Miyake-Apple view of inner side of sclerotomy during microincision vitrectomy surgery

Abstract

Purpose: To examine the inner surface of the sclerotomy during microincision vitrectomy surgery by Miyake-Apple view. Methods: The anterior half of porcine eyes was attached to a transparent acrylic plate with cyanoacrylate glue. Then, either a 23-gauge or a 25-gauge trocar-cannula was inserted through the sclera obliquely. The inner surface of the entrance site was observed posteriorly by Miyake-Apple view. These images were compared with the endoscopic view of two patients who underwent vitreous surgery for an epiretinal membrane. Results: When the trocar-cannula was inserted obliquely, the Miyake-Apple view showed that the ciliary epithelium at the sclerotomy site was stretched. When the trocar-cannula was inserted vertically, the ciliary epithelium was folded, and the folds remained even after the trocar was removed. Vitreous strands were seen incarcerated into the sclerotomy site. In human eyes, a folding of the ciliary epithelium was not clearly seen with the endoscopic view but the incarcerated vitreous was seen. Conclusion: The Miyake-Apple view provided a precise, in vivo, observation of the inner surface of the entry site. It disclosed the morphological stress on the ciliary epithelium by the sclerotomy. Vitrectomy with 23-gauge and 25-gauge instruments, termed microincision vitrectomy surgery (MIVS), was developed to permit transconjunctival sutureless vitrectomy (Fujii et al. 2002; Eckardt 2005). The vitreous incarcerated in the wound after the surgery was believed to plug the sclerotomy, i.e., a self-sealing wound. However, postoperative complications such as hypotony, choroidal detachments and endophthalmitis (Ibarra et al. 2005; Lopez-Guajardo et al. 2006; Kunimoto & Kaiser 2007, Acar et al. 2008; Scott et al. 2008) have been reported after MIVS. Because of these complications, the conventional vertical insertion of the trocar was changed to an oblique entrance to try to reduce the incidence of hypotony and wound leakage after MIVS (Eckardt 2005; Shimada et al. 2006; Inoue et al. 2007; Taban et al. 2008; Woo et al. 2009). In 1985, Miyake & Miyake (1985) documented the effects of haptic fixation of implanted intraocular lenses into enucleated human eyes by posterior cinematography to allow a simultaneous view of the anterior and posterior segments. In 1990, the method was modified by Apple et al. (1990) to allow real-time viewing at high resolution with enhanced magnification. The Miyake-Apple posterior views provided a unique view of the structures in the anterior segment of the eye, which allowed an evaluation of the movement of the lens capsule, the stress on the zonular fibres, the inner surface of the iris, the ciliary processes and the anterior vitreous surface (Pereira et al. 2009). The enlargement of the lens capsular bag during phacoemulsification or capsular stretching during the intraocular lens insertion was clearly seen (Nishi et al. 2000; Auffarth et al. 2004a,b; Ahmed et al. 2005; Werner et al. 2005; Vasavada et al. 2008). The purpose of this study was to examine the morphological changes in the interior surface of the sclerotomy site during 23-gauge or a 25-gauge MIVS instruments and to evaluate the ability of the Miyake-Apple viewing system to study these changes. We shall show how the vitreous gel moved in the vitreous cavity around the cannula during vitreous surgery and how the vitreous was trapped in the inner side of the sclerotomy at the end of the surgery. These observations were compared with those obtained by an endoscope in human eyes. In the initial experiments, porcine eyes were used instead of post-mortem human eyes. The extraocular tissue around the isolated porcine eye was excised. The globe was then placed in a plastic mould where it was marked and bisected near the equator in the corneal plane. The anterior half of the eye including the attached vitreous body was fixed to a transparent acrylic plate with cyanoacrylate glue and placed under an operating microscope (Fig. 1). The plate was adjusted so that a good view of the inner half of the eye was obtained. The schema of the Miyake-Apple view. The anterior half of the eye is fixed to a transparent acrylic plate. For posterior view, a separate microscope connected to a video camera was mounted under the table, inverted and directed upward. The illumination for posterior view came from flexible fibre optic cables in three directions delivered from separate light sources below the table. An operating microscope was used to view the anterior segment as in conventional ophthalmic surgery. For the posterior view, another microscope was mounted under the table, inverted and directed upward. A video camera was mounted to this microscope, and the posterior half of the eye was viewed on a video monitor. The illumination for the anterior (surgeon’s) view was turned off and that for posterior (camera) view came from separate light sources below the table (Fig. 1). Three flexible fibre optic cables delivered the light from the sources, and the anterior segment was viewed by retroillumination. A 23-gauge or 25-gauge trocar-cannula (Alcon lab, Fort Worth, TX, USA) was inserted obliquely through the delaminated sclera. The lengths of intravitreal part of the 25-gauge and 23-gauge cannula were both 4 mm. The entrance of the trocar-cannula was observed with the operating microscopic from above and with the camera-monitor from inside the eye by the Miyake-Apple view. The 23-gauge vitreous cutter and instruments were inserted into the vitreous cavity through the 23-gauge cannula and the 25-gauge instruments through the 25-gauge cannula. The posterior view was observed with the Miyake-Apple view. To make the transparent vitreous more visible, triamcinolone crystals (Kenacort-A®, 40 mg/ml; Bristol Myers KK, Tokyo, Japan) were injected intravitreally, and the movements of the vitreous was observed while the vitreous cutter was used. During these procedures, the intraocular pressure was maintained by frequent injections of intravitreal balanced salt solution. For comparison, a 23-gauge endoscopic system was used through the 23-gauge cannula to observe the inner site of the sclerotomy during the oblique insertion of a 25-gauge trocar-cannula and also when the cannula was removed at the end of the surgery. This was carried out in two eyes that underwent vitrectomy for an idiopathic epiretinal membrane after a written informed consent had been obtained. The surgical procedures included core vitrectomy, peripheral vitrectomy and removal of the epiretinal membrane. Peripheral vitrectomy with scleral indentation was not performed. A 23-gauge endoscope system was used with a solid fiber catheter (AS-611; FiberTech Co., Ltd, Tokyo, Japan) and a 3CCD (charge-coupled device) imaging system (FT-203F; FiberTech Co., Ltd). The findings in both the pig and human eyes are presented in Table 1. When the trocar-cannula was inserted obliquely, the ciliary epithelium at the site of the sclerotomy was stretched on the proximal side (direction of insertion) of the entry site (Fig. 2A). With a vertical insertion of the trocar-cannula, the stretched ciliary epithelium was folded at the entry site (Fig. 2B). After the insertion of the cannula and the surgical instruments, the folds of the ciliary epithelium remained on the proximal side around the cannula (2, 3). In addition, vitreous strands were seen incarcerated from the anterior edge of the vitreous base into the entry site of the cannula towards the distal side of the cannula where the entry site in the ciliary epithelium was enlarged (Fig. 3B). Miyake-Apple view for 23-gauge trocar-cannula. (A) Ciliary epithelium at the site of inner sclerotomy is stretched (white arrow) to the distal side of entry site when the 23-gauge trocar is inserted obliquely. (B) Ciliary epithelium is folded (white arrows) when the direction of inserting trocar is vertical. (C) After placement of cannula, the folds of ciliary epithelium are still present (white arrows). (D) Vitreous strands can be seen incarcerated into the gap between the cannula and the 23-gauge vitreous cutter (white arrowheads). (E) After the 23-gauge vitreous cutter is removed, vitreous strands (white arrowheads) are incarcerated into the cannula. (F) Incarceration of the vitreous (white arrowheads) into the defect of the ciliary epithelium (white arrow) is seen after the cannula is removed. Miyake-Apple view for 25-gauge trocar-cannula. After placement of 25-gauge cannula, the folds of ciliary epithelium (white arrow) are present when the surgical instruments have not been inserted (A) and inserted (B). The incarcerated vitreous strands (white arrowheads) from the anterior edge of the vitreous base to the entry site of the cannula are stretched towards the cannula. After removal of the 25-gauge cannula (C), the folds (white arrow) of ciliary epithelium decrease and the defects (black arrow) of the ciliary epithelium remain (black arrowheads: pieces of scleral tissue). The vitreous cutter and the forceps were inserted and moved forward and backward after it was seen in the vitreous after the injection of the triamcinolone crystals. The vitreous gel was trapped on he inside of the cannula. Vitreous strands were seen trapped in the opening port of the vitreous cutter and into the gap between the cannula and the cutter shaft (Fig. 2D) or the forceps. After the vitreous cutter was removed, a stream of the vitreous gel was observed to flow into the cannula (Fig. 2E). These observations were made even in hypotonic eyes without continuous replacement of intraocular irrigating fluid. When the cannula was removed, vitreous strands became incarcerated in the defect of the ciliary epithelium (Fig. 2F). When an endoscope was used, the oblique insertion of a 25-gauge trocar three millimetre posterior to the corneal limbus of human eyes was found to pass through the pars plana posterior to the pars plicata (Fig. 4A). The folds of ciliary epithelium that were seen in the porcine eyes with the Miyake-Apple view were not visible (Fig. 4B) possibly because of the lower resolution of the endoscope. Vitreous strand from the anterior side of the vitreous base was seen trapped in the cannula (Fig. 4C) even after the peripheral vitreous was removed. The vitreous was incarcerated in the defect of the ciliary epithelium after removal of the cannula at the end of the surgery (Fig. 4D). Endoscopic view of 25-gauge trocar-cannula in a human eye. (A) Trocar has been inserted into the eye through the pars plana (white arrow) behind the pars plicata from the endoscopic view. (B) The folds of the ciliary epithelium (white arrowheads) are not apparent possibly due to lower resolution of the endoscopic image. (C) The incarcerated vitreous strands from the anterior side of the vitreous base into the cannula can be seen. (D) Vitreous strands are incarcerated (white arrowheads) at the ciliary epithelium defect (white arrow) after removal of the cannula. The oblique insertion technique was developed to avoid postoperative hypotony, and the incidence of endophthalmitis was expected to be lower because of the self-closure of the outer side of the sclerotomy (Singh et al. 2008). A comparison of the oblique and vertical incisions by histology showed that the wound closure was better following an oblique insertion (Singh et al. 2008; Taban et al. 2008; Gupta et al. 2009). Nagpal et al. (2009) reported that endoscopy showed a large amount of vitreous blocking the inner lip of the sclerotomy ports after 23-gauge conventional vitrectomy for uncomplicated cases of vitreous haemorrhage because of diabetic retinopathy. The view from the vitreous side can be made by endoscopy or cross-sectional images of the inner sclerotomy by optical coherence tomography (Nagpal et al. 2009; Taban et al. 2009). However, the resolution of 20-gauge and 23-gauge intraocular endoscopes is limited because of the number of optical fibres that can be contained in a small diameter probe (10 000 optical fibres in 20-gauge endoscopic probe and 6000 fibres in 23-gauge probe). The advantage of Miyake-Apple view is the higher resolution of the images on the video camera-monitor attached to an operating microscope (Pereira et al. 2009). The disadvantage of the Miyake-Apple view is that the observation is on eyes that are hypotonic. Even with frequent irrigation, the water tightness of the dissected eye and cyanoacrylate glue was not completely sealed to prevent the leakage of the intraocular fluid. In addition, the video camera was located beneath the eye and leakage of the fluid can interfere with the image of the eye. This may be why we could detect a limited blockage by vitreous at the inner wound of the sclerotomy port in the Miyake-Apple view, although we observed massive vitreous blocking with the 23-gauge endoscope. Our findings suggested that the folds of the ciliary epithelium that were displaced to the distal side of the sclerotomy and the enlargement or defect of the ciliary epithelium may be the cause of the intraoperative complications including temporary hypotony and misdirection of intraocular irrigating fluid into the suprachoroidal space (Ooto et al. 2008). The porcine sclera is thicker than that of the human eyes which would induce a greater shift of the inner incision of the sclerotomy than that of the human eyes when the direction of the trocar-cannula was changed from oblique to the vertical directions. The enucleated porcine eye may have persistent hypotony before the experiment and subsequent ciliary epithelium detachment that should easily cause the stretching and folding of the ciliary epithelium when the trocar was inserted. In addition, the sclerotomy in porcine eyes was made at the pars plicata because the area of pars plana was narrower than that in human eyes. This may enhance the folds of ciliary epithelium that were observed more clearly in the porcine eyes. We conclude that the Miyake-Apple view is a useful method to evaluate the inner structures of the posterior segment of the eye, and it showed the morphological stress on the ciliary epithelium by the sclerotomy. The potential of damaging the ciliary epithelium at the inner side of sclerotomy should be considered when the sclerotomy is created with the oblique insertion. The corresponding author (MI) has full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.