Piggyback posterior chamber intraocular lenses (IOLs) for correction of high hyperopia in eyes having cataract extraction or refractive lensectomy have become increasingly popular.1–4 However, little has been published regarding long-term follow-up in these cases. We present a series of 6 eyes of 3 patients who had a late hyperopic shift associated with Elschnig pearl proliferation in the peripheral interface between piggyback lenses. To our knowledge, this is the first reported complication of piggyback IOLs.
A 67-year-old woman developed moderate nuclear sclerotic cataracts in both eyes, impairing her reading ability. Ocular history was remarkable for exotropia. Best corrected visual acuity (BCVA) was 20/50 in both eyes, with a refraction of +6.75 sphere in the right eye and +5.75 sphere in the left. Motility examination was remarkable for 70 prism diopters (PD) of exoptropia. Anterior segment examination was normal except for moderately dense nuclear sclerotic cataracts in both eyes. The remainder of the comprehensive examination was normal. Average manual keratometry was 44.62 diopters (D) in the right eye and 45.93 D in the left; axial lengths measured 19.92 and 19.95 mm by contact A-scan and 20.19 and 20.45 mm by immersion.
In February 1996, the right eye had phacoemulsification by one of us (J.K.S.) with implantation of 17.5 and 17.0 D acrylic IOLs (model MA60BM) through scleral tunnel incisions; the 2 IOLs were placed in a piggyback configuration in the capsular bag with the haptics aligned parallel. The left eye had an identical procedure 1 week later in which 2 16.5 D IOLs were implanted. One month postoperatively, uncorrected visual acuity (UCVA) was 20/40−1 in the right eye and 20/30−2 in the left. Refraction of −1.75 +0.75 × 90 yielded a visual acuity of 20/30−2 in the right eye, while −1.50 +1.00 × 90 yielded 20/30−2 in the left. This patient was described in the first report of stacked acrylic IOLs.3
In May 1997 at routine follow-up, the patient was asymptomatic and happy with her vision. Her visual acuity with the spectacles prescribed at 1 month refraction was 20/30 in the right eye and 20/25−3 in the left. In November 1997, the patient complained of a drop in visual acuity during the preceding week. Her acuity with glasses had decreased to 20/200 in the right eye and 20/40−1 in the left. On dilated examination, Elschnig pearls were noted in the peripheral interface between the IOLs in the right eye, and the central interface between the IOLs appeared to contain a brownish material in a circular distribution. The patient returned for retinoscopy and refraction through undilated pupils. Despite the biomicroscopic appearance of the interface between the IOLs in the right eye, an excellent retinoscopic reflex was present in both eyes without distortion. With a correction of +2.25 +0.25 × 165, the right eye achieved a visual acuity of 20/30, while the left eye also corrected to 20/30 with a refraction of +2.00 sphere. The patient was closely observed over the ensuing 15 months; visual acuity and refraction remained stable without further intervention.
To determine the cause of the patient's hyperopic shift, keratometric and biometric measurements were repeated in February 1999 using the same keratometer and A-scan unit as in the initial measurements. Mean keratometry was 45.25 D in the right eye and 45.62 D in the left. By immersion A-scan, anterior chamber depth measured 3.68 mm and 3.33 mm, respectively, while preoperative immersion scans showed anterior chamber depths of 2.76 mm and 2.64 mm, respectively. Preoperative phakic lens thickness measured 4.68 mm in both eyes. Assuming an ultrasound speed of 2120 m/s in acrylic material, the distance from the anterior surface of the anterior IOL to the posterior surface of the posterior IOL was 3.02 mm in the right eye and 2.97 mm in the left. Postoperative axial length was 20.58 mm and 20.80 mm, respectively.
A 66-year-old woman complained of poor distance vision, impairing her ability to drive. Both eyes were best corrected to 20/50, with a refraction of +7.00 +1.25 × 180 in the right eye and +7.00 +2.00 × 10 in the left. Comprehensive examination was remarkable for moderately dense nuclear sclerotic and cortical cataracts in both eyes. Mean keratometry was 44.62 D in the right eye and 44.88 D in the left, and axial lengths were 20.74 mm and 20.47 mm, respectively.
In June 1996, one of us (J.K.S.) performed phacoemulsification with implantation of piggyback 16.5 and 16.0 D acrylic posterior chamber IOLs (model MA30BA) through a clear corneal incision, along with simultaneous arcuate keratotomy in the patient's right eye. An identical procedure was performed in the left eye the next week using IOL powers of 16.5 and 15.5 D. The surgeon attempted to leave all haptics 90 degrees apart after each procedure.
One month postoperatively, UCVA was 20/20 in the right eye and 20/40 in the left. In the right eye, refraction with −0.50 +0.50 × 15 yielded a visual acuity of 20/20, while the left eye corrected to 20/20+1 with −1.25 +1.25 × 177. The IOL haptics in the right eye were positioned 90 degrees apart, while those in the left were 30 degrees apart. Vision and refraction were stable at the patient's last routine follow-up examination in January 1997.
The patient was lost to follow-up, but she returned in August 1998 complaining of decreased visual acuity in both eyes during the previous month. With glasses, the patient's visual acuity was 20/200 in both eyes. Refraction of +1.75 +0.25 × 20 yielded an acuity of 20/20−3 in the right eye, while the left eye corrected to 20/20−1 with +1.25 +0.75 × 178. Dilated examination showed Elschnig pearls in the peripheral interface between the IOLs in both eyes (Figure 1).
In May 1997, a 72-year-old man with a history of nanophthalmos and high hyperopia presented to one of us (T.S.) complaining of poor visual acuity interfering with reading and driving. Review of records from 1978 showed a spherical equivalent (SE) of +4.75 in the right eye and +4.87 in the left. Until the patient began to develop cataracts, the SEs had been stable for many years. Best corrected visual acuity was 20/40− in the right eye and 20/30− in the left. Anterior segment examination showed the anterior chambers to be shallow. Gonioscopy showed the angles to be A0S by the Spaeth classification in the right eye and AB5S in the left; intraocular pressure (IOP) was 16 mm Hg and 14 mm Hg, respectively. Corneas showed few guttata, and anterior chambers were shallow. Brunescent nuclear sclerotic cataracts were present. Asteroid hyalosis was present in the right eye. Funduscopic examination was unremarkable with cup-to-disc ratios of 0.4 to 0.5 in both eyes.
In May 1997, the patient had successful laser peripheral iridotomies in both eyes. The angles were graded as C10S in both eyes postoperatively, and IOP was 8 mm Hg and 7 mm Hg in the right and left eye, respectively. Mean keratometry was 47.68 D and 48.42 D and axial length, 20.01 mm and 19.60 mm, respectively. In July 1997, the right eye had phacoemulsification with implantation of 2 poly(methyl methacrylate) (PMMA) IOLs (ORC U251F), 16.0 and 15.5 D, placed in a piggyback configuration in the capsular bag. An identical procedure was performed in the left eye 1 week later; both IOL powers were 16.5 D and the haptics were positioned 90 degrees apart. Two weeks after the second eye had cataract extraction, the right eye corrected to 20/25−2 with a refraction of plano sphere, while BCVA in the left eye was 20/25 with +0.50 +0.50 × 20.
The patient did well until January 1998, when he complained of decreased visual acuity in both eyes. Best corrected visual acuity had decreased to 20/50+2 in both eyes, with 1 to 2+ posterior capsule opacification, greater in the right eye. The patient was referred for vitreoretinal consultation to exclude a posterior segment etiology for the decreased visual acuity. Mild epiretinal gliosis was noted in both eyes, worse in the left eye, and choroidal folds were noted in the superior retina in the right eye. The consultant concurred with the indication for neodymium:YAG (Nd:YAG) laser posterior capsulotomy, which was performed in the right eye in March 1998 and in the left in April 1998. Refraction 1 day after the second capsulotomy showed the right eye corrected to 20/30−1 with +1.00 +1.25 × 15, while the left eye corrected to 20/30 with +1.00 sphere. When the patient was re-evaluated 1 month later, spectacles were prescribed and provided a visual acuity of 20/30+1 in the right eye and 20/30 in the left.
The patient was asymptomatic until July 1998, when he returned complaining of decreased vision in the right eye for the preceding 2 weeks. Best spectacle-corrected visual acuity had decreased to 20/200 in the right eye, but it corrected to 20/30 with a refraction of +4.25 +0.75 × 10. Vitreoretinal evaluation was done again, and in the right eye, Elschnig pearls were noted in the interface between the 2 IOLs, with a confluent haze centrally. A double interface was identified with the slit beam, and there was no obvious separation of the lenses. Funduscopic examination was unchanged, and fluorescein angiography was unremarkable. In August 1998, refraction in the right eye remained stable, but the left eye refracted to +2.25 +0.50 × 22. In September 1998, the patient noted decreased visual acuity in the left eye, which now required +5.50 +0.25 × 20 refraction to achieve a visual acuity of 20/25−.
Each of the 6 eyes receiving piggyback posterior chamber IOLs experienced a clinically significant hyperopic shift between 1 and 2 years postoperatively. All eyes had proliferating Elschnig pearls visible in the peripheral interface between the IOL optics. Neodymium:YAG laser posterior capsulotomy was performed in eyes with PMMA IOLs prior to hyperopic shift, while eyes with acrylic IOLs had no secondary procedure.
The cellular material proliferating in the peripheral interface between the lenses appears to cause posterior displacement of the posterior IOL, explaining at least part of the hyperopic shift. The combined optic thickness of the IOL pair is about 1.5 mm for this lens model and power (Richard Lambert, PhD, DVM, Alcon Laboratories, Inc., personal communication, February 1999), while the thickness of each pair was about 3 mm measured by ultrasound. Although no visible separation between the central optic surfaces has been observed biomicroscopically, the late postoperative ultrasound biometric measurements in Case 1 show about 1.5 mm of separation between the 2 optics centrally. Late Elschnig pearl formation in the interface between the lenses appears to force the IOL optics apart. The greater magnitude of late hyperopic shift in the 2 eyes with prior capsulotomy further supports the hypothesis that the optic of the posterior IOL had migrated posteriorly. However, a 1.5 mm posterior shift of a 16.5 D IOL should induce only about 2.4 D of hyperopia, so a second mechanism may explain the remainder of hyperopic shift, at least in the 2 eyes available for postoperative immersion biometry.
A second possible cause of the hyperopic shift is separation of the 2 optic surfaces peripherally, which could affect zonular tension and consequently cause posterior displacement of the IOL/capsular bag complex. As the IOL optics are spread farther apart by material proliferating in the interface between them, tension is taken off the loops and the bag equator, and the capsulor–zonular apparatus could move posteriorly, similar to a nonaccommodated state. Finally, a third possible cause is Elschnig pearl material proliferating under the capsulorhexis and displacing the pair of IOL optics posteriorly. However, this should not be more likelywith piggyback implantation than with monopseudophakic eyes, and the first 2 mechanisms may explain this phenomenon.
Repeated measurement of Case 1's keratometry and axial length also excluded the possibility of late flattening of the cornea or shortening of the eye. Previous reports5–7 have described uveal effusion in nanophthalmic eyes, especially associated with intraocular surgery; therefore, it was important to ensure stability of axial length measurements.
When Case 1 initially presented, the first postulated explanation of the hyperopic shift was compression of the IOLs against each other with flattening of the optics. At the time of this report, one of us (J.K.S.) had placed piggyback acrylic IOLs in 75 eyes and observed a contact zone of compression between the IOLs that is probably responsible for an increased depth of focus observed in many of these polypseudophakic eyes (Joel K. Shugar, MD, MSEE, “Piggyback Acrylic Posterior Chamber Lenses: Omniopia?”, presented at the Symposium on Cataract, IOL and Refractive Surgery, Boston, Massachusetts, USA, April 1997). Subsequent to IOL implantation in the first 4 eyes in the initial report, lens model MA30BA has been used in all piggyback acrylic implantations. In each of the eyes with this model, the surgeon has tried to leave the haptics almost exactly 90 degrees apart after surgery rather than obliquely, as described by Masket.5 Perpendicular positioning may increase the contact zone size between the lenses by sandwiching them more tightly together, thereby increasing the magnitude of the omniopia effect. However, compression of the optic surfaces could not explain the hyperopic shift in the eyes with PMMA IOLs, and once ultrasound biometry was performed in Case 1, late compression of the optics against each other was excluded as a possible mechanism.
Two eyes with late hyperopic shift have stacked lenses (model MA60BM) with parallel haptics, 1 eye has model MA30BA with haptics in a parallel orientation, and 1 eye has model MA30BA with haptics in a perpendicular orientation. One eye with stacked PMMA IOLS has haptics positioned parallel, while the other's haptics are positioned perpendicularly. The effect of haptic orientation on the incidence and magnitude of Elschnig pearl ingrowth and late hyperopic shift requires further study. The eye with stacked MA30BAs, whose haptics were noted to be parallel during the late postoperative period, corroborates Masket's observation that the 2 lenses can sometimes rotate with respect to one another during the early postoperative period.5 One of us (J.K.S.) has observed such late rotation in 6 eyes with piggyback acrylic IOLs. None of these eyes has required an Nd:YAG capsulotomy, so its effect in these eyes is speculative. We are trying to follow this cohort regularly, and data from these eyes will be reported periodically.
The long-term incidence of Elschnig pearl ingrowth and late hyperopic shift in eyes with piggyback IOLs is unknown. Interpseudophakos Elschnig pearl formation has been observed in 1 of 14 eyes with piggyback plate-haptic silicone IOLs, but to date no hyperopic shift has been observed in that series (Harry Grabow, MD, personal communication, December 1998). Therefore, the complication of late Elschnig pearl ingrowth is not material specific, and late hyperopic shift may be common to all styles of looped lenses implanted in a piggyback configuration. The 3 patients in this series, having spent many years with high hyperopic correction, accepted the refractive outcome and were pleased with their vision after appropriate spectacle prescription. All patients receiving piggyback IOLs should be counseled about the possibility of Elschnig pearl ingrowth and late hyperopic shift.
More worrisome than the refractive shift is the possibility of Elschnig pearl proliferation in the interface between lenses, causing a reduction in BCVA. Fortunately, to date this complication has not been observed in our combined experience, with all interface material remaining outside the central 4 mm region. Should Elschnig pearl proliferation impinge upon the visual axis, it would not be amenable to Nd:YAG laser treatment and would require surgical aspiration of the cellular material. This could be technically difficult, particularly in eyes with piggyback acrylic IOLs, since the capsule firmly adheres to the IOL optic material postoperatively. If it became necessary during the late postoperative period, deliberately fashioning the capsulorhexis larger in diameter than the IOL optic or placing the more anterior IOL optic in the ciliary sulcus are 2 approaches that would allow subsequent surgical access to the space between IOLs. Intuitively, a capsulorhexis diameter larger than the IOL optics should serve to sequester the interface between the optics from proliferating Elschnig pearls better than interposing the anterior capsule remnant between the optics, as would occur with sulcus implantation with a standard-size capsulorhexis. For this reason, we currently favor the larger capsulorhexis approach, but long-term results with this technique will not be available for several years.
Since the long-term incidence of this complication in eyes with piggyback IOLs is unknown, it is recommended that piggyback IOL implantation be approached with caution. This is particularly true when excellent alternatives to piggyback implantation exist, such as in eyes requiring less than +30 D of total IOL power. Extreme caution is also urged in younger patients and in highly elective situations, such as clear lensectomy for hyperopia, in which spectacle or contact lens correction is a viable alternative to surgery. The possibility of this complication occurring during the late postoperative period had been postulated when the initial case series on multiple acrylic IOL implantation was reported,3 and this potential risk has been considered in patient selection for piggyback implantation. However, truly nanophthalmic eyes are prone to uveal effusion and spontaneous suprachoroidal hemorrhage during cataract surgery,5–7 so minimizing wound size by using foldable acrylic IOLs in a piggyback configuration may currently represent the best option for this group of patients. Piggyback IOL implantation can be particularly helpful in special situations such as after penetrating keratoplasty9 or in keratoconus.10 Further study is needed to better define the role of piggyback IOL implantation8.
1. Gayton JL, Sanders VN. Implanting two posterior chamber intraocular lenses in a case of microphthalmos. J Cataract Refract Surg 1993; 19:776-777
2. Holladay JT, Gills JP, Leidlein J, Cherchio M. Achieving emmetropia in extremely short eyes with two piggyback posterior chamber intraocular lenses. Ophthalmology 1996; 103:1118-1123
3. Shugar JK, Lewis C, Lee A. Implantation of multiple foldable acrylic posterior chamber lenses in the capsular bag for hyperopia. J Cataract Refract Surg 1996; 22:1368-1372
4. Fenzl RE, Gills JP, Cherchio M. Refractive and visual outcome of hyperopic cataract cases operated on before and after implementation of the Holladay II formula. Ophthalmology 1998; 105:1759-1764
5. Masket S. Piggyback intraocular lens implantation. J Cataract Refract Surg 1998; 24:569-570
6. Brockhurst RJ. Nanophthalmos with uveal effusion; a new clinical entity. Arch Ophthalmol 1975; 93:1289-1299
7. Ryan EA, Zwaan J, Chylack LT Jr. Nanophthalmos with uveal effusion; clinical and embryologic considerations. Ophthalmology 1982; 89:1013-1017
8. Jin JC, Anderson DR. Laser and unsutured sclerotomyin nanophthalmos. Am J Ophthalmol 1990; 109:575-580
9. Gayton JL. Secondary implantation of a double intraocular lens after penetrating keratoplasty. J Cataract Refract Surg 1998; 24:281-282
10. Gills JP. Piggyback minus-power lens implantationin keratoconus. J Cataract Refract Surg 1998; 24:566-568