Br J Ophthalmol 96:193-196 doi:10.1136/bjo.2010.200766
  • Clinical science
  • Original article

A novel technique of tangential, circumferential, scleral tunnel in 20-gauge transconjunctival sutureless vitrectomy: optical coherence tomography-aided analysis of wound integrity and clinical outcome

  1. Zachariah R Koshy2
  1. 1Tennent Institute of Ophthalmology, Gartnavel General Hospital, Glasgow, UK
  2. 2Department of Ophthalmology, Ayr Hospital, Ayr, UK
  1. Correspondence to Zachariah R Koshy, Department of Ophthalmology, Ayr Hospital, Ayr KA6 6DX, UK; zkoshy{at}
  • Accepted 13 April 2011
  • Published Online First 7 June 2011


Aim To describe a novel technique of sclerotomy construction to facilitate 20-gauge transconjunctical sutureless vitrectomy (TSV) along with the evaluation of the wound integrity.

Methods The surgical technique is described. One hundred consecutive patients who underwent TSV were evaluated for wound leaks, postoperative hypotony, endophthalmitis and any other complication related to surgery. The sclerotomies of eight patients (24 ports) were analysed by imaging with anterior segment optical coherence tomography immediately after surgery, and on the first postoperative day and after the first postoperative month.

Results 104 eyes of 100 patients were evaluated with a mean follow-up of 9.6 months. All cases underwent surgery with standard 20-gauge instrumentation and vitrectomy techniques. Five sclerotomies were found to leak at the end of surgery, requiring a suture. Anterior segment optical coherence tomography images were obtained from eight eyes with good apposition of the tunnel noted in all the cases. Mean intraocular pressure was 18.7 mm Hg on the first postoperative day. One patient had hypotony without leak and this patient had pre-existing hypotony due to chronic panuveitis. There was no instance of postoperative endophthalmitis.

Conclusions This technique of 20-gauge TSV achieves good wound apposition with a low incidence of complications while using standard 20-gauge instrumentation and vitrectomy techniques.


Sutureless sclerotomies provide a number of advantages over sutured sclerotomies for pars plana vitrectomy, including less postoperative inflammation, faster wound healing, shorter duration of surgery and improved patient comfort.1–5 Sutureless self-sealing sclerotomies for 20-gauge instruments in pars plana vitrectomy were first described by Chen in 1996.3 Several modifications of this technique have ensued,2 4 6–8 with some authors reporting low complication rates.

Fujii et al introduced a 25-gauge transconjunctival sutureless vitrectomy (TSV) system in 2002.5 However, problems relating to the flexibility and fragility of the new set of instruments, the low infusion and aspiration rates, and the increased risk of endophthalmitis have limited its use.1 9 10 In 2005 a 23-gauge TSV system was introduced that offers the advantages of the minimally invasive TSV system and the benefits of a sturdier and larger instrumentarium, though hypotony still remains a concern.11 These systems use a trocar and cannula for the ports to allow easy passage of instruments and to prevent conjunctival swelling. This necessitates a unique set of instrumentation specific to the size of the trocar.

Benefits of 20-gauge TSV include the ability to use conventional vitreoretinal instruments and standard flow rates, decreased cost compared with the smaller gauge systems, as well as the ability to perform standard vitrectomy techniques including close peripheral vitreous shave.

The primary concern with TSV is wound integrity and the incidence of postoperative hypotony, which can potentially risk endophthalmitis, choroidal effusion and maculopathy.

We describe a novel technique of sclerotomy construction for 20-gauge TSV and evaluate its integrity.

Methods and materials

One hundred and seven consecutive patients who underwent the modified 20-gauge TSV were included. We excluded seven patients who required silicone oil as a tamponade agent as these sclerotomies were sutured. Two patients had a prior history of conventional 20-gauge sutured vitrectomy.

A single, right-handed surgeon (ZK) performed all surgeries at a single hospital between February 2007 and July 2009. The following preoperative information was obtained for each patient: age, sex, indication for surgery, preoperative visual acuity (VA) and intraocular pressure (IOP). Per-operative data recorded included number of sclerotomies performed with this technique, number of sclerotomies sutured at the end of surgery due to a perceived wound leak, and per-operative complications. Seventy-eight per cent of the patients underwent surgery under retrobulbar anaesthesia with a 5 ml mixture of 2% lidocaine and 0.75% bupivacaine, and the remaining under general anaesthesia. We recorded the following postoperative information: IOP at day 1, month 1 and month 3, the presence of any conjunctival bleb, complications related to the sclerotomies, other postoperative complications and final VA. Hypotony was defined as IOP <8 mm Hg. The sclerotomy wounds of eight consecutive patients (24 ports) were imaged with anterior segment optical coherence tomography (AS-OCT) (Visante Model 1000; Carl Zeiss Meditec, Dublin, California, USA) with an axial resolution of 18 μm, transverse resolution of 60 μm and scan speed of 2000 scans per second. By creating standard and high-resolution, cross-sectional and longitudinal images of the scleral tunnel immediately after surgery, on the first postoperative day and after the first postoperative month, we assessed the integrity of the sclerotomy sites.

Surgical technique

The 20-gauge TSV sclerotomies were fashioned in the following way. A circumcorneal, tangential, transconjunctival incision was performed at 3.5 mm from the limbus in the infero-temporal quadrant with a 20-gauge microvitreoretinal (MVR) blade. Once the MVR blade had travelled 3 mm into the sclera, it was then rotated to enter the vitreous cavity perpendicular to the scleral surface (figures 1 and 2). A 4 mm infusion cannula was inserted; the tubing was taped to the cheek and the intraocular positioning of the tip was confirmed. Supero-temporal and supero-nasal incisions were made in a similar manner, but the sclero-conjunctival end was widened while removing the MVR blade to create a funnel-shaped incision such that the internal aspect of the scleral tunnel maintained 20-gauge dimensions and the external aspect was broad enough to allow easy introduction and passage of the various 20-gauge instruments. The conjunctiva was stretched around the wound to allow easy reflux of fluid, thus preventing iatrogenic chemosis. A standard vitrectomy was performed with a high-speed cutter (2500 cuts per min) and any additional adjunctive procedures as required with standard 20-gauge instruments. A non-contact wide-angled viewing system was used. Scleral plugs were not required during the procedure due to the self-sealing nature of the sclerotomies. At the end of the procedure superior ports were closed simultaneously with pressure and massage using cotton-tipped buds, followed by closure of the infusion port in a similar manner as the infusion cannula was withdrawn. The simultaneous closure of the superior ports allows the IOP to build up and activate the valve mechanism of the scleral tunnel to effect a seal. Finally, subconjunctival injections of cefuroxime were given around the ports to create blebs that overhung the sclerotomies. The eye was patched and taped overnight.

Figure 1

Illustration of incision technique (on cross section).

Figure 2

Illustration of the incision technique.


One hundred and four eyes of 100 patients were included and the mean follow-up was 9.6 (range 3–28) months. Fifty-eight patients were men and 42 were women; and the mean age was 64 (range 30–78) years. General anaesthesia was used in 22 patients and 78 patients had local anaesthesia in the retrobulbar space.

The most common indications for surgery were proliferative diabetic retinopathy including non-resolving vitreous haemorrhage in 25 eyes (24%), rhegmatogenous retinal detachment (RRD) in 24 eyes (23.1%), epiretinal membrane in 19 eyes (18.2%) and macular hole in 17 eyes (16.3%). Other indications are given in table 1. The preoperative VA was equal to or better than 20/40 in 14 patients (13.4%) and worse than 20/200 in 42 patients (40.3%).

Table 1

Indications for vitreo-retinal procedures in which 20-guage transconjunctical sutureless vitrectomy (TSV) was performed

Two hundred and eight superior sclerotomies and 104 infusion line sclerotomies were made with this technique. Fifty-three patients had combined cataract extraction and 20-gauge TSV. In addition to the posterior vitrectomy, 40 patients (38.5%) had C2F6 gas (concentration of 14%) and 17 patients (16.3%) had SF6 gas (concentration of 20%) injected in the vitreous cavity at the end of surgery.

One patient developed per-operative suprachoroidal haemorrhage and had suturing of all three sclerotomies. Of the remaining 309 sclerotomies, five (1.6%) were leaking at the end of surgery and required a single 8-0 vicryl suture across the external scleral lips of the sclerotomy (four of these had gas). Eight entry site breaks (2.5%) were discovered at the end of surgery and treated with cryotherapy and gas injection.

AS-OCT images were obtained from eight eyes (24 ports) immediately after surgery and good apposition of the tunnel was noted in all the cases (figure 3). Conjunctival cover of the entry site was noted from the first postoperative day in all cases. Immediate postoperative images showed a slender cyclodialysis cleft in the superior sclerotomy on the right side of four eyes (none of them developed hypotony) that resolved completely by the first month. A small ridge of scleral thickening was noted at the entry site in the images after a month.

Figure 3

Optical coherence tomography (OCT) image of entry site immediately following surgery (A) and after 1 month (B).

Mean IOP was 18.7 (range 4–45) mm Hg on the first postoperative day and 17.3 mm Hg after 1 month. Seven patients (four eyes received long-acting gas) had IOP >30 mm Hg on the first postoperative day requiring anti-glaucoma management in the short term; the IOP was within normal limits at 1 month without any treatment. Only one patient had hypotony without leak and this patient had pre-existing hypotony due to chronic panuveitis. No cases of subconjunctival gas or fluid leaks were recorded, and none of the patients had a choroidal detachment. Postoperative images of the sclerotomies on day 1 and month 1 are shown (figure 4).

Figure 4

Photograph of entry site on the first postoperative day (A) and the first postoperative month (B).

There was no instance of postoperative endophthalmitis. Other postoperative complications included worsening of cataract in six of 29 phakic eyes (20.6%), vitreous cavity haemorrhage in two eyes (1.9%) and RRD in seven eyes (6.7%).

All 304 self-sealing sclerotomies healed well postoperatively. All cases injected with gas maintained good intraocular tamponade in the postoperative period. No cases of late mechanical and tractional complications related to the sclerotomies were recorded.

The best corrected VA at the last follow-up was equal to or better than 20/40 in 47 patients (45.1%), 45 patients (43.3%) had VA between 20/50 and 20/200, and VA was worse than 20/200 in 12 patients (11.5%).


We describe a novel technique of a tangential, circumferential, scleral tunnel in 20-gauge TSV. The novel features are the transconjunctival, funnel-shaped, circumferential wound construction, the minimal recession of the conjunctiva, anchoring of a standard 4 mm infusion cannula, simultaneous closure of the superior ports to allow the IOP to effect a seal and the conjunctival blebs overhanging the entry sites. This modified technique shortens operative time, demonstrates less postoperative inflammation and hastens wound healing. The funnel-shaped superior scelerotomies allow easy utilisation of existing 20-gauge instrumentation and all the advantages they confer. In certain surgical situations such as complex retinal detachments with proliferative vitreo-retinopathy, removal of retained lens fragment or silicone oil; this versatility may make 20-gauge TSV preferable. The use of 20-gauge instrumentation is more cost-effective and the fluidics of a 20-gauge system allows a more rapid vitrectomy.

Optical coherence tomography (OCT) is a non-contact, non-invasive, high-speed, high-resolution imaging modality. The AS-OCT is an evolution of the retinal OCT. It uses a longer wavelength (1310 nm) than the retinal OCT (820 nm). This allows greater penetration through tissues that scatter light highly, such as the sclera and limbus, and can therefore acquire and analyse cross-sectional tomograms of the anterior segment.12 We are unaware of prior in vivo human investigations using AS-OCT to evaluate sclerotomy wounds and their anatomical structure in 20 g TSV in the immediate postoperative period. In addition to the clinical evidence of wound closure and lack of hypotony, the OCT images confirmed complete closure of the scleral tunnel.

Three hundred and four of our 309 sclerotomies self-sealed without any evidence of vitreous or retinal incarceration and any wound gaping or leaking at the end of surgery. We encountered wound leak in the five superior sclerotomies (1.6%) that required suturing. However, our incidence of wound leakage is less than that reported by Saad and Assi (5.5%)6 and Theelan et al (5%).8 The construction of the sclerotomies facilitated a valve-like mechanism, which obviated the need for scleral plugs and allowed secure seals at the end of the procedure.

Postoperative leakage and endophthalmitis10 13 also occur with 25-gauge TSV, with up to 7.1% of cases requiring a suture13 and up to 17.1% of cases demonstrating hypotony on the first postoperative day. In our study, only one patient (0.9%) had hypotony (IOP <8 mm Hg) on day 1 and the patient had pre-existing hypotony due to chronic panuveitis. Even if transient, hypotony may increase the risks of serious postoperative complications, such as endophthalmitis, retinal or vitreous incarceration, and suprachoroidal haemorrhage. The lack of postoperative hypotony in our series addresses this risk. The circumferential approach allows the incision track to remain within the pars plana as well as reducing the risk of inadvertent damage to the lens or retina. This also results in a radial cut through the superficial scleral fibres, which we believe will have a minimal astigmatic effect, although this was not measured in this study. The creation of conjunctival blebs near the entry sites by subconjunctival injections of antibiotics followed by patching of the eye appears to provide conjunctival cover of these sites. There was no evidence of sequelae from epithelial ingrowth.

Seven of our patients including four eyes with long-acting gas tamponade had IOP >30 mm Hg at day 1 that required anti-glaucoma management. None of the patients required long-term anti-glaucoma treatment.

Eight entry site breaks (2.5%) occurred in our patients that necessitated per-operative cryotherapy and internal gas tamponade. None of these patients developed RRD later. This incidence is lower than that described in the literature for a traditional stab incision sclerotomy.14 The seven eyes that developed postoperative retinal detachments did not have entry site breaks. The cyclodialysis cleft noted in four eyes on the right-sided sclerotomies is probably because of the right-handed surgeon's increased manipulation on that side. These had disappeared in all four cases by the 1-month examination.

Our technique aims to create a sclerotomy that achieves a good seal: this is demonstrated by the above results. The incision track created by a standard MVR blade has an elliptical cross-section that results in an effective valve mechanism when the IOP is allowed to rise. This technique also facilitates the smooth progression of surgery by allowing easy introduction of standard 20-gauge instruments due to the funnel-shaped incision, lack of conjunctival swelling and dispensing with the need for scleral plugs. The principle of wound closure is similar to that of small incision cataract surgery where the IOP is allowed to rise to create an effective seal of the valve-like incisions.

In conclusion, this modified technique of 20-gauge TSV is reliable and achieves good wound construction and apposition. It allows easy introduction and manoeuvring of standard 20-gauge vitreo-retinal instruments with a low incidence of complications.


  • Funding None.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.