Various intrascleral IOL fixation techniques have been developed in recent years. Although safety and good fixation of the IOL with these intrascleral fixation methods were reported, there is room for improvement in terms of simplifying complicated techniques and reducing the high levels of skill required. Totan and Karadag described a technique for making scleral tunnels prepared by insertion of 25-gauge transconjunctival sutureless vitrectomy microcannulas using trocars, and this technique provided good IOL stabilization with a shorter surgical time [14]. We believed that using 27-gauge trocars to form the scleral tunnels for intrascleral IOL fixation with the flange technique [7] might be safer and more easily performed due to the smaller surgical wound and direct approach to the IOL loop with the intraocular forceps. Similarly, Diamint and Giambruni reported a sutureless IOL scleral fixation technique using a 27-gauge trocar. That study found good postoperative visual acuity over a 12-month observation period [15].
In the series of 19 patients who underwent our novel intrascleral fixation method, we determined the detailed IOL fixation results using anterior segment-OCT. The mean preoperative logMAR UCVA in our patients improved 1 month postoperatively compared with baseline. To the best of our knowledge, this study was the first to measure intrascleral IOL fixation tilt and decentration with anterior segment-OCT. The tilt and decentration of the IOL were measured at the corneal topographic axis because the corneal vertex is not affected by the shape of the pupil, and therefore that axis is a better reference for the measurement than the pupil center.
In this study, the mean postoperative IOL tilt was 3.52 ± 3.00° and decentration was 0.39 ± 0.39 mm. Kimura et al. reported that the crystalline lens and IOL showed an average tilt of 4–6˚ toward the inferotemporal direction relative to the corneal topographic axis and an average decentration of less than 0.12 mm [16]. Therefore, our technique resulted in greater IOL tilt and decentration than occurs in normal cataract surgery. We could not compare our results with other reports of intrascleral IOL fixation because the IOL tilt and decentration were not determined in a similar manner. There was no significant difference between the mean number of corneal endothelial cells before and after surgery, although a declining trend was noted in this study. It was reported that 1 month postoperatively, corneal endothelial cell loss was 10.1% in eyes that underwent planned extracapsular cataract extraction with posterior chamber IOL implantation [17]. Yamane et al. found that the corneal endothelial cell density with the double-needle technique was 2341 ± 481 cells/mm2 preoperatively and 2313 ± 462 cells/mm2 at 6 months postoperatively, showing no significant decrease [7]. The difference between Yamane et al.’s [7] and our results could be related to the observation time point. In addition, the method used to grasp and pull out the IOL haptics directly might have affected the corneal endothelial cells, although no direct attachment occurred in our series of patients.
IOL tilt and decentration were not correlated with BCVA. Some groups reported that tilt and decentration after IOL implantation impaired visual quality and led to higher-order aberrations [18, 19]. However, it was also found that IOL decentration did not influence visual acuity in eyes with monofocal IOLs [20] and that in-the-bag IOL that maintained a decentration of <1 mm and an angle of <4° did not influence BCVA [21]. The present study also suggested that IOL tilt and decentration did not affect visual acuity. However, the correlation could not be confirmed because of the small number of eyes and the short observation period.
The prediction error of the target refraction did not show a myopic shift trend and postoperative refraction was not stable in this study. After transscleral fixation of the IOL in aphakic vitrectomized eyes, a –1.0 D myopic shift was seen 6 months postoperatively by Ahn et al. [13]. In the transscleral procedure reported, the IOL haptic is sutured 1.5 mm from the corneal limbus [13, 22], with scleral tunnels formed 2 mm from the limbus, and thus no myopic shift occurred. The needle insertion angle is also important in fixing the IOL [7]. The unstable postoperative refraction seen in our patients was likely due to the gap that occurred between the insertion angle and insertion position when forming scleral tunnels.
Vitreous hemorrhage and hypotony occurred more frequently among our patients than in previous studies of transscleral IOL suture and intrascleral IOL fixation [7–13, 23]. The inner diameter of the 27-gauge trocar is 0.4 mm (27 gauge), but the outer diameter is 0.5 mm (25 gauge). Previous studies used 9-0 polypropylene and 27- or 30-gauge needles, and thus the cause of hemorrhage is believed to be related to the surgical wound size. Our technique was based on the method using a 25-gauge transconjunctival sutureless vitrectomy trocar as reported by Totan and Karadag [14]. However, our postoperative complications could not be compared because they did not report them. We also performed surgery by referring to the handshake technique, which Agarwal et al. originally reported, grasping the IOL haptic with forceps and pulling it out of the eye [23]. It is essential to hold the haptic at the tip so that it does not snag on the sclerotomy wound during externalization. Because we did not note this point in some cases, further study is necessary to assess it.
Rizzo et al. reported a study showing that 27-gauge sclerotomies resulted in good wound closure with a 1-step insertion at an angle of 30° [24]. Suturing of the sclerotomy site was unnecessary, and postoperative hypotony did not occur [24]. Mitsui et al. reported that after 27-gauge vitrectomy for epiretinal membrane, the scleral wounds closed at 7.7 ± 4.7 weeks [25]. Recently, we have reported that 27-gauge trocar wound closing occurred by postoperative day 10 [26]. Therefore, we thought that IOL fixation in the sclera was completed within 1 month in our study. Since hypotony was observed in a few patients, pulling out the IOL haptics with a trocar might affect wound closure even with a 27-gauge system. We need further detailed analysis of wound closure using OCT to assess this point.
We created a flange to prevent the haptics of the IOL from detaching. The outer diameter of the 27-gauge trocar is 0.5 mm, which ensures that the flange is large enough to fix the haptic securely [7]. All IOLs used in this study were the NX-70 type, and the flanges were made by heating the haptics. Scleral tunnels were formed at the 4 o’clock and 10 o’clock positions to prevent damage to the long ciliary nerves at the 3 o’clock and 9 o’clock positions [11].
Recently, Diamint and Giannbruni have reported a similar surgical procedure and found that the technique is safe, minimally invasive and relatively easy to perform [15]. Our procedure is a modification of the handshake technique, which involves grasping the IOL haptic with forceps and pulling it out of the eye [23]. An advantage of the handshake technique is that it does not require the insertion of a haptic, which is the most difficult element of the double-needle technique, into the needle. The handshake technique can also be applied in numerous clinical situations. For example, it is possible to cope with the IOL fall reported as a complication during intrascleral fixation, since the IOL that has fallen into the vitreous can be pulled out by grasping and lifting the haptic using vitreous forceps. Also, in the case of the double-needle technique, the needle tip may reach the retina and ciliary body after the puncture, but with the present method, that risk is lower because the forceps are manipulated under direct vision.