Patient selection
Ours was a single-center retrospective comparative consecutive case series study. Patients with RRD treated with 25g PPV with gas tamponade without scleral buckling between October 2015 and June 2018 in the Department of Ophthalmology of the University Hospital Kralovske Vinohrady, Prague, were included in the study. Both pseudophakic and phakic patients were included. We excluded patients with proliferative vitreoretinopathy (PVR) grade C and higher, patients with a previous perforating eye injury and patients with follow-up periods shorter than 3 months. Patients were then divided into two groups. In the first group (the experienced surgeon group [ESG]) include patients treated by an experienced surgeon (MV, 1909 PPV at the beginning and 2438 PPV at the end of the inclusion period). In the second group (the inexperienced surgeon group [ISG]) included patients treated by two inexperienced surgeons (MP and ZS, 28 and 22 PPV, respectively, at the beginning and 172 and 161 PPV, respectively, at the end of the inclusion period). Both inexperienced surgeons were trained by the same surgeon (MV) and assisted in PPV operations for 2 years prior to their first solo vitrectomy. Both had no previous experience with intraocular surgery. We compared the anatomical success rate, change in visual acuity, and complication rate between groups. All patients signed an informed consent before the surgery. The study protocol adhered to the tenets of the Declaration of Helsinki Principles. Approval was granted by the Ethics Committee of Third Faculty of Medicine, Charles University, Prague and the Kralovske Vinohrady University Hospital, Prague under the number EK-R/01/0/2020.
Pre- and postoperative assessment
In all patients, the pre- and postoperative best-corrected visual acuity (BCVA) was assessed using ETDRS charts, results of which were then converted to decimal values for statistical analysis. BCVA of counting fingers, hand motion, or light perception were converted do decimal values using the chart published by Holladay[11]. Pre- and postoperative slit-lamp examination and fundus biomicroscopy were performed to determine the extent of RRD, the location and number of retinal breaks (RB), grading of the preoperative PVR, as well as to assess the postoperative state of the retina. Intraocular pressure (IOP) was measured using non-contact tonometry using a NT-530 (Nidek, Aichi, Japan) preoperatively, on the first postoperative day, and on all visits during follow-up. Hypotony was defined as an IOP < 10 mmHg and hypertension as an IOP > 25 mmHg. The primary endpoint was the single surgery retinal reattachment rate. The secondary endpoints were the postoperative change in BCVA (in decimal) and complication rates. All postoperative endpoints were assessed on the last day of follow-up
Surgical technique
All patients underwent a three-port 25g PPV, using the oblique cannula insertion technique described previously[6], using a Constellation® vitrectomy machine (Alcon, Fort Worth, TX, USA), with an Ultravit® vitrectomy probe with a cutting rate of 5000 cuts/min. Valved cannulas were used in all patients starting in August 2016. A Resight® 500 (Zeiss, Germany) wide-angle visualization system was used to visualize the fundus. Cannulas were placed 3.5 mm posterior to the limbus in pseudophakic eyes and 4 mm posterior to the limbus in phakic eyes. A core vitrectomy was performed, followed by peripheral vitreous removal assisted by scleral indentation with light probe. Shaving of the vitreous base was performed around RB and suspicious lesions. Anterior hyaloid dissection was not performed in phakic patients. Perfluorodecalin (Arcaline, Arcadophta, France) was used in some patients to immobilize a detached retina and facilitate peripheral vitreous removal. If an epiretinal membrane (ERM) or macular hole (MH) was present in the macula, brilliant blue (Ocublue, Aurolab, India) dye was used and ERM and internal limiting membrane (ILM) peeling was performed. Perfluorodecalin was used in patients with a detached macula when membrane peeling was indicated. Fluid-air exchange with subretinal fluid (SRF) drainage through a peripheral RB was performed using a Charles Flute Cannula (Alcon, Fort Worth, TX, USA) or a vitrectomy probe. Perfluorodecalin was used or posterior retinotomies were performed to achieve complete SRF drainage in patients where the surgeon was concerned about the risk of a retinal fold forming in the macula. Posterior retinotomies were also performed in patients where pre-existing RB were not identified during vitrectomy. Perfluorodecalin was used and retinotomies were performed at the discretion of the surgeon. Complete SRF drainage was not required. Retinopexy of the margins of RB, lattice degenerations and other suspicious peripheral lesions were performed under air using an endolaser or cryotherapy probe. In some patients, retinopexy to the extent of the detached retina or a 360° retinopexy was performed. The extent of retinopexy depended on the number and location of RB and the lattice degenerations and was also at the discretion of the surgeon. A non-expansive 20% mixture of sulfur hexafluoride (SF6) (Alchimia, Italy) or a 15% mixture of perfluoropropane (C3F8) (Alchimia, Italy) was used as a tamponade in all patients. The decision on which gas to use depended largely on the locations of RB. In patients with superior RB, 20% SF6 was generally used; in patients with inferior RB, 15% C3F8 was generally used. However other factors were taken into consideration, like the presence and location of lattice degenerations and other suspicious peripheral lesions, patients’ ability to posture after the operation etc. The final decision on the gas tamponade was made solely by the surgeon. After the air-gas exchange, cannulas were removed and the tightness of the sclerotomies was checked. If leakage was present, a digital massage of the sclerotomy was performed. If leakage persisted after digital massage, the sclerotomy was sutured using Vicryl 8-0 (Ethicon, Johnson & Johnson Int). After all the cannulas were removed, digital palpation was used to check the IOP at the end of the operation. If the IOP was considered low, additional gas mixture was injected through the sclera using a 30-gauge needle. No scleral buckling was performed. Depending on the locations of break, patients were instructed on proper head positioning for the one-week period after the operation.
Statistical analysis
Quantitative variables - age, follow-up period, pre- and postoperative BCVA, BCVA change, number of RB and extent of retinal detachment were tested by means of Kolmogorov-Smirnov test on normality. Age and BCVA change were found to have normal distribution. They are given as means and standard deviations. To compare the age of patients and BCVA change after operation between both groups, the independent Student’s t-test was calculated. Follow-up period, pre- and postoperative BCVA, number of RB and extent of retinal detachment were found to have non-normal distribution and are given as median and interquartile range (IQR). Mann-Whitney U test was used to compare these variables between both groups. To compare the BCVA before and after surgery, the paired samples Wilcoxon test was used for all three surgeons together and for each surgeon separately. Retinal attachment success rates, lens status, state of the macula and the number of patients with inferior RB in the ESG and ISG are given as total frequency and percentages. The differences in retinal attachment success rates, lens status, state of the macula, complication rate and the number of patients with an inferior RB between the ESG and ISG were analyzed using contingency tables – the Pearson’s chi-square test or in the case of small sample numbers, the exact Fisher test was calculated. Statsoft STATISTICA version 9 was used for statistical analysis. P-values less than 0.05 were considered to be statistically significant.