Evaluation of Intrastromal Lipid Deposits After
Intacs Implantation Using In Vivo Confocal
Linda T. Ly, B.S., James P. McCulley, M.D., Steven M. Verity, M.D.,
H. Dwight Cavanagh, M.D., Ph.D., R. Wayne Bowman, M.D., and
W. Matthew Petroll, Ph.D.
Purpose. To assess the structure and location of intrastromal lipid
deposits after implantation of Intacs by using in vivo confocal
microscopy. Methods. Seven eyes of six patients were examined
by in vivo confocal microscopy 5 years (n ? 6) or 2 months (n ?
1) after uncomplicated implantation of Intacs for the correction of
mild myopia. Selected images from all corneal layers were qual-
itatively evaluated for structural changes, with special attention
paid to areas surrounding the Intacs implants. Results. In the
peripheral cornea of eyes examined 5 years after surgery, epithelial
and endothelial cell layers appeared normal. Tandem scanning
confocal microscopy showed stromal haze surrounding the im-
plants in all eyes examined, but no keratocyte activation was seen.
Reflective amorphous or crystalline structures consistent with lipid
deposition were detected in all eyes with long-term implantation of
Intacs. Deposits were localized to the inner and outer edges of
Intacs segments and to the region anterior to the implant. Confocal
microscopy did not show any deposits in the eye examined 2
months after surgery, although the region anterior to the implant
appeared hazy and edematous. Areas central to the implant ap-
peared normal in all eyes. Conclusions. The mechanical and
physiologic stresses introduced by the implantation of Intacs lead
to the accumulation of lipid deposits in the extracellular matrix. By
using in vivo confocal microscopy, the location and structure of
these deposits can be determined.
Key Words: Cornea—Confocal microscopy—Intacs—Wound
Intracorneal ring segments, now called Intacs (Addition Technol-
ogy, Inc., Des Plaines, IL), are two microthin, 150° crescent-
shaped segments made of polymethylmethacrylate (PMMA) used
to correct mild to moderate degrees of myopia. Intacs are surgi-
cally placed at approximately two thirds of the stromal depth in the
mid peripheral cornea and work to reshape the curvature of the
anterior surface while maintaining prolate corneal asphericity.1,2
Insertion of the ring segments does not necessitate removal of
corneal tissue from the central optical zone, unlike most other
refractive surgeries. The degree of correction with Intacs depends
on the thickness of segments, ranging from 0.25 to 0.45 mm in
0.05-mm increments. Currently, only the 0.25- to 0.35-mm seg-
ments have been approved by the U.S. Food and Drug Adminis-
tration for the correction of low myopia.
Early findings have generally suggested that implantation of
Intacs in the corneal stroma offers a safe and effective means to
reducing refractive error.3,4A long-term study that assessed the
efficacy and safety of Intacs in patients enrolled in U.S. Food and
Drug Administration clinical trials found that mild myopia was
reduced or eliminated in most cases, and segments were generally
well tolerated by patients.4Moreover, Chan and Khan5showed
that refraction and visual acuity essentially return to preoperative
levels with the removal of Intacs, and segments can be exchanged
successfully. These findings are consistent with a recent study that
showed preoperative refractive error was rapidly reestablished in
most patients after the removal of Intacs segments.6Thus, the near
reversibility of the procedure allows for the implants to be replaced
or removed if indicated by a change in visual needs or if the
removal is requested by the patient. In addition to myopia, Intacs
have also been used to treat patients with keratoconus7,8and
iatrogenic corneal ectasia after excimer laser surgery.9
Although serious complications from the Intacs procedure are
uncommon, reports of visual problems after implantation of the
ring segments have been noted and include mild discomfort,
photophobia, and optical aberrations.3Bourges et al.10reported a
case of anterior stromal necrosis of the cornea associated with
implantation of Intacs in a woman 5 years after an uneventful
surgical insertion. Earlier studies have reported the appearance of
extracellular intrastromal lipid deposits along the lamellar chan-
nels of implants,11–13and clinical characteristics of these deposits
after Intacs implantation in humans have been described.11Similar
intrastromal corneal deposits have been observed in primates after
From the University of Texas Southwestern Medical School (L.T.L.) and
the Department of Ophthalmology (J.P.M., S.M.V., H.D.C., R.W.B.,
W.M.P.), University of Texas Southwestern Medical Center, Dallas, TX.
Supported in part by NIH grants R01 EY13322 (W.M.P.) and infrastruc-
ture grant EY016664, a Lew R. Wasserman Merit award (W.M.P.), and an
unrestricted grant from Research to Prevent Blindness, Inc.
Presented in part at the American Society of Cataract and Refractive
Surgery Symposium and Congress (April 15–19, 2005) and the Association
for Research in Vision and Ophthalmology annual meeting (May 1–5,
Address correspondence and reprint requests to Dr. W.M. Petroll,
Department of Ophthalmology, University of Texas Southwestern Medical
Center, 5323 Harry Hines Boulevard, Dallas, TX 75390-9057; e-mail:
Accepted October 12, 2005.
Eye & Contact Lens 32(4): 211–215, 2006
© 2006 Contact Lens Association of Ophthalmologists, Inc.
the insertion of intracorneal lenses synthesized from PMMA or
polysulfone and hydrogel implants.14,15Although it is believed
that these deposits are not associated with anatomic or physiologic
corneal deterioration11and resolve once Intacs segments are
removed,6their formation suggests an active tissue repair
response.16A previous confocal study showed the appearance of
intrastromal deposits in eyes evaluated 2 to 15 months after Intacs
implantation.12In the current study, in vivo confocal microscopy
was used to evaluate the structure and location of lamellar channel
lipid deposits in the cornea of human eyes 2 months and 5 years
MATERIALS AND METHODS
Seven eyes of six patients (three men and three women) with
mild myopia were included in this retrospective study. At the time
of surgery, patients were between 22 and 57 years of age (mean,
45 ? 13 years). Spherical equivalents ranged from –1.75 diopters
(D) to –3.75 D, with up to 1.00 D of astigmatism. Implantation of
the Intacs segments was performed by three experienced surgeons
at the University of Texas Southwestern Medical Center at Dallas
(J.P.M., R.W.B., S.M.V.). The details of this surgical procedure
have been described previously.17The thicknesses of the implants
ranged from 0.25 to 0.35 mm. No complications related to the
Intacs procedure occurred.
Six eyes were examined with in vivo confocal microscopy 5
years after Intacs implantation, and one eye was examined 2
months postoperatively. The 5-year patients had channels created
mechanically, whereas the 2-month patient had channels created
with a femtosecond laser (IntraLase).
In Vivo Confocal Microscopy
A tandem scanning confocal microscope (model 165A, Tandem
Scanning Corporation, Reston, VA) with a 24? surface-contact
objective (0.6 NA, 1.5 mm working distance) was used to evaluate
corneal structure. The system design and use of this microscope have
been previously described in detail.18,19Before examination, a drop of
a topical anesthetic, proparacaine hydrochloride 0.5%, was adminis-
tered to the cornea, and a drop of 2.5% hydroxypropyl methylcellu-
ages taken in the corneal periphery at
the Intacs region after 5 years of im-
plantation. (A) Intact superficial epithe-
lial cell layer. (B) Subepithelial nerves.
(C) Anterior stromal layer showing
dense population of keratocytes with a
slightly nonuniform distribution. (D)
Stromal layer anterior to Intacs seg-
structure, but a nonuniform distribu-
tion. (E) Stromal haze anterior to Intacs
segment. (F) Endothelial cell layer
showing typical honeycomblike struc-
ture (horizontal field width, 281 ?m).
Representative confocal im-
212L.T. LY ET AL.
Eye & Contact Lens, Vol. 32, No. 4, 2006
lose was placed at the tip of the objective to serve as an immersion
Waldwick, NJ) to allow for the alignment of the objective to the
region of interest. Sections of the peripheral cornea near the implant
were imaged. Real-time images of all layers of the cornea were
detected through the use of a low-light camera (VE1000 SIT, Dage-
videotapes were evaluated, and images were digitized and stored into
computer memory. Qualitative assessment of the corneal images was
performed, and special interest was paid to the stromal areas sur-
rounding the implant.
Six eyes underwent in vivo confocal microscopy follow-up 5
years after surgery. A normal epithelium was observed anterior to
the implant region in all six eyes (Fig. 1A), and subepithelial
nerves were detected (Fig. 1B). Anterior and posterior keratocytes
were morphologically intact (Fig. 1C, D), although cells often
appeared less evenly distributed than they would normally. No
keratocyte activation was observed, but diffuse haze (increased
reflectivity between cells) was detected in areas surrounding the
Intacs in all cases (Fig. 1E). A normal endothelial cell layer was
observed under the Intacs in all eyes (Fig. 1F).
The most striking findings were shiny, crystalline, or amorphous
deposits, coalesced along the central and peripheral curvature of
the implants and directly anterior to the segments (Fig. 2). Amor-
phous deposits appeared as punctate spots or larger objects with
varying structures (Fig. 2A–C). Refractile crystals, scattered
widely and distributed randomly, appeared as thin, linear, or flat,
rectangular structures (Fig. 2D–F). The severity of deposition
varied significantly; two corneas had only minor deposits, whereas
others showed large aggregates of unevenly dispersed crystals.
Deposits were detected in all six eyes by confocal microscopy, but
were detected in only five of the six eyes by clinical slitlamp
examination. Keratocytes surrounding the implant appeared to be
quiescent (Fig. 2B), and the implants themselves remained clear,
with no crystalline or amorphous deposits visible in the device.
Images digitized directly from confocal scans approximately
300 to 400 ?m central to the implant were also evaluated for
structural changes in the corneal tissue. Representative images of
the sublayers are shown in Figure 3. Overall, tissue in this area
remained fairly quiet in all corneas. Normal structure of superficial
epithelial cells was maintained, as indicated by visible cell borders
and bright nuclei (Fig. 3A). The epithelial nerve plexus was intact
(data not shown). In general, the corneal stroma and keratocytes
appeared quiescent, as indicated by weakly reflective nuclei with-
out evidence of keratocyte activation or nuclear fragmentation
ages of lipid deposits in lamellar chan-
nels. (A) Amorphous lipid deposits
along the edge of the Intacs (implant
surface on right). (B) Stromal kerato-
cytes of normal structure (arrows) were
seen just central to the Intacs (depos-
its on the right side of the image). (C)
Reflective material indicative of lipid
deposit formation near the posterior
surface of the Intacs. (D) Amorphous
lipid deposits and formation of crystals
at the anterior surface of the Intacs. (E
and F) Thin, needlelike, and flat, rect-
angular crystal deposits at the edge of
the Intacs (horizontal field width, 308
Representative confocal im-
213 INTRASTROMAL LIPID DEPOSITS WITH INTACS
Eye & Contact Lens, Vol. 32, No. 4, 2006
(Fig. 3B). Nerves of normal structure were detected in the middle
stroma (Fig. 3B). Normal endothelial cell organization was ob-
served, as evidenced by the typical hexagonal appearance (Fig.
3C). Thus, the principle stromal changes were localized to areas
directly surrounding the Intacs.
Confocal microscopy was performed on a single eye 2 months
after the Intacs implantation. Evaluation of corneal tissue sur-
rounding the segments showed few remarkable findings. Corneal
stromal layers generally appeared normal (Fig. 4A), although
evidence of slight edema and haze was observed in the stroma
anterior to the surface of the implants (Fig. 4B). Neither deposits
nor keratocyte activation was detected in the vicinity of the
As with all refractive procedures involving alloplastic implan-
tation in the cornea, long-term changes in the stroma or overlying
epithelium may occur. Intrastromal lipid deposition is commonly
observed in the cornea of patients with Intacs implantation.3,4
These deposits appear to fill in the potential spaces between the
implant surface and lamellar channels and do not seem to interfere
with the Intacs performance. In this study, in vivo confocal
microscopy showed that lamellar deposits occurred to some degree
in all eyes that were examined after 5 years of Intacs use. It is
noteworthy that deposits were detected in only five of six eyes by
slitlamp examination. Deposits of two types were observed adja-
cent to the edges of implants: amorphous deposits and highly
refractive, crystalline structures. Crystals were observed in four of
six eyes and had a flat rectangular or a needlelike structure. The
latter deposits resemble lipid crystals found in Schnyder’s crystal-
line stromal dystrophy,20,21whereas the rectangular crystals have a
structure resembling that of cholesterol deposits.15,22Parks et al.15
previously identified similar linear and rectangular structures in
the corneal stroma of monkey eyes implanted with a hydrogel
lenticule by using specular microscopy and also concluded that the
structures were lipid crystals. It is noteworthy that the lipid crystals
were highly localized to the implant surface; the cornea 300 to 400
?m central to the Intacs appeared normal by confocal microscopy.
A previous study using confocal microscopy to examine the
cornea of patients 2 to 15 months after Intacs implantation also
reported the appearance of amorphous deposits and linear, needle-
like structures adjacent to the segments. The authors characterized
the linear deposits as migratory fibroblasts,12cells most often
observed during the initial phases of wound healing. However, the
current observation of morphologically similar structures in pa-
tients 5 years after surgery arguably shows that the needlelike
structures more likely represent acellular aggregates of lipid crys-
talline deposits. Furthermore, no evidence of keratocyte activation
was seen at this late time point. It is noteworthy that the authors
also observed highly reflective basal cell nuclei in some corneas.12
Basal cells are not generally visible with our microscope, but we
noted a nonuniform distribution of keratocytes anterior to the
implant in some corneas. The endothelium posterior to the implant
appeared normal in both studies.
The presence of lipid deposits has been verified in rabbits and
primates after insertion of polymers into the stroma. Twa et al.16
evaluated the histologic changes in the corneal stroma after intra-
corneal ring segment implantation in rabbit eyes. The corneal
tissue response included activation of keratocytes, lipid deposition,
segments. (A) Normal superficial epithelial cell layer. (B) Normal stroma showing weakly reflective
keratocyte nuclei and a stromal nerve (arrow). (C) Endothelial cell layer showing typical hexagonal
structure (horizontal field width, 281 ?m).
Representative confocal images of corneal tissue taken 300 to 400 ?m central to the Intacs
ages taken 2 months after Intacs im-
plantation in the corneal periphery of a
patient. (A) Normal anterior stromal
layer. (B) Apparent edema and haze
anterior to the Intacs surface (horizon-
tal field width, 301 ?m).
Representative confocal im-
214 L.T. LY ET AL.
Eye & Contact Lens, Vol. 32, No. 4, 2006
and new collagen formation. Light microscopy and transmission Download full-text
electron microscopy showed intracellular accumulations of osmo-
philic and saturated lipid material, suggesting keratocyte synthesis
or phagocytosis of lipids.16Rodrigues et al.14identified the pres-
ence of neutral fat and unesterified cholesterol in deposits in the
corneas of Rhesus monkeys implanted with PMMA inserts and
proposed that corneal keratocytes produce lipid as a nonspecific
response to stress. Cogan et al.23,24have shown that rabbit corneal
keratocytes can synthesize fat under certain culture conditions.
They also performed immunostaining in the human cornea and
found that some stromal keratocytes showed apolipoprotein A-1
staining, a marker for high-density lipoprotein. They proposed that
the association of apolipoprotein A-1 with keratocytes could rep-
resent local synthesis of the lipoprotein or reflect its exogenous
uptake by keratocytes.25Deposit formation after Intacs implanta-
tion is generally characterized clinically by complete development
within the first 6 months after surgery. That deposits do not
increase in size or number after that argues against a chronically
active process. Taken together, the data suggest that lipid deposits
may result from alterations in lipid synthesis or metabolism by
activated corneal keratocytes during the initial wound healing
response after implantation.
We examined one patient 2 months postoperatively, and this
cornea did not show any signs of deposit formation or keratocyte
activation. It is noteworthy that the tunnels for Intacs in this patient
were created using IntraLase. It has been suggested that the use of
IntraLase to create thinner, more precise tunnels for Intacs inser-
tion may produce smaller unoccupied spaces within the lamellar
channels and thereby reduce deposit formation along the surface of
the implant segments. Long-term follow-up will determine whether
lipid deposits eventually develop in these corneas.
1. Holmes-Higgin DK, Baker PC, Burris TE, et al. Characterization of
the aspheric corneal surface with intrastromal corneal ring segments. J
Refract Surg 1999;15:520–528.
2. Burris TE, Holmes-Higgin DK, Silvestrini TA, et al. Corneal asphe-
ricity in eye bank eyes implanted with the intrastromal corneal ring. J
Refract Surg 1997;13:556–567.
3. Asbell PA, Uc ¸akhan O¨O¨. Long-term follow-up of Intacs from a single
center. J Cataract Refract Surg 2001;27:1456–1468.
4. Schanzlin DJ, Abbott RL, Asbell PA, et al. Two-year outcomes of
intrastromal corneal ring segments for the correction of myopia.
5. Chan SM, Khan HN. Reversibility and exchangeability of intrastromal
corneal ring segments. J Cataract Refract Surg 2002;28:676–681.
6. Clinch TE, Lemp MA, Foulks GN, et al. Removal of INTACS for
myopia. Ophthalmology 2002;109:1441–1446.
7. Hellstedt T, Makela J, Uusitalo R, et al. Treating keratoconus with
Intacs corneal ring segments. J Refract Surg 2005;21:236–246.
8. Boxer Wachler BS, Chandra NS, Chou B, et al. Intacs for keratoconus.
9. Lovisolo C, Fleming J. Intracorneal ring segments for iatrogenic
keratectasia after laser in situ keratomileusis or photorefractive kera-
tectomy. J Refract Surg 2002;18:535–541.
10. Bourges J-L, Trong TT, Ellies P, et al. Intrastromal corneal ring
segments and corneal anterior stromal necrosis. J Cataract Refract
11. Ruckhofer J, Twa MD, Schanzlin DJ. Clinical characteristics of
lamellar channel deposits after implantation of Intacs. J Cataract
Refract Surg 2000;26:1473–1479.
12. Ruckhofer J, Bohnke M, Alzner E, et al. Confocal microscopy after
implantation of intrastromal corneal ring segments. Ophthalmology
13. Fink AM, Gore C, Rosen ES. Corneal changes associated with intras-
tromal corneal ring segments [photo essay]. Arch Ophthalmol 1999;
14. Rodrigues MM, McCarey BE, Waring GO, et al. Lipid deposits
posterior to impermeable intracorneal lenses in Rhesus monkeys:
Clinical, histochemical, and ultrastructural studies. Refract Corneal
15. Parks RA, McCarey BE, Knight PM, et al. Intrastromal crystalline
deposits following hydrogel keratophakia in monkeys. Cornea 1993;
16. Twa MD, Ruckhofer J, Kash RL, et al. Histologic evaluation of
corneal stroma in rabbits after intrastromal corneal ring implantation.
17. Linebarger EJ, Song D, Ruckhofer J, et al. Intacs: The intrastromal
corneal ring. Int Ophthalmol Clin 2000;40:199–208.
18. Petroll WM, Jester JV, Cavanagh HD. In vivo confocal imaging:
general principles and applications. Scanning 1994;16:131–149.
19. Petroll WM, Jester JV, Cavanagh HD. Quantitative three dimensional
confocal imaging of the cornea in situ and in vivo: System design and
calibration. Scanning 1996;18:45–49.
20. Barchiesi BJ, Eckel RH, Ellis PP. The cornea and disorders of lipid
metabolism. Surv Ophthalmol 1991;36:1–22.
21. Weller RO, Rodger FC. Crystalline stromal dystrophy: Histochemistry
and ultrastructure of the cornea. Br J Ophthalmol 1980;64:46–52.
22. Fine BS, Townsend WM, Zimmerman LE, et al. Primary lipoidal
degeneration of the cornea. Am J Ophthalmol 1974;78:12–23.
23. Cogan DG, Kuwabara T. Experimental aberrant lipogenesis I. Serum
factor. Arch Pathol 1957;63:381–386.
24. Cogan DG, Kuwabara T. Lipogenesis by cells of cornea. Arch Pathol
25. Ashraf F, Cogan DG, Kruth HS. Apolipoprotein A-1 and B distri-
bution in the human cornea. Invest Ophthalmol Vis Sci 1993;34:
215 INTRASTROMAL LIPID DEPOSITS WITH INTACS
Eye & Contact Lens, Vol. 32, No. 4, 2006