Trimanual vitrectomy versus bimanual vitrectomy in tractional proliferative diabetic retinopathy

ABSTRACT

Objective

To evaluate the impact as well as safety of trimanual vitrectomy versus bimanual vitrectomy in cases with complicated proliferative diabetic retinopathy (PDR) necessitating surgical intervention.

Methods

We designed a randomized, interventional, prospective study on 40 pseudophakic eyes suffering from tractional retinal detachment (TRD). 20 eyes of 16 cases with trimanual 23-gauge Parsplana vitrectomy (PPV) were categorized as Group A ”trimanual group.” Twenty eyes of 18 cases with bimanual 23-gauge PPV were categorized as Group B” bimanual group” in the duration between May 2021 and December 2022. All cases were under continual observation for ≥ 6 months postoperatively.

Results

A total of 40 eyes out of 34 cases with the mean age of cases 50.5 years (ranged between 29–69 y) in group A and 47.9 (ranged between 33–60 y) old in group B were assessed. Retinal attachment, total elimination of the pre-retinal proliferation attained in all cases of the two groups. Surgical time was statistically significant shorter in the trimanual group compared to bimanual group (p value < 0.05). Both groups gained statistically significant better vision. The mean pre-operative VA (Visual acuity) in group A was 1.75 (Log MAR) and 1.59 in group B, whereas the mean final visual outcome 6 months following operation was 0.73 in the 1^st group & 0.82 in the 2^nd one (p value < 0.001*).

Conclusion

Trimanual PPV is a safe as well as efficient approach that may ameliorate the result of vitrectomy in tractional PDR.

KEYWORDS:

• Tractional retinal detachment
• perfluorocarbon liquids
• vital dyes
• pars plana vitrectomy

1. Introduction

Since advocation of vitreous operation for the 1^st time, it’s one of the basic for managing the complicated cases of proliferative retinopathy. Diabetic vitrectomy might be considered the only treatment available for such cases to prevent the contraction effect caused by the posterior vitreous surface on the retina [1]. The essential step of this maneuver necessitates precise dealing with the fibrovascular membrane as marked barriers are often expected [2].

Dissection of epiretinal membranes (ERMS) is actually the essential challenge of tractional retinal detachment (TRD) operation, yet fortunately results showed a dramatic improvement over the years due to novel tools and approaches. High-speed 23 g & 27 g vitreous cutters help conformal cutter delamination along with segmenting the bridging tissue located between various ERMS epicenters [3] Bimanual operation term has been suggested to detect vitrectomy surgery where devices utilized with the two hands of the surgeons, this method for fibrovascular tissue elimination is in particular of great benefits for eyes with marked fibrovascular proliferation [4]. Bimanual vitrectomy is utilized in advanced diabetics suffering TRD for the total and safe elimination of ERMS [5]. Generally, the principle of bimanual operation is to prevent the undesirable pulling and pushing-out forces of scissors and pics that happen for the retina during dissecting the ERM [6]. In complicated diabetic patients suffering diabetic TRD and proliferative vitreoretinopathy, grasping of the membrane is easier on using bimanual vitrectomy via the use of forceps and separating them from the retina via dissecting them by scissors along with cutters. The advantage of a wide range of particularly designed devices has accelerated the method of membrane dissection [7,8]. Chandelier illumination is pivotal for the bimanual operation. Despite beneficial, there’s still some concerns as regard the glare faced the surgeons with the such system, in particular after fluid-air exchange. since the illumination is statically produced, it does not allow accurate detection of the planes needed to carry out dissection. The released light remains inadequate for illuminating the field completely through one incision [9]. Recently, modification involves adding of various light probes simultaneously or reposition the light probe at time of operation [10–12]. Tri-manual operation had been suggested by current published studies, in which perfluorocarbon liquids (PFCLS) permitting a free hand for the operation to carry out a safe dissection of membranes for more challenging patients.

The distinct physical criteria render it an optimal intra operative method to enhance the efficacy as well as the safety of surgical maneuvers in difficult diabetic patients. The essential function of perfluorocarbon liquids in vitreoretinal operation includes the fixation of the detached retinal portions, causing displacement of the sub-retinal & sub choroidal fluid in front, and helping to remove foreign body. The related clinical application includes retinal detachment (RD) with marked proliferative vitreoretinopathy (PVR), giant tears, Diabetic retinopathy (DR), retinopathy of prematurity (ROP), and posterior dislocation of the crystalline lens and intraocular lens (IOL). The utilization of PFCLS actually increased in the last decades. Various PFCLS related inflammation of the eyes were determined in in vitro research, animal models, and clinical studies [13].

The total elimination of PFCLS is better to be done at the end of the operation in the vast majority of patients. PFCLS have increased specific gravity that ranges between 1.76 and 2.03, reduced surface tension, decreased viscosity, optical clarity, and transparency, these physical properties render these substances reduced intraoperative methods for vitreoretinal operation [14], various types of PFCLS are used in ophthalmology in various nations including perfluoro octane (PFO), perfluorodecalin (PFD), perfluorotributylamide (PFTB), and perfluorooctylbromide [15,16]. The specific gravity of PFCLS entitles the liquid to flatten the detached retinal portion and displacement of the subretinal fluids, the transparency of PLFCS provide a clear media and allow intraoperative photocoagulation, the surface tension of PFCLS proves the liquid staying proportionally cohesive following injection in the vitreous cavity, the decreased viscosity enables PFCLS to be easily handled during injection and removal [17–20].

Zimmerman & Faris utilized PFCLS as an intraoperative equipment that relocates the detached retina in 1982 [21]. Chang et al. used PFCLS in vitreous operation of RD cases with marked PVR. The specific gravity of PFCLS used is ~2 times more compared to perfusion solution. Thus, during injection into the vitreous at time of vitrectomy, the gravity of PFCLS produces a force opposite to the interface downward. The use of PFCL during vitreous operation for PDR was documented by many researchers, often in very complicated states. The function of PFCLS during operation is the same as cases of RD with marked PVR. PFCL is considered a helpful assistant during vitrectomy for extensive PDR, in particular to make the shrunken retina flat, it enables easy access to membrane dissection. In addition, PFCL is efficient in flattening RD that occurs during relieving tight vitreoretinal adhesions [22]. Besides, PFCL use offers a better condition to carry out pan retinal photocoagulation when required with lower energy. To date, comparative studies that evaluated such methods with other vitrectomy procedures are few in the published studies. The vital dyes facilitating good detection of the membrane allow precise dissection with subsequently improved outcomes of surgery. The combined 0.15% trypan blue, 0.025% Brilliant Blue G (BBG), and 4% Polyethylene glycol (PEG) (Membrane Blue Dual, DORC International) can efficiently stain the internal limiting membrane (ILM) in conjunction with the ERMs [23].

2. Aim of the work

Our work aimed to evaluate the effect along with the safety of trimanual vitrectomy versus bimanual vitrectomy in patients with complicated PDR requiring surgical intervention.

3. Subjects and methods

This study was performed according to the Declaration of Helsinki along with the regulations of good clinical practice. We informed the patients about the possible risks as well as beneficial effects of the management and they wrote an informed consent prior to the approaches.

Our work is a randomized, interventional, prospective, comparative study. It was performed on pseudophakic eyes with TRD that involved ≥3 Disc Diameter (DD) fibro- vascular membranes threatening and/including the macula, with or without vitreous hemorrhage(VH) and combined tractional/rhegmatogenous detachment.

Twenty eyes out of 16 patients who had trimanual 23-gauge PPV were included in Group A “trimanual group.” Twenty eyes of 18 cases with bimanual 23-gauge PPV were included in Group B “bimanual group” in the period between May 2021 and December2022.

3.1. Patient examinations

Medical record for each patient was assessed for age, sex, type of Diabetes Mellitus (DM), the condition of the lens, nature of TRD, coexisted complications as well as any previous treatments. Argon laser photocoagulation was clinically cateogrized into 3 grades that included: 1-none (there’s no photocoagulation), 2- incomplete (photocoagulation was <3 sessions or photocoagulated retina was <50% of the peripheral retinal portion), and complete (>3 sessions of photocoagulation were carried out, and photocoagulation in the vast majority of the peripheral retinal portion was complete).

Every patient was examined before the surgery and, on the 1^st day, 1^st week, 1^st month, and 6^th month following vitrectomy (postoperative assessment). At each visit, examination of the eye was done completely and included Best corrected visual acuity (BCVA) using Snellen’s chart, slit-lamp biomicroscopy, Intraocular Pressure (IOP) measurement by Goldmann applanation tonometry, and stereoscopic biomicroscopy of the retina using 90-Diopter(D) lens. The macular state and the degree of the TRD were determined through fundus examination, fluorescein angiographyn (FA), and ocular coherence tomography (OCT) following operation. In cases of vitreous hemorrhage, B-scan US (Ultrasound) before operation, and fundus examination during the operation were conducted as well.

The recorded intraoperative data such as the operating time and membrane removal time, fluid – air exchange, gas or silicone oil (SO) tamponades, intra operative complications, and suture place on terminating the procedure. The operating time refers to the time consumed from inserting till removing the lid speculum, served as a key metric. The surgical approach of membrane removal starts following removing the vitreous core and hemorrhage. The membrane removal time refers to the total duration of complete fibro – vascular membrane elimination and included intraoperative bleeding. Intraoperative bleeding was classified based on the magnitude: Grade I: mild (didn’t hinder visualization and was easily aspirated), grade II: moderate (hindered visualization for small period yet was still easily aspirated), and grade III: marked (hindered visualization totally, blood present at the interface between PFCLand balanced salt solution and was hardly aspirated or even inconceivable to be aspirated). Bleeding was controlled via either intraocular compression or diathermy. The parameters were evaluated by the surgeon and his assistant and recording of the data in the operative report of the patient was done.

3.2. Post- operative complications were recorded including

The extent of inflammation inside the eye was evaluated clinically by the existence of temporary cells, flare and fibrin inside the Anterior chamber(AC) or vitreous cavity and/or an increase IOP reaching >21 mmHg. Early VH postoperatively refers to VH that occurred within 1 month following the operation while late postoperative VH refers to VH that occurred >1 month following the operation.

3.3. Surgical procedures

The eyes were divided on 1:1 basis by stratified randomization according to the age, gender, and type of diabetes (type 1 versus type 2) and the indication of PPV into one of two groups. Randomization sequence was created using Excel 2016 (Microsoft, Redmond, WA, USA) by an independent computer operator.

An intravitreal injection of anti-Vascular Endothelial Growth Factor (VEFG) was administered in the involved eye 48 Hour (H)–72 H prior to vitrectomy. The PPV operation was done under general anesthesia. All procedures were done by the same experienced surgeon at a single hospital. In both groups, three oblique sclerotomies were done by 23-G trocar-cannula system (CONSTELLATION® Vision System, Advanced ULTRAVIT® ALCON. United States.) in the Pars Plana (PP) at 3.5 mm. Placing of the infusion trocar in the inferior lateral quadrant was performed. The two upper insertions superiorly were utilized for fibreoptic and vitreous cutter attachment, then insertion of two self-retaining cannula-dependent 25 G disposable Eckardt twin light chandelier endoilluminator (DORC, Zuid land, Netherlands) at the 6 & 12 o’clock positions. Visualization of the fundus was done via the use of a non-contact wide-angle viewing system (Resight; Carl Zeiss Meditec AG, Jena, Germany) during vitrectomy. The approach starts by removing the cortical vitreous along with the VH, then removal the posterior hyaloid membrane between the retinal portion and the base of the vitreous. Performing Peripheral vitrectomy was done aiming at relieving the front-back tractions. The Vitreous base was shaved (360 degrees) underneath the scleral depression aiming at removing the peripheral cortical gel and releasing the epicenters of fibro vascular proliferation anteriorly.

3.4. The subsequent steps of the operations were different in the 2 groups

In the 1^st group, injection of ~ 2-mL size PFCL bubble (Arcaline; Arcadophtha, Toulouse, France) at the posterior pole aiming at stabilizing the retinal portions (Figure 1). Thereafter, staining of the ILM as well as the pre-retinal proliferation with combined 0.15% trypan blue, 0.025% BBG, and 4.00% PEG (Membrane Blue-Dual, DORC International).

Figure 1. Surgical steps of case # 6 (right eye with TRD and ERM of 49-year-old female patient suffering type 1 diabetes for 14 years) in (group A) (A) Injection of the vital dye under the PFCL. (B) Bimanual dissection of fibrovascular membranes. (C) after removal of fibrovascular proliferations.

Group B (the 2^nd group), complete elimination of the posterior hyaloid along with the fibro- vascular proliferations around the retina was carried out by bimanual dissection, using vital dyes only as an adjunct in the operation. Removal of the fibro-vascular membranes in the two groups was carried out via bimanual manipulation approaches like membrane peeling, segmentation, and delamination. insertion of 23-gauge equipment was done through the microcannulas and were utilized for vitrectomy that included the vitreous cutters, endoilluminators, microscissors, forcepses, endolaser probes, endodiathermy probes, and flute needles with back flush handles. Removal of the Tangential tractions were done via scissors and elevation of the proliferative tissues via end-gripping forceps. Elimination of the fibro-vascular proliferations were carried out via blunt along with sharp dissection. The excision of residual as well as free fibrovascular tissues was carried out via scissors that could be totally & safely removed using a vitreous cutter (Figure 2). Controlling of hemorrhage was done at the same time via intraocular diathermy & a soft-tip cannula. Endolaser photocoagulation was carried out so as to complete Panretinal Photocoagulation (PRP) up to the anterior retina. A demarcation laser was used to treat retinal breaks. Complete PFC removal was done in group A, following fluid – air exchange, Air, Sulfur hexafluoride (SF6), or silicone oil (Silicone mate” Polydimethylsiloxane”1300 Centistoke. GERMANY) was used as a tamponade in all cases.

Figure 2. surgical steps of case # 5 (left eye with TRD and ERM of 51-year-old male patient suffering type 1 diabetes for14 years) in bimanual PPV group (group B): (A) Injection of the vital dye over the posterior pole. (B) Stained fibrovascular proliferations. (C) Bimanual dissection of the preretinal proliferation with intraoperative bleeding.

Lastly, the microcannulas used were cautiously removed from all scleral incisions via specifically designed cannular forceps without applying any sutures to the sclera or conjunctiva. The sclerotomy sites were firmly pressed using a cotton-tip applicator to promote closure of the sclerotomies, and to return back the conjunctiva, we displaced to its normal site. The suturing of the sclerotomy sites were done using 7/0 vicryl if a sign of leaking was determined (for instance, bleb formation or hypotony). Following applying 20-second compression three times, injection of dexamethasone in inferior subconjunctival space was performed. Following surgery, topical antibiotic and steroid therapy were prescribed for a month. Cases who had gas or silicone oil tamponades were guided to keep their faces down for 7–10 days. During the follow-up duration, antiglaucoma eye drops like β-blockers, carbonic anhydrase inhibitors (CAIs), or prostaglandins (PGs) analogue were administered when IOP was >21 mmHg.

3.5. Statistical analysis of the data

Results were collected, fed to the computer for analysis via IBM SPSS software package version 20.0. (Armonk, NY: IBM Corp). Description of the qualitative data was done using numbers and percentages The Kolmogorov-Smirnov test was utilized for verifying the normality of distribution Description of the quantitative data was done using range (minimum & maximum), mean standard deviation (SD), median, and Inter quartile range (IQR). Significant level of P-value equals 5% indicates significant difference.

The Tests we utilized included:

1. 1- Chi-square test

   For categorical variables, for comparison between different groups.

2. Fisher’s Exact or Monte Carlo correction

   Correction for chi-square when >20 percent of the cells have an expected count <5.

3. Student t-test

   For normally distributed quantitative variables, for comparing between 2 studied groups.

4. Mann Whitney test

   For abnormal distribution of the quantitative variables, for comparing between 2 studied groups.

5. Friedman test

   For abnormal distribution of the quantitative variables, for comparing >2 periods or stages and Post Hoc Test (Dunn’s) for pairwise comparisons.

4. Results

This study was a randomized, interventional, prospective, and comparative conducted on pseudophakic eyes were categorized into three groups according to fundus examination as in group A:16 eyes with TRD and fibro-vascular proliferation, three eyes with TRD and sub hyaloid hemorrhage, and one eye with combined tractional and rhegmatogenous RD. In group B,12 eyes, 7 eyes, and one eye presented with TRD with fibro-vascular proliferation, TRD with sub-hyaloid hemorrhage, and combined tractional and rhegmatogenous detachment, respectively. This study was conducted between May 2021 and December 2022.

Twenty eyes out of 16 cases with trimanual 23-gaugePPV using PFC and vital dyes were assigned to Group A “trimanual group.” Twenty eyes of 18 cases with bimanual 23-gauge PPV done with only vital dyes were assigned to Group B “bimanual group.”

The mean age of the cases was 50.5 (ranging from 29 to 69 y) old in group A and 47.9 (ranging from 33 to 60) years old in group B (p value = 0.42). More females (25) than males (15) were assigned in our study.

Concerning the type and duration of DM, group A contained 6 patients with insulin dependent (IDDM), and 14 patients with non-insulin dependent (NIDDM). On the other hand, group B contained 18 and 2 patients with IDDM and NIDDM, respectively (p value > 0.05), with a mean duration of DM :14 y in group A and 15 y in group B (p value = 0.57).

Both groups were compared regarding intraoperative feasibility, ability to completely remove pre retinal membranes, operative time of PPV, and complications (intraoperative bleeding and iatrogenic retinal tears).

The results were tabulated and were subjected to statistical analysis.

4.1. Surgical procedure

Both groups were compared as regards the time of surgery, intraoperative bleeding, intraoperative breaks.

As regards the time of membrane removal, statistically significant differences were observed between both groups, with a shorter mean time of membrane removal (13.35 ± 1.79) minute (MIN)in group A while (14.55 ± 1.36) MIN in group B. The mean times of the surgery were (31.80 ± 4.32) MIN and (35.50 ± 5.36) MIN in groups A and B, respectively. Operative time showed statistically significant reduction in the trimanual group compared to bimanual group (p value > 0.05), as shown in Table 1.

Table 1. Comparison between the two studied groups according to time of membrane removal and time of surgery.

	Group “A” (n = 20)	Group”B” (n = 20)	t	P
Time of membrane removal (minutes)
Min. – Max.	10.0–16.0	13.0–17.0	2.394*	0.022*
Mean ± SD.	13.35 ± 1.79	14.55 ± 1.36
Median (IQR)	14.0(12.50–14.50)	14.50 (13.0–15.50)
Time of surgery
Min. – Max.	25.0–43.0	30.0–45.0	2.404*	0.021*
Mean ± SD.	31.80 ± 4.32	35.50 ± 5.36
Median (IQR)	30.0(30.0–34.50)	35.0(30.0–40.0)

IQR: Inter quartile rangeSD: Standard deviation t: Student t-test.

p: p value for comparing between Group “A” and Group “B”.

Intraoperative bleeding was assessed in both groups, with no bleeding in 13 eyes in the 1st group and 9 eyes in the 2nd group. Grade I and II bleeding was detected in 7 eyes in the 1st group and 11 eyes in the 2nd group, with no statistically significant differences between both groups. Demonstrated results are shown in Table 2.

Table 2. Comparison between the two studied groups according to intraoperative bleeding.

Bleeding	Group” A” (n = 20)		Group”B” (n = 20)		χ^2	P
	No.	%	No.	%
No	13	65.0	9	45.0	1.616	0.204
Grade I to Grade II	7	35.0	11	55.0

χ²: Chi square test.

P: p value for comparing between Group A and Group B.

As regards iatrogenic breaks, there were no breaks in 11 eyes, while 9 eyes had one or more breaks in group A. No statistically significant differences were noted in group B, where 4 eyes had no breaks, and 16 eyes had one or more breaks. The data are shown in (Figure 3).

Figure 3. Comparison between the two studied groups according to iatrogenic breaks.

Regarding the kind of tamponade used after fluid air exchange, gas tamponade (SF6) and Air were used more in group A, while silicone (Silicone mate” Polydimetilsiloxane”1300 Centistoke. GERMANY) was used more in group B (p value < 0.05).

Patients with silicone oil were arranged for another intervention for removal of silicone oil within [3.88 ± 0.64] months in group A and [4.08 ± 0.95] in group B (p value < 0.05).

Patients were examined at postoperative 1st day, 1st week, 1st month and at the 6th months following vitrectomy. During all visits, BCVA & IOP measurement were done, early vitreous hemorrhage, and early inflammation were recorded as well.

All cases included in the study achieved totally attached retina, and none of them needed a second intervention for post-operative complications.

4.2. Postoperative complications

All post-operative complications were medically treated without the need of re-intervention. Early Grade I to Grade II vitreous hemorrhages were noticed in 6 eyes (1st Group) and 7 eyes (2nd Group). Mild inflammation was noticed in 5 eyes in the 1st group and 6 eyes in the 2nd group. Increased IOP was found in 6 eyes (in the 1st group) and 4 eyes (in the 2nd group).

Statistical analysis was done with no statistically significant differences between both groups as analyzed in Table 3.

Table 3. Comparison between the two studied groups according to post-operative complication.

Post-operative complication	Group”A” (n = 20)		Group “B” (n = 20)		χ^2	P
	No.	%	No.	%
Early vitreous hemorrhage	(n = 6)		(n = 7)
Grade I	6	100.0	6	85.7	0.929	^FEp= 1.000
Grade II	0	0.0	1	14.3
Inflammation
No	15	75.0	14	70.0	0.125	0.723
Grade I	5	25.0	6	30.0
Increased IOP
No	14	70.0	16	80.0	0.533	0.465
Controlled Medically	6	30.0	4	20.0

χ²: Chi square test FE: Fisher Exact.

p: p value for comparing between Group “A” and Group “B”.

4.3. Visual acuity

Visual acuity (VA) was measured pre-and post-operatively following complete resolution of gas and removal of silicone in silicone-filled eyes.

The pre- and post-operative VA outcomes in group A were more or less similar to group B. In both groups, there was a statistically significant improvement in VA after surgeries. The mean pre-operative VA in group A was 1.75 (Log MAR) and 1.59 in group B, while the mean final visual outcome six months after surgery was 0.73 in the 1st group and 0.82 in in the 2nd group. There was non- statistically significant difference between initial visual acuities and final visual outcomes in both groups. Statistical analysis of visual acuity is demonstrated in Table 4.

Table 4. Comparison between the different studied periods according to BCVA in each group.

Distance Snellen VA	Preoperative	PO Day 1	PO 1 week	PO 1 month	PO 6 months	Fr	p
Group “A” (n = 20)
Min. – Max.	0.69–2.30	1.0–2.30	0.12–1.85	0.12–1.85	0.12–1.85	55.661*	<0.001*
Mean ± SD.	1.75 ± 0.44	1.95 ± 0.49	0.99 ± 0.42	0.89 ± 0.45	0.73 ± 0.43
Median (IQR)	1.81 (1.77–1.85)	2.30 (1.62–2.3)	1.0 (0.69–1.3)	0.85 (0.69–1.09)	0.69 (0.47–1.0)
p0		0.881	0.001^*	<0.001^*	<0.001^*
Group”B” (n = 20)
Min. – Max.	0.69–2.30	1.0–2.30	0.69–1.85	0.30–1.40	0.39–1.30	61.815*	<0.001*
Mean ± SD.	1.59 ± 0.53	1.81 ± 0.46	1.18 ± 0.34	0.98 ± 0.37	0.82 ± 0.34
Median (IQR)	1.77 (1.3–1.85)	1.77 (1.39–2.3)	1.30 (1.0–1.4)	1.0 (0.69–1.3)	0.69 (0.47–1.12)
p0		0.803	0.002^*	<0.001^*	<0.001^*
p1	0.265	0.369	0.127	0.327	0.314

IQR: Inter quartile range SD: Standard deviation.

Fr: Friedman test, Sig. bet. periods was done using Post Hoc Test (Dunn’s).

PO: postoperative.

p: p value for comparing between the different studied periods in each group.

p₀: p value for comparing between Preoperative and each other periods.

p₁: p value for Mann Whitney test for comparing between Group “A” and Group “B”.

*Statistically significant at p ≤ 0.05.

5. Discussion

This research aims to assess the trimanual PPV, a technique that enables the combination of vital dyes, self-retained chandelier lighting, and PFCL to facilitate the management of the most challenging patients of diabetic vitrectomy. In all of medical literature, the term (trimanual PPV) was documented only by Moon et al. [24] who mentioned the diabetic TRD 23-gauge sutureless PPV aided via using an auxiliary – controlled light probe illumination along with wide-angle viewing system.

The decreased operating time (31.80 ± 4.32 min), which was statistically significant with a p value less than 0.05, was one of the main advantages of the trimanual PPV that at least improved the non – inferiority of trimanual PPV in comparison to bimanual PPV, decrease the risk of phototoxicity of the light source to the macula and increased the turnover in the operating theater. Furthermore, the occurrence of iatrogenic RBs produced, silicon oil tamponade was less in the eyes managed by tri-manual vitrectomy (however, it wasn’t of statistical significance). It is suggested that the good bimanual handling of glial tissues in addition to the opposing traction via PFCL bubble and the better detection of the dissection planes in association with the vital dyes led to such differences. Both groups achieved statistically significantly better vision following six months of follow up. In none of our cases, PFCL passed deep to the underneath the retina. It might be because of the cautious injection and elimination of the PFCL bubble in addition to the good determination of the planes to be dissected owing to the auxiliary usage of the vital dyes.

Consistent with previous reports, iatrogenic retinal breaks(RBs) is an essential complication of PDR cases intraoperatively, it may occur through various mechanisms including that occurs during dissecting the proliferative fibro-vascular membranes or the induction of posterior vitreous detachment (PVD) [25,26]. the current work, comparing the incidence of iatrogenic RBs in trimanual vitrectomy Vs. bimanual vitrectomy for TRD was carried out. The results indicated that trimanual group was accompanied by fewer cases suffering from intraoperative iatrogenic RBs (45%) compared to bimanual group (70%) during PPV.

However, the rate of iatrogenic breaks was more in this work rather than other studies [24,27]. it might be explained by the patients managed in our study were the very complex cases of DM vitrectomy.

Imamura et al. [28] demonstrated the usage of PFC on 18 eyes with advanced PDR with TRD and & or rhegmatogenous retinal detachment (RD), similar to our study PFCL help easy flattening of the folds of the retina that helped successful membrane dissection and efficient end photocoagulation. The rates of improvement in the vision (VA of 20/200 or better in 9 eyes (50%)) was the same as the current study (mean VA 6 months’ post-operative in trimanual group “0.73 ± 0.43”). Considering postoperative complications there’s one eye with RD and another eye with subretinal PFC. Compared to our study, our results are better than the results demonstrated in this study with none of our cases having RD or subretinal PFC. Both studies concluded that PFCL is actually a helpful auxiliary during vitrectomy for treating excessive PDR.

In addition, Wafapoor et al. [29] conducted a study on 40 cases with PDR and RD. PFC liquid was utilized as an adjunct to PPV and membranectomy. Preoperative diagnosis included combined traction and rhegmatogenous RD in 23 eyes (57.5%), TRD in 13 eyes (32.5%), and recurrent rhegmatogenous RD in 4 eyes. VH was detected in 17 eyes (42.5%). PFC was mainly utilized to flatten the retina along with managing intraoperative complications like iatrogenic tears. The retina flattened during the operation in all cases, helping the administration of laser photocoagulation. Cases were followed up for at least 6 months. On termination of follow-up, the macula was still attached in 37 eyes (92.5%), including 31 (77.5%) in which the retina was fully attached. RD was in 3 eyes (7.5%). VA showed postoperative improvement in 20 cases (50%), was unchanged in 13 cases (32.5%), and worsened in seven cases (17.5%). Considering these results and comparing them to our current study, our study on 20 eyes [group A] with TRD in 16 eyes, combined tractional and rhegmategenous RD in one eye and subhyaloid hemorrhage was present in 3 eyes. PFC was utilized as a 3rd hand to remove epiretinal membranes and flatten the retina intraoperatively. In contrast, in our study, all patients achieved totally attached retina at the end of follow- up periods, with statistically significantly improved VA where the mean preoperative VA (1.75 ± 0.44) in comparison to the mean postoperative VA (0.73 ± 0.43).

The novel trimanual PPV procedure has many benefits: 1- the PFCL offers a counter-power on the hind pole that serves as a surgeon (3rd hand), hence permitting good retinal stabilization when manipulating the tissue bimanually. 2- it helps to identify the sites of tractions and remaining proliferations enabling better elimination of the pre-retinal proliferations with subsequently decreased risks of VH after surgery. 3- the PFCL might offer good intraocular hemostasis, visualization, and laser photocoagulation of the retina. 4- good achievement of peripheral vitreous shaving with a stabilized retina via PFCL. And lastly, the auxiliary usage of the vital dyes helps good determination of the proliferation surrounding the retina and ILM that reduces the risks of retinal break.

Limitations of this current study that may be overcome in the future by including larger number of cases and longer duration of follow up periods. Besides, the study included severely complicated patients of diabetic PPV, and the outcomes couldn’t be suitable for less complicated patients. Furthermore, one of the limitations that it didn’t include groups for DM vitrectomy assisted with the PFCL only as encouraging outcomes were actually published in other studies [28,30,31]. Nevertheless, utilization of vital dyes for staining DM pre-retinal proliferation underneath the PFCL was not documented before. Despite the staining of the fibro-vascular proliferation already helped the surgeons in good determination and dissection of fibro vascular proliferation, future studies should assess the efficacy of vital dyes in diabetes vitrectomy performed underneath the PFCL. We anticipate that this research will provide light on the trimanual PPV’s potential as a surgical approach that might improve surgical results in difficult diabetic vitrectomy situations. Finally, we are hoping that our research will contribute to bringing attention to the outcomes of these procedure.

Informed consent

It was taken from all patients participated in the current study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Mariam Magdy Ghobrial

Mariam Magdy Ghobrial Department of Ophthalmology, Faculty of Medicine, Alexandria University. MD Email: [email protected]

Faten Ali Hussin

Prof. Dr. Faten Ali Hussin Department of Ophthalmology, Faculty of Medicine, Alexandria University. Professor of ophthalmology, MD.PHD. E-mail: fatenhussein99@gmail. Com

Samir Mohammed El Baha

Prof. Dr. Samir Mohammed El Baha Department of Ophthalmology, Faculty of Medicine, Alexandria University. Professor of Ophthalmology, MD.PHD. Email: [email protected]

Mohamed Hassan Said

Dr. Mohamed Hassan Said Department of Ophthalmology, Faculty of Medicine, Alexandria University. Lecturer of Ophthalmology, MD.PHD. Email: [email protected]