that fail initial repair with PR may have good visual outcomes. Data from PIVOT also suggested that visual acuity after PR is superior to PPV in the initial postoperative months (month 3 and 6), then still exceeds but falls closely in line with visual acuity after PPV at 12 months.² Eyes treated with PR also had lower rates of vertical metamorphopsia, possibly due to less retinal displacement.²¹ A post hoc analysis of subjective visual outcomes also showed superior results with PR.²² Of note, this study was noncomparative but did disclose a very low rate of postoperative metamorphopsia. We acknowledge that the current study relied on billing codes, not objective measures of metamorphopsia, and that overall rates were likely underreported.

PR has been touted for its faster time to visual recovery.² Other procedures, such as scleral buckling (which often has slower resolution of subretinal fluid) and PPV (which requires air, gas, or silicone oil tamponade), may have longer delays in visual restoration. Although this study was noncomparative, it disclosed that the mean time to maximal visual acuity occurred at approximately 9 months.

Eyes treated with PR have been found to have a low rate of adverse events. Hilton and Tornambe²³ disclosed that new retinal breaks occurred in 13% of patients and PVR in 4%. Other potential complications include posterior breaks,²⁴ anterior chamber gas,²⁵ subretinal²⁶ and suprachoroidal²⁷ gas, and endophthalmitis.²⁸ The current study did not identify eyes diagnosed with postoperative PVR. However, 11% of eyes receiving an initial failure underwent a secondary complex RD repair, suggesting that PVR was present. This likely represents a limitation in the ability of electronic medical record–based registry studies to identify postoperative complications. However, the rate of endophthalmitis of 0.03% is in the range of that reported with intravitreal injection.²⁹

Our study has limitations. Similar to most big data studies, the current report is limited by the quality of data inferred by diagnosis and billing codes. Data on adverse events may be underreported, as these diagnoses are not always listed reliably in the electronic medical record. Another potential limitation is that this study could not capture the intention of each treating surgeon. For instance, recent data suggest that PR is used more commonly on the weekend when access to the operating room may be more scarce.³⁰ If a significant proportion of surgeons in the registry use PR as a bridge to PPV, then the rate of SOS may be underestimated, since bridge treatments would have been counted as an anatomic failure. As failures were assessed based on reoperation, those patients who redetached but refused further surgery would not have been recognized as anatomic treatment failures using our methodology. Additionally, those patients who sought care outside a registry practice for reoperation would also not have been captured in our database.

Quiz Ref IDSurgical notes are not captured by the IRIS Registry. As such, information on the macula status for this group of detachments was not available. Analysis on the type of gas tamponade selected and whether cryotherapy or laser was used for retinopexy also could not be elicited. Furthermore, retinal drawings were not available for this study, and a detailed analysis of outcomes for macula-involving or macula-sparing detachments could not be undertaken, as macular status was unreliably noted via billing and diagnosis codes. Furthermore, lens status was not reliably obtained in this database and thus was not evaluated.

In conclusion, this study explored the results of PR for noncomplex RRD in a registry encompassing a large number of eyes. Single-operation anatomic success was 68.5%, and eyes with SOS or SOF experienced relatively favorable visual outcomes and a low rate of adverse events.

Accepted for Publication: April 2, 2021.

Published Online: June 17, 2021. doi:10.1001/jamaophthalmol.2021.1860

Corresponding Author: Nicolas A. Yannuzzi, MD, Bascom Palmer Eye Institute at the University of Miami Miller School of Medicine, 900 NW 17 St, Miami, FL 33136 (nay7@miami.edu).

Author Contributions: Dr Yannuzzi had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Yannuzzi, Li, Kelly, Parke.

Acquisition, analysis, or interpretation of data: Yannuzzi, Li, Fujino, Kelly, Lum, Flynn.

Drafting of the manuscript: Yannuzzi, Li, Kelly.

Critical revision of the manuscript for important intellectual content: Yannuzzi, Li, Fujino, Lum, Flynn, Parke.

Statistical analysis: Li, Fujino, Kelly, Flynn.

Obtained funding: Yannuzzi, Parke.

Administrative, technical, or material support: Li, Lum, Flynn.

Study supervision: Kelly, Parke.

Conflict of Interest Disclosures: Dr Yannuzzi has served on advisory boards for Novartis, Genentech, and Alimera Sciences. Dr Parke has received royalties from Vortex Surgical and served as a consultant for Genentech. No other disclosures were reported.

Funding/Support: This work was supported by a grant from the Hoskins Center via the American Academy of Ophthalmology.

Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.