25-gauge vitrectomy with gas tamponade for uncomplicated rhegmatogenous retinal detachment: experienced vs. inexperienced surgeons

Introduction

Rhegmatogenous retinal detachment (RRD) is a separation of the neurosensory retina from the retinal pigment epithelium caused by the influx of subretinal fluid through the retinal break (RB). Vitreous traction plays an important role in the creation of RB and the formation of RRD (Mitry et al., 2010). It is a vision-threatening condition that requires prompt surgical intervention. Several surgical techniques for the treatment of RRD have been developed, and scleral buckling (SB), pneumatic retinopexy (PnR), and pars plana vitrectomy (PPV) are currently used (Escoffery et al., 1985; Hilton & Grizzard, 1986; Chrapek et al., 2013a; Chrapek et al., 2013b; Hejsek et al., 2014; Veith et al., 2018).

PnR is a procedure that involves applying cryotherapy or laser retinopexy around the RB before or after the intravitreal injection of expansile gas. It is a less invasive procedure than PPV or SB; however, it is suitable only for uncomplicated retinal detachments and requires careful patient selection to achieve good results (Hillier et al., 2019).

In SB, silicone exoplants are used to indent the sclera over the RB previously treated with cryopexy or laser retinopexy, thus lessening the vitreous traction. It is an extraocular procedure with advantages over PPV, especially in young, phakic patients with clear optical media and without posterior vitreous detachment (PVD). It can be combined with subretinal fluid drainage and intravitreal injection of expansile gas (Park, Lee & Lee, 2018).

PPV is an intraocular surgery aimed at removing the vitreous, draining the subretinal fluid and treating the RB with cryopexy or laser retinopexy. Air, mixtures of expansile gases or silicone oil are used as a tamponade to block the RB for the time of chorioretinal scar formation (Escoffery et al., 1985). PPV can be combined with SB, although whether this affects the outcome remains uncertain (Nichani et al., 2022; Rajsirisongsri et al., 2024). The development of sutureless, small-gauge vitrectomy and advancements in surgical techniques and equipment have made PPV increasingly popular among surgeons in the management of RRD (McLeod, 2004; Ho et al., 2008). Thus, it is becoming increasingly important even for beginner vitreoretinal surgeons to manage retinal detachment cases successfully.

Several studies have compared the results of PPV for retinal detachment between experienced and inexperienced surgeons (Heimann et al., 2006; Dugas et al., 2009; Ehrlich et al., 2011; Mazinani et al., 2012; Keller, Haynes & Sparrow, 2016; Sallam et al., 2018). In the abovementioned studies, the retinal reattachment success rates were comparable between experienced and inexperienced vitreoretinal surgeons. However, all the abovementioned studies have some limitations. Some of them use older generations of vitrectomy machines or contact lenses for visualisation, and their results do not reflect the advancements in surgical instruments and machines (Heimann et al., 2006; Dugas et al., 2009; Ehrlich et al., 2011; Mazinani et al., 2012). Some focus on operation success rates and visual acuity change, but do not report on complication rates (Dugas et al., 2009; Mazinani et al., 2012; Keller, Haynes & Sparrow, 2016; Sallam et al., 2018). The study by Sallam et al. (2018) is a national clinical database study and does not include details about the surgical techniques or instruments used for the PPV. The study by Heimann et al. (2006) focuses primarily on the risk factors for primary surgery failure, with surgeon experience being just one of the studied variables.

In our study, we compared the anatomical success rate, change in visual acuity, employed surgical techniques, and complication rates in patients with RRD, performed using a 25-gauge (25 g) PPV with gas tamponade without scleral buckling, between an experienced surgeon and two inexperienced surgeons using a high-speed vitrectomy probe and noncontact wide-angle visualisation system. This article was published as a preprint (Pencak et al., 2020).

Materials and Methods

Results

We examined the documentation of 407 PPV for retinal detachment performed on 368 eyes between October 2015 and June 2018 in the Department of Ophthalmology of the University Hospital Kralovske Vinohrady, Prague. 245 PPV were performed by the experienced surgeon, 162 by the inexperienced surgeons. The reason for exclusion from the study was the use of silicone oil tamponade in 53 PPV performed by the experienced surgeon and in 21 PPV performed by the inexperienced surgeons, other inclusion and exclusion criteria were not met in other cases. In the end, 216 eyes of 216 patients met all the inclusion and exclusion criteria and were included in the study, 106 in the ESG and 110 in the ISG. Table 1 shows the baseline demographic and clinical characteristics of the participants.

The date of diagnosis of retinal detachment was recorded in 48 patients in both groups. The median time to surgery was 2.0 (IQR 0–3) days in ESG and 1.0 (IQR 0–3) day in ISG (P = 0.435). The duration of symptoms was recorded in 81 patients in ESG and 79 patients in ISG. The median duration of symptoms was 5.0 (IQR 0–10) days in both groups (P = 0.889). One patient in the ESG underwent unsuccessful pneumatic retinopexy with laser barricade around the RB before the PPV, and one patient had undergone scleral buckling for retinal detachment 11 years before the PPV. In the ISG, two patients underwent unsuccessful laser retinopexy around the RB before the PPV. There were two patients with high myopia in both ESG and ISG. There was no patient with a giant retinal tear in either group. A macular hole was present preoperatively in two patients in the ESG. ILM peeling was performed during PPV for retinal detachment in both patients, and the macular hole was closed in one of them. Table 2 shows the surgical techniques and gas tamponade used in both groups.

The results are shown in Table 3. The difference in single surgery success rates between individual surgeons was not statistically significant. After the second surgery, the success rate was 100% in the ESG and 97.3% in the ISG; a 100% success rate was achieved in the ISG after the third surgery. The causes of failure were RB in the scar after cryopexy in four patients in the ESG and two patients in the ISG, reopening of the original RB in one patient in the ISG, a newly diagnosed RB in three patients in the ISG, and a PVR in two patients in the ESG and in one patient in the ISG.

Of the 13 patients who required a second surgery for retinal redetachment, gas tamponade was used in 10, and silicone oil was used in three. PVR dissection was performed in two patients and retinectomy in one patient during the second surgery. Silicone oil was used as a tamponade for all three patients who required a third surgery. PVR dissection was performed in one patient and retinectomy in two patients during the third surgery. In one patient who required three surgeries, silicone oil was left in the eye as a permanent tamponade due to hypotony. Scleral buckle was not used in any of the patients. The median final BCVA in patients who required two surgeries was 0.417 (IQR 0.250–0.500) decimal. The median final BCVA in patients who required three surgeries was 0.250 (IQR 0.188–0.250) decimal.

The complication rate was similar between both groups. The most common complication was the progression of cataract in phakic patients who required cataract surgery. All these patients underwent cataract surgery before the end of follow-up. The second most common complication was postoperative intraocular hypertension. Three patients in each group required temporary therapy with oral acetazolamide, and one patient in the ESG underwent laser iridotomy. All other cases were resolved with IOP-lowering topical medication. Other complications included intraocular hypotony in one patient in each group that resolved without therapy. Two patients in the ESG and four patients in the ISG underwent further PPV for ERM formation in the macula. A postoperative macular hole occurred in two patients in the ESG and one patient in the ISG. Subretinal perfluorocarbon occurred in one patient in ESG. In ISG, one patient suffered an iatrogenic posterior lens capsule tear, and one patient underwent further PPV for intraocular haemorrhage.

Discussion

In our study, inexperienced surgeons were able to match the success rate of an experienced surgeon in uncomplicated RRD from the very start, suggesting a short learning curve. There is a possible bias in our study in the patient selection in each group. It may be argued that more experienced surgeons would operate on more complex cases. It is impossible to eliminate selection bias in retrospective studies. We tried to minimise it by including only patients with no or only mild PVR. We also excluded patients where silicon oil was used as tamponade and patients operated using 23-gauge vitrectomy, because in our clinic, there is a tendency to use both in more complex cases. The only statistically significant difference between the two groups was in the age of the patients. We do not believe it influenced the results since all the other anatomical and functional characteristics of retinal detachment were similar in both groups. It should also be noted that 68% of all the PPV for retinal detachment performed by inexperienced surgeons met all the inclusion and exclusion criteria and were included in the study. Also, scleral buckling is not performed in our clinic, the only alternative to PPV for the treatment of RRD in our clinic is pneumatic retinopexy.

Another factor that might have influenced the results of the study is the difference in manual skills between surgeons. However, this factor is hard to quantify. Differences in manual skills might affect the time of surgery, the incidence of iatrogenic retinal tears, iatrogenic posterior lens capsule tears, cases of lens touch in phakic patients, and single surgery success rates. We evaluated the incidence of iatrogenic retinal tears, iatrogenic posterior lens capsule tears and single surgery success rates and found no difference between individual surgeons.

Single operation success rates in both groups were comparable to previously published figures for 25 g PPV with gas tamponade (Miller et al., 2008; Mura, Tan & De Smet, 2009; Kunikata & Nishida, 2010; Gotzaridis et al., 2016). We believe this is due to the simplified operation technique used in our clinic. We try to limit surgical techniques that have not been shown to improve uncomplicated RRD surgery outcomes. These include the use of perfluorodecalin (Schneider, Geraets & Johnson, 2012; Vidne et al., 2018), complete subretinal fluid drainage (Yamaguchi, Ataka & Shiraki, 2014; Gotzaridis et al., 2016; Chen et al., 2017), and 360° retinopexy (Bilgin et al., 2019). Limiting the use of these techniques may also prevent certain postoperative complications (Wilbanks et al., 1996; Cauchi, Azuara-Blanco & McKenzie, 2005; Bouheraoua et al., 2015; Liu, Gao & Liang, 2018).

It should be noted that there is no widely accepted definition of an experienced and an inexperienced surgeon. Different definitions were used in studies comparing the results of experienced and inexperienced surgeons. In the study by Dugas et al. (2009), less experienced surgeons were defined as having performed less than 50 PPV, and more experienced as having performed more than 3,000 PPV. In the study by Heimann et al. (2006), three groups of surgeons were defined. Surgeons with more than 100 PPV were labelled as “specialists”, with 31 to 100 PPV as “non-beginners”, and those with 30 or fewer PPV as “beginners”. In the study by Ehrlich et al. (2011), beginner surgeons were vitreoretinal fellows, in the study by Sallam et al. (2018), the results of consultant surgeons, independent non-consultant surgeons and trainees (fellows and specialist registrars) were compared. In the study by Keller, Haynes & Sparrow (2016), trainees with less than 12 months’ experience were compared with senior surgeons who had completed 30 or more operations in the audit period. In our study, both beginner surgeons performed less than 30 PPV before the start of the inclusion period. Both were trained in PPV two years prior; however, it should be noted that in the Czech Republic, there are no institutionalised subspecialty education programs that would be comparable to fellowship programs.

The final success rate was 100% in both groups. This might have been thanks to the low number of reoperations—only 13 patients had retinal redetachment after the first surgery. Also, the time to surgery from initial diagnosis of retinal detachment was very short in both groups (median 2.0 days in ESG, 1.0 day in ISG). Another factor might have been the short duration of retinal detachment in both groups, with a median duration of symptoms of only 5.0 days. Also, all patients were instructed on the normal postoperative course of events and were encouraged to come for a check-up immediately in case of abnormal development. The first check-up was usually scheduled one month after the surgery. We believe these factors led to early diagnosis of retinal redetachment and prompt reoperation, which lowered the risk of complications that might have led to unfavourable outcomes, like the development of PVR.

Surgical techniques employed by the experienced and inexperienced surgeons in our study were similar, with the exception of the selection of intraocular tamponade and the frequency of sclerotomy suturing. Inexperienced surgeons used C3F8 gas more frequently, a longer-lasting and more “secure” option. We believe this was due to inexperience and subsequent lack of self-confidence, which led inexperienced surgeons to use C3F8 gas even when it might not have been necessary. In our opinion, the more frequent use of C3F8 in ISG was also the reason for the higher cataract surgery rate after the PPV between the ESG and ISG, although the difference was not statistically significant. The higher rate of sclerotomy suturing could possibly be explained by the longer operating times in the ISG; however, we do not have data to support this claim, as the duration of the operation was not recorded.

The complication rate was similar in both study groups. There was a notable difference in cataract surgery rate after the PPV between the ESG and ISG, although the difference was not statistically significant. Apart from the more frequent use of C3F8 gas for tamponade, the patient age difference between the ISG and the ESG might have been a contributing factor. The low postoperative intraocular hypotony rate can be explained by the oblique cannula insertion technique used by all surgeons (Acar et al., 2008; Bourgault & Tourville, 2012) and the use of digital palpation to assess IOP after the removal of cannulas. However, it should be noted that some complications were not evaluated. These include postoperative cystoid macular oedema, as optical coherence tomography (OCT) was not performed regularly in all patients during the follow-up visits.

Other studies have shown comparable results in PPV for RRD between experienced and inexperienced surgeons (Heimann et al., 2006; Dugas et al., 2009; Ehrlich et al., 2011; Mazinani et al., 2012; Keller, Haynes & Sparrow, 2016; Sallam et al., 2018). Ehrlich et al. (2011) also compared the surgical techniques employed by vitreoretinal fellows and consultants. Compared to our study, they used perfluorocarbon more frequently—in 28.3% of eyes in the vitreoretinal fellow group and 32% of eyes in the consultant group, compared to 9.1% in the ISG and 5.7% in the ESG in our study. Interestingly, the vitreoretinal fellows in their study tended to use SF6 gas more frequently than the consultants, as opposed to our study. The study does not report the extent of retinopexy, the number of eyes requiring sclerotomy sutures, nor the number of retinotomies.

Dugas et al. (2009) and Ehrlich et al. (2011) compared the success rates of sutureless PPV for RRD between experienced and inexperienced surgeons and used similar exclusion and inclusion criteria as our study. In both these studies, the combined single operation success rate was 75% (75.4% and 80.9%, respectively, for experienced surgeons and 74.8% and 70.0%, respectively, for inexperienced surgeons). This was significantly lower than in our study. This can be explained by the improvements in surgical instruments, vitrectomy machines, and surgical techniques. For example, high-speed vitrectomy has been shown to lower the number of iatrogenic retinal tears during PPV (Rizzo, Genovesi-Ebert & Belting, 2011). Other reasons could be a low incidence of different factors that make RRD surgery more challenging, such as RRD in highly myopic eyes, giant retinal tears, etc., in our study. On the other hand, it is worth noting that in a study by Ehrlich et al. (2011), less experienced vitrectomy surgeons were recruited from fellows who had extensive experience in other intraocular procedures. In contrast, in our study, both inexperienced surgeons had no previous experience with intraocular surgery.

As for the impact of our study on clinical practice and surgical training curricula, we believe that with a supportive teaching environment, close observation by the teaching surgeon at the beginning of the training period and constructive feedback, the learning curve for PPV can be quite short. In our study, both inexperienced surgeons without any previous experience with intraocular surgery spent 2 years in training before performing their first complete vitrectomy. However, this was relatively low-intensity training with only one day a week spent at the operating theatre. We believe that with higher-intensity training, this period can be significantly reduced. The training time can be reduced further with modern technological means, both in the operating theatre and outside. In the operating theatre, 3D visualisation systems can significantly speed up and simplify teacher-trainee interaction and communication (Razavi et al., 2023). Outside the operating theatre, virtual reality training and modern simulators can prepare trainees for different scenarios they can encounter during surgery (Deuchler et al., 2016).

Both inexperienced surgeons matched the results of an experienced vitreoretinal surgeon in PPV for uncomplicated RRD. This fact can benefit the patients, because uncomplicated RRD usually benefit most from prompt intervention and surgeon unavailability is one of the factors that can lead to delays.

The limitations of our study include a low number of evaluated surgeons, the retrospective nature of our study and a subsequent lack of randomisation. Besides the selection bias, other problems associated with retrospective studies could have influenced our results. These include incomplete reporting on some observed variables, namely, time from diagnosis to surgery and duration of symptoms. Furthermore, the minimal follow-up interval in our study was 3 months, which might have caused the underestimation of the number of subsequent retinal detachments. However, according to previously published reports, most retinal redetachments occur within 3 months after the initial surgery (Lee et al., 2014; Heath Jeffery et al., 2020).

Conclusions

A 25 g PPV with gas tamponade for the treatment of RRD yields excellent anatomical results and improvement in BCVA. With good technique and the use of modern vitrectomy machines and instruments, some inexperienced surgeons can achieve a high single-surgery success rate, suggesting a short learning curve. The complication rate is comparable between experienced and inexperienced surgeons.

Supplemental Information