A comparative study of transscleral sutured intraocular lens fixation and sutureless flanged intraocular lens fixation

In patients with insufficient capsular support, intraocular lens (IOL) can be implanted in the anterior chamber, fixed on the iris or fixed in the ciliary sulcus with transscleral sutures [1]. Since anterior chamber IOL can accelerate corneal endothelial cell loss and iris fixed IOL can cause recurrent iritis, transscleral sutured IOL fixation technique was preferred by many surgeons. However, suture-related complications remained a major concern, such as suture exposure and subluxation of the IOL after suture breakage. In a 6-year study involving 61 eyes, suture breakage was observed in 26.2% of cases after scleral fixation of IOLs using 10–0 polypropylene sutures [2].

In 2007 and 2008, Gabor and Agarwal reported the technique of sutureless intrascleral IOL fixation, which was modified by others in later studies [3‐7]. The basic principle of this technique is incarceration of the IOL haptics in a scleral tunnel parallel to the limbus. In 2017, Yamane et al. developed a technique named “flanged IOL fixation” [8]. This procedure does not require peritomy or scleral flap dissection. The IOL can be fixed firmly to the sclera with the enlarged tips of the haptics.

Previous studies have compared the IOL positions and the visual outcomes between sutured transscleral fixation and sutureless intrascelral fixation techniques, but there are limitations [9‐12]. First, most of these studies used ultrasound biomicroscopy (UBM) or Scheimpflug photography to measure IOL tilt and decentration, which had relatively low accuracy and reproducibility due to alignment issues or image quality. Second, the astigmatism measured by optometry comes from both the cornea and the IOL. The influence of IOL malposition on internal astigmatism could not be decided accurately. Third, higher-order aberrations (HOAs) was not included in the evaluation of visual outcomes. In this study, we aimed to compare the two IOL fixation techniques in a more detailed manner, using a second generation swept-source anterior segment optical coherence tomography (AS-OCT) to measure the IOL tilt and decentration and using the iTrace Visual Function Analyzer to measure internal astigmatism and HOAs.

The study consisted of 27 eyes (24 patients) from the transscleral sutured IOL fixation group and 26 eyes (26 patients) from the sutureless flanged IOL fixation group. The mean ages were 56.5 ± 10.1 years (range, 36–71 years) for the sutured group and 56.5 ± 13.9 years (range, 28–82 years) for the flanged IOL fixation group. The two groups did not differ significantly in age, sex, UDVA, CDVA, IOP, axial length or corneal endothelial cell density at baseline. There were 22 dislocated crystalline lenses, 22 dislocated posterior chamber IOLs and 9 aphakic eyes. The two groups did not differ significantly in the distributions of preoperative diagnosis (P = 0.461). The patient characteristics are shown in Table 1.

The CDVA in both groups improved significantly at 1, 3, 6 and 12 months postoperatively compared with baseline (P = 0.002, P < 0.001, P < 0.001 and P < 0.001 for sutured group, and P = 0.011, P = 0.005, P = 0.002 and P = 0.002 for sutureless flanged fixation group, respectively). However, there was no difference in CDVA between the two groups at all time points (P > 0.05).

There was no significant difference in astigmatism measured by subjective refraction between the two groups at 3 months postoperatively (P = 0.272). Refractive difference was defined as the difference between the spherical equivalent at 3 months postoperatively and the intended refractive error. Corneal endothelial cell loss was defined as the difference between the endothelial cell density 3 months postoperatively and at baseline. There was no significant difference between the two groups in refractive difference (P = 0.618) and corneal endothelial cell loss (P = 0.250) (Table 2).

There was no significant difference between the two groups in vertical, horizontal or 3D tilt from the corneal topographic axis (P > 0.05). The differences in vertical, horizontal or 3D decentration from the corneal topographic axis between the two groups were not significant either (P > 0.05) (Table 3).

For vertical decentration, 70.4% in the sutured group and 88.5% in the flanged IOL fixation group decentered downward, 11.1% in the sutured group and 11.5% in the flanged IOL fixation group decentered upward. For horizontal decentration, 51.9% in the sutured group and 38.5% in the flanged IOL fixation group decentered temporally, 29.6% in the sutured group and 19.2% in the flanged IOL fixation group decentered nasally.

There was no significant difference between the two groups in all internal optical aberrations, including total internal optical aberration, internal astigmatism, total internal HOA, internal coma, horizontal and vertical internal coma, internal trefoil and internal spherical aberration (P > 0.05). The two groups did not differ in average height of modulation transfer function (aMTF) or Strehl ratio (SR) either (P > 0.05) (Table 4).

Vertical IOL decentration significantly correlated with total internal optical aberration (r = 0.288, P = 0.036), total internal HOA (r = 0.440, P = 0.001), internal coma (r = 0.348, P = 0.001), vertical internal coma (r = 0.388, P = 0.004), aMTF (r = − 0.364, P = 0.007) and SR (r = − 0.297, P = 0.031). Horizontal IOL decentrations significantly correlated with horizontal internal coma (r = 0.312, P = 0.023). IOL tilt did not correlate with any internal HOA. IOL tilt or decentration did not correlate with spherical aberration (Table 5).

Table 6 shows the postoperative complications of the two groups. There was no difference between the two groups in early complications including hypotony (IOP ≤ 6 mmHg), vitreous hemorrhage or IOL capture. No late complication including cystoid macular edema, IOL dislocation or severe complications like endophthalmitis and retinal detachment happened during the follow-up time.

In eyes with insufficient capsular support, visual rehabilitation could be achieved with several IOL implantation techniques. Among them, transscleral sutured IOL fixation and intrascleral sutureless IOL fixation had been proved to be safe and effective [1, 15]. In recent years, intrascleral sutureless IOL fixation technique was favored by many surgeons because it was free of suture-related complications. Different modifications of intrascleral sutureless IOL fixation technique had been proposed since its advent in 2007 [3‐8]. The flanged IOL fixation technique developed by Yamane et al. was a sutureless technique with the advantages of transconjunctival approach, short operating time and firm incarceration of the IOL haptics in the sclera [8].

The methods to measure IOL tilt and decentration had evolved in the past few decades, including Purkinje image technique, Scheimpflug photography, UBM and AS-OCT [14, 16‐21]. None of the above methods could measure IOL tilt and decentration directly except for second generation AS-OCT (Casia2). Image processing software and manual measurement were necessary when using these methods. Moreover, pupillary axis was frequently used as the reference axis in the initial studies based on these methods [19]. The deviation of pupil center had a negative effect on the accuracy of measurement. Compared to the pupil center, corneal vertex was considered to be a better reference to assess IOL tilt and decentration because it was not affected by pupil shape [22]. The present study used Casia2 to measure IOL tilt and decentration. Casia2 was a second generation AS-OCT with deep scan depth, good penetration and higher resolution. It could automatically measure the tilt and decentration of IOL using the corneal topographic axis as the reference. It had been proved to have very high inter-observer and intra-observer reproducibility in the measurement of IOL tilt and decentration [22].

There was a paucity of studies comparing sutured and sutureless IOL fixation techniques [9‐12]. Marianelli et al. used UBM to compare the IOL tilt between sutureless and sutured scleral fixated IOLs. No significant difference was seen in the vertical tilt and horizontal tilt [10]. Later on, two studies had compared sutured IOL fixation technique and flanged IOL fixation technique. In Sül et al.’s study using a Scheimpflug camera, flanged IOL fixation group had lower IOL tilt in both meridians and lower lenticular astigmatism than the sutured group. The two groups did not differ in IOL decentrations [11]. In another study conducted by Do et al., however, the IOL tilt and decentrations in both meridians measured by Casia2 did not differ between the two groups. The postoperative visual acuity and refractive differences did not differ either [12]. Similar to Do et al’s study, the two groups in the present study did not differ in IOL tilt and decentrations in both meridians.

Good alignment of IOL is fundamental for optimal visual function. However, a significant amount of tilt and decentration can sometimes be induced after sutured or sutureless fixation of IOLs in the absence of capsular support. IOL tilt and decentration compromise the optical performance of eyes by inducing astigmatism or HOAs [23, 24]. To our knowledge, there was previously no study comparing the HOAs between transscleral sutured IOL fixation and intrascleral sutureless IOL fixation. The present study used iTrace Visual Function Analyzer to measure the internal astigmatism and HOAs and showed that the two groups did not differ in internal astigmatism or internal HOAs. The present study also investigated the relationship between IOL tilt and decentration and internal HOAs. Among sutured and sutureless fixed IOLs, vertical decentration was found to be correlated with total internal HOA, internal coma, vertical internal coma, internal aMTF and internal SR. Horizontal decentration only correlated with horizontal internal coma. Vertical and horizontal IOL tilt did not correlate with any internal HOA. In some other studies, however, IOL tilt instead of IOL decentration was often found to be correlated with coma-like aberrations [24‐26]. Same as in other studies, IOL tilt and decentration did not correlate with spherical aberration.

According to the present study, flanged IOL fixation and sutured IOL fixation had similar IOL alignment and visual outcomes. Since flanged IOL fixation could avoid suture-related complications and had shorter operating time [12], this technique could be a promising alternative to treat patients with insufficient capsular support. However, compared with sutured technique, it is more challenging to ensure good centration of IOL for flanged IOL fixation. First, the internal opening of the tunnels could not be visualized when penetrating the sclera with the needle. Second, once the haptics are externalized onto the sclera, it is almost impossible to adjust the position of the IOL. Inappropriate positioning of the insertion points and angles of the two needles may cause considerable amount of IOL decentration and tilt in flanged IOL fixation. To solve this problem, we used a corneal marker for radial keratotomy to help ensure correct positions of the insertion points of the needle. The surgeon’s experience was important to ensure correct angle of the needle and correct position of the internal opening of the tunnel. In 2019, a “needle stabilizer” was invented by Yamane et al., which was likely to be useful for surgeons with less experience in flanged IOL fixation [27]. For sutured IOL fixation technique, a corneal marker could help to reduce IOL decentration effectively.

When the results of our study are discussed, its limitations should be mentioned. First, the recruitment of the participants did not follow a random pattern, so that a confounding effect owing to a potential selection bias might have been possible. However, the two groups did not differ in baseline demographic and clinical characteristics. All of the vitrectomy was done using a 23G system and the IOL fixation techniques were not selected based on the status of the patients. Second, the follow-up time was only 12 months, so that longer-term complications could not be seen in the present study. Since breakage of 10–0 polypropylene sutures can happen 6 or more years after surgery, the advantage of flanged IOL fixation technique might have been underestimated in this study. In addition, the sample size was small. A larger sample size is needed to support our argument that the two procedures had similar visual outcomes and IOL positions. In a word, a large, randomized study with longer follow-up time is therefore necessary to confirm our results.

In conclusion, sutureless flanged IOL fixation had similar IOL position and visual outcomes compared with transscleral sutured IOL fixation within 1 year postoperatively. Vertical and horizontal IOL decentrations were associated with increased internal HOAs and thus should be avoided in particular.