Foldable phakic iris-claw lenses: results and complications during the surgeon´s learning curve.

In hands of experienced surgeons, the foldable Artiflex phakic iris-claw IOL showed to be predictable, effective and safe for the correction of high myopia in a prospective multicentre study conducted in 299 eyes and with follow-up visits for 2 years [10]. However, to the best of our knowledge, there is no information published regarding the outcomes and complications of this IOL during the surgeon´s learning curve. This study is our contribution to assess the impact of the surgeon learning curve in the results, complications and surgical trauma with this phakic IOL.

The refractive parameters and visual acuity improved significantly after the surgery (Table 1). The predictability analysis showed that after 3 months and 1 year, the SE was within 0.5 D in 86.36% and 80.00% of the cases respectively; and within 1.0 D of SE in 95.45% and 90.00% of the eyes respectively (Fig. 1). This result is similar to those reported by other authors. Dick HB et al.(10), reported in their multicentre prospective study 75.20% of eyes within 0.5D and 94.30% within 1.0 D. Pjano MA et al.[11], 90.00% of eyes within 1.0 D; Qasem Q et al. [12], reported 86.00% of eyes within 1.0 D at 1 year of follow up and Albarrán et al.[13], 89.10% of eyes within 0.5 D of SE. The median efficacy index in our study was 1. Again, this is comparable with the efficacy index reported by other authors [10,12,14]. Considering all these comparisons, we might conclude that the refractive results accuracy does not seem to be affected by the surgeon learning curve.

In terms of safety, the median logMAR BCVA improved significantly from 0 (p25: 0; p75: 0.02) to 0 (p25: 0; p75: 0) in the last visit (p=0.005). Indeed, 20.00% of eyes gained at least 1 line of vision whereas no eye lost lines of vision. In high myopia, it is well known that, the likelihood of gaining lines of vision is greater with phakic IOLs than with laser corneal procedures [13]. This is because of the magnification effect of the intraocular lens. However, other authors reported a higher proportion of eyes gaining lines of vision [10-13], being as high as 50.00% in the series published by Albarrán et al. [13]. One possible explanation for this could be that the vast majority of our patients had an excellent preoperative BCVA with almost no presence of amblyopic eyes in our series. In fact, all the eyes included in our series had a preoperative logMAR BCVA of 0.1 or even better, therefore, the chance of gaining lines of vision could be limited by this factor.

A median safety index of 1.02 (p25: 1; p75: 1.14) was obtained. A safety index over 1 is considered a good result as it indicates that there is some degree of improvement in the BCVA after the surgery. Dick HB et al. [10], reported a safety index of 1.09, 1.11, and 1.09 at 6 months, 1 year, and 2 years after surgery, respectively. Castro de Luna et al.[14], calculated a safety index of 1.1 and 1.02, 1 and 2 years postoperatively. According to those results regarding the surgeon´s learning curve, our safety index does not differ in a relevant way with those published in previous studies which were performed by experienced surgeons.  

Endothelial cell monitoring is one of the most important aspects when a phakic IOL is implanted as phakic IOLs could trigger an endothelial cell loss acceleration in some cases [5,6]. Endothelial cell loss is a common issue for all kinds of phakic IOLs and it might also occur in posterior chamber phakic lenses [15,16]. Indeed, in a long series in which the reasons for phakic IOL explantation were analysed, endothelial cell loss was the second most important reason only after cataract development [17]. Basically, there are two main mechanisms of endothelial cell damage in patients with phakic IOLs. First mechanism is the endothelial cell loss due to the surgical trauma and the second one is the long-term loss that can be higher than the physiological due to the IOL presence itself. Hence, it is advisable to perform an ECC analysis 3 months after the surgery to assess the endothelial damage due specifically to the surgery. In this investigation, this was one of the main outcomes measures as we hypothesised that, during the learning curve, the surgical trauma could be more intense leading to a higher endothelial cell loss in the short term. The median preoperative ECC was 2967 (p25: 2341; p75: 3296) cells/mm2 and decreased to 2924 (p25: 2670; p75: 3124) cells/mm2 at the 3 months follow up visit. Thus, there was a median loss of 43 cells (1.44%) due to the surgical trauma (P= 0.867). Other authors have previously reported endothelial cell loss results with this Artiflex foldable phakic iris-claw lens. We detected that the cell reduction was worse than the published by Dick HB et al. [10], who reported a mean cell loss of 0.05%, 6 months after the surgery. However, other investigators showed greater endothelial reductions; Pjano et al.[11] reported an endothelial cell loss of 6.10% 1 month after the surgery, Castro de Luna et al.[14] had a 8.34% reduction in the cell count 3 months postoperatively, Shajari and colleagues[18], detected a 4.50% cell loss 1 year after the surgery. Karimian et al.[19], reported an endothelial cell loss of 9.00% but after a mean follow up of 30±11 months. They did not provide data from earlier follow-up visits. Hashemi et al.[20], showed an endothelial cell loss of 3.04% 1 year after the IOL implantation. Finally, Jonker et al.[21], reported an endothelial cell loss 6 months after the implantation of artiflex myopia of 2.16%, similar to our findings. Hence, the endothelial cell loss due to the surgical trauma during the learning curve was acceptable and consistent with most of the previous published papers. The ECC reduction from the 3rd month to the last follow up visit (39th month), was 83 cells/mm2 (2.83%). The follow up time of our study was not long enough to assess the impact of this IOL in the endothelium survival in the long-term. However, this long-term loss should not be related to the surgical learning curve, and because of that, it was not a goal of this study.

IOL centration may be challenging in the first iris-claw implantation surgeries. In fact, in two eyes (9.09%) of our series (case 6th and 8th), the IOL was not properly centred (Fig. 2) leading to bad quality of vision and dysphotopsia. Dysphotopsia is related to bad IOL centration or to an insufficient optic diameter for the pupil in low mesopic or scotopic conditions. Those two eyes required a second surgery to improve the IOL centration which was finally achieved. Other authors did not report any incidence of early repositioning surgery to improve IOL centration [10-14]. However, Dick et al.[10], reported glare and halo in 3.40% and 3.00% of cases respectively. It was remarkable that those two eyes with poor IOL centration occurred within the first 8 cases with no more incidences after that. Therefore, achieving good IOL centration is, in our opinion, related to the surgeon´s learning curve. We advocate to check the IOL position at the slit lamp before the patient leaves the operating room. This should help to detect and resolve those cases with insufficient IOL centration before the patient is discharged the day of the surgery.

IOL inflammatory deposits (Fig. 3) were detected in 3 eyes (13.63%). In all the cases these deposits developed early, within 3 months postoperatively. These deposits were located in the IOL optic and a drop of at least 2 lines of vision was detected in all of them. Fortunately, the inflammatory deposits resolved completely after an intensive steroid drop course (Fig. 3). Even though inflammatory reaction has been described after Artisan phakic IOL implantation [22,23], it seems to be more relevant with Artiflex IOL. Dick et al.[10], reported pigment precipitates in 4.80% of the eyes and non-pigment precipitates in 1.40% of eyes. Pigment precipitates did not cause any loss of vision whereas non-pigment ones did. Visual acuity was restored after steroid treatment. Hashemi et al.[20], also detected deposits in 3 eyes in the early postoperative period which resolved with steroid drops. Similarly, Ruckhofer et al.[24], reported giant cell precipitates on the IOL surface in 12.00% of the eyes but these precipitates diminished over time without treatment. In a study performed in children (who are well known to be a population at high risk of postoperative inflammation) one eye out of 11 (9.09%) developed inflammatory deposits which were successfully treated with prednisolone drops[25].  Other investigators reported subclinical inflammation without proper formation of precipitates in 7.50% of cases [11]. In a study by Zuberbühler et al.[26], inflammation was reported in 5.80% of eyes. The reason why those precipitates developed may be due to multifactorial causes. A convex iris or too much iris enclavated, may cause and increased contact between the iris and the IOL breaking the blood-aqueous barrier [27,28]. Other authors suggested that the more demanding surgery of Artiflex compared with Artisan or the silicone material itself may play a role as well [10,11]. After having those cases with inflammatory precipitates, we decided to extend the topical steroid drops for 5 weeks (as described in the methods section) instead of only 3 weeks. Since then, we have not detected inflammatory precipitates again. Thus, an adequate iris configuration selection, avoiding excessive iris tissue enclavation and long enough steroid drops treatment are strategies that may prevent inflammatory precipitates to develop.

We did not find other complications previously described such as hyphema, synechiae, or cataract development in our series.

In summary, foldable phakic iris-claw IOL implantation was an effective, predictable and safe technique during the surgeon´s learning curve. The most common issue related to the learning curve was the correct IOL centration, which was found in 2 eyes, both within the first 8 implantations. As pupil is usually slightly nasally deviated, nasal enclavation is more challenging as there is less room and less iris tissue. We recommend starting by enclavating the nasal haptic as we have found that this helps to improve IOL centration. As above mentioned in the text, we also advocate to check the IOL position at the slit lamp in the operating room. This helps to detect and resolve those cases with insufficient IOL centration the same day of the surgery.