A Pioneering Study of Retinal Pigment Epithelial Photocoagulation to Seal Retinal Breaks and Prevent Hypotony during Pars Plana Vitrectomy

Background: During vitrectomy, a relatively high level of accuracy is required when using retinal laser energy, which is difficult to control in the presence of subretinal fluid and other conditions. We explored the clinical effects of retinal pigment epithelium (RPE)photocoagulation in closing retinal breaks and preventing hypotony during vitrectomy. We describe the possible indications of RPE photocoagulation. Methods: This was a prospective, non-random, uncontrolled case series study. We recruited 20 patients, with retinal detachment in 20 eyes. In this study, RPE photocoagulation was performed under the detached retina and retinal holes. RPE laser photocoagulation was performed in 3-4 rows at the edge of the periphery around the hole and the tear, replacing the traditional retinal photocoagulation and closing part or all of the retinal break. The energy of RPE photocoagulation was 100-150mJ for 120-200ms to seal the breaks, and the same energy but little gaps for the RPE without covering of retina. Results: After vitrectomy, in 19 eyes, there was a visible pigmentation reaction around the hole and tear, except for 1 eye the retinal hole located in the myelinated nerve fibre area. The mean best-corrected visual acuity (BCVA) at 6 months after surgery was significantly higher than that before surgery (p=0.002). At the last follow-up, the mean BCVA remained significantly higher than that before surgery (p=0.001).There was no significant difference in BCVA between the 6th month and the last follow-up (p=0.806).The thickness of the neurosensory layer in RPE photocoagulation and retinal photocoagulation area at 1 month after surgery was 216.33±54.42μm and 87.67±34.65μm, respectively. By the end of the follow-up, there were no serious complications, and the retinas of all 20 eyes were reattached. No hypotony occurred after surgery. Conclusions: The effects of RPE photocoagulation and retinal photocoagulation are recurrent retinal detachment, did not occur in any eye. Haemorrhage at the retinal incision gradually disappeared within 1 month after the initial follow-up. Silicone oil emulsification was observed in 4 eyes throughout the follow-up period. After removing silicone oil from eyes in which silicone oil was used, the retinal reattachment success rate reached 100%.

3 formation on the RPE, resulting in adhesion between the neurosensory layer and the RPE layer. Laser photocoagulation could be considered as comparable to glue instead of welding when closing the retinal breaks. Trial registration ChiCTR1900021504,2019-02- 24. Background Rhegmatogenous retinal detachment(RRD) is formed by liquefied vitreous entering the subretinal space through a retinal tear [1]. Closure of the retinal tear is the key to the treatment of retinal detachment. Laser can be absorbed by the melanin tissue in the fundus and the haemoglobinin the blood vessels, which produces a thermal effect. This causes the tissue to coagulate and become necrotic, forming a retinal choroidal scar adhesion, which can result in closing of the retinal tear [2][3][4].The method with which laser photocoagulation seals the hole is simple, resulting in a minor wound and the treatment can be repeated. Laser photocoagulation can prevent the occurrence of retinal detachment for dry holes, promote the absorption of subretinal fluid and prevent the development of lesions [5,6].
Retinal laser photocoagulation is not considered ideal in conventional vitrectomy surgery and may even cause complications in the following situations:(1)It is easy to cause iatrogenic tearing in patients with high myopia, especially in the thinner part of the peripheral retina [8]. (2)Absorption of high-intensity laser energy in areas with hyperpigmentation or retinal thinning can cause damage to nerve fibres [8,9].(3) For patients with diabetic retinopathy oedema or haemorrhage, it is difficult to produce ideal results using ordinary laser methods [10,11]. (4)If the retinal tear is located in the middle of the myelinated nerve fibre layer, it is difficult for the laser to penetrate. (5)It is difficult to form an effective laser spot when there is subretinal fluid.
In clinical retinal reattachment surgery, a small amount of subretinal fluid residue is often 4 found without using 'heavy liquid'(perfluorodecalin or perfluorocarbon), especially on the posterior edge of the retinal tear. It is difficult for laser spots to form on the retina. In this case, it may be necessary to increase the laser energy [12][13][14].Based on our clinical experience and references, we hypothesised that retinal pigment epithelial (RPE) photocoagulation is effective in patients with closed retinal tears during vitrectomy. Therefore, in this study, we reported the results and recovery of RPE photocoagulation closure of retinal breaks during vitrectomy.

Methods
Twenty patients, with 20 eyes affected by retinal detachment were enrolled; patients were diagnosed by the Lixiang Eye Hospital of Suzhou University from January 2015 to October 2017. We specifically included the eyes with subretinal fluid which was difficult to remove, or the transparency of the retina was seriously affected (e.g.intraretinal hemorrhage, severe retinal edema or myelinated nerve fibers) during surgery. The exclusion criteria for this study were the eyes with RPE photocoagulation range less than 50% around the breaks, or the eyes with good results after retinal photocoagulation. The follow-up time was 11-33months, and the average follow-up time was (22.1±8.0) months. This study complies with the Hospital Ethics Committee and the World Helsinki Declaration. Informed consent was signed for each patient prior to their recruitment to this study. All patients underwent a comprehensive eye exam including non-contact intraocular pressure(IOP), best corrected visual acuity (BCVA), slit lamp, B-ultrasound and fundus examination.
The postoperative follow-up time was scheduled for 1 week, 1 month, 3 months,6 months and 1 year after surgery. At each follow-up visit, the best corrected visual acuity, anterior segment and intraocular pressure examination were performed. After fully dilating the pupils, the patient's fundus was observed with a +90.0D anterior lens, and binocular indirect ophthalmoscopy was performed. When necessary, optical coherence tomography(OCT,Stratus; ZEISS; Germany) and fundus photography were used to record postoperative fundus recovery and postoperative complications. All Snellen visual acuity was converted to logarithm of the minimum angle of resolution (LogMAR) for statistical analysis. The counting fingers in visual acuity are defined as 2.3logMAR, the hand movements are 2.6logMAR, and the light perception is 2.9logMAR.

Surgical procedure
All procedures were performed by the same doctor (X.Y). After the patient was subjected to retrobulbar anaesthesia or general anaesthesia, a three-port closed vitrectomy was used to completely remove the vitreous with a surgical microscope or ocular endoscope.
The proliferating membrane, vitreous traction membrane and part of the neovascular membrane were removed,whilst minimising mechanical retraction, the retina was loosened, and if necessary the 'heavy liquid' was used to reattach the retina. Retinotomy or resection was performed on areas with severely fixed folds, retinal stiffness, or shortened, thickened or curled areas that could not be flattened. The extent of retinotomy was dependent on the size of the lesion and varied in this group from 45° to 220°.In this study, 11 patients underwent retinotomy, and RPE photocoagulation was performed at the posterior margin of the retinal break. One patient with retinal detachment had a myelinated nerve fibre on the edge of the optic disc, and the hole was located in the middle of it, and RPE photocoagulation was performed under the detached retina.RPE laser photocoagulation was performed in 3-4 rows at the peripheral edge around the hole and the tear, resulting in partial or complete closure and replacing the traditional retinal laser 6 photocoagulation. Partial RPE laser range refers to ≥180°circumference. The remainder of the range is performed by conventional retinal photocoagulation. The energy of RPE photocoagulation was 100-150mJ for 120-200ms. The light spot reaction on the RPE results in shows a paler yellow colour compared to the reaction when laser is not used. An attempt was made to avoid the occurrence of bubbles or blasting during the operation.
After gas-liquid exchange or heavy liquid use, the hole was observed during retinal reattachment, and the pale yellow RPE laser area could still be seen through the retina. If the RPE laser near the edge of the hole was insufficient due to the displacement of the hole, the conventional retinal laser was used to supplement treatment. In this study, RPE photocoagulation in case 19 was performed under endoscopy-assisted vitrectomy. After the operation, the vitreous cavity was filled with silicone oil or gas (C3F8 or SF6). The light spot reaction on the RPE results in shows a paler yellow colour compared to the reaction when laser is not used. Patients who underwent silicone oil injection had silicone oil removal 3 to 6 months after the operation.
Additionally, following retinal reattachment in the retinotomy, if necessary, the pigment epithelial layer exposed to the incision zone was also photocoagulated using a dense spot with the same energy and little gaps. The continuous photocoagulation can be applied to the wider area, so that the pigment epithelial layer is completely covered by photocoagulation.

Statistical analysis
We used SPSS 17.0 statistical software (IBM/SPSS, Inc., Chicago IL) for statistical nonparametric analysis. Visual acuity, intraocular pressure, and thickness of the neurosensory layer were expressed as mean ± standard deviation in all continuity data. A P-value of less than 0.05 was considered statistically significant.

Results
There were 20 patients with 20 affected eyes in this study, including 12 males and 8 females, average age47.4 ± 15.0 years (range, 24-69 years). The basic characteristics of the statistical data for all patients are listed in Table 1.
There were 11 eyes with rhegmatogenous retinal detachment(cases 1,2,4,7,8,9,12,13,14,17 and 20), including 3 eyes with macular holes and rhegmatogenous The mean BCVA at 6 months after surgery was significantly higher than that before surgery (p = 0.002). At the last follow-up, the mean BCVA was significantly higher than that before surgery (p = 0.001). There was no statistically significant difference between BCVA at 6 months postoperatively and at the last follow-up (p = 0.806).
In 3patients, neurosensory thickness was measured after RPE photocoagulation and retinal hole between the sensory retina and RPE [21].
In many cases, during retinal detachment surgery, retinotomy promotes a significant increase in the success of retinal reattachment. For example, cases of RRD [20] with high myopia and a marked increase in the axial length [22,23], retinal detachment with severe proliferative vitreoretinopathy (PVR) [24,25] or a giant tear. In cases of ocular trauma, the majority of patients are young and middle-aged, and the vitreous proliferation of the injured eye is often very intense. In order to loosen the severe traction, especially the peripheral traction, extensive retinotomy is required [26][27][28]. Our innovative surgical approach to RPE photocoagulation was first applied in endoscopic-assisted vitrectomy, such as in case 19.In the cases with severe retinal oedema and bleeding, conventional retinal photocoagulation is difficult to perform, however endoscope assisted RPE photocoagulation is clear and simple.Part of the pigment epithelial layer is exposed in retinotomy and retinectomy, which leads to an increase in intraocular fluid outflow and a postoperative chronichypotony. During the operation, photocoagulation of the pigment epithelial exposed area and coagulative necrosis of the pigment epithelial cells eventually cause scarring, thereby reducing the choroidal outflow of the intraocular fluid. To a certain extent, the stability of the intraocular pressure is maintained, and the occurrence of hypotony is prevented [29].Chronic hypotony did not occur in this study series. From this study, it was determined that RPE photocoagulation can be performed as necessary during surgery. Pigment epithelial photocoagulation has an irreplaceable advantage in certain cases. This procedure may also reduce the migration and evolution of pigment epithelial cells to a considerable extent, reducing the chances of PVR development [30][31][32][33][34].
In this study, the visual acuity of patients at 6 months after surgery was significantly higher than that before surgery. After 6 months, there was no significant further improvement in visual acuity. The postoperative retinal reattachment rate was 100%.
There are many possible advantages of RPE photocoagulation to seal retinal breaks

Consent for publication:
Not Applicable.

Availability of data and materials
Data are available upon request due to concerns about potential breach of confidentiality.
The dataset is only available upon request to qualified researchers. Please contact the first author, XY, for access to the data set.
assisted the operations; YYZ and XC analyzed the data. XC drafted the paper; XY contributed to the revision of the paper. All authors have read and approved the manuscript.  Figure 1 The graph shows changes in the vision of patients before and after PPV.

Tables
Postoperative visual acuity was significantly improved compared with preoperative. There was no significant change in visual acuity during postoperative follow-up.