Safety and Efficacy of Colored Iris

Reconstruction Lens Implantation

HONGXING WANG, JOOYEON JUNG, SHAWN R. LIN, MICHAEL D. OLSON, AND KEVIN M. MILLER

P

PURPOSE: We sought to evaluate the 1- to 9-year safety ATIENTS WITH PARTIAL OR TOTAL ABSENCE OF IRIS
and efficacy of colored iris reconstruction lens implanta- tissue from trauma, surgery, severe inflammation,
tion in eyes with visual disturbances caused by partial or congenital defects often suffer debilitating light
or complete aniridia. and glare sensitivity. In addition, they experience reduced

DESIGN: Prospective, interventional case series. depth of field, reduced contrast sensitivity, and decreased

METHODS: Thirty-eight patients were implanted with overall visual quality and acuity.1 These functional impair-
Ophtec 311 colored iris reconstruction lenses at the Uni- ments, along with the cosmetic concerns of patients with
versity of California, Los Angeles as part of a larger U.S. light-colored irises, can lead to significant quality-of-life is-
Food and Drug Administration clinical trial. Patients in sues and numerous economic, social, and psychological
group 1 lacked corneal pathology. Patients in group 2 pa- consequences.2
tients had corneal pathology, such as endothelial failure, In recent years, artificial iris implants have gained popu-
previous transplants, or scarring. Safety measures larity for treating visual and cosmetic disturbances in these
included loss of corrected distance visual acuity eyes. The first devices were designed by Peter Choyce in the
(CDVA), surgical complications, adverse events, second- 1950s. They are the forerunners of the devices reported
ary interventions, and corneal endothelial cell loss. Effi- herein.3 In 1994, Sundmacher and associates4,5 first
cacy measures included improvement in uncorrected described the implantation of a 2-piece black iris diaphragm
distance visual acuity (UDVA) and subjective visual intraocular lens (IOL), comprised of a clear polymethyl
disturbances. methacrylate (PMMA) IOL and a black PMMA annulus.

RESULTS: Groups 1 (n [ 8) and 2 (n [ 30) showed A modified PMMA capsular tension ring with a black sec-
improvements in CDVA (P [ .155 and .038), UDVA toral occluder was subsequently developed by Volker Rasch
(P [ .002 and P < .001), and subjective visual distur- and Kenneth J. Rosenthal.6 Various modifications have
bance scores at year 3. Median CDVA and UDVA been made to these devices since the early years.7,8 In the
declined slightly for both groups after 1-2 years. Group United States, artificial iris implants have been available
2 experienced more adverse events, surgical complica- through Compassionate Use Device Exemptions issued by
tions, and secondary interventions. Endothelial cell loss the U.S. Food and Drug Administration (FDA) for several
was greater for group 2 (19.7%) than group 1 (8.05%), decades.9–14 Morcher GmbH (Stuttgart, Germany)
although this difference was not statistically significant produces a variety of black iris diaphragm intraocular
(P [ .067). lenses with clear central optics of various diameters, a

CONCLUSIONS: Colored iris reconstruction lens im- black iris diaphragm, and haptics with suture-fixation islets.
plantation improved CDVA, UDVA, and subjective vi- HumanOptics AG (Erlangen, Germany) produces the
sual disturbances 3 years postoperatively and beyond. CustomFlex artificial iris, which is the first artificial device
Adverse events, complications, and subsequent declines to be approved by the FDA. Their device lacks a central op-
in visual acuity were common, however, in these eyes tic. Reper (Nizhny Novgorod, Russia) also manufactures
with complex medical and surgical histories. (Am J custom artificial irises.
Ophthalmol 2020;216:174–185. Ó 2020 The The first clinical trial of an artificial iris in the United
Author(s). Published by Elsevier Inc. This is an open States was initiated by Ophtec USA (Boca Raton, Florida)
access article under the CC BY-NC-ND license (http:// in 2002. Its model 311 iris reconstruction lens is a rigid 2-
creativecommons.org/licenses/by-nc-nd/4.0/).) piece device with an opaque, 9-mm colored outer ring, 2 in-
tegrated C-loop haptics, and a separate 4.0-mm clear cen-
tral optic. The optic is available in a range of powers
(Figure 1). The device is manufactured from clinical-
Accepted for publication Mar 24, 2020.
From the Stein Eye Institute and the Department of Ophthalmology, grade, ultraviolet-absorbing PMMA and is available in
David Geffen School of Medicine at UCLA, Los Angeles, California, brown, blue, and green. The dioptric powers include 0.0
USA. Dr. Wang is now affiliated with the Department of diopter (D; Plano) and þ10.0 D to þ30.0 D in 0.5-D incre-
Ophthalmology, Beijing Chuiyangliu Hospital, Tsinghua University,
Beijing, Peoples Republic of China. ments. This iris reconstruction lens can be suture-fixated to
Inquiries to Kevin M. Miller, Kolokotrones Chair in Ophthalmology, the sclera or passively fixated in the ciliary sulcus or
Stein Eye Institute, University of California, Los Angeles, 300 Stein capsular bag. The 1-year follow-up report from the phase
Plaza, Los Angeles, CA 90095-7000, USA; e-mail: [email protected]

174 © 2020 THE AUTHOR(S). PUBLISHED BY ELSEVIER INC. 0002-9394

https://doi.org/10.1016/j.ajo.2020.03.036
FIGURE 1. Ophtec model 311 colored iris reconstruction lenses.

1 trial of the Ophtec device showed significant improve- study was scheduled for 1-year follow-up and excluded
ments in both uncorrected distance visual acuity eyes with corneal pathology. After the study commenced,
(UDVA) and subjective visual disturbances in 10 eyes the majority of patients presenting to study sites had
with partial or total iris defects.15 This phase had a limited corneal pathology, so the study was amended to include
patient enrollment and follow-up duration. The study was corneal pathology and the FDA extended the follow-up in-
expanded shortly thereafter to include multiple investiga- terval to 3 years.
tors at multiple locations. At UCLA, we recruited 40 patients who underwent sur-
We aim to document the 1- to 9-year safety and efficacy gical implantation of an Ophtec model 311 iris reconstruc-
of Ophtec 311 colored iris reconstruction lens implanta- tion lens between December 2002 and December 2009.
tion by evaluating the outcomes of 38 patients implanted Two patients were excluded, 1 who was lost to follow-up af-
at the University of California, Los Angeles (UCLA) dur- ter 3 months and another who died 1 month after surgery
ing phase 3 of the trial and followed to their last recorded from leukemia that had not been diagnosed before surgery.
visit, with a focus on the effects of ocular comorbidities The remaining 38 patients (38 eyes) completed > _1 year of
on the clinical course. Although this device was denied follow-up. The patients were divided into 2 groups by study
FDA approval for largely regulatory reasons, it remained design as mentioned previously based on pre-existing
available in Europe and elsewhere in the world. This report corneal comorbidity. Patients in group 1 (n ¼ 8) had clear
will provide valuable safety and efficacy information for corneas, while patients in group 2 (n ¼ 30) had corneal pa-
current and future generations of surgeons who work with thologies, such as corneal endothelial failure, previous
artificial irises and the device manufacturers who produce corneal transplantation, or visually significant scarring. Pa-
them. tients who were enrolled before the study was opened to
subjects with corneal pathology were followed for 1 year.
All other patients were followed for 3 years. No attempt
was made to segregate groups based on comorbidities other
PATIENTS AND METHODS than corneal pathology.
Patients were examined preoperatively, intraoperatively,
THIS PROSPECTIVE INTERVENTIONAL CASE SERIES WAS and postoperatively at day 1; week 1; months 1, 3, and 6;
approved by the UCLA Institutional Review Board. Chart years 1, 2, and 3; and at the last recorded visit. At each visit
review and data collection were conducted in compliance up to year 3, patients completed a survey grading the
with the Health Insurance Portability and Accountability severity of their visual disturbances including daytime
Act of 1996. Written informed consent for participation and nighttime glare, starbursts, and photophobia, scoring
in the clinical trial, surgical management, and publication 0 ¼ none, 1 ¼ mild, 2 ¼ moderate, or 3 ¼ severe. In addi-
of medical information including photographs was ob- tion to demographic information, recorded data included
tained from all participants. the operated eye, ocular comorbidities, cause of the iris
UCLA was 1 study site in a larger, multicenter FDA defect, and extent of the iris defect as 0-24%, 25-49%,
investigational device trial. The larger study did not pub- 50-74%, or 75-100%. Postoperative data collection
lish its final findings, so we decided to publish ours sepa- included UDVA, corrected distance visual acuity
rately because the data are clinically relevant to the (CDVA), intraocular pressure (IOP), endothelial micro-
general topic of artificial iris implantation. The original scopy, surgical complications, adverse events, and

VOL. 216 COLORED IRIS RECONSTRUCTION LENS IMPLANTATION 175

secondary interventions. Safety was assessed by loss of
CDVA, surgical complications, adverse events, secondary TABLE 1. Demographic Information and Length of Follow-
interventions, and reduction in corneal endothelial cell Up
count (ECC) as assessed by a Konan CellChek XL (Konan
Group 1 Group 2 Total
Medical USA, Irvine, California, USA) specular micro-
scope. Cell counts could not always be obtained, especially n (%) 8 (21.1) 30 (78.9) 38 (100)
in group 2. Efficacy was assessed by improvement in UDVA Age (y)
and subjective visual disturbance scores. Mean 6 SD 66.8 6 14.6 55.5 6 13.2 57.9 6 14.2
Median (range) 70 (29-78) 54 (26-79) 56 (26-79)
Details of the surgeries varied considerably from patient
Sex, n (%)
to patient based on pre-existing ocular pathologies. Gen-
Male 8 (100) 21 (70.0) 29 (76.3)
eral anesthesia was administered most of the time, but local Female 0 (0) 9 (30.0) 9 (23.7)
anesthesia with a retrobulbar block was administered occa- Race/ethnicity, n (%)
sionally. Scleral pockets were fashioned as needed. Cata- White 8 (100) 18 (60.0) 26 (68.4)
ract, if present, was removed by the phacoemulsification Hispanic 0 (0) 5 (16.7) 5 (13.2)
technique. Anterior or pars plana vitrectomy was Asian 0 (0) 3 (10.0) 3 (7.9)
performed as needed. Corneal trephination was performed Other 0 (0) 4 (13.3) 4 (10.5)
when corneal scarring was significant. Otherwise, an 11- Follow-up (y), n (%)
mm limbal incision was fashioned to accommodate the 1 8 (100) 30 (100) 38 (100)
nonfolding Ophtec 311 implant. IOL removal was 3 7 (87.5) 26 (86.7) 33 (86.8)
6 3 (37.5) 18 (60.0) 21 (55.3)
performed, as needed. The Ophtec 311 was passively
9 1 (12.5) 5 (16.7) 6 (15.8)
fixated in the ciliary sulcus if there was adequate capsule
Mean 6 SD 4.2 6 1.9 5.2 6 2.0 5.0 6 2.0
and zonular support; otherwise, it was fixated to the sclera
with 9-0 Prolene sutures. Gore-Tex sutures (W. L. Gore SD ¼ standard deviation.
& Associates, Elkton, Maryland, USA) were not available
at this time. Finally, a replacement corneal graft was su-
tured in place, if planned, or the limbal incision was closed
with interrupted 10-0 nylon sutures. 2 was blunt or penetrating trauma (73.3%). Group 2 had
For statistical analyses, Snellen acuities were converted significantly more preoperative ocular comorbidity,
to logarithm of minimal angle of resolution (logMAR) including previous corneal transplantation (43.3%), previ-
values. The following logMAR conversions were used for ous retinal detachment repair (36.7%), and glaucoma
nonnumeric visual acuity values: count fingers (CF) ¼ (30.0%). There was no view of the fundus for 10 eyes in
1.7, hand motion (HM) ¼ 2.0, light perception (LP) ¼ group 2 secondary to ocular pathologies, such as dense
2.3, and no light perception (NLP) ¼ 3.0. Visual acuity me- corneal scars and corneal edema. Thirteen eyes in group
dians and means were calculated from logMAR values. Re- 2 required concurrent penetrating keratoplasty.
sults were graphed as median 6 the 25th-75th interquartile Changes in median CDVA after surgery are graphed in
range. Changes in the visual acuity and ECC were analyzed Figure 2. The error bars represent the 25th and 75th percen-
using the Student t test. P values < .05 were considered sta- tiles. Error bars were not plotted for the total. Loss of CDVA
tistically significant. was the primary safety outcome of the study. Over 3 years,
group 1 experienced a 27.6% improvement in logMAR
CDVA (P ¼ .155) and group 2 experienced a 35.7%
improvement (P ¼ .038). The overall improvement for
RESULTS both groups was 34.1% (P ¼ .027). The greatest improve-
ment occurred between postoperative months 3 and 6 for
THE STUDY ANALYZED OUTCOMES FROM 38 PATIENTS, group 1 and between postoperative months 1 and 3 for
including 29 men and 9 women, who ranged in age from group 2. Both groups experienced a slight worsening of
26 to 79 years. Table 1 shows demographic information. CDVA after years 1 and 2, respectively. At the last visit,
The minimum follow-up was 1 year. The mean follow-up in- no patient in group 1 experienced a decrease in CDVA of
terval, defined as the interval between surgery and the last >
_2 lines compared with baseline and 1 patient experienced
recorded visit, was 5.0 6 2.0 years. Several patients were an improvement from 20/80 at baseline to 20/25 at 8.5 years
followed many years after the conclusion of the formal study. postoperatively. In group 2, 7 patients experienced a
Table 2 shows the preoperative condition of the study decrease in CDVA of > _2 lines at their last visit 3-8.5 years
eyes. All 8 eyes in group 1 had iris defects measuring postoperatively and all but 1 was a result of corneal graft
<50%, whereas nearly half of the eyes in group 2 failure or rejection. For the 1 patient who was the excep-
(46.7%) had defects involving >50% of the iris. The tion, the loss of CDVA from CF at baseline to NLP at the
most common cause of iris defect for group 1 was surgical 2-year follow-up examination was caused by protracted
trauma (62.5%), whereas the most common cause for group intraocular inflammation and end-stage glaucoma. By

176 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2020

 TABLE 2. Preoperative Information

Group 1 (n ¼ 8) Group 2 (n ¼ 30) Total (N ¼ 38)

Operative eye, n (%)

Right 4 (50.0) 13 (43.3) 17 (44.7)
Left 4 (50.0) 17 (56.7) 21 (55.3)
Extent of iris defect, n (%)
0-24% 7 (87.5) 6 (20.0) 13 (34.2)
25-49% 1 (12.5) 10 (33.3) 11 (28.9)
50-74% 0 (0) 9 (30.0) 9 (23.7)
75-100% 0 (0) 5 (16.7) 5 (13.2)
Etiology of iris defect, n (%)
Blunt or penetrating trauma 3 (37.5) 22 (73.3) 25 (65.8)
Surgical trauma 5 (62.5) 6 (20.0) 11 (28.9)
Congenital 0 (0) 1 (3.3) 1 (2.6)
Other 0 (0) 1 (3.3) 1 (2.6)
Lens status, n (%)
Aphakic 4 (50.0) 23 (76.6) 27 (71.1)
Pseudophakic 3 (37.5) 5 (16.7) 8 (21.1)
Cataract 1 (12.5) 2 (6.6) 3 (7.9)
Corneal pathologies, n (%)
Previous corneal transplant 0 (0) 13 (43.3) 13 (34.2)
Epithelial abnormality 0 (0) 6 (20.0) 6 (15.8)
Corneal edema 0 (0) 4 (13.3) 4 (10.5)
Stromal anomaly 0 (0) 3 (10.0) 3 (7.9)
Endothelial defect 0 (0) 2 (6.6) 2 (5.3)
Keratopathy 0 (0) 1 (3.3) 1 (2.6)
Corneal decompensation 0 (0) 1 (3.3) 1 (2.6)
Other ocular comorbidities, n (%)
Previous retinal detachment 1 (12.5) 11 (36.7) 12 (31.6)
Glaucoma 1 (12.5) 9 (30.0) 10 (26.3)
Maculopathy 1 (12.5) 7 (23.3) 8 (21.1)
Synechiae 0 (0) 3 (10.0) 3 (7.9)
Iritis 1 (12.5) 0 (0) 1 (2.6)
Nystagmus 0 (0) 1 (3.3) 1 (2.6)
Other 0 (0) 3 (10.0) 3 (7.9)

comparison, 9 patients in group 2 experienced an improve- the 1-month postoperative visit because of an IOP increase
ment in CDVA of > _2 lines at the last examination 3-6 years from 15 to 22 mm Hg and was maintained on the drop with
postoperatively, with the greatest changes being seen in 3 excellent IOP control. Two patients in group 2 were started
patients whose CDVAs improved from CF at baseline to on timolol and dorzolamide, and another 3 patients were
20/20, 20/30, and 20/50, respectively. started on timolol, dorzolamide, and brimonidine; all but
Figure 3 and Table 3 show how mean IOP varied over 2 of them were eventually taken off drops. One patient in
time. Table 3 contains the P values associated with IOP group 2 began timolol at postoperative day 1 with good
changes at each postoperative time point. The mean preop- IOP control, was then monitored off the drop, but experi-
erative IOP was 14.0 6 3.1 mm Hg and 14.3 6 4.4 mm Hg enced a spike in IOP to 35 mm Hg. The patient was started
for groups 1 and 2, respectively. At the last visit, the mean on travoprost, timolol, dorzolamide, and brimonidine at
IOP was 12.3 6 1.1 mm Hg for group 1 (a 12.9% reduction this time and eventually required tube shunt placement.
compared with the preoperative baseline, P ¼ .15) and 12.6 Two other patients in group 2 underwent tube shunt place-
6 4.5 mm Hg for group 2 (an 11.9% reduction, P ¼ .18). ment. One patient, also in group 2, experienced hypotony
There was no obvious trend toward higher pressures during with an IOP of 2 mm Hg with positive Seidel sign caused by
the study in either group, but several patients were under loose sutures in the corneal graft. The patient underwent
treatment with pressure-lowering medications during the reoperation to resuture the graft with an eventual return
study. One patient in group 1 was started on timolol at of IOP back to baseline.

VOL. 216 COLORED IRIS RECONSTRUCTION LENS IMPLANTATION 177

FIGURE 2. Changes in median corrected distance visual acuity after surgery. Error bars were not added to the total group. CDVA [
corrected distance visual acuity; m [ month; y [ year.

FIGURE 3. Changes in intraocular pressure after surgery. The error bars represent standard deviations. Error bars were not added to
the total group.

Table 4 shows additional safety outcomes. Fifteen eration was performed to suture fixate the device to the
adverse events were reported in 14 patients during the sclera and it remained centered for the duration of the
formal 3-year study (3 in group 1 and 12 in group 2). study. Another patient in group 1 died of pneumonia be-
Only 1 was considered to be device related. Six months af- tween the second- and third-year follow-up examinations.
ter surgery, 1 patient in group 1 presented with a slight infe- During the 3-year formal study, there were 4 intraopera-
rior displacement of the device in the absence of tive complications and 21 postoperative complications. All
antecedent trauma because of zonular dehiscence. A reop- of the intraoperative complications occurred in group 2

178 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2020

 eyes. In 2 instances, pre-existing Ahmed glaucoma valves
impeded proper positioning of the iris reconstruction

12.3 (P ¼ .15,

12.6 (P ¼ .18,
n ¼ 30)
n ¼ 8)
lens. There was 1 instance of scleral in-folding, which

Last Visit
made suture closure of the surgical wound difficult. Finally,
1 eye with an opaque cornea was found to have a retinal
detachment that had not been present on ultrasound exam-
ination several weeks earlier. The detachment was treated

17.0 (P ¼ .63,
N/A (n ¼ 1)
by intraocular gas tamponade and scleral buckling. The

n ¼ 3)
7-9
most common postoperative complications during the
formal study were corneal edema (7 eyes), followed by
increased IOP (5 eyes), iritis (4 eyes), and posterior
capsular opacification (3 eyes).
14.0 (P ¼ .70,

13.1 (P ¼ .34,

During the 3-year formal study, there were 11 secondary

n ¼ 17)
n ¼ 2)

surgical interventions including repeat corneal transplan-

4-6

tation (4 eyes), glaucoma tube shunt placement (2 eyes)

or removal (1 eye), Ophtec 311 repositioning (1 eye), laser
capsulotomy (2 eyes), and anterior chamber washout (1
12.0 (P ¼ .019,
12.4 (P ¼ .25,

eye). There were also 11 nonsurgical interventions, all of

No. of Years

n ¼ 25)
n ¼ 7)
TABLE 3. Mean Intraocular Pressure in mm Hg at Each Study Visit

which consisted of the administration of additional eye

3

drops. One patient in group 1 developed iritis, an epiretinal

membrane, and macular edema 1 year postoperatively. The
cause was felt to be uveal irritation and the eye was treated
with corticosteroid eyedrops, resulting in complete remis-
14.0 (P ¼ 1.0,

12.9 (P ¼ .29,
n ¼ 26)
n ¼ 8)

sion of the iritis and edema. One patient in group 2 devel-

2

oped idiopathic iritis with vitreous hemorrhage at

postoperative week 2. Symptoms completely resolved after
2 months with continued use of postoperative medications.
After the formal study there were another 20 late com-
14.0 (P ¼ .66,

12.7 (P ¼ .21,

plications and an additional 15 late surgical interventions.

n ¼ 30)
n ¼ 7)

Seven patients experienced 10 graft failures (1 patient had

1

multiple repeat grafts) from 4-6.5 years after device implan-

tation, 2 of which were caused by allograft rejection. Dur-
ing this time period, 4 patients underwent a total of 6
12.9 (P ¼ .60,

15.8 (P ¼ .50,

procedures, including 4 penetrating keratoplasties and 2

n ¼ 29)
n ¼ 8)

Descemet stripping endothelial keratoplasties. These

6

events occurred beyond the final 3-year study visit, and

therefore they were not reported to the IRB. No patient
in group 1 experienced corneal complications requiring
13.6 (P ¼ .83,

14.5 (P ¼ .88,

additional surgery during the study.

No. of Months

n ¼ 30)
n ¼ 8)

Specular endothelial microscopy images were obtained

3

when corneal visibility permitted at the preoperative visit

and following surgery at months 6, 12, 24, and 36. A total
of 21 patients (7 in group 1 and 14 in group 2) had > _2 reli-
able specular microscopy images. Figure 4 shows the
16.9 (P ¼ .13,

14.0 (P ¼ .87,
n ¼ 29)
n ¼ 8)

changes in mean ECC over the duration of follow up. On

1

average, there was a 27.7% reduction in ECC per eye be-

tween the first and final specular microscopy measure-
ments. The mean decrease in ECC per year was 8.05%
for group 1 (P ¼ .047), 19.7% for group 2 (P < .001),
N/A ¼ not applicable.
14.3 (n ¼ 30)
14.0 (n ¼ 8)
Preoperatively

and 16.8% overall (P < .001). The difference in mean

percent decrease in ECC per year between groups I and II
was not statistically significant (P ¼ .067).
Changes in median UDVA over time are plotted in
Figure 5. Visits occurring after 3 years fell outside the
Group

formal study and were not subject to the rigorous data

1

collection protocol used during the study. Thus, there is

VOL. 216 COLORED IRIS RECONSTRUCTION LENS IMPLANTATION 179

 TABLE 4. Secondary Safety Outcomes of Colored Iris Reconstruction Lens Implantation

Complications Secondary Interventions

a a a b a
Adverse Events Intraoperatively Postoperatively Late Surgical Nonsurgicala Late Surgicalb Decrease in ECC/Yeara

Group 1 (n ¼ 8) 3 0 4 1 1 2 1 8.05% (P ¼ .047)

Group 2 (n ¼ 30) 12 4 17 19 10 9 14 19.7% (P < .001)
Total (N ¼ 38) 15 4 21 20 11 11 15 16.8% (P < .001)

ECC ¼ endothelial cell count.

a
Collected prospectively during the U.S. Food and Drug Administration clinical trial.
b
Collected retrospectively after patients exited the clinical trial.

FIGURE 4. Changes in mean endothelial cell count after surgery. Error bars were not added to the total group. ECC [ endothelial
cell count; m [ month.

discrepancy between the number of UDVAs and CDVAs Patient T.M. was 29 years of age at the time of colored iris
reported at the 4- to 6-year and 7 to 9-year time points. reconstruction lens implantation. He was assigned to group
Other efficacy measures are tabulated in Table 5. The 1. The photographs shown in Figure 6A were taken preop-
improvement in mean UDVA after 3 years was 54.1% for eratively. His ocular history was notable for blunt trauma to
group 1 (P ¼ .002), 36.3% for group 2 (P < .001), and the right cheek from a golf club 5 years earlier. He under-
40.2% overall (P < .0001). For both groups, there was an went immediate repair of cheek fractures, followed
improvement in all subjective visual disturbance scores 6 months later by a lensectomy and evacuation of intraoc-
compared with preoperative values. No patient reported ular clot. Three subsequent strabismus operations were
worsening of daytime glare, daytime or nighttime star- performed. At the time of study surgery, he was aphakic
bursts, or daytime photophobia. Two patients reported an and he had a markedly deformed iris and pupil. Under our
increase in nighttime glare, which changed from mild to care, he underwent anterior vitrectomy, implantation of a
moderate in both cases. Three patients reported increased green iris reconstruction lens, and partial iridodialysis
nighttime photophobia by 1 scale point. repair. The photographs in Figure 6B were taken 3 months
after surgery. Three years after surgery, his UDVA was 20/

CASE EXAMPLES: The following 3 case examples were 50 and his CDVA with spectacles was 20/20 2.
selected to highlight typical preoperative presentations and Patient P.H. was 48 years of age at the time of colored iris
postoperative clinical courses of patients in groups I and II. reconstruction lens implantation. He was assigned to group

180 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2020

FIGURE 5. Changes in median uncorrected distance visual acuity after surgery. Error bars were not added to the total group. m [
month; UDVA [ uncorrected distance visual acuity; y [ year.

TABLE 5. Mean Subjective Visual Disturbance Scores

Preoperative 3 Months 1 Year 3 Years

Glare (day)
Group 1 2.4 6 0.70 1.5 6 0.50 1.0 6 0.76 1.7 6 0.94
Group 2 2.6 6 0.55 1.5 6 0.67 1.3 6 0.70 1.3 6 0.67
Total 2.6 6 0.59 1.5 6 0.64 1.3 6 0.74 1.3 6 0.76
Glare (night)
Group 1 2.5 6 0.71 1.5 6 0.50 1.0 6 0.53 1.2 6 0.37
Group 2 2.1 6 0.85 1.0 6 0.66 0.9 6 0.68 1.0 6 0.69
Total 2.2 6 0.84 1.1 6 0.66 0.9 6 0.67 1.1 6 0.64
Starbursts (day)
Group 1 1.6 6 1.22 0.6 6 0.70 0.3 6 0.45 0.7 6 0.75
Group 2 1.9 6 1.14 1.0 6 0.89 0.8 6 0.87 0.8 6 0.78
Total 1.8 6 1.16 0.9 6 0.87 0.7 6 0.85 0.8 6 0.77
Starbursts (night)
Group 1 2.3 6 1.30 1.3 6 0.97 0.9 6 0.99 1.2 6 0.90
Group 2 2.2 6 1.08 1.1 6 0.88 1.0 6 0.93 0.9 6 0.78
Total 2.2 6 1.13 1.2 6 0.90 1.0 6 0.94 1.0 6 0.81
Photophobia (day)
Group 1 2.8 6 0.43 1.3 6 0.66 1.4 6 0.73 1.8 6 0.90
Group 2 2.7 6 0.60 1.4 6 0.76 1.3 6 0.77 1.1 6 0.80
Total 2.7 6 0.57 1.4 6 0.74 1.3 6 0.78 1.3 6 0.87
Photophobia (night)
Group 1 2.0 6 0.87 1.3 6 0.66 1.4 6 0.73 0.8 6 0.90
Group 2 1.9 6 0.92 1.0 6 0.68 0.8 6 0.65 1.0 6 0.78
Total 1.9 6 0.91 1.1 6 0.69 0.9 6 0.71 1.0 6 0.81

2. The photographs shown in Figure 7A were taken preop- and went through his eye, traumatizing the retina in the
eratively. His ocular history was notable for a nail injury to temporal macula. He lost his iris and lens in the process.
the left eye 35 years earlier. The nail penetrated the cornea He underwent primary repair of the globe injury followed

VOL. 216 COLORED IRIS RECONSTRUCTION LENS IMPLANTATION 181

FIGURE 6. (A) Preoperative and (B) 3-month postoperative composite photographs of a 29-year-old man with a history of blunt
trauma to his right cheek 5 years earlier from a golf club.

FIGURE 7. (A) Preoperative and (B) 3-month postoperative composite photographs of a 48-year-old man who suffered a nail injury
to the left eye 35 years earlier.

by corneal transplantation 20 years later. He had strabismus orbital fractures. After primary repair of the rupture and
surgery subsequently and an unrelated operation for a right fractures, his examination was notable for aphakia with
cholesteatoma, which left him with a partial right VII cra- some retained lens capsule, nearly complete aniridia, and
nial nerve paresis and poor orbicularis oculi tone. Under our corneal scarring. Under our care, he underwent penetrating
care, he underwent repeat penetrating keratoplasty and keratoplasty, residual lensectomy, anterior vitrectomy, and
scleral suture fixation of a green iris reconstruction lens. scleral suture fixation of a brown iris reconstruction lens.
The photographs in Figure 7B were taken 3 months after The photographs in Figure 8B were taken 3 months after
surgery. Three years after surgery, his UDVA was 20/50 surgery. Three years after surgery, his UDVA was 20/
and his CDVA with spectacles was 20/20 2. 60 1 and his CDVA with spectacles was 20/50 þ1. The
Patient K.G. was 35 years of age at the time of colored iris remaining corneal sutures were removed and he was
reconstruction lens implantation. He was assigned to group referred for rigid contact lens fitting.
2. The photographs shown in Figure 8A were taken preop-
eratively. His ocular history was notable for radial and
astigmatic keratotomy in the left eye. He had been
assaulted 4 years earlier. The blunt injury to his left globe DISCUSSION
resulted in globe rupture and expulsion of intraocular con-
THE OPHTEC 311 COLORED IRIS RECONSTRUCTION LENS IS A
tents including the iris and lens. He also suffered several left
2-piece implant that is designed to correct visual

182 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2020

FIGURE 8. (A) Preoperative and (B) 3-month postoperative composite photographs of a 35-year-old man who was assaulted 4 years
earlier. The blunt injury to his left globe resulted in globe rupture and expulsion of intraocular contents including the iris and lens.

disabilities associated with partial or complete aniridia. It (n ¼ 8). This discrepancy is related to the higher preva-
consists of an opaque 9.0-mm diaphragm and a central, lence of uncontrolled blunt or penetrating injury in group
fixed 4.0-mm optic, which blocks excess light from entering 2 as compared with the surgical trauma sustained by most
the eye while allowing for examination of the peripheral patients in group 1.
retina via indirect ophthalmoscopy. The 2004 paper by The primary safety measure of the study was loss of
Price and associates,15 which reported 1-year outcomes of CDVA. Patients in both groups showed improvements in
a phase I clinical trial of the Ophtec device in the United mean CDVA 3 years postoperatively compared with base-
States, showed improvements in UDVA, visual distur- line, although only group 2 patients achieved statistical sig-
bances, and cosmetic appearance after device implanta- nificance. The worsening of CDVA observed in both
tion. In 2007, Miller and associates16 confirmed those groups at 1 and 2 years, respectively, represented progres-
results in a 3-year follow-up study of 9 eyes that underwent sion of ocular comorbidities. Group 2 patients showed a
combined penetrating keratoplasty and Ophtec 311 im- relatively greater loss. Eleven patients experienced corneal
plantation. All of the patients in the 2007 report are decompensation and 1 patient with advanced retinitis
included in this long-term follow-up study. pigmentosa developed cystoid macular edema. Progression
Two other sets of artificial iris devices are similar to the of glaucoma may have played a partial role in 3 patients. It
Ophtec 311. One is a series of black iris diaphragm IOLs is possible that device implantation accelerated these
manufactured by Morcher GmbH. Only 1 prospective study changes, but there is no way to know from the study design
has been published on these devices to date, specifically the because there was no control group. Despite the overall
model 67B. The authors of the study found that black iris improvement in CDVA experienced by patients as a
diaphragm IOL implantation in aphakic eyes with large whole, 2 patients (5.26%) lost CDVA compared with their
iris defects was relatively safe and effective at improving preoperative status by 3 years. This is a low number given
CDVA and reducing light and glare sensitivity.17 That the trauma history and comorbidity load observed in the
study had a follow-up interval of 1 year. The results of the study population. Other safety measures, including surgical
current study are similar in terms of overall outcomes at 1 complications, adverse events, and secondary interven-
year. The other is a series of acrylic artificial iris devices tions, were more prevalent in group 2, as expected.
with integrated optics made by Reper. There are only anec- Efficacy measures included improvements in UDVA and
dotal reports of this device in the published literature. subjective visual disturbance scores. By 3 years postopera-
The current report focuses on 1- to 9-year outcomes of 38 tively, both groups achieved a statistically significant
patients who underwent implantation of the Ophtec 311 improvement in UDVA with group 1 showing a greater
device at the Stein Eye Institute between 2002 and 2009. improvement than group 2. Unlike CDVA, the inflection
These patients were part of a larger US FDA clinical trial. in the graph of median UDVA over time was seen much
The patients were divided into 2 groups based on the pres- earlier and to a greater degree than in group 1. This may
ence of preoperative corneal pathology. Group 2 patients be attributable in part to fluctuations in regular and irreg-
(n ¼ 30), who had pre-existing corneal pathology, had ular corneal astigmatism, which ranged from þ0.25 D
significantly greater iris involvement and a higher inci- to þ10.50 D. Another efficacy measure, the subjective vi-
dence of other ocular comorbidities than those in group 1 sual disturbance score, showed an overall improvement in

VOL. 216 COLORED IRIS RECONSTRUCTION LENS IMPLANTATION 183

daytime and nighttime glare, starbursts, and photophobia of the process, the FDA asked to inspect the facility
for both groups 3 years postoperatively. where the raw materials were manufactured. That facil-
Although cosmesis was not evaluated as part of this ity was closed years earlier, but considerable material
study, it should be noted that 1 patient in group 2 requested was warehoused for future device manufacture. Because
explantation of his colored iris reconstruction lens 1 year the FDA was unable to inspect the manufacturing fa-
postoperatively because of cosmetic concerns. As previ- cility, the Ophtec PMA application request was de-
ously described, this device is only available as a flat, untex- nied. The device remained CE marked. Ophtec BV
tured diaphragm in 1 of 3 colors: brown, blue, or green. This more recently announced at the 2019 meeting of the
necessarily limits the extent to which the device can match European Society of Cataract and Refractive Surgeons
the natural color and texture of the iris in the fellow eye. that they would discontinue sales of the Ophtec 311
Other concerns with the implant are its relatively small on December 31, 2019 and begin sales of the Reper
outer diameter of 9 mm and relatively large pupil diameter. acrylic artificial iris instead on January 1, 2020.
Recently, the CustomFlex, a custom-printed artificial iris Despite the cessation of future sales, lessons learned
prosthesis developed by HumanOptics AG (Erlangen, Ger- from this clinical trial should benefit patients already
many), received FDA premarket approval (PMA).18 While implanted and provide guidance and benchmarking
the CustomFlex device is foldable and offers a more natural data against which to compare future artificial iris
and aesthetic alternative to the Ophtec model 311, factors designs.
such as cost and the lack of a built-in IOL need to be In summary, the Ophtec model 311 colored iris recon-
considered when comparing and contrasting different types struction lens was found to be relatively safe and effective
of iris implants. With the HumanOptics device, a lens must at improving visual acuity and subjective visual distur-
be implanted separately or sutured to the iris if the eye to be bances caused by partial or total iris defects. As evidenced
implanted will be aphakic at the time of implantation. In by the modest percentage of eyes that experienced adverse
addition, the HumanOptics device has been implicated safety events, the risk of performing complicated surgery on
in residual iris retraction syndrome if a fiber-containing eyes with significant comorbidities must be thoroughly
model is trephined and contacts uveal tissue.19 assessed. The potential benefits of iris reconstruction lens
Data from the 3-year Ophtec 311 clinical trial were implantation should be carefully weighed against the possi-
presented to the FDA in a PMA application. As part bility of complications and loss of CDVA.

ALL AUTHORS HAVE COMPLETED AND SUBMITTED THE ICMJE FORM FOR DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST.
Funding/Support: The authors indicate no financial support. Financial Disclosures: Dr Miller is a consultant for Alcon Research and Johnson & Johnson
Vision. All other authors indicate no financial conflict of interest. All authors attest that they meet the current ICMJE criteria for authorship.

REFERENCES tions, and techniques. J Cataract Refract Surg 2000;26(6):

898–912.
1. Slamovits TL, Glaser JS, Mbekeani J. The pupils and accom- 8. Srinivasan S, Ting DS, Snyder ME, Prasad S, Koch HR. Pros-
modation. In: Tasman W, Jaeger E, eds. Duane’s Clinical thetic iris devices. Can J Ophthalmol 2014;49(1):6–17.
Ophthalmology, 2 (Chapter 15). Philadelphia, PA: Lippin- 9. Osher RH, Burk SE. Cataract surgery combined with implan-
cott Williams & Wilkins; 2013. tation of an artificial iris. J Cataract Refract Surg 1999;25(11):
2. Mayer CS, Reznicek L, Hoffmann AE. Pupillary reconstruc- 1540–1547.
tion and outcome after artificial iris implantation. Ophthal- 10. Tanzer DJ, Smith RE. Black iris-diaphragm intraocular lens
mology 2016;123(5):1011–1018. for aniridia and aphakia. J Cataract Refract Surg 1999;
3. Choyce P. Intraocular lenses and implants. London, UK: 25(11):1548–1551.
Lewis and Co. Ltd.; 1964. 11. Thompson CG, Fawzy K, Bryce IG, Noble BA. Implantation
4. Sundmacher T, Reinhard T, Althaus C. Black diaphragm of a black diaphragm intraocular lens for traumatic aniridia. J
intraocular lens in congenital aniridia. Ger J Ophthalmol Cataract Refract Surg 1999;25(6):808–813.
1994;3(4-5):197–201. 12. Reinhard T, Engelhardt S, Sundmacher R. Black diaphragm
5. Sundmacher R, Reinhard T, Althaus C. Black-diaphragm aniridia intraocular lens for congenital aniridia: long-term
intraocular lens for correction of aniridia. Ophthalmic Surg follow-up. J Cataract Refract Surg 2000;26(3):375–381.
1994;25(3):180–185. 13. Mavrikakis I, Casey JM. Phacoemulsification and endocapsu-
6. Rosenthal K. A Tale of 2 Irides: Contrasting Treatment of 2 Pa- lar implantation of an artificial iris intraocular lens in trau-
tients with Iatrogenic Iridodialysis [film]. Presented at the Amer- matic cataract and aniridia. J Cataract Refract Surg 2002;
ican Society of Cataract and Refractive Surgery Symposium, 28(7):1088–1091.
San Diego, California, April 27–May 2, 2007. 14. Karatza EC, Burk SE, Snyder ME, Osher RH. Outcomes of
7. Menapace R, Findl O, Georgopoulos M, Rainer G, Vass C, prosthetic iris implantation in patients with albinism. J Cata-
Schmetterer K. The capsular tension ring: designs, applica- ract Refract Surg 2007;33(10):1763–1769.

184 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST 2020

15. Price MO, Price FW, Chang DF, Kelley K, Olson MD, with large iris defects: report 4. J Cataract Refract Surg 2018;
Miller KM. Ophtec iris reconstruction lens United States clin- 44(6):686–700.
ical trial phase I. Ophthalmology 2004;111(10):1847–1852. 18. Rana M, Savant V, Prydal JI. A new customized artificial iris
16. Miller AR, Olson MD, Miller KM. Functional and cosmetic diaphragm for treatment of traumatic aniridia. Contact Lens
outcomes of combined penetrating keratoplasty and iris Anterior Eye 2013;36(2):93–94.
reconstruction lens implantation in eyes with a history of 19. Mayer CS, Laubichler AE, Masyk M, Prahs P, Zapp D,
trauma. J Cataract Refract Surg 2007;33(5):808–814. Khoramnia R. Residual iris retraction syndrome after
17. Miller KM, Kuo A, Olson MD, Masket S. Safety and efficacy artificial iris implantation. Am J Ophthalmol 2019;199:
of black iris diaphragm intraocular lens implantation in eyes 159–166.

VOL. 216 COLORED IRIS RECONSTRUCTION LENS IMPLANTATION 185