Outcomes of Pneumatic Retinopexy for the Management of Rhegmatogenous Retinal Detachment at a Tertiary Care Center.

PURPOSE: The purpose of the study was to report the outcomes of pneumatic retinopexy (PR) for rhegmatogenous retinal detachment (RRD).
METHODS: A retrospective chart review of all patients with RRD who were managed with PR at the King Khaled Eye Specialist Hospital between 2000 and 2014. Data were collected on preoperative ocular history, characteristics of the RRD, postoperative anatomical and visual outcomes, and secondary surgeries.
RESULTS: Sixty-five eyes comprised the study sample with mean a follow-up of 42 months. Eighteen (27%) eyes had a history of ocular surgery and 5 (7.6%) had previous ocular trauma. Thirty-nine (60%) eyes had macula on RRD. Retinal detachment (RD) was caused by a single break in fifty (76.9%) eyes. Superior breaks were found in 56 (86.1%) eyes. Twelve (18%) eyes had posterior vitreous detachment. Fifty-one (78.5%) eyes underwent PR under local anesthesia and 9 (13.8%) underwent conjunctival peritomy. Octafluoropropane gas was used in 49 (75.4%) eyes and sulfur hexafluoride in 16 (24.6%) eyes. Cryotherapy was performed in 48 (73.8%) eyes and only laser photocoagulation in 8 (12.3%) eyes. Nine (13.8%) eyes underwent both cryotherapy and laser photocoagulation. Primary anatomical healing was achieved in 45 (69.2%) eyes, of which 20 (30.8%) required a second intervention. The final anatomical outcome was achieved in 100% of the eyes. At final visit, vision improved in 72% of eyes, was maintained in14.5%, and decreased in 9.7% (6).
CONCLUSION: PR is a safe procedure for RRD repair. The final anatomical outcome was excellent in all cases including the recurrent RD after primary failure.

Introduction

Pneumatic retinopexy (PR) is an office-based procedure that is a good alternative to invasive procedures such as vitrectomy and scleral buckles (SBs) for managing selected cases of retinal detachment (RD). PR is a two-step procedure that first requires nonexpansile tamponade using air or an expanding gas bubble such as sulfur hexafluoride (SF6) or octafluoropropane (C3F8), that is injected into the vitreous cavity. The patient is positioned so that the bubble closes the retinal break, permitting resorption of subretinal fluid (SRF). The second step is induction of a chorioretinal adhesion around all retinal breaks with cryopexy, laser, or both. The most common gases used for PR are SF6 and C3F8.[12]

Bubble expansion of SF6 is achieved between 24 and 36 h after injection and lasts for 6–12 days.[1234] However, C3F8 bubble expansion is achieved 3 days after injection and lasts for 38 days. Optimal tensile strength is achieved within 7–14 days following cryopexy or laser retinopexy.[5]

The primary success rate of PR is (60.7% to 80.5%).[126] However, the final success rate was >96% in cases that required two or more surgeries.[126] Success in these cases required the following criteria: (1) presence of a single or multiple retinal breaks occupying 1-clock hour of the retinal arc; (2) superior retinal break at 8 clock hours of the globe; and (3) ability of the patient to maintain the proper head position for at least 16 h/day for 5 days or longer.[126] The contraindications for this procedure were the presence of retinal breaks within the lowest 4 clock hours in the inferior quadrant, presence of proliferative vitreoretinopathy (PVR) Grade C or D, patient noncompliance with head positioning, severe glaucoma, extensive lattice degeneration, and vitroretinal traction and hazy media.[678910]

The advantages of PR over SB for rhegmatogenous RD (RRD) include a decreased risk of globe perforation, no change in refractive error, decreased risk of postoperative strabismus and conjunctival scarring, and decreased risk of cosmetic defects such as ptosis and enophthalmos. In addition, PR, can be performed under topical anesthesia, is a relatively noninvasive, quick, office-based procedure requiring less equipment, staff, and other resources compared to procedures performed in an operating room.[12346]

The most common complication of PR is failure of the primary attachment secondary to missed or new breaks. Other complications include progression of cataract, subconjuctival gas vitreous prolapse, intravitreal gas migration into the anterior chamber, persistent SRF, endophthalmitis, macular hole, epiretinal membrane (ERM), macular edema, macular hole, intraocular hemorrhage, uveitis, vitreous haze, and cataract progression.[1411121314151617181920212223242526] In the current study, we evaluate the outcomes of PR at a tertiary care center and compare it to previous studies.

Methods

This retrospective study was conducted at King Khaled Eye Specialist Hospital (KKESH), Riyadh, Saudi Arabia. This study adheres to the tenets of the Declaration of Helsinki. The medical records were reviewed for all cases of RD that were managed by PR at KKESH from 2000 to 2014. A total of 65 patients were identified, and data were collected on age, gender, past medical and ocular history, and history of trauma. Clinical examination data were collected including visual acuity (VA), intraocular pressure, clinical features of RRD, macular involvement, location and the number of breaks, the presence of posterior vitreous detachment (PVD), and the presence of PVR. Surgical data were collected on the PR technique including type of gas used. Data were collected on postoperative complications, visual and anatomic outcomes, and long-term complication. All data were entered into an Excel spreadsheet (Microsoft Corp., Redmond, WA, USA).

Anatomical outcomes were divided into primary outcome defined as the achievement of an anatomically flat retina after a single operation or secondary outcome defined as the achievement of anatomically flat retina at final follow-up.

Improvement in VA was defined as a gain of two or more lines or a final VA better than or equal to 20/30. The risk factors for recurrent RD and the correlation of variables with final visual outcome were evaluated using Chi-Square or Fisher's exact tests. Data were analyzed with Statistical Package for the Social Sciences version 19.0 (IBM Corp., Armonk, NY, USA).

Results

The study sample comprised of 65 eyes of 65 patients. The mean age at presentation was 50.72 years (range: 12–85 years). Follow-up ranged from 4 weeks to >13 years with a mean of 42 months. Eighteen eyes (27%) had a history of previous ocular surgery, of which 7 (10.8%) had undergone cataract extraction, 45 (69.2%) eyes had undergone refractive surgery, 6 (9.2%) had RD repair done elsewhere, 1 (1.5%) eye had undergone glaucoma surgery, and 3 (4.6%) eyes had undergone other ocular surgeries. Five (7.6%) eyes had a history of previous ocular trauma. Thirty-nine (60%) eyes had macula on RRD and fifty (76.9%) eyes had a RD caused by single break. Superior breaks were noted in 56 (86.1%) eyes with horse shoe tear (HST) in 28 (43%) eyes, atrophic holes in 14 (21%) eyes, and retinal dialysis in one eye. RD was limited to one quadrant in sixty (92.3%) eyes. Twelve (18%) eyes had PVD. There were no cases of PVR. PR was performed with local anesthesia in 51 (78.5%) eyes and conjunctival peritomy was performed in 9 (13.8%) eyes. C3F8 was injected in 49 (75.4%) eyes and SF6 was injected in 16 (24.6%) eyes. Cryotherapy was performed in 48 (73.8%) eyes; laser photocoagulation in 8 (12.3%) and 9 (13.8%) eyes received both cryotherapy and laser photocoagulation.

The only intraoperative complication was hyphema in one eye. Postoperative complications included two eyes that developed cataract, one eye developed a full thickness macular hole, one eye with persistent SRF, and one eye developed and ERM. The full thickness macular hole was managed with pars plana vitrectomy (PPV) with improvement in vision from counting fingers to 20/30. The eye with persistent SRF resolved spontaneously over 14 months.

Primary retinal attachment was achieved in 45 (69.2%) eyes; however twenty (30.8%) eyes had recurrent RRD. In eyes with recurrent RRD, 9 out of 20 (45%) cases were caused by a new break and a single break occurred in 14 out of 20 (70%) eyes. Seven (35%) eyes with recurrent RRD had a HST. These cases were managed with SB and PPV in 50%, PPV in 40%, and SB in 10% of cases. Vision improved in 47 (72%) eyes, vision remains the same in 9 (13.8%) eyes, and vision decreased in 6 (9.2%) eyes. Three (4.6%) eyes were lost to follow up soon after surgery.

There was a higher rate of recurrent RD in pseudophakes (57%) compared to phakic eyes (26.6%) (P = 0.189). A higher rate of recurrent RRD was noted in 6 (50%) eyes that had PVD at presentation compared to 14 (26.4%) with no PVD (statistically insignificant; P = 0.165). There were more cases of recurrent RD caused by multiple breaks (6; 40% eyes), compared to single breaks (14; 28% eyes) (statistically insignificant; P = 0.524). Other risk factors such as lens status, history of trauma, PVR, type of intraocular gas, type of retinopexy, and macular status were also not correlated with recurrent RRD [Table 1]. The final anatomical outcome in 62 eyes was 100% (excluding three eyes that lost their follow-up).

Table 1

Correlation of the risk factors with the primary attachment and recurrent rhegmatogenous retinal detachment

Final visual outcome was not correlated to a number of factors including lens status, history of trauma, macular involvement, and type of intraocular gas [Table 2; P > 0.05, all correlations]. The causes of poor vision were end-stage glaucoma in three (4%) eyes and ERM in 1 (1.5%) eye. The poor vision in the remaining two eyes was related to recurrent RRD.

Table 2

Correlation of visual outcome with risk factors

Discussion

The importance of PR as an office-based and cost-effective procedure has been previously reported. Table 3 presents the pooled data from previous studies that showed the success rate after a single procedure ranged from 60.7% to 80.5% and the final anatomical outcome was 96.1% to 100%.[67891027] These outcomes are comparable to our success rate of 69.2% with a single procedure and the final anatomical was 100% (the primary success rate is considered to be low in comparison to other modalities of surgical intervention. Schaal et al.[28] compared functional and anatomical outcome of SB, PPV, combined SB, and vitrectomy or PR. He reported that the initial success rate for retinal reattachment was 86% for SB, 90% for PPV, 94% for combined SB vitrectomy, and 63% for PR surgery heal so found that the difference in the final reattachment rates and final visual acuity among all four groups were insignificant.

Table 3

Success rate from multiple studies

Our study showed that the highest failure rate was in pseudophakic eyes, and multiple retinal breaks yet neither factor were statistically significant. Tornambe[29] also showed that the failure rate in aphakic/pseudophakic eyes was higher compared to phakic eyes. We also investigated the following factors: trauma, type of intraocular gas, and the use of cryotherapy versus laser; however, they did not showed correlation with the rate of recurrent RRD.

The average reported postoperative complications in previous studies was 11.7% (ranged from 3.7% to 32.78%). 6 which is comparable with our study (13.8%) of eyes.

The most common complication we found was cataract (3%) which is comparable with the previous study (4%).[4]

One case developed macular hole which may be related to PVD induced by mechanical tractional force during the gas injection this was also previously proposed by Wirostko et al. We also had one case of ERM may not be directly related to PR complication as showed in previous reports.[222530] This patient had a history of prophylactic laser for multiple retinal breaks before PR which we consider it as additional risk factor for ERM formation.

The limitations of our study are being the retrospective design, in addition to small sample size that we consider it very low for a large tertiary care center that manages thousands of RRD with surgical intervention, other than PR. This indicates that PR is not the primary choice for retinal repair in our institute. Furthermore, we could not find a significant correlation of the studied risk factors with the outcomes; therefore, we were not able to recommend the best selected criteria for PR procedure for RD repair as mentioned in the previous reported studies.[136] However, it is worth to mentioned that our study proved the fact that, despite the low success rate after a single PR procedure (69.2%), the final outcome was still high (100%); therefore, we consider PR is a safe procedure for RD repair that had a final successful outcome unaffected by primary failure.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.