Scheimpflug Analysis of Centration and Bending of Posterior Chamber Phakic Intraocular Lenses

Transcription

1 ORIGIL CLINICAL STUDY Scheimpflug Analysis of Centration and Bending of Posterior Chamber Phakic Intraocular Lenses Marie-José B. R. Tassignon, MD, PhD,*Þ Jos J. Rozema, PhD, MSc,*Þ and Sorcha Ní Dhubhghaill, PhD. MB* Purpose: The aim of the study was to present a novel technique for the assessment of postoperative centration and bending in phakic intraocular lens (PIOL), illustrated through 4 patient examples. Design: This was a prospective observational study. Methods: Four patients (7 eyes) who presented with complaints of reduced quality of vision after implantation with a Staar Collamer plate haptic PIOL were recruited for assessment. Scheimpflug images were obtained and analyzed with respect to the apical axis and the crystalline lens by means of an in-houseydeveloped algorithm. The program provides an estimate for the PIOL centration and bending through an analysis of the distance between the PIOL and the crystalline lens. Results: The algorithm detected PIOL bending in 5 of the 7 eyes and PIOL decentration in 3 eyes. The PIOL bending or decentration detected corresponded, in all cases, with the axis of the astigmatism-like complaints in the patient. In 2 eyes of the same patient, the analysis could not be performed because the distance between PIOL and crystalline lens was too short for reliable analysis. Conclusions: This algorithm is a useful tool to objectively assess the postoperative complaints of patients implanted with a posterior chamber PIOL. Bending and decentration due to anatomical variations of the sulcus lens position can be measured and in many cases correlated with patients complaints. Key Words: plate haptic phakic intraocular lens (PIOL), PIOL vaulting, Scheimpflug imaging, PIOL-induced astigmatism, plate haptic posterior chamber IOL (Asia-Pac J Ophthalmol 214;3: 136Y14) Phakic intraocular lenses (PIOLs) have been introduced in the last decade for the correction of the patient s refractive error while keeping the natural crystalline lens untouched. Different implantation techniques and methods of IOL positioning have been proposed with a variable incidence of complications. 1 The incision size required depends on the IOL type, ranging from 6 mm for the polymethyl methacrylate angle- or iris-supported PIOLs, to 4 or 3.8 mm for the foldable iris-supported, anglesupported or posterior chamberysupported PIOLs. Of these lenses, the Staar Collamer plate lens (Staar Surgical Company, Monrovia, Calif ) is particularly popular because of its small corneal incision requirement. Subjective clinical outcomes have been reported as very good for myopic to highly myopic patients 2,3 and good for the hyperopic patients. 4 The adverse From the *Department of Ophthalmology, Antwerp University Hospital, Edegem; and Faculty of Medicine and Health Science, Antwerp University, Wilrijk, Belgium. Received for publication May 31, 213; accepted August 24, 213. The authors have no funding or conflicts of interest to declare. Reprints: Marie-José B. R. Tassignon, MD, PhD, Department of Ophthalmology, Antwerp University Hospital, Wilrijkstraat 1, 265 Edegem, Belgium. marie-jose.tassignon@uza.be. Copyright * 214 by Asia Pacific Academy of Ophthalmology ISSN: DOI: 1.197/APO.4 effects of this lens have been widely studied and include cataract formation, 5,6 IOL dislocation, 7,8 pupillary block, 9,1 and retinal detachment. 11 The purpose of this article was to report more subtle complications arising from Collamer plate haptic lens implantation that are related to lenticular bending or decentration in patients referred from refractive surgical centers to a tertiary university hospital setting. The aim was to develop an objective means of detecting phakic IOL bending, vaulting, and decentration in dissatisfied patients by means of Scheimpflug imaging supplemented by an in-houseydeveloped analysis method. 12 MATERIALS AND METHS Patients Participants were recruited from the clinics of Antwerp University Hospital. All patients were referred to the tertiary center for a second opinion regarding postoperative phakic IOL dissatisfaction in at least 1 eye. Four participants (7 postoperative eyes) were recruited, and of this group, not all postoperative eyes were symptomatic. Patients underwent 3- dimensional Scheimpflug measurement using the Pentacam (Oculus Optikgeräte GmbH, Wetzlar, Germany). All patients gave informed consent, and the research was conducted according to the tenets of the Declaration of Helsinki. Analysis of the Images The calculation method has been described in a previous article 12 and uses the distance d (in micrometers) between the PIOL and the crystalline lens along a circle with a diameter of 5 mm, centered on the corneal apex, which is the standard measurement axis of the Pentacam. When a PIOL is well centered with respect to the crystalline lens, the distance between them is shorter in the center than in the periphery. This relationship should be the same along all meridians, which means that the distance d should be constant along the measurement circle. Shifting the PIOL laterally in any direction would cause d to increase in the direction of the shift and to decrease in the opposite direction. In this case, the pattern of d along the measurement circle would result in a sine wave with a single maximum and a single minimum. When a PIOL is folded, one would expect 2 regions of the IOL to be closer to the crystalline lens and 2 regions farther away, aligned at a 9-degree angle to one another. Here, the pattern of d along the measurement circle would appear as a sine wave with 2 maxima and 2 minima. In practice, the Pentacam software was used to manually determine the distance d between PIOL and crystalline lens at 2 points 2.5 mm to the left and right from the apical axis for all 25 Scheimpflug images (Fig. 1A). Next d is plotted as a function of the meridian 5 along which it was measured, after which a least-squares fit is done with the function: dð5þ ¼> 1 cosð5 þ E 1 Þþ> 2 cosð25 þ E 2 Þþd ð1þ where a 1 and E 1 are the amplitude and direction of the Asia-Pacific Journal of Ophthalmology & Volume 3, Number 3, May/June 214

2 Asia-Pacific Journal of Ophthalmology & Volume 3, Number 3, May/June 214 Centration and Bending of a PIOL FIGURE 1. Scheimpflug images and vaulting analysis of patient 1 for (A, B) the RE before PIOL explantation; (C, D) the LE at the same moment; (E, F) the RE after PIOL exchange. The points indicate the measured distance d (Hm), the black line gives the corresponding fit using Eq. (1), and the red line shows the fit of Eq. (1) after subtraction of its first term to correct for PIOL decentration. effect of PIOL decentration; a2 and E2, the amplitude and direction of PIOL bending; and d, the average distance between the lenses. In mild or moderate cases, decentration and PIOL bending occur independently of each other, which makes their effects on d independent as well. It is therefore possible to study decentration and PIOL bending separately by subtracting the decentration component as defined in Eq. (1), to yield the IOL bending component. RESULTS Patient 1 A 35-year-old myopic woman with a spherical equivalent (SE) of j11 diopters (D) in the right eye (RE) and j6.25 D in the left eye (LE) underwent a bilateral Staar (ICM12) phakic IOL implantation. The preoperative anisometropia was due to a difference in axial length between both eyes: 29.7 mm (RE) and mm (LE). Preoperative distance correct visual acuity (DCVA) was.8 RE and 1. LE. The patient underwent an uncomplicated implantation of a phakic Staar IOL of j14.5 D (RE) and j9 D (LE) and a haptic diameter of 12. mm. Postoperative corneal curvatures and astigmatism were similar for both eyes (Table 1). In the early postoperative period, the patient complained of a reduced quality of vision in her RE, corresponding with a stretching of the image suggesting an astigmatism-like effect. The direction of this stretching was estimated by the patient at the 13- to 31-degree meridian. The measured corneal astigmatism was minimal and did not account for this. Although this may respond to standard refractive correction, the patient did not wish to return to spectacle or contact lens wear and decided to seek a second opinion. To objectively assess the complaints, the distance d was analyzed, which demonstrated the presence of a minor decentration (a1 = 7.17 Hm) in the direction E2 = degrees and a PIOL bending (a2 = Hm) along the to 231.5degree meridian (Figs. 1A, B). The PIOL bending can cause image stretching, which corresponded with the 13- to 31degree meridian as indicated subjectively by the patient. For the LE (Figs. 1C, D), only a minor PIOL decentration and TABLE 1. Overview of Patient Biometry Patient 1 Preoperative SE, D PIOL parameters Sphere, D Cylinder, D Haptic size, mm Postoperative SE, D K flat, D Axis 1, degrees K steep, D Axis 2, degrees Axial length, mm DCVA, decimal Patient 2 Patient 3 Patient 4 j11. j6.25 j7.75 j7.75 j8.25 j7.5 j3.5 j4.75 j j j j j j j j j j j j j indicates not available. * 214 Asia Pacific Academy of Ophthalmology 137

3 Tassignon et al Asia-Pacific Journal of Ophthalmology & Volume 3, Number 3, May/June 214 FIGURE 2. Slit-lamp photographs and vaulting analysis of patient 2 for (A, B) the RE; (C, D) the LE. The black line shows distance d (Hm) without correction; the red line shows distance d (Hm) after correction for decentration. bending were seen (respectively, a 1 =2.99Hm and a 2 = 1.59 Hm). The minimal bending and decentration in the asymptomatic LE may account for her high degree of satisfaction on this side. Based on these measurements, her primary surgeon removed the PIOL in the RE and exchanged it for another Staar PIOL with a power of j13 D, but a larger haptic diameter of 12.6 mm. Objective analysis of the Scheimpflug images after the exchange showed that the average d had increased along all meridians from d = Hmtod = Hm; the amplitudes of the decentration and bending had increased to a 1 = Hm and a 2 = Hm, respectively, and with bending along the to degree meridian (Figs. 1E, F). Unfortunately, the patient s visual acuity declined further after the exchange because of the appearance of posterior subcapsular cataract. Patient 2 A 29-year-old woman presented with a history of an axial myopia of SE j7.75 D in both eyes. Her preoperative DCVA was.7 (RE) and.6 (LE). She underwent an uncomplicated implantation of a phakic Staar (ICM12) IOL of j15.5 D, cylinder +4.5 D (RE) and j17.5 D, cylinder +5.5 D (LE), combined with a large superior iridectomy. She was unsatisfied postoperatively with her quality of vision, which involved monocular diplopia in both eyes, haloes, increased glare, and reduced contrast sensitivity. Her postoperative SE refraction was j1.25 D (RE) and j1. D (LE), her corrected distance visual acuity was 1. in both eyes, and her corneal curvatures and astigmatism are shown (Table 1). Analysis of the Scheimpflug images showed that the average distance between the PIOL and the crystalline lens was considerably larger in the RE than in the left (d,right =513.6Hm, d,left = Hm) and that the PIOL was decentered superiorly in both eyes (Figs. 2A-D). In both eyes, PIOL bending was also observed: in the RE with an amplitude a 2 =21.1Hm along the to 29.6-degree meridian (red line, Fig. 2B) and in the LE with an amplitude a 2 = Hm along the 8.- to 26.-degree meridian (red line, Fig. 2D). In both eyes, the meridian of the bending was perpendicular to the meridian of the PIOL haptics. The PIOLs were removed from both eyes, and the iridectomies were closed surgically. Her complaints of diplopia and glare disappeared soon thereafter under full refractive correction. Although it is difficult to directly attribute the complaints to the pathology, it is likely that the haloes and glare were due to the large iridectomies, whereas the monocular diplopia was likely caused by the bending of the lens. Patient 3 A 26-year-old woman with an axial myopia of SE j8.25 D (RE) and j7.5 D (LE) underwent bilateral Staar (ICM12) phakic IOL implantation. Preoperative DCVA was 1. in both eyes. The patient underwent an uneventful implantation of a phakic Staar IOL of j9 D (RE) and j8.5 D (LE) and a haptic size of 12.6 mm. Her corneal curvatures and astigmatism measured by Pentacam are presented (Table 1). Three months postoperatively, both eyes were emmetropic and achieved an uncorrected visual acuity of 1.2. This patient was satisfied with her postoperative outcome. The Scheimpflug images taken at that time however showed an inferior PIOL decentration in both eyes (Figs. 3A, B). After correction for the decentration the analysis of the distance between the lenses showed that the PIOL was folded with an amplitude a 2 = 26.5 Hm along the to degree meridian for her RE (red line, Fig. 3B). Three months later, she came back for a follow-up, and at that time, the bending amplitude was in her RE a 2 =24.13Hmalong the to degree meridian (red line, Fig. 3D). Note that the bending was again perpendicular to the direction of the decentration (Figs. 3A, C). Wavefront aberrations in the RE were low, however, leading to the result that the unintended decentration and deformation of the PIOL managed to correct for the corneal astigmatism present at the - to 18-degree meridian. Patient 4 A 37-year-old man was treated for astigmatic myopia 2 years previously with bilateral toric Staar (ICM12) PIOLs. The patient was amblyopic in the LE. RE was j1. D (j5.25 D at 156 degrees) and LE was j2.d (j5.5 at 22 degrees) * 214 Asia Pacific Academy of Ophthalmology

4 Asia-Pacific Journal of Ophthalmology & Volume 3, Number 3, May/June 214 Centration and Bending of a PIOL FIGURE 3. Scheimpflug images and vaulting analysis of patient 3 for (A, B) the RE at the first visit and (C, D) the second visit. The black line shows distance d (Hm) without correction; the red line shows distance d (Hm) after correction for decentration. Two years later, the patient presented for second opinion regarding complaints of glare. Clinical examination revealed a posterior subcapsular cataract, which would account for the glare symptoms. Scheimpflug analysis illustrated a PIOL/ crystalline lens gap profile that was uniformly reduced to 4 Hm in the RE (Fig. 4). In this case, the distance between the PIOL and the crystalline lens was too short for a reliable analysis. We include this case as a demonstration of the main limitation in the proposed algorithm. DISCUSSION The popularity of the phakic IOLs is related to their good refractive outcomes in high ametropia and a higher postoperative quality of vision compared with laser-assisted in situ keratomileusis for the same range of refraction error. 13,14 Safety of the procedure was also found to be superior to refractive lens exchange. 15 As mentioned previously, however, phakic IOL implantation has been associated with postoperative complications such as retinal detachment and cataract. Chen s metaanalysis of 6338 eyes implanted with different types of PIOLs showed that the incidence of cataract was 9.6% for sulcus-sited PIOLs compared with 1.29% and 1.11% for anterior chambery and iris-fixated PIOLs, respectively. 16 Sulcus-sited PIOLs therefore confer a significant risk of postoperative cataract formation. The postoperative position of a PIOL is crucial to longterm quality outcomes. We have described how this can be monitored using anterior segment optical coherence tomography or Scheimpflug imaging. It may also have applications in determining the best possible shape for the phakic PIOL in order to reduce cataract formation. 17 Although both devices are capable of measuring along meridians 36 degrees around the optical axis, usually only 1 meridian is chosen by the surgeon to define the postoperative position of the PIOL within the eye. 1,18 The complaints of patient 1 were suggestive of astigmatism. The measured postoperative corneal curvature showed only a minimal astigmatism, which would not account for the symptoms. The analysis of the PIOL over 36 degrees showed a bending of the PIOL. The bending corresponded to the patient s complaint effectively causing secondary lenticular astigmatism. This vaulting may be related to an IOL plate size that was too large with respect to the sulcus space available along that particular meridian. The patient returned to her initial surgeon who exchanged her PIOL for a lens with a lower power and an even larger plate size. This was implanted along a meridian almost perpendicular to the orientation of the first PIOL. The treatment resulted in a minor increase in vaulting and a rotation of the bending meridian by about 7 degrees, aligning to the meridian of the corneal astigmatism leading to improved vision. Patient 2 presented with more complex complaints, with disabling glare symptoms that were likely due to the very large iridectomies in both eyes. Analysis of the Scheimpflug images demonstrated that the PIOL bending was smaller for the RE than the left and might have given an acceptable quality of vision, provided the iridectomy was closed. The LE, however, demonstrated a more significant degree of bending along the 8.- to 26.-degree meridian. Even after closure of the iridectomy, it is likely that her quality of vision would have remained relatively poor in comparison with the RE. Patient 3 had an excellent postoperative outcome and no postoperative complaints. This case is included to illustrate a fortuitous feature of her lens implant. Scheimpflug analysis indicated bending and inferior decentration of the spherical IOL in her RE. Typically, this would lead to considerable ocular aberrations. We did not detect any large postoperative aberrations indicating that they had been unexpectedly compensated. The vaulting along the horizontal meridian canceled out her with-the-rule corneal astigmatisms of 1.3 D. The case illustrates how coincidence can sometimes play a major role in a surgical outcome and that this is not necessarily undesirable. Finally, patient 4 was added to show that in case the gap between the PIOL and the crystalline lens can disappear over time. The complaints of the patient were due to coincidental FIGURE 4. Scheimpflug image of patient 4. * 214 Asia Pacific Academy of Ophthalmology 139

5 Tassignon et al Asia-Pacific Journal of Ophthalmology & Volume 3, Number 3, May/June 214 cataractous changes of the crystalline lens rather than due to a suboptimal PIOL. This technique, however, is unable to predict the onset of cataract after PIOL. In conclusion, we have developed a simple in-house algorithm that is a useful tool to better understand the patients complaints after uneventful PIOL implantation. The results obtained illustrate that PIOLs may bend along a particular meridian. This case series displays the potential for PIOLs to change in shape and centration over time after implantation. One patient in the series was asymptomatic as the degree of bending was fortuitously compensated for by the corneal astigmatism providing an unplanned but beneficial toric phakic IOL effect. The degree to which this lens bending occurs in the postoperative population is as yet unknown, although the technique described may assist in a prospective study to assess this. The data in this case series provide promising insight into a potential source of postoperative visual symptoms in this population and may be a beneficial tool in the hands of refractive surgeons investigating such complaints. A large-scale prospective clinical trial in collaboration with a high-throughput refractive centre will be required to validate this technique in patients with and without postoperative visual complaints. REFERENCES 1. Kohnen T, Kasper T, Terzi E. Intraocular lenses for the correction of refraction errors. Part II. Phakic posterior chamber lenses and refractive lens exchange with posterior chamber lens implantation. Ophthalmologe. 25;12:115Y Zaldivar R, Davidorf JM, Oscherow S. Posterior chamber phakic intraocular lens for myopia of j8 toj19 diopters. J Refract Surg. 1998;14:294Y Menezo JL, Peris-Martinez C, Cisneros A, et al. Posterior chamber phakic intra-ocular lenses to correct high myopia: a comparative study between Staar and Adatomed models. J Refract Surg. 21;17:32Y Davidorf JM, Zaldivar R, Oscherow S. Posterior chamber phakic intraocular lens for hyperopia of +4 to +11 diopters. J Refract Surg. 1998;14:36Y Brauweiler PH, Wehler T, Busin M. High incidence of cataract formation after implantation of a silicone posterior chamber lens in phakic, highly myopic eyes. Ophthalmol. 1999;16:1651Y Wiechens B, Winter M, Haigis W, et al. Bilateral cataract after phakic posterior chamber top hat-style silicone intraocular lens. JRefractSurg. 1997;13:392Y Mastropasqua L, Toto L, Nubile M, et al. Long-term complications of bilateral posterior chamber phakic intraocular lens implantation. J Cataract Refract Surg. 24;3:91Y Al-Swailem SA, Al-Rajhi AA. Decentration and cataract formation 1 years following posterior chamber silicone phakic intraocular lens implantation. J Refract Surg. 26;22:513Y Smallman DS, Probst L, Rafuse PE. Pupillary block glaucoma secondary to posterior chamber phakic intraocular lens implantation for high myopia. J Cataract Refract Surg. 24;3:95Y Park IK, Lee JM, Chun YS. Recurrent occlusion of laser iridotomy sites after posterior chamber phakic IOL implantation. Korean J Ophthalmol. 28;22:13Y Domènech NP, Arias L, Prades S, et al. Acute onset of retinal detachment after posterior chamber phakic intraocular lens implantation. Clin Ophthalmol. 28;2:227Y Tassignon M-J, Rozema JJ, Ni Dhubhghaill S. 3D Scheimpflug measurement of posterior chamber plate haptic phakic intraocular lens/crystalline lens gap profile. Acta Ophthalmol In press. doi: Tsiklis NS, Kymionis GD, Karp CL, et al. Nine-year follow-up of a posterior chamber phakic IOL in one eye and LASIK in the fellow eye of the same patient. J Refract Surg. 27;23:935Y Barsam A, Allan BD. Excimer laser refractive surgery versus phakic intraocular lenses for the correction of moderate to high myopia. Cochrane Database Syst Rev. 21;12:CD Huang D, Schallhorn SC, Sugar A, et al. Phakic intra-ocular lens implantation for the correction of myopia: a report by the American Academy of Ophthalmology. Ophthalmology. 29;116:2244Y Chen LJ, Chang YJ, Kuo JC, et al. Metaanalysis of cataract development after phakic intraocular lens surgery. J Cataract Refract Surg. 28;34:1181Y Baikoff G. Anterior segment OCT and phakic intraocular lenses: a perspective. J Cataract Refract Surg. 26;32:1827Y Baumeister M, Bühren J, Kohnen T. Position of angle-supported, iris-fixated, and ciliary sulcus-implanted myopic phakic intraocular lenses evaluated by Scheimpflug photography. Am J Ophthalmol. 24;138:723Y731. "Let food be thy medicine and medicine be thy food." V Hippocrates 14 * 214 Asia Pacific Academy of Ophthalmology