Néstor I. Carreño

Retropupillary iris-claw intraocular lens in aphakic eyes.

Reply : I agree with Dr. Peposes suggestion that this was a study designed to look at the real world performance of accommodating IOLs available and approved for use at the time of the study. Although the refractive target for the Crystalens HD may have been amended to “select the first available plus for the dominant eye and the first available minus for the nondominant eye,” this was not the recommendation at the time of the study. The A constant had been adjusted by the manufacturers as the study was being planned, and the recommendation was to target C0.25 D. Dr. Pepose comments that “despite this 1.0 D advantage for the mini-monovision nonaccommodating monofocal and Tetraflex groups, the mean near visual acuity for the Crystalens HD was 1.5 lines (ie, 7.7 letters) better than the monofocal minimonovision and equivalent to the Tetraflex set for mini-monovision.” He fails to take into account that the CrystalensHDhas a bispheric designwith a central 1.50 mm zone that is 3 mm thicker. This modification of the optic provides an increaseddepth of focus of 1.50D, as stated in the paper. Thus, the Crystalens HD underperforms since it has a 0.50 D advantage, and the near visual performance should be that much better. Dr. Pepose is correct that this study was underpowered todetect a 7.5-letter difference innearvisual acuity, as it was designed to detect a 10-letter difference in acuity. Given the design of the study, no significant difference was found in the performance of the IOLs for near vision. It would be inappropriate to draw conclusions on a difference taken in isolation without considering the standard deviation and without increasing the size of cohorts to possibly detect the lower difference. I look forward to the publication of a controlled prospective comprehensive study incorporating a protocol that evaluates not only the visual performance but also the subjective assessment of accommodating IOLs.dGeorge Beiko, BM, BCh, FRCS

Aphakic retropupillary iris-claw intraocular lens (IOL) pseudophakic accommodation

Dear Editor, We read with great interest the article by Schopfer et al. showing the changes of the objective refraction related to the patients’ position [1]. In the “Abstract” the authors indicated: “...the phenomenon of pseudophakic accommodation is explained by pseudomyopia and pseudo-hyperopia...”. Actually, the pseudophakic accommodation phenomenon refers to myopia induced by the displacement of the IOL in response to contraction of the ciliary muscle, as the authors correctly stated later, in the “Discussion” section of their article [2, 3]. Thus, in these eyes with a retropupillary iris-claw IOL, where themovement of the IOL is not related to the ciliary muscle contraction, but to gravity, there is no such true pseudophakic accommodation. Moreover, the terms “pseudomyopia” and “pseudo-hyperopia” have no place in this definition. In “Results”, data indicating the percentage of eyes achieving a postoperative spherical equivalent refraction within 0.5 D (diopter) (8 %) and 1.0 D (12 %) of plano are confusing. According to the information provided, it seems that correct data are 26 % and 58 %. These percentages are significantly lower than benchmark standard of 55 % and 85 % that has been suggested for cataract surgery [4]. It would be important to know what A-constant the authors used in their biometric calculations. Authors indicated that the “mean amplitude of accommodation” was 4.96 D in backward position, 5.70 D in primary position and 5.18 D in forward head position. Those data of “pseudophakic accommodation” of this non-accommodative IOL are significantly higher than published for other IOLs. Tsorbatzoglou, Nemeth et al., using the defocusing technique, reported mean results of pseudophakic accommodation between 0.82 and 1.00 D [2]. Later, the same authors measured accommodation amplitudes in pseudophakic eyes using three different methods, and all mean values were lower than 1 D [5]. Uthoff et al., using the same device than Schopfer and coauthors (accommodometer), reported significantly more distant near points even for an accommodative IOL (0.6 m) as opposed to between 0.18 and 0.20 m measured by Schopfer et al. [3]. Kuchle et al., also using the accommodometer and an accommodative IOL, reported a postoperative mean of accommodative range of 2.02 D at 12 months [6]. Thus, data and the technique used to obtain them by Schopfer et al. [1] must be verified. Our experience using the iris-claw IOL (Artisan®) fixated in the posterior surface of the iris has also been positive. In the last 5 years we have implanted 22 lenses in combination with penetrating keratoplasty or DSAEK, and 54 lenses as a procedure to correct aphakia (41 as planned secondary implantation and 13 as contingency procedure in case of posterior capsule rupture). We used the A constant of 117.5 (SRK/T formula). A BCVA of 20/40 or better 6months after surgerywasmeasured in 36.6%of the eyes that underwent keratoplasty and 74.1 % of eyes that underwent only iris-claw IOL implantation, with a mean spherical equivalent of 0.05±2.6 D (range from −3.50 to +3.50 D) for the former group, and −0.62±1.06 D (range from −2.75 to + 1.25 D) for the latter group. BCVAworse than 20/40 was related with concurrent macular disease or ocular surface problems. We have abandoned suturing IOLs to sclera or to iris, since we feel that in these challenging cases, with absence of This work did not have public or private financial support.

Comment on: Fungal keratitis: The Aravind Experience

We have mentioned in our paper that the phaco technique used was direct phaco chop technique. However, we agree that cumulative dissipated energy could also have been additionally analyzed. Although we did match the grade of cataracts in the two groups, we have not analyzed the endothelial cell loss by cataract grade because the cataract subgroups were unequally distributed and our study did not had enough statistical power for analysis. These are aspects which could be looked at in further studies.

Aphakic iris-claw intraocular lens pseudophakic pseudoaccommodation

We read with interest the article on pseudophakic pseudoaccommodation by Lincke et al. that included 13 eyes with fixated retropupillary intraocular lenses (IOLs). The authors failed to cite a similar study published in 2012 by Sch€opfer et al. that comprised 51 eyes with a retropupillary fixated iris-claw IOL (Artisan or Verisyse). In the supine position, the mean anterior chamber depth was 4.01 mm G 0.24 (SD) in the group by Lincke et al. (measured using ultrasound biomicroscopy) versus 4.31G 0.44 mm in the group by Sch€opfer et al. (measured using A-scan). In the prone position, the mean values were 3.57 G 0.41 mm and 4.15 G 0.57 mm, respectively. Undoubtedly the statistically significant differences (PZ .0214 and PZ .0010, respectively) could be explained by the different device used and, in the prone position, by a difference in the angle with respect to the horizontal plane. (In the study by Sch€opfer et al., the patient was seated in forward tilted-head position.) However, it would be interesting to have the data obtained in retropupillary aphakic IOLs by Lincke et al. using A-scan in this subgroup of eyes. An additional interesting issue is the refractive impact of these IOL shifts. Lincke et al. found that in eyes with a retropupillary IOL, the mean difference in spherical equivalent (SE) in the supine position versus the prone position was 0.30G 0.53 diopter (D), which was not statistically significant (P Z .5823). In addition, as the authors explained, reading position does not equal the prone position used in the experiment because in down gaze, the eyes are declined by only approximately 30 degrees and the change in SE would be approximately one half the difference between the sitting position and the prone position (ie, approximately 0.15D in this subgroup of eyes with retropupillary implanted IOL). In contrast, in the study by Sch€opfer et al., there was a statistically significant difference ( 0.37 D; P Z .003). A myopic shift between 0.15 D and 0.37 D seems too small to have a real impact on the near-vision capabilities of the patients. Sch€opfer et al. measured the amplitude of accommodation with an accommodometer (Clement Clarke Ltd.) and found amplitudes of 4.96 D in the supine position, 5.70 D in the sitting position, and 5.18 D in the prone position. However, those results were not congruent with their findings on SE changes and were much higher than those published for other IOLs (including accommodating models), as we pointed out. Thus, we suggested that the data and the technique used to obtain them by Sch€opfer et al. should be verified. In summary, we believe the study by Lincke et al. adds to the body of evidence that aphakic iris-fixated IOLs (Artisan or Verisyse) shift with changes in position but that it seems

Small-incision lenticule extraction and corneal collagen crosslinking in keratoconus.

Small-incision lenticule extraction and corneal collagen crosslinking in keratoconus We read with interest the article by GraueHernandez et al. on combined small-incision lenticule extraction and intrastromal corneal collagen crosslinking (CXL) to treat mild keratoconus. The authors indicated in the first paragraph: “Keratoconus is a bilateral, asymmetrical, noninflammatory, and progressive ecstatic disorder of the cornea.” First, undoubtedly they meant ectatic and not ecstatic. Second, the topic of inflammation in keratoconus has been intensively researched in the past decade, and the formerly accepted noninflammatory nature of the condition seems to be challenged. A significant role of proteolytic enzymes, cytokines, and free radicals has been found and although keratoconus does not meet all the classic criteria for an inflammatory disease, the lack of inflammation has been questioned. The majority of studies of the tears of patients with keratoconus have found increased levels of interleukin-6, tumor necrosis factor-a, and matrix metalloproteinase 9. With regard to the results showed by GraueHernadez et al. in a group of 15 eyes with mild keratoconus treated with the Aztec protocol (as described by the authors), they are very impressive. The postoperative uncorrected distance visual acuity 12 to 24months after surgery was 0.12 logMAR G 0.20 (SD) (Snellen 20/26). The authors stated, “Therefore, using smallincision lenticule extraction to correct the refractive myopic and astigmatic error while minimizing biomechanical weakening, and combining this technique with intrastromal CXL to increase stability while overcoming the potential side effects of epithelial removal, appeared to be an ideal therapeutic approach to treat refractive error in mild keratoconic patients.”However, as they correctly concluded in the last section of the article, their results must be interpreted cautiously because the study sample is small and, we would add, the follow-up length is not long enough. Recently, 2 cases of bilateral ectasia after femtosecond laser small-incision lenticule extraction in cases of subclinical keratoconus have been reported. El-Naggar described a case of bilateral ectasia in a 33-year-old patient 6 months after the surgery, and Wang et al. reported the same complication in a 19-year-old patient. In the series by Graue-Hernandez et al., before surgery the eyes reached a corrected distance visual acuity (CDVA) of 20/40 or better and a mean CDVA of 20/20. They had on average a moderate cylinder (mean 2.3 G 1.4 diopters). Taking those data into account, the procedure performed could be considered primarily refractive surgery. (No information on

Intracameral Cyclosporine Drug Delivery System in High-risk Keratoplasty

RADHIKA TANDON, MD RASIK B. VAJPAYEE, FRCSED, FRANZCO Dr. Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi, India; Ocular Pharmacology Services, Dr. Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi, India; Ocular Microbiology Services, Dr. Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi, India; Centre for Eye Research Australia, Royal Victorian Eye and Ear Hospital, University of Melbourne, Australia

Iris-clip versus iris-claw intraocular lenses

We read with interest the case series on anterior megalophthalmos by Messina et al. They indicated that in 4 eyes, an iris-clip anterior chamber intraocular lens (IOL) was implanted. There is a difference between the terms iris-clip IOL and iris-claw IOL, which appears to have been overlooked by Messina et al. The term iris-clip refers to a model of IOLs that is no longer in use. On the other hand, iris-claw IOLs (Artisan, Ophtec BV, and Verisyse, Abbott Medical Optics, Inc.) have a totally different design and are available today. Designed by Binkhorst in 1957 and called the iris-clip IOL since the first publication on this topic in 1959 to the last reports of its use in the 1970s, the iris-clip IOL had 2 wire loops, bent at right angles and attached to its posterior surface close to the equator. This allowed it to be fixated at the pupil margin with one wire loop located anteriorly to the iris and the other one located posteriorly, in a fashion similar to how a paper clip works. On the other hand, the iris-claw IOLs have a different principle based on fixation to the anterior peripheral iris stroma. A loose fold of iris tissue is grasped by resilient claw-shaped haptic tips. Iris-claw IOLs were designed by Worst, who began implanting them in aphakic eyes in 1979. These iris-claw IOLs have been used in some countries in Europe since at least 1980 and today are available almost globally. However, some ophthalmologists in many countries have limited knowledge of the properties and indications of this IOL for aphakia and this IOL has not been used to its full potential. A surgeon in our group (V.G.) was the first to implant these types of IOLs in Colombia in 1998. Still, some surgeons in our country perform complex procedures (eg, suturing IOLs to sclera in eyes with a normal iris) even though fixation of an iris-claw IOL on the anterior surface or posterior surface of the iris is much safer and easier and has yielded excellent results. We noted that in Case 2, the authors performed a pars plana vitrectomy with lensectomy. Ten months later, an aphakia iris-claw IOL was implanted in the anterior chamber. Undoubtedly, to perform the 2 procedures in the same surgical setting would have been preferable. Six years later, loss of enclavation of the iris-claw IOL at the 9 o’clock position was observed. As the authors suggested, enclavation of a generous fold of iris tissue is recommended to prevent this complication. We had a positive experience, as cited by Messina et al., of fixating an iris-claw Artisan aphakic IOL to the posterior surface of the iris in an eye with anterior megalophthalmos; the results and long-term stability were very good. This approach has the advantage of placing the IOL in the posterior chamber, farther from the endothelium. On the other hand, it can be technically more challenging for the surgeon.

Severe pigment dispersion after iris-claw phakic intraocular lens implantation

A 23-year-old female patient presented 3 months after the implantation of an Artisan® phakic intraocular lens with a severe depigmentation of the iris and peripheral anterior synechiae. Explantation of the intraocular lens and goniosynechialysis were performed. Eleven months after the explantation appearance of the iris significantly improved. There was no loss of lines of corrected distance visual acuity. Severe pigment dispersion after the implantation of an Artisan® phakic intraocular lens may happen and may require explantation of the lens. Iris depigmentation may improve with time.

Impacts of Implantable Collamer Lens V4c Placement on Angle Measurements Made by Optical Coherence Tomography: Two-Year Follow-up

WE READ WITH INTEREST THE ARTICLE ON IMPLANTABLE collamer lens (ICL) V4c by Fernández-Vigo and associates. It would be interesting to know if eyes with a vault larger than 750 mm had a larger reduction of trabecular-iris angle (TIA).Also, as vault seemed to be a significant determinant of TIA reduction, it would be important to know the way the authors defined the size of the lens to be implanted. In addition, 14.8% of the eyes showed trabecular-iris contact (TIC). Did those eyes have narrower values of TIA? In a previous study performed by the same authors, they found that the mean of TIA was around 35 degrees for emmetropic eyes with a standard deviation of approximately 2.5 degrees. Assuming a normal distribution, around 99.85% of the eyes had an anterior chamber angle equal to or wider than 27.5 degrees. It is noteworthy that in the group of eyes implanted with ICL, assuming a normal distribution, only around 84% of the eyes had 2 years after surgery TIA nasal, TIA temporal, and TIA inferior equal to or wider than 18.5, 18.7, and 21.9 degrees, respectively. In addition, among the approximately 16% of eyes with narrower angles at least some had 0 degrees of TIA nasal, 5.3 degrees of TIA temporal, and 12.2 degrees of TIA inferior. It would be important to determine if those eyes that finished with such narrow angles also had narrower preoperative angles, in order to establish a minimal value of acceptable preoperative TIA. One eye showed high intraocular pressure (IOP) 1month postoperatively (45 mm Hg), but apparently related to steroid response. Similarly, Rodrı́guez-Uña and associates found in a larger group of 763 eyes with the same model of ICL 1 case with a significant increase in IOP (26 mm Hg) 1 week after surgery, also related to steroid response. Both studies, however, had only 2 years as the longest follow-up time. A current limitation when selecting an ICL for a given eye is the scarcity of size options. As several experts have indicated (J.F. Alfonso and C. Lisa, personal communication, October 4, 2017), it would be convenient to have at least 6, instead of 4. In addition, assuming a normal distribution, around 16% of the eyes in the studied group

Wavefront-Guided LASIK and Preoperative Higher Order Aberrations.

We read with interest the article by Kung and Manche in the April 2016 issue.1 In the first part of the article, they explained that most studies seemed to suggest that the wavefront-guided excimer laser procedures were slightly superior than wavefront-optimized approaches, particularly in patients with preoperative root mean square (RMS) of higher order aberrations (HOAs) less than 0.3 μm. However, the two cited publications by the authors (the prospective, open-label, multicenter study conducted by Stonecipher and Kezirian2 and the meta-analysis by Feng et al.3) reported exactly the opposite: if the magnitude of preoperative RMS was greater than 0.3 μm, wavefront-guided ablation had a significantly better postoperative aberration profile than wavefront-optimized ablation and, on the other hand, wavefront-guided treatments had no clear advantage over wavefront-optimized treatments in eyes with preoperative RMS lower than 0.3 μm.2,3 Kung and Manche found that the two platforms produced similar self-reported symptoms in patients with RMS aberrations greater than 0.3 μm but, in eyes with RMS aberrations less than 0.3 μm, the wavefrontguided platform resulted in higher self-reported “excellent vision” and significantly fewer adverse effects (eg, problems with daytime and nighttime clarity and visual fluctuation).1 However, the authors did not report an analysis on the postoperative aberration profile comparing the two subgroups: those with HOAs higher than 0.3 μm and those with less than that magnitude of aberrations. The findings on visual symptoms are counterintuitive because one would expect the impact of customized aberration correction to be higher in the group with higher preoperative aberrations, as shown in the other studies. We think that the numbers deserve to be rechecked to ensure that there are not any inaccuracies in the capture or analysis of the data. In addition, in Table 3 there seems to be confusion in the presentation of the information. It shows in the first column of each group, under a heading that says logMAR, data on manifest sphere, cylinder, and spherical equivalent, which units are in diopters. However, they are converted to “Snellen visual acuities.” Undoubtedly it is a mistake. The same occurs with the aberrometry data. The text referring to the Table 3 is not clear either. It says: “Postoperative measurements of visual acuity showed that manifest sphere (wavefront-guided vs wavefront-optimized: -0.32 vs -0.56; P = .0001, significant) and manifest spherical equivalent refraction (wavefront-guided vs wavefront-optimized: -0.18 vs -0.41; P = .0001, significant) were superior in the wavefront-optimized group (Table 3).” Manifest sphere and spherical equivalent are not visual acuity measurements, but refractive error determinations. If the postoperative refractive error was smaller, as is shown, in the wavefront-guided group of eyes, the statement made in the text is not correct. It is essential that the authors clarify these critical points so that we as readers can have a better understanding of the results of this interesting study.